Annual Report 2003-Project IP-3: Improved cassava for the

Transcription

Annual Report 2003-Project IP-3: Improved cassava for the
CONTENTS
Page
OUTPUT 1
Genetic base of cassava and related Manihot species evaluated and
available for cassava improvement
Effect of inbreeding in carotene content.
Sampling variation among and within cassava roots for carotene content.
Effect of processing cassava roots on their carotene content.
Effect of storage conditions on carotene content in cassava roots.
Planning workshop within the HarvestPlus initiative.
OUTPUT 2
Genetic stocks and improved gene pools developed and transferred to
national programs
Selection of progenitors based on previous cycle results and information from
other outputs (i.e., resistance/tolerance, root quality traits, etc.).
Establishment of crossing blocks and production of recombinant seed from
previously established blocks.
Generation and distribution of advanced breeding materials for National
Programs.
Selection of recombinant progenies for broad and specific adaptation within
major agro-ecosystems.
Selections for the Sub-Humid Tropical Environment
Selections for the Acid-Soil Savannas Environment
Selections for the Mid-altitude valleys
Selections for the Highlands
Selections for the Middle Magdalena Region
Selections for the Tolima-Huila Departments.
OUTPUT 3
Collaboration with other institutions
Support national programs that have traditionally collaborated with CIAT in
the development and improvement of cassava
Development of collaborative projects with partners in Africa, Asia and Latin
America and the Caribbean
The collaboration with Colombia.
OUTPUT 4
breeding
integrating
Developing
new
approaches
for
cassava
biotechnology tools and quantitative genetics principles
Development of a breeding scheme based on the use of doubled-haploids.
Cassava as a crop and breeding objectives
State of the art of using haploid technology for breeding major food crops.
State of the art for the development and use of haploid technology in self and
crossbreeding species.
Research Strategy for the Use of Inbreeding in Cassava Genetic Improvement.
Research Strategy for the Development of Haploid Technology in Cassava.
Conclusions.
List of participants
Acknowledgement.
Development of a novel approach for the analysis of diallel mating designs for
better understanding the inheritance of quantitative traits.
Index-I
1-1
1-3
1-5
1-6
1-9
2-1
2-4
2-6
2-8
2-14
2-31
2-52
2-61
2-64
2-68
3-1
3-3
3-5
4-1
4-1
4-3
4-4
4-5
4-8
4-13
4-14
4-15
4-16
2003 Annual Report
cont. OUTPUT 4
Page
Selections for the Mid-altitude valleys.
Selections for the Sub-Humid Tropical Environment
Selections for the Acid-Soil Savannas Environment
An alternative way to use the information from diallel studies.
Introduction of inbreeding in cassava.
4-18
4-19
4-20
4-21
4-27
OUTPUT 5
Activities related with the maintenance of the germplasm bank of cassava
and other Manihot species
Maintenance of Manihot germplasm bank in the field.
Evaluation of M. esculenta and related species from the germplasm collection
for useful traits, particularly for higher protein content in the roots.
OUTPUT 6
Breeding for insect and other arthropods resistance and development of
alternative methods for their control
Evaluation of cassava germplasm for resistance to whiteflies (Aleurotrachelus
socialis) during 2002-2003.
Manihot wild species and hybrids evaluated for whitefly (A. socialis)
populations damage at Santander de Quilichao during 2002-03 crop cycle.
Evaluation of cassava dialelic crosses for whitefly (A. socialis) resistance,
using MECU 72 as the resistant female parent (at Santander de Quilichao).
Evaluation of cassava germplasm in several breeding and genetic trials for
whitefly (A. socialis) damage at Santander de Quilichao (Cauca).
Studies on whitefly (Aleurotrachelus socialis) resistance mechanisms in
selected cassava genotypes.
Studies on the biology and development of biotype B of Bemisia tabaci on
cassava, Manihot esculenta and the wild species, Manihot carthaginensis.
Cassava germplasm evaluations for arthropod pest damage at several
localities on the Colombia Atlantic Coast.
Evaluation of cassava germplasm for resistance to cassava green mite,
Mononychellus tanajoa.
Wild Manihot species as a source of resistance to cassava arthropod pests.
Identification of genomic regions responsible for conferring resistance to white
fly (Aleuroptrachelus socialis) in cassava
Dissemination of the generated knowledge: Training, thesis, publications,
conferences, seminars and proposals
OUTPUT 7
Disease Resistance in Cassava
Characterization of family GM 315 (MNGA 19 x CM 9208-13) for resistance to
cassava bacterial blight under greenhouse conditions.
Evaluation of cassava genotypes for their resistance to superelongation disease
(SED) in the greenhouse.
Evaluation of cassava genotypes for resistance to CBB and SED in
Villavicencio
Project IP3: improving cassava for the developing world
5-1
5-2
6-1
6-9
6-11
6-12
6-15
6-25
6-30
6-33
6-38
6-47
6-53
7-1
7-3
7-4
Index-II
cont. OUTPUT 7
Page
Characterizing F1 progeny of four backcross families (MNGA 19 crossed with
each of CM 9208-13, CM 9208-26, CM 9208-31, and CM 9208-73) for
resistance to CBB in Villavicencio.
Evaluating a diallel assay in Villavicencio and Santander de Quilichao.
Evaluation of cassava genotypes for resistance to SED in Santander de
Quilichao.
Using polymerase chain reaction (PCR) with degenerate primers to search for
resistance gene analogs associated with resistance to CBB.
Identification of gene analogs for resistance to cassava (Manihot esculenta
Crantz) diseases, and their relationship to resistance to three Phytophthora
species.
Evaluating a cassava family for resistance to Phytophthora root rot.
Evaluating simple sequence repeats markers linked to bacterial blight
resistance in cassava
Detection of a phytoplasma associated with cassava Frogskin Disease (FSD) in
Colombia.
Transmission of a phythoplasma affecting cassava seedlings and identification
of indicator plants. FSD symptoms recede after chlortetracycline treatment.
Identification of cultural practices and strategy to control frogskin disease in
cassava.
Evaluation of the influence of the soil as a source of FSD vectors.
Isolation and characterization of Agrobacterium tumefaciens from soil and
cassava roots.
Selection and treatment of cassava stakes trial for the management of
diseases and evaluation of genotypes at Sincelejo (Sucre).
Multiplying promising cassava genotypes to ensure sufficient planting
material for both greenhouse and field experiments.
Evaluation of different treatments with ScoreÒ 250 E.C (Difenoconazole) for
the management of SED, Sincelejo (Sucre)
Multiplying promising cassava genotypes to ensure sufficient planting
material for both greenhouse and field experiments
Genetics of resistance to rot caused by Phytophthora tropicalis in two
segregating populations of cassava (Manihot esculenta Crantz).
Training farmers, technicians, and extension agents in participatory research,
cassava and other crops management, and disease control strategies.
OUTPUT 8
Development and use of biotechnology tools for cassava improvement
Molecular Marker-Assisted Breeding for Resistance to the Cassava Mosaic
Disease in Latin American Cassava Gene pools.
Molecular Marker-Assisted and Farmer Participatory Improvement of Cassava
Germplasm for Farmer/Market Preferred Traits in Tanzania
Molecular Marker-Assisted Selection (MAS) for Breeding Early Root Bulking in
Cassava
Genetic Mapping of Beta-Carotene Content
Genetic Mapping of Dry Matter Content (DMC)
Genetic Mapping of Cyanogenic Potential (CNP)
Genetic Mapping of Leaf Retention
Index-III
7-6
7-7
7-9
7-11
7-17
7-21
7-22
7-26
7-31
7-35
7-38
7-39
7-45
7-49
7.47
7-48
7-48
7-49
8-1
8-1
8-4
8-8
8-11
8-13
8-17
8-19
2003 Annual Report
cont. OUTPUT 8
Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local
Cassava Varieties from Guatemala
Characterization of genetic diversity:
Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local
Cassava Varieties from Ghana and Predictability of Heterosis
Characterization of genetic diversity
Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local
Cassava Varieties from Uganda
Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local
Cassava Varieties from Sierra Leone
Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local
Cassava Varieties from Cuba
Report of the Second Tri-Annual Workshop of the Molecular Genetic Diversity
Network of Cassava (MOLCAS)
New Collaborative Arrangements, Networks, Databases, Training and
Workshops
Development of molecular techniques for assessing genetic diversity and
mapping of useful genes
Mining the Primary Gene Pool of Cassava: Introgression of High Root Protein
from Accessions of Manihot esculenta sub spp Fabellifolia and Manihot
Tristis into Cassava
Mining the Primary Gene Pool: Green Mites (CGM) Resistance Genes from
Manihot esculenta sub spp Fabellifolia.
Positional Cloning of CMD2 the Gene that Confers High Level of Resistance to
the Cassava Mosaic Disease (CMD)
Progress in the Anti-Sense Mediated Silencing of the Granule Bound Starch
Synthase I (GBSS I) for the Production of Waxy Cassava Starch
Evaluation of S1 Families for Waxy Mutants
Irradiation of Sexual Seeds for the Production of Waxy Cassava and other
Mutants
Saturation of the Molecular Genetic Map of Cassava with PCR-based Markers:
Progress on the Mapping of a New Set of 140 New SSR Markers
Embryo Rescue of Sexual Seeds from Breeding Populations for Molecular
Assisted Selection (MAS) of CMD Resistance
Dissemination of Improved Cassava Varieties as Tissue Culture plantlets
Updating CassavaDB, a ACEDB-type data base for Results of Genome
Mapping
Associating Horn Worm Resistance in 60444 (MNG11) with Introgression from
Manihot Glaziovii.
Development of Populations Tolerant to Inbreeding Depression in Cassava
A Simple Method for the Rapid Multiplication of Clean Cassava Planting
Material
OUTPUT 9
Integrated cassava-based cropping systems in Asia: Widespread adoption
of farming practices that enhance sustainability
Soil fertility maintenance through the application of chemical fertilizers, or the
use of intercropping, green manuring, alley cropping and crop rotations
Project IP3: improving cassava for the developing world
Page
8-21
8-28
8-36
8-36
8-43
8-46
8-48
8-50
8-52
8-58
8-65
8-68
8-71
8-75
8-79
8-83
8-85
8-89
8-93
8-94
8-96
8-98
9-1
9-1
Index-IV
cont. OUTPUT 9
Page
Long-term NPK trials
Soil Improvement trial in Hung Loc Center
Development of efficient and economical soil preparation practices.
Determination of the response to micronutrients and Mg, and the most
effective way of applying Zn in calcareous soils.
Evaluation of the use of plastic mulch for weed and erosion control.
Evaluation of cassava varieties and determination of optimum NPK
fertilization, plant spacing, cutting height and frequency for cassava leaf
production.
Conducting FPR trials on varieties, fertilization, weed control, green manures,
intercropping, erosion control and pig feeding in Thailand, Vietnam and
China.
Enhancing adoption of new varieties and improved management practices
through farmer participatory research (FPR) and extension (FPE) activities.
Index-V
9-1
9-3
9-9
9-11
9-12
9-16
9-23
9-28
2003 Annual Report
OUTPUT 1
Genetic base of cassava and related Manihot species evaluated and available for
cassava improvement.
The overall objective of this output is to generate genetic stocks and knowledge about genetic
variability for a more efficient cassava genetic improvement. The main activities focus in
improving the nutritional status of people living in marginal environments of the tropics, by
selecting and promoting cassava genotypes with high and good bio-availability of
micronutrients and vitamins. Related traits are the need for a better understanding of the
biochemical and genetic basis of post-harvest physiological deterioration and starch quality
traits. Because of the scarcity of institutions conducting research on cassava new methods
for assessing and exploiting genetic variability are required and need to be developed by
CIAT. A partnership with scientists conducting research on forestry (Uppsala University,
Sweden) is proving to be very useful.
Because of the nature of the research described in this output, it is one of the many
collaborative activities between projects SB2 and IP3. To maintain some coherence through
this report some of the activities reported herein may also be used by project SB2 as well.
Activity 1.1. Effect of inbreeding in carotene content.
Inheritance of carotene content in cassava has been described as relatively simple (Iglesias,
et al. Euphytica 94:367-73), but a definite mode of inheritance has not been clearly specified.
A set of 43 S1 lines from the cross between MBRA 337 (very strong orange coloration in the
roots) and CM 523-7 was produced. S1 means that the germplasm is the result of a selfpollination. Results from the first evaluation are presented in Table 1.1. As expected there
was excellent segregation for color intensity (rating from 1 to 6), which is very closely related
to carotene content. Six stakes per line were obtained but three S1 clonal lines did not
survive for one reason or another. The 40 remaining S1 clonal lines were planted at
Santander de Quilichao station (Cauca Department, Colombia), so adequate amount of roots
will be produced from each line.. Molecular analysis will be made on these plants in search of
a molecular marker that could be used for identifying the region(S) where the gene(s) for
yellowness in the roots are located.
Table 1.1. Results from the 43 S1 lineages derived from the cross between MBRA 337 (yellow
roots) and CM 523-7 (white roots) from which several plants were planted at
Santander de Quilichao for proper measurement of carotene contents.
Average
St.Dev
Min
Max
Harvest
Index
(0 to 1)
0.45
0.10
0.18
0.62
Dry matter
Fresh
content
root yield
(%)
(t/ha)
34.47
37.21
6.02
24.91
16.07
0.00
45.73
131.00
Dry root
yield
(t/ha)
12.90
9.49
0.00
49.64
Root
color
(1 to 9)
3.02
2.02
1.00
6.00
HCN
(ppm)
PPD
(%)
8.54
1.07
5.00
9.00
1.73
2.92
0.00
13.83
HCN= cyanogenic potential; PPD= post-harvest physiological deterioration
Output 1-1
2003 Annual Report
In addition 16 S1 lines obtained from the elite clone MTAI 8 (Rayong 60, derived from the
cross between MCOL 1684 and Rayong 1) were evaluated. MTAI 8 (= Rayong 60) has roots
that are almost white in color. The S1 progenies, surprisingly, exposed a large range of
variation for carotene content. In Table 1.2 the results of these 16 S1 lines, and the parental
material MTAI 8, are presented. One of the observations made after the use of HPLC is that
the coefficient of variation tends to be higher in HPLC than in colorimetry. Further
refinement of the HPLC methodology is required to reach acceptable repeatability and/or
precision of the results.
Improvement of carotenoid quantification in cassava roots.
Moreover, the overestimation of HPLC (or underestimation by the colorimetric method) seems
to be much higher at high values of carotene content. It is possible that the increased
difference between the colorimetric and HPLC methods at high levels of carotene is due to
lack of adequate fit of the standardization curves for one or both methods. This explanation
makes sense because the curves were originally devised for a range of carotene content lower
than 1 mg / 100 g tissue. It is not clear yet if it is a matter of overestimation by HPLC or
underestimation by colorimetry.
Table 1.2. Dry matter and carotene content in roots from 16 S1 lines derived from MTAI 8
(Rayong 60) measured by the colorimetric and HPLC methods on fresh and dry tissue.
AM 320-140
AM 320-136
AM 320-135
AM 320-133
AM 320-123
AM 320-143
AM 320-147
AM 320-127
AM 320-124
AM 320-146
AM 320-139
AM 320-144
AM 320-142
AM 320-138
AM 320-130
AM 320-121
MTAI 8
Average
St.Dev
Min.
Max.
Dry matter
(%)
37.47
33.62
34.47
32.61
37.36
36.89
42.69
37.20
38.86
40.74
33.78
33.91
36.41
37.63
31.53
40.82
39.98
36.63
3.17
31.53
42.69
Colorimetric method
Carotene content (mg/100 g)
Fresh tissue
Dry tissue
1.01
2.71
0.83
2.48
0.87
2.54
0.81
2.49
0.44
1.19
0.36
0.98
0.37
0.86
0.41
1.11
0.31
0.80
0.32
0.78
0.26
0.77
0.26
0.77
0.15
0.41
0.14
0.37
0.14
0.44
0.14
0.35
0.27
0.69
0.43
1.19
0.29
0.85
0.14
0.35
1.01
2.71
Project IP3: improving cassava for the developing world
HPLC method
Carotene content (mg/100 g)
Fresh tissue
Dry tissue
1.76
4.71
1.66
4.93
1.78
5.17
1.49
4.58
0.66
1.78
0.25
0.68
0.40
0.93
0.49
1.32
0.40
1.04
0.42
1.04
0.21
0.63
0.26
0.77
0.10
0.27
0.10
0.27
0.11
0.34
0.10
0.24
0.31
0.78
0.64
1.79
0.64
1.87
0.10
0.24
1.78
5.17
Output 1-2
Results presented in Table 5 clearly suggest that carotene content can indeed be increased
over the natural ranges so far obtained. It is important to emphasize that the highest
concentration of carotenes based on fresh tissue so far found had been 1.06 mg / 100 g
fresh tissue. Four of the sixteen S1 clonal lines evaluated showed much higher levels of
carotene concentration (around 1.70 mg / 100 g fresh tissue). Also promising is the fact that
these high values were obtained from a clon (MTAI 8 or Rayong 60) whose actual
concentration of carotene was rather low (0.31 mg / 100 g fresh tissue).
Activity 1.2. Sampling variation among and within cassava roots for carotene
content.
Rationale
Considering that the HarvestPlus initiative has already been launched, several steps have
been taken to standardize the sampling and measuring procedures to be able to obtain
uniform and comparable data. This activity aimed at determining variation in carotene
content from different samples of roots from a single clone.
Specific Objectives:
a) To measure carotene content in roots samples from different plants, as well as for the same
plant, from a single genotype.
b) To determine carotene content in samples root taken at different but specific positions within
the root.
Materials and methods
Harvesting and sampling. Root samples from 11 plants from the clone CM 2772-3 were taken
at about 10 months of age. Three commercial size roots per plant were randomly selected.
From each root, samples were taken from the proximal, central and distal sections of the
root. For each section 5 g samples were taken from the outside, intermediate and core of the
along the radial axis of the root (Figure 1.1.)
Carotenoid quantification by the colorimetric method. The extraction procedure outlined by
Safo-Katanga et al. (1984), was modified by extracting root parenchyma with petroleum
ether. The extraction protocol for leaves had to be modified due to the presence of tannins
and chlorophylls. The modified protocol included several extractions with petroleum ether
35-65 ºC and washing steps with methanol in order to minimize the interference from the
others pigments. A sample of 5 g was taken out of the root or leaves, randomly selected 10 to
11 months after planting. The quantification was done by ultraviolet spectrophotometry
using a Shimadzu UV-VIS 160A recording spectrophotometer. UV detection was done at l =
455nm for root extracts and l = 490 nm for leaf extracts.
Results
A total of 279 carotene measurements were taken. Three roots were obtained from 9 of the
plants, whereas for the two remaining plants only two roots of commercial size and health
could be obtained. Therefore a total of 31 roots were analyzed. For nine samples (along and
across each root), there are four different sources of variation in carotene measurement that
could be determined in this study, as follows:
Output 1-3
2003 Annual Report
a) Variation among plants within the same clone. Mean carotene content of the whole
experiment was 0.355 mg / 100 g FT with a standard variation of 0.028 mg. The range for
averages was from 0.293 to 0.391 mg / 100 g FT.
b) Variation among roots from the same plant. The standard deviation for variation among
roots within each plant was 0.034 mg / 100 g FT. The largest difference between roots from
the same plant was 0.131 mg / 100 g FT and the lowest was 0.007 mg / 100 g FT. Mean
difference between the root with highest and lowest average carotene content (within the
same plant) was 0.061 mg / 100 g FT.
c) Variation along the longitudinal axis of the root There was a trend for higher levels of
carotene in the proximal sections of the roots. For each section position the data from the
external, intermediate and internal samples were averaged.
Mean carotene contents for the proximal central and distal sections were, respectively, 0.406
(+ 0.045), 0.387 (+ 0.063) and 0.370 (+ 0.049) mg / 100 g FT. Therefore carotene content
tended to be higher in the proximal sections and lower in the distal parts of the roots. Based
on the standard deviations of these values, the central section showed the highest variation.
d) Variation across the radial axis of the root. There was a trend for higher levels of carotene in
the internal samples of the roots. For each sample position the data from the proximal,
central and distal sections were averaged. Mean carotene contents for the external,
intermediate and internal samples were, respectively, 0.351 (+ 0.048), 0.404 (+ 0.074) and
0.409 (+ 0.069) mg / 100 g FT. Therefore carotene content tended to be higher in the internal
and lower in the external parts of the roots. Based on the standard deviations of these
values, the intermediate section showed the highest variation.
Proximal section
Central section
Distal section
External sample
Intermediate sample
Internal sample
Figure 1.1. Distribution of samples taken along the longitudinal and radial axis of the
cassava root for the analysis of distribution of carotene concentration in different
positions.
Project IP3: improving cassava for the developing world
Output 1-4
These results are useful for supporting the need for a standardized procedure for sampling
the roots in carotene measurements. Although the averages obtained were not different
enough to reach statistical significance, individual samples were different enough (0.214 to
0.742 mg / 100 g FT) to justify the need to define the best sampling procedure. At this point
different approaches to reduce the variance for carotene content among samples are being
analyzed.
Activity 1.3. Effect of processing cassava roots on their carotene content.
Rationale
For the improvement of the nutritional status of people it is not enough to produce cassava
clones with high carotene content. The carotenoids found in the roots need to be absorbed
and transformed by the human body and turned into vitamin A (retinal), and the effect of
processing needs to be taken into account. The original levels of carotenes in unprocessed
roots may not be the same after the roots are processed.
Specific Objectives:
a) To determine the amount of carotenes lost after different processing procedures.
Materials and methods
Roots from four varieties were used for this study. Different root processing methods were
evaluated (boiling, sun drying, oven drying, gari and lyophilized roots). Carotene content was
measured following the methods described above. For each variety/processing method four
different measurements were obtained. Carotene content was also measured in fresh roots.
Results
Table 1.3 Carotene content in fresh and processed roots from four cassava clones and true
retention (%) upon processing (compared with fresh root contents). Results are the
average of four replications.
Processing
method
Fresh
Boiled
Oven dried
Sun dried
Gari
Liophilized
Boiled
Oven dried
Sun dried
Gari
Liophilized
Output 1-5
Clone 1
2.250
1.577
0.830
0.584
0.499
1.690
68.50
39.71
25.49
22.07
73.74
Clone 2
Clone 3
Clone 4
Carotene content (mg / 100 g FT)
1.683
4.506
7.143
1.607
2.445
3.055
1.426
0.937
1.055
1.572
1.138
1.340
0.430
0.550
0.682
2.170
3.367
5.347
True retention (%)
76.08
47.41
45.49
62.11
28.75
18.60
42.20
34.08
28.82
23.11
19.54
16.63
95.56
96.28
86.58
Average
3.896
2.171
1.062
1.159
0.540
3.143
59.37
37.29
32.65
20.34
88.04
2003 Annual Report
Table 1.3 presents the averages for carotene content in fresh and processed roots from the
four cassava clones employed in this study. Also true retention (%) values, upon processing,
are presented. Results agree with preliminary data obtained before. Among the processing
methods liophilization was tested, not because it is a way for cassava consumption, but
because it is one alternative method for storing cassava roots for delayed measurement after
harvest. Among the processing methods for cassava consumption boiling resulted in the
highest values of retention (about 60%). Drying cassava roots reduced carotene contents to
lower levels than boiling, with retention levels ranging from 37.3 (oven drying) down to 32.7%
(sun drying). Gari (a popular processing method for cassava consumption in Africa) yielded
the lowest retention levels (20%). The effect of silage on carotenoid content will be probably
measured in 2004, because of its relevance in animal feeding in Asia which ultimately will
also favor the nutritional status of people, but indirectly.
Activity 1.4. Effect of storage conditions on carotene content in cassava roots.
Rationale
Cassava is currently harvested within a period of two months (May and June) every year at
CIAT experimental station. This creates a bottleneck because all the germplasm need and
can only be evaluated during this period. Laboratory facilities and protocols allow for the
determination of only a limited number of clones per day (up to 40 samples for the
colorimetric method and 8 samples for HPLC). Therefore, studies were conducted to evaluate
the possibility of harvesting cassava roots at the normal dates and storing them under
special conditions that will allow measurements 2-6 months later without substantial
modification in carotene contents or quality.
Specific Objectives:
a) To analyze the effect of different storage conditions on carotene content.
Materials and methods
Stability of carotene content was measured on root samples from the clone MCOL 2508 by
the colorimetric and HPLC methods on fresh roots. This information was used as a check for
data obtained after different periods of storage at – 80 °C. Samples of roots were stored for
22, 73, 84, 96, 126, 210 and 230 days and then, carotene contents evaluated using the same
measuring methods.
In a separate experiment roots from 20 different clones were analyzed by colorimetry soon
after harvest and then stored for six months at –20 °C. The root samples were then recovered
and analyzed again for carotene content.
Results
Storage at –80 °C.
Results from this experiment are summarized in Figure 1.2. As it was usually the case, total
carotene content measured by HPLC was slightly higher than when measured by the
colorimetric method (1.07 vs. 0.88 mg / 100 g fresh tissue). Samples of roots analyzed by the
colorimetric method averaged 0.88 mg across the different dates (the same value as the
original samples) with a standard deviation of 0.038. Values ranged from 0.81 to 0.92 mg /
100 g fresh tissue. Measurements based on HPLC showed larger variation. The average
content was 1.01 mg / 100 g fresh tissue, with a standard deviation of 0.156. Individual
Project IP3: improving cassava for the developing world
Output 1-6
measurements ranged from 0.81 to 1.26 mg. There was no clear trend suggesting that
carotenes measurements will decrease or increase as a consequence of storage. Therefore, it
is valid to conclude that root samples can be stored for as long as seven months until
carotenes are finally measured. The logistic advantages of this conclusion are very important
for different projects involving cassava and carotenoid compounds.
Carotene content (mg/100 g fresh tissue)
1.40
1.20
B
1.00
Measurement by HPLC
A
0.80
Measurement by colorimetry
0.60
0.40
Total-Colorimetry
Total-HPLC
0.20
Linear (Total-Colorimetry)
Linear (Total-HPLC)
0.00
0
50
100
150
200
250
Days under storage
Figure 1.2. Stability of carotene content measured on root samples from the clon MCOL
2508.Original levels of carotenes were measured by the (A) colorimetric and (B) HPLC
methods on fresh roots. Samples of roots stored for 22 to 230 days at – 80 °C were
taken and carotene contents evaluated using the same measuring methods.
Storage at –20 °C.
Carotene content was measured with the colorimetric (Figure 1.3) and HPLC (Figure 1.4)
methods. During the storage period the roots were maintained at – 20 °C in open petri
dishes. There was a differential dehydration process of root samples from different clones,
depending on their position in the storage container. This may explain the obvious lack of
correlation between non-stored and stored roots samples of some clones (MCOL2435,
MCOL2508, MCOL 2266, AM 273-7 and AM 273-23).
Similar results were obtained when carotenes were evaluated by the HPLC method (Figure
1.4). In this case, the lack of association between data from non-stored and stored root
samples was found in clones MCOL 2580, MCOL 2410, MCOL 1468, and MCOL
2596.Because of the errors associated after storing roots at - 20 °C using both the
colorimetric and the HPLC methods the potential of maintaining root samples at that
temperature needs further confirmation under closed containers. The next evaluations
should consider the potential of liophylization as well. Until then, roots samples should be
maintained at –80 °C.
Output 1-7
2003 Annual Report
4
3.5
Non-stored roots
Carotenes (mg/100 g dry tissue)
3
2.5
2
1.5
1
0.5
Roots stored at -20 C
GM 319-13
GM 319-40
AM 273-23
AM 273-7
GM 319-22
GM 319-20
GM 319-11
GM 319-8
GM 319-28
GM 319-6
MCOL 2596
MCOL 2474
MCOL 2266
MCOL 1468
MPER 600
MBRA 501
MCOL 2410
MCOL 2580
MCOL 2508
MCOL 2435
0
Clon
Figure 1.3. Comparison of carotene content (measured by the colorimetric method) on roots
from 20 clones at harvest time and then after 6 months of storage at –20 °C.
5
4.5
Roots stored at - 20 C
Carotene content (mg / 100 g dry tissue)
4
3.5
3
2.5
2
1.5
1
0.5
Noon-stored roots
GM 319-13
GM 319-40
AM 273-23
AM 273-7
GM 319-22
GM 319-20
GM 319-11
GM 319-8
GM 319-28
GM 319-6
MCOL 2596
MCOL 2474
MCOL 2266
MCOL 1468
MPER 600
MBRA 501
MCOL 2410
MCOL 2580
MCOL 2508
MCOL 2435
0
Clon
Figure 1.4. Comparison of carotene content (measured by the HPLC method) on roots from
20 clones at harvest time and then after 6 months of storage at –20 °C
Project IP3: improving cassava for the developing world
Output 1-8
Activity 1.5. Planning workshop within the HarvestPlus initiative.
Rationale
The Biofortification Global Challenge Program now renamed Harvest Plus will start its
activities in 2004. Several actors in Asia, Africa and Latin American and the Caribbean will
participate in the development and promotion of cassava clones with high-carotene content
in the roots. A Planning Workshop was held near Kampala (Uganda) where most of these
actors could meet each other for developing a coherent and realistic work plan for the next
ten years. A total of 42 people from the CG System, NARs, NGOs and research institutions
from developed countries participated in this dynamic and productive meeting.
Specific Objectives:
a) To present updated information regarding nutritional quality issues in cassava.
b) To allow scientists in the area of biotechnology, human and animal nutrition, breeders and
development organizations to meet and interact among themselves.
c) To develop a general work plan for the next ten years and a detailed one for year 1 and 3
within the initiative.
Results
All the objectives of the workshop were met and the required documents were produced and
distributed. A formal proposal to be presented at the Harvest Plus board was delivered on
time for its evaluation by mid-November.
1.6 References
Beeching, J.R., Yuanhuai, H., Gómez-Vázquez, R., Day, R.C., Cooper R.M. 1998. Wound and
defense responses in cassava as related to post-harvest physiological deterioration. In:
J.T.Romeo, K.R. Downum and R. Verpporte (Eds.) Recent Advances in Phytochemistry.
Phytochemical Signals in Plant-Microbe Interactions. Plenum Press, New York-London. Vol.
32:231-248.
Brücher, H. 1989. Useful plants of neotropical origin and their wild relatives. Springer-Verlag.
Berlin and New York. 296 p.
CIAT, 2001. Improved cassava for the developing world. Annual Report, 2001.
Safo-Katanga, O., Aboagye, P., Amartey, S.A., Olaham, J.H., 1984. Studies on the content of
yellow-pigmented cassava. In: Terry, E.R. et al. (Eds). Tropical Roots Crops Production and
Uses in Africa. IDRC, Ottawa, Canada. pp.103-104.
van Oirschot, Q. E. A., O’Brien G.M., Dufour D., El-Sharkawy M.A. and Mesa E., 2000. The
effect of pre-harvest pruning of cassava upon root deterioration and quality
characteristics. J. Sci. Food Agric. 80:1866-1873.
Wheatley, C., Lozano, C. Gomez G., 1985. Post-harvest deterioration of cassava roots, In:
Cock, J. H., Reyes, J. A. (Eds). Cassava: Research, Production and Utilization. UNDPCIAT, Cali pp 655-671.
Zapata, G., 2001.Disminución de deterioro fisiológico postcosecha en raíces de yuca (Manihot
esculenta Crantz) mediante almacenamiento controlado. B.S. Thesis, Universidad de San
Buenaventura, Facultad de Ingeniería Agroindustrial. Cali, Colombia. 84 pp.
Output 1-9
2003 Annual Report
OUTPUT 2
Genetic stocks and improved gene pools developed and transferred to national
programs.
The overall objective of this output is to produce genetically improved cassava
germplasm, by recombining selected parental genotypes and then evaluating the
segregating progenies under adequate environmental conditions. Recombinant seed
and/or vegetative propagules from elite clones are then shipped to our collaborators in
Africa, Asia and Latin America. The activities described below may not follow the exact
order used to describe them in the respective work plan. This change has been made
for being more logical and, hopefully, to make it easier to understand the description
of the research carried out. In addition to germplasm we are also producing knowledge
and developing technologies that will make the breeding process more efficient.
Activity 2.1 Selection of progenitors based on previous cycle results and
information from other outputs (i.e., resistance/tolerance, root quality
traits, etc.).
Rationale
The selection of parents to build populations for future breeding work represents the
core of our improvement efforts, since it will be the source of the genetic progress we
will achieve in the future. There are two types of populations developed: open
pollinated and controlled crosses. We generally employ open pollination (polycrosses)
to develop populations for target ecosystems.
We have consistently developed
polycrosses for the sub-humid tropics, acid soil savannas, semi-arid tropics, midaltitude and highland tropics, and sub-tropics. In the case of controlled crosses, yhey
are to develop progenies for specific traits, special studies or the combination of elite
experimental material with local landraces that need to be improved, but they can also
be used for adaptation to target ecosystems as well.
Specific Objectives
a) To identify, a set of elite clones, based on information from evaluation trials at
several locations, and new objectives defined for the project. These clones are
recombined to start a new cycle of selection.
b) To include as progenitor, for each agro-ecological zone, at least one genotype with
high-carotene, yellow roots
c) To base the selection of parental lines increasingly on information from the
performance of their progenies (≈ general combining ability or breeding value).
Materials and Methods
Only genotypes that have been selected over 2 consecutive years in Advanced Yield
Trials are selected to participate as parents for the following generation. Among those
genotypes, clones with outstanding performance for the most important agronomic
traits are selected. After the analysis of results is conducted with data across two
years, those genotypes exceeding at least one standard deviation from the overall
mean are considered as parents for the next generation. Sometimes, landraces or
already released cultivars, that can contribute special features to the progenies
Output 2-1
2003 Annual Report
generated are also included. Lately, thanks to the modifications introduced to the
evaluation process selection of parents is greatly affected by data of the progenies they
produce (≈ general combining ability). It is envisioned that about 15-20% of the
parental lines will be changed, eliminating those with poor general combining ability
and introducing new clones that have had outstanding performance per se in
Advanced Yield Trials to assess their breeding value.
The information provided by pathologists, entomologists and quality specialists in
relation to sources of resistance or special traits is used to select genotypes for
controlled crosses. These controlled crosses are developed upon specific requests from
National Programs that want their main landrace, or released varieties, crossed to
genotypes with specific traits; or requests from CIAT scientists that want to pyramid
genes, or develop segregating progenies for gene tagging.
As will be described below, one of the major changes introduced in the cassava
breeding scheme at CIAT has been to take and record data on all progenies starting at
the first evaluation stage (Clonal Evaluation Trials). The kind of information obtained
allows a gross estimation of general combining ability (simply defined, it is the capacity
of an individual to produce a good progeny) of parental lines employed in generating
the clones included in those trials. This information is increasingly influencing the
decisions of materials that will continue to be used as parents and those that will not.
Significant changes were introduced during the 2002-2003 growing season by
blocking the Clonal Evaluation Trials to reduce the large effects that the
environmental variation within these large trials had on the average performance of
each family. Basically this chages follow the ideas described by Gardner in 1961 for
stratified phenotypic mass selection.
Results
The parents selected for the development of gene pools targeted to specific ecosystems
is presented in Table 2.1. The agronomic performance of these materials is described
further down in this document. Seed will be harvested from July, 2004 through
December, 2004. F1 plants will grow until the planting of the trials early in 2005. A
major decision to take in the genetic improvement of crops is how to choose materials
for use as parents that will produce new varieties with increased production potential
and adequate adaptation to the environmental conditions under which they will be
cultivated.
The principal criterion for selecting parents to date has been their performance per se.
Unfortunately, however, good clones do not necessarily give rise to good progeny,
hence the need to precisely estimate the traits that the progeny of each individual will
produce. Until now, data was recorded starting at the Preliminary Yield Trials, which
meant that no balanced information was available on all progeny produced by a given
individual, but only on those that had passed the first stages of selection. The new
modality implies taking data for all and each clone evaluated, whether or not it will be
eventually selected. This permits the development of a solid database for selecting
parents in terms of the progeny they produce (which, from the genetic viewpoint, is
what really matters) and not merely based on their innate traits, as was done in the
past.
Project IP3: improving cassava for the developing world
Output 2-2
Table 2.1. Parental lines to be used in crosses for different ecosystems, relevant for
cassava production in the world.
Sub-humid
tropics
CG 1141-1
CM 2772-3
CM 3306-4
CM 4365-3
CM 4919-1
CM 6119-5
CM 6754-8
CM 6758-1
CM 7514-8
CM 8027-3
CM 8475-4
CM 9067-2
SGB 765-2
SGB 765-4
SM 805-15
SM 1411-5
SM 1433-4
SM 1438-2
SM 1511-6
SM 1521-10
SM 1565-17
SM 1650-7
SM 1669-5
SM 1669-7
SM 1759-29
SM 1778-45
SM 1973-25
MTAI 8
MTAI 16
MVEN 25
Acid-soil
Mid-altitude High-altitude
High
ACMD
savannas
valleys
environments
carotene
Resistance
CM 523-7
CG 489-1
CM 7138-7
AM 273-23
C-4
CM 2177-2
CM 2772-3
CM 7138-12 AM 320-52
C-6
CM 2772-3
CM 6740-7
CM 7190-2
AM 320-80
C-18
CM 4574-7
CM 7514-7
CM 7438-14 AM 320-120
C-19
CM 5306-8
CM 7951-5
CM 7595-1 AM 320-145
C-24
CM 6055-3 CM 8370-10
CM 8106-4
CM 2772-3
C-33
CM 6438-14 CM 8370-11
SM 707-17 CM 1585-13
C-39
CM 6740-7
SM 909-25
SM 998-3
SM 1433-4
C-41
CM 6921-3
SM 1219-9
SM 1053-23
MBRA 337
C-43
CM 7052-3
SM 1460-1
SM 1958-13 MCOL 1734
C-54
CM 7073-7 SM 1642-22 SM 1495-22 MCOL 2199
C-101
CM 7951-5
SM 1660-4
SM 1703-17 MCOL 2318
C-127
CM 9459-13 SM 1741-1
SM 1937-1
MCR 87
C-227
CM 9460-15 SM 1779-7
SM 1946-2
MPER 297
C-243
CM 9461-15 SM 1855-15 SM 2227-21
C-373
CM 9464-36 SM 1871-33 SM 2229-36
C-377
SM 1219-9
SM 1965-1
SM 2233-11
C-400
SM 1460-1
SM 2052-4
MCOL 2261
C-413
SM 1565-15 SM 2058-2
SM 1741-1
SM 2085-7
SM 1821-7
SM 2211-3
SM 1859-26
MBRA 383
SM 1864-10 MCOL 2737
SM 2632-4
MECU 72
SM 2638-13
MPER 183
SM 2730-1
Inter-regional crosses
SM 2786-10
CM 4919-1
CM 6438-14 CM 7514-7
SM 2792-32
SM 1438-2
CM 6740-7
CM 7951-5
MBRA 502
SM 1565-17
SM 1219-9 CM 8370-11
MCOL 2737
MTAI 16
MCOL 2737
MBRA 383
During this year, the genotypes listed in Table 2.1 were selected to produce a new
generation of crosses. These materials had stood out for their excellent performance
per se, and for demonstrating good levels of general combining ability in relation to the
results observed in the respective Clonal Evaluation Trials (see sections 2.4.1, 2.4.2,
and 2.4.3 for more detail). The agronomic performance of some of these materials per
se is also described below. At the bottom of the table parental lines for special purpose
crosses have also been listed. The seed produced from the current crossings will be
harvested until December 2004.
Planting materials were also selected from these parents to seed the F1 in July 2003.
Output 2-3
2003 Annual Report
In addition to crossing, these lines were also self-pollinated to begin an S2 recurrent
selection scheme to improve each of them for tolerance to inbreeding. The justification
for this approach is given later when the description of a cassava-breeding scheme
based on the production of doubled-haploids described in Output 4.
Because project activities expanded to areas where CIAT had not previously worked
intensely (e.g., Middle Magdalena River and Tolima/Huila Departments, in Colombia),
hybridizations for these areas will, this year, be conducted as follows: (1) polycrosses
and crosses for the two most important cassava-producing regions (Sub-humid and
Acid Soil Savannas). Similar needs exist for inter-Andean valleys that can be fulfilled
by materials for the Acid Soil Savannas. (2) For new regions, for which the project had
not developed specifically adapted materials, production of interregional crosses,
combining the best five materials of the North Coast with clones adapted to the Acid
Soil Savannas and vice versa. These progenies are also expected to produce
germplasm with broad adaptation.
For environments affected by white flies a source carrying resistance to whitefly
(MECU 72) has been included. This pest has become the one true constraint to
cassava cultivation in that region of Colombia. For the high-altitude tropics, crosses
will be carried out within a group of clones recently identified as excellent based on
their cooking quality and good acceptability to farmers.
Activity 2.2 Establishment of crossing blocks and production of recombinant
seed from previously established blocks.
Rationale
Populations developed for specific ecosystems represent the basis for our cooperation
with National Programs and IITA (International Institute of Tropical Agriculture,
Ibadan, Nigeria). The development of genetic stocks is gaining importance through the
years. Genetic stocks are produced based on the recombination of a set of genotypes
that excel for a particular trait, and we would like to upgrade that trait beyond its
natural range of variation (i.e. look for transgressive segregation in broader
adaptation).
Stocks developed for inheritance studies or to support molecular
mapping of specific traits are constructed by the recombination of contrasting
genotypes (i.e. resistance to ACMV, African Cassava Mosaic Virus). Often times our
aim is to pyramid genes responsible for different sources of resistance (i.e. bacterial
blight). As we shift our emphasis from applied breeding to more basic research
supporting breeding (i.e. molecular marker assisted selection or MAS) genetic stocks
will become even more important.
Parental population development in the future will concentrate more in targeting
specific crosses between genotypes selected by NARS and complementary sources of
genetic information from our genetic enhancement program or our global germplasm
collection.
Project IP3: improving cassava for the developing world
Output 2-4
Table 2.2. Production of recombinant cassava seed at CIAT, Palmira, Valle del Cauca,
Colombia, between July 2002 and October 2003.
Purpose of cross
Controlled
Poli-
Crosses
crosses
Total
Inter-specific crosses
Novel starch types
69
69
Hihg dry matter content
6111
6111
High Protein and dry matter content
6155
6155
Yellow roots and high dry matter content
733
733
Yellow roots and high protein content
417
417
1277
1277
495
495
Resistance to mites
Resistance to white flies
Crosses among M. esculenta germplasm
Self-pollinations
7028
7028
Increased carotenes in yellow roots
1632
1632
Resistance to ACMD
2174
4806
6980
Sub-humid environment
1423
21509
22932
Acid soil savannas
2557
15284
17841
Mid-altitude valleys
990
5117
6107
31061
46716
77777
Specific adaptation to:
Total
Specific Objectives
a) To produce large number of seed by sexual crosses (either polycrosses or controlled)
recombining desirable traits from selected parental materials, and deliver them to
NARS in Africa, Asia and Latin America.
Materials and Methods
For polycrosses we use the design developed by Wright 1965 for polycrosses in forage
species. For this type of design there is a need to have a number of clones equal to a
prime number minus one (i.e. 12, 16, 18, etc.). The design allows for each genotype to
have the same probability of being surrounded by any other genotype of the selected
group. Knowledge on flowering capacity is important in order to select a group of
materials with synchronized flowering. When there are considerable differences we
Output 2-5
2003 Annual Report
have to implement delayed planting and/or pruning of the earliest flowering
genotypes. At harvest the seed from different plants of the same genotype are
combined together and named as a half-sib family (SM).
For controlled crosses, we plant 10 to 20 plants depending on the flowering capacity of
the genotype in question. The fruit developed from each flower has the potential to
produce 3 seeds, but in average we obtain no more than 1 seed per pollination. This is
due to the sensitivity of the stigma to the manipulation during pollination. Seeds from
the same cross are mixed together and name as a full-sib family (CM). Because the
number of CM families produced in the last few years has reached 10,000, we began
utilizing a new code for full-sib families (GM).
Results
Almost 80,000 recombinant cassava seeds were produced at CIAT’s Experiment
Station, Palmira, during July 2002 to October 2003 period (Table 2.2). Although the
recombinant seed was produced at CIAT, the generated seedlings used to be
transplanted to fields outside the Experiment Station and under conditions of isolation
from other cassava crops. Thus, the generated F1 plants grew and were maintained
under conditions where possibilities of contamination from frogskin disease were
minimized. This strategy, as can be seen in the description of results from different
Clonal Evaluation Trials, has been highly successful in virtually eliminating the
incidence of this disease from the nurseries for cassava improvement at CIAT. The
production of botanical seed within the CIAT Experiment Station did not represent
high risk because this disease, which is probably induced by a virus, viroid or
phytoplasm, is not likely to be transmitted through botanical seed.
This year, however, F1 plants were transplanted for the first time in many years at
CIAT Experimental Station in Palmira. This change was decided upon the excellent
results observed in the control of white flies (one of the suspected vectors of frogskin
disease) and a reduced incidence in frogskin disease in CIAT’s breeding populations.
Activity 2.3 Generation and distribution of advanced breeding materials for
National Programs.
Rationale
Breeding for Asia has mainly centered on the issue of increased productivity of dry
matter per hectare. Yield and root dry matter concentration have been the primary
traits for selection, with almost no emphasis given to pests and diseases, or cooking
quality. The results obtained in Asia for 15 years, has revealed the possibility to select
for broader adaptation of genotypes. We have the case of Rayong 60 and Kasetsart 50
with good performance in a range of Asian countries. The production of germplasm for
Asia has been moved from Thailand to Colombia due to budget constraints. However,
because of the development attained by several NARS in Asia, the provision of
recombinant material from Colombia can satisfy their needs. A CIAT soil scientist
based in Thailand still coordinates the cassava network for Asia, but covering a
broader spectrum of activities.
Project IP3: improving cassava for the developing world
Output 2-6
For Africa, our breeding efforts have been traditionally channeled through our
collaboration with the International Institute of Tropical Agriculture (IITA) in Nigeria.
As a result extensive germplasm with Latin American “blood” has been introduced to
Africa in a long introgression project financed by the International Fund for
Agriculture Development (IFAD). The purpose of this special project was, among
several others, to introgress Latin American cassava germplasm into Africa, in order to
increase the genetic base of the crop in that continent, particularly for drought
tolerance. This introgression process requires crosses to combine the desirable traits
of Latin American germplasm, with resistance to the African Cassava Mosaic Virus
(ACMV) disease.
Materials and Methods
The same approaches as the ones implemented for other regions of the word
(polycrosses and controlled crosses) have been implemented, but a greater proportion
of segregating progenies from controlled crosses is usually produced. Elite germplasm
identified from the evaluations across the Asian region is periodically sent back to
Colombia, to be used as a parental material in new cycles of selection.
Results
A considerable fraction of the seed produced by the project has been transferred to
National Programs in different regions of the world. As shown in Table 2.2, close to
80,000 recombinant seeds were produced between June 2002 and October 2003 and
about 30% of that seed (23000) has been shipped to our collaborators (Table 2.3). The
retirement of our cassava breeder stationed in Thailand, implied that since 1998 an
increasing proportion of recombinant seed originated in CIAT-HQ. In the future, we
foresee that the flux of improved germplasm between CIAT-HQ, and the Thai and other
Asian breeding programs will continue, and it will be through CIAT that other National
Programs will receive progenies involving the latest selections of elite germplasm from
Asia. By the end of 2003, two scientists from India will came to CIAT to receive
training in molecular markers and to be exposed to the breeding scheme we are now
following.
Because of a self-imposed restriction for in-vitro shipments of cassava germplasm
CIAT shipped a limited number of vitro-plants in the last two years. This restriction,
however, has been gradually eliminated and therefore CIAT will increase the shipment
of vitro-plants. To recover the lost time, the project has set up a tissue culture
laboratory that produces large quantities of vitroplants for our colleagues. The Genetic
Resources Unit previously carried out this activity but the number of clones to be
produced and shipped far exceeds the capactity and function of that Unit. Several
plants from each clone have been or will be sent before the end of the year to countries
in Asia, Latin America and the Caribbean and to IITA. As a result of this
comprehensive on-station participatory evaluation and selection with the farmers, and
NARS partners of the various countries, promising improved genotypes with desirable
characteristics for end users will be identified (as has been the case in the past) under
the local environmental conditions in each of the participating countries.
Output 2-7
2003 Annual Report
Table 2.3. Shipments of recombinant seed produced within the project from
September 2002 through September 2003.
Continents
Genotypes
in-vitro
Crosses
(families)
Plants
(in-vitro)
Seeds in the
shipment
Latin America
In-vitro
Hybrid seed
Africa
230
In-vitro
Hybrid seed
Asia
73
445
159
718
In-vitro
Hybrid seed
Europe + USA
In-vitro
Hybrid seed
Total
In-vitro
Hybrid seed
3402
119
7382
182
40
11983
52
9
502
3228
4617
310
22593
Activity 2.4. Selection of recombinant progenies for broad and specific
adaptation within major agro-ecosystems (sub-humid; semi-arid;
highland and acid soil savanna).
Rationale
Our strategy for cassava germplasm development is centered on the development of
improved gene pools for specific edapho-climatic zones with importance for cassava
production, as defined in Table 2.4. The most relevant ecosystems are the semi-arid
and sub-humid tropics, for which we devote the majority of our efforts. The main
selection activity is conducted in sites selected to represent the conditions of the target
ecosystem. For every genotype that was tested in those sites, a copy was maintained
at CIAT-HQ. This location is considered to be free of bacterial blight and some
important viruses, and to maintain that condition, the introduction of vegetative
material from other areas is restricted. In case vegetative material has to be brought to
HQ, then it has to pass through quarantine, which usually takes more than a year.
Specific Objectives
a) To modify the evaluation procedure to make it more efficient and to adapt it to the
new breeding objectives.
b) To develop and evaluate superior germplasm adapted to particular ecosystems.
c) To develop genetic stocks useful for other CIAT projects.
Project IP3: improving cassava for the developing world
Output 2-8
Table 2.4. Main ecosystems for cassava production, representative production regions,
and main breeding sites.
Description
Representative Countries / Regions
Evaluation Sites
Sub-humid tropics
(rainfall: 800- 1500
mm /year, bimodal
rainfall distribution)
Colombia (Atlantic Coast &
Santanderes); NE. Brazil; NE. Thailand;
Dominican Republic, Haiti; N. and W.
Venezuela; Mexico (Yucatan Peninsula);
subhumid belt of Africa.
Caracolí
Santo Tomás
Huila
Barrancabermeja
Acid soil savannas
(rainfall: 1500 – 3000
mm/year, short dry
period, low pH)
Plains of Colombia & Venezuela; Brazil
(Cerrado); Mexico (Tabasco); Cuba; W.
African savannas; Philippines; Panama
(Ocu)
La Libertad
Matazul
Sder de Quilichao
Barrancabermeja
Humid tropical
lowlands
(rainfall: above 3000
mm/year, no clear dry
period)
Amazon basin (Brazil, Colombia, Peru);
W. Java & Sumatra; Malaysia; S.
Vietnam; Equatorial West Africa
La Libertad
Putumayo
Urabá
Mid-altitude tropics
(800-1400 masl)
Andean zone; central Brazilian
highlands; mid-altitude areas of Nigeria,
Cameroon, East Africa
Palmira
Sder de Quilichao
Barrancabermeja
Tolima-Huila
High-altitude tropics
(1400-2000 masl)
Andean zone; Rwanda; Burundi
Popayán
Mondomo
Armenia
S Brazil; Argentina; China; N Vietnam;
Cuba; Paraguay; S Africa
Sta Catarina
(Brazil)
NE Brazil; NE Colombia; (Guajira)
semiarid belt of West Africa; Tanzania;
Mozambique; Ecuador (Coast)
Guajira
Santo Tomas
NE Brazil
Huila
Subtropics
(latitudes higher than
the tropics)
Semiarid
(rainfall: below 800
mm/year, unimodal)
masl; meters above sea level
Materials and Methods
For each of the zones we conduct a recurrent selection program, with a progressive set
of stages as described in Figure 2.1. As the stages progress, we give more emphasis to
traits of lower heritability, because we have more planting material for each genotype,
and the evaluation can be conducted in bigger plots with replications. Certain
Output 2-9
2003 Annual Report
selection criteria are of general importance across ecosystem (i.e. yield potential, dry
matter content), while others are specific for each ecosystem (i.e. pest and diseases).
Traditionally, the progenies generated from the crossing blocks (F1) were planted in
screen houses and transplanted to the field after 2 months at CIAT. At 6 months after
planting, 2 stakes were harvested from each plant and given a consecutive number
according to the plant. One of the stakes was planted at CIAT, the other one, was
planted at the main selection site (F1C1). Selection was conducted at harvest on
individual plants at the main selection site. Planting material taken from the selected
genotypes, at CIAT, was used subsequently to establish a non-replicated, 6-plant plot,
both at CIAT and at the main selection site (Clonal Evaluation stage). Evaluation
was done using the central 3 plants. Selections were transferred to the following stage
(Preliminary Yield Trial) and planted in non-replicated, 20-plant plots. Evaluation
was done in the central 6 plants, and selections were then passed to the Advanced
Yield Trials at 1 or 2 sites, with 3 replications of 25-plant plots. Genotypes selected
over 2 consecutive years at the Advanced Yield Trial level were considered as “elite
genotypes” and incorporated in the germplasm collection and the crossing blocks.
Since each year a new breeding cycle was initiated, all the stages were simultaneously
being conducted in each site.
Some modifications have been already implemented. A major constraint of the
traditional evaluation methodology was that the first three stages of selection (F1C1,
Clonal Evaluation, and Preliminary Yield Trial) were based on non-replicated plots. In
addition large amount of material was maintained at HQ just to have duplicates of the
very few materials that would reach the status of “elite genotype”, in each cycle.
Therefore, the changes introduced will speed up the selection process, allow for the
evaluation of larger number of progenies and, hopefully, will increase the efficiency of
the selection process. The main changes are as follows:
1) The F1 plants were grown for 10 months rather than 6. At that age they can
produce up to 8-10 stakes. The stakes will be sent to the proper evaluation site for
the Clonal Evaluation. This implies that the F1C1 stage is eliminated and that no
duplicate of each genotype is necessarily maintained at CIAT-HQ.
2) The Clonal Evaluation Trials were based on up to eight plants, rather that six as
before.. Few other traits will also be taken using visual scores: plant architecture,
foliar health (for insects and diseases separately), above ground biomass (for an
estimate of harvest index), and root aspect. A selection index was used to make an
efficient and fast selection of the approximately 1000-2000 genotypes evaluated at
this stage, for each ecosystem.
3) The changes described above allow taking stakes from eight plants (except for
those cases were stakes did not germinate or plants died), rather than three, as in
the past. These eight plants will produce more than 30 cuttings, which will be
used for the first replicated trial based on three replications and two row plots
with ten plants per plot. It is recognized that this evaluation will result in some
competition effect among neighboring plots. However, it is hoped that the number
of replications will neutralize most of these effects. Also, row spacing between plots
can be increased and the plant-to-plant distance within the plot reduced. This will
Project IP3: improving cassava for the developing world
Output 2-10
maintain the density unchanged, while favoring competition among plants from
the same genotype.
4) An important modification to the evaluation process is that data will be taken and
analyzed for all the progenies evaluated. In the past, data was taken only for those
families that went beyond the Clonal Evaluation stage. Therefore it was difficult to
estimate combining ability of parental materials, because most of the crosses did
not produce data (they had been discarded in the field before any data was taken).
The changes introduced allow us to base the selection of the parental materials on
its breeding value (general combining ability) rather that its performance per se, or
empirical appreciation of their potential as progenitor.
5) One final modification introduced this year was the field planting arrangement for
the Clonal Evaluation Trials. In previous years, all the clones belonging to the
same family were planted together one after the other. Because of the large size of
theses trials environmental factors had a relatively large effect on the performance
of the genotypes evaluated. Therefore, each family into three groups of clones each
group having approximately the same size. The experimental plot was also divided
into three “blocks”. Each group of clones from a given family was randomly
allocated to one of these “blocks”. This modification allowed for a replicated
presence for each family. The individual clones, of course, could not be replicated.
On the other hand, the family means are based on three replications and
therefore, more precisely estimated. Selection of individual clones was done within
each “block”, following the ideas behind stratified mass selection proposed by
Gardner in the 1960s.
The main advantages of the new evaluation scheme can be summarized as follows:
The duplication of materials maintained at CIAT-HQ is avoided until they reach
status of “elite genotype”.
The selection of large number of segregating progenies, at the F1C1 stage, which
was based on single plant observations, is avoided.
The time required to reach the stage of replicated trials is minimized.
The total length of each cycle of selection is reduced by almost a year.
Data records will allow for selecting parental material based on general combining
ability.
The total cost for each cycle of selection should be reduced.
Selection will be less subjective by using appropriate software (specifically
developed for that purpose).
Stratification of selection should improve heritability of traits selected (reducing the
phenotypic variance in the denominator in the formula)
For environments with rains concentrated in one season, there is a possibility of
selecting clones able to maintain high dry matter upon the arrival of the rains.
Output 2-11
2003 Annual Report
Time¶
Stage (old system)
Stage (new system)
0
Crossing of selected parental
genotypes
Crossing of selected parental
genotypes
6
F1 (3000-5000)
[6 months]
1 plant / 1 site / 1 rep
Time
¶
0
F1 (3000-5000)
[10 months]
1 plant / 1 site / 1 rep
10
18
F1C1 (2000-4000)
[1 year]
1 plant / 1 site / 1 rep
Clonal evaluation (1000-1500)
[1 year]
6-8 plants / 1 site / 1 rep/3rep§
22
30
Clonal evaluation (500-1000)
[1 year]
6 plants / 1 site / 1 rep
Preliminary yield trial (150-300)
[1 year]
10 plants / 1site / 3 rep
34
42
Preliminary yield trial (100-200)
[1 year]
20 plants / 1-2 sites / 1 rep
Advanced yield trial (40-80)
[2 years]
25 plants / 2-3 sites / 3 reps
58
66
Advanced yield trial (30-60)
[2 years]
25 plants / 2-3 sites / 3 reps
ELITE GERMPLASM
Germplasm
collection
¶
§
Regional
trials
Crossing
blocks
Participatory
research
Time in months after germination of botanical seed.
One replication for clones within each “block” but three replications for families.
Figure 2.1. Basic cassava breeding schemes applied for each of the priority
ecosystems. On the right is the new scheme currently under implementation
(shaded area). Later stages of selection are made following the old system
(shaded area on left).
Project IP3: improving cassava for the developing world
Output 2-12
Preparing new F1 field
About 17,000 recombinant, botanical seeds were germinated early in 2003, and
approximately 11,764 of the resulting plantlets were transplanted at CIAT
Experimental Station in Palmira for the first time in four years (Table 2.5). This
material represents the F1 stage described in Figure 2.1.
Table 2.5. Cassava seed processed for producing F1 plants for various purposes at
CIAT, Palmira, Valle del Cauca, Colombia. F1 nursery was transplanted in July
2003.
Planted seed
Transplanted seedlings
Sub-humid environment
4075
2789
Acid soil savannas
4093
3110
Mid-altitude valleys
4215
2446
Self-pollinations (Double Haploids)
4649
3419
Total
17032
11764
Purpose of crosses
Basic description of the selection index used for ranking the segregating clones
in different types of trials.
Below, results for each agroecological area are presented, together with results of the
best genotypes according to a selection index. This index is a tool for genetically
improving crops, and integrates, into a single value, information on various relevant
traits. In most cases, the index was estimated according to the following formula:
Selection index = [FRY ∗ 10] + [DMC ∗ 8] - [PT ∗ 3] + [HI ∗ 5]
where,
FRY = fresh root yield
DMC = dry matter content
PT = plant type using a 1(excellent) to 5 (very poor) visual scale
HI = harvest index
In this formula, the weighting of each variable is evident. Fresh root yield is
multiplied by 10 to maximize the influence of this trait on the end-result. Dry matter
content is multiplied by 8, also to increase its relevance in the selection process. This
is important because roots with high dry matter content can be dried more quickly, or
else, starch extraction made significantly easier. In both cases, processing costs are
reduced.
Plant type integrates several important aspects for cassava: (1) plant health,
Output 2-13
2003 Annual Report
inasmuch as a plant with a lot of foliage is not likely to have been severely attacked by
leaf diseases and pests (at least, not during evaluation); (2) photosynthesis was
functioning up to evaluation time; and (3) general plant architecture, as on this
depends the quantity of vegetative seed (stakes) produced and the ease with which the
farmer can care for the crop. Because a 1 to 5 score is used (where 1=excellent and
5=very poor plant type), the formula uses a negative term for this trait.
Finally, the harvest index estimates how much of the plant biomass represents the
product with economic value. For now, the index is estimated in terms of the ratio of
root production to the plant’s total biomass.
A technical clarification: these indexes are severely affected by the unit by which each
trait is measured, for example, dry matter content, which fluctuates around 35.0%,
would have a much greater effect than does the harvest index, which fluctuates
between 0 and 1. To avoid this problem, each variable is converted into what are
statistically known as standardized values, which obviate the issue of units.
The most relevant results obtained in six major cassava-producing regions of
Colombia during the cycle that finished with harvests during March to June, 2003 are
summarized below.
2.4.1. Selections for the Sub-Humid Tropical Environment
For logistic reasons, improvement activities developed for several regions of the
Northern Coast of Colombia were centralized initially in Barranquilla. Many of the
materials evaluated there can then be transferred to the more humid region in the
Departments of Córdoba and Sucre, and to the Middle Magdalena (Department of
Santander). The results for this eco-region are described in Tables 2.6 to 2.20 Table
2.6 lists all trials, whereas the other tables show results specific to each one.
A total of 4075 seeds were germinated and 2789 seedlings from these botanical seeds
(targeting this particular environment) were transplanted at CIAT-Palmira in an
isolated field (Table 2.5). The planting of the F1 stage is isolated to reduce as much as
possible infection by diseases that can be found at later stages of the evaluation
process. Seedlings from botanical seed are considered to be disease-free and efforts
are made to maintain this condition for as long as it can possibly be done. Enough
vegetative cuttings from 1158, 10-months old plants from the F1 nursery planted the
previous year could be obtained and planted in the Clonal Evaluation Trial for the subhumid environment that will be harvested in April 2004 (Table 2.7).
A summary of the results from the Clonal Evaluation Trial for the Sub-Humid
environment harvested this year is presented in Table 2.8. The 2200 clones included
in the Clonal Evaluation Trial were allocated in three blocks with 749, 746 and 705
clones each one, respectively. Checks were also included in each block. Table 2.8
provides information on the averages for each of the three blocks. The variation among
the three blocks is an error that eventually affects the selection process. By selecting
within each block, however, this environmental effect could effectively be eliminated.
Project IP3: improving cassava for the developing world
Output 2-14
Table 2.6. Trials conducted in the sub-humid ecosystem (North Coast of Colombia) in
the 2002-2003 cycle.
Trial
F1
Clonal
Evaluation
Preliminary
Yield Trials
Advanced
Yield Trial
Regional
Trials §
Site
N° of
genotypes
N° of reps
Observations
CIATPalmira
2148 (1)
1
Plants are left growing in the field
for 10 months. See Table 2.7.
S. Tomás
2267 (8)
237 (8) †
1 or 3‡
Pitalito
300 (10)
3
Santo
Tomás
Urabá
Sub-humid
Humid
64
(25)
40 (25)
42 (25)
30 (25)
3
3
3
3
See Table 2.8 to 2.13
See Tables 2.14 to 2.16
See Table 2.17.
See Table 2.18 (3 new locations)
See Table 2.19 (two locations)
See Table 2.20 (three locations)
Values in parentheses refer to the number of plants per plot. § A total of 11 locations were
involved in the Regional Trials. ‡ One replication for clones within each “block” but three
replications for families. † Clonal evaluation trial produced from results of previous year
evaluation of diallel crosses.
¶
Because all the clones from each family were divided into three groups allocated to the
three blocks in which the Clonal Evaluation Trials were dvidied, the average
performance of each family was more precisely estimated, since each family was
scattered in three different parts of the field, whereas before it was concentrated in
just one sector (Figure 2.2). As a consequence, the estimates of GCA for each family is
much more reliable now.
A
B
Block 1
Block 2
Block 3
Figure 2.2. Advantage of splitting each family of clones in three groups that were
randomly assigned to each of three blocks in the Clonal Evaluation Trial.(A=
current procedure; B= previous situation).
Output 2-15
2003 Annual Report
Table 2.7. Clonal Evaluation Trial for sub-humid environments. Vegetative cuttings from a total of 1158 clones derived
from 55 families were obtained and planted in May 2003.
Family
CM 8379
CM 8488
CM 9106
CM 9832
CM 9904
CM 9906
CM 9907
CM 9910
CM 9913
CM 9914
CM 9923
CM 9924
CM 9926
CM 9945
CM 9946
CM 9955
CM 9957
CM 9958
GM 248
GM 259
GM 262
GM 266
GM 288
GM 383
GM 385
GM 389
GM 406
GM 408
Female
Progenitor
MTAI 8
CM 4843-1
MTAI 8
SM 643-17
CM 7514-8
CM 6754-8
SM 1565-17
SM 805-15
SM 1438-2
SM 1565-17
SM 1411-5
SM 1433-4
SM 1565-17
SM 805-15
SM 805-15
SM 1433-4
SM 1565-17
MTAI 8
CM 8027-3
SM 1219-9
SM 2192-6
SM 1219-9
SM 1657-12
CM 3555-6
CM 6758-1
SM 805-15
CM 4843-1
CM 8027-3
Male
Progenitor
CM 2772-3
MTAI 8
CM 6754-8
CM 6070-1
CM 6754-8
SM 1438-2
CM 6754-8
CM 7514-8
CM 7514-8
CM 7514-8
CM 8027-3
CM 8027-3
CM 8027-3
CM 6754-8
SM 1411-5
SM 1411-5
SM 1411-5
SM 1411-5
SM 1665-2
SM 1665-2
SM 1219-9
MTAI 8
SM 2192-6
CM 2772-3
CM 3555-6
CM 3555-6
CM 6758-1
CM 4843-1
Seeds
Germ. Transpl.
93
24
100
99
87
61
80
96
93
123
11
55
150
67
100
147
116
50
99
100
68
130
100
68
63
79
86
72
Project IP3: improving cassava for the developing world
22
11
32
23
33
27
19
38
17
44
2
32
40
46
56
39
34
20
28
57
19
66
3
18
24
27
71
62
Clones
Sel.
8
7
4
18
24
19
12
22
9
17
2
15
11
22
39
11
18
8
12
43
12
42
2
10
15
8
45
28
Female
Family
Progenitor
GM 409 CM 4843-1
GM 410 SM 1433- 4
GM 413 CM 4843-1
GM 428 CM 6754-8
GM 436 CM 6758-1
GM 439 SM 805-15
GM 443 SM 1565-17
GM 451 CM 7514-8
GM 456 CM 7985-24
GM 462 CM 7985-24
GM 465 CM 7985-24
GM 466 CM 7985-24
GM 468 MTAI 8
GM 521 SM 1433- 4
GM 546 SM 1565-17
GM 549 SM 1657-12
GM 578 SM 1789-20
GM 579 SM 1789-20
SM 2621 SM 643-17
SM 2750 SM 643-17
SM 3052 CM 4843-1
SM 3054 CM 6754-8
SM 3058 CM 8027-3
SM 3061 SM 1411-5
SM 3062 SM 1438-2
SM 3063 SM 1565-17
SM 3067 MTAI 8
TOTAL
Seeds
Male
Germ.
Transpl.
Progenitor
SM 805-15
71
40
CM 4843-1
74
52
SM 1789-20 31
25
CM 2772-3
90
61
CM 2772-3
66
40
CM 6758-1
90
52
CM 6758-1
123
42
CM 2772-3
73
51
CM 2772-3
105
51
SM 1411-5
114
41
SM 1565-17 122
40
SM 1789-20 114
50
CM 7985-24 60
43
CM 2772-3
55
38
CM 2772-3
78
52
CM 2772-3
27
9
CM 2772-3
70
16
CM 8027-3
128
56
Unknown
75
58
Unknown
50
25
Unknown
54
21
Unknown
105
64
Unknown
130
57
Unknown
106
47
Unknown
126
59
Unknown
100
82
Unknown
87
66
4741
2148
Output 2-16
Clones
Sel.
16
18
13
23
31
27
22
25
38
34
30
39
26
16
36
7
10
26
19
11
5
44
29
26
37
32
35
1158
Table 2.8. Results of the selection carried out in the Clonal Evaluation Trial at Santo
Tomás, Department of Atlántico, from 2267 clones harvested in April 2003.
Clones from the same family were randomly allocated to one of three blocks
within the trial. Therefore, there was no replication for individual clones.
Parameter or
Genotye
Yield (t/ha)
Harvest Index
Fresh roots Dry matter
(0 to 1) ¶
Plant type
(1 to 5) §
Dry matter Selection
content (%)
Index
Results from the 2200 clones (plus checks) evaluated across the three blocks
Maximum
56.9
20.3
0.79
5.00
41.83
84.38
Minimum
0.0
0.0
0.00
0.72
4.15
-62.76
Mean
13.8
3.7
0.47
2.87
26.50
-1.05†
Std. Dev.
6.77
1.96
0.09
0.84
4.36
18.70
Averages of the 749, 746 and 705 clones in Blocks 1, 2 and 3, respectively
Block 1
14.19
3.70
0.50
2.87
26.09
0.00†
Block 2
14.37
3.91
0.46
2.88
27.21
0.00†
Block 3
12.89
3.38
0.44
2.87
26.26
0.00†
Results from the 300 clones selected
Maximum
56.9
16.8
0.78
5.00
41.83
84.38
Minimum
10.9
4.0
0.34
1.00
20.66
19.25
Mean
23.2
6.8
0.55
2.28
29.62
28.02
Std. Dev.
5.37
1.40
0.09
0.80
3.10
7.43
Best 5 clones selected from each block in the Clonal Evaluation Trial
GM 288-49
56.9
16.76
0.67
1
29.47
84.38
CT 59-25
41.8
11.94
0.78
3
28.59
55.04
GM 288-53
40.5
12.28
0.61
2
30.31
54.46
CT 59-18
37.2
11.13
0.71
2
29.91
54.03
GM 281-74
43.3
11.13
0.61
2
25.71
49.36
SM 2962-19
36.8
10.37
0.60
3
28.16
44.43
SM 2779-56
28.8
8.94
0.58
2
30.98
42.57
SM 2964-31
34.8
9.06
0.55
2
26.06
41.25
SM 2964-36
27.6
8.30
0.59
2
30.07
39.42
SM 2547-95
26.1
7.87
0.63
2
30.16
39.09
SM 2547-111
36.7
8.56
0.49
1
23.34
44.38
SM 2948-44
23.4
6.54
0.78
2
28.02
42.81
CT 57-32
31.6
8.54
0.62
3
27.02
38.44
SM 2618-60
15.4
5.85
0.46
1
38.03
38.11
SM 2964-46
35.0
7.31
0.61
2
20.86
37.76
¶ The harvest index is obtained by dividing the production of commercial roots by total biomass
(roots + aerial parts). Preferred harvest indexes are > 0.5.
§ Plant type integrates under one value, plant architecture, leaves health, and capacity to
produce stakes on a scale where 1 = excellent and 5 = very poor is used.
†
Average election index within blocks must be zero, because it is based on a combination of
standardized variables. However, when averaged across the three blocks there is a slight
deviation because selection indices were estimated for each block separately.
Selection of the 300 clones that passed to the next stage of evaluation was based on
the Selection Index described above. A fundamental difference now was that the
point of reference for the selection was the mean performance of all the clones within
each block, and not the whole experiment. The superiority of the selected fraction is
Output 2-17
2003 Annual Report
apparent. Mean fresh root production was 23.2 t/ha in the selected fraction against
13.8 from the entire population. Converted to dry matter productivity these figures
were 6.8 and 3.7 t/ha, respectively. Likewise, the average harvest index was
considerably higher in the selected fraction (0.55) than in the whole trial (0.47) and
plant type score was also better in the selection fraction (2.28) than in the whole
population (2.87). Finally dry matter content was about 3% higher in the selected
fraction than in the total of clones evaluated (29.62 versus 26.50%). These results are
not as good as those reported the previous year, because all the harveting was done at
the end of the season not at two different times. The change is justified because of the
excellent correlation (ρ=0.706) between dry matter contents in the middle of the dry
season (February-March, when dry matter content is at a maximum) and after the
initiation of the rains in May. Harvesting was planned just before the beginning of the
rains, which this year started earlier. That is why dry matter contents are lower than
those reported the previous years.
The contributions of the Ministry of Agriculture and Rural Development of Colombia
and of the Fondo Nacional Avícola (FONAV) of the Federación Nacional de Avicultores
de Colombia (FENAVI) have been fundamental to the project’s significant growth,
enabling the project to now identify, with more certainty, outstanding germplasm.
Table 2.9 describes the average for each of the 55 families planted in the Clonal
Evaluation Trial harvested in May 2003. The total number of clones making up the
family and the proportion that was selected is also provided. All this information is
very valuable for identifying progenitors that tend to produce better (or worse)
progenies. The best three families CT 59, CT 57 and CT 54 had 61.6, 53.1 and 40.6%
of their clones selected (across the three blocks of the trial). It should be emphasized
that about 13.6% of the clones were selected in the entire experiment. It is obvious,
therefore, that these three families had performed particularly well for so many of their
clones to be selected. Worth mentioning is the fact that these three families all come
from crosses made with outstanding germplasm from Thailand (Rayong 5, Rayong 60,
Rayong 90 and Kasetsart University 50). These results clearly indicate the
discriminating power of the procedures employed on one hand, and the genetic
superiority of these materials, evaluated through their progenies, in the sub-humid
environment of the northern coast of Colombia. On the other hand no clone from
families SM 2949, SM2835, SM2667 and CM 9775 was selected.
Information from Table 2.9 is useful not only to identify the best families based on
their proportion of selected clones, but also to understand the reasons for this
success. For instance two out of the three best families (CT 54 and CT 59) showed the
highest dry matter contents among the 55 families. Also, fresh root productivity and
harvest indices are outstanding. Plant type score, on the other hand, was average in
these families. Low dry matter content and mediocre fresh root yields characterized
the worst families listed in Table 2.9.
If the parents involved in the generation of the families listed in Table 2.9 is considered
(one particular clone can be used in generating more than one family), a consolidated
information regarding all the progenies from a particular clone can be obtained.
Project IP3: improving cassava for the developing world
Output 2-18
Table 2.9. Results from the Clonal Evaluation Trial (Santo Tomás, Atlántico), of 1967 genotypes. The averages of each of
the 52 families, number of clones representing them and proportion of selected clones is presented.
Family
%
select.
#
clones
Fresh DM
(t/ha) (t/ha)
%
DM
Plant
Type
(1-5)
HI
(0-1)
¶
SI
§
Family
0.5 23.5 SM 2958
CT 59
73
61.6 21.4
6.52 30.5 3.0
0.6 20.6 CM 9794
CT 57
32
53.1 20.7
5.89 28.4 2.8
0.5 16.4 SM 2773
CT 54
32
40.6 17.5
5.27 30.1 2.9
0.4
GM 288
22
36.4 17.2
4.65 27.0 2.6
7.8
SM 2618
0.4
SM 2547
68
29.4 14.9
4.37 29.3 2.3
10.7 SM 2626
0.4
SM 2779
56
26.8 15.6
4.35 27.8 2.1
8.4
SM 2629
0.4
SM 2783
46
26.1 13.8
3.81 27.6 2.3
5.9
SM 2828
0.5 10.6 SM 2954
CM 9912
64
25.0 15.1
4.48 29.7 2.7
0.4
SM 2834
41
19.5 14.4
3.96 27.5 2.7
3.6
SM 2960
0.4 -0.2
SM 2621
32
18.8 12.0
3.46 28.9 2.8
SM 2956
0.5
GM 281
47
17.0 16.2
3.48 21.5 2.8
-2.8
SM 2957
0.4
SM 2964
53
17.0 15.7
4.09 26.1 3.0
1.6
SM 2948
0.4
SM 3000
33
15.2 13.7
3.80 27.8 3.0
1.8
SM 2951
0.4 -2.7
SM 2620
54
14.8 12.9
3.30 25.6 2.9
SM 2955
0.4
SM 2546
69
14.5 11.8
3.17 26.7 2.9
-2.6
SM 2733
0.4 -1.4
SM 2962
28
14.3 14.0
3.70 26.4 3.2
SM 2836
0.4
SM 2963
40
12.5 13.8
3.47 25.1 3.0
-2.9
SM 2982
0.4
SM 2952
51
11.8 16.6
4.61 27.7 2.9
5.9
SM 2612
0.4 -2.3
SM 2882
52
11.5 14.5
3.65 25.1 3.0
SM 2832
0.4 -4.0
SM 2839
39
10.3 15.0
3.73 24.9 3.1
CM 9797
0.4
SM 2947
49
10.2
9.9
2.75 27.8 3.0
-1.2
SM 2829
0.5 -4.1
CM 9791
60
10.0 12.7
3.13 24.6 3.0
CM 9775
0.4
SM 2777
50
10.0 12.2
3.11 25.5 2.7
-6.6
SM 2667
0.5 -7.7
SM 2959
10
10.0 11.7
2.84 24.2 3.2
SM 2835
0.4
SM 2615
41
9.8
11.4
3.15 27.6 2.7
-1.3
SM 2949
¶ HI = Harvest Index (Root production / total biomass).
§ SI = Selection Index (combines several variables of economic relevance)
Output 2-19
%
select.
#
clones
21
58
83
49
37
51
54
42
28
29
30
46
37
22
51
52
28
61
38
41
42
56
53
33
35
9.5
8.6
8.4
8.2
8.1
7.8
7.4
7.1
7.1
6.9
6.7
6.5
5.4
4.5
3.9
3.8
3.6
3.3
2.6
2.4
2.4
0.0
0.0
0.0
0.0
Fresh
(t/ha)
DM
(t/ha)
11.5
15.6
12.4
9.8
11.6
11.9
11.6
15.6
12.0
11.8
12.5
14.9
15.3
13.9
13.5
12.6
13.8
14.0
12.1
14.5
10.7
12.4
11.0
11.9
11.9
3.03
3.81
3.48
2.77
3.12
2.90
3.32
4.09
3.07
3.27
3.25
3.91
4.04
3.13
3.29
3.47
3.16
3.90
3.21
3.43
2.92
2.97
2.42
3.13
3.03
2003 Annual Report
%
DM
Plant
Type
(1-5)
HI
(0-1)
¶
SI
§
26.3
24.5
28.2
28.4
26.8
24.4
28.6
26.2
25.5
27.7
26.0
26.2
26.3
22.6
24.5
27.5
22.9
27.9
26.6
23.7
27.4
23.9
21.9
26.2
25.4
3.0
3.1
2.9
2.9
3.1
2.7
2.7
3.0
2.7
3.3
3.0
3.2
3.0
2.9
2.7
2.9
2.9
2.5
3.1
3.2
3.0
3.1
2.6
2.9
3.3
0.4
0.5
0.4
0.4
0.4
0.4
0.4
0.4
0.3
0.4
0.4
0.5
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.5
0.4
0.4
0.3
0.4
0.4
-7.3
0.8
0.1
-2.8
-4.9
-6.9
0.4
2.1
-7.1
-2.7
-4.0
1.7
-1.4
-8.5
-5.6
-3.0
-8.2
2.7
-5.9
-4.5
-6.2
-7.7
-16.0
-5.5
-10.4
Table 2.10 summarizes the results from all the progenies generated by a given clone. Progenies
from Rayong 90 (73 clones in total) tended to have a much higher probability of being selected
(61.6 %). The same was true for the other clones originated in Thailand: KU50 (46.9%), Rayong
60 (46.2%) and Rayong 5 (40.6%). Following these good parets came a group of relatively new
clones SM 1068-10, SM 2192-6, SM 1411-5 and CM 7514-8 all with average selections above
20%. It is interesting to note the advantage of introducing new germplasm among the parental
lines compared with other clones (such as CG 1141-1), which have been used already for more
than ten years as progenitors for this particular agro-ecological zone. Worth mentioning is the
poor performance of CM 74514-7 (a sister clone from CM 7514-8), with no clone selected out of
the 56 progenies derived from it. This kind of results indicates the kind of heterotic response
that is likely to be found in a crop like cassava.
Table 2.10. Number of progenies evaluated and selected from each progenitor. Data from the
Clonal Evaluation Trial (Santo Tomás, Atlántico) described in Table 2.10.
Progenitor
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
R 90
KU 50
MTAI 8 (R 60)
R5
SM 1068-10
SM 2192-6
SM 1411-5
CM 7514-8
SM 1657-12
SM 643-17
MVEN 25
SM 1665-2
CG 1141-1
SM 1511-6
SM 890-9
SM 1433-4
SM 1565-17
CM 3372-4
CM 6754-8
SM 1438-2
Family
size
73
64
73
32
68
50
97
118
52
32
53
57
33
87
69
213
108
52
49
109
Progenies
Selected
(number)
45
30
34
13
20
12
23
24
10
6
9
9
5
13
10
26
13
6
5
11
Progenies
Selected
(%)
61.6
46.9
46.2
40.6
29.4
24.0
23.7
20.3
19.2
18.8
17.0
15.8
15.2
14.9
14.5
12.2
12.0
11.5
10.2
10.1
Progenitor
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Family
size
CM 4365-3
41
SM 1657-14
21
SM 1210-10
83
SM 1201-5
37
SM 1422-4
51
CM 7389-9
103
SM 1521-10
42
SM 1754-21
28
SM 1210-10
101
SM 1619-3
29
CM 8027-3
46
MNGA 19
215
CM 2772-3
28
SM 1600-4
61
CM 7395-5
42
SM 805-15
73
CM 6438-14
53
CM 7514-7
56
SM 1431-2
33
Average
Progenies
Selected
(number)
4
2
7
3
4
8
3
2
7
2
3
12
1
2
1
1
0
0
0
Progenies
Selected
(%)
9.8
9.5
8.4
8.1
7.8
7.8
7.1
7.1
6.9
6.9
6.5
5.6
3.6
3.3
2.4
1.4
0.0
0.0
0.0
13.6
If selection were conducted ignoring the differences among the three blocks (as done in the
past) 29 out of the 300 clones would have been different (10% difference approximately).
Most of the difference was in relation to the obviously deficient conditions in Block 3 (Table
2.8) which would have resulted in a fewer number of clones from this area of the
experimental plot being selected.
Project IP3: improving cassava for the developing world
Output 2-20
During the previous growing season a diallel study was conducted (CIAT, Annual Report
2002). This was primarily a quantitative genetics study, but nevertheless 237 clones were
selected and organized in a second Clonal Evaluation Trial for the sub-humid environment.
Table 2.11 summarizes the results from this trial. The diallel study was developed from the
crosses among nine parents. Therefore, a total of 36 full-sib families were produced (p(p1)/2), most crosses represented by 30 clones. Each progenitor, was involved in 8 full-sib
families and 270 clones. Table 2.12 lists the nine progenitors, the number of clones selected
at the diallel study stage and the number of clones selected at the Clonal Evaluation Trial
stage. Then correlation between the two selection stages was ρ = -0.15, indicating that the
mean performance of the progenies from each parental line (as per the proportion of selected
clones) was different. This lack of coincidence may be due to environmental differences
between the two different seasons in which selections were carried out, or to the fact that the
relative performance of individual clones within each family (related to dominance or
heterotic effects) played a more important role that the additive effects inherited from the
common progenitor.
Table 2.11. Results of the selection from the Clonal Evaluation Trials (Santo Tomás,
Atlántico), based on 237 genotypes from the diallel studies conducted the previous
year.
Parameters
Maximum
Minimum
Mean
St. Dev.
Maximum
Minimum
Mean
St. Dev.
GM 291
GM 266
GM 259
GM 238
CM 9106
GM 266
CM 9921
CM 9921
CM 9703
GM 246
Output 2-21
Yield (t/ha)
Dry
Plant
matter
type
Fresh
Dry
roots
matter
(%)
(1 a 5)
Statistics of the 237 clones from the evaluation
28.13
7.94
31.71
5
0.24
0.00
16.30
1
10.61
2.45
24.41
3
6.38
1.78
3.48
0.7
Statistics of the 55 clones selected
28.13
7.94
31.71
3
9.62
2.76
18.71
1
17.93
4.78
26.88
2
4.51
1.16
2.63
1
Statistics of the best 10 clones selected
24.28
7.007
28.9
1
27.04
7.518
27.8
2
28.13
7.935
28.2
3
23.56
6.606
28.0
2
18.03
5.466
30.3
2
24.04
5.998
25.0
2
22.36
5.748
25.7
2
20.91
5.469
26.1
1
24.88
6.128
24.6
2
25.72
7.053
27.4
3
Harvest
index
(0 a 1)
0.70
0.12
0.49
0.11
0.70
0.41
0.56
0.07
0.53
0.60
0.60
0.60
0.58
0.63
0.63
0.50
0.59
0.57
2003 Annual Report
Table 2.12 Mean performance of the progenies derived from each parental line evaluated in
the second Clonal Evaluation Trial for the sub-humid environment. The number of
clones listed in the first column reflect the relative advantages of each parental clone,
since each of them had originally been represented by approximately the same
number of clones in the diallel study ( ≈ 270).
# clones
Progenitor
CM 6754-8
CM 8027-3
MTAI 8
SM 1219-9
SM 1411-5
SM 1565-17
SM 1657-12
SM 1665-2
SM 805-15
¶
¶
41
49
54
68
53
58
55
67
28
%
selected
clones¶
29.27
32.65
42.59
22.06
24.53
17.24
23.64
13.43
17.86
Plant
Type
(1-5)
2.85
2.89
2.68
3.02
2.84
2.98
2.94
3.09
3.08
Dry
matter
(%)
25.42
25.79
24.84
24.04
25.42
22.89
24.77
23.11
23.97
Harvest
Index
(0-1)
0.48
0.50
0.50
0.52
0.48
0.45
0.51
0.50
0.43
Fresh
roots
(t/ha)
42.74
45.30
53.13
43.05
38.68
46.44
40.96
39.07
39.70
Selection
Index
2.71
5.53
7.36
-0.59
1.32
-5.54
0.90
-6.23
-6.36
Correlation between selections in 2002 and 2003 was –0.15.
Results from Tables 2.10 and 2.12 show some agreements. MTAI 8 (Rayong 60) showed in
both analyses an outstanding performance through its progenies. SM 1411-5 also showed
better than average results in both trials. SM 805-15 did not produce good progenies in
either study. CM 8027-3, on the other hand had a large number of progenies selected in the
second Clonal Evaluation Trial (Table 2.12) but had poor results in the main trial (Table 2.10)
One additional comparison that results of interest for the cassava-breeding project are that of
the mean performances of full-sib (GM, CM or CT codes) versus the half-sib (SM code)
families. Table 2.13 summarizes the mean selection indexes of both types of families. It is
obvious that full-sib families were much better than the half-sib families, at least based on
their respective mean selection indexes. This finding may be the result of the excellent
performance of the CT families from Thailand, and not necessarily a reflection of the relative
merit of one type of family over the other. Similar comparisons will be made using the results
from the Clonal Evaluation Trials for other target environments, where there was no
equivalent case to the CT families involved in the trial for the sub-humid environment.
Table 2.13. Comparison between the selection indexes of full- and half-sib families in the
Clonal Evaluation Trial for the sub-humid environment.
Full sib families
Half sib families
Total Average
Block 1
Block 2
Block 3
Number
Average
Number
Average
Number
Average
of clones Sel. Index of clones Sel. Index of clones Sel. Index
164
8.30
163
2.94
159
5.92
585
-2.33
583
-0.82
546
-1.73
749
0.00
746
0.00
705
0.00
Project IP3: improving cassava for the developing world
Output 2-22
Table 2.14. Results of the Experiment 1, where 97 clones selected in the Clonal Evaluation
Trial from the 2001-2002 season were evaluate in three replications in the Atlantico
Department. The mean performance of the best 15 clones is presented.
Yield (t/ha)
Clon
Harvest
Index
Fresh roots Dry matter
(0-1)
GM 290-18
CM 9958-32
GM 290-11
CM 9958-52
CM 9958-44
GM 290-20
CM 9957-73
GM 290-52
CM 9923-53
CM 9907-47
CM 9958-42
CM 9923-50
CM 9958-38
CM 8209-61
CM 9946-31
34.6
33.8
20.2
21.1
34.2
25.9
28.8
28.9
22.2
22.9
24.3
20.3
18.6
17.1
19.6
Maximum
Minimum
Mean
St.Deviation
34.6
17.1
24.8
5.9
Maximum
Minimum
Mean
St.Deviation
34.6
0.5
15.4
7.5
CG 1141-1
SM 1438-2
MTAI 8
11.2
7.9
21.0
Output 2-23
Dry matter
Content
(%)
9.6
0.7
27.5
9.4
0.6
27.8
6.4
0.7
31.7
6.6
0.6
31.5
9.5
0.6
27.7
7.5
0.6
28.7
8.7
0.6
30.1
8.0
0.8
27.3
6.6
0.7
29.5
6.7
0.6
29.0
7.2
0.7
29.9
5.9
0.6
29.5
5.7
0.6
30.5
5.5
0.6
31.8
6.0
0.6
30.8
Statistics of the 15 clones selected
9.6
0.8
31.8
5.5
0.6
27.3
7.3
0.6
29.6
1.4
0.1
1.5
Statistics of the 97 clones evaluated
9.6
0.8
31.8
0.3
0.2
20.0
4.3
0.6
27.5
2.1
0.1
2.4
Checks included in the evaluation
3.7
0.6
32.4
2.4
0.6
29.5
6.0
0.7
28.5
Plant
Score
(1-5)
Root
Score
(1-5)
Selection
Index
2.3
1.0
2.0
1.3
2.0
1.3
2.3
3.0
2.0
1.3
2.7
1.3
1.7
2.0
2.7
2.0
2.3
2.0
2.0
2.7
2.0
2.0
2.7
2.3
2.3
2.3
2.7
2.7
2.3
1.7
34.87
32.71
29.86
29.02
28.95
26.95
26.03
25.27
23.74
22.64
20.98
20.00
18.59
18.16
15.43
3.0
1.0
1.9
0.6
2.7
1.7
2.3
0.3
34.9
15.4
24.9
5.6
4.0
1.0
2.5
0.7
5.0
1.0
2.5
0.6
34.9
-45.2
0.0
16.6
2.7
3.0
2.7
1.7
2.7
1.7
2003 Annual Report
Table 2.15. Results of the Experiment 2, where 97 clones selected in the Clonal Evaluation
Trial from the 2001-2002 season were evaluate in three replications in the Atlantico
Department. The mean performance of the best 15 clones is presented.
Yield (t/ha)
Clon
Harvest
Index
Fresh roots Dry matter
(0-1)
GM 211-51
GM 247-43
GM 239-5
GM 302-25
GM 239-14
CM 9966-61
CM 9966-55
GM 247-44
CM 9958-76
GM 214-60
GM 247-32
GM 302-16
CM 9966-57
GM 246-55
CM 9966-41
35.1
23.6
27.2
22.9
24.9
25.4
23.8
22.9
22.0
14.3
18.3
21.3
22.5
21.9
21.5
Maximum
Minimum
Mean
St.Deviation
35.1
14.3
23.2
4.5
Maximum
Minimum
Mean
St.Deviation
35.1
1.0
11.6
7.5
CG 1141-1
SM 1438-2
MTAI 8
18.9
13.4
20.5
Dry matter
Content
(%)
9.7
0.74
27.6
6.9
0.73
29.3
8.2
0.63
30.1
6.7
0.65
29.3
6.8
0.77
27.4
7.6
0.61
30.0
6.5
0.66
27.4
6.5
0.71
28.3
6.1
0.67
27.4
4.5
0.69
31.5
5.2
0.71
28.7
6.7
0.53
31.4
6.2
0.71
27.5
6.4
0.68
29.3
5.8
0.64
26.9
Statistics of the 15 clones selected
9.7
0.8
31.5
4.5
0.5
26.9
6.7
0.7
28.8
1.2
0.1
1.5
Statistics of the 97 clones evaluated
9.7
0.8
33.3
0.3
0.5
20.7
3.3
0.7
28.4
2.1
0.06
2.25
Checks included in the evaluation
6.2
0.61
32.8
3.7
0.65
27.5
5.5
0.67
26.1
Project IP3: improving cassava for the developing world
Plant
Score
(1-5)
Root
Score
(1-5)
Selection
Index
2.0
2.5
2.7
2.0
3.0
2.7
1.7
2.7
1.7
2.7
2.3
2.0
2.7
3.3
1.7
2.0
3.0
1.7
2.7
2.7
2.7
2.0
2.3
2.3
2.3
2.0
2.7
3.0
3.0
2.7
43.54
29.22
27.06
25.14
24.03
22.85
22.72
22.23
21.19
20.48
19.82
19.46
18.44
17.02
16.35
3.3
1.7
2.4
0.5
3.0
1.7
2.5
0.4
43.5
16.3
23.3
6.6
4.7
1.7
3.2
0.71
4.0
1.7
2.9
0.43
43.5
-29.9
0.0
14.97
2.3
3.0
3.0
1.7
3.0
2.7
Output 2-24
Table 2.16. Results of the Experiment 3, where 97 clones selected in the Clonal Evaluation
Trial from the 2001-2002 season were evaluate in three replications in the Atlantico
Department. The mean performance of the best 15 clones is presented.
Clon
CM 9958-40
CM 9958-65
CM 9958-54
GM 259-69
CM 9958-61
GM 239-2
GM 273-38
GM 288-19
CM 9907-41
GM 273-50
GM 262-28
GM 288-8
GM 258-51
GM 288-17
GM 274-10
Maximum
Minimum
Mean
St.Deviation
Maximum
Minimum
Mean
St.Deviation
CG 1141-1
SM 1438-2
MTAI 8
Yield (t/ha)
Harvest Dry matter
Plant
Content
Score
Fresh roots Dry matter Index
(0-1)
(%)
(1-5)
32.7
10.1
0.6
30.9
2.0
28.4
8.5
0.6
30.0
1.7
30.5
8.5
0.7
27.8
2.0
31.7
8.5
0.7
26.8
3.0
22.2
6.6
0.6
29.9
2.0
25.5
7.0
0.7
27.1
3.0
21.7
5.6
0.7
25.9
2.0
23.7
5.7
0.7
24.1
2.3
20.6
5.9
0.6
28.9
2.3
21.7
6.5
0.7
29.7
3.3
20.4
6.1
0.6
29.8
2.7
18.9
5.7
0.5
30.2
2.3
19.7
5.7
0.6
29.2
2.7
20.2
5.8
0.6
29.0
2.3
18.4
5.6
0.6
29.3
2.7
Statistics of the 15 clones selected
32.7
10.1
0.7
30.9
3.3
18.4
5.6
0.5
24.1
1.7
23.8
6.8
0.6
28.6
2.4
4.8
1.4
0.1
1.9
0.5
Statistics of the 97 clones evaluated
32.7
19.1
0.8
33.4
4.3
0.6
0.1
0.4
16.8
1.7
12.6
3.6
0.6
27.1
3.1
7.2
2.7
0.1
2.9
0.6
Checks included in the evaluation
12.8
4.2
0.6
32.6
2.7
9.7
2.7
0.7
27.6
3.0
21.7
5.6
0.7
25.6
3.3
Root
Score
(1-5)
1.7
2.0
2.3
1.7
2.3
1.7
1.7
2.7
1.7
2.0
1.7
2.3
1.7
2.0
2.3
Selection
Index
2.7
1.7
2.0
0.3
46.5
14.9
24
9.6
4.0
1.5
2.6
0.5
46.5
-59.7
0.00
17.1
46.50
39.22
37.05
27.55
25.64
21.75
21.15
20.52
20.25
18.98
18.27
16.34
16.16
15.91
14.86
1.7
2.7
2.3
The best 291 clones selected from the Clonal Evaluation Trial harvested the previous year
were planted in three different Preliminary Yield Trials identified as Experiments 1, 2 and 3.
Each of these trials included 97 experimental clones and three checks evaluated in 10-plants
plots with three replications. The most relevant results of these trials are presented in Tables
2.14, 2.15, and 2.16. One of the advantages of using the selection index with standardized
values, described above, is that by definition an average performance has a value of zero.
This can be confirmed in the right column of these Tables, where the population mean of the
Output 2-25
2003 Annual Report
100 clones was, in every case, zero. A positive selection index suggests better than average
performance. A negative value, on the other hand, indicates an undesirable general
performance.
Clones selected from Preliminary Yield Trials in 2001 were then evaluated in the following
phase of selection in Advanced Yield Trial (See Figure 2.1). A total of 64 experimental clones
were evaluated in this phase of the selection process with three replications in each of two
locations in the Sucre and Atlántico Departments. Table 2.17 presents the most relevant
results from this trial. The mean fresh root yield across the two locations was around 26
t/ha, but the average of the best 15 clones (Table 2.17) was as high as 31.70 t/ha, which
represented about 11.14 t/ha of dry matter. These materials will be incorporated into the
regional trials for further evaluation and in a higher number of different locations.
Table 2.17 Result of the Advanced Yield Trial conducted in locations in the Sucre and
Atlántico Departments. A total of 64 experimental materials were evaluated with three
replications per location and 20 plant-plots. Data was taken from the 6 central plants.
Results of the best 15 clones are presented.
Averages across the two locations
Clon
SM 2629-36
SM 2620-1
SM 2615-25
CM 9560-1
SM 2616-11
SM 2775-2
SM 2771-5
CM 9456-12
SM 2782-4
SM 2769-11
SM 2773-32
SM 2619-4
SM 2625-1
SM 2782-9
SM 2780-17
Mean 15 clones
Maximum
Minimum
Mean
St.Dev
Correlation
Plant
type
(1-5)
1.83
1.50
2.67
2.83
3.33
1.50
2.83
2.17
2.67
2.17
2.83
2.33
2.67
2.50
2.33
2.41
4.50
1.50
2.61
0.56
0.491
Ranking
Fresh
Dry matter Dry matter
Harvest Selection
content
yield
root yield
Atlantico Sucre
Index
Index
(%)
(t/ha)
(t/ha)
35.30
34.00
12.16
0.63
23.50
2
7
34.45
34.05
11.83
0.59
22.34
4
5
27.66
39.30
10.64
0.57
20.90
3
9
35.73
35.14
12.73
0.57
20.46
9
2
33.46
35.57
11.83
0.64
18.95
6
14
35.49
31.37
11.17
0.62
16.57
8
18
36.93
35.92
13.15
0.50
16.40
13
6
31.98
33.06
10.71
0.67
15.79
12
10
31.11
35.09
10.97
0.58
14.72
1
31
31.72
32.16
10.33
0.66
10.82
20
11
23.34
39.86
9.31
0.47
9.77
5
43
25.75
37.46
9.75
0.52
9.65
26
12
27.77
35.47
9.94
0.58
8.74
24
16
29.65
36.26
10.95
0.51
8.27
36
4
35.22
31.80
11.66
0.59
7.93
44
1
31.70
35.10
11.14
0.58
14.99
-.-.36.93
39.86
13.15
0.67
23.50
9.97
19.27
2.91
0.23
-44.96
26.21
34.02
9.11
0.53
-0.08
5.47
2.88
1.89
0.08
12.75
0.243
0.657
0.203
0.535
0.385
0.318
Project IP3: improving cassava for the developing world
Output 2-26
Only two clones, among the best 15, were from the same family (SM 2782). Table 2.17 also
provides the ranking of each clone in each environment. The first four clones were among the
best ten clones at both locations, suggesting that they are well adapted to the different
conditionts where they were grown and that they may also possess good stability
characteristics. The best clone in the Atlántico Departmnet (SM 2782-4) was ranked 31st in
the trial at Sucre Department. Its sister clone (SM 2782-9), also showed strong interaction
effects, but in the other direction (36th and 4th ranking for Atlántico and Sucre Departments,
respectively). A similar situation was observed for clon SM 2780-17, which was the best one
in Sucre Department, but 44th at Atlántico. As mentioned in previous reports this strong
benotype x environment interaction effects are common in cassava, but not all the genotypes
show this trend. The emphasis in the cassava project at CIAT will concentrate on clones such
as SM 2629-36, SM 2620-1, SM 2615-25 and CM 9560-1 based on their excellent
performance at both locations.
The phenotypic correlations between mean performances in both locations is also presented
in Table 2.17. In spite of the examples of strong genotype x environment interaction
mendioned above, correlations are in general acceptable. As expected dry matter content,
harvest index and plant type had high correlations (ρ= 0.657, 0.535l, and 0.491 respectively).
The correlation for fresh root yield in both locations was much lower (ρ= 0.243) and for dry
matter yield even lower (ρ= 0.203), which is not surprising because the later is a combination
of two variables (fresh root yield and dry matter content). Phenotypic correlation for selection
index was intermediate ρ= 0.385).
An extensive study is currently underway with Regional Trials evaluated across 11 locations
and harvested by May 2001. Three additional locations (all in the Urabá region) were added
and their results presented in Table 2.18. The complete results of the 14 locations will be
submitted for publication in a scientific journal with emphasis on the stability of
performance. One surprising result is the excellent performance of clon SGB 765-2 which
was released as a variety three years ago by CORPOICA. This clon was selected trhough a
farmers’ participatory approach. The enexpected result being that this clon had the highest
regression coefficient indicating that is particularly adapted to high-yielding environments,
very different to those where it was selected. A sister clone (SGB 765-4) was seventh in
ranking across the five Urabá locations listed in Table 2.18. It may be worth analyzing if this
trend (selection for clones adapted to high-yielding conditions) repeats in other experiences
with farmers’ participatory selections. The good performance of clon SGB 765-2 was clearly
favored by its very high dry matter yield (29.2) at Carepa in the second harvest (Carepa-02).
This yield fresh came from 79.9 t/ha of fresh roots.
One other interesting information from Table 2.18 is the excellent yields obtained in the
Urabá region. Carepa-02 had an average dry matter yield of 19.1 t/ha (53.7 t/ha of fresh
roots across the 42 clones evaluated). The worst location was Mutatá-02 with 10.1 t/ha of
dry matter, which is still an excellent average yield and highlights the potential of cassava as
source of raw material for the industry. This is an important banana and plantain producing
region in Colombia, and this industry is currently evaluating the industrial feasibility of
using refined cassava flour in the production of boxes for the packing of the fruits as well as
for the adhesives for gluing them. It should be pointed out, that the Urabá region is not truly
sub-humid. Water availability is lower in the Atlántico and Magdalena Departments (nearby
Output 2-27
2003 Annual Report
Barranquilla) and gradually increases toward the Urabá region. That explains the poor
performance of clon MTAI 8 which is outstanding in drier environments.
Table 2.18. Results of five Regional Trial for the Urabá Region planted in 2001 and 2002.
Ranking of clones is based on selection index estimated from across locations average.
Dry matter yield (t/ha)
Clone
Carepa-01 Necoclí-01 Carepa-02 Necoclí-02 Mutatá-02
1 SGB 765-2
13.6
13.4
29.2
8.7
11.5
2 SM 1511-6
14.5
11.1
23.3
12.5
14.4
3 M VEN 25
14.0
13.6
19.4
14.2
15.0
4 SM 1411-5
12.6
15.8
20.0
12.5
13.7
5 SM 1565-17
20.4
9.9
21.6
11.6
11.8
6 SM 1665-2
13.8
10.5
20.9
13.4
14.2
7 SGB 765-4
16.6
10.7
20.6
11.1
12.4
8 SM 1973-25
15.6
10.0
22.2
11.9
11.1
9 SM 1973-23
14.7
10.3
24.2
13.0
8.8
10 CM 7514-8
12.9
12.8
20.7
11.1
13.0
11 CM 4919-1
15.5
8.5
25.1
10.5
10.8
12 SM 1650-7
12.8
10.9
19.0
11.7
13.6
13 SM 805-15
15.9
12.4
18.6
10.9
10.5
14 SM 1778-53
11.8
10.9
20.7
12.2
10.5
15 SM 1624-2
13.7
10.7
19.9
10.0
10.2
16 SM 1627-16
10.4
11.2
18.6
17.3
7.3
17 CM 6754-8
12.4
7.4
24.1
9.1
12.8
18 SM 1669-7
11.8
9.5
19.2
11.7
11.9
19 SM 1759-29
13.5
7.7
20.9
13.1
10.3
20 M COL 1505
12.8
12.6
19.9
13.6
7.3
21 SM 643-17
13.7
7.9
16.6
11.7
13.0
22 CM 3306-19
13.9
11.7
21.1
9.9
9.1
23 M COL 2215
14.0
10.2
18.0
12.0
8.3
24 SM 1669-5
14.3
11.1
15.4
11.9
10.6
25 CG 1141-1
12.6
11.2
18.9
11.5
8.0
26 SM 1438-2
13.8
12.6
16.0
9.9
9.1
27 CM 3306-4
11.7
12.1
18.0
7.9
11.7
28 CM 8027-3
n.a.
n.a.
16.4
10.0
11.0
29 SB 0216-9
11.9
14.2
14.5
11.9
8.3
30 CM 523-7
14.1
9.2
18.7
10.9
7.3
31 CM 6119-5
12.5
8.5
17.0
12.7
6.0
32 SM 1778-45
10.3
8.2
17.7
13.7
6.1
33 SM 1657-14
10.0
9.9
15.2
11.6
7.5
34 MBRA 384
8.0
7.9
21.1
10.9
6.5
35 MTAI 8
10.8
8.3
16.7
11.0
6.7
36 CM 6758-1
9.2
11.0
11.0
13.7
8.6
37 CM 4843-1
12.3
9.7
11.1
9.5
10.6
38 CM 8475-4
12.8
5.3
18.5
7.9
6.6
39 MPER 183
10.5
9.1
16.4
9.1
6.5
Maximum
20.4
15.8
29.2
17.3
15.0
Minimum
8.0
5.3
11.0
7.9
6.0
Mean
13.0
10.5
19.1
11.5
10.1
¶ DMC Dry matter content in %
Project IP3: improving cassava for the developing world
Yield (t/ha)
Fresh
Dry
42.1
15.3
41.0
15.1
42.1
15.2
41.4
14.9
45.0
15.1
40.6
14.5
39.1
14.3
39.0
14.2
39.8
14.2
38.7
14.1
40.3
14.1
36.6
13.6
39.9
13.6
36.4
13.2
34.2
12.9
34.8
13.0
36.7
13.2
33.6
12.8
36.5
13.1
39.6
13.2
32.6
12.6
39.6
13.1
33.5
12.5
35.6
12.7
35.2
12.5
33.2
12.3
33.7
12.3
35.3
12.4
34.2
12.2
34.5
12.1
31.5
11.3
31.6
11.2
30.7
10.8
31.6
10.9
30.9
10.7
31.2
10.7
31.7
10.6
27.1
10.2
34.6
10.3
45.0
15.3
27.1
10.2
36.0
12.8
DMC
(%)
36.3
37.0
36.3
36.2
33.2
35.9
36.8
36.3
35.8
36.5
34.4
37.1
34.2
36.4
37.7
37.2
35.4
38.2
35.9
33.7
38.4
33.1
37.3
35.8
35.7
37.2
36.9
35.1
35.5
34.9
35.7
35.7
35.1
34.7
34.6
34.1
33.7
37.0
30.0
38.4
30.0
35.7
Output 2-28
Table 2.19. Results of the Regional Trial in La Unión (Sucre Department) and Ciénaga de Oro
(Córdoba Department). Ranking of clones is based on dry matter yield (t/ha).
Clone
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
SM 1637-22
CM 9067-2
CM 4919-1
SM 1656-7
TAI 8
SM 1438-2
SM 1411-5
SM 1521-10
SM 2192-6
SM 1669-7
CM 9067-4
SM 1669-5
SM 1665-2
SM 1511-6
SM 805-15
SM 1127-8
SM 1650-7
SM 1427-1
CM 6119-5
SM 1565-17
PAN 135
SGB 765-4
SM 1759-29
VEN 169
SM 1969-31
M VEN 25
SM 2081-34
CM 3306-19
SM 2450-5
CM 4843-1
SM 2449-22
SM 1624-2
CM 6754-8
CM 8475- 4
SM 1973-25
SM 1433-4
CM 8472-5
SM 1201-5
SM 1619-3
CM 7985-16
SM 643-17
SM 1948-33
Fresh root yield
(t/ha)
C. de Oro Corozal
36.9
36.5
30.4
24.9
27.3
32.8
33.2
29.7
30.8
24.9
25.1
31.2
27.1
25.2
27.5
17.4
27.3
26.4
19.3
35.8
21.6
25.8
24.1
23.3
25.8
24.3
24.5
13.5
17.2
27.6
23.8
28.3
24.9
36.8
6.7
17.2
21.8
11.1
3.7
23.2
14.8
18.7
36.9
3.7
24.5
Maximum
Minimum
Mean
¶ DMY Dry matter yield.
Output 2-29
30.8
40.1
29.3
37.7
29.8
29.2
34.7
48.3
34.3
31.6
37.2
40.1
45.8
29.9
38.9
33.6
43.7
32.1
24.2
36.1
30.1
27.3
28.2
25.9
26.3
37.8
40.8
44.1
31.1
36.4
38.2
29.6
31.7
27.3
26.1
34.3
18.8
29.5
19.7
21.8
20.3
17.3
48.3
17.3
32.1
Dry matter
content (%)
C. de Oro Corozal
34.3
31.5
32.0
33.8
33.2
33.9
33.4
30.6
32.5
32.8
32.3
31.2
30.2
33.3
32.6
34.0
32.6
32.9
33.6
28.9
32.2
34.7
33.0
31.5
32.0
30.8
31.0
29.5
33.0
31.0
27.8
32.2
31.8
31.3
34.1
31.1
30.9
32.6
33.8
30.8
31.1
31.2
34.7
27.8
32.1
38.2
36.9
37.1
39.2
37.2
39.8
37.2
35.8
37.1
38.4
35.9
35.3
35.2
39.0
35.6
36.6
37.2
37.3
36.3
32.8
37.1
40.4
39.5
36.9
36.7
35.2
34.4
33.4
35.1
34.4
34.2
37.2
35.2
34.1
37.4
32.9
35.1
37.4
39.2
36.5
37.3
33.4
40.4
32.8
36.5
Harvest Index
(0-1)
C. de Oro Corozal
0.65
0.69
0.68
0.55
0.70
0.61
0.66
0.71
0.64
0.68
0.68
0.67
0.74
0.62
0.66
0.61
0.59
0.65
0.61
0.70
0.62
0.54
0.49
0.59
0.58
0.60
0.61
0.76
0.63
0.70
0.66
0.52
0.64
0.57
0.60
0.55
0.63
0.60
0.57
0.52
0.46
0.49
0.8
0.5
0.6
0.65
0.66
0.70
0.57
0.67
0.61
0.62
0.63
0.63
0.68
0.72
0.71
0.75
0.63
0.67
0.66
0.54
0.66
0.62
0.78
0.60
0.42
0.57
0.62
0.59
0.57
0.58
0.66
0.61
0.66
0.63
0.45
0.59
0.50
0.60
0.63
0.58
0.51
0.52
0.40
0.49
0.38
0.8
0.4
0.6
DMY¶
(t/ha)
Across
Plant type
(1-5)
Across
12.2
13.1
10.3
11.6
10.1
11.4
12.0
13.2
11.4
10.2
10.7
11.9
12.2
10.0
11.4
9.1
12.6
10.4
7.6
11.1
9.0
10.0
9.5
8.4
8.9
10.4
10.8
9.3
8.3
10.6
9.8
10.0
9.5
10.4
6.0
8.3
6.7
7.3
4.5
7.5
6.1
5.8
13.2
4.5
9.8
1.17
2.00
1.00
1.50
1.33
2.17
2.33
2.33
2.17
2.33
2.17
2.33
2.67
2.17
2.67
1.83
3.00
2.83
1.00
2.17
2.17
3.17
2.50
1.67
2.17
2.33
2.83
2.33
2.50
3.83
2.33
3.17
3.50
3.33
3.00
2.17
2.17
3.50
3.33
3.00
3.33
2.33
3.8
1.0
2.4
2003 Annual Report
Table 2.19 presents the resuls of two Regional Trials for the Sucre and Córdoba
Departments, which are intermediate between the sub-humid conditions in the Northern
Coast (around Barranquilla) and the Urabá Region. Out of the 42 clones evaluated, SGB 7652 ranked 22d. Unfortunately there was no seed available from SGB 765-2 for this trial.
Among the best clones are genotypes that have had consistently reliable performance in the
last few years (CM 4919-1, SM 1656-7, MTAI 8, SM 1438-2 and SM 1411-5) as well as new
ones (SM 1637-22 and CM 9067-2). Most of these clones have been included as parents for
this environment (see Table 2.1).
Table 2.20. Results of the Regional Trial conducted at Montelíbano (Córdoba Department).
Clone
1 CM 4919-1
2 CM 523-7
3 MTAI 8
4 SM 1511-6
5 CM 3306-19
6 SM 643-17
7 CM 3306-4
8 SM 1565-17
9 SM 1411-5
10 CM 4843-1
11 MBRA 384
12 M VEN 25
13 CM 6119-5
14 CM 6754-8
15 SM 1669-7
16 SM 805-15
17 MCOL 1505
18 SGB 765-2
19 SM 1669-5
20 MCOL 2215
21 SM 1973-25
22 SM 1438-2
23 SM 1650-7
24 CG 1141-1
25 SM 1778-45
26 CM 6758-1
27 SGB 765-4
28 SM 1758-53
29 SM 1759-29
30 CM 1627-16
Maximum
Minimum
Mean
Fresh
roots
(t/ha)
22.2
13.3
21.7
21.7
20.0
21.3
19.4
21.1
16.7
19.4
20.8
19.4
15.6
16.1
18.9
15.9
14.4
18.3
15.6
12.8
9.4
16.1
15.0
15.6
11.1
11.1
10.0
10.0
10.0
8.9
22.2
9.4
16.3
Dry
matter
(%)
32.9
44.3
34.5
34.5
31.6
32.5
34.8
30.5
34.3
30.3
30.0
32.1
34.5
32.5
33.0
33.5
34.8
32.3
33.0
34.7
37.4
33.7
29.6
34.0
33.1
34.7
35.7
33.0
31.5
32.2
44.3
29.6
33.6
Dry
Matter
(t/ha)
7.3
4.3
7.5
7.5
6.3
7.0
6.8
6.4
5.7
5.9
6.2
6.3
5.4
5.2
6.2
5.2
5.0
5.9
5.1
4.4
3.5
5.4
4.5
5.3
3.7
4.0
3.6
3.3
3.2
2.8
7.5
3.2
5.4
Harvest
index
(0 a 1)
0.77
0.58
0.67
0.66
0.78
0.67
0.62
0.74
0.69
0.73
0.70
0.65
0.67
0.69
0.57
0.64
0.63
0.60
0.65
0.63
0.63
0.56
0.75
0.53
0.68
0.59
0.57
0.53
0.58
0.54
0.78
0.53
0.65
Selection
Index
21.61
18.96
17.97
17.05
13.21
10.52
9.85
9.34
6.04
4.25
4.00
3.27
2.44
-0.32
-1.08
-1.55
-1.83
-2.05
-3.29
-6.31
-6.94
-6.70
-7.29
-8.80
-11.78
-13.53
-14.63
-25.87
-26.55
-29.71
21.61
-26.55
0.00
na = not available
Project IP3: improving cassava for the developing world
Output 2-30
Finally, Table 2.20 presents the results of yet another Regional Trial evaluated at
Montelíbano (Córdoba Department). In this trial both SGB 765-2 aned SGB-4 were included
but they did not perform as well as in the Urabá region (18th and 27th rankings,
respectively). Other clones that were outstanding based on results from Table 2.19 showed
good performance in Montelíbano (CM4919-9, SM 1511-6, and MTAI 8), but others showed
strong genotype x environment interaction (SM 643-17, CM 3306-19 and SM 1565-17).
2.4.2 Selections for the Acid-Soil Savannas Environment
As for the Caribbean coastal eco-region, only the most relevant experiments conducted for
this environment are described below (Table 2.21) followed by the respective results for each
type of evaluation. As for Barranquilla, many of the improvement activities developed for the
Villavicencio area also benefiting other regions.
The F1 stage was planted in CIAT-Palmira, this was a result of the measures taken to control
both the white flies and frog skin disease. Many botanical seeds were germinated (4093) but
only 3110 produced vigorous enough seedlings to be transplanted to the F1 plot (Table 2.5).
Plants that produce at least seven stakes in May next year will be used for the Clonal
Evaluation Trial of year 2004.
Table 2.21. Trials conducted in the Acid Soil Savannas in the 2002-2003 cycle¶.
Trial
Site
F1
Clonal
Evaluation
Diallel study
CET from
diallel
Preliminary
Yield Trial
Advanced
Yield Trials
Regional
Trials
N° of
N° of reps
genotypes
Observations
CIATPalmira
2605
1
See Table 2.22
La Libertad
1235 (7)
1
See Table 2.23 and 2.26
La Libertad
1360 (6)
3
Data not presented
La Libertad
238
1
See Table 2.27
La Libertad
180 (10)
3
See Tables 2.28-2.32
La Libertad
64 (25)
56 (30)
3
3
See Table 2.33 to 2.36
See Table 2.37 to 2.40
Four locations
27 (25)
3
See Tablea 2.40 and 2.41
Values in parentheses refer to the number of plants per plot. § Genotypes involved in the
diallel experiment.
¶
Output 2-31
2003 Annual Report
A description of the origin of the 1071 clones of the Clonal Evaluation Trial planted in June
2003 is provided in Table 2.22. In the year 2001, a total of 4358 botanical seeds had been
planted for the Acid Soil Savannas region. Many did not germinate or else, produced weak
seedlings that died early before or soon after transplantation to the field. Eventually, only
1071 of the 2605 plants transplanted in August 2002 were vigorous enough to produce the
seven stakes required for the Clonal Evaluation Trial planted for this cycle (Table 2.22). Most
of the progenitors utilized to generate the Clonal Evaluation Trial have had an outstanding
performance (as it will be demonstrated later in this Section); or else had been included for
specific purposes.
Because of the prevalence of foliar diseases such as cassava bacterial blight (Xanthomonas
axonopodis pv. manihotis) and superelongation (induced by the fungus Sphaceloma
manihoticola), evaluations must ensure optimal disease pressure. It is desirable to eliminate,
as early as possible in the improvement process, those genotypes susceptible to these
diseases. Thus, in the Clonal Evaluation Trial, the furrows were located, one behind each
other, in a single row and separated by plants that served as spreaders of these diseases.
These spreader plants permitted not only high pressure, but also ensured uniform
distribution of the diseases. Planting material for spreader plants were stakes chosen from
plants that had been discarded, precisely for being susceptible to these diseases, during the
previous cycle. The high mortality of spreader plants observed 1-2 months after planting
demonstrates that their spreading role has been adequately fulfilled early in the season.
Figure 2.2 illustrates the way the spreader plants were located through the trial.
There is a slight difference in the way the Clonal Evaluation Trial is harvested in the Acid
Soils Savannas, compared with the drier environment of the Northern Coast of Colombia.
There is no marked period without rains, therefore, the harvest can be carried out in one
step. This is the reason why only seven (rather than eight) plants were used to represent each
clone. All the plants were harvested together in May. One other difference for this trial is
that Plant Type incorporates a heavy component of reaction to foliar diseases mentioned
above. The large number of progenies that could be evaluated in this trial are the results of
the financial support by the Ministry of Agriculture of Colombia and the Poultry Growers
Association of Colombia (FANAVI). It also reflects the relative success that the measures
taken to control Frog Skin Disease at the F1 stage.
As in the case of the previous region, selection in the Acid Soils Savannas was also
conducted through a selection index:
Selection index = [FRY ∗ 10] + [DMC ∗ 8] - [PT ∗ 8] + [HI ∗ 5]
where,
FRY = fresh root yield
DMC = dry matter content
PT = plant type using a 1(excellent) to 5 (very poor) visual scale
HI = harvest index
The weight given to plant type has been increased to 8 (it was 5 in the sub-humid ecoregion), because of the heavier pressure to select for materials resistant to foliar diseases
present in the acid soils savannas.
Project IP3: improving cassava for the developing world
Output 2-32
Table 2.22. Origin of the 1071 clones for the Clonal Evaluation Trial planted in May, 2003 at
CORPOICA – La Libertad (Villavicencio, Meat Department). SM families have unknow
fathers, but it is certain that they come from a elite clones in the policross plots.
Family
Mother
Father
CM 9460
CM 9901
CM 9942
GM 220
GM 221
GM 223
GM 224
GM 229
GM 233
GM 241
GM 243
GM 256
GM 275
GM 276
GM 277
GM 305
GM 371
GM 396
GM 400
GM 507
GM 512
GM 514
GM 515
GM 517
GM 536
GM 538
GM 542
GM 543
GM 545
SM 2634
SM 2792
SM 2967
SM 2980
SM 3019
SM 3022
SM 3026
SM 3029
SM 3031
SM 3032
SM 3068
SM 3069
SM 3073
SM 3074
SM 3075
SM 3076
SM 3077
SM 3081
SM 3083
SM 3084
TOTAL
CM 6740-7
CM 6740-7
SM 1741-1
CM 4574-7
CM 4574-7
CM 4574-7
HMC 1
SM 1565-15
CM 6740-7
SM 1565-15
SM 2219-11
SM 1565-15
SM 2058-2
SM 2219-11
SM 1565-15
SM 2219-11
SM 1219-9
SM 1219-9
SM 1859-26
SM 1219-9
SM 1219-9
SM 1219-9
SM 1219-9
SM 1219-9
SM 1565-15
SM 1565-15
SM 1565-15
SM 1565-15
SM 1565-15
CM 6438-14
SM 1565-15
CM 7033-3
SM 2219-11
CM 523-7
CM 2772-3
CM 6740-7
SM 1219-9
MCOL 2737
MCOL 2758
CM 4574-7
CM 6740-7
SM 1219-9
SM 1565-15
SM 1741-1
SM 1779-8
SM 1820-8
SM 1862-25
MBRA 383
MCOL 2737
CM 4574-7
SM 1219-9
SM 653-14
SM 1219-9
SM 1565-15
SM 2219-11
CM 4574-7
CM 6740-7
SM 2219-11
CM 7033-3
CM 7033-3
SM 1219-9
SM 1565-15
SM 1565-15
HMC 1
MTAI 8
CM 2772-3
CM 4574-7
CM 4574-7
SM 985-9
SM 1779-8
SM 1859-26
SM 1862-25
MCOL 2737
CM 4574-7
SM 985-9
SM 1859-26
SM 1862-25
MCOL 2737
Output 2-33
Planted
Transplanted
Harvested
12
14
16
13
34
23
21
28
29
25
15
19
28
39
31
22
28
17
24
5
17
14
24
25
53
24
10
34
26
20
28
15
20
13
16
24
27
5
10
31
14
22
34
27
21
30
15
9
20
4358
17
43
59
29
51
42
42
55
53
35
48
73
51
67
60
62
71
64
48
43
45
55
66
49
80
60
44
75
32
41
52
36
55
78
44
69
56
9
46
65
37
71
53
71
70
83
73
29
48
2605
12
14
16
13
34
23
21
28
29
25
15
19
28
39
31
22
28
17
24
5
17
14
24
25
53
24
10
34
26
20
28
15
20
13
16
24
27
5
10
31
14
22
34
27
21
30
15
9
20
1071
2003 Annual Report
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
Spreader plants
O
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
O
O
O
O
O
O
O
O
O
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Alleyway for circulation
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Spreader plants
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Figure 2.2. Illustration of the way spreader plants were positioned in the Clonal Evaluation
Trial at CORPOICA – La Libertad, to provide uniform and high disease pressure.
Results of the Clonal Evaluation Trial are presented in Tables 2.23 and 2.24 Good
development of leaf diseases could be observed early, together with a wide range of variation
for both cassava bacterial blight and super-elongation. The fraction selected (Table 2.23),
reacted well to leaf diseases (average for plant type = 2.26), compared with the average for the
whole population (3.39). Similarly, good selection pressure was achieved for dry matter
productivity (10.34 versus 6.94 t/ha), as for dry matter content (34.32 % versus 31.76 %),
and harvest index (0.55 versus 0.50).
Data from Table 2.24 is useful for understanding why some families fail to contribute with
clones worth to be selected, while others are outstanding. For instance, none of the 30 clones
from family GM115 was selected. These clones had fresh-root yield (19.8 t/ha) lower than the
average (21.6 t/ha), and poor dry matter content (29.6 % versus an average of 31.8%).
Moreover, the average for plant type was 4.3. This means that they had a very poor reaction
to foliar diseases and/or very undesirable plant architecture. Other families with overall poor
performance were SM 2842, SM 2972, and SM 2973, with more than 20 clones, none of
which was selected.
In contrast families SM2642, SM 2658 and CM 9903 had more than 30% of their clones
selected (compared with 14.6% for the entire trial). The 46 clones from family GCM 9903
Project IP3: improving cassava for the developing world
Output 2-34
yielded 24.2 t/ha of fresh roots, with good level of dry matter content (33.4 %) and plant type
(2.80). Harvest index was 0.52, slightly avobe the average of 0.50.
Table 2.23. Results of the selection carried out in the Clonal Evaluation Trial CORPOICA-La
Libertad, Department of Meta, from 1235 clones evaluated during May 2002 to May
2003. Clones from the same family were randomly allocated to one of three blocks
within the trial. This blocking allowed each family to be replicated three times, but
individual clones were planted only in one of these blocks. Therefore, there was no
replication for individual clones.
Parameter or
Genotye
Yield (t/ha)
Fresh roots
Dry matter
Harvest Index
(0 to 1) ¶
Plant type
(1 to 5) §
Dry matter
content(%)
Selection
Index
Results from the 1235 clones evaluated across the three blocks
72.57
15.52
0.95
5.00
41.48
63.48
0.35
0.06
0.05
1.00
16.75
-95.90
21.64
6.94
0.50
3.39
31.76
0.00†
7.53
2.59
0.09
1.09
3.57
21.03
Averages of the 412, 412 and 411 clones in Blocks 1, 2 and 3, respectively
Block 1
20.88
6.66
0.50
3.33
31.59
0.00†
Block 2
21.73
6.88
0.49
3.35
31.24
0.00†
Block 3
22.30
7.28
0.50
3.48
32.44
0.00†
Results from the 180 clones selected
Maximum
72.57
15.52
0.95
5.00
41.48
63.48
Minimum
15.97
6.32
0.40
1.00
17.90
19.60
Mean
30.41
10.34
0.55
2.26
34.32
28.87
Std. Dev.
6.73
1.83
0.06
0.87
2.68
7.38
Best 5 clones selected from each block in the Clonal Evaluation Trial
CM 9953- 55
45.14
15.2
0.61
1
33.7
63.48
GM 240- 61
35.42
11.0
0.65
1
30.9
45.63
GM 276- 72
33.68
11.3
0.52
1
33.5
41.11
CM 9903- 59
28.47
9.9
0.59
1
34.8
40.90
SM 2634- 29
32.99
11.7
0.58
2
35.4
40.62
SM 2634- 49
35.76
13.3
0.54
2
37.1
46.06
SM 2642- 50
29.79
10.8
0.53
1
36.2
42.87
SM 2968- 28
38.00
12.7
0.56
2
33.4
41.51
CM 9903- 70
29.17
10.9
0.61
2
37.3
41.40
SM 2642- 45
36.46
12.8
0.53
2
35.0
41.29
SM 2972- 16
72.57
13.0
0.95
3
17.9
57.55
CM 9953- 81
33.61
12.3
0.63
1
36.6
46.89
SM 2965- 34
36.91
12.9
0.62
1
34.9
46.75
SM 2855- 13
34.43
12.4
0.61
1
36.0
45.73
SM 3022- 18
60.24
15.2
0.71
3
25.2
45.30
¶ The harvest index is obtained by dividing the production of commercial roots by total biomass (roots
+ aerial parts). Preferred harvest indexes are > 0.5.
§ Plant type integrates under one value, plant architecture, leaves health, and capacity to produce
stakes on a scale where 1 = excellent and 5 = very poor is used.
†
Average election index within blocks must be zero, because it is based on a combination of
standardized variables. However, when averaged across the three blocks there is a slight deviation
because selection indices were estimated for each block separately.
Maximum
Minimum
Mean
Std. Dev.
Output 2-35
2003 Annual Report
Table 2.24. Results of the Clonal Evaluation Trial grown at CORPOICA-La Libertad, presented as average for each
family represented in the trial which involved 49 families and a total of 1235 clones. Each family of clones
was divided in three groups each group randomly assigned to a block with the trial.
%
%
% Plant HI
SI
# select. Fresh DM
DM Type (0-1)
§
Family
# select.
(1-5)
clones
¶
clones
(t/ha) (t/ha)
CM 6787
32
18.8 22.6
7.2 31.7 3.0 0.54
5.7 SM 2848
25
4.0
CM 9474
44
11.4 22.4
6.8 30.3 3.4 0.58
2.1 SM 2850
9
0.0
CM 9831
18
22.2 23.8
7.7 32.3 3.4 0.52
5.5 SM 2852
7
14.3
CM 9903
46
32.6 24.2
8.1 33.4 2.8 0.52 12.9 SM 2853
4
0.0
CM 9918
29
3.5
22.6
7.2 31.6 3.8 0.57
1.7 SM 2854
28
10.7
CM 9940
33
21.2 23.7
8.2 34.7 3.1 0.52 13.0 SM 2855
13
15.4
CM 9953
33
24.2 24.8
8.1 32.7 2.8 0.56 14.0 SM 2857
6
0.0
GM 115
30
0.0
19.8
5.9 29.6 4.3 0.54 -11.6 SM 2870
32
6.3
GM 240
14
28.6 23.0
7.1 31.3 2.7 0.54
8.9 SM 2899
38
2.6
GM 276
36
27.8 24.7
8.4 33.7 2.4 0.50 16.0 SM 2965
38
23.7
SM 2366
47
17.0 21.4
7.1 32.7 3.4 0.46
-0.8 SM 2966
12
8.3
SM 2610
60
6.7
17.5
6.0 33.3 3.8 0.44
-8.4 SM 2968
42
19.1
SM 2632
45
8.9
19.4
6.1 31.1 3.1 0.46
-4.5 SM 2970
19
15.8
SM 2634
48
22.9 22.8
7.8 33.9 2.9 0.47
8.4 SM 2971
4
0.0
SM 2640
38
18.4 22.5
7.8 34.8 3.4 0.48
7.3 SM 2972
20
10.0
SM 2642
24
37.5 22.6
7.6 33.3 2.5 0.44
9.4 SM 2973
20
10.0
SM 2649
16
0.0
19.3
5.8 29.9 4.2 0.53 -11.3 SM 2974
29
17.2
SM 2658
15
33.3 24.1
7.4 30.5 3.5 0.56
3.5 SM 2976
27
25.9
SM 2739
20
25.0 24.4
7.9 31.7 3.4 0.52
4.8 SM 2977
21
19.1
SM 2792
13
7.7
18.4
5.7 30.2 3.2 0.44
-8.8 SM 2978
19
21.1
SM 2841
19
10.5 20.9
6.3 29.9 4.1 0.51 -10.0 SM 2979
8
12.5
SM 2842
22
0.0
17.1
5.0 27.8 4.3 0.48 -22.5 SM 2980
27
3.7
SM 2844
15
0.0
19.0
5.8 29.0 4.2 0.49 -16.1 SM 2982
15
13.3
SM 2846
31
3.2
19.9
6.2 30.9 4.4 0.47 -13.0 SM 3022
18
11.1
SM 2847
26
7.7
22.5
7.2 31.6 3.8 0.50
-2.1 Total
1235 14.6
¶ HI = Harvest Index (Root production / total biomass).
§ SI = Selection Index (combines several variables of economic relevance)
Family
Project IP3: improving cassava for the developing world
%
Fresh DM DM
(t/ha) (t/ha)
20.3
6.4
31.4
16.8
5.4
31.3
24.7
8.1
33.0
18.6
6.4
34.3
18.2
5.5
30.3
20.9
7.1
32.5
22.3
6.7
30.0
17.1
5.6
31.9
22.9
7.1
31.1
23.3
7.5
31.6
19.9
6.3
31.2
22.4
7.4
32.5
21.0
6.7
31.9
17.8
5.1
27.5
22.1
6.3
29.6
21.0
6.5
30.6
18.3
5.9
31.6
23.3
7.7
33.0
23.6
7.7
32.3
24.3
7.4
30.5
20.0
6.2
30.5
22.0
6.7
30.2
24.9
7.4
29.2
25.3
7.2
28.8
21.6 6.94 31.8
Output 2-36
Plant
Type
(1-5)
3.6
3.8
3.3
3.3
3.3
3.5
3.5
3.1
3.8
3.4
3.8
3.5
3.2
3.3
3.5
3.5
3.6
3.1
3.5
2.9
3.4
3.4
2.9
3.2
3.4
HI
SI
(0-1)
§
¶
0.47
-6.2
0.39 -16.7
0.51
7.8
0.45
0.0
0.42 -11.1
0.47
-2.3
0.48
-4.7
0.47
-5.4
0.53
-0.8
0.48
0.8
0.48
-7.4
0.52
3.1
0.45
-2.4
0.43 -18.1
0.47
-7.3
0.53
-1.9
0.43
-8.9
0.52
9.4
0.49
2.8
0.50
3.8
0.49
-5.3
0.53
-1.2
0.52
3.7
0.56
2.7
0.50
0.0
The information from Table 2.24 can be used to produce information of all the
progenies from a given parent (parents are used in generating more than one family).
The averages of the progenies derived from the best and worst five progenitors are
presented in Table 2.25. Progenitor SM 1565-15 produced a large progeny of 73 clones
and 27.4% of them were selected. Plant type and super elongation disease score were
excellent for progeny of this clone (2.7 and 2.18, respectively), but harvest index was
below average (0.46). This is a reflection of the characteristics of this clone that tends
to have large volume of foliage at harvest time (as a result of its disease resistance)
which results in below average harvest index. As found last year clone CM 7033-3
produced progenies with a performance that was better than the average: selection
index of 8.86, 23.o t/ha of fresh roots, the lowest super elongation disease score
(1.93), but about average dry matter content (31.3%).
The progeny from SM 1583-8, on the other hand, performed poorly with an average
selection index of –22.5 and none of the 22 clones derived from it was selected were
were marginally superior to the mean of the population (selection index= 0.44). Dry
matter content was very low (27.8%) suggesting that the progenies from this clone
were under stress, a hypothesis supported by the high super elongation disease (3.5)
and plant type (4.3) scores. These are but examples of the usefulness of the data
generated by the Clonal Evaluation Trial, in addition to the selection process of the
best clones to pass to the following step.
Table 2.25. Averages of all the progenies derived from a common progenitor based on
the data produced by the Clonal Evaluation Trial at CORPOICA – La Libertad
(Villavicencio, Meta Department).
Number Seleced Selected Plant
Fresh
Super- Harvest Dry Selection
Progenitor of clones Clones Clones
type
roots elongation Index matter Index
(1-5)
(%)
(#)
(0 - 1)
(%)
(1 - 5)
(t/ha)
Average performance of the best progenies and their respective female progenitor
SM 1460-1
15
5
33.33
3.5
24.1
3.13
0.56
30.5
3.51
CM 7033-3
14
4
28.57
2.7
23.0
1.93
0.54
31.3
8.86
SM 1565-15
73
20
27.4
2.7
21.9
2.18
0.46
32.4
5.52
SM 1822-5
20
5
25
3.4
24.4
2.80
0.52
31.7
4.83
SM 1861-18
34
8
23.53
3.2
24.0
2.47
0.52
33.0
8.63
CM 6438-14
48
11
22.92
2.9
22.8
2.21
0.47
33.9
8.43
Average performance of the worst progenies and their respective female progenitor
SM 2059-7
38
1
2.63
3.8
22.9
2.8
0.53
31.1
-0.8
CM 6370-2
16
0
0.00
4.2
19.3
3.5
0.53
29.9
-11.3
SM 1583-8
22
0
0.00
4.3
17.1
3.5
0.48
27.8
-22.5
SM 1690-13
4
0
0.00
3.3
17.8
3.5
0.43
27.5
-18.1
SM 2075-1
30
0
0.00
4.3
19.8
4.0
0.54
29.6
-11.6
MCOL 2298
6
0
0.00
3.5
22.3
2.8
0.48
30.0
-4.7
Table 2.26 presents phenotypic correlation values among different relevant traits.
Worth mentioning is the high correlation between root productivity and root aspect
score (ρ= -0.52); foliage production (ρ= 0.53); and harvest index (ρ= 0.58). Plant type
Output 2-37
2003 Annual Report
also had, as expected a negative correlations with root productivity (ρ= -0.26) and
super elongation disease score (ρ= 0.69). It should be remembered that the scale for
these scores mean range from 1=excellent and 5=very poor.
Table 2.26. Phenotypic correlacions for relevant traits evaluated in the Clonal
Evaluation Trial (CORPOICA – La Libertad) from 1235 genotypes.
Plant type
(1-5)
Root score
(1-5)
Super elongation
disease (1-5)
Fresh root yield
(t/ha)
Fresh foliage yield
(t/ha)
Harvest Index
(0-1)
Dry matter
content (%)
Root
score
(1-5)
Super
elongation
(1-5)
Fresh
root yield
(t/ha)
Fresh
foliage
yield
(t/ha)
Harvest
Index
(0-1)
Dry
matter
content
(%)
0.14
0.69
-0.26
-0.18
-0.13
-0.26
1.00
0.11
-0.52
-0.33
-0.27
-0.15
1.00
-0.23
-0.18
-0.09
-0.24
1.00
0.53
0.58
0.24
1.00
-0.29
0.29
1.00
0.07
1.00
A second planting of the diallel study for the acid-soil savannas environment was
evaluated this year. Results from this trial will be combined with those from 2002 for
publication. The ten parents involved in this study were crossed among themselves
with each cross represented by 30 clones. Taking advantage of the data from 2002 a
second Clonal Evaluation Trial was planted with the best 238 performing clones. Table
2.27 presents the number of clones from each progenitor included in this second
Clonal Evaluation Trial (the total number of clones will appear to be twice as many as
236 because each clone is derived from two parents). Since each parental line was
represented by about 270 clones in the dialle study a larger number of clones in the
Clonal Evaluation Trial (second column in Table 2.27) means outstanding performance
in the diallel evaluation (i.e. CM 4574-7 was represented by 74 clones derived from it).
Table 2.27 also presents the selections after the Clonal Evaluation Trial set up from the
diallel study (third column in Table 2.27). For instance, 37.84% of the 74 clones
derived from CM 4574-7 were selected in this phase. The correlation between selection
at the diallel stage (represented by figures in second column) and at the Clonal
Evaluation Trial (third column) was relatively high (ρ= 0.60), indicating consistency of
family performance across the two consecutive evaluations in years 2002 and 2003.
Project IP3: improving cassava for the developing world
Output 2-38
Table 2.27. Results from a Clonal Evaluation Trial derived from the evaluation of a 10parents diallel crosses the previous year, where each progenitor was
represented by a total of 270 clones. The number of clones listed in the second
column indicate how many of these clones were selected for the trial conducted
this year. The proportion of clones selected in 2003 agrees well with that of
2002.
#
clones
Progenitor
CM 4574-7
CM 7033-3
SM 1219-9
CM 6740-7
SM 1565-15
SM 2058-2
SM 2219-11
MPER 183
MTAI 8
HMC 1
74
48
54
61
58
33
74
6
47
21
%
selected
clones
37.84
31.25
29.63
29.51
25.86
21.21
17.57
16.67
12.77
4.76
Fresh
roots
(t/ha)
14.26
14.80
12.99
13.48
12.47
15.04
13.61
15.21
11.73
10.22
Fresh
foliage
(t/ha)
14.62
10.87
9.72
13.90
13.63
13.05
11.86
12.66
10.46
9.76
Dry
matte
r (%)
32.03
30.85
31.46
30.58
33.29
30.09
29.79
28.38
29.92
30.81
Harvest
Index
(0-1)
0.49
0.57
0.57
0.48
0.50
0.54
0.54
0.54
0.52
0.51
Plant
Type
(1-5)
2.24
2.98
3.33
2.75
3.10
3.67
3.35
3.83
3.89
4.00
Selection
Index
7.23
4.80
1.59
-0.20
2.08
-1.96
-2.72
-6.14
-9.21
-10.65
The correlation coefficient between selections this year and in the previous year was 0.60
The second stage of evaluation according to the diagram illustrated in Figure 2.1 is the
Preliminary Yield Trial. Similar to what was done in the northern coast, the 180 clones
selected the previous year from the Clonal Evaluation Trial, were split in three different
experiments with 64 entries each. The trials had 10-plant plots and three replications
each. Four checks (Brasilera, ICA-Catumare or CM 523-7, CORPOICA-Reina or CM
6740-7 and CM6438-14) were also included in each trial. A square lattice
experimental design was used. Tables 2.28, 2.29, and 2.30 present the most relevant
results from these trials. Clones derived from a given family were distributed in the
three experiments. The results from the best 12 experimental clones and the four
checks from each experiment or Preliminary Yield Trial are listed.
Brasilera (MCOL 2737) was by far the best check ranking 4th, 2nd, and 5th in
Experiments 1,2, and 3, respectively. This clon has been included as parental line in
the crosses for the Acid Soil Savannas (see Table 2.1). CM 6740-7 also showed a good
performance and was among the best 13 clones in the three experiments.
The analysis of the combined selections across the three Preliminary Yield
Experiments is summarized in Table 2.31. Clones from a total of 31 families were
included in the experiments, and clones from 20 families were selected. The proportion
of selected clones varied considerably. The three clones evaluated from familiy GM 275
were selected (100%) as well as the only clon from family GM 235. Families GM 241,
GM 261, GM256 had a high proportion of their clones also selected. Interestingly,
these families (with the exception of GM 235, had a higher than average selections at
the Clonal Evaluation Trial phase harvested the previous year: GM 275 (25.0%), GM
235 (7.4%), GM 241 (53.8%), GM 261 (23.8%), and GM256 (42.5 %). On the other
hand, few clone from families GM 219 and GM 211were selected in the Preliminary
Output 2-39
2003 Annual Report
Yield Trial, although a high proportion (75 and 84.2 %) were selected last year at the
Clonal Evaluation Trial phase.
Rank
Table 2.28. Results of the best twelve clones and the four checks in Experiment 1 of
the Preliminary Yield Trials conducted at CORPOICA – La Libertad (Villavicencio,
Meta Departament). Statistical parameters of the 64 entries evaluated are also
provided.
CLON
1 GM 235-55
2 CM 9901-56
3 GM 219-48
4 GM 227-38
5 GM 221-59
6 GM 241-35
7 GM 301-47
8 GM 256-40
9 GM 233-52
10 GM 220-59
11 GM 223-67
12 GM 223-34
4 Brasilera
13 CM 6740-7
14 CM 6438-14
57 CM 523-7
Maximum
Minimum
Mean
Std.Deviation
Roots (t/ha)
Fresh
Dry
matter
30.9
11.0
18.9
6.5
33.8
10.1
28.0
8.5
31.6
9.6
31.1
9.9
18.9
6.1
19.2
6.1
25.3
7.6
16.9
5.4
11.4
3.7
15.5
4.5
34.1
11.1
20.8
6.4
20.7
6.6
6.5
1.9
34.1
11.1
2.6
0.5
15.1
4.4
8.3
2.7
Foliage
(t/ha)
22.1
11.3
22.4
17.6
18.8
33.3
10.9
9.8
17.4
11.0
9.6
10.9
29.9
15.5
15.9
12.1
34.2
1.6
11.5
7.4
Harvest
Index
(0-1)
0.59
0.63
0.60
0.61
0.63
0.49
0.64
0.66
0.58
0.60
0.54
0.58
0.53
0.57
0.57
0.34
0.69
0.34
0.57
0.08
Dry
matter
(%)
35.6
34.3
30.0
30.4
30.3
32.0
32.4
31.8
30.1
31.9
32.1
28.8
32.7
30.8
31.9
29.8
35.6
17.4
28.6
3.6
Plant
type
(1-5)
2.0
1.3
2.3
2.3
3.3
3.0
2.7
3.0
3.0
2.7
1.3
1.7
3.0
3.0
3.3
3.7
5.0
1.3
3.2
0.7
Selection
index
44.2
34.1
33.8
28.2
26.4
23.3
21.5
19.1
17.6
15.3
15.1
12.8
31.1
13.4
13.4
-25.0
44.2
-40.9
0.0
18.7
The average fresh root production of the best 12 experimental clones in Experiment 1
was 23.5 t/ha (compared with 15.1 t/ha for the entire trial) and their average dry
matter content was 31.6 % (against 28.6% for the 64 entires). The average harvest
index for the selected clones was 0.60 and for plant type score 2.39.
Similar results were observed in Experiment 2 (Table 2.29). Mean fresh yield
productivity was 24.6 t/ha (against 15.9 t/ha for the 64 clones) with average dry
matter content of 31.6 % (versus 28.7 % in the entire trial). The selected clones had an
average harvest index of 0.57 and plant type score of 2.83 (compared respectively with
0.56 and 3.4 for all the entries in the experiment)
Finally the comparisons between averages for the selected clones and the entire
Experiment 3 (Table 2.30) were 23.6 versus 16.0 t/ha of fresh root yield, 31.1 versus
28.3 % of fry matter content, 0.63 in contrast with 0.56 for harvest index and 2.68
against 3.4 for plant type score.
Project IP3: improving cassava for the developing world
Output 2-40
Rank
Table 2.29. Results of the best twelve clones and the four checks in Experiment 2 of
the Preliminary Yield Trials conducted at CORPOICA – La Libertad (Villavicencio,
Meta Departament). Statistical parameters of the 64 entries evaluated are also
provided.
CLON
1 GM 277-34
3 GM 275-32
4 GM 226-69
5 GM 221-65
6 GM 256-37
7 GM 261-39
8 GM 261-52
9 GM 275-44
10 GM 256-33
11 GM 256-68
13 CM 9460-76
14 GM 298-39
2
Brasilera
12 CM 6740-7
25 CM 6438-14
40
CM 523-7
Maximum
Minimum
Mean
Std.Deviation
Roots (t/ha)
Fresh
Dry
matter
27.6
9.2
26.3
8.4
32.6
10.6
23.0
8.1
20.8
6.5
29.4
8.8
23.0
6.7
22.1
7.3
21.4
6.9
21.9
6.8
28.7
8.0
18.4
5.7
36.0
11.7
20.6
6.4
14.6
4.6
8.7
2.6
36.0
11.7
8.7
2.0
15.9
4.7
6.7
2.3
Foliage
(t/ha)
17.1
22.4
27.3
17.7
12.0
23.7
10.3
22.9
17.1
21.5
23.9
14.3
31.8
17.1
16.7
5.2
31.8
5.2
12.9
6.6
Harvest
Index
(0-1)
0.62
0.54
0.61
0.56
0.64
0.56
0.69
0.49
0.55
0.52
0.53
0.56
0.53
0.55
0.47
0.63
0.69
0.44
0.56
0.06
Dry
matter
(%)
33.4
31.8
32.6
35.3
31.5
29.9
29.2
33.2
32.3
30.9
28.0
31.1
32.4
31.3
31.9
29.6
35.3
22.2
28.7
3.1
Plant
type
(1-5)
2.5
1.5
4.0
2.7
2.5
3.0
3.5
2.5
3.0
2.5
3.3
3.0
3.5
3.0
2.5
4.0
5.0
1.5
3.4
0.7
Selection
index
41.1
35.4
34.0
32.9
27.2
25.6
22.0
21.3
19.0
16.9
14.3
11.9
35.5
14.9
4.9
-7.6
41.1
-34.1
0.0
17.3
The clones selected at the Preliminary Yield Trial phase (Experiments 1, 2 and 3) will
be combined in one Advanced Yield Trial (see Figure 2.1), also with three replications
and at one location, but with larger plots (25 plants/plot).
One additional analysis that can be performed thanks to the modifications in the
evaluation scheme depicted in Figure 2.1, is the comparison of selections within
families at different stages of the selection process. Table 2.32 presents the parental
clones whose progenies were evaluated in the Clonal Evaluation Trial harvested in May
2002. Total number of progenies evaluated and selected is presented for each
progenitor both at that Clonal Evaluation Trial and at the Preliminary Yield Trial stages.
For clones SM 1219-9, SM 1565-15, S M 2219-11, CM 6740-7 and MPER 183 the proportion
of selected clones in 2002 and 2003 remained approximately constant (Table 2.32). For SM
2058-2 and MTAI 8 the proportion of selected clones in 2003 was considerably higher than in
2002. On the other hand the proportion of selected clones for progenies from CM 7033-3, CM
4574-7 and HMC 1 decreased substantially in 2003 compared with 2002. These results can be
due to differences in the environmental conditions in 2002 and 2003, or just the result of more
precise screening in the replicated trial (2003). However, it can also be due to genetic
characteristics inherited from the progenitors. It is well known that the performance of certain
clones decline over a few years. Perhaps clones from CM 7033-3 are showing this tendency, for
example. It is too early to reach a conclusion in this regard, but this kind of observations are
useful and need to be followed up.
Output 2-41
2003 Annual Report
Rank
Table 2.30. Results of the best twelve clones and the four checks in Experiment 3 of
the Preliminary Yield Trials conducted at CORPOICA – La Libertad (Villavicencio,
Meta Departament). Statistical parameters of the 64 entries evaluated are also
provided.
CLON
1 GM 229-57
2 GM 261-37
3 GM 305-51
4 GM 305-59
6 GM 275-39
7 GM 256-42
8 GM 276-46
9 GM 220-54
12 GM 241-34
13 GM 241-39
14 GM 233-59
15 GM 233-68
5 Brasilera
10 CM 6438-14
11 CM 6740-7
43 CM 523-7
Maximum
Minimum
Mean
Std.Deviation
Roots (t/ha)
Fresh
Dry
matter
30.5
9.8
34.6
9.7
25.4
8.5
29.0
9.2
23.6
7.8
18.2
6.0
24.3
7.0
12.8
4.3
25.3
7.9
23.4
7.4
12.0
3.8
24.1
6.1
30.3
9.9
20.2
6.4
26.7
8.2
11.7
3.3
34.6
9.9
8.2
1.9
16.0
4.6
6.6
2.2
Foliage
(t/ha)
17.8
15.4
10.9
14.0
17.7
10.4
17.4
6.9
19.1
18.5
5.1
12.8
25.3
12.3
18.3
10.7
25.3
5.1
11.7
5.0
Harvest
Index
(0-1)
0.64
0.69
0.69
0.68
0.57
0.64
0.58
0.65
0.57
0.55
0.71
0.65
0.60
0.63
0.59
0.52
0.71
0.43
0.58
0.06
Dry
matter
(%)
32.1
28.0
33.6
31.6
33.2
33.0
28.8
33.6
31.1
31.6
31.9
25.3
32.6
31.6
30.7
28.4
33.6
20.3
28.3
3.4
Plant
type
(1-5)
2.3
2.7
3.7
4.0
2.0
2.0
2.0
2.0
3.5
3.0
2.5
2.5
4.0
3.0
4.0
4.0
5.0
1.5
3.6
0.9
Selection
index
42.6
42.4
36.1
33.7
30.6
28.1
22.5
22.2
20.2
19.8
19.2
17.5
31.6
21.7
20.8
-13.3
42.6
-32.7
0.0
20.0
Table 2.31. Results of the selections in the Preliminary Yield Trials for the acid soil
savannas. The numer of clones representing each family (across the three
experiments) and the numer and proportion of selected clones is provided.
Family
GM 275
GM 235
GM 241
GM 261
GM 256
CM 9901
GM 277
GM 301
GM 305
GM 298
GM 233
GM 226
GM 276
GM 223
GM 227
GM 220
Clones evaluated/selected
Family
Clones evaluated/selected
Evaluated Selected % Select.
Evaluated Selected % Select.
3
3
100.0
GM 229
8
1
12.5
1
1
100.0
CM 9460
11
1
9.1
4
3
75.0
GM 221
25
2
8.0
5
3
60.0
GM 219
22
1
4.6
9
5
55.6
GM 303
8
0
0.0
2
1
50.0
CM 9642
4
0
0.0
2
1
50.0
GM 224
3
0
0.0
2
1
50.0
GM 240
3
0
0.0
5
2
40.0
GM 243
2
0
0.0
3
1
33.3
GM 225
1
0
0.0
12
3
25.0
GM 232
1
0
0.0
4
1
25.0
GM 263
1
0
0.0
4
1
25.0
GM 264
1
0
0.0
9
2
22.2
GM 266
1
0
0.0
7
1
14.3
GM 279
1
0
0.0
16
2
12.5
TOTAL
180
36
20.0
Project IP3: improving cassava for the developing world
Output 2-42
Table 2.32. Results of the selections in the Preliminary Yield Trials for the acid soil
savannas. The numer of clones representing each family (across the three
experiments) and the numer and proportion of selected clones is provided.
Clonal Evaluation Trial - 2002
Evaluated
Selected
E2002 /
(E2002)
(S2002)
S2002
SM 2058-2
148
21
14.2
MTAI 8
209
31
14.8
SM 1219-9
270
71
26.3
SM 1565-15
225
85
37.8
SM 2219-11
300
86
28.7
CM 6740-7
300
68
22.7
CM 7033-3
162
56
34.6
CM 4574-7
293
141
48.1
HMC 1
183
37
20.2
MPER 183
198
10
5.1
Progenitor
Preliminary Yield Trial - 2003
Evaluated Selected
E2003 /
(E2003)
(S2003)
S2003
15
8
53.3
13
5
38.5
38
11
28.9
53
15
28.3
44
9
20.5
48
9
18.8
40
6
15.0
91
9
9.9
13
0
0.0
5
0
0.0
E2002 /
S2003
5.4
2.4
4.1
6.7
3.0
3.0
3.7
3.1
0.0
0.0
A group of 64 clones were evaluated for their root dry matter content at 7, 8, 9, and 10
months after planting. This is an important economic trait that shows a relatively high
heritability based in different studies. However, it shows sharp changes depending on
the age of the plant and the environmental conditions at the time of harvesting. As it
has been demonstrated from data in the sub-humid conditions (CIAT Annual Report
2001) a drastic change in dry matter content occurs when the rains arrive after a long
dry spell. Restults of this study are presented in Table 2.33.
Environmental variation for dry matter content and age of the crop can be analyzed
from the means presented at the bottom in Table 2.33. The highest dry matter content
was observed at seven months after planting with a sharp decrease at eight and nine
months to then increase at the normal age of harvest (ten months). The largest
standard deviation value was observed at 9 months of age and this particular
sampling age was the one showing the lowest correlation coefficients with other ages
(Table 2.33). Otherwise correlation coefficients were relatively high, around the value
0.60.
The genetic variation among the 64 clones evaluated is illustrated by the average
values for the best 12 clones as well as for the four checks. Two clones had an average
dry matter content above 38% with the checks averaging about 30%. In addition to a
high mean dry matter content, the second best clone (SM 2727-12) showed a uniform
performance across the different sampling ages. The fourth and fifth clones (SM 27394 and SM 2786-7) however, showed a much more uniform dry matter content which is
worth further analysis. The industry and market need a stable quality for cassava roots
and the capacity to maintain a stable dry matter content is a major trait affecting root
quality in cassava. The four checks used in the study had an erratic dry matter content
with average or larger than average variation among the sampling dates employed in the
study.
Output 2-43
2003 Annual Report
Table 2.33. Results of the best twelve clones and the four checks in an evaluation of
dry matter content at different ages of the crop. A total of 64 clones were
sampled, including four standard checks. Trials conducted at CORPOICA – La
Libertad – Loma Plot (Villavicencio, Meta Departament). Correlation coefficient
between different ages are also provided.
Rank
Clon
Dry matter content (%)
7 months
8 months
9 months 10 months
1
SM 2452-13
43.4
38.2
36.2
36.3
2
SM 2727-12
40.4
37.2
37.3
38.1
3
CM 9460-15
39.4
36.6
34.2
36.0
4
SM 2739-4
36.6
37.0
36.7
35.4
5
SM 2786-7
37.4
35.6
35.7
36.0
6
SM 2792-32
38.0
36.5
34.9
35.2
7
SM 2730-1
36.6
34.0
36.4
35.3
8
SM 2727-31
37.9
34.0
33.7
36.5
9
CM 9461-1
36.5
34.3
37.9
33.4
10
CM 9463-15
36.7
34.8
34.2
35.9
11
CM 9460-41
36.5
33.7
34.3
36.4
12
CM 9464-30
36.1
36.6
32.1
35.5
30
CM 6740-7
34.4
33.8
31.5
35.9
36
CM 6438-14
33.3
31.0
37.4
32.9
37
Brasilera
36.2
33.5
29.8
33.7
63
CM 523-7
28.9
31.9
27.4
29.3
Maximum
43.4
38.2
37.9
38.1
Minimum
28.9
25.2
22.9
27.9
Mean
35.1
32.8
32.3
33.7
Std.Deviation
2.5
2.9
3.0
1.9
Phenotypic correlation coefficients among sampling dates
7 months
0.56
0.35
0.68
8 months
0.45
0.57
9 months
0.40
Average
38.5
38.2
36.5
36.4
36.2
36.2
35.6
35.5
35.5
35.4
35.2
35.1
33.9
33.7
33.3
29.4
38.5
29.1
33.5
2.0
Following with the new selection scheme (Figure 2.1) the following step in the selection
process after the Preliminary Yield Trials is the Advanced Yield Trials. Table 2.34
presents the results of such a trial for the acid soil savannas based on a total of 64
genotypes (60 experimental clones and four checks). The experimental clones were
first evaluated and selected in the harvest of at the Clonal Evaluation Trial on May
2001 and then at the Preliminary Yield Trial harvested in May 2002.
CM 6740-7 (CORPOICA-Reina) was among the best four checks in the study with a
positive selection index (11.47), which indicates its general avove average performance.
However, based on the selection index Reina ranked 15th in the experiment. Four of
the families involved among the best 15 experimental clones was represented by more
than one clone: CM 9459 (4th and 6th), SM 2792 (1st and 13th), SM 2636 (7th and 16th)
and CM 9460 (8th, 9th and 10th). These would suggest that, at least for these families,
there is indeed a genetic superiority behind the outstanding performance of their
respective clones. On average, dry matter yield across the entire experiment was about
Project IP3: improving cassava for the developing world
Output 2-44
the target of 10 t/ha. At this level of productivity and with current production costs,
cassava can compete with imported maize as soure of energy in animal diets. However,
at the individual clon level many yielded more than 11 t/ha, including CORPOICAReina (CM 6740-7) with an average yield of 11.4 t/ha of dry matter.
Rank
Table 2.34. Results of the best 15 clones and the four checks in the Advanced Yield
Trial. A total of 64 clones were evaluated, including the four checks. Trial
conducted at CORPOICA – La Libertad – Loma plot (Villavicencio, Meta
Departament).
Clon
SM 2792-31
SM 2632-4
CM 9464-30
CM 9459-2
SM 2730-1
CM 9459-13
SM 2636-42
CM 9460-15
CM 9460-41
CM 9460-12
SM 2791-16
SM 2786-10
SM 2792-32
CM 9461-5
SM 2636-6
CM 6740-7
CM 6438-14
Brasilera
CM 523-7
Maximum
Minimum
Mean
Std.Deviation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
16
15
27
49
64
Root yield
(t/ha)
Fresh
Dry
37.9
13.4
36.6
12.5
36.3
12.8
35.7
11.4
31.5
11.1
37.4
11.6
39.5
13.5
30.7
11.0
31.1
11.4
33.9
11.1
28.7
10.5
27.6
10.0
31.3
11.1
33.6
11.2
29.3
9.4
33.9
11.4
27.4
9.8
28.1
9.3
11.5
3.5
39.5
13.5
11.5
3.5
29.0
9.8
4.8
1.7
Fresh
Foliage
(t/ha)
30.9
27.3
27.8
18.8
22.6
18.6
36.6
27.0
27.8
27.4
23.8
23.2
28.5
28.5
17.3
25.0
28.4
21.0
13.9
38.2
13.9
25.0
4.9
Dry
Harvest
matter
Index
(%)
(0-1)
35.4
0.55
34.3
0.57
35.5
0.57
32.0
0.65
35.3
0.58
31.1
0.67
34.0
0.52
36.0
0.53
36.4
0.53
32.8
0.55
36.8
0.54
36.2
0.54
35.2
0.52
33.5
0.53
32.3
0.62
33.7
0.58
35.9
0.49
32.9
0.58
29.3
0.44
38.1
0.67
27.9
0.44
33.7
0.54
1.9
0.04
Plant
type
(1-5)
1.7
1.3
3.0
2.7
2.3
3.0
3.0
2.0
2.3
1.7
2.3
2.0
2.3
2.3
2.0
3.3
2.3
4.0
4.0
4.3
1.3
2.8
0.7
HCN Selection
Index
(1-9)
7
5
8
7
8
9
8
7
7
7
4
5
5
7
4
5
3
7
5
9
3
6.03
1.69
35.67
33.70
25.18
22.28
21.11
20.91
19.52
18.88
18.82
17.16
16.94
14.71
13.11
11.53
10.96
11.47
4.57
-8.71
-74.99
13.45
3.51
9.77
1.69
The same Advanced Yield Trial was also conducted at CORPOICA-La Libertad Lote
Porcinos (Table 2.35). Six experimental clones (CM 9459-13, CM 9460-15, SM 26366, SM 2636-42, SM 2791-16, and SM 2792-32) were among the fifteen best clones at
both locations. CORPOICA-Reina (CM 6740-7) also showed an outstanding
performance at both sites. Table 2.36 presents the results of the combined analysis
including the mean performances across the two sites for the Advanced Yield Trial.
Four families had more than one of its clones among the best 15 experimental
genotypes evaluated in this trial (CM 9459, CM 9460, SM 2636, and SM 2792) further
Output 2-45
2003 Annual Report
confirming not only their genetic superiority but also their stability across different
environments. Table 2.35 also presents the phenotypic correlation coefficients for the
measurements of each variable at both locations. As expected, high heritability traits
such as dry matter content, harvest index and HCN had high correlation coefficients (ρ
= 0.53, 0.46, and 0.42 respectively). Fresh root yield had a much lower correlation (ρ =
0.21). Dry matter yield and plant type scores at both locations were not associatied (ρ
= 0.06 for both traits). The low correlation for plant type may be due to the fact that
disease development is different in both environments leading to an uncorrelated score
for this trait. Selection indices at both locations also showed little association (ρ =
0.07). It is important to stress that no negative correlation was found, suggesting that
germplasm performing well at both locations can still be found. A suggestion
supported by the about 30% of best clones found in common for the two trials.
Rank
Table 2.35. Results of the best 15 clones and the four checks in the Advanced Yield
Trial. A total of 64 clones were evaluated, including the four checks. Trial
conducted at CORPOICA – La Libertad – Lote Porcinos (Villavicencio, Meta
Departament).
Clon
SM 2640-6
CM 9460-9
SM 2636-6
SM 2792-32
SM 2452-13
CM 9462-17
SM 2786-1
SM 2792-43
SM 2791-16
SM 2636-42
SM 2787-4
SM 2638-13
CM 9460-15
CM 9464-29
CM 9459-13
Brasilera
CM 6740-7
CM 523-7
CM 6438-14
1
2
4
6
7
8
9
10
11
12
13
14
15
16
17
3
5
52
58
Maximum
Minimum
Mean
Std.Deviation
Root yield
(t/ha)
Fresh Dry
32.2
11.9
44.3
13.4
30.4
10.2
23.7
8.3
27.4
9.7
30.1
9.9
26.0
9.1
30.7
10.2
25.0
8.3
24.7
8.4
28.9
9.4
22.6
7.6
20.4
7.2
30.5
10.1
30.3
9.0
35.2
12.2
30.7
10.2
11.6
3.9
12.6
4.1
44.3
13.4
7.8
2.6
21.3
7.0
7.3
2.4
Fresh
Foliage
(t/ha)
28.9
22.6
25.8
20.3
26.3
25.4
23.2
30.9
24.2
20.7
29.6
22.5
21.0
31.5
15.9
27.2
29.9
13.9
14.1
31.5
5.8
19.9
6.5
Project IP3: improving cassava for the developing world
Dry Harvest
matter Index
(%)
(0-1)
36.8
0.53
30.3
0.66
33.5
0.54
35.1
0.54
35.4
0.52
32.8
0.55
34.8
0.53
33.1
0.52
33.1
0.51
34.1
0.55
32.5
0.51
33.7
0.50
35.1
0.49
33.2
0.49
29.7
0.65
34.6
0.55
33.4
0.51
33.5
0.48
32.7
0.49
37.6
0.66
28.4
0.32
32.8
0.52
1.8
0.06
Plant
type
(1-5)
2.0
2.0
1.7
2.0
2.7
2.0
2.3
2.7
2.0
3.0
2.3
2.0
2.3
3.0
2.7
3.0
1.7
3.3
3.7
4.3
1.7
2.9
0.6
HCN Selection
(1-9)
Index
4
7
3
8
4
9
8
4
5
8
3
8
7
8
9
3
3
9
5
9
3
6.70
1.96
40.43
38.69
26.75
22.37
21.69
21.33
20.89
15.94
12.78
12.44
12.40
11.70
11.68
11.30
10.63
29.02
24.59
-16.92
-20.63
40.43
-31.55
0.00
16.77
Output 2-46
Rank
Table 2.36. Results combined across two locations of the best 15 clones and the four
checks in the Advanced Yield Trial. A total of 64 clones were evaluated,
including the four checks. Trial conducted at CORPOICA – La Libertad
combined across the two soil conditions of plots Loma and Porcinos
(Villavicencio, Meta Departament).
Clon
1 SM 2792-31
2 SM 2632-4
3 SM 2636- 6
5 SM 2792-32
6 SM 2640-6
7 SM 2636-42
8 CM 9459-13
9 CM 9460-15
10 SM 2791-16
11 SM 2786-1
13 CM 9460-12
14 CM 9464-29
15 SM 2727-12
16 SM 2638-13
17 CM 9459-2
4 CM 6740-7
12 Brasilera
48 CM 6438-14
64 CM 523-7
Maximum
Minimum
Mean
Std.Deviation
Correlations
Root yield
(t/ha)
Fresh Dry
30.2 10.5
30.9 10.4
29.9
9.8
27.5
9.7
29.2 10.3
32.1 10.9
33.9 10.3
25.5
9.1
26.9
9.4
28.4
9.8
29.3
9.4
32.0 10.8
20.0
7.4
28.0
9.2
28.5
9.0
32.3 10.8
31.7 10.7
20.0
7.0
11.5
3.7
36.3 10.9
11.5
3.7
25.1
8.4
4.8
1.5
0.21 0.06
Fresh
foliage
(t/ha)
25.9
24.9
21.5
24.4
26.1
28.7
17.3
24.0
24.0
26.5
25.2
32.5
14.6
26.3
14.9
27.4
24.1
21.2
13.9
33.8
12.8
22.4
4.7
0.34
Dry
Harvest Plant
matter
Index
type
(1-5)
(%)
(0-1)
34.6
0.53
2.2
33.4
0.55
1.8
32.9
0.58
1.8
35.2
0.53
2.2
35.3
0.53
2.3
34.1
0.53
3.0
30.4
0.66
2.8
35.6
0.51
2.2
34.9
0.53
2.2
34.7
0.52
2.7
32.0
0.53
2.0
33.7
0.50
2.8
36.8
0.57
3.0
32.9
0.51
2.0
31.5
0.65
3.2
33.5
0.55
2.5
33.8
0.56
3.5
34.3
0.49
3.0
31.4
0.46
3.7
36.8
0.66
4.2
29.1
0.40
1.8
33.2
0.53
2.9
1.6
0.05
0.5
0.53
0.46 0.06
Average
Ranking Selection
Loc1 Loc2 Index
7.0
1
25
21.93
6.5
2
20
21.59
3.5
16
4
18.85
6.5
13
6
17.74
4.0
44
1
17.64
8.0
7
12
15.98
9.0
6
17
15.77
7.0
8
15
15.28
4.5
11
11
14.86
6.5
31
9
11.98
7.0
10
31
9.73
8.0
21
16
9.62
4.0
17
23
9.59
6.5
22
14
9.55
8.0
4
38
8.79
4.0
15
5
18.03
5.0
49
3
10.15
4.0
27
58
-8.03
7.0
64
52
-45.95
9.0
64
64
21.93
3.5
1
1
-45.95
6.4
-.-.0.00
1.5
-.-.12.68
0.42 -.-.0.07
HCN
(1-9)
Several clones who have had an outstanding performance in trials conducted
previously were evaluated in a different Advanced Yield Trial. These lines, for example,
may have not been able to produce enough planting material at a certain stage in
previous years and that is the reason why they were not included in the main selection
process cultminating the Trials described in Tables 2.34 to 2.36. Several clones that,
for one reason or another, were considered good enough to have a second opportunity
were planted in and additional Advanced Yield Trials whose results are presented in
Tables 2.37 and 2.38 for the two sites where they were evaluated. There were 51
experimental lines in this second Advanced Yield Trial and five checks for a total of 56
entries (Table 2.37). The best clones from the second Advanced Yield Trial will be
combined with those from the first one for a larger that usual Regional Trial (see Figure
2.1), which was planted in June 2003. It is relatively common that some genotypes fail
to produce enough planting material through the selection process. Under certain
circumstances they are given a second opportunity, although it is recognized that the
capacity to produce enough planting material year after year is a criteria taken into
consideration by farmers, a fact incorporated into the main breeding scheme from
farmers’ participatory research.
Output 2-47
2003 Annual Report
Rank
Table 2.37. Results of the best 15 clones and the five checks in an evaluation of
several clones that have shown potential over the last few years. A total of 56
clones were evaluated, including the five checks checks. Trial conducted at
CORPOICA – La Libertad – Loma Plot (Villavicencio, Meta Departament).
Clon
1 SM 1694-2
2 SM 2454-6
3 SM 2219-11
4 SM 2456-3
5 SM 1871-32
6 SM 2425-3
7 SM 1960-1
8 SM 1812-69
9 SM 1794-18
10 SM 1812- 29
11 SM 1674-1
12 SM 2371-1
14 SM 1363-11
16 SM 2452-6
17 SM 1353-3
13 CM 2177-2
15 CM 6438-14
23 CM 6740-7
37 Brasilera
54 CM 523-7
Maximum
Minimum
Mean
Std.Deviation
Root yield
(t/ha)
Fresh
Dry
25.1
8.2
25.0
8.1
22.8
6.5
24.0
7.7
20.5
6.3
29.5
7.4
20.1
6.2
22.7
6.3
23.0
7.5
12.7
4.4
19.6
6.2
22.8
6.8
17.1
5.6
17.8
5.4
22.5
7.1
24.7
7.1
19.9
6.2
17.4
5.2
22.1
6.0
7.2
2.1
29.5
8.2
5.3
1.6
17.3
5.1
5.4
1.6
Fresh
foliage
(t/ha)
26.4
24.5
11.5
21.5
18.1
20.5
19.6
14.5
25.0
13.4
16.4
22.0
14.9
17.0
24.1
15.3
21.1
14.4
21.5
15.5
26.4
2.1
15.2
5.6
Harvest
Index
(0-1)
0.48
0.51
0.67
0.53
0.53
0.59
0.51
0.61
0.48
0.49
0.55
0.51
0.53
0.51
0.48
0.62
0.49
0.54
0.51
0.31
0.85
0.31
0.54
0.08
Dry
Plant
matter type
(%)
(1-5)
32.5
1.7
32.5
2.0
28.4
1.0
32.0
2.0
30.6
1.0
25.2
1.3
31.0
1.3
27.7
1.3
32.8
2.7
34.2
1.0
31.8
2.3
29.8
2.0
32.5
2.3
30.6
1.7
31.5
3.0
28.9
3.0
30.9
2.0
29.8
2.0
27.2
3.0
28.6
2.0
34.2
3.7
24.4
1.0
29.5
2.4
2.3
0.6
HCN
(1-9)
6
7
7
8
5
9
6
8
6
9
5
8
8
7
8
5
6
6
7
7
9
4
6.7
1.2
Selection
Index
27.1
26.2
26.1
24.0
20.5
19.4
17.2
16.9
16.6
16.1
14.0
12.8
10.6
9.0
8.7
12.2
9.8
4.9
-5.8
-33.4
27.1
-34.7
0.0
15.0
More than 50% of the best 15 clones at each location involved the same genotypes (SM
2456-3, SM 2454-6, SM 2452-6, SM 1871-32, SM 1812-29, SM 1694-2, SM 1674- 1,
SM 1363-11). Indicating not only the genetic superiority of these materials but also
suggesting a good stability, at least considering the two environments employed in this
evaluation. Table 2.39 also presents the phenotypic correlations for the different traits
evaluated among the two locations. HCN, as expected showed an excellent correlation
(ρ = 0.97) and considerably better than that found in the previous Advanced Yield Trial
(Table 2.36). A similar situation was observed for plant type score which was also
higher here than in the previous Advanced Yield Trial (ρ = 0.49 versus 0.06). Dry
matter content and harvest index showed relatively high positive correlations similar
to those described in Table 2.35 (ρ = 0.45 and 0.31, respectively). Correlations for
Project IP3: improving cassava for the developing world
Output 2-48
fresh and dry matter yield among the data from Porcinos and Loma plots (ρ = 0.20 and
0.10, respectively) were almost identical to those found before. A remarkable difference
in this second Advanced Yield Trial was the correlation for Selection Index between the
two locations, wich was considerably higher (ρ = 0.27 versus 0.07 in Table 2.26).
Rank
Table 2.38. Results of the best 15 clones and the five checks in an evaluation of
several clones that have shown potential over the last few years. A total of 56
clones were evaluated, including the five checks checks. Trial conducted at
CORPOICA – La Libertad – Porcinos plot (Villavicencio, Meta Departament).
Clon
SM 1697-1
SM 1363-11
SM 1812-29
SM 1694-2
SM 1674- 1
SM 1871-32
SM 2452-6
SM 2454-6
SM 2375-13
SM 2456-3
SM 1773-2
SM 667-1
CM 8748-2
SM 1862-25
SM 1241-12
CM 6740-7
Brasilera
CM 523-7
CM 6438-14
CM 2177-2
Maximum
Minimum
Mean
Std.Deviation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
33
41
42
44
50
Root yield
(t/ha)
Fresh Dry
46.3
15.5
28.2
10.5
28.4
10.3
33.7
11.6
35.2
12.4
31.3
10.4
27.4
10.2
35.9
11.6
35.3
10.8
34.3
10.9
28.6
10.2
32.4
10.7
35.3
10.9
31.9
11.0
30.0
9.7
23.0
7.4
20.8
6.9
23.7
8.3
24.4
7.9
21.2
6.4
46.3
15.5
9.4
2.9
26.9
8.8
6.4
2.2
Fresh
foliage
(t/ha)
45.0
24.3
33.2
35.1
32.8
25.5
26.4
40.6
35.4
29.6
27.7
32.4
33.6
33.8
18.9
22.7
22.8
31.6
29.1
21.5
45.0
9.5
27.8
7.4
Harvest Dry
Index matter
(0-1)
(%)
0.51
33.5
0.55
37.2
0.46
36.3
0.48
34.5
0.52
35.2
0.55
33.4
0.53
37.2
0.47
32.4
0.50
30.7
0.54
31.9
0.51
35.7
0.50
32.9
0.51
31.0
0.49
34.3
0.61
32.3
0.53
32.1
0.50
33.0
0.44
35.0
0.47
32.4
0.51
30.4
0.63
40.1
0.37
20.6
0.50
32.6
0.06
2.8
Plant
type
(1-5)
1.3
2.0
1.0
1.7
2.7
1.7
2.3
1.7
1.3
2.0
2.0
1.7
2.0
2.3
2.7
2.0
2.0
2.7
2.3
3.0
4.0
1.0
2.3
0.6
HCN
(1-9)
Selection
Index
6
8
9
6
5
5
7
8
6
8
5
6
6
7
9
6
7
7
6
5
9
4
6.7
1.3
41.14
21.83
19.61
19.45
18.92
18.51
16.11
15.83
15.23
15.10
14.50
14.17
11.99
11.38
10.70
-2.94
-5.99
-6.51
-7.75
-19.75
41.14
-58.81
0.00
16.63
Itìs also worth mentioning the excellent performance of these clones in contrast with
the five checks used in the Trial. The best ranking check (based on the average
selection index across the two environments) was CORPOICA Reina, which occupied
33d.
Output 2-49
2003 Annual Report
Table 2.39. Results from two locations combined of the best 15 clones and the five
checks in an evaluation of several clones that have shown potential over the
last few years. A total of 56 clones were evaluated, including the five checks
checks. Trial conducted at CORPOICA – La Libertad, combined across the Loma
and Porcinos polots (Villavicencio, Meta Departament).
Rank
Root yield
(t/ha)
Clon
Fresh Dry
1 SM 1697-1
29.2
9.4
2 SM 1694-2
29.4
9.9
3 SM 1363-11 22.7
8.0
4 SM 1871-32 25.9
8.3
5 SM 1812-29 20.5
7.3
6 SM 2454-6
30.4
9.9
7 SM 1674-1
27.4
9.3
8 SM 2456-3
29.1
9.3
9 SM 2452-6
22.6
7.8
10 SM 2375-13 26.9
8.0
11 SM 1812-69 27.3
8.3
12 SM 2219-11 25.6
7.5
13 SM 667-1
26.9
8.0
14 SM 1960-1
25.1
8.0
15 SM 1241-12 23.3
7.4
33 CM 6740-7
20.2
6.3
36 CM 6438-14 22.2
7.0
39 BRASILERA 21.4
6.4
48 CM 2177-2
23.0
6.8
49 CM 523-7
15.5
5.2
Maximum
32.4
9.9
Minimum
9.2
2.7
Mean
22.1
6.9
Std.Deviation
4.6
1.4
Correlations
0.20
0.10
Fresh
foliage
(t/ha)
23.6
30.7
19.6
21.8
23.3
32.6
24.6
25.6
21.7
23.6
22.5
17.3
23.4
25.5
14.4
18.5
25.1
22.2
18.4
23.6
32.6
8.8
21.5
4.7
0.04
Harvest Dry Plant HCN
Ranking
Index matter type (1-9)
(%) (1-5)
(0-1)
Loc1 Loc2
0.68
30.3 2.2
6
1
32
0.48
33.5 1.7
6
4
1
0.54
34.9 2.2
8
2
14
0.54
32.0 1.3
5
6
5
0.48
35.2 1.0
9
3
10
0.49
32.5 1.8
8
8
2
0.53
33.5 2.5
5
5
11
0.53
31.9 2.0
8
10
4
0.52
33.9 2.0
7
7
16
0.55
29.4 1.7
6
9
22
0.56
29.8 1.7
8
17
8
0.60
29.2 1.2
7
20
3
0.55
29.0 1.8
6
12
28
0.50
31.7 1.5
6
18
7
0.62
31.7 2.7
9
15
18
0.53
31.0 2.0
6
33
24
0.48
31.6 2.2
6
44
15
0.51
30.1 2.5
7
41
38
0.56
29.7 3.0
5
50
13
0.37
31.8 2.3
7
42
55
0.68
35.6 3.7
9
56
56
0.37
23.7 1.0
4
1
1
0.52
31.0 2.3
6.7
-.-.0.05
2.2
0.6 1.27 -.-.0.4
0.31
0.45
9
0.97 -.-.-
Average
Selection
Index
30.33
21.36
19.03
19.01
18.73
18.42
17.69
17.33
14.32
12.73
11.40
11.20
10.41
10.24
10.13
-0.98
-3.35
-5.94
-11.75
-13.24
30.33
-43.59
0.00
13.97
0.27
Table 2.40 present the results of the four Regional Trials that have been harvested and
analyzed for the acid soil savannas. The best clone was CM 6438-18, ranked first in
two of the four locations. The outstanding performance of the best five clones was very
consistent across the four environments, which indicates their good stability. In
addition to tolerance to soil acidity these clones must have tolerance to the prominent
diseases in the region: super-elongation disease and bacterial blight. CORPOICA-Reina
(CM 6740-7) showed an intermediate performance with a 16th overall ranking.
Project IP3: improving cassava for the developing world
Output 2-50
Table 2.40 Results from the Regional Trials conducted at four locations in the acid soil
savannas (Meta Departament). The 27 genotypes are ranked according to their
respctive average selection index across the four environments.
CM 6438-14
CM 2177-2
SM 1363-11
CM 6921-3
SM 1143-18
SM 1353-3
SM 1859-26
CM 6975-14
AM 261-1
CM 7052-3
SM 1864-10
CG 165-7
CM 6055-3
SM 1697-1
MBRA 502
CM 6740-7
SM 1241-12
SM 2219-11
CM 4574-7
CM 523-7
SM 1812-29
SM 1565-15
CM 7073-7
SM 667-1
Brasilera
CM 5306- 8
SM 1810-6
Fresh
root
(t/ha)
34.9
32.9
29.8
30.6
27.2
27.2
30.5
28.4
26.3
28.2
29.3
28.1
24.4
25.5
27.7
29.6
24.9
26.0
22.8
20.7
21.2
22.4
21.1
23.7
23.2
20.6
19.5
Maximum
Minimum
Mean
St. Dev.
34.94
19.48
26.19
3.98
Clon
Dry
Harvest
Plant
matter
index
type
(%)
(0-1)
(t/ha)
38.8
0.53
1.83
36.7
0.58
2.13
40.1
0.57
3.00
39.5
0.56
2.75
40.5
0.57
2.71
38.4
0.60
3.08
36.4
0.58
2.92
38.1
0.53
2.83
37.6
0.57
3.08
36.8
0.59
3.46
37.5
0.54
3.36
35.0
0.63
3.08
36.8
0.58
2.75
37.3
0.55
2.92
38.7
0.52
3.58
35.0
0.57
3.42
36.8
0.58
3.50
36.1
0.56
3.50
37.6
0.53
3.08
39.1
0.52
3.25
37.6
0.56
3.57
37.3
0.48
3.13
37.2
0.57
3.65
34.8
0.55
3.21
36.4
0.52
3.33
35.8
0.45
2.75
35.4
0.52
3.24
Parameters for each trait
40.52
0.63
3.65
34.75
0.45
1.83
37.32
0.55
3.08
1.52
0.04
0.42
Selection
Index
29.5
21.3
19.7
19.1
18.7
9.8
8.3
4.8
3.4
3.1
1.9
1.8
1.1
0.9
0.4
-1.2
-3.5
-6.6
-6.9
-6.9
-10.3
-11.4
-11.7
-13.3
-13.3
-23.5
-24.3
29.52
-24.29
0.40
13.39
RANKINGS
Puerto
Gaitán
La
Puerto
Libertad Turpial López
1
3
7
1
5
1
5
7
6
2
1
11
7
8
4
3
3
9
2
5
8
7
12
12
4
16
9
10
2
24
3
18
15
15
6
16
16
17
11
8
11
6
19
19
17
13
10
15
18
10
15
14
9
19
21
9
12
22
14
4
22
21
16
2
27
5
8
20
13
26
17
13
14
12
27
17
10
27
23
6
21
11
22
25
25
4
13
27
23
14
20
23
20
18
26
21
19
23
18
22
24
20
24
26
26
25
25
24
Dry matter yield at each location
14.90
10.09 16.57 17.16
6.22
4.47
7.85 5.36
9.81
7.08
11.40 11.09
2.29
1.71
2.43 2.59
Table 2.41 presents the correlation coefficients for six relevant characteristics and
between the means for each clone (across three replications used in each location)
among the four environments where the Regional Trials were conducted. In general,
the average correlation coefficients for the six traits considered were relatively high
(around 0.35, with the exception of plant type which was 0.23). The most consistent
associations were found for Harvest Index and Selection Index (ρ=0.38). These results
suggest that the environmental conditions in which the Regional Trials were conducted
allowed for a similar response from the genotypes evaluated. Most important, in only
few cases negative correlations were found (Pto López - La Libertad for dry matter
content, Pto López – Turpial and Turpial – Pto Gaitán for plant type scores).
Output 2-51
2003 Annual Report
Table 2.41. Correlation coefficients for the most relevant characterisitics between the
four locations where the Regional Trials for the acid soil savannas were
conducted. A Total of 27 clones were evaluated and the coefficients are based on
means across replications within each location.
Locations involved in
each correlation
Pto Lopez
Pto Lopez
Pto Lopez
Turpial
Turpial
La Libertad
Fresh
Root
(t/ha)
Turpial
La Libertad
Pto Gaitan
La Libertad
Pto Gaitan
Pto Gaitan
Max
Min
Mean
0.45
0.17
0.50
0.20
0.56
0.25
0.56
0.17
0.35
Dry matter Dry matter
Yield
content
(t/ha)
(%)
0.44
0.09
0.48
0.22
0.61
0.27
0.61
0.09
0.35
0.47
-0.08
0.45
0.48
0.32
0.22
0.48
-0.08
0.31
Harvest
Index
(0-1)
Plant
Type
(1-5)
Selection
Index
0.67
0.47
0.25
0.40
0.38
0.12
0.67
0.12
0.38
-0.07
0.29
0.56
0.31
-0.15
0.44
0.56
-0.15
0.23
0.38
0.08
0.66
0.40
0.55
0.24
0.66
0.08
0.38
2.4.3. Selections for the Mid-altitude valleys
As for other eco-regions, the most relevant experiments conducted in the Valle del
Cauca will be listed first (Table 2.42) followed by results specific to each type of
evaluation.
Table 2.42. Trials conducted in the Department of Valle del Cauca, Colombia, during
2000-2001.
Trial
Site
F1
Clonal
Evaluation
Preliminary
Yield Trial
CIAT-Palmira
Regional
Trials
¶
N° of
N° of reps
genotypes
4514/879
1
Observations
See Table 2.43
Palmira
1761 (7)
1
See Table 2.44 to 2.46
S. Quilichao
210 (10)
3
See Table 2.47
26 (25)
3
See Tables 2.48 and 2.49
20 (25)
3
See Table 2.50
Caloto,
S.Quilichao
Dolores
Montelindo
Values in parentheses refer to the number of plants per plot.
involved in the diallel experiment.
Project IP3: improving cassava for the developing world
§
Genotypes
Output 2-52
As for the other regions already discussed, many improvement activities developed
here also benefited other areas. For the mid-altitude valleys the selection index utilized
is identical to the one used for other environments and was as follows:
Selection Index = [Fresh root yield * 10] + [Dry matter content * 8] - [Plant type * 5] +
+ [Harvest Index * 5]
The vegetative cuttings for the Clonal Evaluation Trial planted in July, 2003 were
harvested in the F1 nursery planted during the previous season. A total of 4514
botanical seeds were germinated from which only 2160 were vigorous enough to be
transplanted. At harvest time 879 clones had enough plant vigor to produce the seven
stakes required for the Clonal Evaluation Trial to be harvested next year. Table 2.43
describes the origin of these 879 clones. A total of 49 families are involved and, as it
can be seen, five of them involve the source of resistance to white flies (MECU 72) as a
progenitor.
The large Clonal EvaluationTrial for the mid-altitude valleys included a total of 1761
clones. As in the other regions the clones from a given family were randomly allocated
to each of three groups within the trial. The purpose of this procedure is to obtain a
replication for each family and to reduce the environmental effect on the selection
process. Table 2.44 summarizes the most important results from the Clonal Evaluation
Trial. The stratification, as was the case in the other regions already described, was
justified since the mean performance in the three “blocks’ (represented by fresh root
yield) were 24.05, 28.08 and 27.51 t/ha respectively. This variation which is mainly
environmental in origin, has been eliminated in the selection process, since selections
are performed within each ‘block’. In general root productivity across the Clonal
Evaluation Trial was excellent (Table 2.44) with an average dry matter production of
about 10 t/ha. However, the average for the 450 selected clones was obviously much
higher (14.65 t/ha).
Table 2.45 presents the mean performance for each family (across the three ‘blocks’ in
which their respective clones had been allocated). There was a large variation among
the families with the best showing 48.9% of its clones selected and the worst family
with only 2.7% selection success. Analyzing the origin of selected clones there is a
slight advantage of controlled crosses (families with the CM or GM codes) over those
coming from the polycrosses (SM code). For the former the % of selected clones was
29.72% (170 selected clones from a total of 572), whereas for the latter the % was
24.14 (287 clones selected from a total of 1189). Since the parents involved in both
types of crosses are the same, the difference may be more logically explained by the
occurrence of self-pollinations in the polycross nurseries. Although the planting is
done to favor outcrossing, the occurrence of self-pollinations is not totally avoided and
the differential response for directed versus open-pollinations may be a result from it.
Output 2-53
2003 Annual Report
Table 2.43. Origin of the 879 clones for the Clonal Evaluation Trial planted in July,
2003 at CIAT – Palmira (Valle del Cauca Department). SM families have unknow
fathers, but it is certain that they come from a elite clones in the policross plots.
Family
CM 9901
CM 9903
CM 9919
CM 9920
CM 9953
GM 228
GM 230
GM 254
GM 260
GM 266
GM 268
GM 269
GM 284
GM 291
GM 292
GM 295
GM 297
GM 306
GM 308
GM 309
GM 314
GM 370
GM 372
GM 373
GM 374
GM 375
GM 473
GM 501
GM 502
GM 503
GM 509
GM 555
GM 556
SM 2802
SM 2859
SM 2860
SM 2982
SM 2983
SM 2985
SM 3085
SM 3087
SM 3090
SM 3091
SM 3092
SM 3094
SM 3096
SM 3097
SM 3098
SM 3099
TOTAL
Mother
SM 1219-9
CM 6740-7
CM 7951-5
CM 7951-5
SM 1741-1
CM 6740-7
CM 6740-7
SM 1219-9
SM 1673-10
MTAI 8
SM 1278-2
SM 1741-1
SM 1741-1
SM 1665-2
SM 1741-1
SM 1741-1
SM 1741-1
MECU 72
MECU 72
MECU 72
MECU 72
CM 2772-3
SM 1460-1
SM 1557-17
CM 2772-3
SM 1689-18
CM 8151-1
SM 1219-9
SM 1210-4
SM 1557-17
SM 1557-17
SM 1660-4
SM 1660-4
SM 1219-9
CM 6740-7
CM 7951-5
CM 2772-3
CM 6740-7
SM 1219-9
CM 2772-3
CM 6740-7
SM 1210-4
SM 1219-9
SM 1460-1
SM 1660-4
SM 1741-1
MECU 72
MPER 183
MTAI 8
Father
CM 6740-7
SM 1741-1
SM 1565-17
SM 1741-1
SM 1219-9
SM 1278-2
SM 1636-24
SM 1278-2
SM 1219-9
SM 1219-9
SM 1673-10
SM 1278-2
SM 1636-24
MTAI 8
SM 1673-10
SM 2219-11
MPER 183
MPER 183
CM 6740-7
SM 1219-9
HMC 1
SM 1210-4
CM 2772-3
CM 2772-3
SM 1660-4
CM 2772-3
CM 8370-11
SM 1210-4
SM 1460-1
SM 1210-4
SM 1219-9
CM 8370-11
SM 1210-4
Project IP3: improving cassava for the developing world
Planted
100
83
56
93
100
50
65
50
72
65
50
75
70
100
100
120
75
70
100
70
125
65
64
73
117
75
115
90
106
107
112
68
55
100
50
80
150
140
125
140
97
109
110
93
125
139
132
103
85
4514
Transplanted
61
53
38
21
64
35
32
34
42
38
45
44
37
65
80
49
53
42
85
27
31
18
30
51
60
38
39
57
31
31
44
19
32
68
14
15
30
14
73
67
27
42
79
37
47
70
35
60
56
2160
Harvested
27
24
6
3
21
18
15
10
4
11
15
9
20
18
16
20
36
26
35
17
16
4
17
26
34
21
13
11
14
23
22
7
12
20
3
12
13
7
21
34
28
11
29
26
14
23
16
29
22
879
Output 2-54
Table 2.44. Results of the selection carried out in the Clonal Evaluation Trial at
Palmira, Department of Valle del Cauca, from 1761 clones evaluated during
September 2002 to June 2003. Clones from the same family were randomly
allocated to one of three blocks within the trial. This blocking allowed each
family to be replicated three times, but individual clones were planted only in
one of these blocks. Therefore, there was no replication for individual clones.
Parameter or
Genotye
Yield (t/ha)
Harvest Index
Fresh roots Dry matter
(0 to 1) ¶
Plant type
(1 to 5) §
Dry matter Selection
content (%)
Index
Results from the 1761 clones evaluated across the three blocks
Maximum
Minimum
Mean
Std. Dev.
66.63
0.22
26.51
12.21
24.70
0.00
9.50
4.63
0.95
0.02
0.58
0.12
5.00
1.00
2.75
1.12
44.87
16.91
35.92
3.27
51.19
-80.24
0.01†
18.92
Averages of the 605, 588 and 568 clones in Blocks 1, 2 and 3, respectively
Block 1
Block 2
Block 3
24.05
28.08
27.51
8.86
10.21
9.76
0.63
0.57
0.54
2.68
2.63
2.97
36.61
36.02
35.09
0.00†
0.00†
0.00†
44.87
29.10
37.83
2.32
51.19
11.16
20.87
7.03
Results from the 450 clones selected
Maximum
Minimum
Mean
Std. Dev.
66.63
18.08
38.96
9.07
24.70
7.41
14.65
3.18
0.95
0.45
0.64
0.06
5
1
2.05
0.90
Best 5 clones selected from each block in the Clonal Evaluation Trial
SM 2858-4
SM 2858-2
GM 295-7
SM 2998-5
SM 2983-13
SM 2804-28
SM 2863-20
SM 2871-23
SM 3047-19
GM 295-18
GM 264-145
GM 297-79
SM 2864-21
GM 295-33
SM 2804-45
¶
§
†
62.50
47.99
52.57
50.78
52.01
53.01
66.29
55.47
46.54
56.58
64.62
58.59
63.17
46.99
58.04
24.51
19.19
19.23
20.83
19.08
21.92
24.37
21.79
19.06
21.39
24.35
23.5
23.36
18.22
22.93
0.65
0.67
0.72
0.67
0.74
0.69
0.68
0.73
0.65
0.67
0.62
0.57
0.72
0.75
0.55
1
1
1
3
2
1
1
2
1
1
1
1
2
1
2
39.21
39.99
36.59
41.02
36.69
41.34
36.76
39.29
40.95
37.8
37.69
40.12
36.98
38.79
39.51
51.19
41.10
39.03
35.67
34.44
45.69
42.87
39.42
38.00
37.93
48.49
47.15
45.78
41.69
40.20
The harvest index is obtained by dividing the production of commercial roots by total biomass
(roots + aerial parts). Preferred harvest indexes are > 0.5.
Plant type integrates under one value, plant architecture, leaves health, and capacity to
produce stakes on a scale where 1 = excellent and 5 = very poor is used.
Average election index within blocks must be zero, because it is based on a combination of
standardized variables. However, when averaged across the three blocks there is a slight
deviation because selection indices were estimated for each block separately.
Output 2-55
2003 Annual Report
Table 2.45 Results from the Clonal Evaluation Trial (CIAT-Palmira, Valle del Cauca), of 1761 genotypes. The averages of
each of the 54 families, number of clones representing them and proportion of selected clones is presented.
Family
%
select.
#
clones
Fresh DM
(t/ha) (t/ha)
GM 295
45
48.9
30.4
10.9
SM 2858
59
47.5
31.6
11.2
SM 2652
17
47.1
24.7
9.4
GM 297
49
46.9
35.0
12.5
CM 9903
43
46.5
28.4
10.8
SM 2983
41
46.3
29.7
11.0
CM 9953
38
44.7
25.6
9.7
SM 2801
14
42.9
27.4
10.6
GM 269
27
37.0
22.1
8.1
SM 2799
20
35.0
22.1
8.3
SM 2804
43
34.9
30.9
11.2
SM 2651
27
33.3
29.5
11.0
SM 2656
3
33.3
29.2
11.8
SM 2802
37
32.4
25.2
9.2
SM 2862
53
32.1
27.6
10.3
SM 2865
58
31.0
26.7
9.8
SM 2863
30
30.0
26.5
9.5
CM 6979
24
29.2
26.8
10.0
GM 234
28
28.6
30.2
10.7
SM 2860
56
28.6
24.1
8.6
SM 2988
43
27.9
26.1
9.1
SM 2913
40
27.5
26.5
9.9
GM 128
33
27.3
24.5
9.4
SM 2861
33
27.3
24.6
9.3
CM 8885
26
26.9
30.2
11.1
SM 2869
43
25.6
29.4
10.7
SM 2870
35
22.9
28.0
10.1
¶ HI = Harvest Index (Root production
%
DM
35.8
35.5
38.1
35.6
37.8
36.9
37.9
38.8
36.8
37.6
36.3
37.3
40.3
36.3
37.4
36.7
36.0
37.2
35.4
35.6
35.1
37.5
38.5
37.7
36.7
36.4
36.2
/ total
Project IP3: improving cassava for the developing world
Plant
Type
(1-5)
HI
(0-1)
¶
SI
§
Family
%
select.
#
clones
1.8 0.61
8.3 GM 228
2.1 0.57
6.1 GM 264
2.5 0.56
3.7 SM 2866
2.1 0.62
11.0 GM 265
2.4 0.66
11.0 SM 3047
2.6 0.60
6.7 GM 260
2.3 0.66
9.7 GM 254
3.0 0.53
4.6 CM 9901
3.0 0.61
-1.0 SM 2864
2.5 0.55
0.7 SM 2992
2.3 0.61
7.9 CM 9642
2.6 0.58
6.6 SM 2803
3.0 0.61
11.2 SM 2871
2.8 0.66
3.0 SM 2657
2.7 0.56
6.8 SM 2658
2.9 0.61
3.0 SM 2868
3.4 0.62
-0.6 SM 3045
3.2 0.58
1.0 SM 2982
2.7 0.63
4.5 CM 8884
2.9 0.60
-2.3 SM 2998
3.1 0.55
-5.3 SM 2654
2.5 0.54
3.8 SM 2661
2.9 0.63
6.3 SM 2867
2.8 0.52
-0.8 SM 2663
2.7 0.59
5.7 GM 270
2.9 0.62
4.9 SM 2665
2.6 0.61
4.1 SM 2999
biomass).§ SI = Selection Index
Fresh
(t/ha)
DM
(t/ha)
%
DM
Plant
Type
(1-5)
HI
(0-1)
¶
SI
§
14
21.4
25.4
9.7 38.3
3.1 0.61
4.1
38
21.1
29.4
10.6 36.3
2.6 0.65
7.0
43
20.9
26.4
8.8 33.5
3.1 0.54
-6.6
48
20.8
25.2
8.7 34.4
2.6 0.58
-4.2
34
20.6
23.7
8.5 36.1
3.6 0.58
-5.7
41
19.5
25.6
9.3 36.1
2.9 0.62
1.1
26
19.2
20.7
7.5 36.2
3.1 0.56
-6.9
21
19.0
25.5
9.5 37.3
2.9 0.62
3.4
21
19.0
25.8
9.8 37.9
3.0 0.55
2.0
21
19.0
25.7
8.9 34.6
3.4 0.56
-7.1
34
17.6
28.8
10.0 34.8
2.5 0.53
-2.5
29
17.2
26.3
9.3 35.5
3.4 0.56
-4.8
35
17.1
23.1
7.6 32.7
3.1 0.55 -14.0
13
15.4
26.6
9.4 35.2
2.8 0.53
-3.9
26
15.4
25.2
8.5 33.8
2.8 0.54
-8.5
26
15.4
22.5
7.5 33.3
2.8 0.55
-7.5
30
13.3
24.8
8.7 34.9
3.5 0.59
-6.7
24
12.5
28.6
9.7 34.0
2.6 0.56
-2.8
17
11.8
25.7
8.8 34.4
2.2 0.58
-2.7
28
10.7
26.7
9.0 33.9
3.5 0.51 -10.5
29
10.3
23.5
8.8 37.4
3.4 0.52
-4.2
40
10.0
20.0
6.4 32.1
2.4 0.48
-9.2
42
9.5
19.7
6.7 34.1
2.4 0.43 -14.8
38
7.9
24.3
8.4 34.6
2.9 0.56
-7.1
20
5.0
20.1
7.3 36.3
3.3 0.68
-2.8
21
4.8
24.0
7.8 32.7
2.0 0.53
-9.1
37
2.7
26.3
8.7 32.9
3.0 0.50 -12.1
(combines several variables of economic relevance)
Output 2-56
Table 2.46. Mean performance of the progenies from parental lines used for the Clonal
Evaluation Trial described in Table 2.45.
Progenitor
SM 2219-11
CM 5655-4
SM 1741-1
SM 643-17
CM 6740-7
SM 653-14
SM 1565-17
CM 7951-5
SM 909-25
CM 523-7
SM 1210-4
SM 1543-16
CM 8602-12
MTAI 8
SM 2075-4
SM 1219-9
MPER 183
MCOL 1505
SM 1278-2
CG 489-4
HMC 1
SM 1673-10
SM 1278-5
SM 1406-1
MBRA 12
SM 1557-17
MCOL 1468
MECU 72
SM 719-6
SM 1565-15
SM 1479-8
CM 2772-3
SM 1677-4
# of
clones
45
59
237
70
181
53
86
83
30
24
38
43
40
33
33
307
168
60
61
43
140
62
26
111
35
26
17
28
29
40
42
45
38
% clones
selected
48.9
47.5
42.2
34.3
33.1
32.1
30.2
30.1
30.0
29.2
28.9
27.9
27.5
27.3
27.3
26.7
23.8
23.3
23.0
20.9
20.0
19.4
19.2
18.0
17.1
15.4
11.8
10.7
10.3
10.0
9.5
8.9
7.9
Yield (t/ha)
% dry
fresh roots dry matter matter
30.4
10.9
35.8
31.6
11.2
35.5
28.3
10.3
36.7
23.8
9.0
37.8
28.0
10.3
36.7
27.6
10.3
37.4
30.1
11.0
36.4
26.8
9.8
36.5
26.5
9.5
36.0
26.8
10.0
37.2
27.5
10.7
39.0
26.1
9.1
35.1
26.5
9.9
37.5
24.5
9.4
38.5
24.5
9.4
38.5
25.5
9.3
36.4
28.8
10.0
34.4
26.9
9.8
36.4
22.5
8.4
37.1
27.9
9.9
35.5
26.2
9.5
36.0
25.7
9.1
35.4
20.7
7.5
36.2
26.1
9.0
34.5
23.1
7.6
32.7
22.5
7.5
33.3
25.7
8.8
34.4
26.7
9.0
33.9
23.5
8.8
37.4
20.0
6.4
32.1
19.7
6.7
34.1
26.3
8.8
33.4
24.3
8.4
34.6
Harvest
Index
0.61
0.57
0.6
0.5
0.6
0.56
0.6
0.6
0.62
0.58
0.6
0.55
0.54
0.63
0.63
0.6
0.6
0.6
0.6
0.6
0.6
0.6
0.56
0.5
0.55
0.55
0.58
0.51
0.52
0.48
0.43
0.5
0.56
Plant
type
1.8
2.1
2.4
2.6
2.7
2.7
2.6
2.7
3.4
3.2
3.0
3.1
2.5
2.9
2.9
2.7
2.5
3.1
3.1
2.4
3.0
3.1
3.1
3.0
3.1
2.8
2.2
3.5
3.4
2.4
2.4
2.3
2.9
Selection
Index
8.3
6.1
7.2
1.2
4.5
6.8
6.4
2.2
-0.6
1.0
5.9
-5.3
3.8
6.3
6.3
2.0
-2.0
0.0
0.1
1.5
0.6
-3.0
-6.9
-5.9
-14.0
-7.5
-2.7
-10.5
-4.2
-9.2
-14.8
-6.0
-7.1
The consolidation of the information from Table 2.45 allows obtaining results of the
performance from the progenies from each parental line, which is presented in Table 2.46.
The best parental line for this year Clonal Evaluation Trial was SM 2219-11 with 48.9% of its
45 clones selected. The progenies from CM 5655-4 and SM 1741-1 were also outstanding,
particularly for the later, which had 237 clones derived from it.
Clones selected in the Clonal Evaluation Trial the previous year were planted in Stantander de
Quilichao (Cauca Department) for further evaluationand selection. This is the equivalent of a
Preliminary Yield Trial and was planted at tha location because of a higher than desirable incidence of
Output 2-57
2003 Annual Report
frog skin disease. A total of 210 clones were included in this trial (see table 2.47) and 64 were
evaluated. These selected clones will be brought back to CIAT-Palmira to be “cleaned” from infectious
diseases through tissue culture and eventually be evaluated in Regional Trials later on.
Table 2.47. Mean performance of the progenies from the Preliminary Yield Trial evaluated in
Santander de Quilichao.
Clone
Mother
GM 254-41
GM 268-40
GM 284-69
GM 228-50
CM 9901-122
GM 271-43
GM 234-107
GM 254-36
GM 264-99
GM 271-33
GM 230-62
GM 309-67
GM 312-34
GM 284-45
GM 235-71
GM 309-61
GM 234-115
GM 312-56
GM 284-70
GM 254-32
SM 1219-9
SM 1278-2
SM 1636-24
CM 6740-7
CM 6740-7
SM 1278-2
CM 6740-7
SM 1219-9
SM 1219-9
SM 1278-2
CM 6740-7
MECU 72
MECU 72
SM 1636-24
CM 6740-7
MECU 72
CM 6740-7
MECU 72
SM 1636-24
SM 1219-9
Maximum
Minimum
Average
Maximum
Minimum
Average
Fresh root Fresh foliage Harvest Dry matter
yield
yield
Index
content
(t/ha)
(t/ha)
(0-1)
(%)
SM 1278-2
73.44
65.6
0.53
35.2
SM 1673-10
60.94
56.3
0.52
38.4
SM 1741-1
40.63
34.4
0.54
44.0
SM 1278-2
55.63
53.1
0.51
39.2
SM 1219-9
65.63
65.6
0.50
36.0
MPER 183
43.75
34.4
0.56
37.6
HMC 1
50.31
43.8
0.53
36.6
SM 1278-2
49.69
46.9
0.51
37.8
HMC 1
52.81
46.9
0.53
34.3
MPER 183
62.50
59.4
0.51
31.8
SM 1636-24
55.00
53.1
0.51
34.8
SM 1219-9
43.75
40.6
0.52
38.5
SM 1673-10
36.88
28.1
0.57
36.4
SM 1741-1
40.00
34.4
0.54
36.3
MTAI 8
50.94
50.0
0.50
34.5
SM 1219-9
44.06
37.5
0.54
33.6
HMC 1
46.25
46.9
0.50
36.4
SM 1673-10
42.81
37.5
0.53
34.5
SM 1741-1
19.69
15.6
0.56
41.6
SM 1278-2
39.69
37.5
0.51
37.2
Statistics of 64 clones selected
73.44
65.63
0.57
44.90
18.75
15.63
0.49
29.69
38.89
35.50
0.52
36.04
Statistics of 210 clones evaluated
73.44
65.63
0.58
44.90
5.31
6.25
0.35
19.71
28.94
26.99
0.52
33.36
Father
Dry
matter
(t/ha)
25.8
23.4
17.9
21.8
23.6
16.4
18.4
18.8
18.1
19.9
19.2
16.9
13.4
14.5
17.6
14.8
16.8
14.8
8.2
14.7
25.82
7.23
13.92
25.82
1.71
9.72
The performance of the best 20 clones from this Preliminary Yield Trial (table 2.47) was
excellent as well as that of the 64 selected clones. The average dry matter yield was 13.92
t/ha in an environment where white flies and frog skin have a strong incidense. Several of
the selected clones were derived from MECU 72 the source of resistance to white flies.
Project IP3: improving cassava for the developing world
Output 2-58
Table 2.48. Results from the Regional Trials planted in two locations (Caloto and S. de
Quilichao) in the Cauca Department. The trail was planted in Dolores (See Table 2.49)
but two clones were missed in that trial.
Clone
SM 1219-9
CM 7951-5
SM 1642-22
SM 1965-1
MBRA 383
CM 8370-11
SM 2141-1
SM 2211-3
CM 6660-21
SM 1855-15
SM 1660-4
HMC 1
SM 2085-7
SM 2052-4
SM 1779-7
MTAI 8
CM 8370-10
SM 2058-2
MPER 183
CM 523-7
SM 1520-16
SM 1520-18
SM 2160-2
SM 2073-1
SM 1871-33
SM 1959-1
CM 7463-2
SM 2198-4
Maximum
Minimum
Average
St.Deviation
Correlations
¶
Fresh
root yield
(t/ha)
34.22
31.27
29.60
31.89
29.62
30.79
26.59
26.11
25.40
26.76
27.78
23.56
27.79
26.23
26.46
21.43
25.66
25.32
24.84
18.08
21.06
21.04
19.64
21.45
23.87
22.97
21.18
18.74
34.22
18.08
25.33
4.15
0.52
Dry
matter
(%)
36.21
36.08
35.00
36.84
35.29
34.18
36.90
36.07
34.40
33.55
32.68
35.55
34.14
31.96
34.61
35.11
34.91
34.10
31.00
37.91
32.48
33.32
33.69
34.36
34.59
31.15
35.47
37.37
37.91
31.00
34.60
1.75
0.54
Harvest
Index
(0-1)
0.73
0.68
0.67
0.57
0.63
0.64
0.56
0.57
0.64
0.64
0.65
0.61
0.58
0.68
0.57
0.64
0.54
0.58
0.68
0.55
0.68
0.65
0.64
0.59
0.51
0.64
0.49
0.45
0.73
0.45
0.61
0.06
0.73
Selection
Index
39.61
26.73
18.20
16.96
13.35
12.69
7.19
3.63
2.47
2.41
2.08
1.42
0.21
-1.31
-3.05
-4.20
-5.27
-5.52
-7.83
-8.50
-9.36
-10.03
-10.69
-11.20
-14.00
-15.67
-19.21
-21.12
39.61
-21.12
0.00
13.92
0.53
Rank
in
Caloto
1
2
4
6
5
10
12
7
13
19
14
3
9
11
17
22
8
15
21
18
26
27
20
23
16
25
28
24
Rank
in
Quilichao
1
5
4
2
6
3
7
12
11
8
10
26
15
17
18
9
24
20
16
25
13
14
22
19
28
23
21
27
Rank
in
Dolores
5
1
24
4
2
10
n.a.
9
15
11
12
25
16
23
7
21
18
17
26
14
19
3
13
20
22
n.a.
8
6
0.46¶
Pearson product moment correlation coefficient, r.
Output 2-59
2003 Annual Report
Table 2.49. Results from the Regional Trials described in Table 2.49 when planted in Dolores
in the Cauca Department. Clones SM 2141-1 and SM 1959-1 were not included in this
trial.
1
CM 7951-5
2
BRA 383
3
SM 1520-18
4
SM 1965-1
5
SM 1219-9
6
SM 2198-4
7
SM 1779-7
8
CM 7463-2
9
SM 2211-3
10
CM 8370-11
11
SM 1855-15
12
SM 1660-4
13
SM 2160-2
14
CM 523-7
15
CM 6660-21
16
SM 2085-7
17
SM 2058-2
18
CM 8370-10
19
SM 1520-16
20
SM 2073-1
21
TAI 8
22
SM 1871-33
23
SM 2052-4
24
SM 1642-22
25
HMC 1
26
PER 183
Maximum
Minimum
Average
St.Deviation
Fresh root yield
(t/ha)
54.7
43.5
32.6
32.9
38.6
38.9
35.5
30.3
32.4
33.9
34.0
31.0
26.9
26.8
28.0
33.1
35.6
27.2
26.2
28.4
21.1
27.0
23.9
25.8
20.7
19.3
54.7
19.33
31.09
7.57
Dry matter
(%)
39.1
38.6
40.0
39.4
35.8
35.6
37.8
38.1
36.1
36.5
35.2
37.0
37.6
38.6
37.1
35.5
33.4
37.1
33.5
35.2
35.7
35.9
35.3
34.3
35.1
31.4
40.0
31.35
36.32
2.02
Harvest Index
(0-1)
0.59
0.53
0.48
0.46
0.41
0.41
0.34
0.36
0.44
0.34
0.42
0.35
0.38
0.31
0.37
0.32
0.39
0.29
0.54
0.37
0.46
0.27
0.33
0.30
0.25
0.27
0.59
0.25
0.38
0.09
Selection
Index
54.02
33.70
21.77
18.68
9.02
8.98
8.88
4.43
3.92
1.96
1.29
0.79
-0.72
-1.05
-1.71
-4.36
-5.02
-7.23
-8.90
-8.97
-11.31
-13.38
-16.88
-19.85
-26.39
-41.67
54.02
-41.67
0.00
18.80
The regional trial for this region was planted in three different locations (S. de Quilichao,
Caloto and Dolores). A total of 28 clones were evaluated, however, two of them could not be
included at the Dolores trial. Because of this the results of this regional trial are presented in
two different tables (2.48 and 2.49). Results from Quilichao and Caloto were very similar with
a Pearson rank correlation coefficient of 0.46 (Table 2.48). Results from these two locations,
however, tended to differ from those observed in Dolores. SM 1219-9, CM 7951-1, SM 19651, MBRA 383, and CM 8370-11 had a good performance across the three environments.
Project IP3: improving cassava for the developing world
Output 2-60
Table 2.50. Results from the Regional Trials planted in Montelindo (Caldas Department). This
is a new testing environment for cassava in the coffee growing area of Colombia.
SM 1219-9
MBRA 383
CM 4843-1
SM 653-14
CM 7951-5
SM 1741-1
MPER 183
SM 1557-17
MTAI 8
CM 523-7
Regional
CM 3306-4
CM 7514-7
SM 1433-4
MVEN 25
HCM 1
CM 6119-5
CM 849-1
CG 1141-1
Manzanita
Maximum
Minimum
Average
St.Deviation
Fresh root
yield
(t/ha)
74.3
66.6
66.9
52.6
53.1
52.2
39.7
45.9
49.1
24.4
48.0
28.9
26.3
32.2
35.2
37.1
35.1
25.9
24.1
18.7
74.3
18.7
41.8
15.85
Dry
matter
(%)
34.3
37.1
36.8
39.8
36.7
38.5
42.6
37.1
35.0
36.6
35.3
39.9
39.2
36.5
36.3
34.9
31.7
33.8
34.4
29.1
42.6
29.1
36.3
3.0
Harvest
Index
(0-1)
0.61
0.50
0.50
0.46
0.68
0.50
0.40
0.40
0.56
0.47
0.56
0.40
0.39
0.32
0.39
0.44
0.44
0.36
0.32
0.29
0.7
0.3
0.45
0.1
Plant type
(1-5)
Thrips
(1-5)
3
3
3
3
3
3
3
3
3
2
4
4
3
3
4
4
4
4
4
3
4.0
2.3
3
0.47
1.0
1.0
1.0
1.7
1.3
1.3
1.7
1.0
2.3
1.7
1.3
1.0
1.0
2.3
2.3
1.0
1.0
1.0
1.0
3.0
3.0
1.0
1.5
0.6
Calidad
Culinaria
(1-5)
5
1.6
5
5
5
2.3
2.3
5
5
3
1.6
1.6
5
2.3
5
3.6
3.6
5
3.6
3
5.0
1.6
3.68
1.36
HCN
(1-9)
5
3
7
3
6
3
3
3
8
5
6
5
3
6
8
4
5
6
7
5
8.0
2.5
5
1.69
In Table 2.50 presents the results from a smaller Regional Trial planted in the coffee growing
Department of Caldas at Montelindo. All the best materials identified in the previous regional
trials (Tables 2.48 and 2.49) that were included in this new trial also had an excellent
performance (SM 1219-9, MBRA 383, and CM 7951-5). It is clear, therefore, the genetic
superiority and stability of performance of these clones which will be further included as
parental lines for new crosses.
2.4.4. Selections for the Highlands
Cassava can be grown up to about 1800 meters above sea level. In some cases these
“highland” regions depend heavily on cassava. This is the case for the production of cassava
in the Cauca Department where its roots are used to produce fermented starch in artisanal
starch processing facilities known as “rallanderos”. Cassava, as it is frequently the case, it is
a fundamental crop from the social point of view for these marginal environments where
Output 2-61
2003 Annual Report
farmers have few options to attain a livelihood. Table 2.51 presents the results of a Clonal
Evalaution Trial harvested this year for this environment, where cassava is typically
harvested at 18-24 months of age.
Table 2.51 Results from a Clonal Evaluation Trial conducted in the highland environment at
Popayán (Cauca Department). The trial involved 237 clones evaluated during an 18
months growing period between 2001-2003.
Clone
Plant
Type
(1-5)
CM 9783-9
SM 2398-39
SM 2718-48
SM 2392-33
SM 2714-33
SM 2392-32
SM 2714-32
SM 2709-51
SM 2564-2
SM 2568-1
CM 9030-18
SM 2568-5
CM 9814-5
CM 9815-12
SM 1843-41
SM 2314-10
SM 2387-56
SM 2709-56
SM 2718-47
CM 9028-36
2
4
2
2
2
2
2
2
2
2
3
4
2
3
2
2
3
2
2
3
Maximum
Minimum
Average
St.Deviation
4.0
2.0
2.4
0.62
Maximum
Minimum
Average
St.Deviation
4.0
2.0
3.0
0.77
Fresh
rotos
(t/ha)
Annual
Dry
Dry
Harvest
Yield
Matter
Matter
Index
(t/ha)
(%)
(t/ha)
(0-1)
Best 20 performing clones
63.8
45.3
38.3
24.4
0.50
71.7
50.9
36.9
26.4
0.44
55.2
39.2
35.7
19.7
0.46
46.5
33.0
36.8
17.1
0.50
49.4
35.1
34.9
17.2
0.55
39.0
27.7
38.6
15.0
0.50
42.5
30.2
37.2
15.8
0.50
22.5
16.0
42.2
9.5
0.62
34.6
24.6
39.0
13.5
0.51
41.0
29.1
36.3
14.9
0.52
40.2
28.6
37.7
15.2
0.65
53.8
38.2
35.5
19.1
0.57
40.0
28.4
35.6
14.2
0.53
43.8
31.1
36.0
15.8
0.61
42.7
30.3
34.5
14.7
0.51
29.6
21.0
38.7
11.4
0.50
42.3
30.0
37.1
15.7
0.41
31.9
22.6
37.5
12.0
0.47
22.5
16.0
40.7
9.2
0.48
44.2
31.4
35.3
15.6
0.51
Statistics of the 60 clones selected
71.7
50.9
42.2
26.4
0.7
19.0
13.5
30.9
7.6
0.3
36.1
25.6
36.1
12.9
0.5
10.73
7.62
2.68
3.71
0.07
Statistics of the 237 clones evaluated
71.7
50.9
42.2
26.4
0.7
1.3
0.9
16.3
0.2
0.1
23.8
16.9
34.0
8.2
0.4
11.20
7.95
3.16
4.03
0.11
CBB
Score
(1-5)
Selection
Index
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
1
55.347
44.248
40.033
36.097
35.191
33.955
33.537
31.482
31.300
30.444
30.170
28.130
28.042
28.032
27.143
25.812
24.224
24.041
24.008
23.991
2.0
1.0
1.0
0.18
55.3
11.1
21.5
8.88
4.0
1.0
1.1
0.49
55.3
-71.4
-0.6
17.48
The scheme of evaluation and selection for the highland does not follow the standar flow
(figure 2.1). Two different Regional Trials were planted at Popayán (Cauca Department) and
Project IP3: improving cassava for the developing world
Output 2-62
its more relevant results are presented in Tables 2.52 and 2.53. Only few clones are common
to both trials, so comparison or across locations evaluations is not feasible. An important
reference point is that the most widely grown cassava clone in the region is “Algodona”
(MCOL 1522). This regional check had a very poor performance in the first trial (Table 2.52),
but was the second best in the second trial (Table 2.53). The best performing clones from
each trial will be merged in a common experiment to be evaluated across several locations
this year.
Table 2.52. Results from the Regional Trial planted in the highland location in the Cauca
Department (Popayán).
Clone
SM 1707-41
SM 1713-25
CM 7595-1
SM 1712-10
SM 1495-5
SM 1834-28
SM 1834-20
SM 1992-1
CM 7138-7
SM 1942-17
SM 1498-4
CM 7438-4
SM 1940-12
SM 1702-23
CM 7436-7
SM 1703-22
MCOL 2061
SM 1835-15
CM 8596-6
SM 1713-26
SM 524-1
CG 402-11
SM 1934-9
SM 1936-4
SM 1846-12
MCOL 2261
CM 8296-4
MCOL 1522
Maximum
Minimum
Average
St.Deviation
Output 2-63
Fresh root
yield
(t/ha)
50.9
47.9
50.5
47.8
53.7
53.0
57.1
48.6
47.3
41.3
48.3
37.8
38.0
39.7
50.8
34.5
24.4
48.6
34.7
33.6
30.6
39.7
31.6
26.6
30.0
32.0
28.0
26.1
57.1
24.4
40.5
9.77
Dry matter
(%)
38.6
38.6
35.8
39.9
36.8
35.4
34.5
35.6
37.6
37.1
35.6
37.4
36.1
35.4
34.9
38.2
34.6
34.0
36.2
36.5
35.3
32.9
36.7
36.4
35.6
33.4
33.5
34.6
39.9
32.9
35.9
1.69
Harvest
Index
(0-1)
0.62
0.41
0.54
0.41
0.43
0.47
0.56
0.43
0.44
0.44
0.48
0.38
0.47
0.44
0.45
0.28
0.60
0.47
0.42
0.35
0.40
0.46
0.32
0.29
0.31
0.37
0.35
0.31
0.62
0.28
0.43
0.09
Plant type
(1-5)
HCN
Selection
Index
4.0
3.3
3.7
2.3
3.0
3.0
2.3
3.3
2.0
3.0
2.7
3.7
3.7
3.3
2.0
3.3
4.3
2.0
2.7
3.0
3.3
3.0
2.7
3.7
3.3
4.0
4.0
2.3
4
2
3
0.65
4.0
4.0
2.5
8.0
4.5
5.5
7.5
3.5
3.0
4.5
4.5
3.0
2.5
3.5
4.0
4.5
3.0
6.0
3.0
7.0
4.0
3.0
4.0
4.0
7.5
2.5
2.0
4.0
8
2
4.3
1.62
41.06
20.90
20.42
19.57
17.28
12.14
11.62
8.69
7.12
6.26
5.80
5.57
4.90
-1.10
-1.39
-2.16
-3.42
-6.79
-8.35
-9.85
-12.86
-14.42
-15.11
-15.60
-17.22
-17.31
-21.85
-33.91
41.06
-33.91
0.00
16.17
2003 Annual Report
Table 2.53. Results from a second Regional Trial planted in the highland location in the
Cauca Department (Popayán).
Clones
SM 524-1
MCOL 1522
SM 2233-11
SM 1061-5
CM 7438-14
CG 402-11
SM 1053-23
SM 1835-28
SM 1058-13
MCOL 2261
SM 2229-36
SM 2227-21
CM 7138-12
SM 850-1
SM 1946-2
SM 2226-48
SM 998-3
CM 8106-4
SM 1495-22
MCOL 2740
SM 1944-10
SM 1937-1
SM 1703-17
SM 1938-12
CM 7190-2
MCOL 2061
SM 1933-5
SM 1833-21
SM 2311-3
Maximum
Minimum
Average
St.Deviation
Fresh root
yield
(t/ha)
56.0
43.6
38.8
56.5
42.2
42.9
36.9
41.9
42.8
35.4
37.5
37.1
32.7
42.4
42.0
37.9
38.2
36.0
41.9
32.3
27.9
30.9
28.6
44.6
23.9
26.1
22.7
21.3
23.7
56.5
21.3
36.71
8.85
Dry matter
content
(%)
36.3
37.0
39.9
30.8
36.1
33.7
34.8
34.5
31.4
34.9
35.1
36.4
37.2
31.7
30.9
35.4
34.5
35.6
32.4
34.9
36.1
34.4
35.3
25.1
35.9
33.1
34.4
32.4
30.8
39.9
25.1
34.14
2.77
Harvest
Index
(0-1)
0.49
0.68
0.44
0.56
0.49
0.63
0.58
0.45
0.62
0.57
0.50
0.41
0.46
0.56
0.61
0.41
0.45
0.37
0.43
0.48
0.34
0.37
0.36
0.56
0.37
0.29
0.29
0.26
0.29
0.7
0.3
0.46
0.11
Cooking
Quality
(1-5)
3.0
2.0
4.0
2.0
1.0
1.0
4.0
3.0
1.0
3.0
1.0
1.0
3.0
3.0
5.0
4.0
3.0
3.0
5.0
2.0
3.0
5.0
2.0
5.0
2.0
2.0
5.0
2.0
2.0
5.0
1.0
2.83
1.34
Selection
Index
29.14
25.62
17.90
17.00
13.10
13.00
7.23
6.44
6.00
5.59
5.31
4.95
4.43
3.62
2.84
2.76
2.34
-0.55
-0.78
-2.08
-9.77
-9.89
-10.24
-12.69
-13.28
-22.38
-22.54
-31.14
-31.93
29.14
-31.93
0.00
15.07
2.4.5. Selections for the Middle Magdalena Region
CIAT recently began evaluating cassava elite materials in this region. Working in the midMagdalena River region is important because of the social relevance of cassava and because
Project IP3: improving cassava for the developing world
Output 2-64
it combines characteristics of the sub-humid and the acid-soil savannas. One of the
objectives for these evaluations was to determine the kind of germplasm that adapts better to
this environment (south of Cesar and Bolivar Departments and Santander and Norte de
Santander Departments). Table 2.54 presents the most important results from the Clonal
Evaluation Trial conducted for this region in San Vicente (Santander Department).
Table 2.54. Relevant results from the Clonal Evaluation Trial for the Medium Magdalena River
Region. The trial involved 700 clones and was planted at San Vicente (Santander
Department).
Clone
or
parameter
Block 1 (243)
Block 2 (235)
Block 3 (222)
Fresh root
yield
(t/ha)
30.04
32.22
32.31
SM 2967-7
CM 9934-1
SM 2963-46
SM 2963-45
GM 235-87
56.12
46.68
63.65
48.47
51.02
GM 266-162
CM 9614-18
CM 9928-12
SM 2967-9
CM 8335-18
75.13
93.37
49.11
72.70
57.78
SM 2963-53
CM 9953-153
CM 9985-27
CM 9985-19
SM 2733-137
57.27
68.24
66.71
53.57
61.22
Maximum
Minimum
Average
St.Deviation
93.37
30.48
49.11
11.99
Maximum
Minimum
Average
St.Deviation
93.37
2.68
31.45
14.99
Output 2-65
Dry matter
Harvest
Plant type
content
index
Score
(%)
(0-1).
(1-5)
32.83
0.58
2.92
32.49
0.58
3.07
33.15
0.57
3.23
Best five clones in Block 1
33.30
0.62
1
36.45
0.65
2
33.47
0.68
3
34.97
0.66
2
33.95
0.67
2
Best five clones in Block 2
32.48
0.63
2
28.08
0.67
3
39.79
0.63
2
31.29
0.61
2
34.87
0.60
2
Best five clones in Block 3
38.77
0.59
1
35.89
0.62
2
37.03
0.68
3
36.18
0.60
1
33.99
0.60
1
Satistics of the 101 clones selected
42.21
0.90
4
26.90
0.45
1
34.87
0.63
2.46
2.85
0.07
0.85
Statistics of the 700 clones evaluated
56.93
0.90
5
22.73
0.23
1
32.85
0.58
3.07
3.28
0.09
0.90
Selection
Index
0.00
0.00
0.00
30.99
29.47
28.79
27.37
26.39
38.29
36.93
38.20
32.89
30.53
880.83
962.63
951.69
823.14
882.17
44.55
16.20
23.38
6.68
44.55
-64.87
-1.02
15.97
2003 Annual Report
Table 2.55 Relevant results for each family participating in the Clonal Evaluation Trial for the
Medium Magdalena River Region. The trial involved 700 clones.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Family
CM 8335
CM 9614
CM 9748
CM 9772
CM 9903
CM 9916
CM 9928
CM 9934
CM 9935
CM 9940
CM 9953
CM 9965
CM 9985
GM 70
GM 212
GM 215
GM 235
GM 266
SM 1521
SM 2733
SM 2826
SM 2830
SM 2832
SM 2859
SM 2861
SM 2864
SM 2882
SM 2963
SM 2965
SM 2967
Mother
SM 1210-4
MTAI 8
SM 494-2
CM 8602-27
SM 2075-1
MTAI 8
CM 6740-7
CG 1141-1
CM 523-7
SM 1219-9
CM 8593-13
MCOL 1515
SM 1219-9
CM 7073-7
CM 4574-7
SM 2075-1
CM 6740-7
CM 6740-7
SM 1219-9
CM 7033-3
CM 4365-3
CM 7951-5
CM 3372-4
SM 643-17
CM 7514-8
SM 1210-10
SM 1741-1
SM 1565-17
CM 3299-4
SM 805-15
Father
CM 6740-7
MTAI 8
SM 2069-26
CM 8602-27
MNGA 19
SM 653-16
SM 805-15
SM 1741-1
SM 2069-29
CM 2177-2
MTAI 8
MNGA 19
SM 2102-36
SM 1741-1
SM 653-14
CM 6740-7
SM 1565-17
CM 523-7
Size
14
23
22
25
31
18
32
26
26
37
22
28
18
32
26
27
35
15
33
21
23
27
16
18
20
22
20
21
8
14
Selected
5
7
6
6
7
4
7
5
5
7
4
5
3
5
4
4
5
2
4
2
2
2
1
1
1
1
0
0
0
0
% Selection
35.7
30.4
27.3
24.0
22.6
22.2
21.9
19.2
19.2
18.9
18.2
17.9
16.7
15.6
15.4
14.8
14.3
13.3
12.1
9.5
8.7
7.4
6.3
5.6
5.0
4.5
0.0
0.0
0.0
0.0
Following the same criteria used for other regions, the Clonal Evaluation Trial for the
Magdalena River was consolidated to present the mean performance for each of the 30
families of clones included (Table 2.55) as well as that for the average of all the progenies
derived from a given parental line (Table 2.56). As expected some well-known clones such as
MTAI 8 produced excellent progenies with high frequency of selection (Table 2.56), whereas
other clones (CM 3299-4) produced progenies that were clearly deficient for one reason or
another.
Project IP3: improving cassava for the developing world
Output 2-66
Table 2.56 Relevant results for each parental clone whose progenies participated in the
Clonal Evaluation Trial for the Medium Magdalena River Region. The trial involved 700
clones.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Progenitor
SM 494-2
MTAI 8
CM 8602-27
SM 2069-26
CG 1141-1
SM 2075-1
MNGA 19
CM 8593-13
CM 2177-2
MCOL 1515
CM 6740-7
SM 1210-4
SM 1219-9
CM 7073-7
CM 4574-7
SM 653-14
SM 2102-36
SM 1741-1
SM 805-15
CM 523-7
CM 7033-3
CM 4365-3
CM 7951-5
CM 3372-4
SM 643-17
CM 7514-8
CM 3299-4
Total # of clones
22
81
56
47
26
58
64
22
28
28
132
36
88
32
26
59
27
92
40
47
21
23
27
16
18
20
8
# selected clones
6
20
13
10
5
11
12
4
5
5
23
6
14
5
4
9
4
12
5
5
2
2
2
1
1
1
0
% selected clones
27.3
24.7
23.2
21.3
19.2
19.0
18.8
18.2
17.9
17.9
17.4
16.7
15.9
15.6
15.4
15.3
14.8
13.0
12.5
10.6
9.5
8.7
7.4
6.3
5.6
5.0
0.0
The best 144 clones from the Clonal Evaluation Trial described in Table 2.54 were planted in
three Preliminary Yield Trials. Each trial had 48 clones with three replications and 10 plants
per plot.
Table 2.57 presents the results of the performance of the best germplasm so far identified for
the Magdalena River region. Eighteen clones were evaluated (two of them regional checks). As
it is frequently the case MTAI 8 had an outstanding performance in this region. The best
regional check occupied the 8th rank within this experiment.
Output 2-67
2003 Annual Report
Table 2.57 Relevant results of a Regional Trial for the Medium Magdalena River Region. The
trial involved 16 clones and two local checks.
Clon
MTAI 8
HMC 1
CG 1141-1
SM 1557-17
CM 6119-5
CM 523-7
CM 849-1
CM 7514-7
SM 653-14
MVEN 25
SM 1433-4
MPER 183
MBRA 383
CM 4843-1
CM 3306-4
CM 1219-9
Maximum
Minimum
Average
St.Deviation
Chamalota
Monablanca
Fresh root
Dry matter
Fresh
Dry matter
yield
yield
foliage
content
(t/ha)
(t/ha)
t/ha
(%)
27.4
9.2
10.7
33.6
24.1
8.5
18.1
35.3
22.7
8.4
9.8
36.9
22.4
7.3
11.1
32.5
22.0
8.1
13.3
36.5
20.4
7.3
10.9
36.0
20.4
7.0
11.5
34.5
19.9
7.5
13.9
37.8
19.8
7.2
11.1
36.4
19.8
6.7
9.6
33.9
17.7
6.4
8.9
36.1
16.9
4.9
9.4
29.0
14.7
5.0
8.7
34.2
13.5
4.5
5.4
33.6
8.3
2.7
7.5
32.1
7.3
2.4
4.4
33.3
Statistics of experimental clones evaluated
27.41
9.20
18.15
37.77
7.33
2.44
4.44
28.99
18.59
6.45
10.28
34.48
5.41
2.00
3.26
2.21
Local checks
20.93
7.03
10.93
33.57
11.67
4.08
17.04
34.98
Harvest
index
(0-1)
0.72
0.57
0.70
0.67
0.64
0.65
0.65
0.58
0.63
0.68
0.63
0.68
0.64
0.74
0.60
0.67
0.74
0.57
0.65
0.05
0.66
0.41
2.4.6. Selections for the Tolima-Huila Departments.
This region shares characteristics of the acid-soil savannas, the mid-altitude valleys and
some of the water deficits present in the North Coast of Colombia (sub-humid environment).
Therefore a special Clonal Evaluation Trial was prepared for this environment. Unfortunately,
there was difficulty of finding an adequate site for planting this experiment in the Target
Environment and not to loose the vegetative cuttings it was decided to plant it at CIAT
Experimental Station in Palmira. Table 2.58 presents the most relevant results of this
evaluation. The trial followed the standard fashion of stratrification on three blocks, which
revealed large differences (particularly for the second one).
Project IP3: improving cassava for the developing world
Output 2-68
Table 2.58 Relevant results from the Clonal Evaluation Trial for the Tolima-Huila Region. The
trial involved 608 clones.
Clone
or
parameter
Block 1 (216)
Block 2 (202)
Block 3 (190)
Fresh
root yield
t/ha
23.43
29.33
24.31
CM 9733-108
CM 9914-6
SM 1521-27
CM 9791-64
CM 9914-8
53.57
44.64
35.49
29.02
25.11
GM 234-165
CM 9912-79
CM 9962-12
SM 2802-76
SM 2802-78
67.97
37.95
51.56
59.71
46.43
SM 2826-40
CM 9914-16
GM 265-191
SM 2802-85
SM 2865-97
86.16
43.30
59.93
40.07
50.22
Maximum
Minimum
Average
St.Deviation
86.16
22.88
39.95
10.33
Maximum
Minimum
Average
St.Deviation
86.16
2.12
25.67
11.26
Output 2-69
Fresh
Harvest
Dry matter
foliage
index
content
t/ha
(0-1)
(%)
12.47
0.65
36.42
19.57
0.60
36.54
16.83
0.59
36.42
Best five clones from Block 1
21.21
0.72
37.42
15.96
0.74
36.90
15.07
0.70
40.28
7.48
0.80
41.55
8.93
0.74
41.85
Best five clones from Block 2
29.46
0.70
36.14
23.44
0.62
44.79
27.12
0.66
37.40
22.99
0.72
32.55
25.78
0.64
40.70
Best five clones from Block 3
31.03
0.74
41.77
24.89
0.64
40.15
39.84
0.60
36.86
15.96
0.72
36.65
22.32
0.69
35.90
Statistics of the 87 clones evaluated
55.69
0.83
44.79
6.14
0.50
32.55
21.76
0.66
38.36
8.76
0.06
2.43
Statistics of the 608 clones evaluated
55.92
0.83
45.52
2.46
0.17
23.77
16.31
0.61
36.44
8.33
0.10
3.09
Plant type
Score
(1-5)
2.58
2.48
3.13
Selection
Index
2.00
1.00
1.00
2.00
1.00
40.72
36.16
34.15
31.05
29.72
1.00
1.00
1.00
1.00
2.00
45.69
35.91
32.07
31.35
29.99
3.00
2.00
3.00
1.00
2.00
73.92
34.28
32.72
32.05
32.08
4.00
1.00
1.92
0.78
73.92
16.29
24.27
7.46
5.00
1.00
2.72
1.01
73.92
-83.73
0.00
18.09
2003 Annual Report
References
CIAT, 2001. Annual Report International Center of Tropical Agriculture (CIAT), 2001. Cali,
Colombia.
CIAT, 2002. Annual Report International Center of Tropical Agriculture (CIAT), 2001. Cali,
Colombia.
Gradner, C.O. 1961. An evaluation of effects of masss selection and seed irradiation with
thermal neutrons on yield of corn. Crop Sci. 1:241-245.
Jennings, D.L. C.H. Hershey, C.H. 1985 Cassava breeding: a decade of progress from
international programs. 89-116. In: G.E. Russell (Ed.) Progress in Plant Breeding.
Butterworths Press. London – Boston.
Van Oirschot, Q. E. A., O’Brien G.M., Dufour D., MEl-Sharkawy M.A. and Mesa E., 2000. The
effect of pre-harvest pruning of cassava upon root deterioration and quality
characteristics. J. Sci. Food Agric. 80:1866-1873.
Project IP3: improving cassava for the developing world
Output 2-70
Personnel and Collaborating Institutions.
Coordinator: Hernán Ceballos (CIAT)
Participating Personnel from CIAT:
Elizabeth Alvarez
Antony Bellotti
Fernando Calle
Lee Calvert
Alba Lucía Chávez
Martin Fregene
Gustavo Jaramillo
María Cristina Guzmán
Jorge Iván Lenis
Javier Lopez
Nelson Morante
Eusebio Ortega
Juan Carlos Perez
Ricardo Sanchez
Teresa Sanchez
Joseph Tohme
Personnel from other CGIAR Center:
Howard Bouis (IFPRI)
Alfred Dixon (IITA)
Willy Roca (CIP)
Personnel from Universidad Nacional de Colombia:
Sara Mejía de Tafur
Yamel López
Guillermo Perez
Jorge Eliecer González
Carmen Tulia Potosí
Personnel from CLAYUCA:
Licímaco Alonso
Luis Fernando Cadavid
Jorge Luis Gil
Amalia Jaramillo
Bernardo Ospina
Personnel from ILTAB-DANFORTH CENTER:
Claude Fauquet
Nigel Taylor
Main contacts from other collaborating institutions
K. Abraham (CTCRI, India)
David Celiz (Corporación Colombia Internacional, Colombia)
Output 2-71
2003 Annual Report
Barry Nagle (National Starch, USA)
Eduardo Flores (Concentrados del Norte, Colombia)
Wania Fukuda (Embrapa, Brazil)
Luis Armando Gerstl (Agropecuaria Mandioca, Venezuela)
Guy, Henry (CIRAD, Brazil)
Johnson, Anthony (DAI Intl., Haiti)
Kaimian, Li (Chinese Academy of Tropical Agricultural Sciences (Hainan, China)
Antonio Lopez (CORPOICA, Colombia)
Teresa Losada (IAC, Brazil)
Kim, Hoang (Institute of Agricultural Sciences of South Vietnam, Vietnam)
José Mantilla (Universidad Central de Venezuela, Venezuela)
Marcio Porto (Embrapa, Brazil)
Palupi Puspitorini (Grat Giant Pineapple Coy, Indonesia)
Carlos Restrepo (Corn Products Intl, Colombia)
Carlos Reyes (Almidones Nacionales, Colombia)
Sergio Rodriguez (INIVIT, Cuba)
Diego Miguel Sierra (Federación Nacional de Avicultores – Colombia)
Erwin Silva (Industrias Protón, Colombia)
Watananonta, Watana (Rayong Field Crops Research Center, Thailand)
Acknowledgement to Donor Agencies financing one or more of the activities
described above:
Federación Nacional de Avicultores (Colombia)
McCain Foods Intl. (Canada, Colombia)
Ministerio de Agricultura y Desarollo Rural (Colombia)
NIPPON Foundation (Japan)
USAID (United States of America)
Rockefeller Foundation
United Nations Development Programme (UNDP)
Project IP3: improving cassava for the developing world
Output 2-72
OUTPUT 3
Collaboration with other institutions.
Activity 3.1 Support national programs that have traditionally collaborated with
CIAT in the development and improvement of cassava.
Rationale:
CIAT has the responsibility to contribute with cassava research worldwide. In the past, this
was achieved through the collaboration of National Agriculture Research Programs (NARs),
and in the case of Africa, with the valuable collaboration with IITA. This scenario has
changed drastically through the last decade, when the NARs in most of the tropical countries
weakened consistently. However, new institutions and partners are assuming a leading role
and CIAT is actively searching for these new partners. In this activity, at least for Latin
America, we are closely collaborating with CLAYUCA. In the implementation of industrial
uses of cassava, because of the convenience of our location, most of the validation and
adaptive research is carried out in Colombia. Once the technology (for instance, for the
artificial drying of cassava roots) is evaluated and offers acceptable results, it can be moved
out to other countries. This strategy implies that a considerable portion of our research is
carried out in Colombia. However, this does not imply that cassava projects at CIAT are
restricting their activities only to Colombia.
Specific Objectives:
a) To promote the use of cassava and the adoption of new technologies and germplasm by
cassava growing countries of the world.
b) To contribute to the training of personnel involved with cassava research.
c) To identify new partners in each country.
Results
A major thrust in CIAT’s strategy to achieve the stated objectives has been through training
and visits to NARs, in addition to the provision of germplasm described in Output 2. In Table
3.1 a list of the most important events in which personnel from the project participated is
provided. Although some of these events were scientific meetings, it should be pointed out
that the list involves only those events leading to the development of research proposal or
else were part of ongoing collaborative efforts.
There are many more specific activities and contributions that cannot be mentioned because
of their informal nature. An important activity in this regard is the continuous consulting
from producers, students, researchers and processors from Colombia and other countries.
An important amount of energy is dedicated to satisfy the demand for information and
products through these requests. For instance from January to the end of October a total of
2946 e-mail messages were sent from the mail box of the Coordinator of the IP3 project.
About 50% of these messages were internal and had mainly an administrative nature. The
rest o the messages are estimated to have been sent outside CIAT, and most of them were in
response of requests made by different people or institutions.
Output 3-1
2003 Annual Report
Table 3.1. Events where personnel from cassava breeding project participated for the
development or execution of research proposals. Additional events were attended by
personnel working in the areas of enthomology, plant pathology, and biotechnology
and are not listed here to avoid duplications.
Date
June
June
June
January
Jan-Aug
September
April
May
September
September
October
Event
Curso en sistemas modernos de producción,
procesamiento y comercialización de yuca
Curso en sistemas modernos de producción,
procesamiento y comercialización de yuca
Curso en sistemas modernos de producción,
procesamiento y comercialización de yuca
Meeting for the promotion of bioavailability studies
on carotenes from cassava roots.
Field trials, promotion of cassava, introduction
hipping machine, etc.
I Taller internacional de conservación, evaluación y
usos de los recursos genéticos de yuca y I Reunión
Clayuca-Perú.
II Congrteso Brasileiro de Fitomelhoramento
II Encontro
Latino-Americano dos Centros de
ecotecnologías para o Desenvolvimento Sustentável.
Fundación Kellogg, Serta
1er Congreso Venezolano de Mejoramiento y
Biotecnología Vegetal
Crop meeting within Harvest Plus
Launching of the participation of Nicaragua
as CLAYUCA member
Location
Anzoátegui,
# particpants
25
Cojedes,
Venezuela
Zulia, Venezuela
25
Port-au-Prince,
Haiti
Haiti
10
Huaral-Perú
Porto Seguro
Gloría do Goitá –
Pernambuco
–
Brasil.
Maracay,
Venezuela
Kampala, Uganda
Managua,
Nicaragua
25
10-30
25
300
80
185
42
10-30
In addition the project has received the visit of different scientists in activities that can be
described as training. However, the nature of these visits was very diverse and varied from
gradute students to visiting scientist. During 2003 the following colleagues have visited the
project for periods longer that two weeks:
Henry Ojulong – Uganda – Ph.D. Student
Elizabeth Baldeyusa – Uganda - Ph.D. Student
Yoel Beavides – Cuba – Training in Molecular Markers
Mary Luz Folgueras – Cuba – Training in plant pathology
Profesor José Branco Miranda Filho – Brazil -Interaction in the area of Quantitative Genetics.
George James – India - General visit to the cassava research project.
Hoang Kim - Institute of Agricultural Science of South Vietnam (IAS).
Project IP3: improving cassava for the developing world
Output 3-2
Activity 3.2 Development of collaborative projects with partners in Africa, Asia and
Latin America and the Caribbean.
Rationale:
There is a clear trend in the last few years for a reduction of core contributions to CIAT and a
simultaneous increase of special projects. Also the trend involves a stronger participation of
NARs as key partners in the execution of different projects. Several proposals have been
developed and submitted during the course of the year and few of them have already been
approved. Below is a brief description of each of these successful research proposals.
High carotene cassava roots.
This is the cassava component of the Biofortification Initiative (now Harvest Plus). After more
than ten years developing the basic data that allowed the initiative to move forward, full
activities will began in 2004. A planning workshop was held in Uganda during September 1819, 2003 (See Output 1, Activity 1.5). A work plan has been developed and submitted to
Harvest Plus administration. The work plan involves a strong participation by key partners in
the different continents. CIAT has assumed the coordinating role for cassava worldwide and
collaborates closely with IITA. Harvest Plus has also allowed for a more direct collaboration
with the Sweet Potato project from CIP. In fact the workshop held in Uganda was organized
in close collaboration with that project coordinator (Regina Kapinga)
In Africa the activities will be coordinated by IITA and several countries will eventually join
the field activities for the development, multiplication, and promotion of elite germplasm with
yellow, high-carotene roots. EMBRAPA-CNPMF (Brazil) and CIAT will produce vitroplants of
elite clones with high carotene, drought resistance and other desirable characteristics and
then ship them to IITA for their introduction and incorporation in the breeding programs in
Africa. Eventually some clones could be released if they prove to have outstanding
performance. EMBRAPA-CNPMF will also produce sexual seed and share it with other
collaborators. EMBRAPA has been a very important and traditional partner for CIAT in the
area of cassava research. This project will benefit and continue with this history of
collaboration between the two institutions.
In Latin America and the Caribbean two key countries will participate in the initiative: Brazil
and Haiti. The target region in Brazil is the North East region where cassava is important and
poverty and vitamin A deficiency are common. EMBRAPA – CNPMF as well as other
EMBRAPA institutes such as CENARGEN in Brazilia and CTAA in Rio de Janeiro.
Haiti is the second target country for the deployment of cassava clones with yellow roots. In
that case CIAT will work together with World Vision in the Central Plateau. A variety (Yema
de Huevo) has already been identified and the initial reaction has been very positive because
of its excellent cooking quality.
In Asia, also two countries have been targeted based on the prevalence of human
consumption of cassava and vitamin A deficiency on one hand and an assessment of the
possibilities of attaining success. The countries are India and Vietnam. India has an excellent
root and tuber program lead by the Central Tuber Crops Research Institute (CTCRI) in Kerala
State, Southern India. CTCRI will participate within Harvest Plus not only in the area of
cassava research but also on sweet potato. In Vietnam the higher human consumption of
Output 3-3
2003 Annual Report
cassava takes place in the central region (for example Hue Province). CIAT will introduce
cassava germplasm with yellow roots into Vietnam since not much such germplasm exists
currently. The main collaborator in Vietnam for this particular initiative is Thai Ngyen
University of Agriculture and Forestry.
Doubled-Haploids for cassava genetic improvement.
The introduction of inbreeding in cassava has been an evolving idea for a few years now.
Between April-July 2003 several meetings took place that lead to the successful development
of a research proposal to develop a protocol for the production of doubled haploids in cassava
from male flowers and to incorporate their use in different cassava genetic improvement
programs. An additional advantage of this system is that will allow for the identification of
useful, commercially beneficial recessive mutants. The main ideas of this research proposal
are further described in Output 4. Here the collaboration with key partners in different
continents is highlighted.
There will be several countries participating in the project mainly in the process of producing
segregating populations with varying levels of inbreeding. The main objective is to improve
tolerance to inbreeding in elite cassava germplasm and to eventually identify useful recessive
traits that could offer commercial advantages.
CIAT has initiated contacts with Uganda (National Agricultural and Animal Research Institue)
in Eastern Africa and Ghana (Council for Scientific and Industrial Research and Crops
Research Institute) in Western Africa. In Latin America EMBRAPA-CNPMF and Instituto
Agronomico Campinas (IAC) in Brazil and INIVIT in Cuba have participated in the initial
meetings and will be invited to participate in the execution of the project. CTCRI in India, the
Institute of Agricultural Science of South Vietnam (IAS) and Khon Kaen Field Crops Research
Center in Thailand will initially be involved in the case of Asia. The problems derived from the
SARS epidemics prevented China to join the events early in 2003, but hopefully there will be
an opportunity for China to also participate.
CLAYUCA
As stated in the Annual Reports from previous year the creation of CLAYUCA has been a very
positive development of the cassava project at CIAT. CLAYUCA is effectively helping in the
technology transfer and in south-to-south collaboration among the participating countries.
Through CLAYUCA, CIAT maintains a close association with many countries in the Region.
Moreover, the creation of a similar consortium (ACCORD) is currently sought for Asia. Initial
steps have been given in this direction.
Development of research proposal within Challenge Programs.
Two new global challenge programs have been approved within the CG system one is the
Water Challenge Program and the second is the Genetic Resources Global Challenge Program
(GRGCP). Because of their very nature these research proposals involve close collaboration
with different NARs and NGOs. The proposal within the Water Challenge program was not
approved. In the GRGCP, however, several proposals have been successfully submitted.
Many of them involve join activities with EMBRAPA and CORPOICA, representing respectively
Project IP3: improving cassava for the developing world
Output 3-4
Brazil and Colombia, two key countries from the cassava genetic biodiversity point of view.
These research proposals were also developed in close interaction with IITA.
Activity 3.3. The collaboration with Colombia.
Project IP3 has a special relationship with Colombia. As host country for CIAT it is in
Colombia where a large proportion of the research on cassava is conducted. But the
relationship with Colombia goes well beyond this point. Colombia has been a key financial
supporter of the projects turning around industrial uses of cassava both from the
Government (Ministry of Agriculture) and the private sector (Poultry Growers Association).
There are too many events developed in Colombia to be listed. Instead, what is happening
with the idea of Trapiches Yuqueros will be used as an illustration of the intensity and
success of the activities developed in the country. A Trapiche Yuquero is basically a
centralized facility for artificial (or mixed) drying of cassava roots and foliage, surrounded by
300-1000 hectares of cassava grown by many farmers. Frequently the farmers growing the
cassava are also part owners of the drying facility.
CIAT and CLAYUCA have been promoting the creation of Trapiches Yuqueros for three years
now. To materialize the idea two main issues had to be dealt with:
a) A production of cassava roots (and foliage) that was economically competitive so the
end product of the Trapiche Yuquero could compete with imported maize in the
production of animal feed. To do this cassava productivity had to be increased
(improved industrial clones, clean seed, appropriate cultural practices, etc.) and
production costs reduced (mechanization, sound fertilization approaches, timely
weeding, etc.). After the initial efforts it is now a reality widely accepted across
Colombia that average fresh root yields above 20 t/ha can be attained with
production costs to allow selling cassava at about 30 US dollars/ton of fresh roots.
b) Development of artificial drying plants at affordable price. The first prototype of an
artificial drying plant was developed at CIAT by CLAYUCA and a private company
(PROTON), and lead to the first quotation of such machinery. The current price of
this artificial drying plant is only 20% of the original price in 2000. This is an
interesting example of how dynamic investment in technology for cassava has
become in a country like Colombia. Although the initial efforts were made by
CLAYUCA and CIAT, the current developments are totally lead by the private sector.
The year 2003 may be a turning point for cassava in Colombia because not only the first
Trapiche Yuquero was created this year, but also because ten additional Trapiches followed.
This idea has been adopted by many different communities, which further developed or
modify the original idea to adapt it to local conditions. Table 3.2 lists the locations where the
eleven Trapiches Yuqueros are or will be built.
CLAYUCA and CIAT are very proud of these developments. Although this is taking place only
in Colombia at this point, the idea has generated enough interest in other countries in the
Region (Nicaragua, Venezuela, Ecuador) as well as outside the Region (Nigeria).
Output 3-5
2003 Annual Report
Table 3.2 List of Trapiches Yuqueros under construction or to be constructed in Colombia
Capacity
(t/hour fresh roots)
3 t/hour
3 t/hour
3 t/hour
3 t/hour
0.5 t/hour
20 t/day
15-20 t/day
3-5 t/hour
1 t/hour
3 t/hour
3 t/hour
Location
Type
San Pablo. South of Bolivar Department.
San Juan de Arama. Meta Department.
Valencia. Córdoba Department (Figure 3.1).
Tamalameque. South of Cesar Department.
Urabá Antioqueño.
Sabana de Torres. Santander Department (Figure 3.2).
Cúcuta. Norte de Santander Department.
Aguazul, Casanare Department.
Berástegui, Córdoba Department (Figure 3.3).
Orito. Putumatyo Department.
La Hormiga. Putumatyo Department.
Artificial
Artificial
Artificial
Artificial
Artificial
Mixed system
Mixed system
Artificial
Artificial/Mixed
Artificial
Artificial
The situation derived from the creation of the Trapiches Yuqueros listed in Table 3.1 is very
interesting and encouraging for the following reasons:
a. There is an aggressive incorporation of the private sector of Colombia and Brazil in the
development of technologies for the artificial drying not only of cassava roots but also the
foliage and even other commodities such as sweet potato.
b. The joining of the private sector has lead to a substantial reduction of the prices of the
machinery required for the drying of cassava products. Within three years the price was
reduced down to 20% of the original price.
c. The participation of new actors in the development of technologies has lead to the
development of mixed systems where cassava roots are dried one day in sun-drying patios
and then the second day they are further dried using an artificial system.
d. The share number of Trapiches Yuqueros created only in 2003 reveals the confidence of
different actors across Colombia in the basic ideas lying behind the concept: cassava is a
competitive crop and can be used as a cash crop for industrial processes.
e. The creation of these Trapiches Yuqueros is also a clear illustration that cassava is an ideal
crop for promoting rural development. Every Trapiche listed in Table 3.1 is located in rural
areas, more frequently than not, in areas with serious social problems with reduced
alternatives for the farmers living there.
d. The whole initiative has created an space where a country like Colombia can even export
technologies to other regions and therefore promoting south-to-south cooperation.
Project IP3: improving cassava for the developing world
Output 3-6
Figure 3.1. Construction of the artificial drying plant for the Trapiche Yuquero in Valencia,
Córdoba Department. Colombia. For further information contact Mario Tovar (Phone
(57-66) 344-6087/344-6365/344-6412)
Output 3-7
2003 Annual Report
Artificial drying phase
Cassava chips
Metal
mesh
Hot air
Natural sun-drying phase
in “patios”
Figure 3.2. Illustration of a mixed system for drying cassava roots used in the Trapiche
Yuquero in San Vicente, Stantander Department.
Hot air return
Dried flour
Fresh roots
Hot air
Figure 3.3. Illustration of a mixed system for drying cassava roots used in the Trapiche
Yuquero in Berástegui, Córdoba Department. For further information contact Ricardo
Muskus (Phone (57-4) 234-7905).
Project IP3: improving cassava for the developing world
Output 3-8
Acknowledgement to Donor Agencies financing one or more of the activities
described above:
Ministerio de Agricultura y Desarollo Rural (Colombia)
Federación Nacional de Avicultores (Colombia)
McCain Foods Intl. (Canada, Colombia)
Ministerio de Agricultura (Perú)
NIPPON Foundation (Japan)
A major factor contributing to the success in the area of collaboration with NARs from Latin
American Countries has been the complementation and collaboration between CIAT and
CLAYUCA, which is dully acknowledged here.
Output 3-9
2003 Annual Report
OUTPUT 4
Developing new approaches for cassava breeding integrating
biotechnology tools and quantitative genetics principles.
4.1 Development of a breeding scheme based on the use of doubled-haploids.
Rationale
A Planning Workshop was conducted in June 11-12, 2003 to elucidate the potential uses that
haploid technology may have to increase the efficiency of cassava breeding and/or the value
of the crop and to describe the process required for the development and adaptation of such
technology in this crop. This event was possible thanks to the financial support of UNDP and
Rockefeller Foundation.
Cassava scientists from Africa, Asia and Latin America visited research programs in different
continents to conduct a Needs Assessment Study (NAS). The aim of this study was to provide
background information on cassava as a crop, breeding objectives of each program,
bottlenecks in cassava research and marketing, and the potential benefits of the haploid
technology for cassava research. The Needs Assessment Study was conducted three weeks
before the Planning Workshop and a summary of its main conclusions was presented at the
Workshop.
The Planning Workshop was structured to first provide the participants with background
information on cassava as a crop and the breeding objectives from CIAT (International Center
for Tropical Agriculture) and National Agriculture Research Institutions. Cassava research by
IITA (International Institute of Tropical Agriculture) was also described. The discussions also
included progress using haploid technology for breeding major food and non-food crops,
background on cassava flower biology and cytogenetics, and the state of the art for the
development and use of haploid technology in breeding self and crossbreeding species,
including those with known inbreeding depression. The workshop finalized with a day of
discussion by disciplines (breeding/genetics and tissue culture) on technical aspects to
outline research strategies for the use of inbreeding in cassava genetic improvement and for
the development of haploid technology in this crop.
Below is a summary of the main points disclosed during the background presentations
followed by the tentative research strategies proposed for i) increasing inbreeding tolerance in
cassava and incorporating inbreeding in cassava genetic improvement; and ii) developing a
protocol for haploid technology in cassava.
4.1.1. Cassava as a crop and breeding objectives
The demand for cassava production is expected to increase in both the domestic and export
markets. In order to satisfy this increased demand, and cope with the consequences of an
Output 4-1
2003 Annual Report
open economy, costs should be reduced by increasing the crop’s yield potential and by
developing improved production and processing technologies. A strong trend for increased
use of cassava in different industrial processes, particularly starch production, was
recognized. The use of cassava for starch is widespread in Asia and in some Latin American
countries, and its importance in Africa is growing.
Many alternatives for human consumption of cassava already exist—fresh boiled roots, gari,
bakery products, etc. — these traditional uses will continue to play an important socioeconomic role. The interest in exploiting cassava foliage is also on the rise; however, the
largest changes in cassava utilization will occur in its industrial uses: starch, flour, and
derivatives (manitol, sorbitol, sodium monoglutamate, dextrins, citric acid, acetic acid, lysine,
etc.). Cassava roots are used also for animal feed or as silage. The leaves are increasingly
utilized in different alternative ways (i.e. protein and vitamin supplement). Therefore, the
relative importance of cassava in tropical and subtropical agriculture can be increased
through new uses and products.
Increased genetic variability will be required to achieve diversified end uses of cassava
products. This implies not only access to existing variability but effective tools for proper
evaluation and screening, as well as the generation of new genetic variability for several key
traits. A fundamental step for broadening the array of uses of cassava products is the
investment in research on processing technologies.
Cassava breeding methods still follow the traditional approach implemented many years ago,
with only minor adjustments being made by different programs. Cassava breeding starts with
the creation of large segregating populations through crosses among a selected group of
parents. The selection of parents and best clones within segregating populations is entirely
based on their per-se performance. The emphasis and largest allocation of resources goes to
the selection and evaluation of clones within segregating populations, which is onerous,
time-consuming, and of doubtful effectiveness. A typical program starts out with no less
than 10,000 botanical seeds and takes about 10 years to eventually release a new variety,
which many times proves not to be much better than the existing varieties. Transfer of useful
genes from one clone to another is cumbersome and inefficient. The backcross scheme
utilized successfully in most crops cannot be implemented in cassava because of the
heterozygous nature of the parental lines. Generally speaking there has been very little effort
to identify elite germplasm based on their breeding value or in introducing inbreeding at
certain stages of the breeding process.
Most cassava breeding programs worldwide pursue an array of traits such as high starch
yield and stability of starch content throughout the crop’s growth cycle and over seasons;
tolerance or resistance to diseases and pests; tolerance to drought and other abiotic stresses;
good quality of vegetative planting material for long lasting storage and good germination;
stability of productivity; and early bulking.
When cassava is used to produce starch, its industrial quality and stability throughout the
growth cycle and over seasons become prevalent goals. Viscosity and transparency of gels,
variations in the amylose:amylopectin ratio, the size of starch granules, among others, are
important traits for which there is little variation available. Certain ways of human
Project IP3: improving cassava for the developing world
Output 4-2
consumption of cassava requires low levels of cyanogens. High protein yield in leaves and/or
roots is very important for human and animal nutrition, but the genetic variability for this
trait is limited. Recently, attention has been given to high carotene content in roots. Good
cooking quality is also essential and the specific physical and chemical characteristics
required will depend on the processing method used (i.e., cassabe, farinha, fufu, gaplek, gari,
sago, etc.). A key trait for which there seems to be very little genetic variation is that of
reduced post-harvest deterioration of roots. The short shelf life of cassava roots continues to
be a major constraint for marketing and utilization. In certain regions, the use of mechanized
cultural practices, for example for planting and harvesting, has been observed. In these
areas, the adaptability of cassava varieties to mechanization is highly desirable.
4.1.2. State of the art of using haploid technology for breeding major food crops.
Breeding of the most important crops always relies on the use of inbreeding at one stage or
another of the process. In the case of self-pollinated crops the end product is an inbred
(homozygous) line. Cross-pollinated crops include inbreeding for an array of reasons ranging
from its advantage in reducing genetic load (undesirable genes) to fixing the genetic make up
of parental lines for the production of hybrid seed. Only in few crops inbreeding has not been
incorporated, mostly because of self-incompatibility barriers or strong inbreeding depression.
The most important advantage of the use of haploid technology is the rapid and complete
achievement of homozygosity. In turn, this implies a reduction of costs and faster genetic
improvement (by reducing the time involved in the production of inbred lines). Doubledhaploids (DH) are also frequently used in the production of segregating populations used for
basic molecular research. The advantage of the technology is higher the longer the time
required to attain complete homozygosity through successive self-pollinations. The relative
efficiency in the production of DH affects the costs of their production and their relative
advantage over materials produced by the traditional method. One disadvantage of the
technology is that the breeder cannot select through the inbreeding process.
For the Planning Workshop, rice and maize were selected as case studies to analyze the
potential and limitations of using haploid technology for developing breeding materials at
large scale. These case studies were selected bearing in mind contrasting and broadspectrum scenarios. Rice is a self-pollinated crop, whereas maize is a cross-pollinated
species showing inbreeding depression such as cassava.
Double-haploid production in rice was first reported in Japan in 1968, and then followed an
adaptation process worldwide for its mass application. More than 100 rice varieties had been
released through this technology. However, the most efficient application has been to develop
breeding populations for broadening the genetic base through the introgression of germplasm
from other regions, as well as inter-specific hybridizations. DH populations have also been
used for molecular mapping analyses. CIAT uses DH in its conventional rice breeding
program since 1988. About 14,000 plants are generated year-round and subjected to
pedigree or recurrent selection breeding methods to increase yield potential, incorporate
resistance to diseases and pests, incorporate tolerance to cold and drought, and to improve
grain quality. This capacity was transferred to WARDA in 1992, and to several Latin
Output 4-3
2003 Annual Report
American National Rice Breeding Programs in 1994-1995 with the support of the Rockefeller
Foundation. International collaboration had included the generation of DH populations used
for cloning the first avirulence gene from Pyricularia grisea, and for QTL mapping of drought
tolerance genes in rice at Texas A&M and IRRI.
DH technology is also used in the production of inbred lines in cross-pollinated crops such
as maize. In this case the end product are the parental lines to be used in the production of
commercial hybrids. In maize, however, there is an alternative to the production of
homozygous lines through tissue culture. Basically the same kind of product can be obtained
taking advantage of the “ig gametophyte mutant system” and without the need of tissue
culture processes, which ultimately, reduces the costs per inbred line produced.
4.1.3. State of the art for the development and use of haploid technology in self and
crossbreeding species
A series of four presentations summarized the knowledge on cassava compared with other
open- and self-pollinated crops with or without inbreeding depression, on ways to overcome
potential bottlenecks for a mass use of DH in breeding programs, and on current approaches
to develop protocols for new species.
Fundamental cytogenetics aspects of cassava, flower biology, and pollen development were
highlighted. Flowering (up to four peaks) is associated with branching. Most flower buds
formed at early growth are abortive. There is variation in flowering time (from 4 mo to > 8 mo)
and in number of flowers between varieties. There is differential genotype-environment
interaction on flowering among varieties, as well as a strong environmental effect on number
of flowering peaks within the same variety. Most varieties show microspore in the tetrad
stage of development when flower buds are between 0.8 and 1 mm in diameter. Flower buds
of 1.2-1.5 mm in diameter contain microspores at the uninucleate stage, whereas those of
1.5-1.8 mm in diameter are at the binucleate. Most flower buds contain mature pollen grains
when they reach 2.0-2.2 mm in diameter. This association between flower bud size and
microspore developmental stages changes with each peak of flowering. Early varieties seem
to show a faster development of pollen grains in smaller flower buds. Few complete
cytogenetic studies are available in cassava. These reports suggest that cassava contain 36
small and similarly shaped chromosomes, showing 18 bivalent pairing at meiosis. Although
three nucleolar chromosomes are noticed as for true diploids, duplication of some
chromosomes is also present suggesting segmental allotetraploidy. Previous work conducted
at CIAT indicated that it is possible to induce callus formation derived from in vitro culture of
isolated microspores. However, this result was not reproducible and no plant regeneration
was obtained, suggesting the need for a more systematic analysis.
Cassava is an out-crossing species prone to inbreeding depression. Therefore, it was relevant
to review the status of using haploid technology in breeding species with similar behavior.
An overview of the results with haploid induction and use with out-crossing species such as
carrots, poplars, ryegrass, timothe and orchard grass were presented by KVL University
(Denmark). For all these species there is strong dependency on genotype for response in
anther culture. For the diploid species like carrot and ryegrass, spontaneous chromosome
Project IP3: improving cassava for the developing world
Output 4-4
doubling frequently takes place during culture to create homozygous DH. This phenomenon
is less frequent among hexaploid timothe and tetraploid orchard grass. Doubled haploids of
these out-crossing species generally show weak growth and considerable sterility, but
homozygous plants with reasonable growth vigor and seed set can be found. Effects of
inbreeding depression are best studied in perennial ryegrass, where a small percentage of the
produced plants may be used for further breeding.
New approaches to develop highly efficient DH production for breeding of polyploids were
reviewed by the company Northwest Plant Breeding (NPB), U.S.A. One of the most recent
success cases is in wheat via induced microspore embryogenesis. (US Patent No. 6, 362,393;
C.S42: 682-692, 2002). The work reported a new chemical formulation for inducing
microspore androgenesis, along with results of practical experiments to induce wheat double
haploids. Their isolation and culture follow treatments of mid-to-late uninucleate
microspores with the chemical formulation on artificial media. Microspore androgenesis is
known to follow a common pathway in many plant species, switches microspore development
from gametogenesis to embryogenesis. Embryoids can be produced from up to 50% of
isolated microspores, potentially yielding thousands of double haploid plants.
Various microspore embryogenesis systems used in breeding (Brassica spp., tobacco, and
barley), as well as current approaches to develop such procedures for new species at
Wageningen University (The Netherlands) were described in detail. New cases were
documented for hot pepper, tulip, Delphinium, Anemone and Zantedeschia. In Brassica
napus, a new system for direct microspore embryogenesis mimicking zygotic embryogenesis
has been developed. This new system is an ideal biological tool to discern genes involved and
controlling embryogenesis in plants. Basic research on the molecular mechanism underlying
initiation of embryogenesis from microspores and on genes isolated so far was presented.
Recently, cDNA micro-array analysis for an extensive gene expression profiling during early
microspore embryo development from Brassica has provided a collection of new embryoexpressed cDNAs, which will be analyzed in future genomics research. Earlier differential
screening methods in the Brassica model led to isolation of the BABY BOOM (BBM) gene,
controlling direct embryogenesis induction from somatic tissues. An interesting approach to
test plants recalcitrant to microspore embryogenesis might be the use of the microsporespecific promoter NTM19 from tobacco to modulate transgenic expression of the BBM gene in
microspores in order to induce haploid plant production.
For cassava, it is recommended first to test the model systems already available as a quick
assessment to elucidate main factors to study in more detail. Finally, it was argued that the
high degree of heterozygosity in the out-crossing species cassava would not be a limitation
for the successful achievement of a microspore embryogenesis protocol in the crop. For
instance in tulip, which is like cassava a heterozygous and vegetatively propagated crop, a
highly efficient microspore embryogenesis system was developed. However, the DH plants
obtained suffered from inbreeding depression, which may also be expected for cassava. Using
inbred lines selected for increased tolerance to inbreeding depression might successfully
solve this temporary disadvantage.
Output 4-5
2003 Annual Report
4.1.4. Research Strategy for the Use of Inbreeding in Cassava Genetic Improvement
The participants in this brainstorm section are listed in Section 5.2.a. The genetic
improvement of cassava could greatly benefit from introducing inbreeding into the process.
However, because inbreeding would require about 9-10 years to attain acceptable levels of
homozygosity, an efficient protocol for production of DH is critical. Inbreeding in cassava is
desirable because: a) Selection among DH would not be affected by dominance effects which
could be exploited through reciprocal recurrent selection; b) additive effects among DH are
twice as large as in the current system; c) homozygous lines are genetically fixed, and
therefore, their genetic superiority (as progenitors) can be better exploited, than genetically
unstable heterozygous parents; d) germplasm exchange based on botanical seed is much
easier than that of vegetative cuttings; e) cleaning planting stocks from viral or other
pathogens could be achieved without the need of meristem culture; f) mutation breeding
would be more easily implemented; g) the identification of useful recessive mutants would be
greatly facilitated; h) the production of genetic stocks for basic and applied research would
now be feasible; i) for those projects exploiting polyploidy in cassava breeding, the availability
of haploids and DH is also desirable; j) the backcross breeding scheme could be implemented
for the transfer of useful genes from one cassava clone to another. But most of all, DH would
allow designing better performing hybrids rather than just finding them by trial and error as
is currently being done. All these advantages were recognized throughout the Needs
Assessment Study and the Planning Workshop.
Short term impact
The introduction of inbreeding in cassava offers several advantages. However, inbreeding is
likely to induce a drastic reduction of vigor in the first few cycles of selection. This
phenomenon is known as inbreeding depression. Maize was severely affected by inbreeding
depression when it was first subjected to inbreeding. However, inbred maize lines yielding as
much as 4 t/ha have now been developed. In other words, tolerance to inbreeding depression
can be built in crops. The first stage of this project will increase tolerance to inbreeding in
cassava to prepare the crop for the full homozygosity attained with the production of DH
lines. Below a brief description of this scheme is provided. The process will produce
interesting products at the end of the third year: a) improved versions of each elite clone
(reduced genetic load); b) identification of useful recessive traits unknown in cassava until
now; c) production of genetic stocks for basic and applied research and production. At the
end of the fifth year further products will be the identification of heterotic patterns and the
production of the first generation of cassava hybrid clones based on partially inbred parental
lines.
Project IP3: improving cassava for the developing world
Output 4-6
In the process of improving tolerance to inbreeding the following products will be obtained:
1. Useful recessive traits identified and shared among participating institutions (starting at
the end of year 2).
2. Inbred families with improved tolerance to inbreeding and, therefore, better breeding
values (end of year 3).
3. Genetic stocks generated, identified and shared among participating institutions (end of
year 3).
4. Preliminary identification of heterotic patterns among cassava lineages (end year 5). First
generation of elite hybrids derived from partially inbred parental lines (end of year 5).
Year
1
2
3
4
5
Activity
Plant 20 to 30 elite clones and selfpollinate each of them to produce 400500 botanical seeds.
Transplant 300 S0 plants/elite clone.
Select best 20% and self-pollinate.
Plant about 300 S1 plants/elite clone.
Select best 20% and self-pollinate.
Recombination of best S1 lines.
Plant about 300 S2 plants/elite clone.
Select best 20% and self-pollinate.
Recombination of best S2 lines. Test
crosses also performed
Evaluation of test
selected S2 lines
crosses
from
Outputs
● This stage produces botanical seed of
segregating material with an average of
50% homozygosity.
● Segregating material with an average of
75% homozygosity.
● Useful recessive traits may be identified.
● Segregating material with an average of
87.5% homozygosity.
● Improved version of each clone with
increased tolerance to inbreeding.
● Identification of genetic stocks with
useful (particularly recessive) traits.
● Segregating material with an average of
93.75% homozygosity.
● Improved version of each clone with
increased tolerance to inbreeding.
● Identification of genetic stocks with
useful (particularly recessive) traits.
● Heterotic patterns eventually identified.
● Useful hybrids identified for release as
clones to farmers.
Mid-term impact
The introduction of inbred parental lines in cassava breeding implies drastic changes in the
way cassava genetic improvement is approached. Below is a brief description of a breeding
scheme based on the use of DH. Although the process described is more complex than the
current system, it has the advantage of overcoming most of its problems and bottlenecks as
well.
Output 4-7
2003 Annual Report
All the advantages justifying inbreeding listed above will be fully operational in this proposed
scheme that could easily be changed to adapt the particular conditions and objectives of
different breeding projects. These changes would take place within 10 years.
Time*
Activity
Purpose – objective
0
Planting of elite cassava clones
To obtain flowers from superior clones.
10 Flowers from elite cassava The flowers contain the micro-spores from which
clones harvested.
DH will be obtained.
24 10 vitroplants obtained from Each DH line, obtained from the micro-spores, is
each of 30-200 DH per elite represented by 10 vitroplants.
clone**.
34 DH lines evaluated per se using Selection to exploit the large additive genetic
10 plants per line.
variance present among these DH lines.
52 Selected DH lines crossed and The hybrid seed produced represents the first
hybrid seed from these crosses generation of clones from inbred parental lines
obtained.
(DH).
62 F1 hybrids selected based on The same hybrids are evaluated using 10 and 10010-plant plots.
plant plots. This stage selects the best clones to be
and
identifies
the
best
hybrid
74 F1 hybrids selected based on released,
100-plant
plots.
Heterotic combinations (heterotic patterns) as well as best
patterns identified. Best DH parental DH.
lines identified.
68 Planting of nursery for next As the selection of hybrids (above) proceeds, a new
cycle of selection.
nursery is planted.
78 Cross the selected DH lines Crosses among selected DH lines, within each
within each heterotic group heterotic group, generate new genetic variability to
begins.
begin a new cycle of selection.
84 Seed
from
crosses
within Each cycle increases the heterotic effects.
heterotic group obtained.
0
Initiation of next cycle of selection leading to reciprocal recurrent selection or breeding
methods similar to those currently utilized in maize hybrid breeding.
* In months from the starting of each cycle of selection.
** Number of DH lines developed from each clone will vary in the first cycles of selection. Later on, as
the protocol is improved the number of DH will be more uniform.
4.1.5. Research Strategy for the Development of Haploid Technology in Cassava.
The discussion centered on identifying the main technical requirements to develop a
reproducible system for the production of DH in cassava, to outline the sequential major
steps required to achieve short term (3-5 years) and long-term (2003-2020) impacts, and to
set up a timeline for achieving these aims. The participants in this brainstorm section are
listed in Section 4.1.7.
Project IP3: improving cassava for the developing world
Output 4-8
The discussion reached the following general conclusions:
x There is a high probability to develop such a technology for cassava.
x Successful protocols have been developed for other species with high level of
heterozygosity as cassava.
x Although perhaps more difficult, protocols have been developed for other species with
inbreeding depression.
x The development of the protocol should start using elite varieties and then be adapted to
others, as tolerance to inbreeding is improved.
As general principles it was decided to:
x Generate an initial tentative list of 25-30 genotypes for developing the cassava haploid
technology.
x Select the potential materials in conjunction with the breeders.
x Request assistance of breeders to obtain information on pollen production in different
environmental conditions, relationship between vigor and pollen production, association
between flowering and root formation, crop management for vigorous growth and
flowering, materials with broad environmental adaptation, identification of different
diversity groups within cassava (i.e. based phenotypic and molecular markers
characterization), include as much as possible profuse flowering materials.
x Include in the list of potential materials for developing the protocol, the genotypes used
for genome mapping in order to facilitate the future use of DH for molecular marker
analysis.
x Test the different generations of partially inbred lines developed by the breeding
component of the project, with more tolerance to inbreeding depression, as they become
available, in order to evaluate their response in vitro and aid the development of breeding
populations.
The discussion of the technical aspects for the development of the in vitro protocol and a
tentative timetable and work-plan are presented below:
Isolated microspore vs. anther culture systems.
It was recommended that an isolated microspore system be developed for cassava.
Nowadays, anther culture protocols are generally less efficient.
In establishing the research steps, first efforts will be centered on isolated microspore
culture, and if unsuccessful, a pollen shed based system using anther culture will be tested.
Current status of knowledge.
Knowledge concerning cassava microspore production and development was discussed.
Highlights included:
x Only about 100 microspores produced per anther = 1000 per flower bud. This figure
needs to be checked. If this low amount of pollen grains is corroborated, it will be an
Output 4-9
2003 Annual Report
x
x
x
x
important bottleneck for isolating large number of microspores required for isolated
microspore culture in such a case, the pollen shed system should be considered.
Large numbers of flower buds less than 1 mm in diameter must be obtained if sufficient
uninucleate microspores are to be obtained.
The above calculations mean that approximately 40 buds required to produce 1 ml of
microspore suspension.
Previous research at CIAT has identified some aspects of the culture conditions e.g.
osmolarity required to keep microspores alive.
Northwest Plant Breeding (NPB) has achieved indications of cell division in preliminary
studies of microspore culture in cassava (results were not presented).
Status of the mother plants.
Obtaining enough flower buds at the correct developmental stage to allow experimentation
and development of a DH system was recognized as a critical component of any such project
in cassava.
x There is a need to identify prolific flowering varieties.
x Varieties that produce compact inflorescence are considered superior to others due to
more synchronous development of the buds.
x First, second and third flowering peaks may differ in responsiveness e.g. in the first cycle
many flowers are known to abort.
x Time of year when flowering occurs may have importance.
x Soil fertility may affect flower production and development. Recommendations of fertilizer
applications may be beneficial e.g. high P.
x Due to the above and the need to generate large numbers of flowers, production of
sufficient flowers under greenhouse conditions was considered unfeasible, thus CIAT was
selected as the best location for flower production.
x Requirements for virus indexing to ensure that cassava mosaic disease is not imported to
the Americas requires two years testing, thus any African varieties to be screened in the
proposed project best obtained from African accessions already in the CIAT collection.
Technical aspects of starting microspore systems in a new species
x Effects of genotype.
x Culture systems – basal media, osmolarity.
x Cold and heat treatments and starvation.
x Effect of chemical inducers- e.g. NPB have their MEg•.
x Potential use of ovaries as inducers as used in wheat and some cereals.
x Possibility of spraying flower buds with inducing agents while still on the donor plant.
x Need to slow down microspore development and ideally to synchronize this to allow
shipment and more effective induction of embryogenesis.
Use of established DH systems to test responsiveness in cassava.
It was recommended by those who have hands on experience with isolated microspore
systems to apply known protocols to cassava, especially in the early stages of the project, to
determine the relative responsiveness of the crop. These include the protocols for: rape,
tobacco, barley and wheat.
Project IP3: improving cassava for the developing world
Output 4-10
Table 4.1. Outline of five-year plan to develop and apply double haploid technologies to
Cassava
Activity
Protocol for isolating high
yielding and clean microspore
suspension
Selection of best 3-5
experimental genotypes out of
the initial 25-30.
Develop culture system to
induce microspore cell division
Attain capacity for embryo
production
Optimize culture protocol
Apply protocol to a broader
number of genotypes
(including Sn lines)
Establish electronic
communication systems
Collaborators meetings
Year 1
End first
year
Year 2
Year 3
End third
year
Year 4
Year 5
End fifth
year
Training
i. visiting scientists
ii. graduate students
iii.workshops
International shipping of cassava flower buds.
While the vast majority of plant flower production would take place at CIAT, significant
culture work and DH induction would also take place in the partners’ laboratories. It will
thus be necessary to ship immature flowers to these Northern locations. Initial attempts
between CIAT and NPB in Washington State, USA have shown that sterilization of field grown
material and successful international shipment is possible. Healthy materials can, therefore,
be shared between collaborating institutions in the tropical and temperate regions.
Communication systems
In order to facilitate efficient exchange of materials, ideas and results between collaborating
institutions in the South and North, it was agreed that dedicated electronic communications
be established between the institutions involved. This would include Internet based “notice
boards”, teleconferencing and occasional meetings.
Output 4-11
2003 Annual Report
Table 4.2. Detailed activities and timetable for the first three years in the development of a
Double Haploid protocol in cassava
Months 0-3
2.
1.
x
x
x
x
Develop system for:
Shipping clean bud material.
Isolation of microspores.
Separation of microspores.
Maintaining viable microspores in
culture.
Select best 25 –30 potential genotypes based on:
x Flowering characteristics: early, intermediate
and late varieties.
x Diversity: broad environmental adaptation
(Africa, Asia, Latin America, alkaline
vs. acid soils, molecular characterization).
x Cytology: slow maturation process from
uninucleate to mature pollen grain.
x Synchrony of microspore development.
x Inputs from breeders.
Consideration of mapping populations
Above to be carried out using one
genotype selected from the tentative
list of 25-30 genotypes based on its
prolific flowering.
Goal: identification of best experimental materials.
Months 3-12
Select best five genotypes from above based on their flowering capacity and performance of
microspores in culture:
x Send buds of five genotypes to collaborating institutions.
x Screen for microspore performance using established DH systems in rape, tobacco,
barley and wheat.
x Identify best medium and culture conditions (preliminary).
Goal: Achieve cell division from cassava microspores.
Months 12-24
Adapt system for best 1-2 genotypes by studying:
x Pre-culture treatments.
x Temperature.
x Chemical inducers.
x Sub-culture regime.
x Plant regeneration from somatic embryos.
x Determine ploidy level of plants.
x Confirm genetic nature of DH tissues obtained.
Goal: Achieve production of double haploid embryos and embryogenic tissues.
Months 24-36
Optimize culture system for routine production of DH cassava plants
x Perfect systems for recovery of DH plants from embryos
x Optimize system for routine application to model genotype(s)
x Transfer DH capacity to breeder identified important genotypes from America, Asia
and Africa
x Develop and apply DH system for Sn lines
Goal: Ability to apply DH systems to a range of relevant cassava genotypes.
Project IP3: improving cassava for the developing world
Output 4-12
4.1.6 Conclusions
There was a general consensus that cassava is and will be increasingly used by the industry.
Policies in Africa, Asia and Latin America aim at support the industrial uses of cassava to
promote rural development. However, there was also a clear agreement that cassava needs a
more efficient breeding approach to be able to compete with other sources of raw material for
the different industrial processes where it currently has a marginal competitive advantage.
The Needs Assessment Study and the Planning Workshop supported the advantages and
convenience of introducing inbreeding in cassava through the production of DH. Two main
bottlenecks were identified: i) Problems related to strong inbreeding depression in cassava;
and ii) lack of a protocol for the production of DH lines in this crop.
Cassava exposes inbreeding depression, as it once did maize. Participants agreed that
tolerance to inbreeding could be built in cassava populations through traditional breeding
approaches. A S2 recurrent selection approach was suggested as the best alternative for
overcoming this potential problem of inbreeding depression. The process will require as many
as five years for completion, but outputs will be available starting in the third year. This
process will generate populations tolerant to inbreeding which, by definition, have better
breeding values and should produce better clonal progenies for the benefit of cassava
farmers. This process will also lead to the identification of currently unknown useful
recessive traits and the generation of genetic stocks to be shared with the institutions
working with cassava. Finally the process will eventually identify heterotic patterns in
cassava that will further facilitate the production of better clones than those currently
available.
In relation to the development of a protocol for the production of DH in cassava, it was
generally agreed that the technology could be developed within a five years period, within the
first tree years for a restricted group of elite clones and in the last two years for a broader
range of cassava germplasm. Guidelines regarding the selection of initial germplasm to work
with, and a stepwise process for the development of the protocol were listed. This activity
serves as an innovative step towards the merging of traditional breeding and biotechnology
for a crop that has been frequently neglected and which seldom had fluent access to new
technologies. The DH approach will radically change the way cassava breeding is conducted.
In addition strong collaborative links will be developed not only between research institutions
from developed and developing countries but also among developing countries in Africa, Asia
and Latin America.
Output 4-13
2003 Annual Report
4.1.7 List of participants
Needs Assessment Study
Institutions
Central Tuber Crops Research Institute (CTCRI) Thiruvananthapuram, Kerala, India
Council for Scientific and Industrial Research, Ghana
Crop Research Institute (CRI), Ghana
Empresa Brasileira de Pesquisa Agropecuária (EMBRAPA)–CNPMF, Brazil
International Center for Tropical Agriculture (CIAT), Colombia
Institute of Agricultural Science of South Vietnam (IAS), Vietnam
Instituto de Investigación de Viandas Tropicales (INIVIT), Cuba
Kasetsart University, Bangkok, Thailand
Kasetsart University Kampangsan Campus, Nakornprathom, Thailand
Khon Kaen Field Crops Research Center, Thailand
National Agricultural and Animal Research Institute (NAARI), Uganda
Plant Genetic Resources Centre, Ghana
Rayong Field Crops Research Center, Thailand
Thai Tapioca Development Institute, Korat, Thailand
Wenchi Farm Institute, Ghana
Cassava farmers in Brazil, Ghana, India, Uganda, and Thailand
Individuals who collaborated with the Needs Assessment Study
A.Alves, Brazil
M.Anantharaman, India
S.K.Nanda, India
A.Bua, Uganda
M.Fregene, Colombia
S.N.Moorthy, India
A.Vanavichit, Thailand
M.Folgueras M., Cuba
S.Tragoonra, Thailand
A.Limsila, Thailand
M.Opolot, Uganda
S.Ramanathan, India
B.Nambisan, India
M.S.Palaniswami, India
Staff at Pokuase Station,
C.Pons Pérez, Cuba
M.T.Sree Kumari, India
Ghana
C.S.Easwari Amma, India M.Unnikrishnan, India
Staff of Tuber Crops
C.Rojanaridpiched, Thail. O. Molina C., Cuba
Program, Kumasi, Ghana
D.L.González H., Cuba
P.Sarawat, Thailand
S.Tumwesigye, Uganda
D.H.Reinhardt, Brazil
P.Van Bien, Vietnam
S.Charoenrath, Thailand
E.Okay, Ghana
P.Padmaja, India
T.N.Ngoan, Vietrnam
G.A.Mensah, Ghana
P.Wongtiem, Thailand
T.V.R.Naya, India
G.Semakula, Uganda
R.Howeler, Thailand
V.Medero V., Cuba
H.Ceballos, Colombia
R.A. Kirkby, Uganda
W.M.G. Fukuda, Brazil
H.Gomez, Colombia
R.R.Nair, India
W.Watananonta, Thailand
H.Kim, Vietnam
R.Best, Uganda
W.Kositratana, Thailand
J.J.Wang, Vietnam
S.V.Pillai, India
W.S.Sserubombwe,
J.L.Loyola D., Brazil
S.Edison, India
Uganda
K.Abraham, India
S.Rodríguez M., Cuba
Yona Baguma, Uganda
M. García G., Cuba
S.Pillai, India
Individuals participating in the Planning Workshop.
Breeding subgroup
Project IP3: improving cassava for the developing world
Output 4-14
1.
2.
3.
4.
Alfred Dixon. IITA. Nigeria.
Alfredo Alves. EMBRAPA – CNPMF (Brazil, Bahia State) - CBN Coordinator. Brazil.
Carlos Iglesias. Weaver Popcorn Company Hybrid Research. U.S.A
Egesi Chiedozie Ngozi. Cassava Programme, National Root Crops Research Institute.
Umuahia, Abia State. Nigeria.
5. Hernán Ceballos. CIAT. Colombia.
6. Hoang Kim. Institute of Agricultural Science of South Vietnam (IAS), Ho Chi Minh City.
Vietnam.
7. José Branco de Miranda Filho. Escola Superior de Agricultura "Luis de Queiroz" (ESALQ).
Universidad de São Paulo. Brazil.
8. Jonathan Mkumbira. Bvumbwe Agricultural Research Station, Limbe. Malawi.
9. Teresa Lozada. Instituto Agronomico Campinhas. Brazil
10. Wania Maria Gonçalves Fukuda. Embrapa - Mandioca e Fruticultura. Cruz das AlmasBA. Brazil.
Tissue Culture / Biotechnology subgroup
1. Cal Konzac. Northwest Plant Breeding. Pullman, Washington State. U.S.A.
2. Gary Toenniessen. Rockefeller Foundation. U.S.A.
3. Hankoua Bertrand Bachaumond. Cameroon.
4. Jan Custers. Wageningen University and Research Centre. The Netherlands
5. Krit Raemakers, Wageningen University and Research Centre. Wageningen. The
Netherlands.
6. Martin Fregene. CIAT. Colombia.
7. Nygel Taylor. Danforth Center. St. Louis, Missouri.
8. Peng Zhang. Institute for Plant Sciences, ETH-Zentrum, Zurich, Switzerland. (also
representing China).
9. Rafiullah Sahibzada. Northwest Plant Breeding. Pullman, Washington State. U.S.A.
10. Sven Bode Andersen. Royal Veterinary & Agricultural University . KVL, Copenhagen.
Denmark
11. Tatsuo Fujimura. United Nations Development Programme. New York. U.S.A.
12. Tim Loughney. Northwest Plant Breeding. Pullman, Washington State. U.S.A.
13. Zaida Lentini. CIAT. Colombia.
4.1.8 Acknowledgement
The participants of the Needs Assessment Study and the Planning Workshop would like to
express their appreciation to United Nations Development Programme and to the Rockefeller
Foundation for their support in this initiative. The personal commitment of Drs. Tatsuo
Fujimura and Gary Toenniessen was fundamental for the success of these events. The
organizers of the event would also like to thank the participants of the Needs Assessment
Study and the Planning Workshop, as well as their home institutions and personnel, for their
enthusiasm and unconditional support. The efforts by CIAT’s regional office in Uganda and
Thailand were fundamental for the logistical success of the Needs Assessment Study. Finally,
Visitors Office at CIAT-Palmira made a superb work organizing traveling schedules, the
issuing of visas and arranging for accommodations and the meeting facilities. Thank you all
for your commitment.
Output 4-15
2003 Annual Report
4.2 Development of a novel approach for the analysis of diallel mating designs for
better understanding the inheritance of quantitative traits (also in SB2 report).
As stated in the report from the previous year, three diallel mating designs have been
evaluated for the three main target environments (sub-humid environments, acid soil
savannas and mid-altitude valleys). Diallel trials were planted in 2001 and in the northern
coast and acid soil savannas also in 2002. The relevance of these studies certainly lies on
the information they will produce. However, it is important to emphasize that these are diallel
studies that offer new alternatives to extract information and requires development of new
models for their analysis. In this section a brief description of this innovative approach will
be described. This research will serve as thesis work for Ph.D students from Vietnam and
Uganda (a female and a male, respectively), as well as for a M.S. thesis for one of the
assistants working in the project.
4.2.1 Theoretical model for analyzing inter and intra-family variation.
The most important difference between these diallels and the traditional ones is the fact that
thirty clones were used to represent each cross, and individual measurements were made for
each individual genotype making up the full-sib cross. Most diallels will harvest all the
individual genotypes for each cross ignoring the within-family variation. This is because the
main target is identifying good families or parents. However in the case of cassava breeding
the individual genotype (clone) is the central point of attention whereas in other crops such
as maize, the main target is the family or F1-cross. It is important to emphasize the relevance
of this distinction because the current method of selection (particularly in the first stage of
selection, called Clonal Evaluation Trials) the cassava breeding project is trying to develop a
more systematic evaluation process in which we select on the family basis (all the clones
derived from a given parental line), and then the best clones from that family (the individual
clone that will eventually be released 8-10 years later).
One of the several questions these diallels aim at answering is what is the relative genetic
variability between and within families. This is obviously fundamental for a more scientific
process of selection. In other words, should we focus in developing good families of clones or
should we concentrate on developing families of clones that have large genetic variability to
be exploited?
Table 4.3 illustrates the progress made in developing a quantitative genetic model for the
analyses of the diallels. The information presented was exposed to Dr. José Miranda Branco
Filho who is an eminent quantitative geneticist who co-authored with Arnell Hallauer the
most widely used book on the subject (Quantitative Genetics in Maize Breeding, Iowa State
University Press). The issue of the model and alternative ways of using the information is still
underway and there are excellent perspectives that these analyses will in fact be a relevant
contribution in the area of quantitative genetics. That is, the studies will hopefully not only
be useful for cassava but also for other crops as well.
Project IP3: improving cassava for the developing world
Output 4-16
Table 4.3. Theoretical model for the quantitative genetic analysis of diallel crosses in cassava. In addition to the
usual variation among F1 crosses, within family variation has been included. They were considered fixed and
random genetic effects, respectively.
Source of variation
Locations (L)
Rep/locations
Among F1 crosses
GCA
SCA
Within F1 crosses
Among F1 crosses x L
GCA x L
SCA x L
Error (a)
Within F1 crosses x L
Error (b)
Degrees of
freedom
a-1
a(r-1)
[p(p-1)/2]-1
p-1
p(p-3)/2
(p(p-1)/2)(g-1)
(a-1)([p(p-1)/2]-1)
(a-1)(p-1)
(a-1)(p(p-3)/2)
a([p(p-1)/2]-1)(r-1)
(p(p-1)/2)(g-1)(a-1)
a(p(p-1)/2)(g-1)(r-1)
Mean squares expectations
S2e
S2e
S2e
S2e
S2e
S2e
S2e
S2e
S2e
S2e
+
+
+
+
+
+
+
+
+
gS2H + grS2Among F1 x L + graS2Among F1
gS2H + rS2SCA x L + r(p-2) S2GCA x L + raS2SCA + ra(p-2) S2GCA
gS2H + rS2SCA x L + raS2SCA
rS2 Within F1 x L + raS2 Within F1
gS2H + grS2Among F1 x L
gS2H + rS2SCA x L + r(p-2) S2GCA x L
gS2H + rS2SCA x L
gS2H
rS2Within F1 x L
Where:
a = number of locations
r = number of replications within location
p = number of progenitors in the diallel
g = number of clones within each F1 cross
S2Among
S2Within
F1
= Variation among averages for each F1
cross
S2
GCA
F1
= Variation among clones within F1
crosses
= general combining ability
S2H = experimental error type a
S2SCA = specific combining ability
S2e = experimental error type b
S2........ x L = interaction of each effect with the environment
Output 4-17
2003 Annual Report
Selections for the Mid-altitude valleys
Tables 4.4 and 4.5 present the results of the combined analysis of variance for the two diallel
evaluations conducted in the mid altitude valleys environment and the values for GCA
effects, respectively. One striking result in this analysis is the significanse of SCA effects,
which are related to the dominance of heterosis present in the hybrid clones evaluated in the
studies.
Table 4.4. Combined analysis of variance for a diallel study conducted in the mid altitude
valleys at Palmira and Jamundí (Valle del Cauca), Colombia. Mean squares for yield
(kg / plant), Harvest index and dry matter content are presented only for the among
crosses component of the study.
Source of
variation
Locations (L)
Rep/ L
Among F1s
GCA
SCA
Among F1s x L
GCA x L
SCA x L
Error
df
1
4
35
8
27
35
8
27
140
Fresh root yield
(kg / pl )
3.949
17.579
3.125
5.760
2.461
0.645
1.392
0.424
0.447
ns
**
*
**
ns
**
ns
Mean Squares
Harvest Index
(0-1)
0.496
0.028
0.009
0.028
0.004
0.001
0.002
0.001
0.001
*
**
**
**
**
**
*
Dry matter
(%)
563.608
19.009
9.685
26.855
4.597
3.200
11.304
0.799
0.899
**
**
ns
**
**
**
ns
Table 4.5. General combining ability (GCA) effects from the diallel study in two mid-altitude
valley environments. Nine progenitors were involved in the diallel design.
Clon
CM 6740-7
SM 1219-9
SM 1278-2
SM 1636-24
SM 1673-10
SM 1741-1
HMC 1
M ECU 72
MPER 183
St.Dev. Gi
St. Dev. (GI – Gj)
Fresh root yield
(kg/pl)
0.003
0.341
-0.426
-0.314
-0.308
0.052
-0.313
0.341
0.624
0.172
0.257
Harvest Index
(0-1)
-0.009
0.024
0.010
-0.022
0.007
0.037
0.015
-0.048
-0.014
0.0060
0.0097
Dry matter content
(%)
0.608
-0.572
0.914
-0.464
0.641
1.069
-0.444
-1.079
-0.672
0.489
0.734
In Table 4.5 a more precise estimate for GCA effects that the one presented previously in
2002 is presented. The information provided is a good example of the complexities involved in
Project IP3: improving cassava for the developing world
Output 4-18
cassava breeding. The parent generating the best yielding progenies was MPER 183 (the
highest positive GCA value) but it showed very deficient performances regarding Harvest
Index and Dry Matter Content (both with negative GCA estimates). On the other hand CM
6740-7 has an excellent GCA estimate for dry matter but intermediate performance regarding
fresh root yield and harvest index. The clone SM 1741-1 has a good overall performance as a
parent, particularly in relation to dry matter content.
Selections for the Sub-Humid Tropical Environment
Following the same criteria for the previous section Tables 4.6 and 4.7 present the relevant
results from the diallel studies for the sub-humid environments. There are some interesting
differences between the two data sets. The most striking one is that here SCA effects did not
reach statistical significance (Table 4.6). This may be the result of the strong dry spell in this
environment, which has a strong effect on dry matter content. For the other traits both GCA
and SCA showed highly significant statistical differences.
Table 4.6 Combined analysis of variance for a diallel study conducted in the sub-humid
environments at Pitalito and Santo Tomás (Atlántico), Colombia. Mean squares for
yield (kg / plant), Harvest index and dry matter content are presented only for the
among crosses component of the study.
Source of
variation
Locations (L)
Rep/ L
Among F1s
GCA
SCA
Among F1s x L
GCA x L
SCA x L
Error
df
1
4
35
8
27
35
8
27
140
Fresh root yield
(kg / pl )
3.341
0.732
1.140
2.976
0.596
0.366
0.797
0.238
0.157
**
**
**
**
**
Mean Squares
Harvest Index
(0-1)
0.258
0.022
0.009
0.021
0.006
0.003
0.008
0.002
0.001
*
**
*
**
**
**
*
Dry matter
(%)
118.418
21.699
6.744
20.974
2.528
2.356
5.642
1.383
0.677
**
**
ns
**
**
**
Table 4.7 summarizes the GCA effects for each of the nine parents involved in this diallel. As
in the previous cases, it is obvious that there is no perfect parental line. Good GCA for yield
productivity is generally associated with a negative value for dry matter content and/or
harvest index, and vice versa. It was disappointing to see the poor performance of SM 156517 regarding dry matter content (GCA = - 1.467). One of the promising clones for this
environment was SM 1411-5, whose breeding value based on the results in Table 4.7, is
clearly highlighted.
Output 4-19
2003 Annual Report
Table 4.7. General combining ability (GCA) effects from the diallel study in two sub-humid
environments. Nine progenitors were involved in the diallel design.
Clon
MTAI 8
CM 6754 - 8
CM 8027 - 3
SM 805- 15
SM 1565- 17
SM 1411- 5
SM 1219- 9
SM 1657- 12
SM 1665- 2
St.Dev. Gi
St. Dev. (GI – Gj)
Fresh root yield
(kg/pl)
0.004
-0.372
-0.126
-0.438
0.267
0.139
0.220
0.020
0.286
0.130
0.195
Harvest Index
(0-1)
-0.011
-0.001
-0.005
-0.032
0.039
-0.019
-0.010
0.010
0.028
0.013
0.019
Dry matter content
(%)
0.053
0.241
0.858
0.055
-1.467
0.919
-0.032
-0.280
-0.347
0.345
0.518
Selections for the Acid-Soil Savannas Environment
Results from the diallels at CORPOICA – La Libertad for the acid soils environments (Table
4.8) indicate a strong interaction with the environment. In fact this makes sense because the
two locations had sharp contrast regarding their soils. One had more stressful savanna
conditions and the other had alluvial soils, which support excellent yields and not severe
disease development. The large interactions with the environment (error term for the main
genetic effects) resulted in some of them not being statistically significant. Individual location
analysis would yield highly significant statistical differences.
Table 4.8. Combined analysis of variance for a diallel study conducted in the acid soil
savannas at CORPOICA La Libertad (Meta), Colombia. Mean squares for yield (kg /
plant), Harvest index and dry matter content are presented only for the among crosses
component of the study.
Source of
Locations (L)
Rep/ L
Among F1s
GCA
SCA
Among F1s x L
GCA x L
SCA x L
Error
df
1
4
44
9
35
44
9
35
176
Fresh root yield
172.011
3.743
0.631
1.462
0.417
0.476
1.244
0.279
0.189
Project IP3: improving cassava for the developing world
**
**
**
Mean Squares
Harves Index
0.263
0.060
0.015
0.048
0.007
0.008
0.023
0.005
0.003
*
**
**
Dry matter
16.988
0.551
0.076
0.190
0.047
0.042
0.102
0.026
0.019
**
*
*
**
**
Output 4-20
Table 4.9. General combining ability (GCA) effects from the diallel study in the acid soil
savannas environments. Ten progenitors were involved in the diallel design.
Clon
CM 4574 – 7
CM 6740 – 7
CM 7033 – 3
SM 1219 – 9
SM 1565 – 15
SM 2058 – 2
SM 2219 – 11
HMC 1
MPER 183
MTAI 8
St.Dev. Gi
St. Dev. (GI – Gj)
Fresh root yield
(kg/pl)
0.184
0.076
-0.110
0.089
-0.073
0.078
0.276
-0.127
-0.326
-0.067
-0.067
0.216
Harvest Index
(0-1)
-0.001
-0.008
-0.006
0.029
-0.004
0.006
0.041
0.019
-0.077
0.000
0.000
0.029
Dry matter content
(%)
1.106
0.182
-0.348
0.676
1.434
-0.301
0.593
-0.032
-3.422
0.114
0.114
0.062
As expected CM 4574-7 had a good performance as a parent with positive GCA for yield and,
particularly, for dry matter content (Table 4.9). However, better that CM 4574-7 was SM
2219-11 with positive GCA values for the three variables evaluated.
4.3 An alternative way to use the information from diallel studies (also in SB2 report).
Below several examples are given on the way the additional information based on the within
F1 family variation could provide. The model is taking advantage of the fact that a single
genotype can be vegetatively propagated. Therefore each genotype (clone) was planted in two
(mid-altitude valleys) or three (acid soil savannas and sub-humid environments) locations,
with three replications in each location. An error exists for the within family variation which
is not ordinarily available in diallel studies. In other words, it is possible to estimate the
genetic component of the within family variation. That information can be contrasted with
the one obtained from the among-F1 families variation and if differences are large, then
circumstantial evidence of the occurrence of strong epistatic effects can be produced.
Moreover, individual analysis of each variable can help identifying interesting events that
otherwise would go unnoticed. For instance, averages for each F1 family and standard
deviations (within each F1 family) can be plotted. This information can be very useful and as
far as it is known the kind of analysis has never been reported in the literature. The basic
idea is that parental lines that have many dominant alleles affecting a given trait will yield
more homogeneous progenies than a progenitor with recessive alleles. This is the principle in
which the graphic approach for diallel analysis (Hayman 1954a; 1954b; 1958; Mather and
Jinks, 1971).
Output 4-21
2003 Annual Report
The first example turns around the reaction to white flies evaluated in the diallel experiment
evaluated in Jamundí (Valle del Cauca, Colombia). Figure 4.1 shows the result for the
averages of the F1 crosses as well as the variation (standard deviation) among the 30 clones
that made up each F1 cross.
1.3
Standard deviations for within family variation
1.2
MECU72 x SM 1278-2
1.1
1.0
MECU72 x SM 1671-1
MECU72 x SM 1673-10
0.9
MECU72 x HMC1
0.8
MECU72 x CM6740-7
MECU72 x SM1636-24
0.7
MECU72 x MPER 183
0.6
MECU72 x CM1219-9
0.5
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Family averages for reaction to white flies (1=resistant; 5=susceptible)
Figure 4.1. Averages and standard deviations for the reaction to white flies in a diallel study
evaluated in Jamundí, Valle del Cauca, Colombia. Data points for the 8 different
progenies with MECU72 as common parent are identified.
In Figure 4.1 the variation in the response to white flies attacks is clearly illustrated through
the values along the horizontal axis. Lower scores meant more resistant phenotypes. Higher
scores indicate susceptible reaction. It was very interesting to see the location of the
progenies from MECU 72 who has been shown to have resistance (antibiosis) against the
white flies. As expected most of the progenies from MECU 72 where at the left of the figure,
indicating a resistant reaction to the insect. This is as much as a traditional diallel analysis
could go. What is interesting and innovative is the possibility to analyze the variation around
the standard deviations within families. Some crosses had very small standard deviations
(MECU72 x CM1219-9) indicating high uniformity within family. Other progenies (MECU72 x
SM1278-2) have a much higher within family variation. This contrast is extremely interesting
for understanding the genetic structure for relevant traits.
Project IP3: improving cassava for the developing world
Output 4-22
Standard deviation for within family variation
1.80
1.60
Susceptible reaction
Recessive behavior
1.40
1.20
1.00
0.80
Resistant reaction
Dominant behavior
0.60
0.40
1.00
1.50
2.00
2.50
3.00
3.50
Family averages for thrip score (1=excellent; 5=susceptible)
Standard deviation within family variation
1.8
1.6
Susceptible
Recessive behavior
1.4
1.2
1.0
Resistant
Dominant behavior
0.8
0.6
0.4
1.2
1.7
2.2
2.7
3.2
Family averages for thrip score (1=resistant; 5=susceptible
5=susceptible)
Figure 4.2. Averages and standard deviations from a diallel study at Santo Tomás (above) and
Pitalito (below) in the Altántico Department for the reaction to thrips (1=resistant;
5=susceptible). Dominant and recessive “behavior” does not refer to the traditional
meaning for dominance and recessiveness. Dominance refers either to a trait where
non-additive effects are important or cases where the parental lines may have loci at
the homozygous dominant status.
Output 4-23
2003 Annual Report
Continuing with reactions to pests, Figure 4.2 illustrates the results from the diallels
targeting the sub-humid conditions (different progenitors that those for mid-altitude valleys
and acid soil savannas). These results are typical of the trends that can be obtained by
putting together the averages for each family and the standard deviation within families.
First, there is a clear consistency between the results of the two locations, with an upward
trend at the right of the figures. The results agree with the expectations, clones with lower
averages (resistant to mites) show a more uniform progenies (smaller standard deviations)
than the susceptible ones at the right of the figures. The tendencies observed in these plots
could be explained as an statistical artifact: families with larger values are likely to have
larger standard deviations. However, as will be shown below, for other traits the opposite is
observed.
In Figure 4.3 the plots for dry matter content (%) from one of the locations for the diallel for
the sub-humid (Santo Tomás, Atlántico Department) and acid soil savannas environment
(CORPOICA-La Libertad, Meta Department) are presented. Results for the other respective
locations yielded similar results.
In this case families tended to be more uniform with higher averages. In other words the
magnitude of the averages do not necessarily dictate the magnitude of the standard
deviations within each family. Families with high dry matter content tended to be clearly
more uniform. This could be envisioned, perhaps, as a result of the continuous selection for
increased dry matter content. Most of the germplasm would have the key alleles for high dry
matter, the exception being the opposite. When a progeny has an unusually low dry matter
content, then the variability within family tends to increase.
It is interesting to note that similar trends were observed for this variable in the three
different ecosystems and for all the locations in which they were evaluated. There is a
consistency in the way each variable appear in this kind of graphs, which is further suggest
that they may be an interesting new approach for analyzing genetic variability and identifying
relevant cases where further studies are justified.
Harvest index is another variable that consistently showed the same tendency as dry matter
content, with reduced variability within family at higher levels of the index. As in the case of
dry matter content, the long selection process for optimum harvest index (Kawano, 2003).
This process could serve to explain why high harvest indices result in reduced variability. In
other words, high harvest index is the rule at least among the elite parental lines used in
producing the diallel families.
Finally root yield showed a different tendency with increased variability within family in those
F1 crosses that showed the highest averages. As in the case of the previous variables, this
trend was shared by the three different ecosystems targeted and for each of the locations
within each ecosystem. Figure 4.4 illustrates the results for root yield (kg/plant) for one of
the locations at each target environment. It is clear that a common trend exists; with families
with low productivity showing a “dominance behavior” that does not necessarily mean that
root yield is a recessive trait.
Project IP3: improving cassava for the developing world
Output 4-24
5.50
Standard deviation for within family variation
5.00
4.50
4.00
Low dry matter
Additive behaviour
3.50
3.00
2.50
High dry matter
Dominant behaviour
2.00
22.00
23.00
24.00
25.00
26.00
27.00
28.00
29.00
30.00
Average dry matter content
Standar deviations for within family variation
6.5
6.0
5.5
Low dry matter
Recessive
behavior
5.0
4.5
4.0
3.5
Highdry matter
Recessive
behavior
3.0
2.5
25.0
26.0
27.0
28.0
29.0
30.0
31.0
32.0
33.0
34.0
35.0
Family averages for dry matter content (%)
Figure 4.3. Averages and standard deviations from a diallel study at Santo Tomás (above) and
Acid Soil Savannas (below) in the Altántico and Meta Departments, respectively.
Output 4-25
2003 Annual Report
3.5
High yield
Recessive behavior
Standar deviation for within family variation
3.0
2.5
2.0
High yield
Dominant behavior
1.5
Low yield
Dominant behavior
1.0
2.5
3.0
3.5
4.0
4.5
5.0
Family averages for fresh root yield (kg/plant)
2.60
Standar deviations for within family variation
High yield
Recessive behavior
2.10
High yield
Intermediate
behavior
1.60
1.10
Low yield
Dominant behavior
0.60
1.75
2.25
2.75
3.25
3.75
Family averages for fresh rot yield (kg / plant)
4.5
High yield
Recessive behavior
Standard deviations for within family variation
4.0
3.5
3.0
2.5
Low yield
Dominant behavior
High yield
Dominant behavior
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
Family averages for fresh root yield (kg / plant)
Figure 4.4. Averages and standard deviations from a dialllel study at Pitalito (above),
CORPOICA La Libertad (middle) and Palmira (below) in the sub-humid, acid soils and
mid altitude valleys, respectively.
Project IP3: improving cassava for the developing world
Output 4-26
It is clear from the information provided in Figures 4.3 and 4.4 that interesting differences
are found among and within the families involved in the diallel studies. It is also evident the
consistency with which each variable responds to the analysis. Further study will be
conducted by determining the identity of each family in the plots (as done for white flies).
Concluding remarks regarding diallel analysis.
Our knowledge on the inheritance of agronomically relevant traits in cassava is still very
limited. The most interesting result of the analyses on diallels described above can be
summarized as follows:
a. Innovative approach from the quantitative genetics point of view. The studies will
hopefully provide advances in this area of research with a more complete model for
explaining genetic variability, model that can be related to actual data. It is
interesting, therefore, that cassava may eventually come to provide tools for genetic
analyses of other vegetatively propagated crops.
b. The graphic deployment of data facilitates the identification of contrasting families
deserving a more detailed study. That is the case, for example, illustrated in Figure 4.1
among the progenies from MECU 72.
c. In general the statistical and graphic analysis identified an ideal group of genotypes
that can be used for molecular analysis. This is for instance, what CIAT is currently
doing with a group of genotypes that showed a contrasting phenotype regarding leaf
retention.
References:
Hayman, B.I. (1954a). The analysis of variance of diallel tables. Biometrics 10:235-244.
Hayman, B.I. (1954b). The theory and analysis of diallel crosses. Genetics 39:789-809.
Hayman, B.I. (1958). The theory and analysis of diallel crosses: II. Genetics 43:63-85.
Kawano, K. (2003). Thirty years of cassava breeding for productivity-Biological and social
factors of success. Crop Sci. 43:1325-1335.
Mather, K. and J.L. Jinks. 1971. Biometrical Genetics. Cornell University Press, Ithaca, NY.
USA.
4.4 . Introduction of inbreeding in cassava also in SB2 report).
For many years cassava research at CIAT has been interested in introducing inbreeding in
cassava. The advantages of inbreeding can be summarized as follows:
Because no inbreeding is carried out, a sizable genetic load (undesirable or deleterious genes)
is expected to prevent the crop from fully achieving its actual yield potential.
Since there are no clearly defined populations (quantitative genetics sense) allelic frequencies
cannot be efficiently modified.
Because the highly heterozygous nature of the crop, dominance effects are likely to play a
very important role in the performance of materials being selected. The current scheme can
exploit dominance effects because, once an elite clone is identified, it can be propagated
vegetatively (therefore carrying along the dominance effects). However, it is the same elite
Output 4-27
2003 Annual Report
clones that frequently are selected as progenitors for the production of new segregating
material. In that case, the current procedure has a bias because the breeding value of these
clones are unlikely to be well correlated with their performance per se, precisely because of
the distorting effects of dominance. In other words, good clones are just found not designed.
Production of recombinant seed is cumbersome in cassava. Only 0.6 viable seeds per
pollination are produced. It takes about 18 months since a given cross is planned until an
adequate amount of seed is produced.
When a desirable trait is identified, it is very difficult to transfer it from one genotype to
another (even if a single gene controlled the trait). The backcross scheme, one of the most
common, successful and powerful breeding schemes for cultivated crops, is not feasible in
cassava, because of the constant heterozygous state used throughout the breeding process.
Table 4.10. Results of S2 seedlings transplanted to the field. These materials are the result of
a second consecutive self-pollination (average of 75% homozygosity).
Progenitor
Family
AM 247
AM 262
AM 273
AM 298
AM 320
AM 321
AM 322
AM 323
AM 324
AM 326
AM 328
AM 329
AM 330
AM 331
AM 332
AM 333
AM 334
AM 335
AM 336
AM 337
AM 338
AM 339
AM 340
AM 341
AM 342
AM 343
CM 507-37
CM 2772-3
MCOL 72
CM 6754-8
MTAI 8
CG 1141-1
CM 4365-3
CM 4574-7
CM 4919-1
CM 6921-3
CM 7514-8
SM 805-15
SM 909-25
SM 1219-9
SM 1411-5
SM 1438-2
SM 1460-1
SM 1511-6
SM 1565-15
SM 1665-2
SM 1669-5
SM 1669-7
SM 1741-1
SM 1778-45
MTAI 16
CM 3306-4
TOTAL
Seeds
produced
50
80
71
34
444
47
49
14
36
36
23
28
56
520
25
19
519
493
434
506
440
241
362
23
35
62
4649
Seeds
germinated
4
27
37
28
340
29
29
8
19
30
13
13
37
339
11
12
378
418
334
444
355
230
215
21
23
32
3426
Seedlings
transplanted
2
21
36
27
340
29
29
8
19
30
13
13
37
339
11
12
378
418
334
444
355
230
215
21
23
32
3416
%
Germ.
8.0
33.8
52.1
82.4
76.6
61.7
59.2
57.1
52.8
83.3
56.5
46.4
66.1
65.2
44.0
63.2
72.8
84.8
77.0
87.7
80.7
95.4
59.4
91.3
65.7
51.6
62.0
The lack of inbreeding in cassava implies a very restricted genetic variability based on
recessive traits. Commercially important mutants such as those found and exploited in maize
(waxy, floury, high-quality-protein, sweet corn, popcorn, etc.) are not known in cassava. It is
Project IP3: improving cassava for the developing world
Output 4-28
not clear if cassava has or not this kind of useful mutants, but it is obvious that if they
existed the breeding scheme employed did not facilitate their identification since the
heterozygous nature of the crop significantly reduced the chances of the homozygosity
required for the expression of recessive traits.
The use of totally inbred material (as parents in the production of hybrids which will then be
propagated vegetatively) would greatly facilitate the exchange of germplasm among different
cassava breeding projects. Currently this exchange takes place as in vitro plants, which is
expensive, time-consuming, and by its very nature restricted to a few genotypes. Botanical
seed of inbred genotypes will breed true, and therefore the genotype can be transferred in
this way without the genetic segregation that occurs from non-inbred materials. This will
effectively reduce the relative isolation in which cassava-breeding projects currently operate.
CIAT has begun the production of inbred cassava material (currently at 50 and 75%
homozygosity). All the elite clones identified through the cassava-breeding project are going
to be used for recombination (to produce new segregating populations), but also will be selfpollinated to initiate an S2 recurrent selection process to: a) reduce the inbreeding
depression in cassava (§ reduce genetic load); b) identify useful recessive traits of commercial
(i.e. waxy roots), nutritional (acyanogenesis) or agronomic (reduced post harvest
deterioration) relevance. Paralel to this a special project has been approved for the
development of a protocol for the production of doubled-haploids from cassava anthers.
Table 4.10 summarizes the results of transplanted S2 seedlings in the Palmira Experimental
Station in August 2003. A large number of seedlings (expected average of homozygosity of
75%) were produced and transplanted.
During the first semester of 2004 these 3461 plants will be harvested. In the process
extensive evaluations will be performed in search of useful traits. This is could eventually be
a turning point for cassava research at CIAT
Output 4-29
2003 Annual Report
OUTPUT 5
Activities related with the maintenance of the germplasm bank of cassava and other
Manihot species.
CIAT has been trusted with the maintenance of the cassava world germplasm bank, which includes more than
6000 accessions of Manihot esculenta and other Manihot species. In the following pages a summary of activities
related to the germplasm bank will be described.
Activity 5.1. Maintenance of Manihot germplasm bank in the field.
Rationale
The Genetic Resources Unit is officially in charge of the maintenance of the cassava germplasm bank, both in
vitro and in the field. However, for practical reasons, the field operations are coordinated by IP3 project. Since
year 2000 an extensive activity to clean up from frogskin disease, the germplasm bank has been carried out. Plots
from the germplasm bank maintained in the field, because of its very nature, could not be eliminated even if
frogskin disease appeared in some of the plants. Eventually the incidence of the disease increased to unacceptable
levels.
In order to reduce the costs of maintenance of the germplasm collection and because of the problems associated
with frog skin disease it was decided that the collection will be moved from the field and be maintained using the
“bonsai system” under greenhouse conditions. This is an activity described in more detail in the respective report
from the Germplasm Collections and will not be further discussed herein.
Because the Germplasm Collection is no longer in the field during this period, a set of clones has been
regenerated in order to produce enough roots for the evaluation of different traits, particularly novel starch quality
traits. IP3 project is trying to produce plants from as many as 2000 clones from the germplasm collection. These
plants are produced from stocks that have been certified to be frog skin free.
Specific Objectives:
a) To screen the germplasm bank in search for the natural occurrence of novel starch types
Results
We have begun a systematic characterization of the starch properties in the roots of the accessions from the
germplasm bank. Every year up to 2000 accessions are evaluated. So far approximately 4000 clones have been
characterized in the last few years and a group of about 2000 clones will be evaluated next year. To do this
evaluation plants from these clones have been gradually recovered from the in vitro collection. The results of this
evaluation will be published as soon as the data set is completed and an agreement has been reached with the
company financing this research. To produce the required plants the following steps need to be taken.
Regeneration of each accession from the in vitro collection.
From each accession, a plant from the in vitro collection was regenerated and indexed to certify it is free of
diseases. Plants passing this first test are then hardened in conditions that do not allow for the presence of white
flies, and therefore, minimizes the possibility of acquiring the frogskin disease agent again.
2003 Annual Report
Output 5-1
Planting of disease free plantlets outside CIAT, in isolated fields.
Because of the higher incidence of frogskin disease at CIAT (mainly at the germplasm bank collection in the
field), plants that are certified to be disease free, or those developed from botanical seeds (which do not transmit
viral agents to the plants germinating from them), were planted outside CIAT in isolated plots (CEUNP). Only
virus-free plants were planted in those isolated plots. In the meantime, plantings at CIAT were reduced as a higher
proportion of the cassava germplasm is being certified to be disease-free. In short the outside plantings were
certified to be “clean”, whereas the plantings at CIAT were not. This situation was maintained until the middle of
2001, when materials not certified to be disease free moved out of CIAT, and those that are clean, came back to
the station.
Breaking the life cycle of the white flies at CIAT.
In addition of maintaining an ideal reservoir for the agent of the frogskin disease in the germplasm bank, there is a
second factor that facilitated the spread of the disease. In effect, the white flies problem has increased
considerably during the last few years. A major factor for this increment has been the continuous planting of
cassava year round. The insects, therefore, had an ideal condition for maintaining high population densities.
Between July 1 and July 31, 2002, there was no cassava plant in the field at CIAT’s station in Palmira. It is
expected that this measure will reduce population densities for the insect, and in turn, will reduce to a minimum
the already inefficient transmission of the frogskin disease agent to healthy plants.
Harvest of stakes only from asymptomatic plants.
A common procedure to harvest cassava is to first take the stakes (vegetative “seed”) out of the field, and then
harvest the roots. In fact this practice prevents the elimination of stakes from diseased plants, because when the
roots are evaluated for symptoms, the stakes from each plant has already been mixed with other stakes from
different plants. Starting in this year, the harvest protocol has been changed slightly. The whole plant is first taken
out of the ground, so before taking the stakes the roots can be inspected to make sure they are asymptomatic.
Stakes are taken only from plants that do not show the symptoms. This practice will reduce to a very minimum
the “seed” transmission of the disease to only two possible cases: a) when the worker fails to recognize the
symptoms; or b) when the plant has been infected late in the season and, therefore, it does not show the symptoms
but the disease will be transmitted through its stakes.
Results
All the activities were carried out as expected. A large proportion of accessions from the germplasm bank was
evaluated for frogskin disease and, if clean, planted in isolated conditions. Sequential plantings were performed as
the plants were certified to be disease-free. Therefore, harvest of these plants was also done sequentially. The
levels of frogskin were very low, as expected. However, given the results from the previous year, when higher
than acceptable levels of frogskin disease were observed, it has been decided not to plant the entire germplasm
bank in the field, until the vector(s) and pathogen(s) are clearly determined.
Activity 5.2.
Evaluation of M. esculenta and related species from the germplasm collection for useful traits,
particularly for higher protein content in the roots.
Rationale
Many of the activities related to the evaluation of the introgression of useful genetic variability from wild relatives
of cassava is described in Otuput 8. In this activity, however, a particular action will be described because of its
relevance. In a previous Annual Report (2002) it was reported that a few clones more commonly from Central
America had been found to posses high levels of crude protein in the roots. These clones were properly identified
and are currently in the process of recovery from the germplasm collection and will soon be evaluated again to
Output 5-2
Project IP3: improved cassava for the developing world
confirm their protein content in the roots. One important feature that will be evaluated is the stability of that trait
through multi-location evaluations and also the effect of the age of the plant.
Specific Objectives:
a) To produce plants from vitroplants for field evaluation on protein content in the roots.
Results.
Plants from ?? clones are currently in the process of hardening.
2003 Annual Report
Output 5-3
OUTPUT 6
Breeding for insect and other arthropods resistance and development of alternative
methods for their control
Activity 6.1. Evaluation of cassava germplasm for
(Aleurotrachelus socialis) during 2002-2003.
resistance
to
whiteflies
Rationale
Research in host plant resistance (HPR) to whiteflies has increased in recent years, primarily
because of the extensive damage caused by the Bemisia tabaci species complex on a wide
range of agricultural crops. In general, a narrow range of germplasm has been tested and
there are few deliberate breeding programs designed to develop higher levels of resistance in
cultivars. In some cases, related wild species have been evaluated for a source of whitefly
resistance in breeding programs, but examples are limited. Consequently, HPR to whiteflies
in cultivated crops is rare.
The CIAT cassava and IPDM project (IP-3 and PE-1) have placed a special emphasis on our
ongoing efforts to develop whitefly resistant cultivars in cassava. This project is unique
because we are systematically screening a large germplasm bank (|6000 clones), and more
recently, wild Manihot species. We continue to identify resistant genotypes and through a
comprehensive breeding scheme, develop commercial hybrids containing whitefly resistance.
We have also developed a mapping population of genotypes suitable for identifying molecular
markers for a given trait, using MEcu 72 as the resistant parent and MCol 2246 as the
susceptible parent (Fam: CM 8996).
Whiteflies, especially in the Neotropics, cause direct damage to cassava by feeding on the
phloem of the leaves. This causes leaf curling, chlorosis, and leaf fall, which results in
considerable reduction in root yield if prolonged feeding occurs. There are two major species
causing direct feeding damage in the Neotropics, Aleurotrachelus socialis in the Northern
region of South America (Colombia, Ecuador and Venezuela) and Aleurothrixus aepim in
Brazil. Yield losses resulting from A. socialis and A. aepim feeding are common in both
regions. In Colombia cassava field losses as high as 79% are reported caused by A. socialis
and about 40% due to A. aepim in Brazil.
HPR offers a low-cost practical, long-term solution for maintaining lower whitefly populations
and reducing crop losses. This is especially important for cassava, as it has a long growing
cycle and is often grown by resource limited smallholder farmers who cannot afford costly
inputs. During March of 2003, CORPOICA (Colombia, MADR) officially released a whitefly (A.
socialis) resistant cassava cultivar, Nataima-31 (CG 489-31; CIAT Breeding Code). This
cultivar, a progeny of a MEcu 72 x MBra 12 cross, was developed over a 15 year period in a
collaborative CIAT-CORPOICA (Nataima, El Espinal, Tolima) effort.
A field day to
commemorate the varietal released attracted about 200 cassava producers from the El
Espinal/Tolima region.
Output 6-1
2003 Annual Report
Specific Objectives:
A. Evaluation of the family CM 8996 for a genetic study for whitefly (A. socialis)
resistance at CORPOICA, Nataima, Tolima (2002-03).
Materials and methods
Systematic and continuing whitefly (A. socialis) evaluations on cassava germplasm are
carried out at primarily two sites, the CIAT farm in Santander de Quilichao and at
CORPOICA, Nataima, Tolima.
The family CM 8996 was developed from a cross of the whitefly resistant cultivar, MEcu 72
and the susceptible cultivar MCol 2246.
The resulting progeny, approximately 700
genotypes, are being systematically evaluated at the two aforementioned sites with the
objective being to study the genetics and inheritance of whitefly (A. socialis) resistance in
cassava. The Santander de Quilichao plantings were done in April 2002 and harvested in
March, 2003. The CORPOICA, El Espinal plantings were also planted in April 2002 and
harvested during May 2003. Both sites traditionally have moderate to high whitefly
populations although in some years A. socialis populations have been too low for reliable
screening for resistance. At CORPOICA, El Espinal; 654 genotypes (progeny) of family CM
8996 were planted, while at CIAT/Santander de Quilichao 700 genotypes were planted. Each
genotype was planted in rows of five plants and every 20 rows, the susceptible cultivar CMC
40 (MCol 1468) was planted (this “indicator” cultivar provides a record of whitefly population
levels and distribution). Three evaluations for whitefly populations and damage levels were
carried out during the course of the crop cycle at Santander and 4 at Nataima. A 1 to 6
damage and population scale was employed (Table 6.1). Root yield data was recorded at
harvest by harvesting the central 3 plants of each genotype/row.
Table 6.1.
Population and damage scales for evaluating cassava germplasm for resistance
to whiteflies.
Population Scale (Nymphs and Pupae)
1 = no whitefly stages present
2 = 1-200 individuals per cassava leaf
3 = 201-500 per leaf
4 = 501-2000 per leaf
5 = 2001-4000 per leaf
6 = > 4000 per leaf
Damage Scale
1 = no leaf damage
2 = young leaves still green but slightly flaccid
3 = some twisting of young leaves, slight leaf curling
4 = apical leaves curled and twisted; yellow-green mottled appearance
5 = same as 4, but with “sooty mold” and yellowing of leaves
6 = considerable leaf necrosis and defoliation, sooty mold on mid and lower leaves and
young stems.
Results
CORPOICA, Nataima. This is the second cycle of evaluating this family CM 8996 at Nataima.
Last year (Cycle 1, 2001-2002) whitefly (A. socialis) populations were low throughout the crop
cycle, resulting in low damage ratings. Whitefly populations during the second cycle (2002Project IP3: improving cassava for the developing world
Output 6-2
03) were higher and present on all 633 genotypes evaluated (Figure 6.1). The highest
number, 338 (53.4%) genotypes occurred in the 2.0 to 3.0 population range, indicating up to
500 A. socialis individuals per leaf, while 127 (20.1%), genotypes fell into the 1.1 to 2.0
range, or 1-200 per leaf. CMC 40, the susceptible check had an average damage rating of
4.5, indicating about 1000 to 2000 whitefly individuals per leaf and signifying a moderate to
high A. socialis population. This population was fairly evenly distributed throughout the
evaluation field. CMC 40 also had high damage ratings, consistently between 4.0 and 5.0
throughout the field (Figure 6.2). This level of population and damage results in sufficient
selection pressure to do adequate evaluations of germplasm.
400
338
350
317
300
Genotypes
264
250
200
153
127
150
100
50
48
0
4
15
0
0
0
1
1.1 - 2
2.1 - 3
3.1 - 4
Population Grade
4.1 - 5
0
5.1 - 6
Damage Grade
Damage Grade
Figure 6.1. Whitefly (A. socialis) population and damage ratings of genotypes from the
cassava family CM 8996 evaluated at CORPOICA, Nataima (Tolima) during the 200203 crop cycle.
6
5.5
5
4.5
4
3.5
3
2.5
2
1.5
1
4.18
Superiors
4.6
4.2
Mid
4.2
Lower
Damage Grade
CMC 40
Figure 6.2. Average whitefly (A. socialis) damage and population ratings at three plant
levels (superior, middle and lower leaves) of the susceptible control clone, CMC 40 at
CORPOICA, Nataima (Tolima) during 2002-03.
The majority of the genotypes, however, had very low damage scores: 264 (41.7%) genotypes
had a damage rating of 1.0, indicating no visible damage. Three hundred seventeen (50.1%)
Output 6-3
2003 Annual Report
had a damage rating of 2.0 (young leaves slightly flaccid) while 48 (7.6%) had a 3.0 rating
(some twisting of young leaves, slight leaf curling). Only 4 (0.6%) genotypes had, what can be
considered, a high damage rating (4.0). Both A. socialis populations and damage were higher
during the 2002-03 cycle than the 2001-02 cycle, as indicated by the > 4.0 damage rating
(curling and twisting of apical leaves, mosaic-like appearance to the leaves and sooty mold)
for the susceptible control CMC-40 (average of 19 rows evaluate) (Table 6.2).
Table 6.2.
Whitefly (A. socialis) populations and damage ratings on the susceptible cassava
cultivar CMC 40 planted with the cassava Family CM 8996 at CORPOICA, Nataima
(Tolima) during 2002-03.
Population Grade
Upper Bud
Clone
Medium
Nymph Pupa Nymph Pupa
Lower
Damage Grade
Pupa
Upper Medium Lower
Adult
Egg
CMC 40
3.0
4.0
4.0
5.0
4.0
4.0
4.0
4.5
4.0
4.0
CMC 40
3.0
4.0
4.0
4.0
4.0
4.0
4.0
4.5
4.0
4.0
CMC 40
3.0
4.0
5.0
5.0
5.0
4.0
4.0
4.5
4.5
4.5
CMC 40
4.0
3.0
5.0
5.0
5.0
4.0
5.0
4.5
4.5
4.0
CMC 40
4.0
3.0
4.0
4.0
5.0
5.0
4.0
4.0
4.0
4.0
CMC 40
3.5
4.0
4.0
4.5
4.0
5.0
4.0
4.5
4.0
3.5
CMC 40
4.0
3.5
4.5
5.5
5.0
6.0
4.0
4.5
5.5
5.0
CMC 40
4.0
3.0
4.0
4.0
5.0
4.0
4.0
4.0
4.0
4.0
CMC 40
3.0
4.0
4.0
4.0
4.0
4.0
4.0
4.5
4.0
4.0
CMC 40
4.0
4.0
4.5
4.0
5.0
5.0
4.5
4.0
4.5
4.0
CMC 40
3.0
3.0
4.5
4.0
3.0
3.0
3.0
4.0
4.0
4.0
CMC 40
4.0
4.0
6.0
6.0
6.0
4.0
4.0
4.0
4.0
4.5
CMC 40
4.0
4.0
6.0
6.0
5.5
5.5
5.0
4.0
4.0
4.5
CMC 40
4.0
4.0
5.5
5.5
5.5
4.0
4.0
4.0
4.0
4.0
CMC 40
3.0
3.0
6.0
6.0
6.0
5.0
4.0
4.0
4.5
4.0
CMC 40
4.0
4.0
5.0
5.0
5.0
4.0
4.0
4.0
4.0
4.0
CMC 40
3.0
4.0
4.0
4.0
4.0
4.0
4.0
4.5
4.0
4.0
CMC 40
4.5
4.0
4.5
4.5
5.0
4.5
4.5
4.0
4.5
5.0
CMC 40
4.0
4.5
4.0
3.0
4.5
5.0
5.0
4.0
4.5
5.0
These results may be indicating that there exists at least a moderate level of A. socialis
resistance in nearly all of the CM 8996 genotypes. Since the female parent, MEcu 72, is
highly resistant to A. socialis, these results are not completely unexpected. Both parents of
this family are vigorous varieties and the resulting vigor in the progeny should be expected.
At harvest the range of segregation in the F1 can be observed in the yield data. Based on
only the three central plants of each row of each genotype, yields ranged 0.0 T/ha (no roots
produced) for CM 8996-170 to 98.7 T/ha for genotype CM 8996-679 (Table 6.3).
Project IP3: improving cassava for the developing world
Output 6-4
Table 6.3.
Yield parameters (harvest index, % dry matter and cooking quality) of 628
cassava genotypes from the family CM 8996 at CORPOICA, Nataima (Tolima) during
2002-03 crop.
Yield T/ha
Harvest Index
% Dry Matter
Cooking Quality
Range
No.
Genotypes
Range
No.
Genotypes
Range
No.
Genotypes
Grade
No.
Genotypes
0–10
80
0-0.39
75
17–29.96
107
1
128
11–19.8
20–29.9
30–49.6
50–69.2
70–98.7
115
162
206
53
12
0.4–0.49
0.5–0.59
0.6–0.69
0.7–0.86
1.0
102
187
213
51
0
30–32.96
33–35.93
36–40.34
41–48.16
-
228
231
57
2
-
2
3
4
5
-
261
125
97
10
-
About 80 (12.7) genotypes yielded less than 10 T/ha, or below the national average, while
115(18.3%) yielded between 11 and 19.8 T/ha or somewhere in the range of the national
average. This means that 68.9% (433 genotypes) yielded above 20 T/ha; 65 of these (10.4%)
yielded above 50 T/ha (Table 6.3). Since these yield estimates are based on only 3 plants,
this data is presented to indicate the range of yield segregation in the F1. It is interesting to
note that we do not see this type of segregation in whitefly resistance, where most of the F1
genotypes are in the moderate to high level of resistance.
This data however does indicate the possible yield potential of some of the genotypes.
Harvest index show a similar range; 451 genotypes (71.8%) had a harvest index equal to or
above 0.5 and 264 genotypes (42.0%) were above 0.6 (Table 6.3). This data further supports
the production capacity of this family (CM 8996). Root dry matter also displays a range of
segregation; more than half (235=53.3%) of the genotypes had dry matter below 33%, which
is unacceptable for a commercial variety. Nearly 37% (231 genotypes) of the F1’s had a root
dry matter between 33 and 36%, this is acceptable but higher is preferred. Fifty-nine
genotypes (9.4%) had dry matter content above 36% and 2 of these were above 41% (Table
6.3).
More than half of the genotypes (389 or 62.6%) had a cooking quality rating between 1 and 2,
indicating roots that soften or are made edible in 20 to 25 minutes, a high portion of starch
and excellent eating texture (Table 6.3). These results further indicate that it is possible to
combine good whitefly resistance with high yield and excellent commercial qualities.
B. Evaluation of the Family CM 8996 for a genetic mapping study for whitefly (A.
socialis) resistance at Santander de Quilichao (2002-03).
Results
Santander de Quilichao. The methodology for the trial at Santander, is basically the same as
that for CORPOICA, Tolima, described in the previous section. In May 2003, 673 genotypes
of family CM 8996 were harvested. Three whitefly damage and population ratings were made
during the growing cycle. Growing conditions at Santander differ from those at Tolima in
that soils at Santander are acid and clay loam in texture while at Nataima soils are not acid
and more a sandy leam. Planting material for this trial originated from the 2001-02 trial at
Nataima, Tolima. In addition, at Santander several other pests are present and affect
Output 6-5
2003 Annual Report
cassava growth, especially mites and thrips. The presence of these additional pests may
suppress whitefly populations and damage symptoms and also affect yield.
Root yield of the CM 8996 genotypes at Santander ranged from 0.0 to 42.2 T/ha (Figure 6.3)
while at Nataima yields ranged from 0.0 to 98.7 T/ha (a difference of about 57.2%). At
Santander 432 of the 673 genotypes (64.2%) had yields of 10 T/ha or lower, while at Nataima
only 80 genotypes (12.7%) fell into this range (Figure 6.3). At Santander only 50 genotypes
7.4%) had yields above 20 T/ha (Table 6.5) with a maximum production of 42.2 T/ha, while
at Nataima, 68.9% (433) of the genotypes had yields above 20 T/ha (Figure 6.3) and a
maximum of 98.7 T/ha.
Harvest index of Santander was between 0.39 and 0.89 (Table 6.5). The accuracy of this
data is questionable due to the damage caused by mite and thrips attack as well as other
variables such as the presence super-elongation disease (Sphaceloma sp) toward the end of
the crop cycle. Mite, thrips and super elongation attack did not occur at Nataima.
432
450
Tolima
No. Genotypes
350
250
Santamder Quilichao
206
191
162
150
80
115
44
6
50
-50
0 -10
11 - 19.8
20 - 29.9
30 - 42.2
53
0
50 - 69.2
12 0
70 - 98.7
Range of Yields (T/ha)
Figure 6.3. Yields for the cassava family CM 8996 (MEcu 72 x MCol 2246) under whitefly
(Aleurotrachelus socialis) pressure at two evaluation sites, Santander de Quilichao
(Cauca) and CORPOICA, El Espinal (Tolima) during 2002-03 crop cycle.
Results on dry matter content did not differ greatly between the two sites. At Santander, as
in Nataima, more than half the genotypes had a dry mater content below 33.0% (Tables 6.3
and 6.5). At Santander 55% (374) genotypes (53.3% and 335 genotypes at Nataima) had a
dry mater content below 33.0%; 32.4% (218 genotypes) fall in the range of 33 to 36% (231
and 37.0% at Nataima) and 10.5% (71 genotypes above 36% (59 and 9.4% at Nataima).
Cooking quality of the roots harvested at Santander ranged from very good to very poor.
About 32.4% (210) of the genotypes had excellent cooking quality, ranging between 1 to 2
(Table 6.5), while nearly 29% (194) were intermediate at grade 3. A higher percentage of the
genotypes had a poorer cooking quality at Santander than at Nataima (36.4% vs. 17.0%)
(Table 6.3 and 6.5). All genotypes are sweet (i.e. low HCN).
A comparison of genotypes with the highest yields at Nataima, Tolima and Santander de
Quilichao can be found in Table 6.6. It can be observed that those genotypes having the
highest yield in Tolima had very low yields at Santander. For example, CM 8996-79 yielded
the equivalent of 98.7 T/ha at Tolima and yielded nothing at Santander. CM 8996-250
yielded 71.3 T/ha at Tolima and only 4.3 T/ha at Santander (Table 6.6). However, we can
observe that those genotypes that yielded highest at Santander also yielded well at Tolima, in
Project IP3: improving cassava for the developing world
Output 6-6
several cases yielding higher. These results suggest that clones that yield well in a harsher
environment, in most cases will do even better when sown in a more favorable environment.
Table 6.5.
Yield parameters (harvest index, % dry matter and cooking quality) of 673
genotypes from the family CM 8996 at Santander de Quilichao (Cauca) during the 2002-03
crop cycle.
Yield T/ha
Harvest Index
% Dry Matter
Cooking Quality
Range
0-10.9
11-19.8
20-29.9
No.
Genotypes
432
191
44
Range
0-0.39
0.4-0.49
0.5-0.59
No.
Genotypes
88
112
212
Range
20.2–29.9
30–32.9
33–35.9
No.
Genotypes
123
251
218
Grade
1
2
3
No.
Genotypes
42
168
194
30-38.4
5
0.6–0.69
200
36–40.4
62
4
196
40-42.2
1
0.7–0.89
61
41-55.8
9
5
49
Table 6.6.
Highest yielding cultivars of the cassava family CM 8996 evaluated at two sties
for whitefly (A. socialis), Santander de Quilichao (Cauca) and CORPOICA, Nataima
(Tolima) during 2002-03.
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
Clone
8996-79
8996-14
8996-126
8996-290
8996-536
8996-244
8996-401
8996-482
8996-660
8996-250
Yield
Tolima
98.7
95.5
84.5
82.9
80.3
78.7
78.5
76.5
75.5
71.3
Yield
S. de Quilichao
13.5
7.5
20.7
9.9
3.7
14.9
4.8
6.9
4.3
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
Clone
8996-318
8996-596
8996-410
8996-282
8996-404
8996-280
8996-88
8996-81
8996-305
8996-65
Yield
S. de Quilichao
42.3
38.4
37.5
37.3
35.9
30.68
29.9
29.9
29.7
29.5
Yield
Tolima
48.4
52.8
34.5
22.3
42.5
34.73
42.8
35.8
12.93
16.4
C. Evaluation of selected cassava varieties for whitefly (A. socialis) resistance at
CORPOICA, Nataima, Tolima (2002-03).
Rationale
The systematic evaluation of the cassava germplasm bank (>5000 landrace varieties) has
resulted in the identification of numerous cultivars as potential sources of whitefly
resistance. These selected varieties will go through numerous field cycles. The methodology
used for evaluation is similar to that described in the previous activity.
Results
Whitefly populations and pressure was uniformly high for this trial indicating good selection
pressure (Table 6.4). Eleven of the 16 varieties evaluated had population levels of 4.0 or
higher (Figure 6.4). The trial was composed primarily of Peruvian, Colombian and Brazilian
varieties and included Nataima-31 (CG 489-31) and CG 489-34, both whitefly resistant
progeny from the MEcu 72 x MBra 12 cross. The five varieties with the lowest whitefly
population levels and consequently the lowest damage ratings were Nataima 31 (2.5 popl. Vs
1.0 damage), CG 489-34 (3.2 and 1.3), MPer 317 (1.8 and 1.3), MPer 334 (2.4 and 1.3), MPer
Output 6-7
2003 Annual Report
273 (2.0 and 1.7) (Table 6.4). All other varieties had populations’ levels above 4.0 and
damage rating ranging from 2.7 to 4.3. The majority of the varieties were of Brazilian origin
(9) and all appeared more susceptible than the three Peruvian varieties. This again supports
the observation that most whitefly (A. socialis) resistance is concentrated in Ecuadorian and
Peruvian varieties. MPer 317, MPer 334 and MPer 273 have consistently shown low
populations and damage levels through several field evaluations and growth chamber trials.
The two hybrids from the MEcu 72 x MBra 12 cross, Nataima 31 (CG 489-31) and CG 489-34
continue to express good field resistance with low whitefly populations and damage levels.
Table 6.4.
Selected Peruvian, Brazilian and Colombian cassava cultivars and hybrids
evaluated for whitefly (A. socialis) resistance at CORPOICA, Nataima (Tolima) during
2002-03 crop cycle.
Population Grade
Upper Leaves
Middle Leaves
Clone
Nataima-31
CG 489-34
MPer 317
MPer 334
MPer 273
MBra 292
MBra 81
MCol 2025
MBra 29
MBra 442
0MBra 532
CG 936-7
MBra 303
MCol 2246
MBra222
MBra370
Damage Grade
Lower
Average
Adult
Egg
Nymph
Pupa
Nymph
Pupa
Pupa
Upper
Mid
Lower
Pop.
Damage
1.5
4
2
2
2
3
3
4
3
3.5
4
3
3
2
5
4
1
4
2
2
2
3
4
4
4
3.5
3
4
4
3
4
4
2
4
3
2
1
5
5
4
5
5
4.5
5
4
5
4
5
3
4
2
3
1
4
5
5
5
5
4.5
5
4
5
5
5
3
3
1.5
3
2.5
4
5
4
4.5
5
4
5
5
4
5
5
2.5
3
1.5
2
3
4
5
4
4
5
4
4
5
3
5
4
2
2
1
2
2.5
4.5
4
4
4.5
4
4
4
5
3
5
5
1
1
1
1
1
1
3
2.5
1
3
4
4
4
4
4
4
1
1
1
1
1.5
3
3
2.5
3
3
3.5
4
4
4
4.5
4.5
1
2
2
2
2.5
4
2
3
5
4
4
4
4
4
4
4.5
2.5
3.2
1.8
2.4
2
4.3
4.8
4.2
4.6
4.8
4.2
4.6
4.6
4
4.8
4.8
1
1.3
1.3
1.3
1.7
2.7
2.7
2.7
3
3.3
3.8
4
4
4
4.2
4.3
12
11
Genotypes
10
8
5
6
4
4
2
4
2
2
2
1
1
0
0
0
1
1.1 - 2
2.1 - 3
3.1 - 4
Population Grade
4.1 - 5
0
5.1 - 6
Damage Grade
Figure 6.4. Whitefly (A. socialis) population and damage ratings on Peruvian, Brazilian and
Colombian cultivars and hybrids evaluated for resistance at CORPOICA, Nataima
(Tolima) during 2002-03.
Contributors: Bernardo Arias, Anthony C. Bellotti.
Collaborators: Gustavo Trujillo, Gerardino Pérez.
Project IP3: improving cassava for the developing world
Output 6-8
Activity 6.2. Manihot wild species and hybrids evaluated for whitefly (A. socialis)
populations damage at Santander de Quilichao during 2002-03 crop
cycle.
Rationale
Wild Manihot species have provided a potential source of resistance genes for arthropod pests
(see Activity 9, this reports). In collaboration with the cassava genetics and plant breeding
sections, field plantings of wild Manihot species are evaluated when arthropod attacks occur.
During 2002-03 about 549 genotypes, consisting of wild cultivars (CW), GM and CM hybrids
were evaluated for whitefly (A. socialis) damage at Santander de Quilichao.
Results
Whitefly damage ratings, in general, were moderate and of the 549 genotypes, 150 (27.3%)
resulted in no damage symptoms (Figure 6.5). Two hundred fourteen (39.0%) showed light
damage symptoms (1.1 to 2.5 on the 1-6 damage scale, Activity 1, Table 6.1), while 44 (8.0)
had intermediate damage symptoms (2.6-3.0) and 141 (25.7%) genotypes showed damage
symptoms above 3.0. Of these 37 genotypes (6.7%) had severe damage (4.1 to 5.0).
All genotypes were infested with whiteflies with populations ranging from 1 (<200 individuals
per leaf) to 4 (500 to 2000 per leaf). Of the a 549 genotypes, 371 (67.6%) had low whitefly
populations (below 2.5 on the 1 to 6 scale), while 178 (32.4%) resulted in an intermediate
damage rating (2.6 to 4.1) (Figure 6.5).
371
400
350
No. Clones
300
250
214
200
150
121
150
100
50
104
44
56
37
1
0
0
0
0
1
1.1 - 2.5
2.6 - 3.0
Population
3.1 - 4.0
4.1 - 5.0
5.1 - 6.0
Damage
Figure 6.5. Whitefly (Aleurotrachelus socialis) populations and damage on 549 wild Manihot
hybrids and other hybrid at Santander de Quilichao (Cauca) during the 2002-03 crop
cycle.
Of the 296 Manihot hybrids (CW) evaluated, nearly half, 48.6%, had no whitefly damage
symptoms (Figure 6.6). Only 4.7% (13 genotypes) had a damage rating of 3.0 or higher.
Whitefly populations were present on all genotypes but populations were in the low range
(1.0 to 2.5) on nearly all (95.2%) of the genotypes. Whitefly populations and damage were
higher on the GM genotypes (Figures 6.7 and 6.8). In the GM and CM hybrid plantings
(Figure 6.7), whitefly damage ranged from 1.1 to 5.0 on the damage scale. Of the 224
genotypes, 65.6% had damage ratings above 2.5 and about 55% had whitefly populations
above 2.5 (Figure 6.7). In the CM planting of 29 hybrids genotypes, nearly 83% had damage
Output 6-9
2003 Annual Report
Percent (%)
ratings above 2.5 (Figure 6.8) and no genotypes were without damage. Whitefly populations
were moderate.
54.7
60
50
48.6
46.6
42.6
40
30
20
10
0
2.7
0
1
1.1 - 2.5
2.6 - 3.0
2.7
2
3.1 - 4.0
Populaltion
0
0
4.1 - 5.0
0
0
5.1 - 6.0
Damage
Figure 6.6. Percent whitefly (A. socialis) populations and damage recorded on 296 wild
Manihot hybrids at Santander de Quilichao (Cauca) during the 2002-2003 crop cycle.
50
45
37.9
Percent (%)
40
37.5
31.7
30
20
14.3
10
16.5
13.8
2.7
0
0.4
0
0
0
1
1.1 - 2.5
2.6 - 3.0
3.1 - 4.0
Population
4.1 - 5.0
5.1 - 6.0
Damage
Figure 6.7. Percent whitefly (A. socialis) populations and damage recorded on 224 GM
hybrids planted at Santander de Quilichao (Cauca) during the 2002-03 cycle.
60
48.3
Percent (%)
50
44.8
37.9
40
30
20.7
17.2 17.2
20
13.8
10
0
0
0
0
0
0
1
1.1 - 2.5
2.6 - 3.0
Population
3.1 - 4.0
4.1 - 5.0
5.1 - 6.0
Damage
Figure 6.8. Percent whitefly (A. socialis) populations and hybrids recorded on 29 hybrids
(CM) planted at Santander de Quilichao (Cauca) during the 2002-03 cycle.
Project IP3: improving cassava for the developing world
Output 6-10
The fact that the Wild Manihot hybrids had both low whitefly populations and damage is not
completely unexpected. Separate research and observations over the years have indicated
that the wild Manihot species may contain arthropod resistant genes that can provide a
potential source for resistance to the cultivated M. esculenta. Modern biotechnological tools
provide the means to achieve a more effective success in the development of pest resistant
germplasm. The possible whitefly resistance observed in the CW wild Manihot hybrids is
further evidence of this potential and needs to be pursued.
Contributors: Bernardo Arias, Anthony C. Bellotti.
Collaborators: José María Guerrero, Gustavo Trujillo, Gerardino Pérez, Carlos Ñañes.
Activity 6.3. Evaluation of cassava dialelic crosses for whitefly (A. socialis) resistance,
using MEcu 72 as the resistant female parent (at Santander de
Quilichao).
MEcu 72 has consistently displayed high levels of resistance to the whitefly, A. socialis. This
variety was used as the resistant female parent in a dialelic cross with several other
genotypes, including MPer 183. CM 6740-7, SM 1219-9, SM 1278-2, SM 1636-24, SM 167310, SM 1741-1 and HMC-1. The resulting progeny were planted out at Santander de
Quilichao and evaluated for whitefly resistance and damage.
Results
The dialelic cross resulted in 225 progeny. All progeny had whitefly populations but in
general both damage and populations were low (Figure 6.9). Nearly all of the progeny had
whitefly populations between 1.1 to 2.5 (99.1%). One hundred four genotypes (46.2%)
resulted in no damage symptoms, while 111 progeny (49.3%) had a damage rating between
1.1 and 2.5 (Figure 6.9). Ten genotypes had a damage level of 3.0, considered to be
intermediate but susceptible. The genotype GM 310-21 had a high damage rating; the male
parent SM 1278-2, also had a high damage (4.0).
250
223
No. Clones
200
150
104
111
100
50
1
1
10
0
0
0
0
0
0
0
1
1.1 - 2.5
2.6 - 3.0
Population
3.1 - 4.0
4.1 - 5.0
5.1 - 6.0
Damage
Figure 6.9. Whitefly (A. socialis) populations and damage on 225 cassava hybrid progeny
from a dialelic cross using MEcu 72 as the resistant female parent (Santander de
Quilichao, Cauca, 2002-03).
Output 6-11
2003 Annual Report
These results, low whitefly populations and corresponding low damage, of the progeny from
this cross, are not unexpected. MEcu 72, the female parent, is highly resistant and it
appears that this resistance is readily passed on to the progeny.
Contributors: Bernardo Arias, Anthony C. Bellotti.
Collaborators: José María Guerrero, Gustavo Trujillo, Gerardino Pérez, Carlos Ñañes.
Activity 6.4. Evaluation of cassava germplasm in several breeding and genetic trials
for whitefly (A. socialis) damage at Santander de Quilichao (Cauca).
The cassava entomology section participates in the screening of genetic and breeding
materials in close collaboration with breeders and geneticists. The trials evaluated include 1)
Yield trials, 2) Regional trials, 3) Beta carotene varietal selection, 4) Observation trial, 5)
Multiplication trials. A brief description of results from these evaluations follows. Actual
data for all evaluations of each clone is available in the database.
1. Yield trial: Whitefly (A. socialis) populations were generally high; high adult and egg
populations were found on the upper leaves while the middle and lower leaves had high
nymph and pupae populations. Of the 75 clones evaluated, only 1, SM 2583-3 displayed no
damage symptoms (Figure 6.10), while 45 clones (60%) had a damage rating above 3.0 and
26 (34.7%) had a rating between 4.0 and 5.0 (leaf curling, severe clorosis and sooty mold).
Nearly two-thirds of the clones (65.3%) had a population rating of 3.0 or lower and 30 clones
(40%) were rated 2.5 or lower.
35
No. Clones
31
29
30
24
25
21
20
19
16
15
10
5
7
2
1
0
0
0
0
1
1.1 - 2.5
2.6 - 3.0
Population
3.1 - 4.0
4.1 - 5.0
5.1 - 6.0
Damage
Figure 6.10. Average whitefly (A. socialis) population and damage ratings for 82 clones in a
cassava yield trial at Santander de Quilichao (Cauca) during the 2002-03 crop cycle.
2. The regional trial consisted of 29 clones evaluated at Santander de Quilichao for
whitefly (A. socialis) damage. These 29 clones were planted in three replicates. All clones
had average whitefly populations that ranged from 2.5 to 5.0, indicating a moderate to high
selection pressure (Figure 6.11). Two of the clones that consistently presented very low
damage ratings across the three replicates were SM 1871-33 (1.0, 2.0 and 1.7) and SM 20857 (1.3, 1.7 and 1.7). Populations on some clones, especially on the upper and mid leaves
were very high, reaching 5.0 to 6.0 on the damage rating scale. Few clones show low damage
ratings; 5 of the 29 clones (17.2%) had damage ratings below 2.5 (although populations on
Project IP3: improving cassava for the developing world
Output 6-12
some of these reached 4.0 to 5.0). In general many of these are vigorous clones and may
“outgrow” some of the damage symptoms or display some tolerance to whitefly attack.
Twenty-seven of the 29 clones (93.1%) had whitefly population ratings above 3.0, again
indicating high selection pressure (Figure 6.11).
25
20
No. Clones
20
14
15
10
8
7
5
5
2
0
0
2
0
0
0
0
1
1.1 - 2.5
2.6 - 3.0
3.1 - 4.0
Population
4.1 - 5.0
5.1 - 6.0
Damage
Figure 6.11. Average whitefly (A. Socialis) population and damage rating of 29 clones in a
cassava regional trial at Santander de Quilichao (Cauca) during the 2002-03 crop
cycle.
3. Yellow pulp (Beta carotene) varieties. In this trial of 27 clones average whitefly
populations ranged from 2.6 to 4.9 indicating moderate to high selection pressure. Whitefly
(A. socialis) damage levels ranged from 2.0 to 4.0 (Figure 6.12). Twenty-two clones (81%) had
whitefly population levels of 3.0 or higher but only 12 of these had damage levels of 3.0 or
higher. The clones CM 9731-2, CM 9731-11 and CM 9712-7 combined low damage levels
(2.0) and low whitefly populations (2.6). These clones, or this group of clones, should be reevaluated in another crop cycle.
20
No. Clones
16
15
9
10
10
8
8
5
3
0
0
0
0
0
0
0
1
1.1 - 2.5
2.6 - 3.0
Population
3.1 - 4.0
4.1 - 5.0
5.1 - 6.0
Damage
Figure 6.12. Whitefly (A. socialis) population and damage ratings for 27 yellow pulp cassava
clones planted at Santander de Quilichao during the 2002-03 crop cycle.
4. Observational trial: This observational trial of the zone 2 (Colombia, Llanos Orientales)
only 11 clones was planted and evaluated in Santander de Quilichao (Cauca). Whitefly (A.
socialis) populations were low to moderate, not higher than 3.0 on any of the clones.
Output 6-13
2003 Annual Report
Consequently, damage ratings were also low to moderate, ranging from 1.7 to 3.0 (Figure
6.13). Five clones, SM 2730-51, SM 2741-26, SM 2731-9, SM 2740-18 and SM 2743-18 had
damage ratings of 1.7 and whitefly population ratings of 2.3 to 2.4.
10
8
No. Clones
8
6
6
5
4
3
2
0
0
0
0
0
0
0
0
0
1
1.1 - 2.5
2.6 - 3.0
3.1 - 4.0
Population
4.1 - 5.0
5.1 - 6.0
Damage
Figure 6.13. Whitefly (A. socialis) population and damage ratings for 11 clones of zone 2
(Colombian Llanos Orientales) in a cassava observational trial in Santander de
Quilichao (Cauca) during the 2202-03 crop cycle.
5. Regional trial, Atlantic Coast: This regional trial of the zone 1 (Colombian Atlantic
Coast), only evaluated six cassava clones in Santander de Quilichao (Cauca). Whitefly (A.
socialis) population ratings ranged from 3.0 to 4.0, indicating a moderate 2.3 to 3.3, also a
moderate level (Figure 6.14). This trial has not yet been harvested.
4
No. Clones
3
3
3
2
2
2
1
1
1
0
0
0
0
0
0
0
1
1.1 - 2.5
2.6 - 3.0
Population
3.1 - 4.0
4.1 - 5.0
5.1 - 6.0
Damage
Figure 6.14.
Whitefly (A. socialis) population and damage ratings of 6 cassava clones of
zone 1 (Colombian Atlantic Coast), regional multiplication trial in Santander de
Quilichao during the 2002-03 crop cycle.
Contributors: Bernardo Arias, Anthony C. Bellotti.
Collaborators: José María Guerrero, Gustavo Trujillo, Gerardino Pérez, Carlos Ñañes.
Project IP3: improving cassava for the developing world
Output 6-14
Activity 6.5. Studies on whitefly (Aleurotrachelus socialis) resistance mechanisms in
selected cassava genotypes.
As direct feeding pests and virus vectors, whiteflies cause major damage in cassava based
agroecosystems in the Americas, Africa and, to a lesser extent Asia. The largest complex of
whitefly pests on cassava is in the Neotropics, where 11 species are reported. Eight whitefly
species are reported feeding on cassava in Colombia. Aleurotrachelus socialis is the major
species on cassava in Northern South America (Colombia, Ecuador, Venezuela) while
Aleurotrachelus aepim predominates in Brazil and Bemisia tabaci in Africa and parts of Asia.
In whitefly surveys on cassava in Colombia, approximately 92% of the species population is
A. socialis. For this reason, A. socialis receives most of the research effort, especially in the
identification of whitefly resistant cassava genotypes and the development of resistant
varieties.
The first symptoms of whitefly damage are manifested by curling of the apical leaves and
yellowing, necrosis and abscission of lower leaves. This results in plant retardation and
considerable reduction in root yield if feeding is prolonged. Damage and yield losses of this
type are common with A. socialis and A. aepim. There is a correlation between duration of
whitefly attack and yield loss, which has been recorded as high as 79% in prolonged (11
months) attacks and on susceptible cultivars. Cassava farmers will respond to whitefly
attack with frequent applications of toxic chemical pesticides. Pesticide use is costly, often
causes environmental contamination, a hazard to human health and may not provide
effective control.
Stable host plant resistance (HPR) offers a practical long-term low cost solution for
maintaining reduced whitefly populations. Although whitefly resistance in agricultural crops
is rare, several good sources of resistance have been identified in cassava and high-yielding,
whitefly resistant cassava hybrids are being developed. At CIAT we are systematically
evaluating the cassava germplasm bank of more than 6000 accessions. The clone MEcu 72
has consistently expressed the highest level of resistance and is being employed in a breeding
scheme to develop whitefly resistant hybrids (see CIAT 2002, IP-3 Annual Report). Additional
cultivars expressing moderate to high levels of resistance in field trials include MEcu 64,
MPer 335, MPer 415, MPer 317, MPer 216, MPer 221, MPer 265, MPer 266 and MPer 365.
The objective of these present studies is to evaluate several selected genotypes for
mechanisms of resistance to A. socialis under controlled growth chamber conditions.
Methodology
The genotypes selected for evaluation were MEcu 64, MPer 273 and MPer 334; CMC 40 was
the susceptible control and MEcu 72 the resistant control. All genotypes have been field
evaluated during numerous trials at CORPOICA, Nataima (Tolima). As previously mentioned
MEcu 72 has consistently shown resistance to A. socialis and in laboratory controlled
resistance mechanisms evaluations, resulted in a 72% mortality, a lower oviposition rate,
longer development time and reduced size. CMC 40 supports high A. socialis populations
and low mortality. In field trials MPer 273, MPer 334 and MEcu 64 genotypes showed low to
moderate A. socialis populations and few damage symptoms. Four A. socialis development
parameters were evaluated, mortality/survival, duration of the life cycle, nymphal
development size, (Antibiosis), and ovipositional preference (Antixenosis). This was done in
two separate experimental designs.
Output 6-15
2003 Annual Report
1. Antibiosis Experiments: This was done in two parts; in the first, test plants were
infested and evaluated by using A. socialis adults harvested directly from the greenhouse
maintained colony being reared on the susceptible CMC 40 (Figure 6.15A). The second
evaluations were done by first preconditioning A. socialis on the selected test genotypes for
two generations. These individual colonies on the five aforementioned genotypes were reared
in wooden, nylon mesh lined cages (1m x 1m x 1m) in the greenhouse (Figure 6.15B). All
antibiosis experiments were carried out in the growth chamber (28r1qC, 60-70% RH, 12 hrs.
light) by measuring the life cycle development of A. socialis as the aforementioned resistant
and susceptible genotypes. Cassava plants were grown in plastic pots and were 4 to 5 weeks
of age at infestation. Plant infestation was accomplished by introducing 20 whitefly adults
into small leaf cages, supported by plastic straws (Figure 6.16). Each leaf cage has a small
lateral opening and with the aid of a pasteur aspirator, A. socialis adults are encouraged to
enter the leaf cages. Five leaf lobes were infested on each plant (total 180) during a 4-hour
period with 3600 adults (Figure 6.17A). A. socialis adults were allowed to oviposit for 24
hours, thereby assuring a uniform population. Leaf cages and adults were then removed and
egg infested plants were placed in the growth chamber (Figure 6.17B). Each leaf lobe was
sequentially numbered to assure accurate data collection on each of the tested genotypes.
(A)
(B)
Figure 6.15. Antibiosis experiments. (A) Greenhouse colony of Aleurotrachelus socialis on
CMC 40 (not-preconditioned); (B) A. socialis pre-conditioned and reared in nylonmeshed cages on resistant and susceptible genotypes.
Figure 6.16. Thirty-day cassava plants grown in plastic pots and with attached leaf cages
conditioned for Aleurotrachelus socialis infestation.
Project IP3: improving cassava for the developing world
Output 6-16
(A)
(B)
Figure 6.17. (A) Cassava plants conditioned for A. socialis infestation; (B) cassava plants
infested with A. socialis eggs in the growth chamber (28r1qC, 60-70% RH, 12 hrs
light).
To determine the biological cycle of A. socialis on resistant and susceptible genotypes, 200
eggs are selected per plant, and an “infestation map” was designed so that daily evaluations
of immature development can be recorded for instar changes, growth characteristics and
survival/mortality. Daily evaluations were done with the aid of a stereomicroscope on the
leaf undersurface. The potted plants, fastened to an iron support rod that allows upwarddownward movement for optimal positioning, are inverted for easy observance. A rubber disk
inserted at the base of the plant stem at the soil line prevents soil loss or plant movement
and injury when the potted plants are invested (Figure 6.18).
Figure 6.18. Inverted cassava plants fastened to an iron support rod allowing easy
observance of Aleurotrachelus socialis development stages with the use of a
stereomicroscope.
The differences in duration of biological stages, time of development, morphological
measurements of immature stages and adult dry weight were analyzed using the Ryan-EinotGabriel-Welsch Multiple range test (REGW). The rate of survival and relationship between
sexes was analyzed with the Chi-Square (X2) test.
Output 6-17
2003 Annual Report
Morphological measurements were done by removing 10 individuals per leaf lobe (40
individuals total per genotype) and taking measurements of the 2nd and 3rd instar nymphs
and the pupal stage. A stereomicroscope with a digital dispositive for micro measurement
(Wild MMS 225/MMS 2535) (Figure 6.19).
Figure 6.19. Digital micrometric measuring devise to determine morphological size of
Aleurotrachelus socialis immatures.
Dry weight of adult whiteflies was done by placing well-developed pupae in the small leaf
cages to prevent adult escape upon hatching. Sexing was done under the stereomicroscope
using adult anal morphological characteristics to separate male and females. Captured
adults from each of the tested genotypes were placed in plastic vials with cotton stoppers and
dried in a Blue-M stove at 37qC for 72 hours. These were weighted on a CAHN C-30
microbalance, sensitive to 1 Pg.
2. Antixenosis Experiments. These experiments compared and determined the ovipositional
and feeding preferences of A. socialis on the five genotypes. One potted plant of each
genotype was randomly placed in a 1m x 1m x 1m wooden, nylon meshed lined cage. Each
30-day-old plant contained only three leaves, numbered in descending order from the top,
middle and lower portions of the plant. This design allowed measurement of both total and
vertical plant preference of oviposition. All plants were of equal height and distributed in a
circular fashion to provide each genotype with an equal chance for oviposition (Figure 6.20).
Five hundred A. socialis adults of the same age and randomly selected from the whitefly
colony being reared on CMC 40, were introduced into the center of each cage. Recorded data
was logarithmically transformed (Log H+1) and significant differences were determined using
the Ryan-Einot-Gabriel-Welsch multiple F test. The variables were, 1) the number of
whiteflies perched on each genotype at 24 and 48 hours after infestation, and 2) the number
of eggs oviposited on each genotype after 48 hours. A visual count of perched adults was
accomplished by carefully opening each cage without disturbing plants and whitefly adults.
Egg counts were made under the stereomicroscope. The evaluation was done 3 times with
four repetitions using a randomized block design in the growth room (28r1qC, 60-70% RH
and 12 hr. light).
Project IP3: improving cassava for the developing world
Output 6-18
(A)
(B)
Figure 6.20. Cassava genotypes (Resistant and Susceptible) placed in nylon meshed cages
and infested with 500 A. socialis adults for free choice ovipositional preference
evaluations in the growth chamber.
Results
1. Antibiosis: No preconditioning.
A. socialis individuals (adult infestation directly from the greenhouse colony = unpreconditioned) feeding on MEcu 64 and MPer 334 had a significantly longer development
period than on the other genotypes (Table 6.5) with 36.8 and 36.4 days respectively. MEcu
72 and MPer 273 resulted in a duration of 35.2 and 33.6 days, while CMC 40 the susceptible
control had a significantly more rapid development of 32.7 days. The duration of the egg
stage ranged from 10.1 (CMC 40) to 11.1 (MEcu 64) and difference between genotypes were
significant (Table 6.6). The greatest differences occurred in the first nymphal instar. Most
rapid development occurred on CMC 40 (4.9 days) and longest on MEcu 64 (6.4 days); MPer
334, MEcu 72 and MPer 273 had first instar duration of 6.1, 6.1 and 5.6 days respectively
(Table 6.6). Significant differences between genotypes also occurred in the 2nd and 3rd
nymphal instars but they were not as dramatic as in the first instar. The duration of the
pupal stage ranged from 9.6 days (CMC 40) to 10.5 days (MPer 334). The relationship
between sexes was approximately 1:1 in all of the genotypes evaluated (Table 6.5).
Table 6.5.
Average development time of Aleurotrachelus socialis (non-preconditioned)
feeding on five cassava genotypes (resistant and susceptible) in the growth chamber.
Genotypes
MEcu 64
MPer 334
MPer 273
MEcu 72
CMC 40
n.
63
45
94
62
152
Average r SD
36.8 ± 2.09
36.4 ± 2.21
33.6 ± 1.55
35.2 ± 2.56
32.7 ± 1.65
1. Ryan-Einot-Gabriel-Welsch Multiple Range test.
Sex Relation
a1
a
c
b
d
1.0 : 1.0
1.1 : 1.0
0.7 : 1.0
1.2 : 1.0
1.3 : 1.0
X2 : NS2
Columns with the same letter are not
significantly different at the 5% level.
2. Independent Test. Female/Male sex relation is 1:1 in all genotypes.
Output 6-19
2003 Annual Report
Table 6.6.
Duration of Aleurotrachelus socialis developmental stages on whitefly resistant
and susceptible genotypes (non-preconditioned) (n=200).
Genotypes
Egg
Nymph 1
6.4 ± 1.08 a
MEcu 64
11.1 ± 0.57 a1
10.7 ± 0.47 b
6.1 ± 1.08 b
MPer 334
10.4 ± 0.49 c
5.6 ± 1.01 c
MPer 273
10.6 ± 0.58 b
6.1 ± 1.05 b
MEcu 72
10.1 ± 0.50 d
4.9 ± 0.85 d
CMC 40
1. Ryan-Einot-Gabriel-Welsch Multiple Range
significantly different at the 5% level.
Nymph 2
Nymph 3
Pupae
4.3 ± 0.71 a
5.1 ± 0.66 a
10.4 ± 0.93 a
4.3 ± 0.95 a
5.0 ± 0.94 a
10.5 ± 1.07 a
3.7 ± 0.64 b
4.4 ± 0.63 b
10.0 ± 0.84 b
4.4 ± 0.63 a
4.6 ± 0.96 b
9.9 ± 0.87 b
3.8 ± 0.59 b
4.4 ± 0.54 b
9.6 ± 0.83 b
(F) test. Columns with the same letter are not
2. Antibiosis: With preconditioned A. socialis.
Development time for A. socialis in this experiment was significantly longer when reared on
MEcu 64 (34.5 days) compared to the other genotypes. It was shortest on CMC 40 (31.8
days) and intermediate for the remaining three genotypes (Table 6.7). Nymphal duration was
longest during the first instar; MEcu 64 was longest (6.3 days) and CMC 40 was shortest
duration (5.0 days). The remaining three genotypes, MPer 334 (5.6 days), MPer 273 (5.5
days) and MEcu 72 (5.7 days) were significantly different from the susceptible genotype CMC
40 (Table 6.8). Differences in development duration in the second and third instars were not
as dramatic as in the first instar. Duration of the pupal stage ranged from MEcu 64 (10.5
days), the longest, to CMC 40 (9.5 days) the shortest and the remaining genotypes,
intermediate (Table 6.8). Results in this experiment were similar to those in the unpreconditioned experiment, however, the values were lower or of shorter duration, indicating
that preconditioning A. socialis effects development time.
Table 6.7.
Aleurotrachelus socialis development time on cassava genotypes (resistant and
susceptible) during preconditioning phase.
Genotypes
N
Average r SD
Sex Relation
1.0 : 1.0
0.9 : 1.0
1.3 : 1.0
0.8 : 1.0
1.5 : 1.0
X2 : NS2
1. Ryan-Einot-Gabriel-Welsch Multiple Range test. Columns with the same letter are not
significantly different at the 5% level.
MEcu 64
MPer 334
MPer 273
MEcu 72
CMC 40
96
124
127
127
140
34.5 ± 1.94
33.0 ± 1.76
32.8 ± 2.22
33.5 ± 1.82
31.8 ± 1.61
a1
bc
c
b
d
Table 6.8.
Duration of Aleurotrachelus socialis development stages on whitefly resistant
and susceptible genotypes (n=200) (preconditioning phase) in the growth room.
Genotypes
Egg
MEcu 64
9.7 ± 0.54 b1
9.4 ± 0.54 c
MPer 334
10.0 ± 0.72 a
MPer 273
9.7 ± 0.62 b
MEcu 72
9.6 ± 0.71 b
CMC 40
1. Ryan-Einot-Gabriel-Welsch
significantly different at the
Nymph 1
Nymph 2
Nymph 3
Pupae
6.3 ± 1.34 a
4.2 ± 0.86 a
4.2 ± 0.70 a
10.5 ± 1.05 a
5.6 ± 0.92 b
3.8 ± 0.83 b
4.0 ± 0.52 a
10.4 ± 1.04 a
5.5 ± 0.99 b
3.8 ± 0.77 b
4.0 ± 0.61 a
9.7 ± 1.21 bc
5.7 ± 1.11 b
4.2 ± 1.01 a
4.1 ± 0.58 a
10.0 ± 1.02 b
5.0 ± 0.87 c
3.8 ± 1.02 b
4.1 ± 0.614 a
9.5 ± 0.95 c
Multiple Range test. Columns with the same letter are not
5% level.
Project IP3: improving cassava for the developing world
Output 6-20
A. socialis survival on resistant genotypes (MEcu 64, MEcu 72, MPer 334 and MPer 273) is
significantly lower than on the susceptible check, CMC 40 (Figures 6.21 and 6.22). First
instar nymphs are the most effected; they have difficulty adhering to the leaf undersurface
and initiating feeding on resistant genotypes. This is not a problem on the susceptible
genotype CMC 40, where establishment and feeding readily occur (Figure 6.21A). In the two
experiments A. socialis survival remained the same (76 and 75% survival) (Figure 6.22).
Without precondition A. socialis survival on MPer 344, MEcu 72, MEcu 64 and MPer 273
were 22.5, 31.0, 31.5 and 47.0% respectively. For preconditioned A. socialis, the results
were similar but the rate of survival was higher for all of the resistant genotypes (Figure
6.22). For example, in the first experiment MEcu 64 survival was 31.5%, wile in the second
it was 48.0%. In both experiments the resistant genotypes had a significantly lower survival
rate than the susceptible genotypes (P=0.05). These results indicate that constant rearing of
A. socialis on resistant genotypes may reduce the effectiveness of the resistant factors. This
will play a role in the deployment of resistant cultivars in field plantings.
(A)
(C)
(B)
(D)
(E)
Figure 6.21. Aleurotrachelus socialis nymphal survival on cassava (A) Susceptible control
CMC 40; (B) Resistant control, MEcu 72; (C) MPer 273; (D) MPer 334; (E) MEcu 64.
Morphological measurements of A. socialis feeding on resistant and susceptible genotypes
show that 2nd and 3rd instar nymphs and pupae were significantly longer on CMC 40 than on
the resistant genotypes (Figure 6.23) (P=0.05). The results for width were similar although
differences were not always significant. A. socialis adult dry weight was significantly lower
when feeding on MEcu 64, followed by MPer 334, MPer 273 and MEcu 74 (P=0.05) (Figure
6.24). All resistant genotypes were significantly lower than the susceptible check, CMC 40,
for both the non-preconditioned and preconditioned A. socialis.
Output 6-21
2003 Annual Report
100
76
% Survival
80
75.5
63.5
62
60
48
47
31.5
40
63.5
31
22.5
20
0
Not preconditional
MEcu 64
Preconditional
MPer 334
MPer 273
MEcu 72(T.R)
CMC 40(T.S.)
Figure 6.22. Percent survival of Aleurotrachelus socialis feeding on five cassava genotypes
(resistant and susceptible) in the growth chamber (28r1qC, 60-70% RH, 12 hrs. light).
Nymph 2
Nymph 3
0.7
0.7
Average (mm)
0.6
0.6
0.5
0.5
0.4
0.4
0.3
c c ab bc a
Average (mm)
c
bc b
c
a
c bc b bc a
0.3
a
a a
0.2
a a
0.2
0.1
0.1
0
0
Largo
Ancho
Largo
Ancho
Pupa
0.7
0.6
b
b
b
Average (mm)
a
b
0.5
a
0.4
b
b
a
b
0.3
0.2
0.1
0
Largo
MEcu 64
MPer 334
Ancho
MPer 273
MEcu 72
CMC 40
Figure 6.23. Morphological measurements of Aleurotrachelus socialis 2nd and 3rd instar
nymphs and pupal stage on five cassava genotypes in the growth chamber.
Project IP3: improving cassava for the developing world
Output 6-22
0.013
Average (mg)
0.011
c
0.009
c
b
a
b
d
a
c
c
c
0.007
0.005
0.003
0.001
Non Preconditional
MEcu 64
MPer 334
MPer 273
Preconditional
MEcu 72(T.R.)
CMC 40(T.S.)
Figure 6.24. Dry weight of Aleurotrachelus socialis adults reared on five cassava genotypes (2
experiments) in the growth chamber.
3. Antixenosis: Free choice feeding preference.
Under a free choice evaluation where A. socialis adults were offered five randomly placed
genotypes, a significantly higher feeding preference occurred on CMC 40 (Figure 6.25)
(P=0.05). There was no significant difference among the remaining resistant genotype,
although feeding was lowest on MEcu 64 (the data was logarithmically transformed (X+1)).
An interaction was noted between experiment, time (hour) and leaf, where the time of
evaluation influenced results on the first leaf, where preference for A. socialis feeding was the
same at 24 and 48 hours. Leaf one, or the upper most leaf, was the most preferred for
feeding (Figure 6.26) in all three experiments, for all genotypes. There was no significant
difference in feeding preference on leaves 2 and 3, but in general, feeding activity was higher
during the initial 24-hour period (Figure 6.26).
Oviposition was effected by genotype. Oviposition on MEcu 64 was significantly lower
(P=0.05) than on the susceptible check (CMC 40) in all three experiments (Figure 6.27). In
experimental 1, all resistant genotypes were significantly lower than CMC 40; however in
experiment 2, only MEcu 64 was significantly lower, and in experiment 3, both MEcu 64 and
MEcu 72 were lower (Figure 6.27). Total oviposition was significantly higher on the upper
leaf in all three experiments (Table 6.9); 75% of the eggs were oviposited on the upper leaf,
15% on the second and 10% on the third leaf.
The combined results for feeding and ovipositional preference and those for mortality and
nymphal development indicate that MEcu 64 along with MEcu 72 are the most A. socialis
resistant genotypes.
Output 6-23
2003 Annual Report
35
a
No. Adults (Average)
30
25
b
20
b
b
15
b
10
5
0
MEcu 64
MPer 334
MPer 273
MEcu 72 (T.R)
CMC 40 (T.S)
Figure 6.25. Free choice Aleurotrachelus socialis feeding trials on five cassava genotypes (3
leaves per plant and 3 repetitions over a 48hr. period.
100
90
24 Hours
48 Hours
a
No. Adults (Average)
80
bc
70
bcd
60
50
40
ab
30
20
abc
e
cd
fg
10
ef
efg
fg
cd
d
e
e
efg
0
Leaf 1
Leaf 2
Leaf 3
Leaf 1
Experiment 1
Leaf 2
Leaf 3
Experiment 2
Leaf 1
Leaf 2
g
Leaf 3
Experiment 3
Figure 6.26. Free choice Aleurotrachelus socialis feeding preferred trials on five cassava
genotypes on three leaves per plant during a 48-hour period.
Project IP3: improving cassava for the developing world
Output 6-24
Table 6.9.
Aleurotrachelus socialis ovipositional distribution on three cassava leaves of five
genotypes in free choice trials.
Leaf Position
Hour
Experimental 1
Experimental 2
Experimental 3
1
418.8 a
661.0 a
1
48
335.3 a
50.0 b
135.7 b
93.2 b
2
48
46.6 b
111.7 b
32.9 c
3
48
1. Ryan-Einot-Gabriel-Welsch Multiple Range (F) test. Columns with the same letter are not
significantly different at the 5% level. Analysis with transformed data. Log (x=1).
450
a
400
ab
a
Average ovipositon
ab
350
a
300
ab ab
250
ab
200
150
100
50
b
b
b
bc
b
c
b
0
Experiment1
MEcu 64
Experiment 2
MPer 334
MPer 273
MEcu 72 (T.R.)
Experiment 3
CMC 40 (T.S.)
Figure 6.27. Free choice ovipositional preference of Aleurotrachelus socialis on five cassava
genotypes (three experiments).
Contributors: Miller J. Gómez, A.C. Bellotti
Collaborators: Myriam C. Duque, Claudia M. Holguín, Bernardo Arias, Diego F. Múnera
Activity 6.6. Studies on the biology and development of biotype B of Bemisia tabaci on
cassava, Manihot esculenta and the wild species, Manihot
carthaginensis.
Whiteflies are major pests of cassava in the Americas, Africa and Asia. Several species are
involved; Aleurotrachelus socialis predominates in Northern South America (Colombia,
Venezuela and Ecuador), while Aleurothrixus aepim is the major species in Brazil. Bemisia
tabaci, a pantropical species prevails in Africa and parts of Asia (i.e. India) where it is the
vector of Africa Cassava Mosaic virus (ACMV) and related viruses. Until the early 1990’s, B.
tabaci biotypes found in the neotropics did not feed on cassava, and it has been speculated
Output 6-25
2003 Annual Report
that the absence of ACMD in the Neotropics may be related to the inability of B. tabaci to
colonize cassava. Biotype B of B. tabaci, has been collected feeding on cassava in the
neotropics. However, recent research at CIAT indicates that cassava is not a very successful
host (see CIAT Pest and Disease Management Annual Report, 2002, pp. 25-35). B. tabaci
feeding on beans (Phaseolus vulgaris) was successfully transferred to cassava only after
completing several generations on other Euphorbiaceae species such as Euphorbia
pulcherrima (Poinsettia) and Jatropha gossypiifolia (Jatropha). However mean longevity,
female fecundity, oviposition and adult survival were low when compared to other whitefly
species feeding on cassava.
This present study evaluates the potential of B. tabaci to adapt to wild Manihot species, such
as M. carthaginensis and compares this to the development of B. tabaci on cultivated cassava
Manihot esculenta, variety MCol 2063.
Methodology
Life table parameters of B. tabaci were evaluated in the growth chamber on potted plants of
M. carthaginensis and the cassava variety MCol 2063. B. tabaci longevity, fecundity,
development time, survival and demography were calculated. B. tabaci populations
originated from a colony maintained on Jatropha gossypiifolia (Euphorbiaceae), in screened
cages (1m x 1m x 1m) for 9 generations (25±5qC, 70±5% RH and 12/12hr. photoperiod).
Longevity and fecundity were evaluated by placing 40 pairs (1m x 1f) of recently emerged
Biotype B of B. tabaci adults in small leaf cages (2.5cm diameter x 2.0cm deep) on test
plants. Every 48 hours adults were moved to another leaf area and this was repeated until
all (40) females died. When males died, they were replaced until female mortality occurred.
Fecundity was estimated by counting eggs oviposited by each female during the 48 hour
period; longevity was estimated by the maximal survival of each female.
Development time and survival were studied by placing 50 adults (25 males + 25 females) in
the small leaf cages and allowed to feed on the leaf undersurface for 6 hours. Adults were
then removed and 200 eggs were selected to evaluate development time from egg to adult and
record nymphal survival and sex ratio. Life tables for B. tabaci were calculated (Price, 1975)
using net reproduction rate (Ro), generation time (T), intrinsic growth rate (rm) of the
population and employing the formula:
¦exp(-rmx)1xmx=1
where:x = age
lx = age specific survival
mx = proportion of female progeny from female x
For the calculated values of rm the corrected age of x + 0.5 were used (Carey, 1993).
Results
The longevity of B. tabaci on M. carthaginensis and M. esculenta (MCol 2063) were similar. It
was two days longer on M. carthaginensis (12 days) than on M. esculenta (10 days) (Figure
6.28). By the end of 6 days 65% of the females on M. carthaginensis and 82.5% of the
females on M. esculenta had died. The average longevity on the two genotypes differed
significantly (Student-Newman-Keuls P<0.05, after K-Wallis P<0.0001).
Project IP3: improving cassava for the developing world
Output 6-26
Survival
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
M.c
0
2
4
6
8
10
12
M.e
14
Female Age (days)
Figure 6.28. Survival curves of Biotype B of B. tabaci feeding on M. carthaginensis and M.
esculenta (MCol 2063).
Oviposition occurred readily on both genotypes but the range was greater on M. esculenta,
although the difference was not significant (K-Wallis P<0.0001, followed by Student-NewmanKeuls P <0.0.5) (Figure 6.29, Table 6.10). The mean ovipositional rate was significantly
higher on M. esculenta (eggs per female/2 days). All females of B. tabaci initiated oviposition
within 48 hours of eclosion on both genotypes. On M. esculenta 72% of the total oviposition
occurred during this 48-hour period while only 35.5% occurred on M. carthaginensis. These
results indicate a preference of B. tabaci to oviposit on M. esculenta. Highest oviposition on
M. esculenta occurred on day 2, while on M. carthaginensis it was on days 4 to 6.
B. tabaci development time was significantly lower or faster on M. carthaginensis than on M.
esculenta (Table 6.11). The development time or life cycle on M. esculenta was 11 days (44.4
days) longer than on M. carthaginensis (33.3), indicating a more rapid adaptation of the
immatures when feeding on M. carthaginensis. Taking into consideration that fecundity was
higher on M. esculenta (8.6 eggs vs. 5.3) (Table 6.10) and combines this with the faster
development time on M. carthaginensis, it results in the intrinsic growth rate (rm) to be the
same for both genotypes (Table 6.11). These results indicate that populations of Biotype B of
B. tabaci, in spite of a higher fecundity on M. esculenta, will be of equal growth rates on both
genotypes.
Output 6-27
2003 Annual Report
Oviposition/Female//2 days
7
M.c
6
M.e
5
4
3
2
1
0
0
2
4
6
8
10
12
14
Female Age (Days)
Figure 6.29. Reproduction curves of Biotype B of B. tabaci feeding on M. carthaginensis and
M. esculenta (MCol 2063).
Table 6.10. Average longevity, fecundity and ovipositional rate (eggs/female/2 days) of
Biotype B of B. tabaci feeding on M. carthaginensis and M. esculenta (MCol 2063).
Parameters
M. carthaginensis
M. esculenta
Average Longevity
5.1 a
3.25 b
Range
2-12
2-10
No insects
30
40
Average Fecundity
5.35 a
8.6 a
Range
1-35
1-41
Average Oviposition Rate
1.05 a
2.64 b
Range
0.25-3.6
0.5-8
Figures followed by different letters across columns indicate significant differences (KruskalWallis) P<0.001, followed by Student-Newman-Keuls P<0.05).
Survival rates were significantly higher on M. carthaginensis (Table 6.11). Results show that
of 200 eggs oviposited on M. carthaginensis, 120 or 60% survived to the adult stage, while
only 55 eggs (36%) survived to adulthood on M. esculenta (Figure 6.30). Immature survival is
a good indication of the eventual ability of a biotype to develop on a genotype. These results
indicate that M esculenta (MCol 2063) is not an optimal host for Biotype B of B. tabaci (Figure
6.30).
Significant differences in the net reproductive rate were obtained between the two genotypes.
It was estimated that at the end of a generation, populations of Biotype B of B. tabaci would
multiply 8.6 times on M. esculenta (MCol 2063), three times greater than on M.
carthaginensis (Table 6.11). This can be explained by total reproduction was less on M.
carthaginensis. One generation of B. tabaci on M. carthaginensis is 35.6 days vs. 44.8 on M.
esculenta. These results indicate that B. tabaci can complete 10 generations per year on M.
carthaginensis and eight on M. esculenta. Population growth of B. tabaci was the same on
both genotypes (Table 6.11). The difference in development time was a more important
criterion for the population increase of B. tabaci on M. carthaginensis, than were the
differences in ovipositional rate. Population increases of B. tabaci on M. esculenta were more
Project IP3: improving cassava for the developing world
Output 6-28
influenced by changes in reproduction rate. It should be noted that the high rate of
oviposition of B. tabaci on M. esculenta can be independent of subsequent development of the
immature stages.
Table 6.11. Demographic parameters of biotype of B. tabaci feeding on M. carthaginensis
and M. esculenta (MCol 2063).
Parameter
M. carthaginensis
M. esculenta
Development time (d)
33.3 a
44.41 b
Rate of survival (%)
60 a
27.5 b
Proportion of females (%)
50.6
50.9
Intrinsic rate of increase ( rm)
Net reproductive rate (Ro) ¦lxmx
0.048
0.048
5.35
8.63
Generation time (T)
35.6
44.76
Days to duplicate population Ln2/ rm
14.4
14.4
Development time: different letters across columns indicate significant differences (K-Wallis
P0.0001, followed by Student-Newman-Keuls P0.05). Rate of survival: (F2=29.9, 1df, P
0.0001).
Figure 6.30. Pupal capsules, pupae and adults of biotype B of B. tabaci feeding on M.
carthaginensis and M. esculenta (MCol 2063).
It can be concluded that Biotype B of B. tabaci can successfully develop on both M. esculenta
(MCol 2063) and M. carthaginensis. In this case, however, it should be noted that these
populations of B. tabaci had already adapted to related Ephorbiaceae, Jatropha, prior to
being evaluated on the two aforementioned genotypes. Previous research has shown that
when the B. tabaci populations originate on an unrelated genotype, such as beans (P.
vulgaris), they do not readily adapt to M. esculenta. These results, however, do provide
evidence that biotype B of B. tabaci can adapt to Wild Manihot species as well as the
cultivated species, M. esculenta and represents a potential threat to cassava production in
the Neotropics.
Contributor: Arturo Carabalí.
Output 6-29
2003 Annual Report
Activity 6.7. Cassava germplasm evaluations for arthropod pest damage at several
localities on the Colombia Atlantic Coast.
The cassava breeding section carries out considerable germplasm evaluation on the Colombia
Atlantic Coast. These trials include observation fields, dialelic crosses, yield trials, advanced
yield trails and regional trials. The entomology section collaborates with the cassava
breeders by carrying out evaluations on arthropod damage during these trials. Arthropod
pests consist of a complex that includes whiteflies (A. socialis), stemborers (Chilomima
clarkei) and mites (Oligonychus peruvianus). All actual data on these evaluations is available
in the cassava entomology and breeding databases. The experiments evaluated were in
Santo Tomás and Pitalito (Atlántico), La Unión (Sucre) and Ciénaga de Oro (Córdoba).
Climatic conditions during the 2002-03 growing cycle were not always favorable and stake
germination, especially of MTai 8, suffered. Populations of the aforementioned pests were not
high; this is in part due to the pest resistance present in some of the germplasm.
Unfortunately, it appears that in certain trials, pesticides were applied and these further
reduced pest populations. In future plantings it is advisable to include additional rows of
susceptible genotypes to enhance pest populations, resulting in a more efficient and uniform
selection pressure.
1. Observation fields at Santo Tomás; F1 selections from CENICAÑA, Valle del Cauca.
A series of hybrids, CM, SM, CT and GM, were selected (F1’s) from trials at CENCAÑA, Valle
del Cauca, and planted out on the Atlantic Coast. Male parent clones consisted of MNGA 19,
KU 50, R60 and different CM and SM hybrids. Female parent clones were R5, R60, R90 and
SM and CM hybrids.
Selected parents had shown good adaptation to the coastal
agroecosystem during regional trials. Two thousand two hundred forty seven clones were
planted in this observational trial.
Results show that nearly all genotypes had some pest damage (Figure 6.31). The principal
pests observed were the whitefly (A. socialis), stemborers (C. clarkei) and mites (O.
peruvianus). Pest populations were low, primarily between 1.0 and 2.0 on the 1 to 6 damage
scales that were employed for each pet species. A rating of 1, indicates an absence of the
pest while a 2.0 rating signifies very low populations. O. peruvianus, for example, was
observed with a damage rating of 3.0 on only 72 of the 2247 genotypes (3.2%); 12 genotypes
had a 4.0 rating and 2 had a 5.0 rating.
Insecticides were applied to plant stems for stemborer control (C. clarkei) and this
undoubtedly influence the low stemborer incidence observed.
Stemborers cause
considerable damage to, and loss of planting material. Since these are observational trials of
F1’s representing important crosses, planting material of selected genotypes needs to be
protected.
Project IP3: improving cassava for the developing world
Output 6-30
2500
No. Clones
1931
1927
2000
1822
1500
A. socialis
Chilomina
Oligonychus
1000
500
316
318
339
0
2
72
0
0
12
0
0
2
0
0
0
0
1
2
3
4
5
6
Population Scale
Figure 6.31. Whitefly (A. socialis) stemborer (C. clarkei) and mite (O. peruvianus) population
levels on 2247 F1 clones planted in observational fields at Santo Thomas, Atlántico,
during 2002-03 crop cycle. (Population scale; 1=absence of the pest, 2= very low
incidence, 6= very high incidence).
2. Observation field, Santo Tomás; dialelic selections.
The 237 genotypes in this planting were hybrids of numerous families including CM 9106,
CM 9178, CM 9966, CM 9958, CM 266, CM 289, CM 291, CM 9921, CM 9945, CM 9907, CM
9954, CM 236, CM 237, CM 238, CM 9703, and several others. These families were selected
from the dialelic cross harvested in 2001 for their outstanding agronomic qualities. Three
pest groups appeared during the crop cycle, whiteflies (A. socialis), mites (O. peruvianus) and
stemborers (C. clarkei). Pest populations were again very low, mostly between 1.0 and 2.0 on
the population and damage scales. Fifty genotypes (22.1%) had no whitefly populations and
176 (77.6%) had a minimum population and no damage rating above 1.0. Stemborer
populations and damage were also low; only 61 clones (27.0%) presented at least one C.
clarkei damage hole, but not more than 5. O. peruvianus populations were observed on 74
genotypes (32.7%); 8 genotypes (3.6%) had population ratings between 4 to 5. Only 21
genotypes had damage ratings of 2.0 or higher, and 5 of 4.0 or higher.
3. Advanced Yield Trial, La Unión (Sucre).
The trial at this locality contained 64 hybrid genotypes, mostly SM and a few CM, that were
selected from 2001 yield trials on the Atlantic Coast. Control clones were SM 1438-2, MTai
8, CG 1141-1 and CM 3306-4, all of which have been grown for numerous cycles in the
region. Populations of O. peruvianus and C. clarkei (stemborers) were very low; only three
clones had exit holes. Whitefly (A. socialis) populations were also low, never rising above an
average of 1.6 (< 200 individuals per leaf) and no leaf damage was presented. The clone CM
4919-1, noted for high leaf retention was multiplied in this field. Pest damage was minimal
(Figure 6.32).
4. Regional Trial, Ciénaga de Oro (Córdoba), 2002-03 cycle.
The 40 hybrids being evaluated in this trial originated from selection trials on the Atlantic
Coast over the past 8 years. They were grown in 3 repetitions. Although three pest problems
appeared during the crop cycle, mites (O. peruvianus), stemborers (C. clarkei) and whiteflies
(A. socialis), populations remained low. Whitefly populations ranged from 1.3 to 2.1 (< 200
individuals per leaf) and no leaf damage was recorded. O. peruvianus populations ranged
from 1.0 to 4.0 with 26 clones (65%) having no mites (1.0) and only one clone (SM 1127-8)
Output 6-31
2003 Annual Report
with a population rating of 4.0 and a damage rating of 3.0. For all other clones the damage
rating did not rise above 2.0. Stemborer damage was recorded on only 3 of the 40 clones
with one clone, SM 643-17, having 4 exit holes. The aforementioned control clones also all
had low pest populations and damage.
Figure 6.32. Clone CM 4919-1 noted for high leaf retention being grown at La Unión (Sucre)
during 2002-03 crop cycle.
5. Yield Trials (Experiments 1, 2, 3) Santo Tomás (Atlántico) 2002-03.
These are preliminary yield trials (replicated) that originate from selected crosses of
promising parents selected on the Atlantic Coast. The progeny of these crosses are planted
in “observation fields” and may go through several evaluations. One hundred genotypes in
three repetitions were evaluated for pest population and damage. The highest score of the 3
repetitions was recorded for each of the 3 experiments. The pests observed were whiteflies
(A. socialis), mites (O. peruvianus) and stemborers (C. clarkei).
Although whiteflies were present throughout all 3 experiments, populations never reached
2.0 on any of the genotypes and no leaf damage was recorded. O. peruvianus populations
and damage was high in all three experiments.
In experiment 1, populations were
consistently above 3.0 and more often above 4.0 (79 clones, 81.4%). This population level
indicates that 26-50% of the leaf undersurface is covered with mite webbing. In experiment
2, O. peruvianus populations were 4.0 or above in 75.8% of the clones and in Experiment 3,
39% had a rating of 4.0 or higher. Stemborer damage occurred only on one clone in
experiment 2 and on only 3 clones in experiment 3. The recorded evaluations for all
experiments and each clone are available in the database.
6. Advanced Yield Trial, Pitalito (Santo Tomás, Atlántico).
This trial was comprised of 68 SM and CM clones, grown in 3 replicates and previously
selected from observation trials on the Atlantic Coast. Whitefly (A. socialis) populations and
damage were very low with populations not recorded above 2.1 (only on one clone) O.
peruvianus populations ranged from 1.0 to 4.0, with 26 clones (38%) having a rating of 3.0 or
above and 13 clones (19%) with a damage rating of 3.0 or higher. Stemborer (C. clarkei)
damage occurred on 22 clones (29.4%). The clone SM 2550-60 had 6 stemborer exit holes.
Contributors: Bernardo Arias, Anthony C. Bellotti.
Collaborators: Gustavo Trujillo, José María Guerrero, Carlos Ñañes.
Project IP3: improving cassava for the developing world
Output 6-32
Activity 6.8. Evaluation of cassava germplasm for resistance to cassava green mite,
Mononychellus tanajoa.
Rationale
Mites are major pests of cassava, causing considerable foliage damage and yield losses on
two continents, South America and Africa. Of the more than 40 species reported feeding on
cassava, the most frequent are Mononychellus tanajoa (syn=M. progresivus), M. caribbeanae,
Tetranychus cinnabarinus and T. urticae (also reported as T. bimaculatus and T. telarius).
Cassava is the major host for the Mononychellus species, whereas the Tetranychus species
have a wide host range. Other mite species, although numerous, are not important
economically and are usually only sporadic or very localized feeders. Oligonychus peruvianus
may often cause crop damage on the Colombian Atlantic Coast.
The cassava green mite (CGM), M. tanajoa, is reported causing crop losses in the American
and Africa, especially in seasonally dry regions of the lowland tropics. M. tanajoa is native to
the Neotropics, first being reported in 1938, in Northeast Brazil. It first appeared in Africa (in
Uganda) in 1971 and by the mid 1980’s had spread across most of the African cassava belt,
occurring in 27 countries and causing yield losses ranging from 13 to 80%. CGM
populations preferably feed on the undersides of young leaves, which develop a mottled
whitish-to-yellow appearance. Heavy CGM attacks occur only during the dry season, where
high populations can cause severe defoliation and death of the apical bud and shoots.
A substantial effort has been made by CIAT, IITA and regional and national research
programs to identify cassava genotypes resistant to M. tanajoa and develop mite resistant
hybrids. Of the nearly 5000-landrace cultivars in the CIAT cassava germplasm bank, about
300 or 6% have been identified as having low to moderate levels of resistance. The search for
CGM resistance in Manihot esculenta is a continual activity, and, in addition, this search is
now being expanded into the wild Manihot species (see Activity 6.9) in this report. In close
collaboration with the cassava breeding and genetics section, systematic evaluations of
selected progeny from specific crosses are often carried out and are presented in this report.
1. Interspecific crosses with wild Manihot species.
Interspecific crosses were made using the wild Manihot species M. flabellifolia and several M.
esculenta genotypes. The resulting progeny were evaluated in the field at CIAT using a 1 (no
damage) to 6 (severe leaf yellowing and necrosis and apical defoliation) damage scale. Two
groups of genotypes were evaluated; the first group consisted of F1’s high in beta-carotene
from an interspecific cross. The second group of interspecific F1’s was developed for mite (M.
tanajoa) resistance.
Results are presented by the number of F1’s that resulted in mite damage evaluations
between 1 to 3 on the 1 to 6 damage scale. A 1 to 3 rating indicates high to moderate levels
of resistance. In the first trial of the 251 F1 genotypes 84 or 33.5% had damage ratings
between 1 to 3 (Table 6.12). This represents an acceptable level of resistance and selected
progeny will be evaluated in subsequent trials.
In the second grouping of the 593 F1 progeny from the interspecific cross (GY 200248), 533,
or 90% had mite damage ratings between 1 to 3 (Table 6.13). This is a very high percentage
of resistant progeny, indicating that the first year of evaluations (2002) for mite resistance of
these genotypes resulted in a successful or accurate selection.
Output 6-33
2003 Annual Report
Table 6.12. Interspecific crosses with wild Manihot sp evaluated for cassava green mite (M.
tanajoa) damage at CIAT.
Cross
Parents
Plants
No. Plants with
Identification
Evaluated
1-3 Damage Rating*
Female
Male
CW 73
CM1585-13
OW 284-1
12
7
CW 177
OW 132-2
CM 1585-13
63
16
CW 184
OW 180-1
MCol 1734
5
5
CW 186
OW 181-2
CM 1585-13
4
2
CW 188
OW 181-2
MCol 1734
19
19
CW 207
OW 280-1
CM 1585-13
127
21
CW 212
OW 284-1
MCol 1734
7
6
CW 251
MCol 1734
OW 189-1
5
0
CW 256
MCol 1734
OW 280-1
9
8
251
84
* Ratings based on 1 (no damage) to 6 (general leaf yellowing, necrosis and apical defoliation)
mite damage scale.
Table 6.13. Progenies from interspecific crosses with wild Manihot sp and M. esculenta
evaluated for mite damage at CIAT, Palmira.
Cross
Plants
No. Plants with
Parents
Identification
Evaluated
1-3 Damage Rating*
Female
Male
CW 74
CM 2177-2
CW 65-77
1
0
CW 75
CM 3306-4
CW 66-60
7
7
CW 76
CM 3306-4
CW 68-3
9
8
CW 77
CM 7951-5
CW 65-77
5
4
CW 78
CM 7951-5
CW 66-19
4
4
CW 79
CM 7951-5
CW 66-62
1
1
CW 80
CM 7951-5
CW 67-42
5
4
CW 81
CM 7951-5
CW 67-98
3
3
CW 213
SM 805-15
CW 67-39
1
1
CW 214
SM 805-15
CW 67-87
11
7
CW 215
SM 909-25
CW 66-60
7
7
CW 217
SM 1219-9
CW 65-77
17
16
CW 218
SM 1219-9
CW 66-73
12
11
CW 219
SM 1219-9
CW 66-74
3
2
CW 220
SM 1219-9
CW 67-23
12
10
CW 223
SM 1460-1
CW 66-19
11
11
CW 224
SM 1460-1
CW 66-60
17
17
CW 225
SM 1460-1
CW 66-62
29
29
CW 226
SM 1460-1
CW 66-73
11
10
CW 227
SM 1460-1
CW 68-3
3
3
CW 229
SM 1511-6
CW 67-87
16
13
CW 230
SM 1565-15
CW 66-19
4
4
CW 231
SM 1565-15
CW 66-60
24
20
CW 232
SM 1665-2
CW 66-19
23
17
CW 233
SM 1665-2
CW 66-60
1
1
CW 234
SM 1665-2
CW 66-74
31
25
Project IP3: improving cassava for the developing world
Output 6-34
Cross
Identification
CW 235
CW 236
CW 237
CW 238
CW 239
CW 240
CW 241
CW 242
CW 243
CW 244
CW 245
CW 246
CW 247
CW 248
CW 257
CW 258
CW 259
CW 260
CW 261
Parents
Plants
No. Plants with
Evaluated
1-3 Damage Rating*
Female
Male
SM 1665-2
CW 67-87
113
105
SM 1669-5
CW 66-19
30
30
SM 1669-5
CW 66-60
8
8
SM 1669-5
CW 66-62
1
1
SM 1669-5
CW 66-73
4
4
SM 1669-5
CW 66-74
29
26
SM 1669-5
CW 67-23
9
9
SM 1669-7
CW 67-87
4
4
SM 1741-1
CW 66-19
6
6
SM 1741-1
CW 66-60
15
15
SM 1741-1
CW 66-62
1
1
SM 1741-1
CW 67-91
9
8
SM 1778-45
CW 66-19
3
3
SM 1778-45
CW 67-45
2
2
MTAI 8
CW 65-77
21
14
MTAI 8
CW 66-60
25
25
MTAI 8
CW 66-73
37
31
MTAI 8
CW 66-74
4
3
MTAI 8
CW 67-123
4
3
593
533
* All are interspecific crosses - Ratings based on 1 (no damage) to 6 (general leaf yellowing,
necrosis and apical defoliation) mite damage scale.
2. Second Regional trial of Industrial Purpose Cassava Genotypes.
Cassava genotypes developed for industrial (starch) uses are usually high yielding and have a
high starch content. This trial was carried out at Caloto, Cauca using three replicates of 25
plants each. Mite populations (M. tanajoa) during this trial were high and 28 genotypes were
evaluated.
Results indicate that all genotypes were highly susceptible to M. tanajoa (Table 6.14). On all
of the genotypes mite damage ratings ranged from 4.0 to 6.0 in all three replications. The
clones CM 7463-2, SM1965-1, SM2198-4 and MBra 383 had damage ratings of 4.0 in all
three replications and they did not show a generalized or overall yellowing of foliage. This
indicates that there may be some low levels of resistance present. These genotypes will be
further evaluated in the laboratory. The high levels of mite susceptibility in these genotypes,
that are being recommended for industrial use planting, is alarming and needs to be taken
into consideration if these genotypes are to be grown in areas with prolonged (3 to 6 months)
seasonally dry periods.
3. Progeny of MEcu 72 x MCol 2246 cross evaluate for mite resistance.
The variety MEcu 72 is highly resistant to whiteflies (Aleurotrachelus socialis) and has often
displayed low to moderate levels of resistance to mites (M. tanajoa). MEcu 72 was crossed
with the whitefly susceptible varieties to try to determine the inheritance of resistance in the
progeny. A selected group of these progeny were evaluated for mite resistance at the CIAT
farm in Santander de Quilichao. Three replications of 6 plants each, of 700 genotypes were
evaluated.
Output 6-35
2003 Annual Report
Table 6.14. Regional trial of industrial purpose cassava genotypes evaluated for mite
resistance at Caloto, Cauca, Colombia.
Damage Ratings*
Clones
Rep. I
Rep. II
Rep. III
CM 6660-21
5
5
6
CM 7463-2
4
4
5
CM 8370-10
6
5
6
CM 8370-11
5
4
5
SM 1520-16
4
6
6
SM 1520-18
4
4
5
SM 1642-22
4
5
4
SM 1660-4
4
5
5
SM 1779-7
5
6
6
SM 1855-15
5
5
5
SM 1871-33
6
6
5
SM 1959-1
6
6
SM 1965-1
4
4
4
SM 2052-4
4
5
5
SM 2058-2
4
4
5
SM 2073-1
5
5
5
SM 2085-7
5
5
5
SM 2141-1
6
5
5
SM 2160-2
6
6
6
SM 2198- 4
4
4
4
SM 2211-3
6
4
5
MTai 8
6
6
6
MPer 183
5
5
6
CM 523-7
6
6
6
HMC 1
6
6
5
MBra 383
4
4
4
CM 7951-5
6
6
5
SM 1219-9
5
6
5
* All are interspecific crosses - Ratings based on 1 (no damage) to 6 (general leaf yellowing,
necrosis and apical defoliation) mite damage scale.
Results show that of the 700 genotypes evaluated, 10 genotypes had relatively low damage
ratings between 2 to 4 on the 1 to 6 damage scale (Table 6.15). This low to moderate level of
resistance is acceptable in field trials, especially if adequate biological control (mite
predators) is available.
These 10 selected genotypes were evaluated in the laboratory in ovipositional trials. Rate of
oviposition by female mites has been found to be an indication of possible M. tanajoa
resistance. Leaf discs (2.5 to 3.0 cm2) of each tested genotype was placed on a layer of water
saturated cotton in petri dishes. The experimental design consisted of 20 repetitions of each
genotype and oviposition was recorded over a three-day period. The genotype CMC 40 was
employed as the susceptible control. CMC 40 consistently has mite damage ratings between
5 to 6 in field trails. These were no choice ovipositional trials, meaning that M. tanajoa
Project IP3: improving cassava for the developing world
Output 6-36
females were obliged to oviposit on only one genotype.
carried out.
Four separate experiments were
Table 6.15. Selected progeny from a MEcu 72 x MCol 2246 cross evaluated for mite
resistance at CIAT, Santander de Quilichao.
Damage Ratings*
Clone
Rep. I
Rep. II
Rep. III
Average
CM 8996-104
3
2
3
2.7
CM 8996-167
3
3
3
3.0
CM 8996-347
4
3
3
3.3
CM 8996-440
3
2
2
2.3
CM 8996-487
3
4
3
3.3
CM 8996-509
2
2
4
2.7
CM 8996-557
2
3
4
3.0
CM 8996-665
3
3
4
3.3
CM 8996-692
3
3
3
3.0
CM 8996-746
4
2
3
3.0
* All are interspecific crosses - Ratings based on 1 (no damage) to 6 (general leaf yellowing,
necrosis and apical defoliation) mite damage scale.
Results show that oviposition was highest on CMC 40 and lowest on CM 8996-487 and CM
8996-509 (Table 6.16). Oviposition, when compared to the susceptible CMC 40, was reduced
by 70 to 75% on CM 8996-47, by 57 to 67% on CM 8996-509, by 50 to 57% on CM 8996746, by 50 to 57% on CM 8966-440, by 37.5 to 60% on CM 8996-167 and by 43 to 44% on
CM 8996-557. These laboratory ovipositioned results support the field observations and
evaluations, indicating possible low to moderate levels of resistance in these genotypes to M.
tanajoa.
Table 6.16. Mononychellus tanajoa oviposition on ten selected cassava genotypes originating
from a MEcu 72 x MCol 2246 cross in no choice laboratory trials.
Trial I*
Trial II
Trial III
Trial IV
Genotype
CMC 40
10
9
7
8
CM 8996-665
9
4
CM 8996-746
5
3
CM 8996-440
5
3
CM 8996-167
4
3
5
CM 8996-487
4
2
2
CM 8996-692
7
4
CM 8996-104
7
CM 8996-347
6
CM 8996-557
5
4
CM 8996-509
3
3
* Each trial average of 20 repetitions.
4. Ovipositional studies on selected interspecific progeny.
Genotypes displaying low mite damage were selected from an interspecific cross that had M.
flabellifolia as one of the parents. This field trial, carried out at CIAT had high mite
population and damage. Approximately one-half of the interspecific genotypes resulted in
Output 6-37
2003 Annual Report
low mite damage ratings. The selected genotypes were evaluated for ovipositional rates in the
laboratory using the methodology described in the previous activity.
Five separate
experiments involving 19 genotypes were carried out. CMC 40 was employed as the
susceptible control. Results show that the majority of the genotypes had ovipositional rates
below that of CMC 40 (Table 6.17). For most of the genotypes there was a reduction of 37 to
62% in oviposition when compared to the susceptible control, CMC 40.
Table 6.17. Mononychellus tanajoa ovipositioned trials on selected interspecific genotypes
originating from a Manihot flabellifolia x M. esculenta cross in no choice laboratory
trials.
Genotype
CMC 40
CW 67-42.2
CW 67-139.2
CW 67-153.2
CW 67-160.2
CW 67-104.2
CW 66-10
CW 65-35.1
CW 67-43.2
CW 67-69.2
CW 67-5.1
CW 67-112.2
CW 67-29.1
CW 67-128.2
CW 67-57.2
CW 67-79.2
CW 67-56.2
CW 67-87.2
CW 67-129.2
CW 67-20.1
Parents
Female
Male
MFla
MFla
MFla
MFla
MFla
MFla
MFla
MFla
MFla
MFla
MFla
MFla
MFla
MFla
MFla
MFla
MFla
MFla
MFla
437-007(6)
437-007(6)
437-007(6)
437-007(6)
437-007(6)
437-007(3)
437-007(3)
437-007(6)
437-007(6)
437-007(3)
437-007(6)
437-007(3)
437-007(6)
437-007(6)
437-007(6)
437-007(6)
437-007(6)
437-007(6)
437-007(3)
MCol 2215
MCol 2215
MCol 2215
MCol 2215
MCol 2215
CMC2766-5
CG 501-16
MCol 2215
MCol 2215
MCol 2215
MCol 2215
MCol 2215
MCol 2215
MCol 2215
MCol 2215
MCol 2215
MCol 2215
MCol 2215
MCol 2215
Trial I
19
12
11
11
10
9
9
7
Trial II
12
Trial III
7
Trial IV
13
Trial V
10
8
8
7
6
5
7
9
7
5
5
6
4
12
10
8
7
9
* Average of 20 repetitions.
Contributors: José María Guerrero, Adriano Muñoz
Activity 6.9. Wild Manihot species as a source of resistance to cassava arthropod
pests.
Several cassava arthropod pests will significantly reduce root yield. Emphasis is given to two
complementary systems, host plant resistance and biological control, for the effective
environmentally sound and low cost methods of controlling cassava pests. Different levels of
resistance to cassava pests have been identified within M. esculenta. For example resistance
to mealybugs, lace bugs, stemborer and burrower bugs (within low HCN varieties) is very low;
resistance levels to mites is low to moderate; resistance to thrips and whiteflies is moderate
to high, while no resistance to hornworms and white grubs have been identified.
The wild species of Manihot offer a potential “source” of resistance genes for the control of
major cassava pests. This “source” of resistance has already been exploited for the control of
Project IP3: improving cassava for the developing world
Output 6-38
Africa Cassava Mosaic Disease (ACMD) in Africa. ACMD resistance was obtained by
intercrossing cassava varieties with Manihot glaziouii and other species of Manihot.
Interspecies hybrids were backcrossed to cassava and this resulted in varieties highly
resistant to ACMD. This research was initiated in the 1930’s and 1940’s, when modern
biotechnology tools and information were not available.
The development of pest and disease resistant varieties using interspecific crosses with wild
Manihot species was difficult and slow. Therefore, the use of wild Manihot species as a
source of pest and disease resistant genes was not effectively pursued. Recent advances in
genetic mapping, gene transfer, transformation and genetic engineering allow for a more
efficient production of new cassava varieties resistant to pests and diseases (Fregene, 2002).
The objectives of the present research is to evaluate species of wild Manihot to determine
their potential as a source of resistance to three major cassava pests, mites (Mononychellus
tanajoa) whiteflies (Aleurotrachelus socialis) and mealybugs (Phenacoccus herreni). Mites,
whiteflies and mealybugs cause significant yield losses in the Americas, Africa and Asia.
This research was divided into two parts; the first consists of the acquisition and
establishment of vegetative material of the Manihot species. Different methodologies were
used, including rooting techniques, soil sand mixtures and soil source (site or location). The
second part consists of infesting genotypes of the different wild species, as well as control
genotypes with the aforementioned arthropod pets and evaluating population dynamics,
behavior, survival and damage.
Multiplication of Wild Manihot Species
Developing a methodology for the multiplication of wild Manihot species was both difficult
and time consuming (see PE-1 Annual Report, 2002, pp 67-69) (Figure 6.33). Originally,
attempts were made to establish genotypes of four wild Manihot species, M. flabellifolia, M.
carthaginensis, M. peruviana and M. tristis. Several methods and rooting media were tested
for achieving the establishment of the wild species. The rooting media that gave the best
results was a mixture of three parts construction sand (this is a coarse grain of sand that
permits good drainage) and one-part rice husks. However, some genotypes rooted and
established more successfully than others. The genotype that resulted in the highest
percentage of germination were M. carthaginensis (MCTH 37-8); a genotype of M. peruviana
(203-3) and of M. flabelliflora (MFLA 444-033). Although stem rooting was achieved in many
of the genotypes of the wild species, plant establishment was not always successful. Since
vigorous growing potted plants with adequate foliage was needed for pest infestation,
establishment and evaluation, many of the wild genotypes were discarded from this final
phase of the project.
Genotypes from the wild species, M. flabellifolia and M. peruviana were selected for the pest
infestation and evaluation phase. These were compared to several M. esculenta cassava
varieties, including several genotypes from Vaupés Department, that are being cultivated by
indigenous peoples in that region (Tables 6.18a and 6.18b).
Output 6-39
2003 Annual Report
Table 6.18a. Manihot genotypes evaluated for resistance to mites (Mononychellus tanajoa)
and mealybugs (Phenacoccus herreni).
Species
Manihot esculenta
Manihot esculenta (Vaupés)
M. flabellifolia
M. peruviana
Genotype
CMC 40
CM 7395
Ecu 72
Ibacaba
Cassava de Mico (Roja)
Abeja
Cassava de Garza
Cassava de Piña
Abiyú
MFla 444-002
MPer 417-003
MPer 417-005
Table 6.18b. Manihot genotypes evaluated for resistance to whiteflies (Aleurotrachelus
socialis).
Species
Genotype
CMC 40
Manihot esculenta
Manihot esculenta (Vaupés)
M. flabellifolia
CM 7395
Ecu 72
Nupará
Flores
MFla 444-002
MPer 417-003
M. peruviana
MPer 417-005
Mite (M. tanajoa), mealybug (P. herreni) and whitefly (A. socialis) colonies were established in
the greenhouse or screen house on M. esculenta (usually variety CMC 40). Mite infestation of
the test genotypes were made by placing a cassava leaf lobe from the colony on the test plant;
this resulted in an infestation of 150 to 200 individuals. Whitefly infestations were made
with 200 adults (100 males and 100 females), while mealybug infestations were done by
placing ovisacs on plant stems at the leaf axial. All test plants were place in four or six
chambered wooden cages enclosed with a fine nylon mesh to prevent arthropod movement
into or out of the cages (Figure 6.34). Each chamber was .50 L x .50 W x 1.0 H in meters.
Project IP3: improving cassava for the developing world
Output 6-40
Vegetative Multiplication of Wild and Domesticated Manihot species
First Planting
Second Planting
Vegetative Material
from CEUNP
Third Planting
Vegetative Material
from CEUNP
CIAT Methodology
using stem cuttings
Planting in germination
chamber
Planting in
polystyrene cups
x
x
x
x
x
x
x
Rooting in water
Rooting in soil/sand
Rooting in humidity chamber
Rooting in soil from Vivero Marinela
Rooting in construction sand.
Rooting of primary cuttings
Rooting in rice husks
Direct rooting
Planting in black
bags
Vegetative material
from Santander
Vegetative cuttings
from germplasm bank
Vegetative cuttings
from Vaupés
Figure 6.33. Flowchart of methodology for the vegetative multiplication of Manihot species.
Evaluations were carried out periodically for both pest populations and plant damage using a
1 (low) to 6 (high) population and damage scales. Mite evaluations were made every 5 days
during a 4-week period (4 evaluations). Mealybug and whitefly evaluations were carried out
every 10 days and a total of 6 evaluations were done for each pest species (Figure 6.34).
Output 6-41
2003 Annual Report
Whiteflies
Mites
x 4 cages, each with 6
compartments
x 8 Manihot genotypes, 3
repetitions of each
x Experiments in
greenhouse T. 20-25qC,
50-70% RH.
Mealybugs
x 6 cages, each with 6
compartments
x 3 repetitions of each of 12
Manihot sp. Genotypes
x Experiments in cassava
patio: T. 25-30qC, 40-70%
RH.
PEST
INFESTATION
200 Whitefly adults
1/1 Male/Female ratio
Mite infested leaves
placed on genotypes
150-200 individuals
Mealybug ovisacs
placed on stems at
leaf axels
Observations: Mite evaluations made every 5 days during 4-week period.
Whitefly and mealybug evaluations made every 10 days; a total of 6
evaluations.
Criteria: 1-6 Populations and damage scales employed for each pest species; % of leaves
infested noted.
Figure 6.34. Bioassay methodologies for the evaluation of Wild Manihot species and cassava
varieties and three cassava pests, mites, whiteflies and mealybugs.
Project IP3: improving cassava for the developing world
Output 6-42
Results
Mite damage, infestation and percentage of infested leaves were significantly different for the
genotypes evaluated (AOV; F(11.132) = 66.72, F(11.132) = 29.86 and F(11.132) = 5.61 respectively
with a P<0.005, for the three variables).
Significant differences in damage were observed between three wild Manihot genotypes, MFla
444-002, MPer 417-003 and MPer 417-005 (Turkey and Newman-Kewls Multiple
Comparison) and all other genotypes (Figure 6.35). On the 1 to 6 damage scale, these three
genotypes were between 2.0 and 3.0 rating, with MFla 444-002 being the lowest. All the M.
esculenta genotypes had a damage rating of 5.3 to 5.7 after 25 days of infestation. Results
for infestation levels were similar in that the genotypes evaluated separated into two groups.
The three wild Manihot genotypes, MFla 444-002, MPer 417-003 and MPer 417-005 were
significantly lower (Turkey and Newman Kewls Multiple Comparison) than all of the M.
esculenta genotypes (Figure 6.36). The number of leaves infested was similar for all the
genotypes tested and there was no significant difference between the wild Manihot genotypes
and the M. esculenta genotypes. These results indicate that the wild Manihot genotypes may
possess low to moderate levels of resistance to M. tanajoa.
7
MFLA 444-002
Average Damage
6
CM 7395
5
ECU 72
CMC 40
4
MPER 417-003
MPER417-005
3
IBACABA
2
YUCA DE MICO ROJA
ABEJA
1
YUCA DE GARZA
YUCA DE PIÑA
0
10
15
20
25
ABIYU
Days after Infestation
Figure 6.35. Average damage ratings (using a 1 to 6 damage scale) of Manihot species after
25 days of mite (Mononychellus tanajoa) infestation.
Mealybug (P. herreni) damage, infestation level and percentage of infested leaves were
significantly different for the genotypes evaluated (AOV F(11.204) = 2.83, F(11.204) = 8.24, and F
(11.204) = 4.41 respectively, with a P<0.005 for the three variables).
MPer 417-003 had a significantly lower damage level than the other genotypes tested (Figure
6.37). The other two wild Manihot genotypes (MPer 417-005 and MFla 444-002) also had
lower damage ratings than the M. esculenta genotypes but the differences were not always
significant. MPer 417-003 had a damage rating of 2.3 60 days after infestation, while MPer
417-005 and MFla 444-002 had damage ratings of 3.6 and 4.6 respectively. All M. esculenta
genotypes had damage ratings of 5.0 or higher (Figure 6.37). Infestation levels were lower for
MPer 417-003, MFla 444-002 and CM 7395 and were significantly lower than all the
remaining genotypes (Figure 6.38). The three previously mentioned genotypes all had
Output 6-43
2003 Annual Report
infestation levels below 3.0 while all the remaining genotypes had ratings of 5.0 or higher.
MPer 417-003 and MFla 444-002 can be seen as having low to moderate levels of resistance
to P. herreni since both also had low damage levels. However, CM 7395, although it has a
low infestation level had a high damage level of 5.0. MPer 417-003 and MFla 444-002 also
had low percentage of leaves infested.
7
MFLA 444-002
6
CM 7395
5
CMC 40
Average Infestation
ECU 72
MPER 417-003
4
MPER417-005
IBACABA
3
Y UCA DE MICO ROJA
2
ABEJA
Y UCA DE GARZA
1
Y UCA DE PIÑA
ABIY U
0
10
15
20
25
Days after Infestation
Figure 6.36. Average infestation level ratings (1 to 6 scale) of Manihot species after 25 days of
mite (Mononychellus tanajoa) infestation.
7
MFLA 444-002
Average Damage
6
CM 7395
ECU 72
5
CMC 40
MPER 417-003
4
MPER417-005
IBACABA
3
YUCA DE MICO ROJA
ABEJA
2
YUCA DE GARZA
YUCA DE PIÑA
1
ABIYU
0
10
20
30
40
50
60
Days after Infestation
Figure 6.37. Average damage ratings (1 to 6 damage scale) of Manihot species after 60 days
of mealybug (Phenacoccus herreni) infestation.
Whitefly damage, infestation and percentage of infested leaves were significantly different for
the genotypes evaluated (AOV; F(7.136) = 19.54, F(7.126) = 12.3 and F(7.126) = 17.12 respectively,
with a P<0.005 for the three variables).
Project IP3: improving cassava for the developing world
Output 6-44
7
MFLA 444-002
CM 7395
6
Average Infestation
ECU 72
CMC 40
5
MPER 417-003
MPER417-005
4
IBACABA
YUCA DE MICO ROJA
3
ABEJA
YUCA DE GARZA
2
YUCA DE PIÑA
1
ABIYU
0
10
20
30
40
50
60
Days after Infestation
Figure 6.38. Average infestation level ratings (1 to 6 scale) of Manihot species after 60 days of
mealybug (Phenacoccus herreni) infestation.
Results in damage with whiteflies differed from the other two pest species, mites and
mealybugs. The three wild Manihot genotypes, MEcu 72 and NUPARA all had significantly
lower damage ratings than the remaining M. esculenta genotypes (Figure 6.39). The three
wild Manihot genotypes and MEcu 72 had a 1.0 damage rating, and NUPARA had a 2.0 rating
while the three remaining M. esculenta genotypes (CM 7395, CMC 40 and FLORES) all had
ratings of 5.3 or higher. MEcu 72 had been selected as the most resistant M. esculenta
genotype from previous evaluations of the cassava germplasm bank.
The three wild Manihot genotypes (MFla 444-002, MPer 417-003 and MPer 417-005) all had
very low infestation levels (1.3, 1.3 and 1.6 respectively) while MEcu 72 and NUPARA both
had infestation levels of 3.0 (Figure 6.40). The percentage of leaves infested was lowest for
the three wild Manihot species, intermediate for MEcu 72 and NUPARA and highest (nearly
100%) for CM 7395, CMC 40 and FLORES. These results indicate that there exist high levels
of resistance to whiteflies (A. socialis) in the wild Manihot species. Higher levels of resistance
are indicated for whiteflies than for the other two pests evaluated, mites and mealybugs.
Output 6-45
2003 Annual Report
7
MFLA 444-002
6
Average Damage
CM 7395
5
ECU 72
CMC 40
4
MPER 417-003
3
MPER417-005
2
NUPARA
FLORES
1
0
5
15
25
35
45
55
Days after Infestation
Figure 6.39. Average damage ratings (1 to damage scale) of Manihot species after 55 days of
whitefly (Aleurotrachelus socialis) infestation.
Overall results show that the possibility of using the wild Manihot species as a source of
resistance to cassava pests has considerable potential for the future. This line of research
needs to be continued and expanded.
7
MFLA 444-002
Average Infestation
6
CM 7395
5
ECU 72
4
CMC 40
3
MPER 417-003
2
MPER417-005
NUPARA
1
FLORES
0
5
15
25
35
45
55
Days after Infestation
Figure 6.40. Average infestation level ratings (1 to 6 scale) of Manihot species after 55 days of
whitefly (Aleurotrachelus socialis) infestation.
Contributor: Maritza Burbano.
Project IP3: improving cassava for the developing world
Output 6-46
Activity 6.10 Identification of genomic regions responsible for conferring resistance
to whitefly (Aleurotrachelus socialis) in cassava
Rationale
The whitefly (Aleurotrachelus socialis) is one of the most serious pests and disease vectors
that affect agricultural production around the world. In cassava (Manihot esculenta Crantz),
the whitefly causes from 70 to 80 percent economic losses. The most important source of
resistance genes is the genotype M Ecu 72. Due to the whitefly’s importance as a pest, it is
necessary to understand the nature of genes that confer resistance to the whitefly in
genotypes such as M Ecu 72. For this purpose F1 segregation and the genetic expression of
the cross M Ecu 72 (resistant genotype) x a very susceptible genotype (M Col 2246) and
molecular markers are being used. This will help accelerate selection of whitefly-resistant
materials, as well as isolate resistance genes (R genes). Genetic and molecular studies
(Richter and Ronald, 2000) have shown that the R genes are clustered in the genome of
several species. They display an apparent multiallelic structure, or they group as genetically
separate loci. Different genes determining resistance to insects and nematodes have been
reported within the same cluster in tomatoes (Rossi et al., 1998). R genes are thought to be
functionally and evolutionary related. The sequences of several R-gene clusters from rice,
tomatoes and lettuce have now shed light on the molecular mechanisms leading to their
evolution. As suggested by Lefebvre and Chèvre (1995), the genes governing quantitative
resistance could share homologies with the clone’s R genes, making the candidate gene
approach feasible for the study of possible association between resistance gene analogs
(RGAs) and quantitative trait loci (QTLs) controlling pest resistance. In this work M Ecu 72
and M Col 2246 were amplified with RGA primers designed by C. Lopez in cassava (pers.
com.), to find putative loci related with whitefly resistance.
An additional step toward a better understanding of the attack response of the whitefly to
cassava was the establishment of a cDNA library, which was developed with a new, highly
effective method known as differential subtraction chain (DSC). Using this approach, two
mRNA populations, extracted from both resistant and susceptible genotypes, were examined
to elucidate the differential gene expression between them.
Materials and Methods
For this work an F1 cross (family CM 8996, 276 individuals) between M Ecu 72 (as the
resistant parent) and M Col 2246 (as the susceptible parent), elite cassava cultivars from
Ecuador and Colombia, respectively, was used. The parents and their offspring were
evaluated in the field at two sites: Nataima (Tolima) and Santander de Quilichao (Cauca). The
purpose of this evaluation was to identify gene segregation in the offspring and select the
resistant and susceptible materials. Both parents were evaluated with 343 cassava SSRs
(simple sequences repeat) (Mba et al., 2001) including 156 cDNA SSRs (Mba et al.,
submitted).
AFLPs (Vos et al., 1995) are being used to find markers associated to resistance for mapping
and ultimately cloning the resistant genes. Silver staining is being used to visualize the allelic
segregation of the markers. Cassava RGA primers were done in the parentals, and the
polymorphics were mapped in the F1.
For the isolation of expressed sequences, 21 forty-day-old plants were used, 7 of each
genotype (M Ecu 72 and M Per 334 resistant and M Col 2246 susceptible). These plants were
taken to the greenhouse, where they were infested with 300 whitefly adults per plant, for a
Output 6-47
2003 Annual Report
population of 2100 whiteflies per cage. Leaves were collected at six different times for RNA extraction.
For the differential subtraction chain (DSC), the follow strategy was used: Genotype M Ecu 72 was
infested for use as the tester, while genotype M Col 2246 was used as the driver. At present the DSC
technology is being performed according to Luo et al (1999). The representational difference analysis of
cDNA was divided into several phases:
x Generation of a PCR amplicon, which is representative of the original mRNA from M Ecu
72 and M Col 2246
x Subtractive hybridization of this amplicon M Ecu 72 (tester) and M Col 2246 (driver),
during which amplified portions of differentially expressed genes are enriched and
common sequences are depleted
x Cloning and screening of the resulting products
Results
Field evaluation
The field evaluation showed high pressure being exerted by the pest in Nataima, where test
materials had high damage rates; however, some materials had lower levels of damage in the
evaluations. We can conclude that these genotypes show a resistance level similar to parental
M Ecu 72.
AFLP analysis
An analysis was made of 128 combinations of primers with both parentals (M Ecu 72 and M
Col 2246) and both bulks of 10 whitefly-resistant DNA and 10 susceptible DNA. We obtained
53 polymorphic combinations, in which there were 425 polymorphic bands between the
resistant and the susceptible. All combinations were amplified in the F1 (Figure 6.41).
Figure 6.41. Silver-stained polyacrylamide gel showing combination ACA CTT of AFLP of
both parents M Ecu 72 (resistant), M Col 2246 (susceptible) and 25 individuals of
progeny. Note the polymorphic bands # 50 and #54’.
Project IP3: improving cassava for the developing world
Output 6-48
Mapping
Approximately 55 of the SSRs evaluated were polymorphics in the parentals and were
evaluated in the F1 (286 individuals). To construct the linkage map, 103 SSRs were analyzed,
of which 71 were anchored. A genetic linkage map of cassava was constructed with 71 SSR
markers segregating from the heterozygous female parent (M Ecu 72) of an intraspecific
cross. The map consists of 19 linkage groups, which represent the haploid genome of
cassava. These linkage groups spanned 550.2 cM, and the average marker density was 1 per
7.9 cM. The position of the 71 SSRs markers is shown on the framework (LOD = 25, theta =
25) of the molecular genetic map of cassava (Figure 6.42). Map distances are shown in
Kosambi map units. Of these markers, 26 (shown in green, Figure 6.42) had been placed
previously on the cassava framework map (Fregene et al., 1997); the other 45 SSRs are new.
Of the 71 SSRs, 31 were cDNA sequences (Mba, in prep.), while the others were genomic
DNA.
Association between molecular markers and resistance
The molecular data are being analyzed using QTL packages (QTL cartographer Q gene) to
determine linkages between the SSR markers and phenotypic characterization. Preliminary
analysis (X² at the 5% level) was done using SAS. Putative associations were found between
43 SSRs markers and the field phenotypic characterization (score 1.0 to 2.0 of the damage
levels and populations).
Output 6-49
2003 Annual Report
Linkage Group E
Figure 6.42. Preliminary cassava framework map of M Ecu 72 for resistance to whitefly,
consisting of SSRs. (Los = 25 and theta = 25).
Project IP3: improving cassava for the developing world
Output 6-50
Cassava RGAs
We obtained eight polymorphic RGA primers in the parentals (Figure 6.43). To date, we have
mapped three Bac primers in the F1 to find associations with yield QTLs. Polymorphism of
the presence and absence bands between the parentals was found in Bac 9, 31, 35, 45,
Contig 39 and Contig 43d and polymorphism of different bands in Bac 36 and the RT.
Figure 6.43. DNAs of M Ecu 72 and M Col 2246 amplified with RGA primers.
Differential subtraction: RNA extraction
RNA was isolated from young leaves of M Ecu 72 (E), M Per 334 (P) and M Col 2246 (M),
colleted in the greenhouse. To isolate total RNA, the Rneasy Plant Mini Kit QIAGEN¥ was
used. Genomic DNA was removed prior to isolation of poly (A)+ RNA with DNAse I. The SV
Total Isolation System of Promega¥ was used. The generation of cDNA was done using poly
A+ mRNA as the substrate, which was isolated using the protocol Oligotex mRNA Spin
Column of QIAGEN¥. First-strand cDNA synthesis and cDNA amplification were done using
SMART PCR cDNA Synthesis kit” de Clontech¥ (Figure 6.44).
Figure 6.44. M: O digested with Pst I. cDNAs amplified with kit SMART¥.
Output 6-51
2003 Annual Report
The PCR products from the amplification of cDNA, were purified using QIAquick PCR
Purification kit QIAGEN¥. Then digestion ligation was done, where the cDNA was digested
with DpnII, and then adapters (BamI and BamII) were ligated. Finally, the amplicon
generation was done for the hybridization reactions of the subtraction. For “tester” M Ecu 72
(E), 150 ng was obtained; and for “driver” M Col 2246 (C), 15 Pg.
Ongoing activities
x Saturation of linkage map of M Ecu 72, using AFLPs
x Isolation, cloning, sequencing and mapping of AFLPs polymorphic bands between
resistant and susceptible genotypes
x Design of SCARs for marker-assisted selection
x QTL analysis for whitefly resistance
x Mapping of cassava RGA polymorphics (BACs Primers, Gene Resistance Primers) in F1
(276 genotypes)
x Isolation of expressed sequences during the defense response of M Ecu 72 to whitefly
attack
In order to identify differentially expressed sequences, a new technology known as DNA chips
or microarray can scan a significant number of clones. Microarray-expression profiling will be
used to identify putative early response regulatory and/or signaling genes and to test the
function of selected candidate genes using reverse genetics.
Contributors: A. Bohórquez, J. Vargas, A. Bellotti, B. Arias, D.F. Cortés, M.C. Duque, J.
Tohme.
References
Arias B. 1995. Estudio sobre el comportamiento de la “mosca blanca” Aleurotrachellus
socialis Bondar (Homoptera: Aleyrodidae) en diferentes genotipos de yuca, Manihot
esculenta Crantz. Tesis maestria. Universidad Nacional de Colombia, Palmira.
Fregene, M.; Angel, F.; Gomez, R.; Rodriguez, F.; Chavarriaga, P.; Roca, W.; Tohme, J. and
Bonierbale, M. 1997. A molecular genetic map of cassava (Manihot esculenta Crantz).
Theor. Appl. Genet. 95:431-441.
Lefebvre, V. and Chèvre, A.M. 1995. Agronomie 15:3-19.
Luo, J.H.; Puc, J.A.; Slosberg, E.D.; Yao, Y.; Bruce, J.N.; Wright, T.C.; Becich, M.J. and
Parsons, R. 1999. Differential subtraction chain, a method for identifying differences in
genomic DNA and mRNA. Nucleic Acids Research. 27(19)
Mba, R.E.C.; Stephenson, P.; Edwards, K.; Melzer, S.; Mkumbira, J.; Gullberg, U.; Apel,
K.; Gale, M.; Tohme, J. and Fregene, M. 2001. Simple sequence repeat (SSR) markers survey
of the cassava (Manihot esculenta Crantz) genome: Towards an SSR based molecular
genetic map of cassava. Theor. Appl. Genet. 102:21-31.
Richter, T.E. and Ronald, P.C. 2000. The evolution of disease resistance genes. Plant Mol.
Biol. 42:195-204.
Rossi, M., Goggin, F.L., Milligan, S.B., Kaloshian, I., Ullman, D.E. and Williams, V. M. (1998).
The nematode resistance gene Mi of tomato confers resistance against the potato aphid.
Proc. Natl. Acad. Sci. USA. Vol. 95, pp 9750-9754.
Vos, P.; Hogers, R.; Bleeker, M.; Reijans., van de Lee, T.; Hornes, M.; Frijters, A.; Pot, J.;
Peleman, J.; Kuiper, M. and M. Zabeau. 1995. AFLP: A new technique for DNA
fingerprinting. Nucleic Acids Research. 23(21): 4407-4414.
Project IP3: improving cassava for the developing world
Output 6-52
Activity 6.11 Dissemination of the generated knowledge:
publications, conferences, seminars and proposals.
Training,
thesis,
Publications
CALATAYUD P.-A.; POLANÍA M.A.; GUILLAUD J.; MÚNERA D.F.; HAMON J.C.; BELLOTTI
A.C. 2002. Role of single amino acids in phagostimulation, growth, and development of
the cassava mealybug Phenacoccus herreni. Entomol Exp Appl 104/2-3:363-367.
DORN, B., MATTIACCI, L., BELLOTTI, A., DORN, S. 2002. Effects of a mixed species infestation
on the cassava mealybug and its encyrtid parasitoids. Biological Control 27:1-10.
VARGAS, H., BOLIVAR, L., ARIAS, B. and BELLOTTI, A. 2002. Nataima-31 Variedad de yuca
(Manihot esculenta Crantz) resistente a mosca blanca (Aleurotrachelus socialis Bondar)
para el Valle Cálido del Alto Magdalena. Colombian Ministry of Agriculture and New
Zealand Ministry of Trade and Foreign Affairs (MFAT) Corporación Colombiana de
Investigación Agropecuaria (CORPOICA) Regional 6 and Centro Internacional de
Agricultura Tropical - CIAT, Cali, Colombia. 400 copies. [Information leaflets]
[Spanish].
Dorn, B., Mattiacci, L., Bellotti, A.C. & Dorn, S. 2003. Verhalten in einfachen und komplexen
Systemen von tropischen natürlichen Gegenspielern. Mitteilungen der Schweizerischen
Entomologischen Gesellschaft 76 (1-2). 178.
ARIAS, B., A.C. BELLOTTI. 2003. Ciclo biológico, comportamiento e importancia económica
de Amblistira machalana Drake (Hemiptera:Tingidae). Chinche negro de encaje, en el
cultivo de la yuca Manihot esculenta Crantz. Revista Colombiana de Entomología 29 (2).
Publications Accepted
RIIS, L., A.C. BELLOTTI, M. BONIERBALE, and G. O’BRIEN. Cyanogenic Potential in Cassava,
its influence on a Generalist Insect Herbivore. Journal of Economic Entomology.
RIIS, L., A.C. BELLOTTI, O. CASTAÑO. In field damage on cassava clones of high and low
cyanogenci potential due to a generalist insect herbivore. Journal of Economic Entomology.
ALEAN, I., A. MORALES, C. HOLGUÍN, A.C. BELLOTTI. Patogenicidad de diferentes hongos
entomopatógenos para el control de Aleurotrachelus socialis (Homoptera: Aleyrodidae) bajo
condiciones de invernadero. Revista Colombiana de Entomología.
HOLGUIN, C.M., A.C. BELLOTTI. Efecto de la aplicación de insecticides químicos en el
control de la mosca blanca Aleurotrachelus socialis Bondar en el cultivo de yuca Manihot
esculenta Crantz. Revista Colombiana de Entomología.
TRUJILLO, H., ARIAS, B., GUERRERO, J., HERNANDEZ, P., BELLOTTI, A., AND J. E. PEÑA.
Survey of parasitoids of whoteflies Homoptera:Aleyrodidae) in cassava growing
regions of Colombia and Ecuador. Florida Entomologist.
CORTÉS, M.L., T. SÁNCHEZ, L. RIIS, A.C. BELLOTTI, P.-A. CALATAYUD. A bioassay to test
HCN toxicity to the burrowing bug Cyrtomenus bergi. Entomologia Experimentalis et
Applicata 108.
Output 6-53
2003 Annual Report
Posters
BELLOTTI, A., B. ARIAS, A. BOHORQUEZ, J.VARGAS, H.L. VARGAS, G. TRUJILLO, C.MBA,
M.C. DUQUE, J. TOHME. 2002. Recent advances in host plant resistance to whiteflies in
Cassava. [poster]. Congress of Entomological Society of America. Fort Lauderdale, FL.
USA.
ARIAS, B., A.C. Bellotti, H.L. VARGAS. 2003. Nataima-31, variedad de yuca (Manihot
esculentaI Crantz) resistente a mosca blanca, Aleurotrachelus socialis Bondar
(Homoptera: Aleyrodidae), una contribución al manejo integrado de plagas. [poster].
Memorias XXX Congreso Sociedad Colombiana de Entomología, SOCOLEN. Julio 17-19,
Cali, Colombia. p. 100.
Bellotti, A.C., A. BOHÓRQUEZ, B. ARIAS, J. VARGAS, H.L. VARGAS, CH. MBA, M.C.
DUQUE, J. TOHME. 2003. Avances recientes en la identificación de genes de
resistencia a mosca blanca, Aleurotrachelus socialis Bondar (Homoptera: Aleyrodidae) en
yuca (Manihot esculenta Crantz). [poster]. Memorias XXX Congreso Sociedad
Colombiana de Entomología, SOCOLEN. Julio 17-19, Cali, Colombia. p. 99.
Thesis
Burbano M., M. 2003. Multiplicación de material de especies silvestres y domesticadas del
género Manihot y estudio de su resistencia natural a tres plagas de cultivo
(Mononychellus tanajoa, Aleurotrchelus socialis, y Phanacoccus herreni) en condiciones
controladas de temperatura y humedad relativa. Tesis (Biólogo). Universidad del Valle,
Facultad de Ciencias, Cali, Colombia. 107 p.
Thesis in Progress
Carabalí, A. 2003. Evaluación del potencial de resistencia/tolerancia de diferentes genotipos
de yuca M. esculenta Crantz al biotipo B de B. tabaci de mosca blanca Bemisia tabaci
(Gennadius) (Homoptera: Aleyrodidae). Tesis MsC. Ciencias-Biología, Universidad del
Valle, Cali, CO.
Gómez S., M.J. 2003. Caracterización de genotipos de yuca (Manihot esculenta Crantz) por
su resistencia a mosca blanca Aleurotrachelus socialis Bondar (Homóptera: Aleyrodidae).
(Ingeniero Agrónomo). Universidad Nacional de Colombia, Palmira, CO.
Project IP3: improving cassava for the developing world
Output 6-54
Conferences
Bellotti, A.C. 2003. Biological control in the Neotropics; a selective review with emphasis on
cassava. 8th Simposio de Controle Biologic, SICONBIOL. Sao Paulo, Brazil. June 22-26,
2003.
Bellotti, A.C. and J. Tohme. 2003. Host plant resistance to whiteflies in cassava. USDAARS, U.S. Horticultural Research Laboratory, Ft. Pierce, FL, USA. February 12, 2003.
Dorn, B., Mattiacci, L., Bellotti, A.C. & Dorn, S. Verhalten in einfachen und komplexen
Systemen von tropischen natürlichen Gegenspielern. Jahrestagung der Schweizerischen
Entomologischen Gesellschaft. (Comportamiento de antagonistas tropicas en systemas
simples y complejos-Zurich Entomological Society). Zürich. Schweiz. 7.3.2003.
Aart van Schoonhoven, Francisco Morales, and Anthony C. Belloti. 2003. CIAT Role in Research and
Management of Invasive Pests and Pathogens in the Caribbean Basin. USDA, CSREES, TStar Sponsored Symposium Challenges and Oportunities in Protecting the Caribbean, Latin
America, and the United States from Invasive Species. Caribbean Food Crops Society, 39th
Annual Meeting, 13-18 July 2003, Grenada, West Indies. 15 July 2003.
Book Chapter
Bento, J.M.S., G.J. de Moraes, A.P. de Mattos, J.F. Warumby e A.C. Bellotti. 2002. Controle
biológico da cochonilha da mandioca no nordeste do Brasil. In: Controle Biológico No
Brasil. Eds. J.R. P. Parra, P.S.M. Botelho, B.S. Corrêa-Ferreira, J.M.S. Bento. Editora
Manole Ltda. 2002. Sao Paulo, Brasil, pp 395-408.
Proposals Presented and Approved
1. Assessing the Impact of Biotechnology on Diversity: Effect of Transgenic Maize on nontarget Soil Organisms, USAID, US$99,360 (Cornell University, Daniel Peck).
2. Integrated Control of Subterranean Pests in South America, GTZ/BMZ, US$256,560
(Andreas Gaigl).
3. Sustainable Integrated Management of Whiteflies through Host Plant Resistance, NZAID,
New Zealand, US$82,000.
4. Dynamics and IPM Component Research, DFID/NRI, US$39,000.
5. Biological Control of Whiteflies, by Indigenous Natural Enemies for Major Food Crops in
the Tropics, USAID, US$30,000.
6. Understanding the Mechanisms of Plant Resistance to Whiteflies, USAID/USDA,
US$35,000.
7. Manejo Integrado de las Principales Especies Plagas de Espárragos en el Departamento del
Cauca, SENA-Colciencias, $239.337.624 (Pesos Colombianos).
Output 6-55
2003 Annual Report
Output 7
Disease Resistance in Cassava
An important feature of the IP3 project relates to the integration of breeding,
entomology, plant pathology, and the development and use of biotechnology tools.
Despite the “divisions” created by the project structure, these four scientific areas have
maintained as close a relationship as possible. In Output 7, progress related to
cassava diseases is summarized.
Activity 7.1.
Characterization of family GM 315 (M Nga 19 x CM 9208-13)
for resistance to cassava bacterial blight under greenhouse
conditions
Objective
To evaluate resistance to cassava bacterial blight (CBB) in family GM 315 progeny under
greenhouse conditions.
Methodology
The GM 315 family was selected as the most segregant of four BC1 populations
evaluated for reaction to CBB (Xanthomonas axonopodis pv. manihotis – Xam –) in field
conditions at Villavicencio in 2002 (CIAT, 2002). To confirm its response to CBB, 138
individuals (1 plant per individual) of the progeny were planted under greenhouse
conditions at CIAT, and several strains of Xam were isolated from genotypes
presenting exudates and/or necrosis in the field (Table 7.1). The most pathogenic were
selected for later evaluation of the GM 315 family response to CBB under greenhouse
conditions.
Once the affected cassava genotypes were selected, stalk fragments were cut,
including healthy and diseased parts, and placed in sterile distilled water for 1 h at
4 ºC, attempting to mobilize the bacterium outside the tissue. The bacterial
suspension was later seeded by droplets in a petri dish with nutritive agar, incubated
at 30 ºC, and 24 hours later colonies were selected with typical Xam morphology.
Table 7.1.
Individuals of the GM 315 family with cassava bacterial blight (CBB)
symptoms, selected to isolate Xam.
Individual
87
124
141
163
250
291
320
322
aResistant,
Output 7-1
Field evaluation of CBB
5.0
3.5
3.0
2.5
2.5
2.5
2.5
4.5
a
1-2; intermediate, 2.5-3; susceptible, 3.5-5.
2003 Annual Report
The isolated bacteria were inoculated in the cassava variety, M Col 1505, susceptible
to Xam, and the most pathogenic isolate was selected to inoculate the GM 315 family.
Six-week-old plants were inoculated with a Xam suspension (24-h culture) in sterile
distilled water whose absorbance was adjusted in a spectrophotometer to an optic
density of λ600 = 0.5. A 1-ml syringe was used to make the diagonal puncture in the
stem and introduce one to two drops of inoculum.
The inoculated plants were incubated under greenhouse conditions for 5 days with
95% relative humidity (RH), and 28 ºC day/19 ºC night temperature (Tº). The
pathogenicity evaluation was made 12, 19, and 26 days after inoculation, following a
CBB severity scale where 1.0 corresponds to plants without symptoms, and 5.0 to
plant death.
A figure was made of CBB frequency distribution in the 138 individuals of the GM 315
family evaluated, and another of the percentage of resistant, intermediate, and
susceptible individuals. Individuals with values between 1.0 and 2.0 on the CBB
severity scale were considered resistant, between 2.5 and 3.5, intermediate, and
between 4.0 and 5.0, susceptible.
Results
The Xam strain isolated from the genotype GM 315-320 was selected as the most
pathogenic on inoculating the susceptible check M Col 1505, and was used in
evaluating the response of the GM 315 family under greenhouse conditions.
Figure 7.1 shows the distribution of frequency for CBB resistance under greenhouse
conditions. The GM 315 family presents a segregation from resistance to
susceptibility, and is normally distributed, confirming field evaluation results of 2002.
23
25
23
Frequency (%)
20
17
13
15
10
9
10
4
5
0
0
0
1,0
1,5
2,0
2,5
3,0
3,5
4
4,5
5
Scale of disease severity
Figure 7.1.
Output 7-2
Distribution of frequency of cassava bacterial blight (CBB) in the GM 315
family (M Nga 19 x CM 9208-13) under greenhouse conditions.
Individuals with values between 1.0 and 2.0 on the CBB severity scale
were considered resistant, between 2.5 and 3.5, intermediate, and
between 4.0 and 5.0, susceptible.
2003 Annual Report
Frequency
A figure was made of the percentage of individuals resistant, intermediate, and
susceptible to CBB (Figure 7.2). More than 60% of the evaluated individuals showed
intermediate reaction to Xam, with values between 2.5 and 3.5 in the scale of disease
severity, 23% were resistant, and 13.5% were susceptible.
Figure 7.2.
70
60
50
40
30
20
10
0
63.5
23.0
13.5
Resistant
Intermediate Susceptible
Reaction to CBB
Reaction of family GM 315 progeny to cassava bacterial blight (CBB)
under greenhouse conditions.
Reference
CIAT (Centro Internacional de Agricultura Tropical). 2002. IP-3 Annual Report. Cali,
CO.
Activity 7.2.
Evaluation of cassava genotypes for their resistance to
superelongation disease (SED) in the greenhouse
Specific objective
To evaluate SED resistance in promissory genotypes.
Methodology
Thirteen cassava genotypes were evaluated for SED (Sphaceloma manihoticola)
resistance under greenhouse conditions. Because there is wide genetic variation in the
pathogen, two isolates of Sphaceloma manihoticola from Santander de Quilichao were
inoculated by spraying a fungal suspension at concentration of 1.2 x 106 spores/ml. A
four-replications divided plot was used as experiment design, with the isolate as a
main plot, and genotypes as a subplot. Inoculated cassava genotypes were incubated
for 5 days at 98% RH, and at 30 °C. Disease reaction was evaluated at 7, 14, and 21
days after inoculation, using a disease severity scale from 1.0 to 5.0, where 1.0 means
no symptoms, and 5.0 corresponds to plant death. The criteria to define the degree of
sensitivity correspond to the highest value of disease expression obtained from four
replications of the reaction of genotypes.
Results
No genotype was resistant to SED. SM 1479-8 was intermediate to isolate SQ-1, but
susceptible to SQ-2, while M Per 183, elite genotype for inter-Andean valleys, was
Output 7-3
2003 Annual Report
intermediate to both isolates. CM 4919-1 and M Tai 8, two commercial genotypes
previously selected as susceptible in the North Coast, were susceptible in the
greenhouse (Table 7.2).
Table 7.2.
Genotype
CM 4919-1
Reaction of 13 cassava genotypes to two Sphaceloma manihoticola
isolates from Santander de Quilichao (Cauca), under greenhouse
conditions.
Zonea
1
Isolate
Scaleb
Genotype
Zonea
Isolate
Scaleb
c
2
SQ-1
4.0
SM 1460-1
SQ-2
4.5
2
SQ-2
4.0
SM 1479-8
SQ-1
2.5
2
CM 6055-3
SQ-1
4.0
SQ-2
4.0
4
SQ-2
4.5
SM 1557-17
SQ-1
3.5
Control
M Bra 703
SQ-1
3.5
SQ-2
4.0
2
SQ-1
4.0
SQ-2
4.5
SM 1821- 7 c
c
4
SQ-1
3.0
SQ-2
4.0
M Per 183
2
SQ-2
3.0
SM 1871-32
SQ-1
4.0
1
SQ-1
3.5
SQ-2
4.0
M Tai 8 c
2
SQ-2
4.0
SM 2219-9
SQ-1
3.5
2
SM 1144-4
SQ-1
3.0
SQ-2
4.5
2
SQ-2
3.5
SM 985-9
SQ-1
3.5
2
SM 1460-1
SQ-1
4.5
SQ-2
3.5
aAgro-ecological zone: 1 North Coast, 2 Eastern Plains, and 4 Inter-Andean valleys. bDisease
severity scale: Resistant, 1.0 – 2.0; intermediate, 2.5 – 3.5; and susceptible, 4.0 – 5.0. cElite
genotypes.
Activity 7.3.
Evaluation of cassava genotypes for resistance to CBB and
SED in Villavicencio
Specific objective
To evaluate cassava genotypes for resistance to CBB and SED in genotypes from yield
trials.
Methodology
Preliminary Yield Trial (EP). One hundred seventy five genotypes, planted in three
different Preliminary Yield Trials (identified as experiments 1, 2, and 3) from selection
in Clonal Evaluation Trial in 2002, were evaluated for reaction to CBB and SED in
Villavicencio, where these diseases are endemic. Disease reaction of the genotypes,
was evaluated with a disease severity scale from 1.0 to 5.0, where 1.0 corresponds to
no symptoms, and 5.0 corresponds to plant death. Four control varieties were
evaluated for CBB resistance: Brasilera, Ica Catumare, CM 6438-14, and La Reina
(CM 6740-7).
Advanced Yield Trial (ER). In a yield trial established in Villavicencio, 106 genotypes
selected from Preliminary Yield Trials and from ER-2001, and varieties from a Regional
Test (PR) were evaluated. The following cassava varieties were included as controls:
Brasilera and La Reina (CM 6740-7) as susceptible, and Ica Catumare, Ica Cebucán,
and CM 6438-14 as resistant.
Output 7-4
2003 Annual Report
Results
Eighty-nine genotypes (50.6%) from Preliminary Yield Trials were resistant to CBB,
with scores up to 2.0 in the disease severity scale. The susceptible genotypes,
Brasilera (score of 2.5) and La Reina (score of 3.5), used as controls, showed slightly
lower scores than in previous evaluations, indicating CBB pressure was not high in
2003, and there was no disease pressure for SED (Table 7.3).
Table 7.3.
Genotype
CM 9901-56
GM 219-31
GM 240-31
CM 9460-111
CM 9460-74
CM 9460-76
CM 9460-77
CM 9460-78
CM 9460-79
CM 9460-85
CM 9642-67
CM 9642-71
CM 9642-76
GM 219-34
GM 219-35
GM 219-42
GM 219-43
GM 219-47
GM 219-49
GM 219-50
GM 219-51
GM 219-54
GM 219-55
GM 219-60
GM 219-61
GM 220-36
GM 220-41
GM 220-44
GM 220-53
GM 220-54
GM 220-59
GM 220-63
aDisease
Evaluation of cassava genotypes planted in three Preliminary Yield Trials,
for their reaction to cassava bacterial blight (CBB) and superelongation
disease (SED) in Villavicencio.
CBBa SEDa Genotype
1.0
1.0
1.5
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
221-35
221-36
221-40
221-41
221-44
221-53
221-54
223-39
223-48
223-49
223-67
223-70
223-71
226-38
226-69
227-38
227-42
227-66
227-76
229-45
229-73
232-37
233-38
233-44
233-60
241-32
243-55
243-56
256-33
256-37
256-38
256-53
CBBa
SEDa
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
Genotype
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
GM
256-57
261-39
263-66
266-140
275-32
275-44
276-41
277-34
279-54
298-39
301-33
303-31
303-34
303-52
220-34
221-32
221-59
221-64
240-48
241-35
276-42
298-52
303-51
303-54
221-46
Control:
Brasilera
Catumare
CM 6438-14
La Reina
CBBa
SEDa
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
2.0
2.5
2.5
2.5
3.5
2.0
1.0
1.5
1.5
severity scale: Resistant, 1.0 – 2.0; intermediate, 2.5 – 3.0; and susceptible, 3.5 – 5.0.
Output 7-5
2003 Annual Report
Advanced Yield Trial (ER). Genotypes SM 1697-1, SM 1143-18, SM 1565-15,
SM 1807-1, and SM 1864-10, corresponding to 4.7% of genotypes, were highly
resistant to CBB disease, with a score under 1.5 in the disease severity scale. Eight
genotypes presented exudates on stem, and wilting, which correspond to a score of 3.0
in the disease scale (Table 7.4).
Table 7.4.
Reaction to cassava bacterial blight (CBB) of genotypes planted in an
Advanced Yield Trial (ER), in Villavicencio.
Genotype
CBB a
Genotype
CBB a
SM 1697-1
1.0
SM 1405-5
2.0
SM 1143-18
1.5
SM 1483-1
2.0
SM 1565-15
1.5
SM 1674- 1
2.0
SM 1807- 1
1.5
SM 1694-2
2.0
SM 1864-10
1.5
SM 1699-26
2.0
CM 5306- 8
2.0
SM 1773- 2
2.0
CM 8748-2
2.0
SM 1794-18
2.0
CM 8748-4
2.0
SM 1854-23
2.0
CM 9461- 3
2.0
SM 1871-32
2.0
CM 9461-51
2.0
SM 1881- 6
2.0
CM 9464-29
2.0
SM 2112-11
2.0
SM 667-1
2.0
SM 2219-11
2.0
SM 1353-3
2.0
SM 2361-25
2.0
SM 1363-11
2.0
SM 2367-16
2.0
aDisease severity scale: Resistant, 1.0 – 2.0; intermediate, 2.5
Activity 7.4.
Genotype
SM 2371- 1
SM 2375-16
SM 2425-3
SM 2456-3
SM 2640-9
SM 2786-10
SM 2792-42
CBB a
2.0
2.0
2.0
2.0
2.0
2.0
2.0
Control:
Brasilera
2.5
Catumare
2.5
Cebucán
2.5
CM 6438-14
2.5
La Reina
2.5
– 3.0; and susceptible, 3.5 – 5.0.
Characterizing F1 progeny of four backcross families ( M Nga
19 crossed with each of CM 9208-13, CM 9208-26, CM 920831, and CM 9208-73) for resistance to CBB in Villavicencio
Objective
To evaluate resistance to CBB in progeny of four families (BC1) of cassava under natural
disease pressure in Villavicencio.
Methodology
The progeny of four families (BC1) of cassava were characterized by reaction to CBB
under natural disease pressure in Villavicencio 4 months after planting in a second
evaluation cycle. The four families were:
Family
GM 315
GM 316
GM 317
GM 318
M
M
M
M
Nga
Nga
Nga
Nga
Cross
19 × CM
19 × CM
19 × CM
19 × CM
9208-13
9208-26
9208-31
9208-73
Individuals (no.)
357
399
348
238
The four families have a common male parent (MNga 19), resistant to various strains
of Xam. A disease severity scale with values between 1.0 and 5.0 was used, where 1.0
indicates a plant without lesions, and 5.0 plant death due to Xam. Each individual
Output 7-6
2003 Annual Report
was planted in plots of 6 plants. The highest value of disease expressed by plants
within each plot was evaluated, and a figure drawn of the percentage of resistant,
intermediate, and susceptible individuals of each BC1 family. Individuals with values
in the CBB severity scale between 1.0 and 2.0 were considered resistant, between 2.5
and 3.5 intermediate, and between 4.0 and 5.0 susceptible.
Results
At least 48% of individuals in each of the BC1 presented values less than 2.0 (Figure
7.3), which indicates that about half of the progeny of the four families showed high
resistance to CBB in Villavicencio. Only in the GM 315 family did some individuals
present a susceptible reaction to CBB, although the percentage was not representative
(1%). There were no susceptible individuals in the other families (GM 316, GM 317,
and GM 318). The CBB reactions registered in the four families may suggest that the
natural pressure of the inoculum at the time of evaluation was very low compared to
the pressure of the previous year, when the same families were evaluated and showed
a good distribution of the disease in the progeny, the GM 315 family being the most
segregant. The low CBB pressure, and the problems of germination of planted stakes
explain the high percentage of resistant individuals evaluated.
70
Frequency (%)
60
50
GM315
GM316
GM317
GM318
40
30
20
10
0
Resistant
Intermediate
Susceptible
Resistance to CBB
Figure. 7.3. Percentage of individuals resistant, intermediate, and susceptible to
cassava bacterial blight (CBB) in the four BC1 families.
Activity 7.5.
Evaluating a diallel assay in Villavicencio and Santander de
Quilichao
Specific objective
To evaluate diallel crosses for resistance to CBB and SED.
Methodology
Villavicencio. A 10 × 10 diallel study, comprising 45 families with 30 plants each,
was evaluated for reaction to CBB and SED under natural disease pressure, and
according to a disease severity scale of 1 to 5, where 1 corresponds to no symptoms,
and 5 corresponds to plant death.
Output 7-7
2003 Annual Report
The diallel was planted with three replicates at the Corporación Colombiana de
Investigación Agropecuária (CORPOICA) “La Libertad” station. The 10 genotypes
making up the diallel were:
CM 4574-7
SM 2058-2
CM 6740-7
SM 2219-11
CM 7033-3
HMC-1
SM 1219-9
M Per 183
SM 1565-15
M Tai 8
Santander de Quilichao. Eight families with 226 individuals, selected from the diallel
study described above, planted in 6-plants rows for each individual, were evaluated for
resistance to SED in Quilichao, where the disease was highly severe in 2003.
Results
Villavicencio. Although SED pressure was low, CM 4574-7 x SM 2058-2, and
CM 4574-7 x HMC-1 showed higher specific combinatory ability (SCA) for resistance to
CBB and SED, with less disease severity. CM 4574-7 had higher general combinatory
ability (GCA), showing the lowest average for severity of both diseases, taking into
account the nine possible crosses with the other genotypes of the study. CM 6740-7
and HMC-1 showed the lowest GCA (Table 7.5).
Table 7.5.
Reaction to cassava bacterial blight (CBB) and superelongation disease
(SED) of crosses in a diallel study in Villavicencio.
Family
Female parent
Male parent
CBBa
SEDa
CM 9460
GM 219
GM 220
GM 221
GM 222
GM 223
GM 224
GM 225
GM 226
GM 227
CM 9901
GM 229
GM 232
GM 233
GM 234
CM 9642
GM 235
GM 240
GM 241
GM 242
GM 243
GM 244
GM 245
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM
CM 6740-7
CM 7033-3
SM 1219-9
SM 1565-15
SM 2058-2
SM 2219-11
HMC 1
M Per 183
M Tai 8
CM 7033-3
SM 1219-9
SM 1565-15
SM 2058-2
SM 2219-11
HMC 1
M Per 183
M Tai 8
SM 1219-9
SM 1565-15
SM 2058-2
SM 2219-11
HMC 1
M Per 183
2.5
3.5
3.5
2.5
2.5
2.5
2.5
2.5
3.5
4.5
3.5
3.5
2.5
4.5
4.0
5.0
3.5
3.0
4.0
3.0
3.0
4.5
3.0
2.0
2.0
2.0
2.0
1.0
2.0
1.0
2.0
2.0
2.0
2.5
2.0
3.0
2.5
4.0
2.5
2.5
2.0
1.5
1.5
1.5
2.0
1.5
4574-7
4574-7
4574-7
4574-7
4574-7
4574-7
4574-7
4574-7
4574-7
6740-7
6740-7
6740-7
6740-7
6740-7
6740-7
6740-7
6740-7
7033-3
7033-3
7033-3
7033-3
7033-3
7033-3
Female meanb
CBB
SED
2.8
1.8
3.7
2.6
Continued.
Output 7-8
2003 Annual Report
Table 7.5. (Continued)
Family
CM 9127
GM 256
GM 261
GM 263
GM 264
GM 265
GM 266
GM 275
GM 276
GM 277
GM 278
GM 279
GM 298
GM 299
GM 300
GM 301
GM 303
GM 304
GM 305
CM 9733
CM 8055
GM 307
Female parent
Male parent
CBB
SED
Female meana
CBB
SED
3.5
1.9
CM 7033-3
M Tai 8
3.0
3.0
SM 1219-9
SM 1565-15
3.0
2.0
SM 1219-9
SM 2058-2
3.0
2.0
SM 1219-9
SM 2219-11
2.5
3.0
SM 1219-9
HMC 1
4.0
3.5
SM 1219-9
M Per 183
3.0
2.5
SM 1219-9
M Tai 8
3.5
3.5
3.2
2.6
SM 1565-15
SM 2058-2
4.0
2.0
SM 1565-15
SM 2219-11
2.5
2.0
SM 1565-15
HMC 1
3.0
1.5
SM 1565-15
M Per 183
3.0
2.0
SM 1565-15
M Tai 8
2.5
1.5
3.1
1.8
SM 2058-2
SM 2219-11
4.0
1.0
SM 2058-2
HMC 1
3.0
2.0
SM 2058-2
M Per 183
3.0
2.5
SM 2058-2
M Tai 8
4.0
3.5
3.2
2.1
SM 2219-11
HMC 1
4.0
4.0
SM 2219-11
M Per 183
4.0
3.5
SM 2219-11
M Tai 8
4.0
3.0
3.4
2.5
HMC 1
M Per 183
4.0
4.0
HMC 1
M Tai 8
3.0
4.0
3.6
2.9
M Per 183
M Tai 8
3.0
3.5
3.4
2.7
MTai 8
3.3
2.9
aDisease severity on a scale of 1-5, where 1 corresponds to no symptoms, and 5 to plant death.
bAverage for female parent in all possible crosses with the other nine genotypes conforming the
diallel assay.
Santander de Quilichao. Four individuals from three families were selected for
resistance to SED: GM 310-26 (M Ecu 72 x SM 1278-2), GM 312-6, GM 312-23
(M Ecu 72 x SM1673-10), and GM 313-19 (M Ecu 72 x SM 1741-1).
Activity 7.6.
Evaluation of cassava genotypes for resistance to SED in
Santander de Quilichao
Specific objective
To evaluate SED resistance in F1 families and interspecific crosses for root quality under
field conditions at Santander de Quilichao.
Methodology
In Santander de Quilichao, 598 cassava genotypes from 39 families established for
root quality evaluation, with some interspecific crosses with M Fla 437- 007, were
evaluated for their reactions to SED under high natural disease pressure. The nursery
comprised 6-plants rows for each individual. Resistance was determined using a
disease severity scale of 1 to 5, where 1 means without symptoms, and 5 corresponds
to plant death.
Output 7-9
2003 Annual Report
Results
Genotype GM 359-1 (Boyacá x SG 107-35) was resistant to SED, with a score of 2.0 on
the severity scale. Genotype GM 363-1 (CMC-40 x CM 2766-5) was intermediate
resistant, with 2.5 on the severity scale. The interspecific family CW 64 (M Fla 437007 x CG 487-2), with five individuals, scored up to 2.5 on the disease severity scale.
Families GM 348 (CM 4365-1 x SM 627-5), GM 350 (CM 4365-1 x HMC-1), GM 355
(SM 627-5 x HMC-1), and GM 361 (M Bra 97 x CG 501-1), were intermediate. Five
families scored up to 3.5, and 27 families were susceptible, with scores over 4.0 (Table
7.6).
Table 7.6.
Superelongation disease (SED) reaction in 39 cassava families evaluated
in Santander de Quilichao. Disease score on severity scalea, variance of
severity score, and individuals in each family, are presented.
Family
Female parent
Male parent
GM 359-1
GM 363-1
CW 64
GM 348
GM 350
GM 355
GM 361
GM 345
GM 347
GM 349
GM 351
GM 356
CM 5759
CM 6756
CM 7151
GM 335
GM 336
GM 337
GM 337
GM 339
GM 340
GM 341
GM 342
GM 343
GM 344
GM 352
GM 354
GM 357
GM 358
GM 360
GM 364
GM 365
Boyacá
CMC 40
M Fla 437-007
CM 4365-1
CM 4365-1
SM 627-5
M Bra 97
CM 2177-2
CM 3064-4
CM 4365-1
SG 107-35
SM 627-5
CMC 40
M Col 2215
M Col 2215
CG 487-2
CG 487-2
CG 487-2
M Pan 70
CG 501-16
CG 501-16
SM 627-5
CG 501-16
CG 501-16
CG 501-16
SM 627-5
SM 627-5
Boyacá
Boyacá
Boyacá
M Col 2215
HMC 1
SG 107-35
CM 2766-5
CG 487-2
SM 627-5
HMC 1
HMC 1
CG 501-1
CG 501-16
CG 487-2
M Bra 97
HMC 1
M Pan 70
M Col 2215
SG 107- 35
HMC 1
M Bra 97
M Col 2215
M Pan 70
CG 487-2
CM 2766-5
SG 107-35
CG 501-16
M Bra 97
M Col 2215
HMC 1
CG 487-2
SG 107-35
CG 501-1
CM 2766-5
HMC 1
CM 2766-5
M Bra 97
Highest
score
2.0
2.5
2.5
3.0
3.0
3.0
3.0
3.5
3.5
3.5
3.5
3.5
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
4.0
Lowest
score
2.0
2.5
2.0
3.0
3.0
2.5
3.0
3.5
2.5
3.5
3.0
3.5
4.0
3.0
3.0
2.5
2.5
4.0
3.5
2.5
2.5
2.5
2.0
4.0
3.5
3.0
2.5
3.5
2.0
2.0
4.0
2.0
Varianceb Individuals
per family
1
1
0.083
5
1
0.000
2
0.083
3
0.000
4
1
0.500
2
1
0.083
3
1
1
0.109
13
0.090
13
0.382
35
0.411
8
0.000
2
1
0.060
7
0.326
13
0.416
9
0.449
57
0.558
3
0.050
5
0.333
3
0.393
7
0.036
9
0.477
12
0.678
11
1
0.513
24
Continued.
Output 7-10
2003 Annual Report
Table 7.6. (Continued)
Family
Female parent
Highest
Lowest
Varianceb Individuals
score
score
per family
GM 369
M Ven 77
SG 107-35
4.0
3.5
0.125
2
CW 65
M Fla 437-007(6) CG 501-16
4.0
2.0
0.399
83
CW 66
M Fla 437-007(3) CM 2766-5
4.0
2.0
0.161
80
CW 67
M Fla 437-007(6) M Col 2215
4.0
2.0
0.388
150
CW 68
M Fla 437-007(3) HMC 1
4.0
2.5
0.236
9
CM 7334
CMC 40
HMC 1
4.5
4.0
0.083
3
GM 353
SM 627-5
CG 501-1
4.5
2.5
0.255
12
aDisease severity scale: Resistant, 1.0 – 2.0; intermediate, 2.5 – 3.0; and susceptible, 3.5 – 5.0.
bVariance calculated among individuals for each family. No variance was calculated for families
with one individual.
Activity 7.7.
Male parent
Using polymerase chain reaction (PCR) with degenerate
primers to search for resistance gene analogs associated
with resistance to CBB
Specific objective
To develop molecular markers associated with resistance to CBB.
Methodology
A group of five primers, NBS, Pto, XLRR, WipK, and KSU (Table 7.7), corresponding to
conserved gene domains of plant disease resistance, were used to amplify similar
sequences in the cassava genotypes selected as resistant to CBB.
Table 7.7.
Primer
XLRR f
XLRR r
WipK f
WipK r
NBS f1
NBS r1
Pto 1
Pto 2
KSU f
KSU r
Degenerate primers used to amplify DNA from CM 6438-14, CM 777213, CM 3311-4, M Bra 1045, and M Nga 19 cassava genotypes, and a
resistant group from the GM 315 family.
Sequence 5´- 3´
-CCGTTGGACAGGAAGGAG-CCCATAGACCGGACTGTT-GGTCGTGGTGCTTATGGAAT-CCATGAAGATGCAACCGAC-GGAATGGGNGGNGTNGGNAARAC-YCTAGTTGTRAYDATDAYYYTRC-ATGGGAAGCAAGTATTCAAGGC-TTGGCACAAAATTCTCATCAAGC-GGIGGIGTIGGIAAIACIAC-ARIGCTARIGGIARICC-
The DNA of CM 6438-14, CM 7772-13, CM 3311-4, and M Bra 1045, together with
that of M Nga 19 (parent resistant to CBB of the GM 315 family), and a group of
resistant individuals (from the GM 315 family), were amplified by means of unspecific
PCR in conditions of low annealing temperature, to obtain sequences associated with
resistance to Xam, using degenerate primers. Each PCR reaction was carried out in a
Output 7-11
2003 Annual Report
final volume of 25-µL, taking into account the following final concentrations: dATP,
dCTP, dGTP, and dTTP 0.2 mM each; 2.5 mM MgCl2; 0.25X solution Q (QIAGEN kit for
PCRs); 1.5 U of Taq polymerase; 1 µM primer; buffer of Taq polymerase 1X; and 150
ng of DNA. For the control reactions, the DNA was replaced by sterile distilled water.
The NBS primer is a sequence of the conserved domain of the nucleotide union site in
the N gene of tobacco and RPS2 of Arabidopsis (Yu et al., 1996); XLRR is a sequence
based on the region of Leucin-rich repeats of the RPS2 and Xa 21 rice genes (Chen et
al., 1998); Pto is a kinase sequence involved in resistance in potato (Leister et al.,
1996); WipK amplifies the conserved region of MAK kinase of coriander (Y12875),
tobacco (D61377), and Arabidopsis (MPK3) (Ligterink et al., 1997); and KSU is a
sequence recommended by Dr. Hulbert Scot, of Kansas State University, that amplifies
a region within the nucleotide binding site (NBS) domain
The amplification of NBS, Pto, WipK, and XLRR was carried out in an MSJ-Research
PTC-100 thermocycler with the following program: 5 min at 94 °C; 45 cycles with
denaturation for 1 min at 94 °C; pairing for 1 min at 45 °C, and extension for 2 min at
72 °C; and a final extension for 7 min at 72 °C. For amplification with the KSU primer,
the same program was used, but changing the pairing temperature to 42 °C, and the
final extension for 10 min.
The PCR product was submitted to electrophoresis in 2% agar gels in TBE 0.5X buffer,
or in 6% acrylamide gels, together with a molecular weight marker (100-pb or 30-300
pb respectively) to estimate the size of each fragment of amplified DNA. Next, this
amplified product was purified based on CM 6438-14, CM7772-13, CM 3311-4, and
M Bra 1045 with the Concert Rapid PCR Purification System kit or elusion of agar
bands of the PCR product with the KSU primer, using the QIAquick Gel extraction
(fragments of 500 pb) kit. Of the DNA amplification of M Nga 19 (parent resistant to
CBB of the GM 315 family) and the resistant group (of the GM 315 family) visualized
in acrylamide at 6%, band elution was made placing a drop of TE 1X on the selected
band, and once humidified, it was cut with a scalpel. The cut fragment was placed in
an eppendof tube of 1.5 ml, and 20 µl of TE 1X added. The tube with this mixture was
placed in a water bath at 95 ºC for 30 min, and then stored at 4 ºC to be used as a
template in a new PCR reaction with the degenerate primer that gave origin to the
elution band. With this step, we sought to augment the quantity of the sample to
clone.
The PCR products and the elution bands were cloned to the cloning vector PGEM-Teasy (Promega kit), and then introduced into the E. coli DH5-α strain by
electrophoresis at E. coli DH5-α. The transforming colonies were selected by color in
LB/ampicilin/IPTG/X-gal medium. The white colonies carry the plasmids with the
fragments of interest from PCR or band elution.
To confirm the presence of cloned fragments, bands were visualized generated in the
restriction of the vector with Eco RI in a 1.5% agar gel. Next, clones with insertions
greater than 300 pb were selected to be sequenced, and to seek homology with genes
reported for disease resistance in various crops in the National Center for
Biotechnology Information (NCBI) database (www.ncbi.nlm.nih.gov), through
application of the Basic Local Alignment Search Tool (BLASTx), where the sequence
obtained with nucleotides sequences registered in the database was paired.
Output 7-12
2003 Annual Report
Results
Cloning. The PCR product of genotypes CM 6438-14, CM 7772-13, CM 3311-4, and M
Bra 1045 was visualized in agar gels at 2% (Figure 7.4), utilizing the 100 pb molecular
weight marker to estimate the size of each amplified fragment.
M1
21
32
4
3
54 65 76
M1
21
23
34
45
B
B
A
600 pb
1 2 3 4 5 6 7 8 9 10 11 12 13
C
600 pb
Figure 7.4.
Output 7-13
1
2
3
4
5
D
500 pb
DNA of CM 6438-14, CM 7772-13, CM 3311-4, and M Bra 1045
amplified with the primers NBS, Pto, WipK, XLRR, and KSU. (A) Primer
WipK: lane 1 = 100 pb; lane 2 = CM 6438-14; lane 3 = CM 7772-13; lane
4 = CM 3311-4; lane 5 = M Bra 1045; lane 6 = positive control (MCR 81);
and lane 7= negative control. (B) Primer XLRR lane 1 = 100 pb; lane 2 =
CM 6438-14; lane 3 = CM 7772-13; lane 4 = CM 3311-4; and lane 5 = M
Bra 1045. (C) Primers NBS and Pto lane 1 = 100 pb; lane 2 = CM 643814; lane 3 = CM 7772-13; lane 4 = CM 3311-4; lane 5 = M Bra 1045; lane
6 = positive control (MCR 81); lane 7 = negative control of NBS, lane 8 =
CM 6438-14; lane 9 = CM 7772-13; lane 10 = CM 3311-4; lane 11 = M
Bra 1045; lane 12 = positive control (MCR 81); and lane 13 = negative
control. (D) Primer KSU lane 1 = 100 bp; lane 2 = CM 7772-13; lane 3 =
CM 3311-4; lane 4 = CM 6438-14; and lane 5 = M Bra 1045.
2003 Annual Report
The visualization of DNA of M Nga 19, and the resistant group of the GM 315 family,
amplified with primers NBS, Pto, WipK, XLRR, and KSU, permitted the observation of
a pattern of bands between 50 and 400 pb (figure not shown).
Band elution and cloning. Elution was made of a 500-pb band generated by the KSU
primer in the four amplified genotypes, and of some bands between 200 and 400 pb
observed in the acrylamide gel, for later cloning.
The PCR product and eluted bands were cloned, and visualization of cloned fragments
was made in agar gel (Figure 7.5), as described in the Methodology.
1
Figure 7.5.
2
3 4 5 6 7 8
9 10 11 12 13 14 15 16
17 18 19 20 21 22 23 24 25 26
Digestion of plasmids extracted with Eco RI. Lane 1= 100 bp; lane 2 and
10= different clones from amplification of CM 3311-4 with the primers;
lane 9= CM 7772-13 amplified with WipK; lanes 11-15, 17-19= different
CM 6438-14 clones amplified with WipK; lanes 20-25= different CM
3311-4 clones amplified with XLRR; and lane 21= CM 7772-13 amplified
with XLRR.
The number of clones obtained with WipK, XLRR, Pto, NBS, and KSU primers from
DNA amplification of the described genotypes is shown in Table 7.8.
Table 7.8.
Clones obtained by amplification of CM 6438-14, CM 7772-13, CM 33114, M Nga 19, and the resistant group of the GM 315 family with NBS,
Pto, WipK, and XLRR primers.
Clonea
Primer
N30*
N31*
N32*
N33*
N34
N35*
N36*
N37
N38
N40
N41*
N44*
NBS
NBS
NBS
NBS
NBS
NBS
NBS
NBS
NBS
NBS
NBS
NBS
Output 7-14
MW
(bp)b
950
900
650
390
320
1100
600
230
250
250
700
650
Genotype
Clone
Primer
CM 3311-4
CM 6438-14
M Bra 1045
M Bra 1045
CM 3311-4
CM 6438-14
CM 7772-13
M Nga 19
Resistant group
M Nga 19
M Nga 19
M Bra 1045
W11*
W12*
W13*
W14*
W15
X1*
X2*
X3*
X4*
X5*
X6*
X7*
WipK
WipK
WipK
WipK
WipK
XLRR
XLRR
XLRR
XLRR
XLRR
XLRR
XLRR
MW
(bp)b
390
450
360
390
250
530
700
750
310
1150
390
370
Genotype
M Nga 19
Resistant group
CM 3311-4
CM 3311-4
CM 3311-4
CM 3311-4
CM 3311-4
CM 7772-13
CM 7772-13
CM 7772-13
CM 7772-13
CM 6438-14
2003 Annual Report
Table 7.8. (Continued)
Clone
Primer
MW Genotype
(bp)b
P30*
Pto
350 CM 3311-4
X8*
XLRR
P31*
Pto
390 CM 3311-4
X9*
XLRR
P32*
Pto
420 CM 3311-4
X10*
XLRR
P33*
Pto
330 CM 3311-4
X11
XLRR
P34*
Pto
380 CM 3311-4
X12*
XLRR
P35*
Pto
580 CM 3311-4
X13*
XLRR
P36*
Pto
530 CM 3311-4
X14
XLRR
P37*
Pto
420 M Nga 19
X15*
XLRR
P38*
Pto
1150 M Nga 19
X16*
XLRR
P39*
Pto
320 Resistant group
X17*
XLRR
P40*
Pto
400 Resistant group
X18*
XLRR
P41*
Pto
390 M Nga 19
X19*
XLRR
P42*
Pto
310 M Nga 19
X20*
XLRR
P43*
Pto
630 CM 3311-4
X21
XLRR
P44*
Pto
600 CM 3311-4
X22
XLRR
P45*
Pto
450 CM 3311-4
X23*
XLRR
W1*
WipK
1200 CM 3311-4
X24*
XLRR
W2*
WipK
480 CM 3311-4
X25*
XLRR
W3*
WipK
420 CM 3311-4
X26*
XLRR
W4
WipK
230 CM 3311-4
X27*
XLRR
W5*
WipK
610 CM 7772-13
X28*
XLRR
W6*
WipK
300 CM 6438-14
X29*
XLRR
W7
WipK
220 M Nga 19
X30*
XLRR
W8
WipK
250 Resistant group
X31*
XLRR
W9*
WipK
850 M Nga 19
X32*
XLRR
W10*
WipK
400 Resistant group
a(*) clones selected for sequencing. b MW, molecular weight.
Clonea
Primer
MW
(bp)b
500
400
450
280
600
580
200
650
350
430
300
580
400
240
280
500
400
750
600
630
750
400
410
800
600
Genotype
CM 6438-14
CM 6438-14
CM 6438-14
CM 6438-14
CM 6438-14
CM 6438-14
Resistant group
M Nga 19
M Nga 19
M Nga 19
Resistant group
M Nga 19
Resistant group
Resistant group
Resistant group
M Nga 19
Resistant group
M Nga 19
Resistant group
M Nga 19
M Nga 19
M Nga 19
Resistant group
M Nga 19
M Nga 19
Clone sequencing. The following clones showed homology with resistance genes
reported in the NCBI database: X1, X5, X9, X15, X17, X18, and X19 from
amplification with the XLRR primer, W5, W6, W9, and W10 from amplification with
the WipK primer, P32, P41, and P36 from amplification with the Pto primer, and N31
from amplification with the NBS primer.
For each of these clones, the presence of an open reading frame (ORF) was sought in
the amino acid sequences deduced from the DNA, and it was found that all present at
least one within the sequence.
The sequences that standardized with resistance genes reported in the NCBI, which
also presented an ORF, were aligned among themselves, with sequences of known
resistance genes, through the option of multiple alignments of sequences of the
DNAman program version 4.13. A phylogenetic tree was constructed with them,
through Parsimonia and bootstrap statistical analysis with 5000 replications (Figure
7.6), to determine the degree of similarity among the generated sequences and those
reported.
Output 7-15
2003 Annual Report
0.05
X18
78
99
X19
X17
X1
93
53
X9
47
RGA marker (XLRR) Lycopersicon hirsutum
X5
25
X15
R protein (NBS-LRR)Triticum aestivum
84
99
N31
99
R protein (NBS) Glycine max
W5
72
W6
100
W9
68
W10
62
100
99
57
Figure 7.6.
P32
P41
100
Kinase serine/theorine (like
-Pto) Solanum
tuberosum
Kinase protein (PtoR)Lycopersicon
pimpinellifolium
P36
Phylogenetic tree of the sequenced clones that showed homology with
resistance genes, carried out with Parsimonia and bootstrap analysis
(5000 replications). X, clones XLRR; W, clones Wipk; P, clones Pto.
According to the phylogenetic tree, the sequences of clones from amplification with the
XLRR primer (domain of Leucin-rich repeats) present similarity with the sequence of
an RGA marker of the (XLRR) type from Lycopersicon hirsutum. Within the same group
was found the sequence of a clone from amplification of the NBS primer (site of
nucleotides union) that in turn shows high similarity with an R Protein type NBS-LRR
from Triticum aestivum and with an R Protein type NBS from Glycine max. The
presence of XLRR and NBS clones in the same similarity group is explicable, taking
into account that according to the type of resistance gene, the LRR and NBS domains
are found together in the expressed protein, and are involved in pathogen recognition
and signaling respectively, as in the case of the L6, N, RPP5, RPS2, and RPM1 genes,
the first three mediating recognition of the virus and fungi, and the last two the
recognition of bacteria, all at the level of cellular cytosol. Possibly, this group is
indicating that a part of the XLRR sequences is encountering homology not only with
the LRR region of resistance genes, but also with the NBS region of other resistance
genes.
In addition, clones from amplification with the WipK primer grouped 62% of the
bootstrap replicas with clones from amplification with the Pto primer; these last show
similarity with a kinase serine/Threonine of Pto type, from Solanum tuberosum and a
kinase (PtoR) protein from Lycopersicon pimpinellifolium. The presence of clones of
WipK and Pto type within the same group of similarity is explicable, taking into
Output 7-16
2003 Annual Report
account that the two primers are designed based on one kinase that mediates the
transduction of resistance signals gene by gene.
References
Chen, X.M.; Line, R.F.; Leung, H. 1998. Genome scanning for resistance gene analogs
in rice, barley, and wheat by high-resolution electrophoresis. Theor. Appl. Genet.
97:345-355.
Leister, D.; Ballvora, A.; Solamini, F.; Gebhardt, C. 1996. A PCR-based approach for
isolating pathogen resístanse genes from potato with potential for wide
application in plants. Nature Genet. 14: 421-429.
Ligterink, W.; Kroj, T.; Zur Nieden, U.; Hirt, H.; Scheel, D. 1997. Receptor-mediated
activation of a MAP kinase in pathogen defense of plants. Science 176:20542057.
Yu, Y.; Buss, G.; Saghai Maroof, M. 1996. Isolation of a superfamily of candidate
disease-resistance genes in soybean based on a conserved nucleotide-binding
site. Procs Nat. Acad. Sci. 93:11751-11756.
Activity 7.8.
Identification of gene analogs for resistance to cassava
(Manihot esculenta Crantz) diseases, and their relationship
to resistance to three Phytophthora species
Specific objectives
1. To identify disease resistance genes analogs in cassava.
2. To identify QTLs associated to cassava resistance to three Phytophthora species.
3. To analyze homology, in the cassava genome, of disease resistance genes isolated
from monocots through hybridization.
4. To clone and sequence cassava DNA regions associated to disease resistance.
5. To search for homology of cassava genes with disease resistance sequences from
different species through amplification with degenerate primers.
6. To design primers for amplification of regions associated to cassava disease
resistance.
7. To evaluate association of amplified regions with designed primers, and resistance to
three Phytophthora species.
Methodology
Root cylinders of cassava K (MNga 2 x CM 2177-2) and CM9582 (MBra 1045 x MCr
81) families were inoculated with mycelial discs from three species of Phytophthora
species (P. melonis, P. palmivora, and P. tropicalis), to identify homologies of cassava
genome regions to disease resistance genes from different crop species. The resistance
of parents and progenies was also evaluated by measuring the volume of root showing
rot symptoms 2, 4, and 6 days after inoculation. Two strategies were used to find
resistance regions. The first consisted on hybridization with heterologous probes from
maize and rice, using RFLP (restriction fragment length polymorphism), facilitated by
Kansas State University (USA). The DNA of parents of the K family was digested with
six enzymes selected based on previous work presenting polymorphic restriction in
cassava. The probes were labeled with 32P[dATP] and allowed to hybridize overnight,
the film being revealed 15 days afterwards. The second strategy consisted of
amplifying conserved regions of DNA, using PCR with degenerated NBS (nucleotide-
Output 7-17
2003 Annual Report
binding sites) and Pto kinase primers. DNA was accordingly extracted from three
cassava genotypes resistant to Phytophthora spp.— MBra 532, MBra 1045, and MCr
81—obtaining clones that were sequenced and homologated with known resistance
genes, using the Blastx tool, in the NCBI (National Center for Biotechnology
Information) database. Specific primers were designed with the sequences, allowing
DNA regions of parental material and two bulks of 10 resistant individuals and 10
susceptible to be amplified. Bands were separated by denaturing polyacrylamide gel
electrophoresis at 4% and 6%, and non-denaturing polyacrylamide gel or SSCP (single
strand conformation polymorphism) at 10%. Sequencing was used to find single
nucleotide polymorphisms (SNP).
Results
Five QTLs associated to Phytophthora spp resistance were identified in linkage groups
E, G, H and O (Figure 7.7). By hybridization of the heterologous probe Pic 15, from
maize, very faint bands were obtained when parental material was digested with EcoR
V and Hind III. It was concluded that cassava has a very low homology with the genes
from the monocotyledons tested. A total of 28 NBS and 2 Pto kinase clones were
obtained using three degenerate primers and subsequent cloning of the obtained
fragments; of these, five showed homologous sequence with NBS-LRR RGAs
(resistance gene analogs) reported by the NCBI (Table 7.9). Four of them showed open
reading frames (ORF) with conserved motifs P-loop, kin-2, kin-3, and GLPL, of the NBS
region, which means they were considered as RGAs. Three different RGAs were
identified based on the phylogenic tree (Figure 7.8); of these, N-37 was similar to the
Mi gene (non-TIR class); N-38 and K-1 were similar to the TIR genes L6 and RPP5;
whereas N-33 was different from all the above. It was concluded that both TIR and
non-TIR genes, subclasses of the NBS-LRR genes, are found in cassava. No
polymorphism in the RGAs was found between the parents by PCR-RFLP,
electrophoresis in polyacrylamide, SSCP, or SNPs, preventing the identification of
association between the RGAs and resistance to Phytophthora.
Output 7-18
2003 Annual Report
cM
44 P4 P12
0
cM
rHRGP
rGY118
rF19b
rS2
GY217
50
rNS74
rNS260
rNS319
rNS217
rSSRY319
100
rGY176
CBB4
rSSR
SSRY
rNS
SSRY
5
rA
K1
NS
NS
10
NS
rSSR
CPY79(15)
E
cM
44 P4 P12
0
5
10
15
GY
GY1
OJ
SSR
NS3
SSR
rGY17
rGY104
GY12
rGY211
rSSRY3
SSRY
SSR
SSRY
SSRY
GY1
rGY
rGY
44 P4 P12
0
GY137
15
rAB
44 P4 P12
1
cM
0
50
rP1a
rGY156
GLU
rGY220
rNS347
rNS10
SSRY90
rSSRY19
G
Significance (P)
O
< 0.0010
< 0.0050
< 0.0100
< 0.050
Not significant
Increased
effect from
B of parent
Additional
effect
H
Figure 7.7.
Output 7-19
Map of identified QTLs, from the mother map (MNga 2), for resistance of
the K family to Phytophthora tropicalis (44), P. palmivora (P4), and P.
melonis (P12), in different linkage groups (E, G, H, and O). The color
indicates QTL significance, according to the key at lower right of figure.
The distance between molecular markers is shown in centimorgans (cM).
2003 Annual Report
Table 7.9.
Resistance genes analogs showing the highest homology with clones N-23, N-33, N-37, N-38, and K-1
isolated by PCR primers NBS and KSU from DNA of cassava genotypes M Bra 1045 and M Bra 532.
Protein codified by homologous sequences to each clone
(GenBank) a
Clone N-23 (primer T7):
Protein NBS-kinase Z2 (AF281282.1)
Putative resistance gene analog, NBS-LRR
(AF516642.1)
Clone N-33 (primer T7):
Similar to resistance protein NBS/LRR (AF402735.1)
Candidate to resistance protein (AAC02202.1)
Species
Homology
(bits) b
Solanum tuberosum
Malus prunifolia
Theobroma cacao
Lactuca sativa
Clone N-37 (primer T7):
Similar to resistance protein NBS/LRR (AF402735.1)
Homologous to disease resistance protein
(AAD34880.1)
Theobroma cacao
Vigna unguiculata
Clone N-38 (primer T7):
Putative protein similar to NBS-LRR (AF515627.1)
Resistance putative protein OB8 (AF363803-1)
Resistance gene analog protein NBS-LRR-Toll
(AF487946-1)
Malus domestica
Phaseolus vulgaris
Medicago sativa
Probability of
highest value
Identity
(%)c
Positives
(%)d
56
54
1e -07
5e –07
35
40
47
47
103
97
1e -21
1e -19
44
46
59
60
1e -18
5e -17
44
44
53
53
8e -29
2e -28
8e -24
46
45
42
70
66
67
92
86.7
126
125
110
Clone K-1 (primer T7):
Putative resistance gene analog NBS-LRR (AF516642-1) Malus prunifolia
146
1e -34
52
72
KNBS3, similar to resistance protein (AF325686.1)
Glycine max
142
2e -33
50
74
Resistance gene analog protein NBS-LRR-Toll
Medicago sativa
132
2e –30
50
72
(AF487946-1)
aIn parenthesis, National Center for Biotechnology Information (NCBI) code. bHomology (bits): Homology score according to number
of homologous nucleotides and fragment length where there are homologous nucleotides. cIdentity: Homologous genic products.
dPositives: Homologous nucleotides.
Output 7-20
2003 Annual Report
0.05
NBSD-H1
48
N-37
83
98
AF402735-1
1
Mi
RPS2
L6
99
N-38
100
K-1
99
AF515627-1
95
RPP5
AF487946-1
100
77
AF516642-1
N-33
Figure 7.8.
Activity 7.9.
Phylogenetic tree of the amino acid sequences carried out with
Parsimonia and bootstrap analyses (5000 replications) with the DNAman
program, of the resistance gene analogues (RGAs) N-33, N-37, N-38, and
K-1, compared with homologous sequences obtained from the National
Center for Biotechnology Information (NCBI), two Toll protein and
Interleukin-1 receptor (TIR) genes (L6 and RPP5), two non-TIR genes
(RPS2 and Mi), the RGA of soya (NBSD-H1), and homologies found in the
NCBI database (Table 7.9.) AF402735-1, AF487946-1, AF515627-1, and
AF516642-1.
Evaluating a cassava family for resistance to Phytophthora
root rot
Specific objective
To evaluate Phytophthora tropicalis resistance in a segregant population.
Methodology
Thirty-four individuals of the cross CM 9582 (MBra 1045 x MCr 81) were evaluated for
resistance to Phytophthora tropicalis by inoculation of fresh roots with microorganism
discs. Root damage was determined by measuring width and length of lesions 5 days
after inoculation, and root affected area was figured out. Frequency analysis and
resistance distribution were completed.
Results
Frequency analysis resulted in grouping according to percentage of rotten root area.
Individuals with highest resistance to P. tropicalis were CM 9582-8, CM 9582-16,
CM 9582-30, and CM 9582-40. Figure 7.9 shows susceptibility ratings.
Output 7-21
2003 Annual Report
Frequency %
Figure 7.9.
40
35
30
25
20
15
10
5
0
R
MR
I
S
Resistance group
HS
Breakdown of cassava family CM 9582, inoculated with Phytophthora
tropicalis on roots, according to resistance groups based on frequency
analysis. R, resistant; MR, moderately resistant; I, intermediate; S,
susceptible; HS, highly susceptible.
Actividad 7.10
Evaluating simple sequence repeat
bacterial blight resistance in cassava
markers
linked
to
Specific objective
•
To identify SSR markers associated with resistance to cassava bacterial blight in a
segregant BC1 cassava family (GM315)
Introduction
Cassava bacterial blight (CBB), caused by Xanthomonas axonopodis pv. manihotis
(Xam), is a major disease of cassava (Manihot esculenta Crantz) in Africa and South
America causing losses between 12 and 90%. Planting resistant varieties is the
preferred method of disease control. Previous studies on the genetics and number of
disease loci involved in host plant resistance to CBB revealed many QTLs that control
disease resistance (Jorge et al. 2000; Jorge et al. 2001). In the present work, the
evaluation of more than 400 simple sequence repeats (SSR) detected one SSR marker
associated with CBB resistance in a segregating backcross family in which disease
symptoms were evaluated under field conditions. The association between SSRY 65
and CBB resistance in field was obtained with a significance level P < 0.05.
Materials and methods
Characterizing F1 progeny of a backcross family. The progeny of a cassava BC1
familiy, GM 315, was characterized for their reaction to CBB under natural disease
pressure in Villavicencio, Colombia.
This family has a recurrent male parent (M Nga 19), which is resistant to many Xam
strains. The female parent is an F1, originated from M Nga 19 and M Col 1522, a local
variety with high quality starch. The family reaction to CBB was analized, using a
Output 7-22
2003 Annual Report
frequency distribution graph, based on an average of 6 plants per plot, and a disease
severity scale of 1.0 to 5.0 where 1.0 indicates plants with no symtoms and 5.0 death
plants.
Resistance plants scored between 1.0 and 2.0 on the scale, intermediate between 2.5
and 3.0, and susceptible more than 3.5
Evaluating simple sequence repeat markers by bulk segregant Analysis. Towards
the identification of SSR markers linked to CBB resístanse, Bulk Segregant Analysis
(BSA) was made in the family GM 315. Bulks were constituted from 11 resistant
genotypes (a score of 1.0 to 2.0) and 11 susceptible genotypes (a score of 4.0 to 5.0).
BSA provides a method of focusing on regions of interest with molecular markers as
against analyzing the entire genome (Michelmore, 1991).
M Nga 19 (this parent is included because is the resistant parent, and let comparison
with its progeny. The other parent was missing, but for the analysis, it is not
important) and contrasting bulks (11 individuals in each one) were evaluated, using
486 microsatellite primers, the SSR markers have been described elsewhere (Mba et al
2001; CIAT 2001).
Opening of the Bulks. Individuals from resistant and susceptible bulks were
evaluated with candidate primers which showed polymorphism between bulks.
Candidate primers that showed polymorphisms between resistant and susceptible
individuals were considered potential SRR marker associated with resístanse. They
were evaluated in the whole population to confirm association between SSR marker
and CBB resistance in the field.
Association between molecular marker and plant response to CBB. Association
between SSR molecular marker and CBB resistance in the field was determinated by
a Chi-square Independence test (SAS procedure, SAS Institute, 1999-2001). An
association was considered significant if the probability of the null hypothesis (no
association) was less than P<0.05
In order to evaluate marker reliability for resistance detection, its sensibility and
specificity were measured, which let to know the probability for true positive and
negative results, according to the marker.
Output 7-23
2003 Annual Report
Fequency (%)
Results
50
45
40
35
30
25
20
15
10
5
0
45.9
19.9
12.0
7.6
5.6
1-1.5
1.6-2
2.1-2.5
2.6-3
3.1-3.5
3.9
3.6-4
3.1
4.1-4.5
2.0
4.6-5
Disease reaction scale
Figure 7.10 Frequency distribution of disease response to Xanthomonas axonopodis
pv.manihotis in the BC1 family GM 315. SD = standard deviation inside
plot.
According to the frequency distribution graph for CBB resistance, the GM 315 family
(Fig. 7.10) showed segregation for resistance/susceptibility and had high standard
deviation (0.697) for individuals in each class: 46% of individuals of this family had a
score between 2.1 and 2.5, thus showing intermediate resistance, whereas 6% scored
less than 1.5 (i.e., resistant); and 2% scored more than 4.5 (i.e., susceptible).
Susceptible Bulk
Resistant Bulk
M Nga 19
Susceptible Bulk
Resistant Bulk
M Nga 19
Susceptible Bulk
Resistant Bulk
M Nga 19
Bulk Segregant Analysis. Bulk segregant analysis of the family GM 315 revealed
polymorphism between the parents and bulks with SSR markers (Figures 7.11).
Figure 7.11 Different types of polymorphism observed with BSA
Output 7-24
2003 Annual Report
Opening of the Bulks.
Primer SRRY65 showed differences between resistant and susceptible individuals of
each bulk (Figure 7.12) and it was evaluated in the whole family (GM 315) (Figure
7.13)
RP RB SB
Resistant individuals
Susceptible individuals
Figure 7.12 Primer SSRY65 evaluated in each individual that forms the resistant
and susceptible bulks in family GM 315. RP: M Nga 19. RB: Resistant
Bulk. SB: Susceptible Bulk
Figure 7.13 Some individuals from GM 315 family evaluated with SSRY65 marker
The presence of SSRY65 marker in resistant individuals of GM 315 family suggests
association with CBB resistance from field evaluations. A probability of P=0.046 was
obtained for association between marker SSRY 65 and field phenotypic data. Marker
specificity was 66%, while its sensibility reached up to 55%, which was the percentage
of individuals evaluated as resistant in the field, identifying the true positive results.
Conclusion
SSRY65 is a microsatellite marker associated with CBB resistance genes that can be
used to differentiate between resistant and susceptible individuals of GM 315 family to
CBB in field. The utility of this marker in cassava breeding is discussed.
Acknowledgements
√ Cassava Genetics Unit (CIAT)
√ Cassava pathology team (CIAT)
√ Cassava breeding project (CIAT)
√ Myriam Cristina Duque. (Statistician, CIAT)
√ Dr. Martín Fregene
Output 7-25
2003 Annual Report
References
CIAT (Centro Internacional de Agricultura Tropical). 2001. Annual Report. Cali,
Colombia.
Dellaporta, S.L., Wood, J., Hicks, J.R. 1983. A plant DNA minipreparation: version II.
Plant Mol Biol Rep 1:19.
Jorge, V., Fregene, M., Vélez, C.M., Duque, M.C., Tohme, J., Verdier, V. 2001. QTL
analysis of field resistance to Xanthomonas axonopodis pv. manihotis in
cassava. Theor Appl Genet. Vol 102: 564-571
Jorge, V., Fregene, M., Duque, M.C., Bonierbale, M.W., Tohme, J., Verdier, V. 2000.
Genetic mapping of resistance to bacteria bligth disease in cassava (Manihot
esculenta Crantz). Theor Appl Genet. Vol 101: 865-872
Mba, R.E.C., Stephenson. P., Edwards, K., Melzer, S., Nkumbira, J., Gullberg, U.,
Apel, K., Gale, M., Tohme, J., Fregene, M. 2001. Simple sequence repeat (SSR)
markers survey of the cassava (Manihot esculenta Crantz) genome: towards an
SSR-based molecular genetic map of cassava. Theor Appl Genet. Vol 102: 2131.• Michelmore, R. W., Paran, I., and Kesseli, R. V. 1991. Identification of
markers linked to disease-resistance genes by bulked segregant analysis: A
rapid method to detect markers in specific genomic regions by using segregant
populations. Proc. Natl. Acad. Sci. USA. Vol 88: 9828-9832. Genetics.
Hahn S. K., Terry E.R., Leuschner E. K., Akobundu I. O., Okali C., Lal R. 1979.
Cassava improvement in Africa. Field Crop Res 2: 193-226.
SAS Institute. (1999-2001). Software release 8.2 (TSMO). Sun OS 5.8 Platform.
Activity 7.11
Detection of a phytoplasma associated
Frogskin Disease (FSD) in Colombia
with
cassava
Objective
To confirm the presence of a phytoplasma associated with cassava frogskin disease
(FSD).
Introduction
Frogskin disease (FSD) is an important disease affecting cassava roots, whose causal
agent remained unknown for many years. FSD has been reported with increasing
frequency in Colombia, Brazil, and Venezuela. In Colombia, for example, incidences of
up to 70% have been recorded in commercial fields in the production areas of Valle del
Cauca, Cauca, Meta, and the North Coast. Disease symptoms consist of small,
longitudinal fissures distributed throughout the root. As the roots increase in
diameter, the fissures tend to heal, giving the injuries a lip form. Root cortex or
epidermis presents a cork-like appearance that is peels off easily. Depending on the
severity of symptoms, the depth and number of lesions increase until the root becomes
deformed.
This study evidences the existence of an association between FSD and phytoplasma.
By applying molecular tools and microscopy, phytoplasma was successfully detected
in FSD-infected cassava roots, leaf midribs, petioles, and peduncles.
Output 7-26
2003 Annual Report
Methodology
Plant tissue. Roots, stems, petioles, and leaf midribs of both FSD-infected and
healthy cassava plants, grown in the field and greenhouse, were processed.
Microscopic analysis. Small pieces of tissue, about 1 mm × 2 mm, were excised and
then fixed in 2%-3% glutaraldehyde/0.1M phosphate buffer. The samples for electron
microscopy were prepared by ultra thin section (60-90 nm) and viewed with a
transmission electron microscope.
DNA extraction. Total DNA was extracted as described by Gilbertson et al,1991.
Nested PCR analysis. The primer pairs P1/P7 or R16mF2/R16mR1 (Gunderson and
Lee, 1996) were used for the first amplification, with an annealing temperature of
55°C. For the nested PCR, diluted (1:30) PCR products were used for amplification,
with the primer pair R16F2n/R16R2 at an annealing temperature of 50°C. PCR
products were analyzed by electrophoresis on 1.5% agarose gel.
RFLP analyses. The amplified PCR products were digested with the restriction
endonucleases Taq I, Rsa I, and Alu I. The restriction products were analyzed by
electrophoresis on 5% polyacrylamide gel.
Cloning and DNA sequencing. Purified PCR products were ligated in pGEM-T Easy
vector, which was introduced into the Escherichia coli strain DH5-a by electroporation
at 2.4 kV/cm2. Transformants were selected on blue/white color screening by plating
on LB/ampicillin/IPTG/X-gal media. Positive inserts were observed by plasmid
restriction with EcoRI and electrophoresis in 1.5% agarose gel. Different-sized
fragments were selected for sequencing by automated dideoxy sequencing (ABI Prism
377-96 DNA Sequencer), using a DNA-sequencing kit from Applied Biosystems.
Grafting. Cassava cuttings from the highly susceptible genotype Secundina were
grafted on cassava infected plants.
Results
The presence of phytoplasma in different plant tissues of affected plants was
confirmed by electron microscopy (Figure 7.14). The specific primers
R16mF2/R16mR1 and R16F2n/R16R2 were used in a nested PCR assay to detect
phytoplasma. Nested PCR revealed 1.3 kb fragments in root, stem, and leaf samples
from symptomatic plants (Figure 7.15). No fragments were obtained from healthy
plants.
Output 7-27
2003 Annual Report
B
A
Figure 7.14 Electron microscopy of healthy (A) and infected (B) cassava tissue.
M
1
2
3
4
5
6
7
8
9
10 11 12
13 14
15 16 17 M
Figure 7.15 A 1.3-kb fragment was amplified from diseased samples by. Lanes 1-2,
infected stems; 3-4, infected petioles; 5-6, infected leaf midribs; 7,
healthy roots; 8, healthy stems; 9, healthy petioles; 10, healthy leaf
midribs; 11-12, infected roots; 13-14, infected stems; 15 and 17,
periwinkle (Catharanthus roseus); lane 16, negative control; and lane M =
100 pb DNA marker.
Output 7-28
2003 Annual Report
Phytoplasma was also detected by PCR in the leaves of grafted stem fragments on
infected plants under greenhouse conditions, indicating successful transmission of the
pathogen. Sequence analysis of a cloned fragment revealed that the cassava
phytoplasma was similar to the Chinaberry yellows phytoplasma (GenBank acc. no.
AF495657, 16SrXIII Mexican periwinkle virescence group) and Cirsium white leaf
phytoplasma (GenBank acc. no. AF373106, 16SrIII X-disease group), both with a
sequence homology of 100% and 99% in two partial fragments with a total of 1.01 kb
(Figure 7.16).
100%
95%
FSD_PCR-6
FSD_Clone_8-3
Chinaberry_Yellows*
Cirsium_White_Leaf*
Gaillardia_Phyllody*
Dandelion_Virescence*
Figure 7.16 Homology tree of 16S rRNA sequences from 6 phytoplasmas, including the
sequences from cloned and direct PCR fragments obtained from cassava.
* = GenBank accession.
Output 7-29
2003 Annual Report
Digestion with Taq I, Rsa I, and Alu I of amplified products of different samples showed
similar restriction patterns (Figure 7.17).
M 1 2 3 4 5 6 7 8 9 10 11 12 13 M M 1 2 3
A
Figure 7.17.
4 5 6 7 8 9 10 11 12 13 M
B
Restriction enzyme analysis of 16S rDNA after PCR amplification
with primer pair R16F2n/R2, using the endonucleases Rsa I (A) and Alu
I (B). Lane M = 1-Kb DNA marker.
Conclusions
Phytoplasma was successfully detected in all FSD-infected tissues by electron
microscopy, and nested PCR techniques. Among the methods used in this study, PCR
was
the
most
sensitive
for
detecting,
identifying,
and
classifying
phytoplasma. Sequence homology from a cloned fragment, obtained from an infected
cassava plant, was 100% similar to the Chinaberry yellows phytoplasma and 99%
similar to that of Cirsium white leaf. This is the first report of a phytoplasma being
associated with FSD in cassava. These results allow us to infer the possible role played
by the phytoplasma in this disease. Future research will involve the evaluation of
additional samples with other groups of enzymes as well as sequence analysis to
classify the phytoplasmas. Experiments are underway to achieve remission of
symptoms with the antibiotic chlortetracycline. Other research topics will include the
development of specific primers for pathogen detection, vector identification, and
classification of phytoplasmas associated with FSD.
References
Deeley, J., W.A. Stevens, and R.T.V. Fox. 1979. Use of Dienes' stain to detect plant
diseases induced by MLOs. Phytopathology 69:1169-1171.
Dyer, A.T. and W.A. Sinclair. 1991. Root necrosis and histological changes in surviving
roots of white ash infected with mycoplasma-like organisms. Plant Dis. 75:814819.
Gilbertson, R.L., M.R. Rojas, L.D. Russel, and D.P. Maxwell. 1991. The use of the
asymetric polymerase chain reaction and DNA sequencing to determine genetic
Output 7-30
2003 Annual Report
variability among isolates of bean golden mosaic geminivirus in the Dominican
Republic. J. Gen. Virol. 72:2843-2848.
Gunderson, D. E. and I. M. Lee. 1996. Ultrasensitive detection of phytoplasmas by
nested-PCR assays using two universal primer pairs. Phytopath. Medit. 35:144151.
Activity 7.12. a
7.12. b.
Transmission of a phythoplasma affecting cassava seedlings
and identification of indicator plants.
FSD symptoms recede after chlortetracycline treatment.
Objective
To determine the causal agent of FSD in the cassava crop.
Experiment 1:
Materials and Methods
Plantlets from 10 Catumare- and 10 Manzana-affected field plants (Rozo, Palmira,
Valle del Cauca, Colombia), and 20 disease-free plants of the same varieties
(Montenegro, Quindío, Colombia) were treated by chlortetracycline, according to
reports indicating effect of this antibiotic against phytoplasmas (Agrios, 1997).
For all experiments, the following precautions were included. Stakes were selected at
harvest time to ensure FSD was present. CIAT virologists indicated that roots were
affected by FSD according to symptoms. The stakes were planted in pasteurized soil,
free of FSD, in plastic pots (10") or bags placed in isolated glass- or screenhouses at
CIAT-Palmira (except Experiment no. 3). All plants were maintained in anti-aphid
cages, and healthy Secundina plants were included. This was to monitor the presence
of vectors. These plants did not show any symptoms during the experiments. Plants
and cages were fumigated perodically, rotating the following products: Vertimec® 1.8%
CE (0.5 cc/l of commercial product, abamectin), Malathion® (malathion, 1 cc/l of
commercial product), Sistemin® (dimethoate, 3 cc/l of commercial product), and foliar
fertilizers. All results presented in this report were checked by a CIAT Virology
Specialist to make sure symptoms were caused by FSD.
Stem cuttings, with the medulla previously perforated with a drill, were immersed in a
solution of 1500 ppm of chlortetracycline (750 ppm prepared based on a liquid
solution and 750 ppm on capsules) during 10 min. on planting day. Plants of each
treatment were planted at CIAT-Palmira in a glasshouse (temperature and RH:
minimum 19 °C and 31%, maximum 28 °C and 98%), and in a screenhouse
(temperature and RH: minimum 20 °C and 26%, maximum 39°C and 98%). All stem
cuttings were maintained in different cages and other precautions were taken to avoid
infection among plants. During 3 months, the soil was watered monthly with the same
chlortetracycline solution (200 ml/plant). Plantlets were treated twice a week with
1500 ppm of chlortetracycline by foliar applications. After 3 months, the dose was
reduced to 1000 ppm. Twenty plantlets (Manzana and Catumare) were also included
from plants affected with FSD, and 20 plantlets from healthy plants, without FSD.
Output 7-31
2003 Annual Report
The plants were evaluated periodically to detect symptoms on leaves. After 4 months,
stems of germinated plants were grafted with Secundina. Grafts were made directly on
plants or through rootstock cuts, maintained in deionized water.
Results
Table 7.10. shows results obtained after grafting with Secundina.
Table 7.10. Effect of applications of chlortetracycline on phytoplasm of FSD-infected
cassava plants.
Variety
Origin of plant
material
Place
Catumare
Screenhouse
Manzana
Screenhouse
Catumare
Glasshouse
FSD-affected
Manzana
Glasshouse
plants
Catumare
Screenhouse
Manzana
Screenhouse
Catumare
Glasshouse
Manzana
Glasshouse
Catumare
Screenhouse
FSD-free plants
Manzana
Screenhouse
Catumare
Glasshouse
Manzana
Glasshouse
aNo. of plants analyzed in parentheses.
Applications with
chlortetracycline
Yes
No
No. of FSD-affected
plants by grafting with
Secundinaa
3
3
5
5
2
3
4
4
0
0
0
0
(3)
(3)
(5)
(5)
(2)
(3)
(5)
(5)
(4)
(3)
(5)
(5)
All plantlets from affected plants in the field showed foliar symptoms in the Secundina
grafts. No plantlet from the field of healthy plants showed symptoms in Secundina
grafts on Catumare or Manzana. The effectiveness of obtaining grafts in the
screenhouse was less than in the glasshouse. However, foliar symptoms were observed
in both. Foliar applications of chlortetracycline do not reduce the incidence or severity
of FSD.
Phytoplasm transmission to cassava (Secundina) plants susceptible to FSD, produced
in vitro, and free of disease was successful through grafting.
Experiment 2:
Materials and Methods
Two foliar applications of chlortetracycline (1000 ppm, liquid form] were made weekly
during 6 weeks to plants of SM 1219-9 and La Reina that showed leaf symptoms
indicating FSD infection. Plants were located in a glasshouse (temperature and RH:
minimum 19 °C and 31%, maximum 28 °C and 98%) at CIAT-Palmira. The stakes of
these plants were obtained from FSD-affected plants (Jamundí, Valle del Cauca,
Colombia).
Output 7-32
2003 Annual Report
Results and Conclusions
We observed that the affected leaves (curling and mosaic) remained affected through
the applications (six plants). The new leaves also showed a severity similar to the
affected leaves of plants (five) untreated with chlortetracycline. It is concluded that
foliar applications with a high dosage of chlortetracycline do not inhibit leaf symptoms
related to FSD.
Experiment 3:
Materials and Methods
Cuttings of Secundina genotypes, M Bra 383 and La Reina, from affected plants from
the glasshouse (cuttings with FSD leaf symptoms) or from the screenhouse (cuttings
without FSD leaf symptoms) were taken from plants in plastic pots. After cutting the
true leaves of the cuttings, they were rooted in deionized water at different doses of
chlortetracycline (injectable form, capsules can cause greater levels of intoxication of
the plant, and are less effective against FSD). The cuttings were incubated in a
laboratory with a controlled temperature system (min. 20 °C, max. 25 °C), and 12-h
alternate periods of light and darkness. The high humidity (66%-98%) was achieved
through the use of closed boxes for the rooting of the cuttings.
Results and Conclusions
Table 7.11 presents the treatments used and results obtained.
Table 7.11.
cuttings.
Genotype
Secundina
Effect of chlortetracycline on cassava frogskin disease (FSD)-affected
Source
Screenhouse, asymptomatic plants (foliar);
these were previously in a glasshouse and
expressed leaf symptoms
Glasshouse, plants affected with leaf
symptoms
M Bra 383
Output 7-33
Glasshouse, plants affected with leaf
symptoms from field plants with FSDaffected roots
Dose of
oxychlortetracycline
(ppm)
0
2.5
5
10
25
50
0
2.5
5
10
25
50
0
2.5
5
10
25
50
No. cuttings
affected by FSD
and total no.
cuttings
analyzeda
4 (4)
2 (3)
4 (4)
4 (4)
3 (6)
0 (3)
5 (5)
5 (5)
5 (5)
5 (5)
1 (5)
0 (5)
5 (5)
5 (5)
5 (5)
5 (5)
3 (5)
2 (5)b
2003 Annual Report
aEvaluated
26, 32, and 40 days after initiation of chemical treatment. No. of plants analyzed
given in parentheses. bTwo cuttings without FSD, one affected cutting, two cuttings with no leaf
formation.
The inhibition of leaf symptoms caused by FSD was successful in two experiments
using a dosage of 50 ppm chlortetracycline. The leaves of affected plants treated with 0
ppm of chlortetracycline showed presence of phytoplasm through nested PCR. Similar
tests should be carried out with a greater number of cuttings.
This is the first report of FSD symptoms being suppressed by chlortetracycline.
The cuttings of the variety La Reina did not show FSD foliar symptomatology despite
being infected and showing foliar symptoms in a glasshouse where cuttings were
obtained to establish the experiment.
Cuttings treated with gentamicin (50 ppm) did not form leaves to adequately evaluate
the antibiotic’s effect on FSD. A treatment with 10 g/l of sugar to reduce the effect of
phytotoxicity of chlortetracycline on the formation of leaves and roots did not function
because it increased rotting of the stems by microorganisms.
Cuttings place in continuously oxygenated deionized water (by means of an air pump)
with (50 ppm) and without chlortetracycline did not reduce the effect of phytotoxicity
on the plants. Plants without the antibiotic rapidly (3 weeks) formed roots and leaves.
Plants with the chemical treatments formed leaf primordials and rooting callus, but
because of the phytotoxicity, no true leaves were formed.
In preliminary trials treating cuttings with 50 ppm of gentamicin, inhibition of FSD
foliar symptoms was not observed.
Experiment 4:
Stakes with one, two, or three leaf buds from FSD-affected plants (Jamundí) of the
variety SM 1219-9 were treated during 15 min. with 0, 25, and 50 ppm of
chlortetracycline or gentamicin respectively, and planted in sterile sand under
screenhouse conditions at CIAT-Palmira.
The substratum is humidified daily with the antibiotics (renewed every 3 days). This
trial is in progress. Transplanting is scheduled to containers with sand, and applying
nutritive solutions, constantly adding the antibiotics to obtain plants with roots
developed sufficiently large to observe the effect of chlortetracycline on FSD. The
treatments with 50 ppm of chlortetracycline and gentamicin inhibit germination of the
leaf buds; 25 ppm permits germination, and adequate plant development.
Acknowledgments
• Robert Zeigler, Kansas State University
• Jaime Eduardo Muñoz, Universidad Nacional de Colombia, Palmira
• CIAT-Virology Unit facilitated the Secundina grafting, and Tulio Rodríguez
performed the indexing.
• Agrovelez S.A. (Jamundí, Valle).
Output 7-34
2003 Annual Report
References
Agrios, G.N. 1997. Plant Pathology 4th edition.Academic Press. USA. P 458.
Activity 7.13.
Identification of cultural practices and strategy to control
frogskin disease in cassava
Objective
To develop a methodology that includes a process for disinfecting cassava stakes of
FSD.
Experiment 1:
Before planting, infected cassava stakes (M Bra 383, harvested at Jamundí, Valle del
Cauca) were treated with thermotherapy. Temperature in the glasshouse ranged from
19 °C to 28 °C (although the temperature sporadically rose above 25 °C in the antiaphid cage), and RH from 31% to 98%. Plants were evaluated periodically to detect
FSD symptoms in leaves. Table 7.12 presents results.
Table 7.12.
Effect of four hot water treatments on germination of stem cuttings and
frogskin disease (FSD).
Pretreatment
None
54 °C during 5 min
56 °C during 5 min
58 °C during 5 min
60 °C during 5 min
aFor each treatment six
Main treatment
None
54 °C during 10 min
56 °C during 10 min
58 °C during 10 min
60 °C during 10 min
cuttings were used.
Germination of stem
cuttings
(no. of cuttings)a
5
4
2
3
1
Disease incidence of
FSD according to
foliar symptoms (%)
100
50
50
0
0
Stakes of the variety M Bra 383, treated with hot water for 15 min at 60 °C, reduced
germination severely. Germination rates after 54 °C was highly acceptable. Plants
treated at 58 °C or 60 °C apparently were clean of FSD, therefore the following
experiment was designed with the main objective to improve plant vigor of treated
plants to realize analysis for presence of FSD.
Experiment 2:
Fourteen cassava genotypes (M Chn 2, HMC 1, M Arg 2, M Bra 325, M Bra 829, M Bra
839, M Bra 856, M Bra 882, M Bra 886, M Col 634, M Col 1178, M Col 1468, M Cub
74, and M Per 16), infected by FSD, were harvested at CIAT-Santander de Quilichao.
Thirty-two hot water stake treatments in combination with Agrodyne®SL (13.20 g/l
iodine polietoxi-polipropoxi-poloetoxi–ethanol complex, 1.59 g/l iodic acid,
Electroquímica West S.A., Medellín, Colombia) were tested for their effect on
germination of stem cuttings and presence of FSD, applying two methods of detection.
Output 7-35
2003 Annual Report
All stem cuttings were planted in plastic bags, and maintained isolated in an antiaphid cage outdoors.
Germination of treated stem cuttings was better outside the glasshouse (temperature
and RH: minimum 20 °C and 30%, maximum 38 °C and 96%) than inside, because
outside the glasshouse stakes are less affected by saprophytic fungi. Stems of
germinated plants (about 3 months old) were grafted onto Secundina. Table 7.13
presents results.
Output 7-36
2003 Annual Report
Table 7.13.
Effect of hot water treatment and iodine on germination rates of stakes
from cassava infected with frogskin disease (FSD).
Pretreatment
Without
Agrodyne
With
Agrodyne
No
treatment
49 °C/1 h
55 °C/10 m
60°C/5 mc
60 °C/5 m
55 °C/10 m
60 °C/5 m
55 °C/10 m
60 °C/5 m
55 °C/10 m
60 °C/5 m
60 °C/5 m
55 °C/10 m
55 °C/10 m
60 °C/5 m
55 °C/10 m
55 °C/10 m
60 °C/5 m
60 °C/5 m
49 °C/1 h
60 °C/5 m
60 °C/5 m
49 °C/1 h
-
Main treatment
Without
Agrodyne
With Agrodyne
No treatment
25°C/5 m c
25°C/1 h c
-
60 °C/10 m
60 °C/5 m
60 °C/10 mb
60 °C/10 m
60 °C/10 m
60 °C/10 m
60 °C/12.5 m
60 °C/12.5 mb
60 °C/20 m
60 °C/15 m
60 °C/12.5 m
60 °C/10 mb
60 °C/12.5 mb
60 °C/15 m
60 °C/15 mb
60 °C/15 mb
60 °C/17.5 m
60 °C/17.5 mb
60 °C/17.5 m
60 °C/17.5 mb
60 °C/20 m
60 °C/20 m
60 °C/30 m
60 °C/30 m
60 °C/30 m
25 °C/5 m d
60 °C/10 m c
49 °C/1 h c
60 °C/5 m c
-
No.
GermiNo.
ungermi- germi- nation of
nated stakes (%)
nated
stakes stakes
0
0
1
6
12
13
100.0
100.0
92.9
1
6
3
2
1
6
6
24
9
3
7
4
6
9
9
6
6
8
5
5
7
4
8
5
10
10
10
2
10
5
16
8
4
2
4
3
11
3
1
2
1
1
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
83.3
72.7
72.7
66.7
66.7
40.0
33.3
31.4
25.0
25.0
22.2
20.0
14.3
10.0
10.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
No. of plants with
positive response
to phytoplasm
(PCR) or FSD
(indexing), and no.
of plants analyzed
Nested Grafting
PCR
with
Secundina
2 (2)
1 (2)
1 (2)
1 (1)
0 (1)
0
1
0
0
(2)
(2)
(2)
(1)
1
0
1
1
1
1
0
0
0
(2)
(1)
(2)
(1)
(1)
(2)
(1)
(1)
(1)
0 (2)
1 (1)
1 (1)
0 (1)
aAgrodyne®SL (13.20 g/l iodine complex polietoxi-polipropoxi-poloetoxi–ethanol, 1.59 g/l iodic acid,
Electroquímica West S.A., Medellín, Colombia), 1 ml/l Inex A. b6 hours after pretreatment. c1.5 ml/l
Agrodyne. d3 ml/l Agrodyne.
Output 7-37
2003 Annual Report
The treatment of stakes with Agrodyne without thermotherapy does not affect
germination. The use of hot water at 60 °C up to 10 min reduces germination, but the
plants obtained are possibly healthier. At this temperature, the use of iodine can be
included in the hot water treatment. It is unclear whether phytoplasm can be
definitely inactivated through thermotherapy. A trend was observed that treatment is
more important during a relatively long period, rather than at an extremely high
temperature.
The most effective treatment is 49 °C during 1 hour (pretreatment) followed by a main
treatment at 60 °C for 10 min, without using Agrodyne. The germination achieved with
this treatment was most acceptable (72.7%, 17 stakes germinated). Cleaning was
demonstrated as effective through indexing with Secundina.
Experiments 3 and 4:
Stakes with one, two, or three leaf buds of the SM 1219-9 genotype affected by FSD
(Agrovelez, Jamundí), were treated at 60 °C during 5 min (pretreatment), and the
following day at 60 °C during 10 min (main), and planted in sterile sand in a
screenhouse (temperature and RH: minimum 20 °C and 26%, maximum 39 °C y 98%)
at CIAT-Palmira. This treatment was carried out with 1 ml of Agrodyne/l in hot water.
The treatment was too strong, and affected germination of leaf buds. No plant
germinated, although callus was present in the stakes. In another experiment, without
Agrodyne, the temperature for the pretreatment was lowered to 55 °C (treatment 1)
and to 50 °C (treatment 2), and for the main treatment to 60 °C (treatment 1) and 55
°C (treatment 2), but without promising results because no stake germinated.
Stakes of a leaf bud without heat treatment did not germinate either; stakes with two
or three leaf buds germinated satisfactorily. Experiments are being scheduled with a
treatment of leaf buds at 49 °C, with a duration between 30 and 60 min.
Acknowledgements
We thank CIAT-Virology for facilitating the Secundina grafts, and Tulio Rodríguez for
carrying out the indexing. We also thank Agrovelez S.A. (Jamundí, Valle) for access to
infected field materials, and James George (Central Tuber Crops Research Institute,
Kerala, India) for suggestions concerning use of short stem cuttings.
Activity 7.14.
Evaluation of the influence of the soil as a source of FSD
vectors
Objectives
1.
2.
To evaluate the soil as a possible source of microorganism vectors of FSD.
To evaluate whether the presence of aerial vectors is related with dissemination of the
disease.
Output 7-38
2003 Annual Report
Materials and Methods
In the municipality of Sincelejo (Sucre), Chochó basin, the possible influence that the
soil may have as a source of FSD vectors is being evaluated. In the region, the disease
occurs most frequently in lots where the presence of FSD has been reported
previously, unlike nearby lots that remain disease free. For the trial, the variety M Tai
8 was chosen as presenting high susceptibility to FSD, and being one of the
commercial genotypes most cultivated in the region, facilitating the obtaining of seed
from lots where the disease has never occurred. Treatments were:
•
•
•
•
Healthy plants in the screenhouse,
Diseased plants in the screenhouse,
Healthy plants outside the screenhouse, and
Diseased plants outside the screenhouse.
Two muslin cages were constructed, 1.80 m in height, one for healthy plants (10.5 m
long x 5 m wide), and one for affected plants (4.5 m long x 5 m wide). The rest of the
trial was planted outside the screenhouse.
An experiment design was used of divided plots with three repetitions. The experiment
unit for healthy plants inside the screenhouse, consisted of a plot of nine plants,
distributed in three furrows of three plants each planted at 1 m x 1 m. For the
diseased plants, three plants were planted for each repetition.
Outside the cage, the same experiment design was kept—three plots of healthy and
diseased plants, with two furrows for 11 healthy plants and eight diseased.
For the seed cut, the machete was disinfected in a solution of 1% sodium hypochlorite.
As part of the management, insecticide was applied, 1 week Sistemin® (dimethoate, 3
cc/l of commercial product), and the following week Malathion® (Malathion, 1 cc/l of
commercial product). Application was only made within cages, and to half the outside
plots, so that of the three repetitions outside, half received insecticide, and the other
half was conserved without application. A barrier of the commercial genotype M Ven
25 was planted between the two to also evaluate the effect of aerial vectors, or whether
the transmission is carried out through some agent of the soil.
Evaluations of the treatments will be made at time of harvest, observing incidence and
severity in the roots.
Activity 7.15.
Isolation and characterization of Agrobacterium tumefaciens
from soil and cassava roots.
Objectives
1.
2.
3.
To isolate Agrobacterium tumefaciens from soil samples and cassava roots.
To evaluate the pathogenicity of strains isolated from several cassava varieties.
To characterize isolates of A. tumefaciens pathogenic on carrot disks using PCR
with specific primers.
Output 7-39
2003 Annual Report
Materials and Methods
Isolation of Agrobacterium tumefaciens. To isolate A. tumefaciens, samples were
initially taken from both healthy and diseased cassava roots and from soil where
diseased plants were sown.
Handling of soil samples. Of each soil sample, 100 grams were weighed and
dissolved in 100 ml sterile distilled water (SDW), agitated for 30 minutes, and then left
to settle for another 30 minutes. Four serial dilutions in 9 mL of 0.75% NaCl were
performed based on this first mixture (base solution) and 0.1 mL of each was planted
on DIM (D1) media, which is selective and differential for Agrobacterium sp. isolates.
The Petri dishes were incubated at 30 ºC for 24 hours. Colony-forming units (CFUs)
growing on D1 were counted and those of yellowish-orange color were selected and
planted per isolate on the same media for purification.
Handling of plant samples. The healthy and diseased roots collected were washed
with deionized water and processed by separating the plant tissue in 4 layers
depending on their distance to the phloem (layer 1 = external, layer 4 = near the
phloem). The portion obtained from each layer was washed with deionized water for 15
minutes, disinfected in 50% alcohol and then excess alcohol was eliminated with
SDW. Smaller portions of each sample were cut and placed on D1 medium and then
incubated at 30ºC for 48 hours. Agrobacterium colonies of characteristic color
(yellowish orange) in the D1 medium were selected and purified on this media.
Pathogenicity test. Bacterial colonies isolated from the soil and cassava roots (Table
7.14) were planted on nutritive agar with sucrose (5%) and incubated at 30ºC for 48
hours.
To evaluate pathogenicity on carrot disks, fresh tubers were washed in deionized
water, then submerged in sodium hypochlorite (5%) for 10 minutes and subsequently
washed with SDW. Tubers were finally dried with sterile towels and small carrot disks
were cut and placed in sterile Petri dishes.
The colonies obtained were inoculated by puncturing around the vascular cambium of
the carrot disks and incubated for 3 weeks at 25ºC in moist chambers. Each strain
was inoculated by duplicate and gall formation around the vascular cambium was
observed on a daily basis.
Carrot disks inoculated with SDW were included as negative check, and A.
tumefaciens isolate 1182 as positive check.
Two tissue-culture grown varieties Secundina and M Bra 383 were used to test the
pathogenicity of strains isolated from cassava. The in vitro plants were planted on
sterile soil, and kept in cages to avoid the presence of insects as possible disease
vectors. They were fertilized weekly by intercalating the following three NPK fertilizer
solutions at 0.5%: 15-15-15, Coljap Producción 5-15-30, and Coljap Desarrollo 30-76, produced by Industria Agroquímica S.A., Bogotá, Colombia. Inoculation was carried
out 1 month afterwards. Strains selected for a pathogenicity test in cassava were
isolated from cassava roots (isolates 23, 24, and 27) with typical symptoms of frogskin
Output 7-40
2003 Annual Report
disease and from soil samples of infected cassava crops (isolates 28, 29, 30, 33, and
35B).
For inoculation, a bacterial suspension of 1x108 CFU/mL (0.5 absorbancy) of the
strains was prepared to inoculate the soil, roots, or base of corona. The soil was
inoculated by adding the bacterial suspension to the surface. To inoculate the roots,
plants were removed from the soil, their roots washed and tips cut, after which they
were submerged in bacterial suspension for 30 minutes. Plants were then re-planted
and placed within the cage. To inoculate the base of the corona, a puncture was made
diagonally in this part of the plant using a 1-mL syringe to inject the inoculum. Once
the inoculation was finished, all plants were placed in the cages and kept in a closed
growth chamber, with 12 hours daylight, at a constant temperature of 20-25ºC and
80% relative humidity, thus favoring the colonization of the pathogen. Each month the
roots were evaluated under the stereoscope to determine any changes in morphology.
Molecular characterization of isolates. The DNA was initially extracted from 43
strains isolated from soil and roots that resulted positive in pathogenicity tests on
carrot disks and the check strain of A. tumefaciens (1182). Extraction was performed
following the Boucher et al. (1985) protocol and the concentration of all strains was
adjusted to 20 ng/µL to be amplified with the specific primers VCR/VCF paired with
the Vir C region of the Ti plasmid of A. tumefaciens (Sawada, 1995). The Vir C region
forms part of the Ti plasmid virulence genes that measured the transfer of plasmid TDNA to the plant where an excessive proliferation of cells occurs, causing one or more
galls to form. Each PCR reaction was carried out in 25-µL final volume, with final
concentrations of 0.2 mM dATP, dCTP, dGTP, and dTTP; 1.5 mM MgCl2; 1.5 U Taq
polymerase; 0.1 µM of each primer; 1X Taq polymerase buffer; and 100 ng template
DNA. For the reaction of the negative check, the DNA was replaced by SDW. The PCR
was performed in a MSJ-Research PTC-100 thermal cycler with the following
amplification program: 2.5 min at 95°C; 40 cycles of denaturation for 1 min at 95°C,
pairing for 1 min at 55°C, and extension for 2 min at 72°C; and a final extension for 7
min at 72°C.
The amplification product was visualized in 2.0% agarose gel with TBE 0.5X buffer
dyed with ethidium bromide (1 µL disolution at 10 mg/ml for 100 ml agarose gel). A
100-bp molecular weight marker was used to estimate the size of the amplified
fragments.
Results
Pathogenicity test. Fifty-five strains with typical A. tumefaciens morphology were
initially isolated from soil and cassava roots in D1 media. All were submitted to
pathogenicity tests on carrot disks, where gall formation was observed 2 weeks after
inoculation (Figure 7.18) with 43 of the isolated strains (Table 7.14).
The pathogenicity test in cassava has not yet shown changes in root morphology.
Inoculated plants have been periodically examined and no apparent changes have
been observed in the roots. To promote plant development and root enlargement, all
inoculated plants were transplanted to large pots and will be monitored monthly until
roots reach a size in which any morphological change resulting from disease caused by
the inoculated strains can be observed by the naked eye.
Output 7-41
2003 Annual Report
B
A
C
Figure 7.18. Gall formation on a carrot disk. A. Positive control (A. tumefaciens strain
1182). B. Negative control (sterile distilled water). C. Isolate 23 (root,
genotype CM 2772-3).
Once the pathogenicity test was performed, the DNA of positive strains was extracted
and quality visualized. The concentration of all strains was adjusted in 20 ng/µL and
was amplified with the specific primers VCR/VCF. A 730-bp band was observed in the
check strain 1182 and in most of the amplified strains (Figure 7.19).
M 1 2
3
4
5 6 7 8
9 10 11 12 13 14 15 16 17 18 19
Figure 7.19. Amplification of DNA of A. tumefaciens strains obtained with the specific
primers VCR/VCF from samples of infected cassava roots and soil. Lane M =
100-bp marker; lanes 1-19 = Ralstonia solanacearum (18, 23, 24, 26B, 27,
28, 30, 32, 33, 40, 1182, 21, 10, 11, 15, 19, 1, 6); lane 17 = negative
control.
Table 7.14 presents the results obtained in the pathogenicity test and the
amplification of strains with VCR/VCF primers.
Output 7-42
2003 Annual Report
Table 7.14. Pathogenicity of A. tumefaciens isolates and DNA amplification with
specific primers (Vir C).
Isolate
no.
1
2
3
4
5
6
7
8
Source
Root
Root
Root
Root
Root
Root
Root
Root
1
1
2
2
2
4
4
1, diseased
9
Root 1, diseased
10
Root 1, diseased
11
Root 1, diseased
12
Root 2, diseased
13
Root 2, diseased
14
Root 2, diseased
15
Root 2, diseased
16
17
18
19
20
21
Sterilized soil
Soil of a healthy
crop (Catumare)
Root (Manzana)
Root (GM 309-7)
Root (GM 309-7)
Root (GM 309-7)
22
Root (GM 309-7)
23
Diseased root
(CM2772-3)
Diseased root
(MBRA 383)
Diseased root
(Catumare)
Diseased root
(Manzana)
Diseased root
(Manzana)
Diseased root
(Venezolana)
Soil of a diseased
crop
24
25
26A
26B
27
28
Output 7-43
Origin
Growth chamber
Growth chamber
Growth chamber
Growth chamber
Growth chamber
Growth chamber
Growth chamber
Santander de
Quilichao
Santander de
Quilichao
Santander de
Quilichao
Santander de
Quilichao
Santander de
Quilichao
Santander de
Quilichao
Santander de
Quilichao
Santander de
Quilichao
Quindío
Quindío
Identification
Petri dish
Petri dish
Petri dish
Petri dish
Petri dish
Petri dish
Petri dish
Layer 1
1
2
1
2
3
2
3
Pathogenicity Specific DNA
amplification
1
2
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+/+
+
+
+
Layer 2
+
+
+
Layer 3
+
+
+
Layer 4
+
+
+
Layer 1
+
+
+
Layer 2
+
+
-
Layer 3
+
+
+
Layer 4
+
+
+
+/-
+/+/-
+
+
+/+
+/+
+/+
+/+
+
+
+
+
+
+
+
+
+
+
CIAT
Dil 101
Dil 102, 2nd
camping
Layer 2
Layer 1
Layer 2
Layer 4, colony
1
Layer 4, colony
2
Layer 4
Palmira
Layer 1
+
+
+
Palmira
Layer 4
+
+
+
Palmira
Layer 1
+
+
Palmira
Layer 2
+
+
+
Sincelejo, Sucre
Layer 4
+
+
+
Jamundí, Valle
No. 2, colony 2
+
+
+
2003 Annual Report
Table 7.14. Continued.
Isolate
No.
29
30
31
32
Source
Soil
Soil
Soil
Soil
33
34A
34B
35A
35B
36
of
of
of
of
a
a
a
a
diseased
diseased
diseased
diseased
Origin
crop
crop
crop
crop
Soil of a diseased crop
Soil of a diseased crop
Soil of a diseased crop
Soil of a diseased crop
Soil of a diseased crop
Soil of a non-diseased
crop (Catumare)
37
Soil of a non-diseased
crop (Catumare)
38
Soil of a non-diseased
crop (Catumare)
39
Soil of a non-diseased
crop (Catumare)
40
Soil of a non-diseased
crop (Manzana)
41
Soil of a non-diseased
crop (Manzana)
42
Soil of a non-diseased
crop (Manzana)
43
Soil of a non-diseased
crop (Manzana)
1182
Agrobacterium
tumefaciens
Ti
Agrobacterium
plasmid tumefaciens
Identification
Jamundí
Jamundí
Jamundí
Santander de
Quilichao
Sincelejo
Sincelejo
Sincelejo
Sincelejo
Sincelejo
Quindío
No.
No.
No.
No.
3,
4,
5,
1,
colony
colony
colony
colony
2
1
1
2
No. 1, colony
No. 2, colony
No. 2, colony
No. 3, colony
No. 3, colony
Colony 1
1
1
2
1
2
Quindío
Pathogenicity Specific DNA
amplification
1
2
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Colony 2
+
+
+
Quindío
Colony 3
+
+
+
Quindío
Colony 4
+
+
-
Quindío
Colony 1
+
+
+
Quindío
Colony 2
+
+
Quindío
Colony 3
+
+
+
Quindío
Colony 4
+
+
+
+
+
+
Biotechnology Project-CIAT
Miniprep
+
References
Boucher C; Barberis P; Trigalet A; Demery D. 1985. Transposon mutagenesis of
Pseudomonas solanacearum: isolation of Tn5-induced avirulent mutants. J Gen
Microbiol 131:2449-2457.
Sawada H; Ieki H; Matsuda I. 1995. PCR detection of Ti and Ri plasmids from
phytopathogenic Agrobacterium strains. Applied and Environmental Microbiology.
February, p. 828-831.
Output 7-44
2003 Annual Report
Activity 7.16.
Selection and treatment of cassava stakes trial for the
management of diseases and evaluation of genotypes at
Sincelejo (Sucre)
Specific objective
To evaluate different genotypes and treatments for the management of SED and CBB,
and to train farmers in the zone.
Methodology
In Sincelejo, effects of some management practices for SED and CBB, endemic
diseases in the region, were evaluated with the following treatments, utilizing stakes of
the local variety, Venezolana, taken from a field affected by the two diseases:
•
•
•
•
•
•
Fertilization with 300 kg/ha of 15-15-15, 90 days after planting, for the plots with
fertilizer.
Selection of stakes based on health and quality.
Thermotherapy with immersion in hot water at 49 °C for 49 min.
Immersion for 5 min in Koccide® (Copper hydroxide, 5 g/l of commercial product),
and Sistemin® (dimethoate, 3 cc/l of commercial product).
Stake immersion for 5 min in Score® (Difenoconazole, 2.7 cc/l of commercial
product), and foliar application at 60, 90, and 120 days after planting.
Traditional farmer practice, without treatment of the stakes.
The same treatments were made in plots without fertilization.
An experiment design of divided plots was used, with fertilizer as main plot, and
treatment in the subplot. The experiment unit consisted of a plot of 45 plants, planted
at 1 m x 1 m.
Results
Evaluation of planting material management. Table 7.15 shows that thermotherapy
did not affect germination. Fertilization reduced the incidence of bacterial blight,
presenting a significant minimum difference between trials of 7.489 (LSD 5%),
although the disease did not affect yield much because disease increased towards the
end of the wet season, when the cassava yield is assured. The best yield (20.06 t/ha)
was achieved with thermotherapy, in spite of the treatment not affording protection
during the crop cycle, but in germination, the treatment deals with material cleaning,
and for this reason the incidence of CBB was relatively high. The lowest incidence of
CBB was obtained with stake selection, independently of fertilization practices.
Output 7-45
2003 Annual Report
Table 7.15.
Effect of different treatments of ‘Venezolana’ cassava stakes and of soil
fertilization on the incidence of cassava bacterial blight (CBB) and yield.
Fertilizer
Treatment of stakes
Germination
CBB
(%)
incidence
(%)
15-15-15
Selection of stakes
98
7.41
15-15-15
100
11.11
Thermotherapy 49 °C for 49 min
15-15-15
Copper hydroxide x 5 min
99
2.96
15-15-15
Score® and foliar 60, 90, 120 d
100
10.37
15-15-15
Traditional, no treatment
100
8.15
Average
8.27
No fertilizer
Selection of stakes
100
10.37
No fertilizer
100
13.33
Thermotherapy 49°C for 49 min
No fertilizer
Copper hydroxide x 5 min
100
20.74
No fertilizer
Score® and foliar 60, 90, 120 d
100
19.26
No fertilizer
Traditional, no treatment
100
21.48
Average
17.13
7.49
LSD (5%)
Yield
(t/ha)
Dry matter
(%)
19.27
20.06
19.95
19.40
19.85
19.71
19.33
18.44
17.63
18.74
19.33
18.70
2.08
36.0
36.0
36.3
35.0
36.7
36.0
35.7
35.7
35.3
36.0
36.0
35.7
0.8
Genotype evaluation. Resistance was evaluated in the genotypes 6758-1, SM 1665-2,
CM 6119-5, SM1565-17, CM 4843-1, CM 6754-8, CM 4919-1, SM 1438-2, and
M Tai 8, and the local variety ‘Venezolana’ (M Col 2215) to these two diseases.
Presence of SED was not observed during cultivation, and incidence of CBB was very
low. Yield fluctuated between 24 t/ha for M Ven 25, and 47.4 t/ha for SM 1565-17,
while dry matter content in this genotype was the lowest, with 27.6%, and CM 4919-1
showed the highest, with 34.2% (Table 7.16).
Table 7.16.
Yield and reaction to cassava bacterial blight (CBB) of 10 cassava
genotypes, Sincelejo (Sucre).
CBB severitya
Genotype
Dry Matter (%)
Yield (t/ha)
1.0
CM 4843-1
30.7
41.6
2.0
CM 4919-1
34.2
42.6
2.0
CM 6119-5
33.9
25.6
2.0
CM 6754-8
33.2
30.6
2.0
CM 6758-1
30.7
30.3
2.0
SM 1438-2
33.9
32.0
2.0
SM 1565-17
27.6
47.4
2.0
SM 1665-2
30.7
38.6
2.5
M Ven 25
32.6
24.0
2.0
M Tai 8
33.2
42.0
Venezolana
36.7
19.9
aCBB severity: 1.0, no symptoms; 2.0, angular leaf spots; 2.5, leaf scorching.
Output 7-46
2003 Annual Report
Activity 7.17.
Evaluation of different treatments with Score® 250 E.C
(Difenoconazole) for the management of SED, Sincelejo
(Sucre)
Specific objective
To evaluate different doses of the commercial product Score® 250 E.C for control of SED.
Methodology
The effect of the commercial fungicide Score® 250 E.C (Difenoconazole) was evaluated
in Sincelejo for control of SED, which is endemic in the region. Cuttings of two highly
susceptible commercial genotypes, M Tai 8 and CM 4919-1, that had shown
symptoms of the diseases, were utilized. Before planting they were treated with the
following concentrations of Score®:
•
•
•
•
•
•
•
Dose
Dose
Dose
Dose
Dose
Dose
Dose
1:
2:
3:
4:
5:
6:
7:
0.5 cc/l
1.0 cc/l
1.5 cc/l
2.0 cc/l
2.5 cc/l
3.0 cc/l
control without application
An experiment design of divided plots was used, with the varieties as main plot, and
the doses as subplot. The experiment unit consisted of a plot of 27 plants planted at 1
m x 1 m with three repetitions per treatment. Fungicide applications were made when
symptoms of the disease occurred.
Results
Table 7.17 shows a slight reduction in germination for the commercial genotype M Tai
8 and a high one for CM 4919-1, due to the time elapsed between seed taking and
planting, awaiting good rainfall that could favor spore germination in the tissues. With
treatment six (3.0 cc/l), better control was obtained giving good germination of the two
varieties (92.6%) with a low incidence (area under disease progress curves [AUDPC]=
250.25) and severity (AUDPC = 33.54) of the pathogen in presence of the control
(incidence AUDPC = 1262.9; severity AUDPC = 68.8). Treatments 1 and 2 did not show
a reduction in SED incidence or severity compared to the controls. Doses between 0.5
and 1.0 cc/l are not recommended for application in commercial crops. Treatments
3,4, and 5 should be re-evaluated with plots presenting good germination.
Output 7-47
2003 Annual Report
Table 7.17.
Effect of different treatments of Score® 250 E.C in M Tai 8 and
CM 4919-1 on the incidence and severity of superelongation disease
(SED).
Dose Score®250 E.C
(cc/l)
Germination of
commercial genotypes
(%)
M Tai 8
92.33
93.67
84.67
14.67
98.67
92.67
98.67
81.57
SED incidencea
(AUDPC)
SED severity SED severityb
(AUDPC)
(average)
CM 4919-1
77.67
70.33
0.00
0.00
68.00
92.67
100.00
58.38
Dose 0.5
562.33
45.21
1.8
Dose 1.0
909.42
55.71
2.3
Dose 1.5
439.83
31.50
1.2
Dose 2.0
927.50
35.00
1.4
Dose 2.5
680.75
37.04
1.5
Dose 3.0
250.25
33.54
1.3
Control
1262.92
68.83
3.1
45.90
Average
719.10
571.64
19.73
LSD 5%
a AUDPC, area under disease progress curves of the disease. bSeverity scale: 1, healthy; 1.5,
small cankers on leaves, no damage to petioles; 2, cankers on leaves and very few on petioles,
leaves slightly affected; 2.5, higher number of cankers on petioles; 3, cankers on stems, no
elongation of internodes; 3.5, higher number of cankers on stems, leaf deformation; 4,
elongation of internodes; 4.5, dieback of branches; 5, plant death.
Activity 7.18.
Multiplying promising cassava genotypes to ensure sufficient
planting material for both greenhouse and field experiments
In the field. Two hundred and twenty seven promising cassava genotypes are being
propagated in a farm located in Rozo, Palmira (Department of Valle del Cauca,
Colombia) for greenhouse experiments on varietal resistance, genetic studies, and
disease management. The group includes two populations for Phytophthora spp.
resistance studies.
Meristem culture. In order to clean cassava cuttings by thermotherapy and meristem
culture, 55 genotypes are kept in vitro, with eight clones (plants) each. Eighty MBra
383 plants and 96 plants of Secundina genotypes were propagated in vitro to obtain
disease-free plants for use as an FSD indicator in assays searching for vector and
disease management.
Activity 7.19.
Genetics of resistance to rot caused by Phytophthora
tropicalis in two segregating populations of cassava
(Manihot esculenta Crantz)
Elizabeth Alvarez, John Loke, Sandra Rivera, and Germán Llano
Paper published in: Revista Fitopatología Colombiana Vol 26 (2): 61 - 66. Cali,
Colombia.
Output 7-48
2003 Annual Report
Abstract
Varietal resistance is one of the main tools for managing root rot in cassava, caused by
different Phytophthora species, and it is therefore necessary to understand the
complexity of the genetic base of this crop. The resistance of parents and progeny of
cassava families K (M Nga 2 x CM 2177-2) and CM 9582 (M Bra 1045 x M Cr 81) to P.
tropicalis was evaluated accordingly. Fresh roots of 69 individuals of family K and 43
of family CM 9582 were also inoculated, in addition to the parental materials of each
family. Based on the phenotypic evaluation and the molecular map of M Nga 2 (female
parent of family K), QTLs associated to the resistance to P. tropicalis were identified
and mapped, using simple marker analysis. Cassava family K genotypes, evaluated
during 2000 and 2001, showed a percentage of infected root area between 22% and
95%. The correlation between the evaluations of 2000 and 2001 was -0.15. Cassava
family CM 9582 genotypes showed an infected area between 70% and 90%. The
distribution of frequency of cassava family K genotypes, based on root area affected by
P. tropicalis, corresponds to a normal distribution, with one genotype presenting
moderate resistance in both years of evaluation, 50 genotypes susceptible, and 17
genotypes highly susceptible. Eight QTLs were defined, two of which accounted for
8.6% and 9% of phenotypic variance. Minor genes were found to confer resistance to P.
tropicalis.
Activity 7.20.
Training farmers, technicians, and extension agents in
participatory
research,
cassava
and
other
crops
management, and disease control strategies
Training
•
•
•
•
•
•
•
•
Sep, 2002. Alejandro Corredor, Universidad de Caldas, Manizales. Phytophthora
spp. culture management.
September – December 2002. Students from Universidad del Valle. Diana López
y Adriana Arenas. Pathogenicity of Agrobacterium tumefasciens in carrot and
“Coralito”.
Oct, 3 2002. 15 students from Universidad San Buenaventura. Molecular tools
for plant disease research.
Oct, 28 and Nov, 18 2002. 62 students from Universidad de Nariño. Molecular
characterization of Sphaerotheca pannosa, Xanthomonas axonopodis and
Sphaceloma manihoticola; integrated disease management; molecular tools for
plant disease research.
Nov, 26 2002. Universidad Nacional, Sede Medellín. 3 students of Biotechnology
Magister. Molecular trials used in cassava pathology
Dec 2002. Training of two research assistants from CENICAFE working on
phytoplasma and biocontrol agents.
March, 14. Luis Fernando Anzola from Ecollanos. Recommendation for cassava
assessment in Guainía, by using participatory research with indigenous and
settlers.
April, 1. Pedro Molina, Miembro de la Comisión Evaluadora del Programa Redes
de Asociación Cooperativa, Ministerio de Ciencia y Tecnología, Venezuela.
Integrated cassava disease management.
Output 7-49
2003 Annual Report
•
•
•
•
•
•
•
•
•
•
•
•
•
•
April, 2. Universidad del Cauca. Professor Andrés Torres. Information on
cassava pathology.
April, 4. Information on cassava diseases incidence in Valle del Cauca. Carlos
Enrique Gómez and Greicy Andrea Sarria. ICA Palmira.
April 22-May 2. Carlos Alberto Galvis. Cenicafé. Training in detection of
Phytoplasma in coffee.
May 8. Ten students from Universidad San Buenaventura. Disease
management strategies and molecular techniques.
May 20. Seven breeders form Cuba, Uganda, Thailandia, India. Cassava
Disease resistance.
Jun 9 – 13. Technicians and producers of plantain. Course: Integrated
management of Moko disease in plantain.
Jun – Sep. Technical assistance to Arvey Benavides, for Phytophthora
management in lulo (Solanum quitoense).
Jul 29 –31. Gira tecnológica en cultivos de yuca con productores asociados y el
equipo de trabajo de la Compañía Agroindustrial Yuquera. San Pablo (South of
Bolívar Department).
Aug, 12. Rebecca Lee and Santiago Fonseca. Ceniflores (Asocolflores). Ecological
practices for flowers disease management.
Aug, 15. Mario Pareja. Fundación Chemonics. Cassava disease management.
Hugo Martínez, Fundación Universitaria de Popayán (Ecology). Advisory for
production of Pleurotus ostreatus y Pleurotus sajor caju
Sandra Patricia Fajardo Daza and Rosa Judith Aranda Muelas, Universidad
Nacional de Colombia, Ingenieria Agroindustrial, Palmira. Advisory for the
production of Pleurotus ostreatus.
Sep, 8. Most important cassava diseases. Ministry of Agriculture from Nigeria
and eight attendants.
Sep, 10. Manage of Phytophthora in pineapple. Francisco Vargas.
Seminars
Detección de un Fitoplasma Asociado con la Marchitez Letal de Palma de Aceite (Elaeis
guineensis) en Colombia. CIAT. Oct 30, 2002. Elizabeth Alvarez.
Development of ecological practices to manage Phytophthora root rot of cassava
(Manihot esculenta) E Alvarez, JB Loke, GA. Llano. Poster presented at 8th
International Congress of Plant Pathology (ICPP2003), Christchurch,
New Zealand, 2 - 7 February 2003. Vol 2: 133.
Control of bud rot in oil palm, Elaeis guineensis, using resistance inducers
E Alvarez, GA Llano, MC Feris, ML Hernández and S.M. Rodríguez. Poster presented at
8th International Congress of Plant Pathology (ICPP2003), Christchurch, New Zealand,
2 - 7 February 2003. Vol 2: 179.
Characterization and classification of phytoplasmas associated with oil palm (Elaeis
guineensis). E Alvarez and JL Claroz. Poster presented at 8th International Congress of
Plant Pathology (ICPP2003), Christchurch, New Zealand, 2 - 7 February 2003. Vol 2:
284.
Output 7-50
2003 Annual Report
Control of powdery mildew in roses by applying lixiviated plantain rachis compost. E
Alvarez, C Grajales, J Villegas and JB Loke. Poster presented at 8th International
Congress of Plant Pathology (ICPP2003), Christchurch, New Zealand, 2 - 7 February
2003. Vol 2: 272.
Detección de marcadores microsatélites asociados con la resistencia al Añublo
Bacterial de la yuca (Manihot esculenta Crantz) en Colombia. PX Hurtado, E Alvarez, M
Fregene and GA Llano. Presented at XXIV Congress of ASCOLFI. June 25 – 27, 2003.
P. 25.
Caracterización genética y patogénica de Colletotrichum spp. agente causal de la
antracnosis en guanábana (Anona muricata) en el Valle del Cauca. CA Ospina and E
Alvarez. Presented at XXIV Congress of ASCOLFI. June 25 – 27, 2003. P. 6
Detecting the phytoplasm-frogskin disease association in cassava (Manihot esculenta
Crantz) in Colombia. E Alvarez, JF Mejía, JB Loke, L Hernández, and GA Llano.
Phytopathology 93 (6): S4. Poster presented at APS annual meeting August 9 -13.
Charlotte, NC, USA.
Detecting SSR markers associated with resistance to cassava bacterial blight (CBB) in
Colombia. E Alvarez, PX Hurtado, M Fregene. and GA Llano. Phytopathology 93 (6):
S4. Poster presented at APS annual meeting August 9 -13. Charlotte, NC, USA.
Congress.
8th International Congress of Plant Pathology (ICPP2003), Christchurch, New Zealand,
2 - 7 February 2003. Vol 2: 272.
XXIV Congreso Nacional de Fitopatología, held by ASCOLFI, Armenia. June 25-27.
Enfermedades de yuca y perspectivas de manejo sostenible. E. Alvarez. Conference
presented at Congreso Nacional de Fitopatología, held by ASCOLFI, Armenia. June 25,
2003
APS annual meeting. August 9 - 13. Charlotte, NC, USA.
Publications
Alvarez, E., Mejia, J.F., and Valle, T.L. 2003. Molecular and pathogenicity
characterization of Sphaceloma manihoticola isolates from south- central Brazil. Plant
Dis.87: 1322-1328.
Alvarez, E., Loke, J.B, Rivera, S. y Llano, G.A. Genética de la resistencia a pudrición
causada por Phytophthora tropicalis en dos poblaciones segregantes de yuca (Manihot
esculenta Crantz). Revista Fitopatología Colombiana Vol 26 (2): 61 - 66. Cali,
Colombia.
Llano, G.A., Alvarez, E., Muñoz, J. E., Fregene, M. Identificación de genes análogos de
resistencia a enfermedades en yuca (Manihot esculenta Crantz), y su relación con la
Output 7-51
2003 Annual Report
resistencia a tres especies de Phytophthora. Acta Agronómica. Palmira, Colombia. In
Press.
Llano, G.A., Alvarez, E., Assessment of integrated management practices of cassava
root rots, by participatory research with indigenous from Mitú, colombian noreast
amazon. Agren. Agricultural research and extension network. Submited.
Hurtado, P.X., Alvarez, E., Fregene, M. Búsqueda de genes análogos de resistencia
asociados con la resistencia al añublo bacterial de la yuca. Revista Fitopatología
Colombiana. Submited.
Hurtado, P.X., Alvarez, E., Fregene, M., Llano, G.A. Detección de marcadores
microsatélites asociados con la resistencia a Xanthomonas axonopodis pv. manihotis
en una familia de yuca (BC1). Revista Fitopatología Colombiana. Submited.
Two brochures:
The following informative brochures were designed to help train farmers how to
prevent and manage superelongation disease (SED) and bacteriosis (CBB), two
diseases that seriously affect cassava production.
Alvarez, E., Llano, G.A. Cassava Bacterial Blight.
Alvarez, E., Mejia, J.F. Superelongation Disease.
Output 7-52
2003 Annual Report
Concept notes and projects developed.
Identification of a phytoplasm associated to Lethal Wilt in oil palm. Presented to
Palmar del Oriente, Palmas de Casanare and Palmeras Santana. US$29.500. Aproved.
Caracterización molecular de un fitoplasma que está afectando el cultivo del lulo en
Colombia. US$4.000 Aproved.
Detección y manejo de microorganismos asociados con Cuero de Sapo en yuca
Project cost: US$104.500. Funds requested: US$45.000. Aproved.
The mechanisms behind disease resistance in cereals. Presented With Royal
Veterinary and Agricultural University (KVL, Denmark) and CORPOICA (Colombia).
Presented to Cereal Genome. Project cost: US$260.000 Funds requested: US$200.000.
Generación de nueva tecnología para el manejo de Mildeo Velloso en rosa y la
caracterización genética de su agente causal, Peronospora sparsa, en Colombia
Presented to Asocolflores/Ceniflores. Project cost: US$238.000 Funds requested:
US$119.000.
Desarrollo de medidas de manejo del Moko (Ralstonia solanacearum), para evitar la
destrucción de las plantaciones de plátano en Colombia. Presented with ICA and
CORPOICA to ASOHOFRUCOL (Colombia). Project cost: US$188.200 Funds requested:
US$117.200.
Medidas Tendientes a Evitar que el Moko (Ralstonia solanacearum) Destruya las
Plantaciones de Plátano y Banano en Colombia. Presented to Ministerio de Agricultura
y Desarrollo de Colombia. Funds requested: US$71.428.
Confronting a Cassava Root Rot Disease Epidemic in West Africa: Survey,
Management, anf Identification of Host Plant Resistance
Thesis presented
Germán A. Llano. 2003. Identificación de genes análogos de resistencia a
enfermedades en yuca, Manihot esculenta Crantz y su relación con la
resistencia a tres especies de Phytophthora. M.Sc. thesis. Palmira, Colombia:
Universidad Nacional de Colombia. 119 pp.
Theses for Master of Sciences degree in progress
John B. Loke. Análisis genético de la resistencia de yuca (Manihot esculenta Crantz) a
Phytophthora tropicalis, causante de pudrición radical. Universidad Nacional de
Colombia—Palmira. For a Master of Plant Breeding.
Paula X. Hurtado. Evaluación de marcadores microsatélites y genes análogos,
asociados a la resistencia de yuca a Xanthomonas axonopodis pv. manihotis.
Output 7-53
2003 Annual Report
For a Master of Biology with emphasis in Plant Molecular Biology. Universidad
de los Andes—Bogotá
Personnel
Staff
Elizabeth Alvarez
John B. Loke
Herney Rengifo
Martin Fregene
Lina María Tabares
Luis Hernández
Germán A. Llano
Juan Fernando Mejía
Zulma Zamora
Students and Technicians
Universidad de los Andes—Bogotá:
Paula Ximena Hurtado
Universidad del Valle
Sonia Ossa
Universidad Católica de Manizales:
Samira Moreno
Técnico Agroforestal, Mitú, Vaupés:
Gabriel Paiva
Universidad
de
los
Llanos—
Villavicencio:
Sandra Milena Rodríguez
INIVIT (Cuba):
Mariluz Folgueiras
Colegio Bolívar:
Cristina Londoño
Universidad de Caldas— Manizales:
Alejandro Corredor
Linkages with CIAT’s Partner Institutions
CLAYUCA
COLCIENCIAS
Instituto de Investigaciones de Viandas Tropicales (INIVIT, Cuba)
IPRA (based at CIAT, Colombia)
Secretaría de Agricultura del Vaupés (at Mitú)
UMATAs (Mitú, La Tebaida, and Montenegro)
Universidad Nacional de Colombia—Palmira (Valle del Cauca, Colombia)
Donors
Hacienda San José (Palmira)
ICA Palmira
Ministerio de Agricultura y Desarrollo Rural
PRONATTA
Collaborators
Local
CENICAÑA (Drs J. Victoria and F. Angel)
CLAYUCA (based at CIAT, Dr B. Ospina)
Output 7-54
2003 Annual Report
CORPOICA—Bogotá (Dr Jairo Osorio)
CORPOICA “La Libertad” (Villavicencio, Dr. A. Tapiero)
CORPOICA—Palmira (Dr G. Aya)
Corporación BIOTEC (Dr J. Cabra; B. Villegas)
Corporación para el Desarrollo Sostenible del Norte y Oriente Amazónico (CDA,
Vaupés, Dr E. Polo and R. Peña)
CRIVA—Consejo Regional Indígena del Vaupés (Mitú)
ICA—Quindío and Valle (Drs E. Vargas, F. Varón, C. A. Montoya and C. Huertas)
Magro S.A. (Andrea Villegas)
Secretaría de Agricultura del Vaupés (at Mitú, Dr G. Arbeláez)
Special (La Tebaida, Mr S. González)
UMATAs (Drs O. Holguín, L. Muñoz, and W. Ospina)
Unillanos—Villavicencio
Universidad Católica de Manizales
Universidad de Caldas—Manizales
Universidad de los Andes—Bogotá
Universidad del Valle (Cali)
Universidad Nacional de Colombia—Palmira
International
Cooperative Research Center for Tropical Plant Protection, University of Queensland,
Brisbane, Australia (A. Drenth).
Iowa State University (Dr T. Harrington, Dr. T. Pepper)
Kansas State University (Drs S. H. Hulbert and Robert Zeigler)
The Royal Veterinary and Agricultural University (Copenhagen, Denmark, Dr
D.Collinge)
Output 7-55
2003 Annual Report
OUPUT 8
Development and use of biotechnology tools for cassava improvement
Activity 8.1. Molecular Marker-Assisted Breeding for Resistance to the
Cassava Mosaic Disease in Latin American Cassava Gene pools.
Collaborators: J. Marín, C. Ospina, E. Barrera, L. Santos, D. Moretta, Y.
Moreno, M. Fregene (CIAT)
Funding: The Rockefeller Foundation
Important Outputs
1) A process for conducting MAS for CMD resistance breeding, from genetic
crosses, embryo rescue, molecular analysis, to data collection has been
implemented at CIAT. More than 1,100 genotypes were successfully
processed this year
2) A scheme to increase three times the number of genotypes that can be
processed is also being implemented.
Rationale
Molecular marker-assisted selection (MAS) for CMD resistance at CIAT is both a
pre-emptive measure, should in case the disease is accidentally introduced in
Latin America, and a dynamic measure, to enable a true evaluation of the value
of CIAT improved germplasm in India and Africa. MAS using the single
dominant gene CMD2 as source has completed the first year at CIAT. A total of
2315 seeds harvested from more than 3000 controlled crosses between CMD
resistant parents introduced from IITA and elite parents of the 5 cassava gene
pools by agro-ecology or backcross derivatives of M. esculenta sub spp
flabellifolia resistant to the green mite. More than 1,100 genotypes were
germinated as embryo axes and multiplied for molecular analysis with the SSR
marker NS158 closely linked to CMD2. Establishment of breeding populations
in vitro is to aid shipment to collaborators in Africa and India. CMD resistant
genotypes, as revealed by MAS, will be sent to the green house for hardening
and also shipped to partners in India and Africa. We describe here results of
the first year of MAS at CIAT.
Methodology
Sexual seeds of crosses between CMD resistant parents and elite parents of
cassava gene pools were germinated from embryo axes (see Activity 22) and
molecular analysis performed on the plantlets. DNA isolation was using a rapid
mini prep method developed for rice (Nobuyuki et al 2000) and 2 young leaves
from the in vitro plantlets. The leaves were placed in an 1.5 ml eppendorf tube
and 200ul of TE buffer (10 mM tris-Cl, 1 mM EDTA; pH: 8.0 ) added. The
leaves were then squashed in the buffer using a small pestle and incubated in a
water bath at 100 ˚C for 15 minutes. Next, 800ul of TE buffer was added and
the tube inverted gently several times to mix the content followed by
centrifugation in a table-top centrifuge at 14,000rpm for 10 minutes. The
supernatant contains about 10 to 20ng/ul and was used directly in the PCR
amplification reaction. The supernatant was transferred to eppendorf 96-well
plates (Costar) using 8 tip multi-channel pipette for easy dispensation into 96-
Output 8-1
2003 Annual Report
well PCR plates and for long term storage. DNA obtained is sufficient for 100
reactions and can be held in the Costar plates for 2 months at –20˚C without
any degradation. PCR and PAGE gel analysis of the PCR product is as
described by Mba et al. (2001). Gel image from the SSR analysis is entered into
an excel sheet containing other information, such as pedigree, phenotypic
evaluation, number of plants available and where.
Results
More than 1,100 plants representing the first breeding population for CMD
resistance at CIAT were analyzed using the NS158 marker associated with CMD
resistance. Results of the molecular analysis are shown in Table 8.1. An
information management system to handle the MAS data and to make it easily
accessible was developed in Micro soft Excel. The versatility of Excel
spreadsheets make it the appropriate software to handle the diverse information
generated by MAS. We also reviewed the process of MAS as conducted by
CIAT´s cassava molecular lab with respect to time, labor and costs.
The time required to pick leaves from in vitro plantlets, extract DNA and
completely fill a 96-well plate is approximately 9 hours. To set up a 96-well
PCR reaction and complete the temperature cycling, in this case for SSR marker
NS158 associated with CMD2, is 4 hours. Running the amplification product
on a 6% acrylamide gel and silver stain also requires 4 hours. In total it takes
17 hours or 2 working days for a single person to complete DNA isolation and
marker analysis for 96 genotypes.
The cassava molecular marker lab currently has two persons working on MAS
for CMD and together they can process 192 genotypes in 2 days or 480
genotypes per week or over 24,000 samples in a year. We are working on
improving this by doing the grinding and DNA isolation in 96-well plates.
Current costs of a single SSR marker data point analysis for cassava at CIAT is
US$0.30, processing 24,000 samples in a year requires a budget of US$7,200.
Conclusions and perspectives
MAS for CMD resistance breeding have been initiated at CIAT. More than 1,100
genotypes have been processed this year and it is expected that three times that
number will be processed that year as the entire system from crosses to embryo
rescue to molecular analysis is made more efficient. Future perspectives
include development of a 96-well method for grinding leaf tissue and DNA
isolation to eliminate the need for time-consuming transfers from eppendorf
tubes to 96-well plates.
Project IP3: improving cassava for the developing world
Output 8-2
Table 8.1. The cassava MAS data management system in Microsoft excel, the
spreadsheet shows part of a 96-well plate molecular marker (NS158)
analysis, and permits easy access to molecular and other types of data.
MATERIALES CR
Number
Code
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
Output 8-3
CR52A-30
CR52A-31
CR52A-32
CR52A-33
CR52A-34
CR52A-35
CR52A-36
CR52A-37
CR52A-40
CR42-1
CR42-2
CR42-3
CR42-4
CR42-5
CR42-6
CR42-7
CR42-9
CR9A-153
CR9A-154
CR9A-155
CR9A-156
CR20A-1
CR20A-2
CR20A-3
CR20A-4
CR20A-5
CR20A-6
CR14B-1
CR14B-3
CR14B-4
CR14B-5
CR14B-6
CR14B-7
CR14B-8
CR14B-9
CR14B-10
CR14B-11
CR14B-12
CR14B-13
CR14B-14
CR14B-16
CR52A-41
CR52A-42
CR52A-43
CR9B-5
CR25-2
CR23-1
CR23-2
CR23-3
CR23-4
CR23-5
CR23-6
CR23-7
Mother
Father
C-243
SM1219-9
C-243
SM1219-9
C-243
SM1219-9
C-243
SM1219-9
C-243
SM1219-9
C-243
SM1219-9
C-243
SM1219-9
C-243
SM1219-9
C-243
SM1219-9
MCOL 2206
C-18
MCOL 2206
C-18
MCOL 2206
C-18
MCOL 2206
C-18
MCOL 2206
C-18
MCOL 2206
C-18
MCOL 2206
C-18
MCOL 2206
C-18
C-4
MTAI 8
C-4
MTAI 8
C-4
MTAI 8
C-4
MTAI 8
CM3306-4
C-33
CM3306-4
C-33
CM3306-4
C-33
CM3306-4
C-33
CM3306-4
C-33
CM3306-4
C-33
C-4
CM523-7
C-4
CM523-7
C-4
CM523-7
C-4
CM523-7
C-4
CM523-7
C-4
CM523-7
C-4
CM523-7
C-4
CM523-7
C-4
CM523-7
C-4
CM523-7
C-4
CM523-7
C-4
CM523-7
C-4
CM523-7
C-4
CM523-7
C-243
SM1219-9
C-243
SM1219-9
C-243
SM1219-9
MTAI 8
C-4
CM7951-5
C-33
CM7951-5
C-4
CM7951-5
C-4
CM7951-5
C-4
CM7951-5
C-4
CM7951-5
C-4
CM7951-5
C-4
CM7951-5
C-4
TC Media
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
4E
In Vitro
Jars
Tubes
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
2**
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
PCR
SI
SI
NO
SI
SI
SI
SI
SI
SI
SI
SI
SI
SI
SI
SI
SI
SI
SI
SI
SI
SI
SI
SI
NO
SI
SI
SI
SI
SI
SI
SI
SI
NO
SI
SI
SI
NO
SI
SI
SI
SI
SI
NO
SI
SI
NO
SI
SI
SI
SI
SI
SI
SI
MAS
Score
Green house plants
S
R
R
S
R
R
R
R
R
R
R
R
R
S
R
R
S
R
S
S
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
S
R
S
R
S
S
R
R
2003 Annual Report
References
Nobuyuki I. Bautista N.S., Yamada T., Kamijima O, Ishi T. 2000. Ultra simple
DNA extraction method for marker-assisted selection using microsatellite
markers in rice. Plant Molecular Biology Reporter 18:1-6
Mba REC, Stephenson P, Edwards K, Melzer S, Mkumbira J, Gullberg U, Apel
K, Gale M, Tohme J, Fregene MA (2001) Simple Sequence Repeat (SSR)
markers survey of the cassava (Manihot esculenta Crantz) genome:
towards an SSR-based molecular genetic map of cassava. Theor Appl
Genet: 21-31
Activity 8.2. Molecular Marker-Assisted and Farmer Participatory
Improvement of Cassava Germplasm for Farmer/Market Preferred Traits in
Tanzania
Collaborators: Alois Kullaya (ARI Mikocheni, Tanzania), Heneriko Kulembeka
(ARI, Ukiriguru , Tanzania), Ms Kiddo Mtunda (ARI, Kibaha,
Tanzania), Morag Ferguson (IITA-ICRISAT, Nairobi, Tanzania),
Jaime Marin, Cesar Ospina, Edgar Barrera, Andrew Jarvis,
Nelson Morante, Hernan Ceballos, Martin Fregene (CIAT).
Funding: The Rockeller Foundation
Important Outputs
1) A molecular marker-assisted, farmer-participatory, breeding project to
develop improved small cassava farmers germplasm with pest and
disease resistance have been initiated in collaboration with the National
program in Tanzania and IITA.
2) The development of parents that combine resistance to the cassava
mosaic disease and green mites for generating broad-based breeding
populations and subsequent selection by molecular markers and farmer
participatory breeding.
Rationale
Tanzania is the fourth largest producer of cassava in Africa with average yields of
about 8 t/ha (FAO, 2001). This is well below the continent’s average of 10tons/ha
and the average yield of 14 tons/ha of Africa’s (and the world’s) largest producer,
Nigeria. The low yield is caused by many factors including susceptibility of
commonly grown varieties to major diseases and pests such as cassava mosaic
diseases, caused principally by the East African Cassava Mosaic Virus (EACMV),
its Ugandan variant (UgV), and the African Cassava Mosaic virus (ACMV), cassava
brown steak disease (CBSD), cassava bacterial blight (CBB), cassava green mite
(CGM), cassava mealy bug (CM) and nematodes. Previous research at Kibaha and
Naliendele reported as high as 55% loss in the local cultivar “Albert” (Mtunda 2003
pers comm). A recent survey in Tanga has revealed crop losses of up to 74%
(Muhanna and Mtunda, 2002). In severely affected areas, entire fields may be
destroyed.
A farmer participatory, molecular marker-assisted, decentralized, breeding
scheme has recently been approved for funding by the Rockefeller Foundation
to speed up the process of improving local cassava germplasm for resistance to
pests and diseases in Tanzania. The proposed breeding project will take farmer
preferred germplasm by agro-ecology and cross them to improved introductions
Project IP3: improving cassava for the developing world
Output 8-4
that have resistance to Cassava Mosaic Disease (CMD), Cassva Green Mite
(CGM), and Cassava Bacterial Blight (CBB). Given the fairly large number of
parents that will be used, molecular markers associated with pest and disease
resistance will be employed to reduce, in a logical manner, the number of
progeny to a manageable number. The progeny selected by MAS will be
evaluated in a single season in the corresponding agro-ecology and then
evaluated over two cycles in collaboration with end-users (rural communities
and cassava processors). The project will be carried out in a total of six years
divided into 2 three-year phases. A principal objective of the project is the
development of capacity for participatory plant breeding and marker-assisted
breeding. This would be achieved by training 2 national program breeders at
the MSc. and PhD level, and through 2 training workshops on participatory
plant breeding and marker-assisted breeding.
Methodology
Activities of the first year of the RF Tanzanian project includes the collection
and evaluation of local germplasm and the introduction of improved progenitors
for use as parents in the breeding project. Improved progenitors were designed
to have resistance to CMD, CBB, and CGM as well as molecular markers
associated with these genes. The improved progenitors were developed as
described below. The RF funded project “Molecular Mapping of Genes
Conferring Resistance to the Cassava Mosaic Disease (CMD) in African Cassava
Germplasm” has led to the identification of 3 SSR and 2 RAPD markers tightly
linked to a novel source of CMD resistance controlled by a single dominant gene
designated CMD2 (Akano et. al. 2002; Moreno and Fregene unpublished data).
The CMD2 source shows high levels of resistance against a wide spectrum of
strains of the virus in sub Saharan Africa, including the aggressive Ugandan
variant (UgV), ACMV and EACMV (Akano 2002, CIAT 2001). Excellent
resistance to CGM have also been observed in 4 F1 inter-specific hybrid families
obtained by crossing the cassava clones CG487-2, CG501-16, MCol2215, and
CM2766-5 to a genotype of M. esculenta sub spp flabellifolia. (Belloti and
Fregene 2002, unpublished data; CIAT 2002). Bulk segregant analysis (BSA)
was quickly used to identify several SSR markers associated with CGM
resistance in the MCol2215 cross (CIAT 2002). The inter-specific hybrids of M.
esculenta sub spp flabellifolia that carry the novel CGM resistance were crossed
extensively to elite parents of cassava genepools from the 5 agro-ecologies, a
number of these parents have the SG107-35 source of CBB resistance. These
first back cross derivatives were then crossed to CMD 2 donor parents to obtain
more than 600 BC2 progenies.
Progeny from the above cross were established from embryo axes of mature
sexual seeds, multiplied and kept in vitro. Two in vitro plants were used for
marker evaluation to identify progeny CMD, CBB and CGM resistance. Another
5 in vitro plants were sent to the green house for phenotypic evaluation of yield
and yield components. About 200 genotypes that combine resistance to CMD,
CBB and CGM and high productivity are being prepared for shipment to
Tanzania, at least 20 plants per genotype will be shipped. An import permit
has been issued by the Tanzanian phyto-quarantine authorities for the
shipment of this germplasm.
Results
Progenies of inter-specific hybrids crossed to parents of cassava gene pools
adapted to the sub-humid lowland, acid savannah, mid-altitude and semi-arid
Output 8-5
2003 Annual Report
agro-ecologies, with good resistance to CBB were crossed to CMD2 donor
parents to obtain more than 600 BC2 progenies. Table 8.2 summarizes the BC2
families obtained and the number of plants per genotype in vitro. In addition,
parents of the above progenitors were indexed for the frog skin disease (FSD)
and other commonly found diseases in South America as well as checked for
pests. Embryo rescue and multiplication of the BC2 families to obtain 10 plants
per genotype was done as described earlier (CIAT 2002). About 5 in vitro plants
per genotype was used in marker analysis using markers associated with CMD
and CGM according to methods described earlier (CIAT 2002). A special format
in Microsoft excel was developed to display results of the molecular markerassisted evaluation of the BC2progenies as described in the MAS for CMD at
CIAT activity.
Plants selected by molecular marker analysis were further multiplied to obtain
more than 20 plants per genotype and will be shipped to Tanzania once the
import permit have been issued and received at CIAT, at least 20 plants per
genotypes will be shipped. On arrival in Tanzania, 18 plants of all genotypes
will be hardened in the screen house and 6 plants each sent to the 3 different
target agro-ecologies for field establishment and evaluation for use as improved
parents. The remainder 2 plants per genotype will be multiplied and kept in
vitro as backup. At least 30 genotypes of the 200 improved introductions will
be selected for crosses to the20 selected local varieties. Selection parameters
will include harvest index, pest and disease resistance and root quality.
The parents will be crossed in all possible combinations using a polycross
design. The crossing blocks, with 40 plants each per local variety and 20 plants
each of the improved introductions, will be set up at two sites, namely SRIKibaha, Eastern zone, and Maruku, in the lake zone, where cassava flowers
profusely, to maximize the possibilities of getting seeds from all genotypes.
Hand pollination will also be carried out to ensure certain combinations are
obtained. The local land races and introductions will be used as female parents
to achieve a wide base of cytoplasm, therefore open pollinated and controlled
pollinated sexual seeds will be harvested from both. Between 10,000 and
20,000 seeds are expected from crosses for each agro-ecology
Conclusions and perspectives
A molecular marker-assisted breeding project to develop improved germplasm
for small cassava farmers have been initiated. The first year will be spent in the
development of parents and generating broad-based breeding populations for
subsequent selection by molecular markers and in collaboration with farmers.
The project is a first and experience gained is expected to guide the application
of molecular markers in cassava breeding.
Project IP3: improving cassava for the developing world
Output 8-6
Table 8.2. List of BC2 families developed for the introgression of resistance to
CGM from wild progenitors of cassava into CMD resistant genotypes.
Crosses were made to cassava parents adapted to the semi-arid low land
tropics (ZO1) and Acid savannahs (ZO2).
No. of
Family
Mother
Father
Genotypes Zone
CW 74- 1 CM 2177- 2 CW 65- 77
1
Z02
CW 75- 1 CM 3306- 4 CW 66- 60
7
Z01
CW 76- 1 CM 3306- 4 CW 68- 3
9
Z01
CW 77- 1 CM 7951- 5 CW 65- 77
5
Z02
CW 78- 1 CM 7951- 5 CW 66- 19
4
Z02
CW 79- 1 CM 7951- 5 CW 66- 62
1
Z02
CW 80- 1 CM 7951- 5 CW 67- 42
5
Z02
CW 81- 1 CM 7951- 5 CW 67- 98
3
Z02
CW 213- 1 SM 805- 15 CW 67- 39
1
Z01
CW 214- 1 SM 805- 15 CW 67- 87
11
Z01
CW 215- 1 SM 909- 25 CW 66- 60
8
Z02
CW 217- 1 SM 1219- 9 CW 65- 77
17
Z02
CW 218- 1 SM 1219- 9 CW 66- 73
12
Z02
CW 220- 9 SM 1219- 9 CW 67- 123
5
Z02
CW 223- 1 SM 1460- 1 CW 66- 19
11
Z02
CW 224- 1 SM 1460- 1 CW 66- 60
16
Z02
CW 225- 1 SM 1460- 1 CW 66- 62
30
Z02
CW 226- 1 SM 1460- 1 CW 66- 73
15
Z02
CW 227- 1 SM 1460- 1 CW 68- 3
3
Z02
CW 229- 1 SM 1511- 6 CW 67- 87
26
Z01
CW 230- 1 SM 1565- 15 CW 66- 19
4
Z02
CW 231- 1 SM 1565- 15 CW 66- 60
24
Z02
CW 232- 1 SM 1665- 2 CW 66- 19
23
Z01
CW 233- 1 SM 1665- 2 CW 66- 60
32
Z01
CW 235- 1 SM 1665- 2 CW 67- 87
117
Z01
CW 236- 1 SM 1669- 5 CW 66- 19
33
Z01
CW 237- 1 SM 1669- 5 CW 66- 60
8
Z01
CW 238- 1 SM 1669- 5 CW 66- 62
1
Z01
CW 239- 1 SM 1669- 5 CW 66- 73
5
Z01
CW 240- 1 SM 1669- 5 CW 66- 74
28
Z01
CW 241- 1 SM 1669- 5 CW 67- 123
4
Z01
CW 242- 1 SM 1669- 7 CW 67- 87
9
Z01
CW 243- 1 SM 1741- 1 CW 66- 19
6
Z02
CW 244- 1 SM 1741- 1 CW 66- 60
15
Z02
CW 245- 1 SM 1741- 1 CW 66- 62
1
Z02
CW 246- 1 SM 1741- 1 CW 67- 91
9
Z02
CW 247- 1 SM 1778- 45 CW 66- 19
3
Z02
CW 248- 1 SM 1778- 45 CW 67- 45
2
Z02
CW 257- 1 MTAI 8 CW 65- 77
25
Z01
CW 258- 1 MTAI 8 CW 66- 60
26
Z01
CW 259- 1 MTAI 8 CW 66- 73
46
Z01
CW 260- 1 MTAI 8 CW 66- 74
6
Z01
CW 261- 1 MTAI 8 CW 67- 123
4
Z01
Total
621
Output 8-7
2003 Annual Report
References
Akano A., Barrera E., Mba C., Dixon A.G.O., Fregene M. (2002) Genetic
Mapping of a Dominant Gene Conferring Resistance to the Cassava
Mosaic Disease (CMD). Theoretical and Applied Genetics (published
online May 8, 2002).
CIAT (2002) Annual Report Project SB2, Assessing and Utilizing
Agrobiodiversity through Biotechnology, CIAT, Cali, Colombia, pp 239241.
FAO (2001) FAO production yearbook for 2000, Rome, Italy: FAO
Muhanna and Mtunda 2001 A report on a survey of cassava in Tanzania.
Ministry of Agriculture and food security, Tanzania.
Activity 8.3 Molecular Marker-Assisted Selection (MAS) for Breeding Early
Root Bulking in Cassava
Collaborators: N. Morante, J. Marin, M. Fregene (CIAT)
Funding: CIAT core funds
Important Output
1) The selection of parents, from an S1 progeny derived from the genotype
K150 of the cassava map population that has many favorable alleles for
earliness, for a scheme to improve early bulking via marker-assisted
selection (MAS)
Rationale
Early root yield is an important trait of cassava (Nweke et al. 1994), critical to
the crop’s role as food security crop in sub-Saharan Africa. Furthermore, as
globalization picks up speed, new opportunities have arisen for cassava as a
source of industrial raw material which requires a great deal of flexibility of
harvest, that is varieties that attain close to maximum yield at 8-10 months
after planting. Genetic mapping of early root bulking in a full-sib cassava
genetic map population was extended to a S1 family of 268 genotypes from a
progeny, K150, of the same map population. Several QTLs for foliage weight
and harvest index, 2 traits that strongly influence early root bulking, were
found in both studies. They include a major QTL for foliage weight, explaining
31% of phenotypic variance, and 3 QTLs for harvest index, explaining between
15 and 22% of phenotypic variance, the QTL for foliage showed a dominance
gene action while those for harvest index were additive (Okogbenin et al. 2003).
The mapping of major QTLs is an important step towards the development of
molecular markers for breeding of early yield, which like yield is a complex trait.
To identify parents for the development of breeding populations for early root
bulking, 25 of the progenies from the S1 population that had the highest yield at
7 months after planting in the QTL experiments last year were evaluated for a
second year in larger plot sizes and more replications. The best 5 genotypes
will be used as parents for making crosses to elite cassava parents for the
development of breeding populations for molecular marker-assisted selection
(MAS).
Project IP3: improving cassava for the developing world
Output 8-8
Methodology
An S1 family was developed from K150, a progeny of the cassava map
population having a large number of positive QTLs for foliage weight, harvest
index and number of roots, was developed in 1998 and evaluated 2001 in a
replicated trial (Okogbenin 2003). A second year evaluation was conducted in
2002 with the difference that harvest was conducted at 7 months after planting.
Twenty five of the highest yielding genotypes with harvest index more than 0.50
and strong canopy, foliage weight per plant greater than 1.5kg, were selected
and re-established for a third year evaluation at the CIAT station in Santander
de Quilichao a high stress environment. Previous work has shown than
selection for early root bulking in a low stress environment may not be
replicated in a high stress environment (Kawano et al, 2001). Experimental
design and evaluation of the above early bulking genotypes was as described for
the 2002 evaluation (Okogbenin 2003). Ten months after planting, the
genotypes were evaluated for root yield, foliage weight, harvest index, and dry
matter content. The best five genotypes from this experiment will be used as
parents in the development of populations for a molecular breeding scheme for
improved early bulking.
Results
Evaluation of the best 25 genotypes from last year’s evaluation for early yield
confirmed earlier observation that a strong canopy and high harvest index are
essential for early root yield (Okogbenin and Fregene 2003). The 3 genotypes
with the highest foliage weight are also those with the highest root yield (Table
8.3). Another observation is the very low standard deviation for harvest index
in the 25 genotypes, all but one genotype have harvest index of more than 0.6,
again revealing the strong correlation between harvest index and early yield,
selection of these genotypes last year was based only on early yield. The
genotypes selected as parents from this evaluation are AM150-241, AM150332, AM150-185, AM150-150, and AM150-309. Stakes from these materials
have planted in the crossing block for next year.
Conclusions and Perspectives
A scheme to improve early bulking via marker-assisted selection has been
initiated with the selection of parents for the development of breeding
populations. Future perspectives include evaluation of breeding populations
obtained next year with markers associated with QTLs for early bulking in the
S1 family from K150.
Output 8-9
2003 Annual Report
Table 8.3 Some statistics yield and yield components of the 25 most early
yielding genotype from the S1 family from K150
Genotype
AM150-10
AM150-34
AM150-43
AM150-56
AM150-70
AM150-83
AM150-96
AM150-100
AM150-104
AM150-111
AM150-114
AM150-119
AM150-124
AM150-137
AM150-143
AM150-154
AM150-165
AM150-179
AM150-182
AM150-185
AM150-206
AM150-241
AM150-290
AM150-309
AM150-332
Mother
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
CM 7857- 150
Maximum
Minimum
Average
Std. Deviation
Foilage/plant (g) Yield/plant (g) HI
1187.50
3850.00
325.00
1150.00
650.00
1650.00
1550.00
2625.00
1937.50
2750.00
687.50
2600.00
850.00
2750.00
1375.00
5500.00
775.00
1887.50
1191.67
1600.00
950.00
3950.00
1125.00
3750.00
518.75
1856.25
1125.00
2956.25
1237.50
3750.00
1700.00
1000.00
641.67
1875.00
508.33
2483.33
387.50
2087.50
3800.00
6737.50
625.00
1550.00
2350.00
8725.00
825.00
1440.00
1375.00
4275.00
4012.50
9550.00
4012.50
9550.00
325.00
1000.00
1268.42
3293.93
934.26
2228.53
0.76
0.78
0.72
0.63
0.59
0.79
0.76
0.80
0.71
0.57
0.81
0.77
0.78
0.72
0.75
0.37
0.75
0.83
0.84
0.64
0.71
0.79
0.64
0.76
0.70
0.84
0.37
0.72
0.10
% Dry matter Yield (t/ha)
34.11
38.50
27.19
11.50
33.16
16.50
31.53
26.25
32.08
27.50
30.24
26.00
31.78
27.50
32.90
55.00
29.94
18.90
32.40
16.00
33.72
39.50
36.27
37.50
34.63
18.60
34.44
29.60
32.84
37.50
28.03
10.00
36.76
18.80
32.20
24.80
32.02
20.90
36.89
67.40
30.37
15.50
32.55
87.30
31.35
14.40
33.95
42.80
36.00
95.50
36.89
95.50
27.19
10.00
32.69
32.95
2.46
22.29
References
Okogbenin E. and Fregene M (2001) Genetic Analysis and QTL Mapping of
Early Bulking in an F1 Segregating Population from Non-inbred Parents
in Cassava (Manihot esculenta Crantz) Theor Appl Genet 106:58-66
Okogbenin E. (2003). QTL mapping of early root yield, morphology and root
quality in cassava. Ph.D dissertation submitted to the University of Ibadan,
Ibadan, Nigeria.
Nweke FI, Dixon AGO, Asiedu R, Folayan SA (1994). Cassava varietal needs of
farmers and potential for production growth in Africa. COSCA working
paper 10.
Project IP3: improving cassava for the developing world
Output 8-10
Activity 8.4. Genetic Mapping of Beta-Carotene Content
Collaborators: Nelson Morante, Teresa Sanchez, Alba Lucia Chavez, Jaime
Marin, Cesar Ospina, Hernan Ceballos, Martin Fregene (CIAT)
Important Outputs
1) An SSR marker, NS251 that explains 30% of phenotypic variation for
beta-carotene content has been identified in cassava.
2) A search for more markers more tightly associated with beta-carotene in
using known biosynthetic genes of beta-carotene is ongoing.
Rationale
A project to fortify cassava varieties grown by rural communities with higher
levels of beta-carotene has been initiated as a way of combating deficiency of
this key micronutrient in areas where cassava is a major staple. The
experimental approach on increasing cassava beta-carotene content includes
conventional breeding and genetic transformation. The discovery of a wide
segregation pattern of root color in two S1 families from the Colombian land race
MCol72 (AM273) and the Thai variety MTAI8 (AM320), led to the
commencement of molecular genetic analysis of beta-carotene content in
cassava. The cost-effectiveness of breeding for high beta-carotene content in
cassava can be considerably enhanced if the mode of inheritance and the
number of genes involved are known. Regions of the genome associated with
beta-carotene content can also be examined for the presence of known genes
involved in the biosynthesis of beta-carotene content can. This information can
be used for functional diversity analysis of natural variation of beta carotene
content for a more rational exploitation of naturally occurring variability. We
describe here results of bulk segregant analysis and the identification of two
regions of the cassava genome associated with beta-carotene content.
Methodology
Ten white colored and 10 orange/pink colored genotypes were selected from
each of the S1 family to serve as individuals for the high and low beta-carotene
bulks. For DNA isolation, 1-2g of young leaves was harvested from genotypes of
both bulks for each family and from all 38 genotypes of family AM273 and 102
genotypes from family AM320. The leaves were dried for 24h in an oven at 48oC
and then ground into a fine powder using a power drill and washed sand. DNA
was isolated from 200mg using a mini prep version of the Dellaporta (1983)
protocol. The bulks of white and orange/pink roots were then created per
family to give a total of 4 bulks. DNA from the bulks and the parents were then
genotyped with the 650 available cassava SSR markers according to methods
described by Mba et al (2001). Markers polymorphic in the bulks were
employed to analyze individuals of the bulks and, where the polymorphism
remained consistent in the individuals as with the bulks, the markers were
analyzed in the entire family. Association with orange/pink color was
determined by a simple linear regression of phenotypic data on marker
genotype marker class means (single point analysis) using the Microsoft Excel.
The amount of phenotypic variance explained by each marker l was obtained
from the R2 value.
Output 8-11
2003 Annual Report
Results
A total of 6 markers, namely NS189, NS980, SSRY240, SSRY251, SSRY9, and
SSRY63 were polymorphic in the bulks from the S1 family AM320 derived from
MTAI8. Six markers, NS980, SSRY240, SSRY9, SSRY251, common to both
families, and 2 additional markers, SSRY192 and SSRY54 unique to AM273,
were found polymorphic in the bulks. On analyzing individuals of the bulks,
polymorphism was consistent in the individuals for markers SSRY251, NS980,
SSRY240 in both families while the remaining markers did not show a clear-cut
pattern of polymorphism between the individuals of the bulks. All the three
markers are located on linkage group D of the molecular genetic map of
cassava. The 3 polymorphic markers were analyzed in all individuals of both S1
BW
BY
3
7
8
10
12
13
14
15
16
17
18
19
20
21
22
23
25
26
27
29
30
31
33
35
36
37
38
39
41
42
43
44
45
47
48
49
51
Color
TAI8
AM320
4
2
8
4
8
5
4
2
2
2
8
5
8
2
5
2
4
5
1
1
2
8
4
8
8
4
8
1
5
2
4
4
1
8
3
3
3
4
3
8
Fig 8.1. Silver stained polyacrylamide gel of PCR amplification of individuals
from Family AM320 with SSR marker NS251. The color code is as follows: 8 =
orange/pink colored roots, 1-2 = white colored roots, 4-5 = cream colored roots
families and results reveal NS251 has the strongest association with betacarotene content in both families (Fig8.1). This marker explains 30%
of phenotypic variance for beta-carotene. From the segregation of SSR marker
NS251 and scorings of color, known to be highly correlated with beta-carotene
content (r>0.8), of some individuals in the family AM320 in figure 1, it can be
observed that the most intense color, a score of 8, is associated with
homozygosity of the smaller sized allele of NS251 in this family. The same was
pattern was also observed in family AM273.
Conclusions and Perspectives
An SSR marker, NS251 that explains 30% of phenotypic variation for betacarotene content has been identified in cassava. A search for markers more
tightly associated with beta-carotene in the linkage group D region of the
genetic map of cassava that may have been missed in the screening of the bulks
due to a lack of polymorphisms is ongoing. A modified BSA method using
recombinants found with marker SSRY251 and several marker systems,
including RAPDs, AFLPs, and known biosynthetic genes of beta-carotene is
ongoing.
References
Dellaporta SL, Wood J, Hicks JR (1983) A plant DNA minipreparation: version
II. Plant Mol Biol Rep 1: 19-21
Mba REC, Stephenson P, Edwards K, Melzer S, Mkumbira J, Gullberg U, Apel
K, Gale M, Tohme J, Fregene MA (2001) Simple Sequence Repeat (SSR)
markers survey of the cassava (Manihot esculenta Crantz) genome:
Project IP3: improving cassava for the developing world
Output 8-12
towards an SSR-based molecular genetic map of cassava. Theor Appl
Genet: 21-31
Activity 8.5. Genetic Mapping of Dry Matter Content (DMC)
Collaborators: Henry Ojulong, Jaime Marin, Nelson Morante, Juan Carlos
Perez, Hernan Ceballos, Martin Fregene (CIAT)
Funding: Rockefelller Foundation
Important Output
1) The discovery of 2 markers, SSRY160 and SSRY150 that explain about 30% and 18%
respectively of phenotypic variance for DMC.
Rationale
Few key traits in cassava hold potential for increasing cost-effectiveness via
molecular marker assisted selection (MAS) compared to root dry matter content
(DMC). It is measured at the end of the growth cycle and affected by the time of
evaluation, DMC is generally high before the onset of the rains but drops after
the rains begin as the plant mobilizes starch from the roots for re-growth of
leaves (Byrne 1984). A tool to evaluate early and accurately DMC can increase
cost-effectiveness of breeding from DMC, via an elimination of a large number of
genotypes leaving the breeder more time to concentrate on the evaluation of a
reduced number of plants. Every plant with low DMC transplanted to the field
uses up space, resources and time (at evaluation) that considerably lowers
breeding efficiency.
The entry point for developing markers associated with DMC was three diallel
experiments that were began in 2000. The diallels, made up of 90 families, is
an ideal experiment to identify genes controlling DMC that are useful in many
genetic backgrounds. In May 2002, a final evaluation of the three diallel
experiments was done. Estimates of general combining ability (GCA) and
specific combining ability (SCA) for many traits of agronomic interest was
calculated, an emphasis was placed on DMC. Based on general combining
ability (GCA) estimates, parents were selected to generate larger sized progenies
for DMC mapping. Sizes of families in the original diallel experiment were
about 30 progenies, a rather small size for genetic mapping. Parallel to the
development of mapping populations was the search for markers associated
with DMC using 2 F1 families, GM312 and GM313 selected from the diallel
experiment having parents with high GCA for DMC.
Methodology
A crossing block of high and low GCA parents for DMC (Table 8.4) was
established in CIAT, Palmira in October 2002, with the aim of generating larger
sized progenies for QTL mapping of DMC. Crossing is ongoing for all genotypes
except for SM1741-1 x MECU) and SM 1411-5 x CM 4574-7), which have either
few plants or have poor flowering. Bulk segregation analysis (Michelmore et al.,
1991) was conducted for 2 F1 crosses, GM 312 and GM 313, each with MECU
72 (high dry matter and good GCA) as a parent. DNA samples were extracted
from 3g of fresh leaves according to Dellaporta et al., 1983. Between 500Pg to
1000Pg of high quality DNA was obtained from each extraction and quantified
using flourometer. The samples where then diluted to 10ng/Pl for PCR
amplification. DNA from genotypes with percent dry matter of 34.3 - 37.7 and
24.5 - 31.5 were bulked to form high and low bulks respectively. The Bulks
Output 8-13
2003 Annual Report
and parents were screened with polymerase chain (PCR) reaction using 650
SSR primers developed in CIAT. PCR product was denatured and
electrophoresed on 6% polyacrylamide gels according to Mba et al., 2001.
Results
A summary of crosses made so far for the development of QTL mapping
populations for DMC can be seen in Table 8.4.
Table 8.4. Status of the crosses for QTL mapping.
Cross
Number of crosses
Direct
Reciprocal
Target area
SM 1741-1 x MPER 183
SM 1741-1 x SM 1219-9
SM 1741-1 x MECU 72
SM 1411-5 x MTAI 8
CM 8027-3 x CM 6757-8
SM 1665-2 x SM 805-15
SM 1411-5 x CM 4574-7
SM 1219-9 x SM 1565-15
CM 4574-7 x SM 1565-15
269
104
0
36
224
544
2
188
124
Mid altitude
Mid altitude
Mid altitude
North coast
North coast
North coast
“out cross”
Acid savannah soils
Acid savannah soils
7
114
0
97
154
570
0
163
Thirty-one primers were found to be polymorphic between the bulks of high and
low dry matter content and in individuals of the bulks from family GM 313.
These were tested on the remaining genotypes of the cross. Of these eleven NS
701, NS 717, NS781, NS80, NS 9, NS 909, NS 917, NS 955, SSRY 150, SSRY
160 SSRY 88 and NS 371 remained polymorphic. A simple regression of dry
matter content on the genotypic classes of the SSR markers was highly
significant for the 2 markers, SSRY160 and SSRY150, with regression
coefficient (R2) of 29.3 and 18.1 respectively. The phenotypic variance explained
by these two markers, based on their regression coefficient, is enough to
consider them as markers for marker-assisted selection (MAS). All other
markers had R2 of between 0.05 and 0.1. The other family GM 312, yielded no
markers polymorphic in the individuals of the bulks in the BSA.
The markers SSRY 160 and SSRY150 associated with DMC in the GM 313
cross are being tested on about 700 genotypes from 23 crosses of the diallel
experiment. The 23 crosses were selected from crosses with a high standard
deviation for dry matter content and from parents with good general combining
abilities. Theses crosses had earlier been planted in the field in Quilichao for
further evaluation of DMC in a completely randomized block design of 3 blocks
(replicates) and up to10 plant per block. At harvest data was collected on fresh
foliage yield, fresh root yield, number of tubers, harvest index, number of
tubers per plant, yield, percentage dry matter and dry matter yield. Results are
summarized in tables 8.5. Low standard deviation values were estimated for
dry matter content compared to other yield related parameters suggesting that
it is a relatively stable trait. All the parameters were highly correlated (Table
8.6) confirming their importance to yield. Dry matter and number of tubers per
plant were not significantly correlated to the number of plants harvested
Project IP3: improving cassava for the developing world
Output 8-14
suggesting that they are not easily influenced by environment, which is
important to note since there was non-uniform establishment. There was an
incidence of frog skin disease (FSD) in this trial and it was highly and negatively
significant to all yield related traits, and it is likely to influence results of the
study. The affected plants were therefore eliminated from the final results.
Cross GM 311 had the highest standard deviation for dry matter (8.58) followed
by CM 9901 (6.47) while GM 228 had the lowest of 2.40 (Table 8.5).
Table 8.5.Some statistics of yield related parameters for the crosses evaluated
in Quilichao, during the 2002-2003 season.
Cross
CM 9642
CM 9733
CM 9901
GM 228
GM 257
GM 260
GM 265
GM 267
GM 268
GM 269
GM 283
GM 284
GM 285
GM 286
GM 293
GM 294
GM 306
GM 309
GM 310
GM 311
GM 312
GM 313
GM 314
Minimu
m
Max
No of
genotyp
es
Mean
St. Dev
% Dry matter
_______________
_
Mean
St. Dev
27
25
31
6
33
31
34
8
29
25
29
42
13
28
27
26
27
36
30
34
19
30
19
6
0.48
0.52
0.47
0.46
0.55
0.65
0.55
0.45
0.52
0.52
0.59
0.54
0.51
0.48
0.50
0.54
0.40
0.49
0.57
0.45
0.46
0.45
0.47
0.40
0.08
0.13
0.18
0.17
0.15
0.10
0.12
0.17
0.15
0.13
0.17
0.18
0.19
0.13
0.14
0.12
0.11
0.19
0.13
0.19
0.11
017
0.12
0.08
33.46
28.20
28.53
35.39
29.66
30.54
29.03
31.33
33.93
33.79
30.46
31.82
29.45
31.43
32.04
31.74
29.52
29.26
33.98
30.23
32.49
32.21
31.33
28.20
3.41
4.34
6.47
2.40
4.80
4.02
4.03
5.39
3.97
3.47
5.37
5.39
4.27
2.85
3.33
3.18
4.44
5.35
4.08
8.58
4.83
5.13
3.77
2.40
20.74
25.94
14.17
19.85
23.06
20.33
19.60
21.01
22.49
25.74
18.86
17.76
22.74
28.09
22.31
25.15
20.40
21.38
28.20
19.32
27.32
22.14
20.59
14.17
21.30
19.28
12.69
14.58
18.13
13.02
12.25
26.41
19.12
17.98
16.86
12.93
16.49
17.26
17.54
18.93
15.02
16.01
21.44
14.86
15.54
16.05
13.53
12.25
Dry yield (t/ha)
______________
_
Mean
St.
Dev
7.02
7.45
7.68
6.30
4.53
4.32
7.29
5.65
7.13
5.84
6.29
4.55
5.95
4.08
10.30
9.14
7.96
7.03
8.91
6.60
6.13
5.63
5.77
4.45
7.02
5.33
8.94
5.77
7.08
5.47
8.01
6.23
6.12
4.62
6.68
5.42
9.94
7.99
5.86
4.81
9.17
5.70
7.02
5.22
6.68
4.96
4.53
4.08
42
0.59
0.19
33.46
8.58
28.20
26.41
10.30
Output 8-15
Harvest Index (0-1)
_______________
Yield (t/ha)
______________
Mean
St.Dev
2003 Annual Report
9.14
Table 8.6. Correlation table of frog skin disease incidence and yield related
parameters
Plants
FSD
ComRt
TbNo
FolWt
HI
Yield
DM
Dyield
Plantsa
FSDb
ComRtc
TbNod
FolWte
HIf
Yieldg
DMh
-0.06NSi
0.57***j
0.05NS
0.57***
0.02NS
0.58***
0.06NS
0.54***
-0.23***
-0.21***
-0.02NS
-0.24***
-0.20***
-0.33***
-0.22***
0.56***
0.58***
0.36***
0.88***
0.29***
0.87***
0.28***
0.35***
0.52***
0.22***
0.52***
-0.29***
0.63***
0.07NS
0.61***
0.39***
0.26*** 0.27***
0.39*** 0.98*** 0.38***
a
=plants harvested per plot; b=frog skin disease; c=number of commercial roots per plot; d=tubers
per plant; e=foliage weight; f=harvest index; g=fresh yield (t/ha); h=dry matter content(0-1)
Conclusion and Perspectives
A study of genes controlling dry matter content in many full-sib families of
cassava has been initiated towards the development of molecular markers for
increasing the cost-effectiveness of breeding for dry matter content. Initial
marker analysis has led to the discovery of 2 markers, SSRY160 and SSRY150
that explain about 30% and 18% respectively of phenotypic variance for DMC.
These markers are being analyzed in 23 crosses with high standard deviation
and derived from parents with high GCA for DMC o confirm their utility across
genetic backgrounds. Parallel to this, larger families are being developed from
selected parent for QTL mapping of DMC.
References
Byrne, D., 1987. Breeding cassava. In: Russell (Ed), Plant Breeding Reviews
2:73-133.
Mba REC, Stephenson P, Edwards K, Mezer S, Nkumbira J, Gulberg U, Apel K,
Gale M, Tohme J, Fregene MA (2000) Simple sequence repeat (SSR)
marker survey of the cassava (Manihot esculenta Crantz) genome: toward a
SSR-based molecular genetic map of cassava. Theor Appl Genet 101: 2131
Michelmore ,R.W.,I. Paran and R.V. Kesseli, 1991 Identification of markers
linked to disease –resistant genes by bulked segregant analysis : a rapid
method to detect markers in specific genome regions by using segregating
populations. Proc. Natl. Acad. Sci. USA 88: 9828-9832.
Project IP3: improving cassava for the developing world
Output 8-16
Activity 8.6.Genetic Mapping of Cyanogenic Potential (CNP)
Collaborators: Elizabeth Kizito, Urban Gullberg (SLU, Uppsala, Sweden); AnaMaria Corea, Jaime Marin, Edgar Barrera, Nelson Morante,
Teresa Sanchez, Martin Fregene (CIAT)
Funding: BIOEARN, DANIDA
Important Outputs
1) The genotyping of the S1 family AM320 derived from MTAI8, with 2
linamarin biosynthetic genes CYP D1 and D2, SSR, and DarT markers
for the genetic mapping of QTLs controlling cyanogenic potential (CNP).
2) Phenotypic evaluation of root and leaf CNP under field and uniform
conditions in a phytotron
Rationale
The project on genetic mapping of cyanogenic potential in cassava, a
collaborative project between the Swedish Agricultural University (SLU),
Uppsala, the Medical Biotechnology Laboratories (MBL), Kampala, and CIAT, is
in its second year. The project is being conducted as a Ph.D. research project
by Ms Elizabeth Kizito at SLU. A review of the project was conducted at a
meeting in Kampala April this year, where several changes in the project were
decided. The principal change was to discontinue the development of F2
populations derived from the cross between the bitter Ugandan variety, Tongolo,
and the CMD resistant land race, TME4, due to the time limit for the
completion of Elizabeth’s Ph.D. research. It was decided that the S1 family
AM320 derived from the bitter variety MTAI8 available at CIAT should be used
instead. This family is currently being genotyped with more than 800 DArT
makers, at CAMBIA, Australia, and 200 SSR markers, at CIAT, for gene tagging
of beta-carotene content, dry matter and harvest index in cassava.
A new initiative was also embarked during the year on the genetic mapping of
the two cytochrome P450 genes, CYP79D1 and D2 that catalyze the ratelimiting step of the biosynthesis of the cyanogenic glucosides, linamarin in the
S1 family AM320 and a search for association with QTLs for CNP. This project
is collaboration with the Royal Veterinary and Agricultural University
Copenhagen, Denmark (Prof Birger Moller), where the genes were cloned. The
discovery of molecular markers for CNP will provide a tool to efficiently select for
low cyanogenic potential in cassava breeding.
Methodology
The 102 individuals of the family AM320 in the field were transferred in vitro
following methods routinely used in the cassava tissue culture facility (Activity
23 of this report). Another 32 plants were obtained from embryo rescue of
residual sexual seeds of AM320. Ten copies per genotype were shipped to SLU,
Uppsala for evaluation of root and leaf cyanogenic potential under uniform
nutrient, temperature, light and humidity conditions in a phytotron. A field
evaluation of CNP in the roots will also be conducted at CIAT. For genetic
mapping of SSR markers and the CYPD1 and D2 genes, DNA was isolated from
green house plants of all AM320 genotypes using a modified Dellaporta et al.
(1983) DNA isolation protocol. DNA from MTAI8, parental genotype of family
AM320, and 5 S1 progenies was screened with 650 SSR markers to identify
polymorphic markers for genetic mapping. To map the cytochrome P-450
Output 8-17
2003 Annual Report
genes, primers were designed from thes sequence of CYPD1 and D2 using the
“Primer3” primer picking software found at http://waldo.wi.mit.edu/cgibin/primer/primer3 (Whitehead Institute for Biomedical Research). The genes
were then amplified in 50ul volume reactions containing 50-100ng of genomic
DNA from MTAI8, 0.2 µM of each forward and reverse primers, 10mM Tris-HCL
(pH 7.2), 50mM KCL, 1.5 or 1mM MgCl2, 200mM of each dNTP, and about 1U of
Taq DNA polymerase. Temperature cycling profile was: an initial denaturation
step for 5min at 94˚C, followed by 30 cycles of denaturation at 94˚C for 1min,
annealing at 55˚C (D1) or 60˚C (D2) for 2 min and primer extension at 72˚C for
2 min. A final extension cycle of 5 min at 72˚C was added. Between 4 and 5 µl
of the PCR reaction was electrophoresed on 5% ethidium bromide stained
metaphor agarose gels and visualized under UV light. The fragments are to be
mapped either as cleaved amplicon polymorphism (CAPs), RFLPs or SNPs
according to standard methods (Cortes et al. 2003; Fregene et al. 1997).
Results
A total of 130 genotypes from the S1 family AM320 have been shipped to SLU,
Uppsala for phenotypic evaluation of CNP in the leaves and roots. Evaluation of
the same family will be conducted on field grown plants by October at CIAT.
Seventy three polymorphic SSR markers, that segregate in the expected fashion
has been found out of 320 SSR markers surveyed till date.
Primers designed for the CYP79 D1 and D2 genes are shown below. They were
used in amplifying a part of the genes from parent MTAI8 and some of its S1
progenies (Fig1).
CYP79 D1
Forward primer AAAGAGTGCTGCTAACAAGG
Reverse primer CCATTGTTGAATCCTTTCAT
CYP79 D2
Forward primer GGTACAGACCGACGTTTCGT
Reverse primer
AATGGCTTGCCATCTGAATC
M
M
M
Fig 8.2. Ethidium bromide stained gel of PCR amplification product of MTAI
8 and 5 S1 progenies amplified with CYP79 D1 primers (first panel of six
lanes), and CYP79 D2 primers (second panel of six lanes). Lanes labelled M
are molecular weight markers (Lambda DNA digested with Pst I restriction
enzymes).
Project IP3: improving cassava for the developing world
Output 8-18
The search for single dose fragments as CAPs, RFLPs, or SNPs in genes CYPD1
and D2 are ongoing using the genotype in MTAI8. Once these markers have
been identified they will be mapped in the S1 family AM320.
Conclusions and perspectives
The genetic mapping of CNP has continued with the S1 family AM320 derived
from MTAI8. Phenotypic evaluation of root and leaf CNP is being conducted at
SLU, Uppsala, Sweden under uniform conditions in a phytotron. The same
family is being genotyped with SSR (at CIAT) and DArT markers (at CAMBIA,
Australia), root CNP will also be evaluated in the field at CIAT. Parallel to this
effort is the mapping of the CYP D1 and D2 gene in the AM320 family. It is
expected that markers associated with CNP will be identified at the end of the
study.
References
Fregene M., Angel F. , Gómez R., Rodríguez F. , Bonierbale M., Iglesias C.,
Tohme J., Roca W. 1997(a). A molecular genetic map of Cassava (Manihot
esculenta Crantz) Theor. Appl. Genet. 95:431-441.
Activity 8.7 Genetic Mapping of Leaf Retention
Collaborators: J. Marin, C. Ospina, J. Perez, H. Ceballos, M.Fregene (CIAT)
Funding: CIAT core funds
Important Output
1) Phenotypic evaluation of more than 100 full-sib families in the North
Coast of Colombia has led to the identification of 2 full-sib and 1 half-sib
family for genetic mapping of genes controlling leaf retention in cassava
Rationale
Leaf retention or the ability of some cassava genotypes to retain their leaves at
5, 6 and up to 7 months after planting, is a trait that has been shown to be
associated with dry matter yield. An experiment conducted in the northern
coast of Colombia (sub-humid agro-ecology) to measure the effect of leaf
retention revealed that clones that retained leaves at 5-6 months had a 26.4%,
7%, and 32% increase at harvest in fresh root yield, dry matter content and dry
matter yield respectively (Lenis et al 2003). Leaf retention in these trials was
also associated with higher root dry matter content and harvest index (7%
increase). This experiment was repeated during a second season with similar
results. Similarly it has been shown that leaf retention and dry matter content
are negatively correlated making it possible to simultaneously improve both
traits. Preliminary genetic analysis suggests that the trait maybe simply
inherited. Development of markers for carrying out pre-selection for leaf
retention of breeding populations at CIAT before they are sent to the target
environment should therefore reduce the load of evaluation and increase costeffectiveness of breeding.
Methodology
More than 100 full-sib and half-sib families were evaluated for leaf retention in
a clonal field trial in Santo Tomas, the Colombian north coast. Leaf retention
was measured visually, on a score of 1 to 9 at 17, 19, 21, and 23 weeks after
Output 8-19
2003 Annual Report
planting. Two full-sib families, and 1 half-sib family were selected based on
the wide segregation for leaf retention. Ten good leaf retention and 10 poor leaf
retention genotypes were selected from each of the families to serve as bulks for
bulk segregant analysis (BSA) for the discovery of markers associated with leaf
retention. For DNA isolation, 1-2g of young leaves was harvested from
genotypes of all the 6 bulks and from all other genotype in each family. The
leaves were dried for 24h in an oven at 48oC and then ground into a fine powder
using a power drill and washed sand. DNA was isolated from 200mg using a
miniprep version of the Dellaporta (1983) protocol. The bulks of high and low
leaf retention were then created per family to give a total of 6 bulks. DNA from
the bulks and the parents will be genotyped with the 650 available cassava SSR
markers according to methods described by Mba et al (2001), and is just
beginning. Markers polymorphic in the bulks were employed to analyze
individuals of the bulks and, where the polymorphism remained consistent in
the individuals as with the bulks, the markers were analyzed in the entire
family.
Results
Evaluation of leaf retention in the North coast of Colombia in more than 100
families led to the selection of three families for molecular analysis. Table 8.7
describes some statistics of leaf retention in the 3 families. It can be observed
that leaf retention evaluated at 23 weeks after planting has the widest
variability within families and is the most appropriate measurement for use in
molecular analysis of the trait. The choice of 3 families for marker analysis is
to identify markers that are useful across genetic backgrounds that can be used
in a breeding program regardless of the parental genotypes.
Table 8.7 Descriptive statistics of 3 families evaluated for leaf retention in the
Colombian North Coast at 17 (RF1), 19 (RF2), 21 (RF3) and 23 (RF4) weeks
after planting.
Clon
CM 9775- 12
Madre
CM 7514- 7
Padre
RF1 RF2 RF3 RF4
MNGA 19 MAX
9.00 9.00 9.00
9.00
MIN
8.00 6.00 4.00
2.00
AVERAGE 8.87 8.25 7.53
6.51
STD DEV
0.33 0.92 1.91
2.49
SM 2783- 45
SM 1511- 6
MAX
MIN
AVERAGE
STD DEV
9.00
8.00
8.74
0.44
9.00
6.00
7.61
1.04
9.00
4.00
6.55
1.81
9.00
3.00
5.71
2.17
SM 2615- 70
CM 4365- 3
MAX
MIN
AVERAGE
STD DEV
9.00
7.00
8.76
0.55
9.00
7.00
8.06
0.76
9.00
4.00
6.85
1.57
8.00
3.00
5.21
1.81
Conclusions and perspectives
Genetic mapping of genes controlling leaf retention has been initiated using 2
full-sib and one self family and BSA. Markers identified to be associated with
Project IP3: improving cassava for the developing world
Output 8-20
leaf retention will serve as a means to screen populations for the trait at CIAT
before they are sent to the field.
References
Dellaporta SL, Wood J, Hicks JR (1983) A plant DNA minipreparation: version
II. Plant Mol Biol Rep 1: 19-21.
Lenis J.I., Calle F., Jaramillo G., Perez J.C., Ceballos H., and Cock J.H. (2003).
The effect of leaf retention in cassava productivity (submitted to Crop
Science).
Activity 8.8. Simple Sequence Repeat (SSR) Assessment of Genetic
Diversity of Local Cassava Varieties from Guatemala
Collaborators: Cesar Azudia Luis Monte (Facultad de Agronomia, Universidad
de San Carlos de Guatemala), Charles Buitrago, Daniel
Debouck, Martin Fregene (CIAT)
Funding: IPICs, University of Uppsala
Important Output
1) A study to assess the genetic diversity of 128 cassava land races from
Guatemala and revealed a group of accessions with many unique alleles
and high genetic differentiation compared with other accessions from
Guatemala and other parts of the world. These results suggest a second
center of diversity in Guatemala or an introgression from wild species.
Rationale
Two primary centers of diversity, one in South America and the other in MesoAmerica have been postulated for the genus Manihot (Roger and Appan 1973).
Although several studies have demonstrated a likely South American origin for
the cultivar (Allem, 1994; Fregene et.al 1994; Roa et al. 1997; Olsen and Schaal
1999), the diversity of cassava and its wild relatives in Memo- America is great
enough to suggest a second center in Meso-America. Besides, the potential of
Meso-American diversity in cassava improvement has not been properly
assessed. Three recent studies of genetic diversity in land races from South
America and Meso-America (Chavariagga et. al. 1999; Fregene et. al. 2002; Raji
et. al. unpublished data) have revealed unique alleles in land races from
Guatemala at a frequency high enough to suggest a Meso American center of
cassava diversity. The results of the three studies were based upon 6, 4, and
13 Guatemalan land races. The small sample size of the previous study could
distort the allele frequencies and lead to wrong conclusions. A larger collection
and SSR characterization of land races from Guatemala was therefore planned
to confirm preliminary data of a Meso-American center of diversity and to
secure the largely untapped diversity in Guatemala before it becomes extinct.
In addition, a selection from the Guatemalan collection will be crossed to CIAT
elite parents to evaluate the utility of the Meso-American diversity in cassava
breeding.
The present study was to confirm the high genetic differentiation between
cassava land races from Guatemala and Nigeria, Brazil, and Colombia. If the
uniqueness of the Guatemalan germplasm is confirmed, genetic crosses to
CIAT’s elite breeding lines will be made to test hybrid vigor and delineate
Output 8-21
2003 Annual Report
heterotic pools. Plant materials are a collection of cassava from all over
Guatemala and a representative group used in previous studies from Nigeria,
Colombia and Brazil to confirm earlier results. It is hoped that results of the
uniqueness and the utility of the Guatemalan germplasm will give collection
and conservation of this germplasm in regions of Meso-America high priority
(Azurdia and Gomez 2002)
Methodology
A collection of cassava land races was carried out all over Guatemala in May
last year (Azurdia and Gomez 2002). A total of 128 accessions were collected in
the departments of Baja Verapaz, Quiche, Huehuetenango, Alta Verapaz, San
Marcos, Escuintla y Santa in Guatemala. See Guatemalan study on the
MOLCAS web site (http://www.ciat.cgiar.org/Molcas) for names of accessions.
For comparison with results of previous studies, DNA from 6, 11 and 12
cassava land races from Nigeria, Colombia y Brazil respectively were included,
making 4 samples for the analysis of genetic diversity and differentiation. DNA
from the Guatemalan accessions was isolated at the Facultad de Agronomia,
Universidad de San Carlos de Guatemala using a micro-prep protocol of the
Dellarporta (1983) methodology and transferred to CIAT. DNA from the other
accessions was obtained from previous studies at CIAT.
A set of 36 SSR markers, carefully chosen to represent a broad coverage of the
cassava genome with moderate to high polymorphism information content (PIC)
and robust amplification, were used in this study. SSR markers, PCR
amplification, polyacrylamide gel electrophoresis, and silver staining used in
this study have been described elsewhere (Fregene et al 2002). The allele data
was captured using the program “Quantity One” (Bio-Rad Inc) and entered
directly into EXCEL (Microsoft Inc) for statistical analysis. Statistical analysis
on the raw SSR data include: genetic distance analysis using a of a distance
matrix based upon 1-proportion of shared alleles (1-PSA), principal component
analysis (PCA) and cluster analysis (UPGMA) of the distance matrix, and
parameters of genetic diversity and differentiation.
Results
A total of 33 SSR markers were analyzed in the 128 accessions from Guatemala
that includes an accession of the wild relative M.aesculifolia. Unique alleles
were observed in the accessions from Guatemala for the markers SSRY 12
(0.14), SSRY 20(0.383), SSRY 34(0.006), SSRY 38(0.063), SSRY 51(0.1), SSRY
59(0.014), SSRY 63(0.014), SSRY 69(0.023), SSRY 82(0.007), SSRY 103(0.24),
SSRY 108(0.043), SSRY 135(0.13) y SSRY 147(0.013). In parenthesis are the
frequencies of the observed alleles. The first and second principal components
of the PCA, based on upon the genetic distance 1-proportion of shared alleles,
are shown in Fig 8.3. Accessions from Guatemala form two groups, one that
clusters along with land races from Brazil, Nigeria and Colombia in a broad
group and a second group that is clusters separately. The only sample from M.
aesculifolia is located far away from both clusters. The results observed
confirms previous observation of a high genetic differentiation of between
certain groups of cassava land races from Guatemala and these from other
parts of Latin America and Africa (Fregene et al. 2003)
Assessment of genetic diversity was based on samples of cassava land races
from the 4 countries, with an addition that accessions from Guatemala were
divided into two groups G1 and G2 based on clustering from PCA of genetic
Project IP3: improving cassava for the developing world
Output 8-22
distances and UPGMA of FST data. Table 8.8 summarizes the parameters of
genetic diversity observed for accessions from the 5 samples. Genetic diversity,
as assessed by the average gene diversity (HE) was high in the accessions
analyzed 0.5422 r0.2468. The population with the highest diversity was
Colombia followed by the cluster G2 and that with the lowest was the cluster of
the Guatemalan accessions clustered separately from other accessions.
Average number of alleles was 3.8+/-0.0358 for all accessions. Average number
of alleles per locus was highest in the cluster G2 of Guatemalan land races, 5.2,
and the lowest in the Nigerian land races, 2.9. A break down of genetic
diversity parameters by individual SSR markers can be seen in Table 8.9.
Principal Component Analysis
Manihot aesculifolia
25
20
Samples from
Guatemala
15
10
Guatemala
Brazil
5
Colombia
Nigeria
0
-15
-10
-5
0
5
10
15
20
-5
-10
-15
PC1
Fig.8.3. Principal component analysis of genetic distances between cassava
accessions from Guatemala, Brazil, Colombia and Nigeria.
A UPGMA cluster analysis of the genetic distance data also produced 2 clusters
of the Guatemalan accessions similar to that found with the PCA (data not
shown). In addition, 2 sub groups were found within the group G1 that
clustered away from the majority of accessions. A UPGMA of a pair-wise
analysis of genetic differentiation (FST) again confirmed the separation of a
group of accessions from Guatemala (Fig 8.4) as observed with the PCA and
UPGMA analysis of genetic distances. The geographical distribution of
accessions in cluster G1 can be observed in Figure 8.5. The distribution of
accessions closely mirrors the distribution of 2 wild Manihot species in
Gutemala, namely Manihot rhomboide and Manihot aesculifolia (Fig. 8.5). The
majority of accessions in sub group A are found in western Guatemala and they
overlap, as regards geographical origin, with Manihot rhomboide. On the other
hand, genotypes from sub group B are found mostly in the Eastern part of the
country together with natural populations of Manihot aesculifolia.
Output 8-23
2003 Annual Report
Gp1
Gp2
Brazil
Colombia
Nigeria
0.04
0.08
0.12
0.16
0.20
Coefficient
Fig. 8.4. UPGMA tree of pair-wise FST data calculated between samples from the
four different countries.
x Group A
x Group B
M. aesculifolia
ʳʳSʳʳM. rhomboidea.
Fig 8.5. Distribution of accessions from groups A and B, of G1 and Manihot
aesculifolia and Mahihot rhomboide in Guatemala .
The origins of highly differentiated samples of cassava germplasm from
Guatemala can be explained by independent domestication events in
populations of different Manihot species that yet exist or are now extinct. They
can also be explained by an introgression from Manihot species in certain
regions that overlap in geographical spread with cassava. Cassava is an
Project IP3: improving cassava for the developing world
Output 8-24
allogamous crop and natural cross pollinate between cassava and populations
of wild Manihot species has been demonstrated Wanyera et al (1993).
The highly differentiated landraces from Guatemala may represent heterotic
pools, like those for maize (Shull 1952). A principal reason for this study was to
assess genetic diversity in cassava landraces as a first step to delineating
heterotic pools for a more systematic improvement of combining ability via
recurrent reciprocal selection (Keeratinijakal and Lamkey 1993). The heterotic
patterns found in maize populations at the turn of the century is the basis of a
very successful maize hybrid industry and has raised maize yields 500% since
1928 (Shull 1952; Tomes 1998). A high level of genetic differentiation, as
revealed by molecular markers, was later found between these populations
(Melchinger et al. 1990). It is noteworthy that accessions from sub group A
and B, for example accession 405, 478, 332, and 729 have excellent agronomic
characteristics (data not shown)
Conclusions and perspectives
A previous study of the assessment of genetic diversity of cassava land races in
14 South and Central American and African countries revealed a number of
unique alleles in accessions from Guatemala and suggests a second center of
diversity in Guatemala (Fregene et al 2003). Meso America is a center of
diversity for many other food crops including common beans, maize, amongst
others. This study shows unique alleles from Guatemala for a higher number
of SSR markers and provides additional evidence for possible independent
domestication of cassava in Meso America. However, an introgression with
Manihot species that overlap with the geographical origins of these accessions
makes it impossible to rule out introgression with these species. Further
studies are required to clarify which is the most likely scenario via the collection
and characterization of wild Manihot species the eastern and western parts of
Guatemala. Additional future activities include a diallel cross of theses
excellent land races from Guatemala and other regions (initial experiments are
described below in activity 8) .
References
Azurdia, C. y M. Gonzalez. 1986. Informe final del proyecto de recoleccion de
algunos cultivos nativos de Guatemala. FAUSAC, ICTA, CIRF.
Azurdia, C.;K Willeams; D. Williams; A. Jarvis (2003). Inventario de las
especies silvestres parientes de los cultivos natives de Guatemala. FAUSAC,
USDA, IPGRI. En preparación
Chavarriaga-Aguirre, P., M. M. Maya, J. Tohme, M. C. Duque, C. Iglesias, M. W.
Bonierbale, S. Kresovich, And G. Kochert, 1999 Using microsatellites,
isozymes and AFLPs to evaluate genetic diversity and redundancy in the
cassava core collection and to assess the usefulness of DNA-based
markers to maintain germplasm collections. Molecular Breeding 5: 263273
Dellaporta SL, Wood J, Hicks JR (1983) A plant DNA minipreparation: version
II. Plant Mol Biol Rep 1: 19-21.
Dje Y., Mheuertz, Clefebvre, X Vekemans. 1999. Assessment of genetic
diversity within and among germplasm accessions in cultivated sorghum
using microsatellite markers. Theor Appl Genet
Fregene M., Suarez M., Mkumbira J. , Kulembeka H. , Ndedya E. , Kulaya A. ,
Mitchel S. Gullberg U. , Rosling H., Dixon A., Kresovich S. (2001) Simple
Sequence Repeat (SSR) Diversity of Cassava (Manihot esculenta Crantz)
Output 8-25
2003 Annual Report
Landraces: Genetic Structure in a Predominatly Asexually Propagated
Crop (Theor Appl Genet, in review).
Fregene MA, Vargas J, Ikea J, Angel F, Tohme J, Asiedu RA, Akoroda MO, Roca
WM
(1994) Variability of chloroplast DNA and nuclear ribosomal
DNA in cassava (Manihot esculenta Crantz) and its wild relatives. Theor
Appl Genet 89: 719-727.
Melchinger AE, Lee M, Lamkey KR, Woodman WL (1990) Genetic diversity for
restrictionfragment length polymorphisms: relation to estimated genetic
effects in maize inbreds. Crop Sci 30:1033-1040
Olsen K and Schaal B. 1999. Evidence on the origin of cassava:
phylogeography of Manihot esculenta. Proceedings of the National Academy
of Science 96: 5586-5591.
Roa, A.C.; M.M. Maya; M.C. Duque; J. Tohme; A.C. Allem; M.W. Bonierbale.
1997. AFLP analysis of relationship among cassava and other Manihot
species. Theor Appl Genet 95: 741-750.
Shull GF (1952) Beginnings of the heterosis concept. In: JW Gowen. Heterosis.
Iowa Sate College Press, Ames, Iowa, pp 14-48
Tomes D (1998) Heterosis: performance stability, adaptability to changing
technology and the foundation of agriculture as a business. In: K Lamkey, J
E Staub. Concepts and breeding of heterosis in crop plants. CSSA Special
Publication Number 25. Crop Science Society of America, Madison, WI.
Wanyera, N. 1993. Phylogeny of two Manihot species and their natural hybrids.
Ph.D. thesis University of Ibadan, Nigeria.
Table 8.8 Intra-population and inter-population estimates of genetic diversity
parameters of cassava land races from different agro-eclogies of
Guatemala, Brazil, Colombia, Nigeria
Pop.
n
G1
G2
BRA
COL
NIG
24
74
12
8
4
mean
std
No.
No. of
de loci
Loc. Pol.
33
33
33
33
33
28
33
32
33
31
33 31.4
6.28 0.98
Percent of
Pol. Loc.
Average No. Average No.
of
of allles/Loc.
alleles/Loc.
Pol.
84.8
100
97
100
93.9
2.8
5.2
4
4.1
2.9
3
5.2
4.1
4.1
3.1
95.15
0.91
3.8
0.0358
3.87
0.0845
HO
HE
HEc_p
0.6273
0.5895
0.5562
0.5912
0.648
0.41
0.6066
0.5745
0.6111
0.5085
0.419
0.6107
0.6013
0.6555
0.6067
0.6024 0.5422 0.5786
0.0918 0.2468
Ho: observed heterozygosity
He: Average gene diversity
Hec_p: Average gene heterozygosity corrected for small samples sizes
Project IP3: improving cassava for the developing world
Output 8-26
Table 2. Parameters of Genetic diversity, Ho, Hs, Ht, Dst, Gst and Gst’
(correction for differences in sample size) by SSR locus
Locus
SSRY4
Ht
Hs
Dst
Gst
0.7596
0.61
0.7061
0.7698
0.8235
0.6691
0.6554
0.1665
0.7177
0.7533
0.6556
0.7151
0.7428
0.8154
0.7673
0.2887
0.5395
0.687
0.5847
0.6123
0.3628
0.5545
0.6644
0.3403
0.7632
0.6983
0.6889
0.6843
0.5767
0.6427
0.652
0.7471
0.6035
0.4869
0.1585
0.6324
0.5214
0.5402
0.5325
0.6926
0.6638
0.6648
0.2534
0.4988
0.5074
0.5173
0.4013
0.3502
0.5353
0.6218
0.3132
0.7046
0.6141
0.5989
0.0753
0.0333
0.0634
0.1177
0.0764
0.0655
0.1685
0.008
0.0854
0.2319
0.1155
0.1826
0.0502
0.1517
0.1025
0.0354
0.0407
0.1796
0.0674
0.211
0.0126
0.0192
0.0426
0.0271
0.0587
0.0843
0.09
0.0991
0.0546
0.0897
0.153
0.0928
0.0979
0.2572
0.0477
0.1189
0.3078
0.1761
0.2554
0.0676
0.186
0.1336
0.1225
0.0755
0.2615
0.1153
0.3446
0.0346
0.0346
0.0641
0.0796
0.0769
0.1207
0.1306
SSRY169
SSRY171
SSRY179
SSRY180
SSRY181
0.7527
0.6498
0.3128
0.5225
0.6236
0.6305
0.6476
0.5982
0.2801
0.5018
0.565
0.5831
0.1052
0.0516
0.0326
0.0208
0.0587
0.0473
0.1397
0.0794
0.1044
0.0397
0.0941
0.0751
mean
std
0.6244
0.1619
SSRY9
SSRY12
SSRY19
SSRY20
SSRY21
SSRY34
SSRY38
SSRY51
SSRY59
SSRY63
SSRY64
SSRY69
SSRY82
SSRY100
SSRY102
SSRY103
SSRY105
SSRY106
SSRY108
SSRY110
SSRY127
SSRY135
SSRY147
SSRY151
SSRY155
SSRY161
SSRY164
0.5422
0.0822
0.1397
0.0589
95%CI
0.5627
0.4793
95%CI
0.6743
0.587
Ho Average observed heterozygosity within country
Ht Total Heterozygosity in the entire data set
Hs Gene diversity within country averaged over the entire data set
Dst Average gene diversity between populations
Gst Coefficient of gene differentiation.
Output 8-27
0.1252
0.0791
0.064
0.1035
Dm
0.0822
Rst
0.1516
0.1009
0.1514
2003 Annual Report
Activity 8.9 Characterization of genetic diversity:
Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local
Cassava Varieties from Ghana and Predictability of Heterosis
Collaborators: Elizabeth Okai, Dr John Otoo (Crop Research Institute, CRI,
Kumasi, Ghana) Charles Buitrago, Martin Fregene (CIAT) Dr
Alfred Dixon (IITA)
Important outputs
1) Genetic diversity analysis of 320 land cassava varieties collected in
Ghana using 36 SSR markers revealed average gene diversity and low
genetic differentiation. Principal component analysis of genetic distance
based on 1-proportion of shared alleles revealed a sub-structure in
theses collections.
2) To test for heterotic patterns present within the Ghanian collection or
between the collection and Guatemalan and Latin American accessions,
selected genotypes were planted in a crossing block along with
introductions for genetic crosses.
Rationale
Cassava is the number one provider of calories in Ghana. The Portuguese first
introduced cassava from Brazil to Ghana during the 16th and 17th centuries
(Jones, 1959). In the then Gold Coast, the Portuguese grew the crop around
their trading ports, forts and castles. It was a principal food eaten by both the
Portuguese and the slaves. By the second half of the 18th century, cassava had
become the most widely grown crop of the people of the coastal plains (Adams
1957). The spread of cassava from the coast into the hinterlands was very slow.
It reached Ashanti region, Brong Ahafo and the northern Ghana, mainly around
Tamale in the 1930. Until the early 1980s,the Akans of the forest belt preferred
plantain and cocoyams and sorghum and millet in the north. Cassava became
firmly established in most areas after the serious drought of 1982/83 when all
other crops failed completely (Korang-Amoakoh,Cudjoe and Adams 1987).
Cassava ranks first in the area under crop cultivation and utilization. It
contributes 22% of the agricultural gross domestic product AGDP compared to
5% for maize, 2% for rice, and 14% for cocoa (Al-Hassan, 1989;Dapaah, 1996).
According to the Ghana Living Standards Survey (GLSS), 83% of 1.73 million
sampled households were found engaged in cassava production. The spread of
cassava into the upper west and upper east of Ghana is an indication of
growing trend in cassava production through area expansion( MOA, 1990).
In the traditional bush-fallow system, some cassava plants are allowed to grow
during the fallow period, which is long enough to allow cassava to flower and
set seeds. The usual out crossing habit of cassava leads to the production of
numerous volunteer hybrids and a heterozygous gene pool, which creates
phenotypic diversity. Desirable new hybrid combinations from volunteer seed
are often selected by farmers and propagated. This process creates pools of
new land races, which are adapted to the different agro-ecological zones of
Ghana. Farmers have done selection for desirable traits for over 500 years.
Hence the landraces possess higher frequencies of genes required for
adaptation to biotic and abiotic stresses, food quality characteristics than
Project IP3: improving cassava for the developing world
Output 8-28
unadapted materials. Vegetative propagation also leads to the accumulation of
pest and diseases and good varieties susceptible to these biotic stresses
disappear. These factors lead to a fairly high turnover of varieties and have
implications for gene pool structure of cassava in any center of diversity.
Selection is one of the principal factors at work in cassava’s gene differentiation
in Africa. High heterosis for yield components, starch, and number of roots
have been observed in cassava, and hence considered a promising method of
genetic improvement (Easwari Amma and Sheela, 1996). Heterotic groups
identified in maize in the early 20th century (Shull et al 1953) have been the
basis of a very successful hybrid seed industry.
Objectives for the study were:
1. The objective for this study is to assess the genetic diversity and
differentiation in Ghanaian landraces
2. To detect heterotic patterns in the collection and between the Ghanaian
collection and land races from other countries and regions.
3. To generate hybrids between the Ghanaian land races and genotypes from
putative heterotic groups and select together with farmers superior hybrids
from the crosses.
Methodology
In January 2002 a collection of cassava land races from all the agro ecological
zones in Ghana was done. A total of 45 villages visited during the collaborative
study on cassava in Africa (COSCA) were visited. Another 28 villages, important
for cassava production, were also visited. Farmers were assembled and asked
to share information on cassava varieties grown by them, characteristics of
their varieties, and reasons for keeping them. Farmers volunteered to give
mature cassava stems, which were labeled. A total of 320 land races were
collected. For the list of genotypes, passport data and characteristics please see
the MOLCAS home page (http://www.ciat.cgiar.org/molcas)
Fresh young leaf samples of the accessions were collected on ice and used for
DNA extraction. An amount of 0.1g of the fresh young leaf was ground in liquid
nitrogen and the DNA extracted using the Qiagen kit. The extraction was
carried out in IITA, Ibadan, Nigeria. The DNA was carried in absolute ethanol
to CIAT. DNA quantification was done using the fluorometer. The DNAs were
diluted to 10ng/ul and used for SSR reactions. A sub-set of 36 SSR markers,
two from each of the 18 linkage groups of the cassava genome, was employed to
obtain an estimate of genetic diversity and differentiation in the land races.
PCR amplification, automated gel analysis and date collection were as descried
by Fregene et al (2003). Genetic distance, based upon the proportion of shared
alleles (PSA), obtained from the raw allele size data using the computer
microsat (Minch 1993, http://www.lotka.stanford.edu/microsat.html).
Distances between the accessions were subjected to principal component
analysis (PCA) using JMP (SAS Institute 1995) to obtain a structure of
relationship between the land races. Parameters of genetic diversity and
differentiation were calculated from allele data using the computer packages
GENSURVEY (Vekeman et al 1997) and FSTAT (Goudet 1990).
Output 8-29
2003 Annual Report
Results
A total of 320 landraces were collected including 18 genotypes with yellow
roots. Farmers who responded were predominantly women. Among the land
races some were very early bulking with maturity at 3-9 months after planting.
The various local names given suggest a lot of useful traits farmers had
associated with the varieties. Woody stems were cut to 20-30cm sizes and
planted in plastic pots in a nursery. These were sent to the field after 4weeks
and planted in an irrigated field at the Ashiaman office of the Ghana Irrigation
Authority. A copy of the collection was packaged and sent to IITA. Accessions
were planted in single rows at 1m x 1m spacing with improved varieties as
checks.
Data from a total of 33 of 36 SSR loci, 3 markers gave poor quality data, was
used to derive estimate of genetic diversity and differentiation genetic distances
between individual genotypes. The average number of alleles for each locus was
close to 5 and is similar to that found for a study of land races from Nigeria,
Tanzania and 7 Neo-tropical countries (Table 1). The probability that 2
randomly selected alleles in a given accession are different, average gene
diversity, was 0.5245r0.0045 and it is lower than that found in the found for
the previous study. Average gene diversity was comparable across all regions
with an exception of the central region and central savannah region. Genetic
diversity parameters, including total heterozygosity (Ht) and genetic
differentiation (Gst) ranged widely across markers. Genetic differentiation, as
estimated by FST (theta), was very low for samples between regions, overall 0.04,
with the exception of some accessions from Northern Ghana that showed
moderate to high genetic differentiation. (data not shown). The results found
here support previous findings that agricultural practices and the allogamous
nature of cassava produces a large pool of volunteer seedlings that natural and
human selection acts upon to maintain a high level of diversity and low
differentiation (Doyle et al. 2001; Fregene et al. 2002).
Project IP3: improving cassava for the developing world
Output 8-30
Table 1 Intra-population and inter-population estimates of genetic diversity
parameters of cassava land races from different agro-eclogies of Ghana
Population
n
#loc. #loc_P PLP
K
K_P HO_p HE_p
Ashanti
Brong
Ahafo
Central
Eastern
Coastal
Sav.
Greater
Accra
Volta
Northern I
Northen II
HEc_p
11
37
33
33
30
31
90.9
93.9
3.9
5.4
4.1
5.6
0.5285 0.5017
0.5012 0.5267
0.5262
0.5339
8
27
4
33
33
33
29
30
27
87.9
90.9
81.8
3.3
5.4
2.9
3.7
4.9
3.4
0.5082 0.4701
0.5322 0.5123
0.5404 0.467
0.4999
0.5223
0.5405
10
33
29
87.9
3.7
4
0.5016 0.5065
0.5336
28
109
53
33
33
33
31
31
31
93.9
93.9
93.9
5.3
6.9
6.3
5.5
7.2
6.5
0.5133 0.5735
0.5369 0.5779
0.5479 0.5851
0.5839
0.5806
0.5908
mean
90.57
4.69
4.98 0.5234 0.5245
std
4.13
0.36
1.31 0.0176 0.045
deviation
1
PLP: Percentage of polymorphic loci at the 5% level within accessions
K: Mean number of alleles per locus within accessions
K_P: Mean number of polymorphic alleles per locus within accessions
Ho_p: observed heterozygosity
HE-p: Average gene diversity
HEc_p: Average gene heterozygosity corrected for small samples sizes
0.5457
0.0317
Output 8-31
2003 Annual Report
Table 2. Parameters of Genetic diversity, Ho, Hs, Ht, Dst, Gst and Gst’
(correction for differences in sample size) by SSR locus
LocNam
Ho
Hs
Ht
Dst
Dst'
Ht'
Gst
SSRY4
0
0.346
0.439
0.093
0.104
0.45
0.212
SSRY5
0.499
0.474
0.481
0.006
0.007
0.481
0.014
SSRY9
0.463
0.582
0.587
0.005
0.006
0.587
0.009
SSRY12
0.704
0.597
0.598
0.001
0.001
0.598
0.001
SSRY19
0.811
0.738
0.764
0.026
0.029
0.766
0.034
SSRY20
0.79
0.73
0.76
0.03
0.033
0.764
0.039
SSRY21
0.596
0.487
0.514
0.027
0.03
0.517
0.053
SSRY34
0.479
0.428
0.425 -0.004 -0.004
0.424 -0.009
SSRY38
0.043
0.08
0.082
0.001
0.002
0.082
0.018
SSRY47
0.429
0.668
0.739
0.071
0.079
0.747
0.096
SSRY51
0.79
0.694
0.751
0.057
0.063
0.757
0.076
SSRY52
0.752
0.603
0.616
0.014
0.015
0.618
0.022
SSRY59
0.152
0.639
0.701
0.062
0.069
0.708
0.089
SSRY63
0.445
0.484
0.519
0.036
0.04
0.523
0.069
SSRY64
0.725
0.67
0.689
0.02
0.022
0.691
0.028
SSRY69
0.557
0.552
0.568
0.016
0.018
0.57
0.028
SSRY82
0.846
0.84
0.858
0.018
0.02
0.86
0.021
SSRY10
0.725
0.779
0.798
0.019
0.021
0.8
0.024
SSRY10
0.007
0.01
0.009
0
0
0.009 -0.037
SSRY10
0.804
0.76
0.764
0.004
0.004
0.764
0.005
SSRY10
0.829
0.761
0.768
0.008
0.009
0.769
0.01
SSRY10
0.422
0.361
0.373
0.012
0.014
0.375
0.033
SSRY11
0.279
0.272
0.274
0.002
0.002
0.274
0.008
SSRY12
0.814
0.629
0.653
0.024
0.026
0.656
0.036
SSRY14
0.08
0.083
0.086
0.002
0.003
0.086
0.029
SSRY15
0.689
0.79
0.806
0.015
0.017
0.807
0.019
SSRY15
0.072
0.595
0.632
0.037
0.041
0.636
0.059
SSRY16
0.528
0.658
0.668
0.01
0.011
0.669
0.014
SSRY16
0.258
0.316
0.321
0.004
0.005
0.321
0.013
SSRY17
0.544
0.555
0.591
0.037
0.041
0.596
0.062
SSRY17
0.845
0.726
0.776
0.05
0.056
0.781
0.065
SSRY18
0.672
0.532
0.535
0.003
0.004
0.536
0.006
SSRY18
0.605
0.705
0.761
0.055
0.062
0.767
0.073
Overall
0.523
0.55
0.573
0.023
0.026
0.575
Ho Average observed heterozygosity within country
Ht Total Heterozygosity in the entire data set
Hs Gene diversity within country averaged over the entire data set
Dst Average gene diversity between populations
Gst Coefficient of gene differentiation.
0.04
0.23
0.015
0.01
0.001
0.037
0.043
0.058
-0.01
0.02
0.106
0.084
0.025
0.098
0.076
0.031
0.031
0.024
0.027
-0.041
0.005
0.011
0.036
0.009
0.04
0.032
0.021
0.065
0.016
0.015
0.068
0.071
0.007
0.08
0.045
Genetic distances between all pairs of individual accessions was calculated by
the 1-proportion of shared alleles (1-PSA) and presented graphically by a
principal coordinate analysis (PCA) (Fig1). The PC1 and PC2 accounted for 26%
and 16% of the total variance respectively. The PCA shows loose clustering of
the land races by region but of note is the sub-structure of some land races
Project IP3: improving cassava for the developing world
Output 8-32
from Northern Ghana. A similar sub-structure in accessions from Nigeria and
from Tanzania was observed in earlier studies. The presence of a defined substructure in the genetic relationship of cassava land races from Africa appears
to be a common feature of cassava germplasm in a number of countries but it is
yet to be understood the underlying factors for the groupings. UPGMA cluster
analysis of FST estimate of genetic differentiation amongst land races was able to
group the land races into loose clusters according to agro-ecologies, with a
group of genotypes from the Northern region sub-structure being the most
differentiated (Fig 2). (Fregene et al.2002). At least 63 duplicates or closely
related accessions were identified in the collection.
Output 8-33
2003 Annual Report
PCA of geneti c distance s (PS A)
30
25
20
15
AS H
BA
C/ A
CR
Gt .Ac cra
No rth ern Reg ion
Vo lta regi on
ER
NS
10
5
0
Fig. 2: Principal component analysis
of genetic distances, based 1- proportion of
-1 5
-10
-5
5
10
15
20
25
shared alleles,
between
cassava0 accessions
from
Ghana.
-2 0
-5
-1 0
-1 5
PC1
Ashanti
Central
Brong Ahafo
Eastern
Coast Savannah
Volta
Greater Accra
Northern IIII
Northern I
0.02
0.06
0.11
0.15
Coefficient
Figure 2: UPGMA tree of pair-wise FST data calculated between samples from
different regions of Ghana.
One of the principal reasons for this study was to assess genetic diversity in
cassava land races as a first step to delineating heterotic pools for a more
systematic improvement of combining ability via recurrent reciprocal selection.
Project IP3: improving cassava for the developing world
Output 8-34
Based upon clusterings obtained above, genotyopes representative of the
clusters were selected as parents for a diallel experiment to search for heterotic
patterns and established in a crossing block at the CRI experimental station in
Wenchi. Genotypes from other parts of Latin America, for example Guatemala,
that cluster away from African accessions were also shipped from CIAT to
Ghana for the study of heterotic patterns. The Latin American genotypes were
multiplied, hardened and planted in the same crossing block where the land
races are currently planted.
Conclusions
A total of 320 land races were collected and established at the University of
Legon and IITA, Ibadan. Genetic diversity and differentiation in the collection
was assessed using SSR markers. Discovery of a sub group within the land
races, as observed from PCA of genetic distances (proportion of shared alleles)
and UPGMA of pairwise FST between the different regions, that may represent
heterotic groups. To test heterotic patterns present within the collection or
between the collection and Latin American accessions, selected genotypes were
planted in a crossing block for making genetic crosses.
References
Adams, C.D. 1957. Activities of Danish Botanists in Guinea 1738-1850.
Transactions of the Historical Society of Ghana III. Part 1.
Al-Hassan, R. 1989. Cassava in the Economy of Ghana. In : Status of Cassava
research in Africa, COSCA working paper No. 3, Eds, F. I. Nweke, J.
Lynam and C. Y. Prudencio, International Institute of Tropical Agriculture,
Ibadan, Nigeria.
.Dapaah, S. K. (1996). The way forward for accelerated agricultural growth and
development. A paper presented to the Government of Ghana on behalf of
the Ministry of Food and Agriculture. 6pp.
Fregene M., Suarez M., Mkumbira J. , Kulembeka H. , Ndedya E. , Kulaya A. ,
Mitchel S. Gullberg U. , Rosling H., Dixon A., Kresovich S. (2003) Simple
Sequence Repeat (SSR) Diversity of Cassava (Manihot esculenta Crantz)
Landraces: Genetic Diversity and Differentiation in a Predominatly
Asexually Propagated Crop Theor Appl Genet 107:1083-1093
Goudet J, 1995. FSTAT (vers. 1.2): a computer program to calculate F-statistics.
J. Hered. 86: 485-486.
.Jones, W. O. 1959. Manioc in Africa. Stanford University Press 1959.
Korang-Amoakoh, S. , Cudjoe, R. A. and Adams, E. 1987. Biological Control of
cassava pests in Ghana. Prospects for the integration of other strategies.
In
Ministry of Agriculture. 1990. Medium Term Agricultural Development
Programme (MTADP) 1991 - 2000: An agenda for sustained agricultural
growth and development 1991 - 2000. Ministry of Agriculture, Accra.
.National Agricultural Research Strategic Plan, Ghana, 1994. Final Report.
181pp.
National Cassava Task Force. 1996. Final Report on the promotion of cassava
production, processing and marketing in Ghana. 39pp. with appendices.
SAS Institute, Inc., 1995. JMP (version 3.1). SAS Inst. Inc., Cary, NC.
Shull G.F. 1952. Beginnings of the heterosis concept. P 14-48. In: JW. Gowen
(ed) Heterosis, Iowa Sate College Press, Ames.
Output 8-35
2003 Annual Report
Vekemans X. Lefebvre C., 1997 On the evolution of heavy-metal tolerant
populations in Armeria maritima: evidence from allozyme variation and
reproductive barriers. J. Evol. Biol., 10: 175-19
Activity 8.10 Characterization of genetic diversity:
Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local
Cassava Varieties from Uganda
Collaborators: E. Kizito, U. Gullberg (Swedish Agricultural University,
Uppsala); A. Bua, J. Omara, (Cassava program, Namunlonge Agricultural and
Animal Research Institute, Kampala, Uganda); M. Egwang (Medical Biotech
Laboratories, Kampala, Uganda); W. Castelblanco, J. Gutierrez, C. Buitrago, M.
Fregene (CIAT)
Background
Cassava is grown mainly for its starchy roots and the leaves and forms a staple
by an estimated half of Uganda’s population (Bua, pers.comm, 2001). Cassava
was introduced into Uganda after 1850 by Europeans and Arabs (Langlands,
1970). Because of its excellent adaptability to erratic rainfall and low fertility
soils, it became a major dietary staple, a famine reserve crop and a source of
cash to many small-scale farming communities. The impact of a cassava
mosaic disease (CMD) epidemic in Uganda in the late 80’s and early 90’s on
genetic diversity grown by small farmers is thought to be significant. In a survey
carried out in 2000 in Mukono, Soroti and Apac districts in Uganda, the impact
of the CMD epidemic on cassava diversity was clearly seen via the loss of
previously well-known varieties (Kizito and Gullberg 2002, unpublished data).
But it was also observed that additional genetic variability had arisen from the
use of volunteer seedlings by some farmers in Mukono district. Traditional
farming systems of slash and burn followed by 3-15 years of fallow have been
known to encourage the allogamous nature of cassava producing a large pool of
volunteer seedlings that natural and human selection acts on to produce new
varieties which maintains a high diversity, for instance in Tanzania (Fregene et
al, 2003). We assessed the genetic diversity and differentiation based on SSR
markers of landraces from all over Uganda and a small subsection from Latin
America and other African countries. The objectives of this study were to:
assess the genetic diversity and differentiation of cultivars within and between
different districts in Uganda; also to determine how the Uganda cassava
diversity compares with the total genetic diversity of species within Africa and
the cassava collection after the CMD epidemic.
Methodology
A total of 257 local cassava varieties were collected accessions September
through to December 2002 in 17 districts that lie between latitudes N02o 121
and S 00o 441, longitudes E029o 561 and E034o 211, and altitudes of 4451ft and
2177ft above sea level in Uganda. Three counties on average in each district
were selected at random and fields with mature crop were sampled from every
7-10 km along the roads that traversed each of the counties. In each farmer's
field the different varieties were identified according to their morphological
characteristics as well as by the name given by the farmer. Where no single
Project IP3: improving cassava for the developing world
Output 8-36
variety dominated, plants of the co-dominant varieties were sampled, labelled to
be planted and maintained in NAARI. Another 20 accessions, representatives of
diversity, were selected from Tanzania, and 18 accessions South America from a
previous study of cassava accessions from Tanzania and the neo-tropics
(Fregene et al. 2003). In addition to this, 20 from the Ghanaian germplasm
bank, 20 from Nigeria and 20 from Guatemala was included. In all, 350
accessions were studied. DNA isolation was from young leaf tissue harvested
by the CTAB method (Doyle & Doyle, 1987) at Med Biotech Laboratories,
Kampala. A subset of 36 SSR markers with high polymorphism information
content (PIC) routinely used for diversity analysis in cassava was source of
markers, the 36 markers have been chosen to represent a wide coverage of the
genome. PCR amplification, automated gel analysis and date collection were as
descried by Fregene et al (2003). Genetic distance, based upon the proportion
of shared alleles (PSA), obtained from the raw allele size data using the
computer microsat (Minch 1993,
http://www.lotka.stanford.edu/microsat.html). Distances between the
accessions were subjected to principal component analysis (PCA) using JMP
(SAS Institute 1995) to obtain a structure of relationship between the land
races. Parameters of genetic diversity and differentiation were calculated from
allele data using the computer packages GENSURVEY (Vekeman et al 1997)
and FSTAT (Goudet 1990).
Results
A total of 35 of 36 SSR loci was used to provide estimates for the genetic
diversity and differentiation of 350 accessions of cassava accessions from
Uganda and eight other countries, one eliminated for being monomorhphic. The
number of alleles observed at each locus in the data set ranged from 2 to 12
alleles per locus over the 35 loci. The average gene diversity, He, for the entire
350 accessions was more than half 0.5649+ 0.0698, average gene diversity of
accessions from Uganda was 0.5530 (Table 1). However, only 1% (Gst=0.0192+
0.0511) of within district variation in Uganda was due to differentiation. On the
other hand, 10% (Gst=0.1078±0.0502) of the overall heterozygosity (Ht=0.6305+
0.1696) in all the country accessions could be attributed to differentiation
among the samples from both Africa and Latin America, revealing that
germplasm within Uganda is at the moment quite uniform. When accessions
from Uganda were divided according to districts, the least values for average
gene diversity were observed in Lira and Luweero districts, 0.4011 and 0. 4219
respectively, while Kasese stands out with the highest value of 0.6208. This
affirms earlier findings of higher varietal diversity in the western and
southwestern districts of Uganda as opposed to those in the eastern districts
(Otim-Nape et al, 2001). The relatively high level of genetic diversity observed
on the whole in this study is unexpected considering Uganda has reported two
major cassava mosaic disease (CMD) epidemics before. The most recent CMD
epidemic being in the last 10-15 years, since 1988, affected drastically the
cassava varietal composition and saw a decrease in area planted to cassava at
its peak between 1990-1994 (Otim-Nape et al, 2001). This finding continues to
demonstrate the fact of active involvement of Ugandan farmers in continuous
testing and the adaptation of new planting materials to their unique situations.
The importance of volunteer seedlings in the dynamics of cassava diversity has
Output 8-37
2003 Annual Report
been demonstrated in traditional farming systems of the Makushi Indians from
Guyana (Elías et al, 2001)
Genetic distance between accessions based on 1- proportion of shared alleles
(1-PSA) was calculated and presented graphically by a principle coordinate
analysis (Fig. 1). The PC1 and PC2 components accounts for about 39% and
10% of the total variance respectively. The PCA reflects a loose separation
between accessions from Africa and the Neotropics as has been shown in earlier
studies (Fregene et al. 2003). No distinct substructure was found amongst the
Ugandan accessions except for accessions from Nakasongola district. Substructures in diversity have been reported for earlier studies for accessions from
Ghana, Tanzania and Nigeria. Genetic differentiation averaged over all loci
estimated by Fst (theta) was 0.103+ 0.009 (jackknifing) and 0.082+ 0.126
calculated by bootstrapping at 99% confidence interval (table 4). This agrees
with previous diversity studies in Tanzania (Fregene et al, 2003). Pair wise
calculations of Fst (theta) over all loci between pairs of country landraces and
Uganda also revealed lower differentiation between African countries compared
with Latin American countries, the lowest being between Ghana and Uganda
(0.039) and the highest being between a group of accessions from Guatemala
group and Nigeria (0.2631)
A dendogram of landraces for UPGMA of pair wise Fst estimates separates the
African from Neotropical accessions with the group of accessions from
Guatemala being the most genetically differentiated (Fig.2). These results agree
with some previous studies on which a high differentiation has been observed
among certain cassava groups cultivated in Guatemala and those in other parts
of Latin America and África (Fregene et al, 2003; CIAT, 2003). Of particular
interest to cassava breeding programs is the group (G1) from Guatemala,
Accessions from G1 from Guatemala is a representation of the region East and
South of Guatemala and it may represent a heterotic group based on
differences in allele frquncies. The phenomenon of heterosis or hybrid vigor is
an important factor in improvement of heterozygous crops such as cassava, and
in cases like the corn, the patterns found in these populations at the beginning
of the XX century have been the base of a very sucecssful industry of hybrid
corn, elevating productivity by more than 500% (Shull, 1952; Tomes, 1998).
Conclusions
x Characterization of the cassava landraces of Uganda with 35 markers
and estimation of genetic diversity and differentiation
x Assessment of the genetic relationship between different countries of
Africa and Latin America.
x Evidence of high differentiation between some land races fron Guatemala
and the rest of the countries that may represent heterotic pools.
x Evidence of low impact in cassava diversity after of epidemic of CMD and
low structure in cassava landraces of Uganda.
Project IP3: improving cassava for the developing world
Output 8-38
Table 1: Genetic diversity within groups of cassava landraces classified
according to country of origin. Standard deviations (SD) were estimated by
jackknifing over loci (200 replications). Ht, Hs, Dst, and Gsta are given over loci
and over groups (country populations).
Population Sample No.of No.of Percent Mean no. Mean no. Hoc
Size loci pol. Of polb. Alleles alleles/
/locus polb.locus
Locib loci
UGANDA
0.5468
COLOMBIA
5
BRASIL
0.6304
PERU
3
GUATEMALA1
0.4219
GUATEMALA2
0.5906
TANZANIA
19
NIGERIA
20
GHANA
0.5694
He-p
198
35
33
94.3
5.2
35
3
33
34
94.3
33
3.3
97.1
3.4 0.5081 0.5363 0.5963
2.8
2.8 0.5735 0.5069
35
7
33
35
94.3
33
2.7
94.3
2.8 0.5810 0.5218 0.6619
2.5
2.6 0.5290 0.3908
11
35
34
97.1
3.8
35
35
19
32
33
35
91.4
94.3
33
3.9
3.9
94.3
4.1 0.5658 0.5386 0.5536
4.0 0.5002 0.5002 0.5131
4.2
4.4 0.5429 0.5542
94.59 3.59
1.70 0.86
3.71 0.5423 0.5176 0.5649
0.89 0.0285 0.0519 0.0698
Mean
Std
Ht
Mean
Std
95%CI
99%CI
Hed
0.6305
0.1696
0.5713
0.6827
Hs
Dst
Gst
0.5635
0.1606
0.5083
0.6135
0.0670
0.0332
0.0566
0.0767
e
5.4 0.5530 0.5454
3.9 0.5274 0.5640
0.1078
0.0502
0.0916
0.1235
Ht= total heterozygosity in the entire data set; Hs= heterozygosity within
country averaged over the entire data set; Dst= average gene diversity between
populations; Gst= coefficient of gene differentiation.
b pol. =polymorphic
cH = average observed heterozygosity within country
o
dH = average expected heterozygosity within country
e
eH
e-p= average expected heterozygosity within country corrected for small
sample sizes (Nei, 1978)
a
Output 8-39
2003 Annual Report
15
10
UGANDA
TANZANIA
5
PC2(10%)
NIGERIA
GHANA
0
COLOMBIA
BRAZIL
-5
PERU
GUATEMALA 1
-10
GUATEMALA 2
-15
-20
-15
-10
-5
0
5
10
15
20
25
30
35
PC1(39%)
Figure 4. PCA of Genetic Distance (1-PSA) of local cassava varieties within the different districts in Uganda
Project IP3: improving cassava for the developing world
Output 8-40
UGANDA
GHANA
TANZANIA
NIGERIA
COLOMBIA
BRASIL
PERU
G2
G1
0.01
0.07
0.14
Coefficient
0.20
0.26
Figure 2. Unweighted pair group method with arithmetic averaging (UPGMA)
dendogram of the pairwise fixation index (Fst) between cassava
landraces, grouped by country and by source.
Output 8-41
2003 Annual Report
References
CIAT (2003) Assesment of simple sequence repeat Diversity of cassava
landraces in Africa y Latin America: Progress report of the Ghana, malawi,
Uganda, Guatemala, and brazil country studies. Report of activities by the
cassava molecular diversity network (MOLCAS)
Doyle JJ and Doyle JL (1987). A rapid DNA isolation procedure for small
quantities of fresh leaf tissue. Phytochemical Bulletin. 19:11-15
Elias M, Penet L, Vindry P, Mckey D, Panaud O, Robert T (2001) Unmanaged
sexual reproduction and the dynamics of genetic diversity of a vegetatively
propagated crop plant, cassava (Manihot esculenta Crantz), in a traditional
farming system. Molecular Ecology 10:1895-1907
Fregene MA, Suarez M. Mkumbira J, Kulembeka H., Ndedya E, Kulaya A,
Mitchel S, Gullberg U, Rosling H, Dixon AGO, Dean R, and Kresovich S
(2003) Simple sequence repeat marker diversity in cassava landraces:
Genetic diversity and differentiation in an asexually propagated crop.
Theor. Appl. Genet. 107:1083-1093
Goudet J. (1995) FSTAT (Version 1.2): A computer program to calculate Fstatistics. J. Hered 86:485-48
Langlands BW (1966) Cassava in Uganda 1860-1920. Uganda Journal 30:211218.
Mba REC, Stephenson P, Edwards K, Melzer S, Mkumbira J, Gullberg U, Apel
K, Gale M, Tohme J, Fregene MA (2001) Simple Sequence Repeat (SSR)
markers survey of the cassava (Manihot esculenta Crantz) genome:
towards an SSR-based molecular genetic map of cassava. Theor Appl
Genet:21-31
Otim-Nape GW, Alicai T, Thresh M (2001) Changes in the incidence and severity
of Cassava mosaic virus disease, varietal diversity and cassava production
in Uganda. Ann. Appl. Biol 138:313-327.
Rohlf FJ (1993) NTSYS-PC numerical taxonomy and multivariate analysis
system. Version 1.8 Exeter Publ., Setauket, NY.
Shull, G.F. 1952. Beginnings of the heterosis concept. In: JW Gowen. Heterosis.
Iowa State College Press, Ames, Iowa. Pp 14-48
Tomes D (1998) Heterosis: performance stability, adaptability to changing
technology and the foundation of agriculture as a business. In: Lamkey K,
Staub J E (eds). Concepts and breeding of heterosis in crop plants. CSSA
Special Publication Number 25. Crop Science Society of America, Madison,
WI.
Vekemans X., Lefebvre C (1997) On the evolution of heavy-metal tolerant
populations in Armeria maritima: evidence from allozyme variation and
reproducive barriers. J. Evol.Biol 10:175-191
Project IP3: improving cassava for the developing world
Output 8-42
Activity 8.11 Characterization of genetic diversity:
Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local
Cassava Varieties from Sierra Leone
Collaborators: A. Dixon, B. Raji (IITA, Nigeria); J. Marin, C. Ospina, C.
Buitrago, M.Fregene (CIAT)
Background
Cassava is an important staplePg crop in Sierra Leone, second only to rice in
importance. Civil strife in the recent past has lead to an erosion of genetic
resources and loss of some very valuable germplasm. A collection was
conducted by the National Program in collaboration with IITA in the Southern
and Eastern provinces where cassava is or more importance to the populace, to
safe guard local varieties for the future. The germplasm collection from Sierra
Leone is being held at IITA Ibadan pending when peace fully returns to the
country and the collection can be returned to the National root crop program.
The collection was analyzed with SSR markers as part of MOLCAS efforts to
characterize diversity found in local African varieties compared to what exists in
Latin America.
Methodology
Forty villages in the Eastern and Southern province of Sierra Leone were visited
in 2001 for the collection of cassava germplasm. In each village farmers were
invited to share their most important varieties, between 3 and 4 varieties were
collected from each village. The stakes of all accessions collected was
established in the field at IITA. DNA was isolated from leaf tissues using a DNA
isolation kit (QIAGEN Gmbh) at IITA and carried to CIAT for SSR analysis. A
subset of 36 SSR markers with high polymorphism information content (PIC)
routinely used for diversity analysis in cassava was source of markers. The
markers have been chosen to represent a wide coverage of the genome. PCR
amplification, automated gel analysis and date collection were as descried by
Fregene et al (2003). Genetic distance, based upon the proportion of shared
alleles (PSA), obtained from the raw allele size data using the computer
microsat (Minch 1993, http://www.lotka.stanford.edu/microsat.html).
Distances between the accessions were subjected to principal component
analysis (PCA) using JMP (SAS Institute 1995) to obtain a structure of
relationship between the land races. Parameters of genetic diversity and
differentiation were calculated from allele data using the computer packages
GENSURVEY (Vekeman et al 1997) and FSTAT (Goudet 1990).
Results
A total of 127 local cassava varieties were collected in from 40 villages. Thirty
three SSR markers were analyzed in 98 accessions from Sierra Leone, the
remainder accessions were not available at the time of SSR analysis. The
average number of alleles for each locus was close to 5 and is similar to that
found for a study of land races from Nigeria, Tanzania and 7 Neo-tropical
countries (Table 1). Genetic distances between all pairs of individual accessions
was calculated by the 1-proportion of shared alleles (1-PSA) and presented
graphically by a principal coordinate analysis (PCA) (Fig1). The PCA shows a
sub-structure in diversity of the local varieties as have been observed in other
collections from Africa. The presence of a defined sub-structure in the genetic
Output 8-43
2003 Annual Report
relationship of cassava land races from Africa appears to be a common feature
of cassava germplasm in a number of countries but it is yet to be understood
the underlying factors for the groupings. The sub structure observed in the
PCA was the basis of further analysis of genetic diversity parameters carried
out.
Average number of alleles per SSR locus in the collection was roughly 4 and
average gene diversity, was 0.5749r0.0690 it is comparable to that found in
previous studies in several African countries. Genetic differentiation, as
estimated by FST (theta), between the groups ranged from 0.088 to 0.140 which
is low to moderate differentiation. The results found here support previous
findings that agricultural practices and the allogamous nature of cassava
produces a large pool of volunteer seedlings that natural and human selection
acts upon to maintain a high level of diversity and low differentiation (Doyle et
al. 2001; Fregene et al. 2002).
PCA of Genetic Distances from SSR Marker data of Cassava Accessions from Sierra Leone
6
4
2
0
-10
-5
0
5
10
15
20
GRUPO A
-2
GRUPO B
GRUPO C
-4
-6
-8
-10
Figure 1. PCA of Genetic Distance based 1-poportion of shared alleled (PSA) of
local varieties from the Southern and Eastern districts of Sierra Leone
Project IP3: improving cassava for the developing world
Output 8-44
Table 1 Genetic diversity within groups of cassava landraces classified
according to country of origin. Standard deviations (SD) were estimated
by jackknifing over loci (200 replications). Ht, Hs, Dst, and Gsta are given
over loci and over groups (country populations).
Population
A
B
C
n #loc. #loc_P
60 33
9 33
16 33
mean 3 pop.
std
32
33
30
PLP
K
K_P
HO_p
HE_p
HEc_p
Fis_p
97.0
100.0
90.9
4.2
4.3
2.9
4.3
4.3
3.1
0.6425
0.6747
0.7355
0.5639
0.6100
0.4930
0.5686
0.6469
0.5092
-0.1391
-0.0607
-0.4542
95.96
4.63
3.83
0.80
3.93
0.72
0.6843
0.0472
0.5556
0.0589
0.5749
0.0690
-0.2180
0.2083
Ht= total heterozygosity in the entire data set; Hs= heterozygosity within
country averaged over the entire data set; Dst= average gene diversity between
populations; Gst= coefficient of gene differentiation.
b pol. =polymorphic
cH = average observed heterozygosity within country
o
dH = average expected heterozygosity within country
e
eH
e-p= average expected heterozygosity within country corrected for small
sample sizes (Nei, 1978)
a
Figure 1. Silver stained acrylamide gel picture of the Sierra Leonean
accessions analyzed with SSR marker SSRY 82
Conclusions
The SSR analysis of genetic diversity of local cassava varieties from Sierra Leone
reveals the same pattern of high average gene diversity, low differentiation, and
a pronounced sub-structure. Future perspectives include tracing the lineages
of some of the lines in the 3 clusters found to get a better idea of the impact of
germplasm development/introduction and adoption in the country.
References
Fregene MA, Suarez M. Mkumbira J, Kulembeka H., Ndedya E, Kulaya A,
Mitchel S, Gullberg U, Rosling H, Dixon AGO, Dean R, and Kresovich S
(2003) Simple sequence repeat marker diversity in cassava landraces:
Genetic diversity and differentiation in an asexually propagated crop.
Theor. Appl. Genet. 107:1083-1093
Goudet J. (1995) FSTAT (Version 1.2): A computer program to calculate Fstatistics. J. Hered 86:485-48
Output 8-45
2003 Annual Report
Mba REC, Stephenson P, Edwards K, Melzer S, Mkumbira J, Gullberg U, Apel
K, Gale M, Tohme J, Fregene MA (2001) Simple Sequence Repeat (SSR)
markers survey of the cassava (Manihot esculenta Crantz) genome:
towards an SSR-based molecular genetic map of cassava. Theor Appl
Genet:21-31
Rohlf FJ (1993) NTSYS-PC numerical taxonomy and multivariate analysis
system. Version 1.8 Exeter Publ., Setauket, NY.
Vekemans X., Lefebvre C (1997) On the evolution of heavy-metal tolerant
populations in Armeria maritima: evidence from allozyme variation and
reproducive barriers. J. Evol.Biol 10:175-191
Activity 8.12 Characterization of genetic diversity:
Simple Sequence Repeat (SSR) Assessment of Genetic Diversity of Local
Cassava Varieties from Cuba
Collaborators: Yoel Beovides (INIVIT, Cuba), Edgar Barrera, Janneth P.
Gutiérrez, Charles Buitrago, Jaime Marin, Martin Fregene (CIAT).
Funding: Cassava Biotechnology Network (CBN)
Background
Cassava is an important crop of modern tropical economies and an attractive
one for millions off resource poor farmers found in the tropics (Best and Henry
1994). Recently a second center of diversity have been postulated in Central
America based on SSR markers (Monte et al. 2003), in addition to the one in
Brazil (Olsen and Schaal 1999). However, the potential of diversity in the
second center, particularly in the Caribbean is not well documented. Recently
SSR markers have been utilized to study the diversity of cassava from different
countries (Fregene et. al. 2003). SSR markers are particularly attractive to
study genetic diversity due to their abundance in plant genomes, high levels of
polymorphisms and adaptability to automation. These studies revealed a high
amount of diversity in accessions from several neotropical countries, a low level
of genetic differentiation between country samples, with the exception of a
group of accessions from Guatemala, and sub-structure in diversity of
accessions from some African countries.
SSR markers can contribute to a better understanding of genetic diversity
present in a collection of local cassava varieties held in Cuba to permit a more
rational conservation and use of diversity on the island. We present here
preliminary results of SSR study of genetic diversity of cassava from Cuba
compared to a sub-set of accessions from Africa, South and Central America
Methodology
A total of 94 accessions were selected from a collection of cassava held at INIVIT
in Cuba, selection criteria was the economic importance and origin in Cuba. A
set of 54 clones from Africa and the Neotropics, 12 from Nigeria, 10 from
Tanzania, 12 from Guatemala, and 20 from South America, representative of a
large set of accessions from these countries used in previous SSR studies
(Fregene et al 2003) were included for comparisons. A third set of 13 improved
genotypes from CIAT with traits of agronomic interest were added. DNA from
Project IP3: improving cassava for the developing world
Output 8-46
all accessions was obtained using the Dellaporta et al. method (1983).
Concentration and quality of the DNA was checked by flourometry and agarose
gel electrophoresis respectively. The DNA samples were diluted to a working
concentration of 10ng/ul for subsequent PCR amplification.
PCR amplification, automated gel analysis and date collection were as descried
by Fregene et al (2003) and Mba et al. (2003). Statistical analysis to be
conducted include calculations of pair-wise genetic distance, based upon the
proportion of shared alleles (PSA), using the computer microsat (Minch 1993,
http://www.lotka.stanford.edu/microsat.html). Distances between the
accessions will be subjected to principal component analysis (PCA) using JMP
(SAS Institute 1995) to obtain a structure of relationship between the land
races. Other analysis are estimation of parameters of genetic diversity and
differentiation, calculated from the raw SSR allele data using the computer
packages GENSURVEY (Vekeman et al 1997) and FSTAT (Goudet 1990).
Results
A total of 15 SSR markers have been analyzed until now in all accessions. A
high number of alleles and level of polymorphisms have been observed in all
SSR markers analyzed until date (Fig 1). Evaluation of the remainder 21 SSR
markers is ongoing and will be completed by October. Also ongoing is reading
of the gels and determination of allele sizes using the program Quantity one.
This raw SSR data will be used for subsequent analysis as described in the
methodology.
Figure 1 Polyacrilamide gel of a PCR amplification of cassava accessions from
Cuba, Nigeria, Tanzania, Guatemala, and South America using the SSR
primer SSRY51.
Conclusions
A SSR study of genetic diversity of cassava in Cuba has been initiated. The
outcome of the study is expected to provide insights on the significance of the
Caribbean region as a center of diversity for cassava and to guide rational
conservation and plant improvement efforts.
References
Monte L. Azudia C, D. Debouck y M. Fregene. 2003. Simple Sequence Repeat
(SSR) Marker Assesment of Genetic Diversity of Cassava Land Races
from Guatemala. (in preparation)
Fregene M, M. Suárez, J. Mkumbira, H. Kulembeka, E. Ndedya, A. Kulaya, S.
Mitchel, U. Gullberg, A. G. O. Dixon, R. Dean y s. Kresovich. 2003.
Simple sequence repeats marker diversity in cassava landraces: genetic
Output 8-47
2003 Annual Report
diversity and differenciation in an asexually propagated crop.
Theoretical and Applied Genetics 107:1083-1093.
Mba, R.E.C., Stephenson, P., Edwards, K., Melzer, S., Mkumbira, J., Gullberg,
U., Apel, K., Gale, M., Tohme, J. and Fregene, M. (2001) Simple
Sequence Repeat (SSR) Markers Survey of the cassava (Manihot
esculenta Crantz) Genome: Towards an SSSR-Bassed Molecular
Genetic Map of Cassava. Theoretical and Applied Genetics. 102: 21-31.
Olsen K and Schaal B. 1999. Evidence on the origin of cassava:
phylogeography of Manihot esculenta. Proceedings of the National Academy
of Science 96: 5586-5591.
Activity 8.13 New Collaborative Arrangements, Networks, Databases,
Training and Workshops
Report of the Second Tri-Annual Workshop of the Molecular Genetic
Diversity Network of Cassava (MOLCAS)
Collaborators: H. Rosling (Karolinska Institute, Sweden), M. Akerbolm (IPICs,
Sweden), E. Kizito, L. Chiwona-Karltun, U. Gullberg (SLU, Sweden), J. Jiggins
(WAU, Netherlands), G. Muhlen (U. de Rondonia, Brazil), E. Okai (CRI, Ghana),
M.Fregene (CIAT)
Background
Understanding the existing genetic diversity and its distribution within and
among individuals, populations, species and gene pools is crucial for an
efficient management and use of germplasm collections. The large amounts of
cassava genetic resources held by farmers have been demonstrated to represent
a critical resource for the future productivity and stability of production of
cassava. A highly successful breeding program at the International Center for
Tropical Agriculture (CIAT) in the early 1970s began with a great initial
germplasm variation, some 2,218 local clones collected from Colombia, Brazil,
Venezuela, and Peru (KAWANO 2003). That program achieved more than 100%
increase in fresh root yield and more than 20% in dry matter content by the
early 1980s. The improved varieties from Latin America combined with local
varieties in South East Asia formed the basis of a very successful breeding
program for the South East Asia Sub-continent, particularly Thailand. The new
Thai cultivars are now planted on more than one million hectares in Thailand
alone and the economic benefits from the increased productivity is in the order
of one billion US dollars and the rural communities in some of the poorest parts
of Asia captured a large proportion of these economic benefits (Kawano 2003)..
The molecular diversity network of cassava comprises of scientists drawn from
institutes in Africa, Latin America, Europe and the USA. In its four years of
existence, MOLCAS has studied diversity of local varieties in Tanzania, Nigeria,
Ghana, Uganda, Guatemala, , Peru, Sierra Leone and a subset of the World
germplasm bank held at CIAT (MOLCAS 2003). Studies are ongoing of diversity
in Cuba and Brazil. The ultimate aim of these studies is to identify and exploit
useful variability for increased crop productivity and value addition. Between
September 3 and 5 this year, members of the workshop from Brazil, Ghana,
Uganda, Netherlands, and Colombia got together in Uppsala with network
Project IP3: improving cassava for the developing world
Output 8-48
members based in Sweden and the director of the IPICs donor for the 2nd
MOLCAS workshop to discuss progress and draw priorities for the next
application to IPICs for the period 2004 – 2006. The meeting was held at the
Plant biology department of SLU, and was well attended by members of the
department including the head of department, Prof Per Bergmann.
Report of the 2nd MOLCAS meeting
The first day of the meeting saw presentations on SSR assessment of local
cassava varieties from Ghana, Brazil, Nigeria, Sierra Leone, Guatemala, and
Uganda. During second day, members of the plant biology department at SLU
working on cassava presented a summary of their findings. They include
cloning and transformation of starch biosynthetic genes, molecular biology of
the cassava mosaic virus, and breeding cassava for small holders. A discussion
was held in the afternoon of the second day on how an application for the third
phase of BIOEARN, the Sida funded East African training network could be
streamlined with activities being carried out in Africa. Of particular interest
was how the wide gap between upstream biotechnology could be applied to
secure the cassava crop as a food security crop and also provide improved
livelihood through value addition. It was decided that BIOEARN members draw
up a draft of their application and circulate it to MOLCAS members for
comments and their inputs.
The third day, Prof Rosling presented results of an epidemiology of the Konzo
disease in Mozambique conducted in the 1970s as an example of a proper
approach to sampling genetic diversity. Two presentations were also made by
Prof Janice Jiggins of WAU, Netherlands and Dr L. Chiwona Karltun on a
farmer participatory research project taking place at the moment in Malawi.
The afternoon of the third day was spent in a priority setting exercise for
projects to be conducted in the next phase of MOLCAS. From a total of 10
project, the following were prioritized for the next phase beginning 2004 until:
1) Evaluation of highly differentiated gene pools in cassava for heterosis (Ghana)
Highly differentiated cassava gene pools may represent heterotic pools. The
study seeks to evaluate these accessions for heterosis or hybrid vigor by making
genetic crosses between and within representative members of the clusters.
The study will be conducted in CRI, Kumasi, Ghana. Latin American
germplasms were shipped to Ghana as tissue culture plants, hardened, and
established in a crossing block along with African genotypes.
2) Tracing the lineages of local African Cassava varieties: towards a better
understanding of sub-structures in African cassava gene pools (Uganda and
Sweden)
To better understand the diversity sub structures in African local varieties,
records and germplasm from breeding programs that existed in East Africa in
the early part of the 19th century will be examined and genotypes compared to
modern day local varieties. This study will be carried out by SLU in
collaboration with researchers from NARO, Uganda.
3) A search for useful variability in local cassava land races based upon the
structure of SSR marker diversity analysis (Nigeria)
Further evaluation of some genotypes that have been observed during the SSR
marker study to have novel characteristics, for example novel starch quality,
Output 8-49
2003 Annual Report
will be carried out to confirm the earlier observations. Should they be
confirmed, genetic crosses will be made to attempt a transfer into improved and
other local varieties and also for inheritance studies
4) A comparison of clustering of cassava germplasm from Brazil and Malawi
based upon bitter or sweet taste of the roots. (Brazil, Malawi, Sweden)
Studies of genetic diversity in Brazil revealed a clustering along the lines of
taste. A similar result was also obtained for cassava germplasm from Malawi.
The objective of this study is to combine both data sets and how they cluster.
Dr Gilda Muhlen of University of Rondonia, Brazil will be responsible for the
study in collaboration with SLU and Malawian researchers.
The meeting was attended by Prof H. Rosling of the Karolinska Institute,
Stockholm; Dr M. Akerbolm, Ms L. Sjobolm of IPICs, Uppsala; Dr JaniceMs E.
Kizito, Dr L. Chiwona-Karltun, Prof U. Gullberg, Prof P. Bergmann (head of
department), Prof Chris Jassen, Dr Roger Anderson, Mr Yona Baguma, of the
plant biology department, SLU, Uppsala; Dr G. Muhlen of the Universidad de
Rondonia, Brazil; Ms E. Okai of CRI, Ghana, and M.Fregene of CIAT.
References
Kawano. 2003. Thirty Years of Cassava Breeding: Biological, and Social
Factors for Success. Crop Science 43:1325-1335.
Activity 8.14 New Collaborative Arrangements, Networks, Databases,
Training and Workshops
Progress in the Development of a Web Accessible Data Base for SSR Marker
Assessment of Diversity of Local Cassava Varieties
Collaborators: A. Akerbolm (IPICs, Sweden); C. Buitrago, F. Rojas, M. Fregene
(CIAT)
Background
The web accessible database of the molecular diversity network of cassava
(MOLCAS) was set up to make the results of its SSR marker assessment of local
cassava diversity in selected countries of Africa and Latin America widely
available. The database is updated as completed country studies become
available. A total of 8 country studies, 5 in Africa and 3 in Latin America, have
now been completed by the network. The database has been updated with 2
additional country studies, bringing the total on the website now to 5, namely
Tanzania, Peru, Nigeria, Guatemala and Ghana. A very rewarding outcome of
the development of the MOLCAS web is the high number of visits the web has
recorded since its inception last year. We describe here progress made this
year.
Results
The MOLCAS web-based data base (http://www.ciat.cgiar.org/molcas and
Fig 1) was conceived as a mechanism to document and make freely available to
the cassava community information being generated by the network. It is
Project IP3: improving cassava for the developing world
Output 8-50
hosted at CIAT. Results of the SSR analysis of local cassava land races from
Ghana and Guatemala, including names and characteristics of local cassava
varieties, allele size data by SSR marker locus, genetic diversity analysis,
principal component of genetic distance etc, are now available for viewing on
the MOLCAS web data base. Results from 3 other country studies namely
Uganda, Sierra Leone and Brazil are being compiled and will be available on the
web before the end of the year. The MOLCAS web-based data base has since its
launch last year received a total of 34,477 visits with an average of 2700 visits
per month. Table 1 shows the summary of all visits and visits in the last 3
months
Figure 1. MOLCAS home page on the web showing links to the data base of
country studies
Table 1. A record of visits to the MOLCAS web-based data base on the country
studies of SSR assessment of local cassava varieties
Jul
Total
Aug
Hits
Sep-03 2003
2003
Request
/molcas/imagen.jsp
8,003
130
1,303
496
/molcas/locus.jsp
7,605
247
1,228
517
/molcas/alelosp.jsp
5,515
374
1,002
252
/molcas/estudios.jsp
3,250
125
266
370
/molcas/
2,162
88
138
130
/molcas/markers-det.jsp
1,892
37
101
187
/molcas/imagenbioquim.jsp
1,401
37
72
156
/molcas/appendix1.jsp
1,381
35
69
179
/molcas/intrap_data2.jsp
1,377
37
86
149
/molcas/appendix2.jsp
858
16
40
145
/molcas/studies.jsp
482
11
16
38
/molcas/pcr_cond.jsp
448
10
48
41
/webapps/molcas/
103
2
5
4
Total
34,477
1,149
4,374
2,664
Output 8-51
2003 Annual Report
Conclusions and Perspectives
The MOLCAS web accessible database has increased the completed country
studies available for viewing to 5 and 3 more will be added before the end of the
year. Two other country studies, Cuba and Mozambique are ongoing and will
also be added once completed. Future perspectives for the database is to add a
number of useful links, for example, sites where genetic diversity analysis
software can be downloaded freely. Other perspectives is to link the MOLCAS
data base to and genotyping activities on cassava in the genetic resources
challenge program which is expected to begin next year.
Activity 8.15 Development of molecular techniques for assessing genetic
diversity and mapping of useful genes
Development of a Diversity Array Technology (DArT) Chip for Cassava
Collaborators: Andrzej Killian, Peter Wenzel (CAMBIA); Carmen deVicente
(IPGRI) Edgar Barrera, Ana-Maria Correa, Martin Fregene (CIAT)
Background
Genetic resources, mostly held by small farmers represent a critical resource for
the future productivity and stability of production of the crop. How to evaluate
and use in a systematic manner the vast amount of variability present in
cassava is still a challenge to most cassava breeding programs. Genotyping
micro-array technologies offer the highest throughput available up to date. One
of them diversity array technology, DArT ™ (CAMBIA), is sequence-independent
(low-input) and allows the fingerprint of an individual’s genome based on a high
number of polymorphic sites spread over the genome. These screening
procedures should allow testing of thousands of individual samples in a speedy
manner. We describe here a proof of concept on using the DArT tool as a costeffective way for measuring and characterizing genetic diversity of cassava
germplasm.
Methodology
The project was initiated in early October 2002, with the shipment of cassava
DNA samples from CIAT to CAMBIA. Plant materials used for the generation of
the DArT chip was chosen to represent a broad as possible diversity of the
cultivar, a few genotype of its wild progenitors and 2 wild species were included
to capture a large number of polymorphic fragments. They include 14
accessions from Brazil, 14 from Colombia, 4 from Guatemala, 2 each from
Nigeria, Cuba, and Ecuador, Peru and Thailand respectively. Others include,
one accession each from Argentina, Bolivia, Costa rica, Fiji islands, Indonesia,
Mexico, Panama, Venezuela, and USA. Six and 2 improved varieties were
included from CIAT and IITA respectively. The wild species accessions were 29
of Manihot esculenta sub spp Flabellifolia, 7 of M. carthaginensis and 1 of M.
walkerae. DNA isolation was according to Dellaporta et al. (1983) followed by
two washes of phenol/chloroform.
Project IP3: improving cassava for the developing world
Output 8-52
A critical step in DArT is the complexity reduction step. Work at CAMBIA with
several other plant genomes has shown that digestion with PstI restriction
enzyme (RE) in combination with more frequently cutting RE is an efficient
method to reduce complexity. A preliminary experiment was therefore
conducted to determine the best enzyme combinations, mixture of genomic DNA
from twenty cassava genotypes was digested with PstI, ligated to adapters and
further digested with several frequently cutting RE (BstNI, ApoI, TaqI and
BanII), followed by amplification with an adapter-specific primer. The cassava
genomic PstI fragments lacking the recognition site for the frequent cutting RE
(BstNI, ApoI and TaqI) were individualized by transformation into E. coli,
amplified from bacterial colonies and micro-arrayed. From each of the libraries
760 clones were arrayed. Genomic representations prepared in the same way
(RE digestion/ligation followed by amplification) from each of the twenty
genotypes separately, were labelled with Cy3-dUTP and hybridized together with
Cy5-dUTP-labelled reference DNA to these microarrays. Slide preparation,
hybridizations, washes and scanning are as described by Andrzej et al. (2002).
Images generated by the scanner were used to extract quantitative fluorescence
signal data for each array feature using our proprietary software. Same
software was used to binarize the data (score as 0/1) for all slides. Binary
scoring table was used to prepare the Hamming distance matrix and to obtain a
dendogram. Library expansion to obtain more polymorphic clones was carried
out using the enzyme combinations PstI/BstNI, and PstI/TaqI, and 80 DNA
samples from CIAT. About 3000 clones were evaluated for polymorphism from
the 2 libraries. The arrays were printed and DArT analysis carried out as
described earlier.
Results
Among the three libraries /PstI/BstNI, PstI/ApoI, and PstI/TaqI) tested in the
preliminary experimnt, PstI/BstNI gave the largest number of polymorphic
clones (132), followed by TaqI (112) and ApoI (69). In total, 313 candidate
polymorphic clones were obtained in the initial experiment to determine the
best enzyme combination. DArT analysis based on 296 polymorphic clones
without a single missing data point was used to generate a binary matrix and
obtain a dendrogram, based on Hammings distance, representing genetic
relationship between the 20 samples (Fig. 1).
Output 8-53
2003 Annual Report
M. esculenta
M. esculenta subsp. flabellifolia
M. esculenta
M. carthagenensis
M. walkerae
Figure1. Genetic relationship among the cassava accession used to identify
polymorphic clones in the PstI/BstNI, Pst/TaqI and Pst/ApoI array. The
dendrogram was created using the distance table based on 296
polymorphisms and the UPGMA clustering algorithm
The libray expansion with 80 clones yielded 440 polymorphic clones (14.3%), for
the PstI/TaqI array, consistent with the results obtained with the initial,
smaller array (14.6% polymorphic clones). A dendrogram based upon the
analysis of the 80 cassava samples with the polymorphic clones is presented in
Figure 2.
Project IP3: improving cassava for the developing world
Output 8-54
COL1522
MCOL2056
MCOL2061
COL72
ECU79
MGUA62
CUB56
TAI16
MGUA7
NGA19
NGA2
MCOL1505
CM6740-7
CM523-7
COL667A
PER333
TME1
COL2737
HMC53
COL1468
BRA890
BRA522
MCOL2516
PER183
TAI8
C4
AMA28
AMA19
BRA403
BRA1439
BRA124
BRA461
BRA383
OW80-7
OW81-8
M.wal001
OW161-1
OW164-5
Figure 2. Genetic relationship among the cassava accession used to identify
polymorphic clones in the PstI/TaqI array. The dendrogram was created
using the distance table based on 440 polymorphisms and the UPGMA
clustering algorithm.
In the PstI/BstNI array 554 polymorphic clones (18.0%) were identifid, also
consistent with the polymorphism frequency in the smaller PstI/BstNI array in
Output 8-55
2003 Annual Report
the first project phase (17.2%). The dendrogram resulting from this array is
shown in Figure 3.
COL2737
ARG5
MCOL1505
TME1
BRA927
CM6740-7
HMC53
BRA403
BRA931
TAI8
C4
BRA124
BRA1439
MCOL2516
NGA19
NGA2
AMA28
BRA890
PER333
BRA522
AMA19
BRA829
CM523-7
OW80-7
NGA2
MGUA62
MGUA7
TAI16
COL1522
COL1672
COL72
ECU72
ECU79
CUB56
MCOL2061
MCOL2056
BRA383
BRA461
M.wal001
OW164-5
OW161-1
Figure 3. Genetic relationships among the cassava accessions used to identify
polymorphic clones (markers) in the PstI/BstNI array. The dendrogram
was created using the distance table based on 554 polymorphisms and
the UPGMA clustering algorithm. Topology and branch lengths of this
dendrogram are biased due to the presence of significant numbers of
polymorphic clones derived from repetitive sequences.
There is a difference between the two dendrograms obtained with the
PstI/BastNI and the PstI/TaqI array, respectively. A thorough inspection of the
Project IP3: improving cassava for the developing world
Output 8-56
data suggests that the BstNI array contains a higher proportion of clones
derived from repetitive sequences than PstI/TaqI array. Typing using repetitive
sequences introduces a bias in genetic diversity analysis due to overrepresentation. The PstI/TaqI array does not show a high proportion of clones
with repeated sequences and can be used as a routine genotyping tool for
genetic diversity analysis. Work is ongoing to analyze a larger sub set of
accessions from the CIAT core collection with at least 800 polymorphic clones
Conclusion and Perspectives
This feasibility project has resulted in the design and validation of a reliable
complexity reduction method that is uncovering several hundred polymorphic
DArT markers in cassava germplasm. Because DArT markers can be scored in
parallel in a single analysis, a high throughput, cost-effective whole genome
genotyping is now available for determining high-density genome profiles of
cassava. As this study has demonstrated, the availability of such a method has
the potential to improve dramatically the recognition, conservation and
exploitation of cassava genetic diversity. This platform is now available for
research on genetic diversity and genetic mapping studies in cassava. With
current improvements of DArT’s throughput and its cost reduction it is feasible
to molecularly characterize very large germplasm collections within a
reasonable period of time and cost.
References
Kawano K. (2003). Thirty Years of Cassava Breeding: Biological, and Social
Factors for Success. Crop Science 43:1325-1335.
Kawano K. (2001). The role of improved cassava cultivars in generating income
for better farm management. In:Howeler R.H., and Tan S.L. (eds).
Cassava`s Potential in the 21st Century: Present Situation and Future
Research and Development Neds. Proceedings of the Sixth Regional
Workshop, held in Ho Chi Minh City, Vietnam. Feb 21-25, 2000. CIATBangkok, Thailans. Pp5-15.
Output 8-57
2003 Annual Report
Activity 8.16 Mining the Primary Gene Pool of Cassava: Introgression of
High Root Protein from Accessions of Manihot esculenta sub spp
Fabellifolia and Manihot Tristis into Cassava
Collaborators: N. Morante, T. Sanchez, J. Marin, C. Ospina, H. Ceballos, A.
Alzate, S. Moreno, M. Fregene
Funding: CIAT core funds
Important outputs
1) Inter-specific hybrids from wild relatives with high root protein and cassava were
evaluated and results reveal the consistency of the trait.
2) Amino acid profile and preliminary SDS-PAGE analysis was also conducted on a
few accessions with high root proteins towards a proper characterization of the
proteins contained in these genotypes
Rationale
As a major staple food crop across the tropics, cassava can serve as a cheap
means of deploying adequate protein requirement amongst the poor and for
feeding animals. A major effort has therefore been embarked upon to increase
protein content of cassava roots. An advanced back cross QTL (ABC-QTL) to
introgress high protein content from wild relatives into cassava is in its third
year at CIAT. The first year saw the evaluation of wild relatives of cassava for
high protein content and in the second year, inter-specific hybridization
between selected high protein lines and some improved elite parents, including
some yellow varieties. Wild by wild crosses were also carried out to investigate
if combining favorable alleles from different populations or species of the wild
accessions can further increase protein content. The inter-specific hybrids were
evaluated in a seedling trial this year.
Other activities this year include an amino acid profile of root flour from 4 wild
and inter-specific accessions and standardization of the SDS-PAGE
methodology for the determination of size and isolation of root protein from high
protein accessions, towards a proper characterization of the protein. During
this year also, cassava varieties that were found to be high in protein from an
evaluation conducted in 2001 were re-established in the field from tissue
culture plants for another round of evaluations. If the previous results are
confirmed, genetic crosses will be made with elite parents of the cassava gene
pools for breeding high protein content and QTL mapping studies.
Methodology
Sexual seeds of inter-specific hybrids obtained from last year crosses were
evaluated in a seedling trial this year. Total protein was measured in root flour,
from 3 root per plant, using the Kjedahl method, in selected individuals of
about half the total number of families. The decision to evaluate a fraction of
the F1 lines was due to cost considerations and size of roots, only genotypes
with fairly large sized roots were used for protein measurement. In
collaboration with the starch company AVEBE, amino acid profile was obtained
from root flour from 1 genotype each of M.esculenta sub spp flabellifolia and M.
tristis high in protein, as well as from 2 inter-specific hybrids high in protein.
Project IP3: improving cassava for the developing world
Output 8-58
Pooled leaf flour sample from 10 cassava varieties with high protein was also
analyzed.
Several protein extraction protocols were evaluated for SDS-PAGE analysis of
root protein in cassava in order to select the most suitable one (Table 8.15).
About100mg of root flour was used in each case, suspended in a 500Pl volume
of sample buffer. The samples were centrifuged in an eppendorf tube for 5 min
at 14000 rpm and the supernatant was transferred to a new eppendorf tube
and, in the case of some protocols, mixed in a 1:1 (v:v) ratio with SDS-PAGE
disruption buffer (Laemmli, 1970). Proteins were completely dissociated by
immersing the samples in a boiling water bath for 5 min, then briefly
centrifuging at 14000 rpm to pellet cellular debris. The resulting supernatants
(total protein extracts) were stored at –20°C until SDS-PAGE analysis .
One-dimensional SDS-PAGE was run with 4.5% acrylamide (w/v) loading gel and 10%,
12%, 13.8%, 15% acrylamide (w/v) separation gel concentrations were tested (Table
8.16) using a Bio-Rad gel electrophoresis apparatus. The 4.5% acrylamide loading gel
was prepared by mixing 9.0ml water, 2.25 ml acrylamide (30% solution), 3.75 ml Tris
1.5M pH: 6.8, 150µl SDS (10% solution), 150µl ammonium persulfate (10% solution)
and 20µl TEMED and the casting (Laemmli,1970). Ten microliters of each sample was
loaded per lane. Constant voltage of 50V was applied for 1 h at 10 qC and increased to
150V for the remaining duration of the run, until the tracking dye reached the gel mold.
A Phaseolus vulgaris protein sample (Phaseolin type T) was used as a check in every
protocol assayed. After electrophoresis, gels were fixed in 12% tri chloro acetic acid, then
stained in a solution containing 2% phosphoric acid, 10% ammonium sulphate, 0.15%
Coomassie Blue G, and 20% methanol for 12 hours. Gels were then destained in 20%
methanol and scanned.
Table 8.15. Different extraction buffers for isolation of proteins in cassava root
Buffer
Reference
1.
1M Tris-HCl pH: 7.5, 0.5M EDTA,
Gutierrez 2003,Personal communication
1% Ascorbic acid, 0.1M PMS
1a. 1M Tris-HCl pH: 7.5, 0.5M EDTA,
1% Ascorbic acid, 0.1M PMS and boiling
for 5 min.
2. 0.005M Sodium phosphate, sucrose 5%
Suiter, 1988
0.05% E-mercapthoethanol, pH 7.0
3. 0.1M KCl, 20 mM cystein, pH 7.3
Bourdon, 198
4. 0.0625 M Tris-HCl, 2% SDS, 10% Glycerol,Laemmli, 1970
5% (E-mercapthoethanol, 0.001 bromophenol
5. 50mM Sodium phosphate
Shewry, 1992
5a. 50mM Sodium Phosphate, 10mM PMSF,
Shewry, 1992
10% PVP
6. 0.5M NaCl, pH 3.2, 2mM EDTA, 2% SDS,
CIAT,
40% Sucrose, 1% (E-mercapthoethanol,
Gutierrez 2003, Personal communication
0.01% Bromophenol blue in 0.0625M
Tris-HCl pH 6.8
Output 8-59
2003 Annual Report
Table 8.16. Conditions of SDS-PAGE for the visualization of root protein in
cassava
1.5M Tris^;
30% Acrylamide*
3.5M Tris^;
30% Acrylamide*
10%
15%
12%
Water
37.5ml
23ml
33ml
Acrylamide*
30ml
50ml
40ml
Tris pH: 8.8
22.5ml
25ml
SDS (10% solution)
APS (10%solution)
TEMED
0.4ml
0.4ml
0.05ml
0.5ml
0.5ml
0.04ml
25ml
0.5ml
0.5ml
0.04ml
3.5M Tris^;
22.8% Acrylamide*
13.6%
11.88ml
54.72ml
22.5ml
0.4ml
0.2ml
0.02ml
37:1 = w/w ratio of acrylamide to N,N'-methylene bis-acrylamide
Thirty one cassava varieties that were found to have between 6 and 8% of crude
protein in their roots in an evaluation conducted in 2001 (CIAT 2002) were reestablished again, from in vitro culture for a second year evaluation.
Results
A total of 4,271 sexual seeds organized into 58 families were obtained from
inter-specific crosses between high protein accessions of M. esculenta sub spp
flabellifolia, M. tristis and elite parents of the cassava gene pool. An evaluation
of root protein content was made of 579 genotyopes , based on root size, and a
summary of the results are shown in Table 8.17. Results reveal that some wild
genotypes such as OW 231-3, 280-2, OW132-2, and OW284-1 have good
general combing ability for root protein content. It was also observed that
crosses between 2 wild parents, both high in protein (WW), had more uniform
high root protein progenies, compared to wild by cultivated crosses, suggesting
that a number of genes for protein content might be recessive or additive. The
inter-specific hybrids are being evaluated for a second year in a single row trial
(SRT) experiment. Amino acid profile revealed very high amounts of arginine,
about half the total amount of amino acids, and low levels of methionine and
lysine in the roots, but high levels in leaves (Table 8.18).
Project IP3: improving cassava for the developing world
Output 8-60
Table 8.17. Descriptive statistics of % Protein, based on dried root basis, in 31
inter-specific families . Genotypes OW230, 231, 180, 181, 189, are
accessions of M. esculenta sub spp flabellifolia, while Genotype OW280,
284, 132, 131, 146 are accessions of M. tristis
Family
CW 205
WW
WW
WW
WW
14402241-
Mother
OW 231-3
51
1
3
9
OW
OW
OW
OW
181284231284-
Father
MTAI 8
2
1
3
1
OW
OW
OW
OW
280280240146-
2
2
8
1
CW 177-1
OW 132-2
CW 161
CW 56- 5
CM 1585- 13
OW 189- 1
CW 160- 1
CW 179-1
CW 56- 5
OW 132-2
OW 181- 2
MTAI 8
WW 24- 2
OW 231- 3
OW 280- 2
CW 99- 26
CW 30- 29
OW 280- 1
CW 185
OW 180- 1
M TAI
WW 3-1
OW 132-2
OW 240-6
OW 280- 1
M COL 1734
OW 230- 3
OW 230- 3
CM 1585- 13
OW
OW
CW
CW
OW
WW 39CW 256CW 200CW 201CW 73-
3
1
1
2
2
CW 198
CW 212CW 251CW 203CW 184CW 204CW 183
WW 21WW 20CW 202WW 9CW 186WW 19Total
1
2
1
2
1
10
2
2
1
1
2
Output 8-61
8
2802804756284-
2
1
3
5
1
OW 230- 3
CW 30- 65
OW 284- 1
M COL 1734
OW 230- 4
OW 180- 1
OW 231- 3
OW 180- 1
OW 231- 3
OW 231- 3
OW 230- 4
OW 180- 1
OW 181- 2
OW 230- 3
M COL 1734
OW 189- 1
CW 48- 1
M COL 1734
AM 244- 31
CW 48- 1
OW 240- 6
OW 146- 1
CW 30- 73
OW 234- 2
CM 1585- 13
OW 240- 8
Size
8
58
76
82
41
63
3
3
17
35
30
6
6
23
9
5
5
12
31
7
5
3
5
5
2
14
8
4
2
4
7
579
Maximum Minimum
10.96
5.37
10.49
4.39
10.46
4.04
9.72
3.19
9.70
3.26
8.52
3.75
8.03
5.37
7.86
4.63
7.86
3.80
7.80
3.42
7.66
3.85
7.58
3.10
7.32
4.05
7.27
3.54
7.20
2.87
7.10
4.00
7.10
4.00
6.84
4.27
6.77
3.97
6.53
4.40
6.46
5.22
6.37
5.29
6.27
3.01
6.25
4.46
6.05
4.66
5.86
3.11
5.74
3.30
5.66
4.02
5.58
5.39
5.27
3.72
4.80
3.01
Average
8.11
6.42
6.28
5.37
6.21
5.71
6.64
6.06
5.65
5.63
5.27
5.13
5.77
5.46
4.86
5.04
5.60
4.97
5.30
5.25
5.89
5.76
4.73
5.04
5.36
4.74
4.98
4.61
5.48
4.69
3.72
Std. Deviation
2.80
1.47
1.26
1.21
1.51
1.04
1.29
1.64
1.17
1.09
0.87
1.66
1.43
1.06
1.63
1.09
1.15
0.86
0.73
0.66
0.55
0.55
1.33
0.75
0.98
1.00
0.90
0.72
0.13
0.67
0.96
2003 Annual Report
Table 8.18. Total protein and Amino acid profile of root flour from 2 high
protein inter-specific hybrids, 2 high protein M. escuenta sub spp
flabellifolia accessions used as parent and a pooled leaf sample of 10 high
leaf protein cassava varieties
Variety:
OW235-3
g/kg
CW66-18
g/kg
CW66-48
g/kg
OW1322
g/kg
Leaf (pooled)
g/kg
N
CP=N*6.25
15.7
98
13.8
86
14.9
93
18.7
117
35.5
222
cys
met
asp
thr
ser
glu
pro
gly
ala
val
ile
leu
tyr
phe
gaba
his
ornt
lys
arg
trp
sum ex gaba,
ornt
sumAA/CP
sumAA/N
0.3
0.6
2.1
1.1
1
7.7
0.9
0.9
1.5
1.3
0.8
1
0.5
0.7
0.9
1.1
1.3
1.5
27
0.6
0.6
0.2
1.3
0.5
0.6
4.7
0.6
0.7
0.9
0.6
0.5
0.8
0.4
0.5
0.7
0.9
0.6
1.5
28.5
0.3
0.3
0.5
1.6
0.8
0.8
8.6
1
0.8
1.4
1
0.7
1
0.5
0.7
0.8
1
3.8
1.5
23.8
0.4
0.4
0.5
2.7
1.1
1.2
9.7
1.1
1.1
2.2
1.2
0.8
1.4
0.1
0.9
1.5
1.5
1.8
1.8
27
0.4
2.9
3.5
22.2
9.4
10.8
31
10.3
10.7
12.1
12.2
9.5
17.2
8.1
11.4
2
4.7
0.1
12.3
11.1
4.4
50.6
0.52
3.23
44.1
0.51
3.20
46.4
0.50
3.12
55.1
0.47
2.94
203.8
0.92
5.74
To get a better understanding of these unusual amino acid profile preliminary
SDS-PAGE analysis of the protein was conducted on the 4 high protein
samples. Extraction buffers number 3, and 5 gave the best quality protein
extract as demonstrated by the visualization of some bands (Fig.8.17). Gel
concentrations of 13,8 and 12% were more suitable in resolving bands. The
bands obtained are still quite faint and need to be improved before proper band
sizing and protein isolation can be conducted. The success of SDS-PAGE relies
upon selection of the buffer which in turns depends of the characteristics of the
tissue. The ratio of buffer to tissue needs to be optimized in materials with low
content of proteins like cassava root flour. Efforts continue to standardize the
SDS-PAGE protocol for root proteins in cassava.
Project IP3: improving cassava for the developing world
Output 8-62
T
a
b
c
T: Phaseolus (control)
a: samples with buffer 1
b: samples with buffer 1a
T
f
T: Phaseolus (control)
f: samples with buffer 5
g: samples with buffer 5a
d
e
c: samples with buffer 2
d: samples with buffer 3
e: samples with buffer 4
g
h
i
j
h: samples with buffer 6
i: samp les with buffer 6a
j: samples with buffer 5b
Figure 8.17 One-dimensional sodium dodecyl sulphate-polyacrylamide gel
electrophoresis (SDS-PAGE) of total proteins from root of four cassava
cultivars with different extraction buffers to standardardize analysis of
root protein in cassava.
A group of 33 genotypes, 5 plants each, that showed high root protein in an
analysis conducted in 2001 were requested from the genetic resource unit and
sent to the green house for hardening (Table 8.19).
Table 8.19. List of cassava varieties with high protein content in an evaluation
conducted in 2001 that are to be re-evaluated again this year.
Output 8-63
2003 Annual Report
Clone and protein (%)
CM 5620-3
8.31
SM 1406-1
8.13
MCOL 689B
7.75
MCOL 1563
7.38
MGUA 76
6.94
MCR 142
6.63
CM 696-1
6.44
CM 3199-1
6.44
SM 734-5
6.44
MCR 38
6.31
MGUA 86
6.31
Clone and protein (%)
MCOL 2436
6.25
MBRA 26
6.25
MCR 136
6.13
MGUA 9
6.13
MGUA 91
6.06
MMEX108
6.06
SM 629-6
6.00
SM 673-1
6.00
MCOL 2532
6.00
MGUA 19
6.00
CM 3236-3
5.94
Clone and protein (%)
MBRA 101
5.94
MCOL 219
5.94
MGUA 33
5.94
CM 7310-1
5.88
MCOL 678
5.88
MMEX 95
5.81
MGUA 79
5.81
MBRA 300
5.75
MCOL 2459
5.75
MBRA 1384
5.75
MCOL 2694
5.75
Conclusions and perspectives
This year, inter-specific hybrids from wild relatives with high root protein and
cassava were evaluated and results reveal the consistency of the trait, amino
acid profile and preliminary SDS-PAGE analysis was also conducted with the
root proteins. Future perspectives include genetic back crosses, to cassava, of
inter-specific hybrids with high protein and evaluation of putative high root
protein cassava varieties
References
Callahan, F. E., Jenkins, J. N. Creech, R. G. and Lawrence, G.W. 1997.
Changes in Cotton Root Proteins Correlated with Resistance to Root Knot
Nematode Development. The Journal of Cotton Science 1: 38-47.
Carvalho, L.J.C., Cascardo, J.C., Ferreira, M.A. and Loureiro, M.E. 1992.
Studies on proteins and enzymes related to tuberization and starch
biosynthesis in cassava roots. In CBN: 234-238.
Gonzales, P.L., Gutiérrez, M.M. y Fuchs, M. 2001. Estandarizaci¢n de la
metodolog¡a para caracterizar genotipos de algod¢n (Gossypium sp.)
mediante patrones electrofor‚ticos de prote¡nas. Rev.Fac.Agron.(Maracay)
27:95-103
Hovenkamp-Hermelink, J.H.M , Jacobsen, E., Ponstein, A.S., Visser, G.F., VosScheperkeuter, G.H., Bijmolt, E.W., de Vries J.N., Witholt, B. and Feenstra,
W. J. 1987.Isolation of an amylase-free starch mutant of the potato
(Solanum tuberosum L.). TAG 75:217-221
Jasso, D., Romero, J., Rodríguez, R. and Angulo J.L. 2002. Characterization of
Proteins from Sunflower Leaves and Seeds: Relatiopnship of Biomass and
Seed Yield. Reprinted from: Trends in new crops, and new uses. ASHS
press.
Laemmli, U. K. 1970. Cleavage of Structural Proteins during the Assembly of
the Head of Bacteriophae T4. Nature Vol. 227: 680-685
Project IP3: improving cassava for the developing world
Output 8-64
Shewry, P.R., Clowes, A., Tatham, A.S. and Beeching, J. 1992. Oportunities for
manipulating the amount and composition of proteins in cassava tuberous
roots. First International Scientific Meeting of the Cassava Biotechnology
Network. CIAT, CBN. Or (Proceedings CBN: 251-254)
Activity 8.17. Mining the Primary Gene Pool: Green Mites (CGM) Resistance
Genes from Manihot esculenta sub spp Fabellifolia.
Collaborators: N. Morante, J. Guerrero, A. Bellotti, J. Marin, C. Ospina, H.
Ceballos, M. Fregene (CIAT)
Important outputs
1) Evaluation of 4 putative SSR markers associated with resistance to green
mites in 4 F1 inter-specific with hybrid between cassava and Manihot
esculenta sub spp Fabellifolia families revealed regression coefficients of
46%, 30%, 30%, and 5% in the 4 families respectively
2) Development of 45 BC1 families to introgress the high level of resistance
to cassava green mites found in some accessions of Manihot esculenta
sub spp Fabellifolia
Rationale
Following a very dry spell in January 2002 at CIAT Palmira and a subsequent
heavy incidence of the green mites, 4 inter-specific hybrid families, CW68,
CW65, CW67, and CW66, from the wild Manihot accession MFLA 437- 007
showed a very high level of resistance to the green mites with an almost equal
number of susceptible(score of 3-4) and resistant (score of 1-2) genotypes. Bulk
segregant analysis (BSA) was used to identify 4 SSR markers NS74; NS217;
NS260; SSRY330 polymorphic in the bulks and individuals of the inter-specific
families resistant to green mites (CIAT 2002). At the same time, an attempt was
made to transfer the resistance observed in the inter-specific families to elite
cassava parents. A total of 45 BC1 families were developed, planted in a
seedling trial, and a preliminary evaluation for mites conducted last season.
Selected individuals of these BC1 families were extensively crossed to CMD
resistant parents used for MAS at CIAT, to combine CMD and CGM resistance
in progenitors meant for Africa, to produce BC2 families. The BC1 progenies
were also cloned and planted in a single row trial for the evaluation of
resistance to green mites during the dry spell early next year. We describe here
evaluation of the putative markers in the inter-specific families, and the
development of BC2 families from selected BC1 individuals.
Methodology
The SSR markers polymorphic in the bulks were evaluated in the 4 F1 interspecific families and a simple regression analysis conducted using Microsoft
excel. The markers found to explain a significant part of phenotypic variance of
CGM resistance in the analysis of the 4 F1 families were then analyzed in the
parents of the 45 BC1 families.. SSR analysis of the 4 markers was as described
Output 8-65
2003 Annual Report
by Mba et al (2001). The BC1 families were also evaluated for resistance to
mites, a rather preliminary evaluation since only one plant is available per
genotype. Based upon the preliminary CGM resistance evaluation of the 45 BC1
a large number of putatively resistant BC1 progenies were crossed to CMD
resistant parents for the generation of BC2 families from which CMD and CGM
resistant lines can be selected from for the generation of parents for breeding in
African gene pools.
Results
Simple regression analysis of SSR markers NS74; NS217; NS260; SSRY330 in
the 4 inter-specific families had coefficients (R2) of 46%, 30%, 30%, and 5% in
the 4 families respectively. The surprisingly low regression coefficients for
resistance that is apparently controlled by a major gene (CIAT 2002) might be
due to the single year evaluation of these families for CGM resistance upon
which the analysis was based. CGM incidence tends to have focal points
therefore a large part of phenotypic variance in an evaluation is due to the
environment. The best way to avoid this is to evaluate over 2 or 3 growing
cycles. A second more in depth evaluation of these families is being carried out
in Santander the Quilichao over two dry seasons, a period of 18 months
September 2002 until March 2004. The preliminary evaluation of the BC1
families is being repeated this year in a clonal observation trial before further
regression analysis is carried out.
An evaluation of the SSR markers associated with CGM resistance in the interspecific hybrids in the parents of the BC1 families revealed that the bands
associated with CGM resistance were not always polymorphic between the
parents of the back cross populations. An effort was therefore initiated to
identify more markers linked to CGM resistance, BSA, using additional SSR
and RAPD markers is being used to identify additional markers for evaluation of
resistance. Transfer of a gene to other crosses via MAS using a single marker
associated with the gene is fraught with problems of linkage disequilibrum,
based upon the frequency of that allele in the gene pool. Many more markers
are required to eliminate the problems of linkage disequilibrum encountered in
diverse genetic backgrounds.
Individuals of the BC1 families that appeared to possess resistance to CGM in
focal points of great CGM damage in the seedling trial were crossed to CMD
resistant parents to obtain recombinants that carry CMD and CGM resistance.
Table 8.20 shows the number of seeds by families of the new crosses. These
crosses have been established in vitro from embryo axes to enable it to be
shared with collaborators in Africa once CMD and CGM resistance have been
confirmed by MAS and phenotypic evaluation respectively.
Conclusions
Markers associated with CGM resistance in bulks of 4 inter-specific families
were evaluated in the entire families and rather low regression coefficients were
found. This is most likely due to the environmental effect in the phenotypic
data. BC2 families have been generated from BC1 individuals towards an
introgression of this resistance into elite cassava gene pools.
Project IP3: improving cassava for the developing world
Output 8-66
Table 8.20. Sexual seeds of BC2 families produced that combines resistance to
CGM and CMD resistance
ITEM
Father
Mother
Code
No. of Seeds
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
C-4
C-4
C-4
C-4
C-4
C-4
C-6
C-6
C-6
C-6
C-19
C-19
C-33
C-33
C-33
C-33
C-33
C-33
C-39
C-39
C-39
C-127
C-127
C-127
C-127
C-127
C-127
C-127
C-127
C-127
C-127
C-243
C-243
C-243
C-243
CW219-3
CW74-1
CW236-14
CW235-8
CW234-12
CW234-19
CW235-72
CW235-2
CW232-8
CW213-1
CW234-17
CW234-17
CW235-8
CW235-2
CW234-12
CW217-7
CW232-8
CW258-17
CW235-100
CW235-8
CW257-25
CW258-17
CW234-8
CW234-17
CW234-19
CW234-32
CW235-2
CW235-8
CW235-51
CW235-72
CW235-100
CW258-19
CW234-17
CW234-19
CW235-2
CW257-10
C-127
TOTAL
A1
A2
A3
A4
A5
A6
A7
A8
A9
A10
A11
A12
A13
A14
A15
A16
A17
A18
A19
A20
A21
A22
A23
A24
A25
A26
A27
A28
A29
A30
A31
A32
A33
A34
A35
A36
94
102
135
105
6
17
21
11
3
6
6
23
5
38
12
9
12
15
8
54
5
27
12
25
51
9
16
2
2
14
16
2
5
2
10
1
881
References
Output 8-67
2003 Annual Report
CIAT, (2002). Annual Report Project SB2, Assessing and Utilizing
Agrobiodiversity through Biotechnology, CIAT, Cali, Colombia, pp 239241.
Dellaporta SL Wood J, Hicks JR (1983) A plant DNA minipreparation: version II.
Plant Mol Biol Rep 1:19-21
Mba REC, Stephenson P, Edwards K, Melzer S, Mkumbira J, Gullberg U, Apel
K, Gale M, Tohme J, Fregene MA (2001) Simple Sequence Repeat (SSR)
markers survey of the cassava (Manihot esculenta Crantz) genome:
towards an SSR-based molecular genetic map of cassava. Theor Appl
Genet: 21-31
Activity 8.18. Positional Cloning of CMD2 the Gene that Confers High
Level of Resistance to the Cassava Mosaic Disease (CMD)
Collaborators: M. Moreno, Martin Fregene (CIAT)
Funding: Rockefeller Foundation
Important output
1) Identification of a RAPD marker RME-1 tightly linked, at less than 1cM,
with CMD2 the major CMD resistance gene and its conversion into a
SCAR marker
Rationale
Previous work revealed that the SSR markers SSRY 28 and NS158 are the
closest markers to the gene CMD2 that confers resistance to the cassava mosaic
disease (CMD), and are located at distances of 9 and 3 cM respectively (Akano
et. al 2002; Zárate 2002, CIAT 2002). High resolution of the CMD2 region of
the genome was therefore initiated in this region. The experimental approach
involves a search for recombinants between CMD2 and the above markers using
a large full-sib population, followed by an analysis of the recombinants with
thousands of readily assayed markers to identify additional markers more
closely linked to the gene. This requires the use of several types of marker
systems that can achieve whole genome screens with a reasonable level of
effort. Molecular markers identified that are closest to gene CMD2 will be
used to screen a BAC library for the construction of a BAC contig that
transverses the region of CMD2 via successive steps of BAC end sequencing,
mapping and more BAC library screening.
Methodology
The fine-mapping population was 1690 individuals from a cross between
TME3, the source of CMD2 and the improved variety TMS30572. The cross was
evaluated in the 2002 growing season for CMD resistance in the field at IITA,
Ibadan, under heavy natural pressure of the disease. DNA was isolated from
the individuals of the cross, using the Dellaporta et al. (1983) method, and
diluted 10X in TE without quantification for molecular marker analysis was at
CIAT. The population was evaluated with the 2 SSR markers according as
described by Mba et al. 2001 and recombinants between the markers and
CMD2 identified. DNA from 10 resistant recombinants and 10 susceptible
recombinants were combined to form 2 bulks which were then evaluated with
Project IP3: improving cassava for the developing world
Output 8-68
several markers system including AFLPs, ISTRs, RAPDs and SSRs in a modified
bulk segregant analysis (BSA) method.(Michelmore et al. (1991). Evaluation
with AFLP markers (Vos et al. 1995) was using a commercial ALFP (Invitrogen
Life Technologies, Gaithersburg, MD) following the manufacturer’s
instructions.. All 64 possible combinations were used in the evaluation. For
ISTRs (Inter Sequence Tagged Repeat), the method described by Rohde et al
(1996) was used with all possible 64 combinations of the 8 F and 8 B universal
retro-elements (retro-transposons) sequence primers. Evaluation with RAPD
markers was using 768 commercial primers (Operon Technologies Inc, CA) and
a modified Williams et al. (1990) protocol. PCR conditions were 1X PCR buffer,
2.5mM of MgCl2, 0.4mM of DNTPs, 0.8uM of each primer, 1.0U of enzyme taq
polymerase, and 50 ng/ul of DNA template in a of 2oül. The amplification
program was an initial denaturation cycle at 940C for 5min; 35 cycles of 940C
for 30s, 360C for 1 min, 720C for 1 min 30s; and a final extension cycle of 720C
for 5min. The amplified fragments were separated on 1.5% agarose gelsand
visualized by staining with Ethidium Bromide and examination under UV light.
A set of 146 newly developed SSR markers (CIAT 2002) were also used to
evaluate the recombinant bulks using PCR and PAGE electrophoresis
conditions described by Mba et al. (2001). Markers that were polymorphic in
the recombinant bulks were then analyzed in individuals of the bulks. A
polymorphic RAPD fragment in the individuals of the recombinant bulk was
cloned into pGEMT-easy (Promega inc, Madison) and sent for sequencing at the
University of Iowa sequencing facility. Primers were designed from the
sequences and are being used as SCAR marker for MAS.
Results
The evaluation of the fine-mapping population with the SSR markers SSRY28
and NS158 allowed the identification 112 recombinant individuals. The
evaluation of the resistant and susceptible recombinant bulks with AFLP, SSR,
ISTR produced several candidate markers that were polymorphic in the bulks
but the polymorphism was not consistent when individuals of the bulks were
analyzed separately (opened bulks) . However, analysis with RAPD markers
produced 2 polymorphic candidate markers, AC-15 and RME-1 that remained
consistent in the individuals of the bulks (Fig 8.18). Evaluation of the two
markers in the entire fine map progeny revealed that AC-15 is at least 2cM from
CMD2, while RME-1 is less than 1cM from the gene.
The polymorphic
fragment in RME-1, a 800bp fragment was cloned into the pGEMT-easy and
sequenced. Homology comparison between the sequence of the RAPD band and
sequences in public database using BLAST (www.ncbi.nlm.nih.gov) revealed the
sequence is similar to the minor caspid protein of bacteriophage T3 suggesting
that the fragment is single copy gene. This is being verified at the moment via
southern hybridization to total cassava DNA. SCAR primers have been
designed from the sequence for the use of the RAPD marker in MAS since it is
closer than to the gene than NS158, the closest marker to date to CMD2. The
above result provides a molecular marker closely linked to CMD2 that can be
used to screen a BAC library BAC of cassava for the construction of BAC
contigs which is the next stage of the positional cloning of the CMD resistance
gene.
Output 8-69
2003 Annual Report
RB
SB
RPSP/RBSB
Figure 8.18. Ethidium bromide stained agrarose gel of individuals
recombinant bulks evaluated with the RAPD marker RME-1. A
at around 800bp (arrow) can be observed in the resistant parent
resistant bulks (RB) that is absent in the susceptible parent
susceptible bulk (SB)
from the
fragment
(RP), and
(SP) and
Conclusions and perspectives
A high resolution map with 4 markers, one at less than 1cM has been
constructed around the genome region of CMD2. The cloning of this gene is
now proceeding to the next stage of BAC library screening and the construction
of a BAC contig around CMD2.
References
AKANO, A; BARRIER, E; DIXON, A.G.O; FREGENE, M. 2002.Genetic mapping
of to dominant gene conferring resistance to cassava mosaic wished.
Theorical and Applied Genetics. 105:521-525.
CIAT 2002.
Annual Report Project SB2, Assessing and Utilizing
Agrobiodiversity through Biotechnology, CIAT, Cali, Colombia, 250pp.
HAYDEN, M.J & SHARP, P.J. 2001. Targeted development of informative
microsatellite (SSR) markers. Nucleic Acids Research. 29: 8e44.
MBA, REC; STEPHENSON, P; EDWARDS, K; MELZER, S; NKUMBIRA, J;
GULLBERG, U; APEL, K; GALE, M; TOHME, J.; FREGENE, M. 2001.
Simple Sequence Repeat (SSR) markers survey of the cassava (Manihot
esculenta Crantz) genome: Towards to molecular SSR-based genetic map
of cassava. Theorical and Applied Genetics. 102:21-31.
MICHELMORE, R.W; PARAN, I; KESSELI, R.V. 1991. Identification of markers
linked to disease resistance genes by bulked segregant analysis: a rapid
method to detect markers in specific genomic regions by using segregating
populations. Proceeding of National Academic of Science of The United
States of America. 88: 9828-9832.
ROHDE, And 1996. Inver sequence-tagged repeat (ISTR) analysis: Two novel
and universal for PCR-based technique genome analysis in the plant and
animal kingdom. Journal Genetics and Breed.50:249-261.
WILLIAMS, J.G.K; KUBELIK, A.R; LIVAK, K.J; RAFALSKI, J.A; TINGEY, S.V.
1990. DNA polymorphisms amplified by arbitrary primers plows useful
ace genetic markers. Nucleic Acids Research. 18: 6531-6535.
Project IP3: improving cassava for the developing world
Output 8-70
ZARATE, L.A. 2002. Genetic Mapeo of a population F1 of cultivated yucca
(Manihot esculenta Crantz) using Microsatélites. Thesis of predegree.
University of the Tolima. Ibagué. Colombia.
Activity 8.19 Progress in the Anti-Sense Mediated Silencing of the
Granule Bound Starch Synthase I (GBSS I) for the Production of Waxy
Cassava Starch
Collaborators: Yina J. Puentes P., Edgar Barrera S., Paul Chavarriaga.,
Chikelu Mba, Martin Fregene (CIAT)
Funding: Ministerio de Agricultura y Desarollo Rural (MADR), Colombia
Important outputs
1) Development of Full-length sense and anti-sense transformation constructs of the
GBSSI gene
2) Successful transformation of the constructs into the model cassava transformation
variety 60444 (MNg11).
3) Regeneration of transformed calli to be followed by molecular and biochemical
tests to test stable expression of the gene and the waxy starch phenotype.
Rationale
Higher incomes from cassava in the developing world where the crop is
generally found will require the industrialization of the crop and the
development of novel industrial products from cassava. There are several novel
products that can be produced from cassava, including modified starches such
as 100% amylopectin or 100% amylose starches, from the down regulation of
the granule-bound starch synthethase (GBSS) gene or the starch-branching
enzyme (SBE) gene. Industrial application of either pure amylopectin or pure
amylose starches, such as the production of high-value biodegradable polymers
from pure amylose starches or the use of 100% amylopectin in thickeners,
pastes and glues, have a market with unlimited growth potential.
Biotechnology can play a very important role in the production of the above
products in cassava.
With funds from the Colombian Ministry or Agriculture and Rural Development,
a project was been initiated to genetically engineer industrial cassava varieties
for the production of waxy starch using an anti-sense and sense construct of
the GBSSI gene. GBSS catalyses the conversion of ADP-glucose to amylose
through the linkage of an ADP glucose to a preexisting glucan chain. Antisense
disruption of the GBSSI gene has been employed to create potato transformants
with 70-100% amylopectin via the down-regulation of the GBSSI gene
(Salehuzzaman et al.,1993) and the disruption sense in sweet potato of the gene
GBSS (Kimura et al., 2001).
Output 8-71
2003 Annual Report
Methodology
Isolation of a cassava GBSS cDNA clone. More than 87,000 clones of a cassava
root and leaf cDNA library cloned in the vector pCMV SPORT (GIBCO BRL Inc.,
USA) was gridded onto high-density filters (Mba et al., 2000 unpublished data).
The library was screened using a potato GBSS cDNA clones, a gift from
Dr.Christine Gerhardt (Max Planck Institute, Cologne, Germany). The potato
GBSS gene was labeled with [32P] dATP by random primer labeling and
hybridized overnight to the cDNA filters according to standard protocols for
Southern hybridization used in cassava (Fregene et al., 1997). The filters were
washed twice with 2X SSC +0.1% SDS at 60oC for 5 min, and autoradiography
was al -80 oC using 2 intensifying screens.
Construction of transformation cassettes. Primers were designed from published
sequences of a full-length cassava cDNA of the GBSSI gene (Salehuzzaman et
al., 1993). BamHI and XbaI restriction enzyme recognition sites were
incorporated in 5’end of the primers to enable sub-cloning of the cDNA clone in
the sense and antisense orientation into the multiple -cloning site (MSC) of the
vector pRT101. The primers were used to amplify the cDNA clone obtained
above, and the PCR product was cleaned using the QIAGEN PCR Clean Up Kit
(QIAGEN Inc., Los Angeles, CA) and digested with the appropriate enzymes. A
2.1Kb BamHI /XbaI fragment was subcloned in the sense and antisense
between the 35S promoter and the 35S polyadenylated terminator region of
vector pRT101, a gift from Dr. Ryohei Terauchi, Iwate Biotechnology Research
Center, Kitakami, Japan. The 35S promoter, GBSSI gene in pRT101 was
liberated using the restriction enzyme PstI, separated on a agarose gel, eluted
and cloned into the PstI site of the binary vector pCAMBIA 1305.2 having the
GUSPlusR and HPT reporter genes.
Transformation by Agrobacterium tumefaciens, of varieties. Friable
Embryogenic Callus of the cassava genotype TMS60444, Mcol.2215 y CM 33064 was transformed via Agrobacterium tumefaciens with the GBSSI gene in
antisense-sense orientation, mediated technology CIAT.
Results
Three GBSS cDNA clones obtained form screening the cassava library were
sequenced, and one was found to be a complete cDNA clone. The cDNA clone
has the ATG start codon 81 bp down stream from the beginning of the cDNA
sequence and a stop codon about 100 bp from the poly-A tail. PCR
amplification with the designed primers yielded a fragment about 2.1 kb in size
that corresponds to the full-length GBSSI cDNA clone(Figure 8.19).
Project IP3: improving cassava for the developing world
Output 8-72
OPstI
1
2
3
4
GBSSI
antisense
5
6
7
8
GBSSI
2100 bp
Amplification clon 3
Figure 8.19. PCR amplification of the GBSSI cDNA clone using primers to
introduce restriction enzyme sites at the ends of the gene. The first lane
on the right is molecular weight marker Lambda DNA, digested with PstI,
the next three lanes are PCR of the gene GBSSI antisense and next are
PCR of the gene GBSSI sense.
The resulting PCR fragment, digested with BamHI and XbaI restriction enzyme,
was cloned into the MCS of pRT101. Next, the GBSSI gene, promoter and
terminator sequences, excised with PstI and the resulting fragments separated
from the vector fragment (sizes 2.8 and 2.7 kb) by electrophoresis was cloned
into the PstI site of pCAMBIA (Figure 8.20). These are the constructs that were
used in the Agrobacterium-mediated transformation.
OPstI
1
2
Sense band of approx
2800bp cloned in the vector
pCAMBIA 1305.2
Antisense band of approx
2800bp cloned in the
vector pCAMBIA 1305.2
Band of approx 2700bp corresponding to the
vector pRT101
Figure 8.20. PstI digested of pRT101 plasmid containing the cassava GBSSI
gene in anti-sense-sense orientation. The fragments of about 2.8 and 2.7
Kb in size represent the GBSSI gene, flanked by the 35S promoter and
the polyadenylation terminator sequence, and the rest of the pRT101
plasmid respectively.
Output 8-73
2003 Annual Report
Two genetic constructions (Figure 8.21) with the GBSSI gene in anti-sense and
sense orientation in the vector pCAMBIA 1305.2 were made to achieve silencing
of the gene.
(A)
(B)
Figure 8.21. Gene constructs of GBSS in sense (A) and anti-sense (B) in the
binary vector pCAMBIA 1305.2
The constructs were transformed into friable embryogenic Callus (FEC) of the
model transformation genotype MNG11 via Agrobacterium tumefaciens. Results
of GUS transitory assay revealed a successful incorporation of the gene (Fig.
8.22)
Figure 8.22. Positive test of GUS in Cotyledonary embryos of variety TMS 60444
with the GBSSI sense gene constrauct in the vector pCAMBIA 1305.2
Project IP3: improving cassava for the developing world
Output 8-74
Conclusion
We have successfully transformed full-length sense and anti-sense constructs
of the GBSSI gene into the model cassava transformation variety MNG11.
Transformed calli are being regenerated following which molecular and
biochemical tests will be conducted to test stable expression of the gene and the
eventually the waxy phenotype. The project was carried out as an
undergraduate project for the Colombian undergraduate student project, Yina
Jazbleidi Puentes P. of the Universidad Nacional de Colombia -Sede Palmira)
References
CIAT (Centro Internacional de Agricultura Tropical). 2003. Annual Report
Project SB2, Assessing and Utilizing Agrobiodiversity through
Biotecnology. Cali, CO. p.202-205
FREGENE, M.; Angel F.; Gomez, R.; Rodriguez, F.; Chavarriaga, P,; Roca, W.;
Tohme, J., Bonierbale, M. 1997. A molecular genetics map of cassava
(Manihot esculenta Crantz) Theor Appl Genet 95:431-441.
KIMURA, T., M. ONTANI., T. NODA., O. IDETA., T. SHIMADA & A. SAITO. 2001.
Absence of amylose in sweet potato (Ipomea batatas (L) Lam) following
the introduction of granule-bound starch synthase I cDNA Plant cell
report 20: 663-666.
SALEHUZZAMAN, S.N.I.M.; Jacobsen, E.; Visser, R.G.F. 1993. Isolation and
characterization of a cDNA-encoding granule-bound starch synthase in
cassava (Manihot esculenta Crantz) and its antisense expresion in potato.
Plant Mol Biol 23:947-962.
Activity 8.20.
Evaluation of S1 Families for Waxy Mutants
Collaborators: Allison Smith, Chris Hylton (JIC, Norwich), Nelson Morante,
Teresa Sanchez, Hernan Ceballos, Martin Fregene (CIAT)
Funding: CIAT core funds
Important ouputs
1) Development of a standard amylopectin percentage determination curve
based on the use of pure cassava amylopection and amylase samples
2) Use of the curve to analyze more than 500 S1 progenies
Rationale
There is a growing interest from both the private and public sectors to develop a
waxy starch phenotype in cassava. Three approaches have been embarked
upon, via genetic transformation, anti-sense, and sense silencing, irradiation of
cassava seeds, and screening the germplasm bank for waxy starch phenotypes.
The heterozygous nature of cassava however makes the second and third option
difficult, given the low chances that a natural mutant of the waxy gene will be
found in the homozygous state. A decision was made to screen many S1
families developed under a population development effort to for tolerance to
inbreeding. About 14 S1 families were available for evaluation. To further
increase the precision of the chemical assay for percent amylase in the above
evaluations, Prof Allison Smith`s group at the JIC was contacted for assistance
Output 8-75
2003 Annual Report
in preparing pure cassava amylose and amylopectin for making an amylose
determination standard curve. Current assays use commercially available pure
potato amylose and amylopectin in preparing the standard curve, but the chain
lengths of amylose in cassava and potato differ and this may introduce errors in
the percent amylose measured. Pure amylose and amylopectin was isolated
from root starch of the cassava variety MCol 2216 and used in developing an
amylose determination standard curve.
Methodology
14 S1 families, developed under an S2 recurrent selection program to develop
populations tolerant to inbreeding were the plant materials for the above
experimene. They were planted last year at the CIAT station in Santander the
Quilichao and harvested July this year. At harvest starch samples was
collected from 3 roots of all progenies, a total of 514 individuals, and taken to
the laboratory for analysis. Pure cassava amylose and amylopection
preparations were made using a sepharose separation column at the John
Innes center, Norwich following methods described by Zeeman et. al. (2003).
Mixtures of amylose and amylopectin with varying proportions of amylose from
0% to 100% were prepared, the samples were then dispersed in ethanol,
hydrolyzed by acid, iodine added to a final concentration o 2% (v/v) and the
absorbance measured at wavelengths of 700nm and 525nm in a
spectrophotometer. Measuring the sample at this two wavelengths and using a
ratio of 700:525 rather than the traditional method of measurement of a single
wavelength , 620nm, have been found to be more accurate in determining
amylose content (Zeeman et al. 2003). The absorbance and of the samples with
varying proportions off amylose was used in generating an amylose
determination curve.
Results
Pure cassava amylose and amylopection were obtained from fractions of a
sepharose column at maximum absorption at wavelength of 595nm (Fig 8.23).
The preparations were used in making mixtures of different proportions of
amylose and the absorbance measured as describd above. Figure 8.24 and
table 8.21 shows results of the development of an amylose content
determination curve. The curve will be used for determining amylose content in
the starch samples obtained from the S1 families. Analysis of the starch
samples are ongoing and should be completed by November. Any samples that
shows less than 5% amylose will be sent for further analysis, including a
quantitative purification of the amylose and amylopectin, and analysis of the
molecular structure, to ensure that the low amylose content is not due to
extremely long amylose molecules whose absorbance in solution closely mirrors
amylopectin molecules
Project IP3: improving cassava for the developing world
Output 8-76
Cassava starch : Sepharose CL2B (20/08/03)
60ul F + 140ul 60% Lugol
0.20
0.18
0.16
A595 (normalised)
0.14
0.12
0.10
0.08
0.06
0.04
0.02
0.00
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
Fraction
Figure 8.23. Absorbance of a sepharose column fractions in the preparation of
pure cassava amylose and amylopectin from the cassava variety
MCol2216
Output 8-77
2003 Annual Report
Amylose content vs ratio of A700 and A525 of glucan-iodine complex
(27/08/03 20.119 "030822 Cassava - Martin Fregene.xls")
1.0
0.8
Proportion amylose
y = 0.4998x - 0.1056
R2 = 0.999
0.6
0.4
0.2
0.0
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
Ratio (A700:A525)
Figure 8.24. An amylose determination standard curve prepared using pure
cassava amylose and amylopectin
Table 8.21. Data for the preparation of the amylose content determination curve
Amylose
Proportion
am ylose
AP
ul
ug
ul
ug
0
0.000
103.1
17.816
3
0.891
98.0
6
1.782
9
A525
Ratio
A700:A525
A700
I
II
Mean
I
II
Mean
0.00
0.2249
0.2391
0.2320
0.0506
0.0497
0.0502
0.2162
16.926
0.05
0.2427
0.2414
0.2421
0.0748
0.0747
0.0748
0.3088
92.8
16.035
0.10
0.2403
0.2369
0.2386
0.0993
0.0983
0.0988
0.4141
2.672
87.7
15.144
0.15
0.2422
0.2353
0.2388
0.1210
0.1253
0.1232
0.5158
12
3.563
82.5
14.253
0.20
0.2515
0.2533
0.2524
0.1584
0.1593
0.1589
0.6294
18
5.345
72.2
12.471
0.30
0.2537
0.2524
0.2531
0.2046
0.2048
0.2047
0.8089
24
7.127
61.9
10.690
0.40
0.2520
0.2562
0.2541
0.2510
0.2673
0.2592
1.0199
30
8.908
51.6
8.908
0.50
0.2483
0.2512
0.2498
0.2971
0.2974
0.2973
1.1902
36
10.690
41.3
7.127
0.60
0.2489
0.2652
0.2571
0.3526
0.3781
0.3654
1.4213
42
12.471
30.9
5.345
0.70
0.2557
0.2671
0.2614
0.4016
0.4207
0.4112
1.5729
48
14.253
20.6
3.563
0.80
0.2645
0.2558
0.2602
0.4804
0.4519
0.4662
1.7919
54
16.035
10.3
1.782
0.90
0.2666
0.2663
0.2665
0.5313
0.5336
0.5325
1.9983
60
17.816
0.0
0.000
1.00
0.2504
0.2617
0.2561
0.5577
0.6009
0.5793
2.2624
Conclusions and pespectives
A more accurate method for the determination of amylose and amylopection
proportions in cassava has been developed in collaboration with the John Innes
Project IP3: improving cassava for the developing world
Output 8-78
center. It consists of using an amylose determination standard curve that was
prepared using cassava amylose and amylopectin rather than those from
potato. The method is currently being used to evaluate starch samples from S1
families
References
Zeeman S., Tiessen A., Pilling E., Kato L., Donald A., and Smith A. (2003)Starch
synthesis in Arabidopsis thaliana: Granule synthesis, composition and
structure. Plant physiology 129:516-529.
Activity 8.21. Irradiation of Sexual Seeds for the Production of Waxy
Cassava and other Mutants
Collaborators: Dr Chikelu Mba (IAEA, Vienna, Austria), Ms Sarah Tafur
(National University, Palmira), Dr Hernan Ceballos, Nelson
Morante, Martin Fregene (CIAT)
Important output
1) Irradiation of sexual seeds with gamma rays and fast neutrons for the
identificication of useful mutants
Rationale
The globalization of economies has meant a search for local crops that are
competitive and will preserve local agriculture. Economic surveys in the past
ten years in cassava growing regions of Africa, Asia and Latin America have
revealed that cassava is an important factor in the improved livelihoods of the
rural population of these regions (Nweke et al 2001; Kawano 2001; Ceballos
2002). A change in starch quality, for example the elimination of amylose (waxy
starch), via a knockout of the GBSS I gene, implies access to new markets for
cassava growers. For most rural communities, a better standard of living
depends on increasing income from their crop harvest. An important
disadvantage of cassava is the short shelf life of its roots, which have to be
consumed or processed within a few days after harvest. This trait, post-harvest
physiological deterioration (PPD), results in losses and higher marketing costs
and its elimination will lead to higher profit margins for the small producer.
PPD is thought to be a wound response cascade that goes out of control, and
the knockout of certain key genes in the cascade should lead to a reduction in
PPD.
Cassava also possesses a wide range of cyanogenic glucoside content that is
often cited as a health risk and a stigma for its acceptance as raw material in
certain food and feed industry. Cyanogenic glucosides are produced by a
biosynthetic pathway that has a cytochrome P450 as catalyst in the ratelimiting step (Andersen et. al. 2000). Removal of the gene that expresses the
cytochrome P450 gene should lead to low cyanogenic potential (CNP) cassava
plants. Increased productivity or increased value of cassava roots such as novel
starch types and improved marketing through elimination of losses from post-
Output 8-79
2003 Annual Report
harvest physiological deterioration and the removal of the stigma of cyanogenic
glucosides, stands to improve livelihoods of cassava farmers. Mutagenesis has
been applied extensively in the production of novel phenotypes in crop species
(Van Harten 1998). The project will take advantage of simultaneous research,
currently under way, that will facilitate to routine production of inbred
materials, for the first time in a cassava breeding project. Once mutants have
been identified, molecular genetic analysis will be used to track down genes
responsible for the novel traits. The use of these genes as markers or in genetic
transformation will permit an increase in the efficiency of transferring these
traits to other cassava gene pools through conventional breeding.
Specific Objectives were:
i)
ii)
iii)
iv)
Irradiate, using gamma rays (a Cobalt-60 source) and fast neutrons, of
sexual seeds from elite cassava genotypes under improvement for
tolerance to inbreeding depression
Establish plants from the irradiated seeds and non-mutated parental
genotypes and evaluate them for useful traits such as delayed post
harvest physiological deterioration (PPD), low cyanogenic potential (CNP),
high dry matter content (DMC), and novel starch types
Develop selfing of the mutated lines to obtain S0 progenies and their
evaluation for the above root quality traits and any other potentially
useful trait
Carry out DNA analysis for genes known to be involved in biosynthetic
pathways of the above traits to identify mutants.
Methodology
Cassava has seldom been inbred, and the large “genetic load” hidden in its
heterozygous background will likely hinder the production of viable homozygous
plants, a phenomenon known as “inbreeding depression”. Selection for
tolerance to inbreeding depression has therefore been initiated to make cassava
populations amenable to the production of inbred lines. Sexual seeds from
cassava genotypes tolerant to inbreeding were the source of genetic material for
mutagenesis. About 2000 sexual seeds were shipped to IAEA for irradiation,
using gamma rays (a Cobalt-60 source), 1000 seeds, and fast neutrons, 1000
seeds. The level of irradiation with gamma rays was 200Gy.
The irradiated seeds were sent back to CIAT for germination and establishment
of the plants in the field. The heterozygous nature of cassava implies that
mutations in a recessive gene will not be observed in the M0 phenotype. There
is therefore a need to self the M0 plants to permit identification of the recessive
mutants. However, the task of selfing thousands of plants is beyond capacity
at CIAT and a selection of mutants for genes of interest will first be carried out
to identify mutants. DNA analysis that can identify single nucleotide
polymorphisms (SNPs) or insertions/deletions (INDELs) in genes of interest will
be employed. At 10 months after planting, the plants from the irradiated seeds
selected above will be evaluated for ability to produce flowers. Plants that
Project IP3: improving cassava for the developing world
Output 8-80
flower will be cloned and planted the following year in a clonal observation trial
fashion of 10 plants per genotype.
Plants in the clonal observation trial above will be selfed to obtain the M1 (S0)
generation. The seeds from the M1 (S0 generation) will again be established in
the field at CIAT and other key target environments and thoroughly evaluated
for the traits mentioned in the previous section. Progeny identified with useful
root traits will be introduced into the cassavabreeding program.
Results
A total of 2000 full-sib seeds from full- and half- families of MCol1505, HMC-1,
C4 (CMD resistant parent) were sent to IAEA for irradiation with gamma rays
and fast neutrons. About 1000 seeds were irradiated with a dose of 200Gy of
gamma rays. They were moisture equilibrated over a 69% glycerol solution in a
dessicator prior to radiation. These seeds have been sent back to CIAT where
they have been planted in a seedling nursery. Seeds irradiated with fast
neutron experiment, about1000 seeds, are still being expected back.
Conclusions and perspectives
This project seeks to use novel methods of mutagenesis, conventional plant
breeding and molecular genetic analysis to identify cassava genotypes with
value added traits. It will take advantage of the recently initiated research to
produce inbred cassava germplasm. The project will also use tools of genomics
to track down genes responsible for the above traits, markers associated to
these genes can be used to efficiently move these genes around the different
cassava gene pools defined by agro-ecologies. The new methods will not only
accelerate the production of improved germplasm but also be a model for the
development of other traits of interest to the market and farmer.
References
Andersen MD, Busk PK, Svendsen I and Møller BL 2000. Cytochromes P-450
from bcassava (Manihot esculenta Crantz) catalyzing the first steps in the
Biosynthesis of the cyanogenic glucosides linamarin and lotaustralin.
Cloning, functional expression in Pichia pastoris, and substrate
specificity of the isolated recombinant enzymes. J Biol. Chem. 275-19661975.
Ceballos, H. 2002. La yuca en Colombia y el mundo: nuevas perspectivas para
un cultivo milenario. In: Ospina B. and H.Ceballos (Eds.) La Yuca en el
Tercer Milenio: sistemas modernos de producción, procesamiento,
utilización y comercialización. CIAT Publication Number 327. Cali,
Colombia.
CIAT (International Center for Tropical Agriculture), 2001. Annual Report
Project IP3: Improved cassava for the developing world. Cali, Colombia.
FAO (2000). Food Outlook December 2000. FAO, via delle terme di Caracalla,
00100 Rome, Italy.
Output 8-81
2003 Annual Report
Kawano, K. (2001). The role of improved cassava cultivars in generating income
for better farm management. In:Howeler R.H., and Tan S.L. (eds).
Cassava’s Potential in the 21st Century: Present Situation and Future
Research and Development Needs. Proceedings of the Sixth Regional
Workshop, held in Ho Chi Minh city, Vietnam. Feb 21-25, 2000, CIAT,
Bangkok, Thailand, pp 5-15
Kawano, K., K. Narintaraporn, P. Narintaraporn, S. Sarakarn, A. Limsila, J.
Limsila, D. Suparhan, and W. Watananonta. 1998. Yield improvement in
a multistage breeding program for cassava. Crop Sci. 38:325-332.
Nweke, F., Spencer, D., Lynam, J. (2001) The Cassava Transformation: Africa’s
Best-Kept Secret. Michigan State University Press, East Lansing, USA.
273 pp.
Salehuzzaman SNIM, Jacobsen E y Visser RG F. (1993) Isolation and
characterization of a cDNA-encoding granule-bound starch synthase in
cassava (Manihot esculenta Crantz) and its antisense expression in potato.
Plant Mol. Biol. 23:947-962.
Van Harten A. M. (1998). Mutation breeding: theory and practical applications.
Cambridge University Press, Cambridge.
Project IP3: improving cassava for the developing world
Output 8-82
Activity 8.22 Saturation of the Molecular Genetic Map of Cassava with
PCR-based Markers: Progress on the Mapping of a New Set of 140 New
SSR Markers
Collaborators: L. Cano, J. Gutierrez, M. Fregene (CIAT)
Funding: CIAT core funds
Important outputs
1) A new set of 140 SSR markers identified from more than 2000 BAC end
sequences of the white fly resistant variety MECU72 were evaluated in
the parents of the mapping population and
2) Thirty five polymorphic markers were identified and 26 have been
evaluated in the mapping progeny
Rationale
The wide-spread utility of the molecular genetic framework map of cassava
published six years ago (Fregene et al., 1997) was delayed by the preponderance
of the RFLP’s markers used in constructing that map. Over the past 3 years,
efforts have been geared to the development and genetic mapping of SSR
markers to make markers on the cassava map more accessible to cassava
researchers worldwide especially in the NARS of Africa, Asia, Latin America and
the Caribbean. An initial effort led to the mapping of 77 markers (Mba et al.,
200, CIAT 2001), other efforts have led to mapping of 57 SSRs (Zarate 2002)
and 45 SSRs from cDNA sequences (Garcia 2002). We describe here the
screening of another 140 SSR markers, identified in BAC end sequences, in the
parents of the mapping population (Niogeria-2 and CM2177-2) and a
preliminary report of the mapping of 26 polymorphic markers.
Methodology
Primers were earlier designed from a total of 141 BAC end sequences found to
contain SSR motifs (CIAT 2002). A Dellaporta et al., (1983,) modified protocol
was used to extract DNA extraction from the parents and the progeny of the
mapping population. DNA quantification was done using the Dyna Quant
TM.200 Fluorometer. Hoefer Pharmacia Biotech. PCR amplifications were
carried out in 25 Pl reactions containing 50 ng of DNA, buffer 1X, 2 mM of
MgCl2, 0.2 mM of each dNTPs , 0.2 mM of each primer and 0.25 U of Taq
polymerase. The PCR profile was changed for some markers, a reduction of the
annealing temperature, to achieve amplification. The products of amplification
were electrophoresed on 6% polyacrylamide gels and visualized by silver
staining. For the survey of the parents, the following samples were included:
Nigeria-2, CM2177-2, K150, TME3, TMS30555, CMD resistant and susceptible
bulks. SSR markers that showed polymorphism in the survey of the parents,
i.e. having a unique allele in either or both parents of the mapping population,
i.e. polymorphic, were used to screen the 150 progenies of the mapping
population.
Output 8-83
2003 Annual Report
Results
A total of 35 polymorphic markers were found in the survey of the parents
(Fig8.25). Of this 26 markers have been evaluated in the F1 progeny (Fig 8.26).
The level of polymorphism is about 25% and it is much lower than that found
for SSR markers developed from an enriched library (60%) and that from a
cDNA library (about 40%). The low level of polymorphism found here could be
due to the source of the markers, BAC ends, or other unknown reasons.
NS 1141
NS 1040
NS 1072
NS 1019
NS 1035
NS 1008
NS1099
Figure 8.25. Silver stained gel polyacrilamide showing PCR analysis of seven
markers NS used to survey the 2 parents of the mapping population,
including another genotypes : the progeny k150, TME3 and resistant (R)
and susceptible (S) bulks. Each marker have changes in the annealing
temperature (left to right : 52, 45, 52, 60, 60, 53, 550C, respectively. ) The
polymorphism between in the male and female genotypes can be observed.
Figure 8.26. Silver stained gel polyacrilamide showing PCR analysis of NS 1019
(600C) marker in the progeny F1 (cross Nigeria-2 x CM 2177-2). The
segregation of dates for this marker in a ratio 1:1, presence: absence of
the unique parental allele (67/79 for the female and 76/70 for the male
framework respectively) can be observed.
Conclusions and perspectives
A new set of 140 SSR markers identified from more than 2000 BAC end
sequences of the white fly resistant variety MECU72 was evaluated in the
parents of the mapping population and 35 polymorphic markers identified, of
this 26 have been evaluated in the F1 progeny
Project IP3: improving cassava for the developing world
Output 8-84
References
Fregene M, Angel F, Gomez R, Rodriguez F, Chavarriaga P, Roca W, Tohme J,
Bonierbale M (1997) A molecular genetic map of cassava. Theor Appl Genet
95: 431-441.
Garcia, T.. 2002. Mapeo genético de una población F1 de yuca cultivada (
Manihot esculenta Crantz) utilizando Microsatélites que proviene de cDNA.
Tesis de pregrado. Universidad del Valle, Cali. Colombia.
Mba, R.E.C., Stephenson, P., Edwards, K., Melzer, S., Mkumbira, J., Gullberg,
U., Apel, K., Gale, M., Tohme, J. and Fregene, M. (2001) Simple Sequence
Repeat (SSR) Markers Survey of the cassava (Manihot esculenta Crantz)
Genome: Towards an SSSR-Bassed Molecular Genetic Map of Cassava.
Theoretical and Applied Genetics. 102: 21-31.
Zarate, L.A. 2002. Mapeo genético de una población F1 de yuca cultivada (
Manihot esculenta Crantz) utilizando Microsatélites. Tesis de pregrado.
Universidad del Tolima. Ibagué. Colombia.
8.23. Embryo Rescue of Sexual Seeds from Breeding Populations for
Molecular Assisted Selection (MAS) of CMD Resistance
Collaborators: Luis Guillermo Santos, Danilo Moreta, Samy Judy Moreno,
Adriana Alzate, Nelson Morante, Hernan Ceballos, Martin
Fregene (CIAT)
Funding: The Rockefeller Foundations
Important output
1) The establishment CMD resistance breeding populations from embryo
axes at CIAT for molecular marker-assisted breeding of resistance has
become routine. This year more than 1000 plants were processed
Rationale
CMD breeding at CIAT aims to develop Latin America cassava gene pools
adapted to the disease should in case it makes an accidental entry into the
region. A second important objective is to facilitate germplasm shipment of
CIAT’s elite cassava germplasm to regions, such as India and Sub Saharan
Africa where CMD is endemic, via the introgression of CMD resistance into
CIAT’s elite germplasm. To permit marker-assisted selection (MAS) of CMD
resistance at CIAT for Latin America and at the same time fulfill plant
quarantine conditions for the shipment of the CMD resistant CIAT germplasm
to India and Africa, it is necessary to germinate and maintain in vitro breeding
populations. This season more than 3000 controlled crosses were made
between CMD resistant parents introduced from IITA and elite cassava parents
or backcross derivatives of M. esculenta sub spp flabellifolia for resistance to the
green mite. A total of 2315 seeds were harvested as mature seeds and have
been germinated in vitro from embryo axes. A summary of results obtained
from the germination of these seeds is presented below. Once germinated, the
Output 8-85
2003 Annual Report
plantlets were multiplied, molecular-assisted selection (MAS) performed using
the marker NS158 and another SCAR marker RME1, and CMD resistant
genotypes will be transferred to the screen house for further hardening and
evaluation as well as for shippment to collaborators in India and Africa.
Methodology
The culture of embryo axes from mature or immature seed has become routine
in the establishment of cassava CMD breeding populations in vitro. The method
being currently used is that developed at CIAT and modified by Okogbenin
(2003), in summary, seeds are selected by a floatation test in water, seeds that
float (indicating of vain or little developed seeds) are discarded, incubation in
97% sulfuric acid to 97% for 50 minutes, to allow for easy scarification, and
washing with water to eliminate the acid. Following, the seeds are disinfected
with 70% alcohol to for two minutes then with a solution of sodium hypoclorite
(0.5%) and a drop of 20 tween for 12 minutes, finally they are rinsed three
times with sterile deionized water. The embryo axes is then removed together
with the cotyledon and placed on 17N media and left in the dark for 5 days at
temperatures between 28 and 31oC. Next, the embryos are incubated at the
same temperature under a photoperiod of 12h light and 12h darkness until
they grow into full plants which are then multiplied in 4E media.
Results
A total of 2315 mature seeds of cassava originating from 62 F1 families
corresponding to breeding populations for resistance to CMD, denominated
”CR” were received from the breeding section. Of this 214 seeds were rejected
by the floatation test and 2101 seeds were germinated. Of this number a total
of 1128 fully formed plants were recovered and multiplied, a little more 54%
recovery (Table 8.22). The low % recovery of plants, down from the normal 80%
was due to a very severe attack of green mites in the growth room this year.
Plants that are observed to be contaminated with green mites are immediately
removed and destroyed. A concerted effort of weekly application of acaricide in
the growth room, elimination of infected plants, and cleaning of individual
tubes with 96% ethanol has been initiated to control the mite infestation. A
number of other cassava tissue culture facilities at CIAT also had mite problems
this year. The source of the mite problem is not known but it is thought they
may have gained entrance via infested in vitro plants transferred from one
facility to another.
One mature plant of each of the 1128 genotypes was sent to the cassava
molecular marker lab for DNA extraction and marker analysis for MAS.
Genotypes that have the CMD2 resistance gene after marker analysis are sent
to the green house for hardening and further field evaluation and to partners in
Africa and India. At the moment 633 genotypes resistant to ACMD are being
transferred to greenhouse, a total off 125 have been established to far, and also
being multiplied, at least 10 copies per shipment, to make a shipment to
Nigeria, Tanzania and India. Also in the green house are 96 genotypes of 3 S2
families developed under the S2 recurrent selection for tolerance to inbreeding
and 47 genotypes of inter-specific crosses established last year from immature
sexual seeds and propagated in vitro this year. These immature seeds are from
Project IP3: improving cassava for the developing world
Output 8-86
plants that had to be removed from the field due to the “zero-cassava” rule of no
plants for one month on the experimental station at Palmira.
Table 8.22. Sexual seeds of cassava that were embryo rescued and multiplied In vitro in the year
2003.
j–‹Œ
t–›Œ™G
mˆ›Œ™
wœ™—–šŒ
jyTX
jyTY
jyTZ
jyT[
jyT\
jyT]
jyT^h
jyT^i
jyT_h
jyT_i
jyT`h
jyT`i
jyTXWh
jyTXWi
jyTXXh
jyTXXi
jyTXY
jyTXZ
jyTX[h
jyTX[i
jyTX\h
jyTX\i
jyTX]
jyTX^
jyTX_
jyTX`
jyTYWh
jyTYWi
jyTYX
jyTYYh
jyTYYi
jyTYZ
jyTY[
jyTY\
jyTY]
jyTY^
jyTY_h
jyTY_i
jyTY`h
jyTY`i
jyTZW
jyTZX
jyTZY
jyTZZ
jyTZ[h
jyTZ[i
jyTZ\
jyTZ]
v~X_ZT[
jT[
jT[
jT[
jT[
jT[
jT[
tjvsGX^Z[
jT[
tjvsGYYW]
jT[
t {hpG_
jTXY^
tjvsGX^Z[
jTXY^
tjvsGYYW]
jTXY^
tjvsGX^Z[
jt \YZT^
jT[
jt \YZT^
jTZZ
jt \YZT^
jt \YZT^
jtZZW]T[
jtZZW]T[
jtZZW]T[
jTZZ
jtZZW]T[
jt]^\[T_
jTZZ
jt^`\XT\
jt^`\XT\
jt^`\XT\
jt^`\XT\
jt^`\XT\
zt`W`TY\
jT[
zt`W`TY\
jTZZ
zt`W`TY\
zt`W`TY\
ztX]]\TY
ztX^[XTX
ztX^[XTX
jTX_
ztX^[XTX
ztX^[XTX
jTXY^
ht Y[[TZX
jn[_`TZ[
j~]]T]W
j~]]T^Z
j~]^T[Y
tjvsGX^Z[
jT[
tjvsGYYW]
jT[
t{hpG_
jT[
tjvsGX^Z[
jTXY^
tjvsGYYW]
jTXY^
t{hpG_
jTZZ
jT[
jt\YZT^
jTZZ
jt\YZT^
jTZ`
jTY[Z
jT[
jTX_
jTZZ
jtZZW]T[
jTY[Z
jTZZ
jt]^\[T_
jT[
jTX_
jTZZ
jTZ`
jTY[Z
jT[
zt`W`TY\
jTZZ
zt`W`TY\
jTZ`
jT[XZ
jTZZ
jT[
jTX_
ztX^[XTX
jTZZ
jTZ`
hs~Thjtk
hjtk
hjtkTt zi
hjtkThjy
hjtkThjy
hjtkThjy
hjtkT€y{
hjtkT€y{
hjtkT€y{
hjtkT€y{
hjtkX
hjtkX
hjtkT€y{
hjtkT€y{
hjtkT€y{
hjtkT€y{
hjt kTWX
€y{Thjt k
hjtkY
hjtkY
hjtkY
hjtkY
hjtkY
hjtkY
hjtkX
hjtkX
hjtkX
hjtkX
hjtkX
hjtkX
hjtkX
hjtk[
hjtk[
hjtk[
hjtk[
hjtk[
hjtkY
hjtkY
hjtkY
hjtkY
hjtkY
hjtkY
hjtkY
hjtk[
hjtk[
hjtk[
hjtk[
hjtk[
Output 8-87
u–UG–G u–UG–GšŒŒ‹šG {–›ˆ“Gu–U–G u–U–Gw“ˆ•›šG LG™ŒŠ–Œ™ G
šŒŒ‹šG
™ŒŠŒŒ‹G ‹šŠˆ™‹Œ‹
šŒŒ‹š
™ŒŠ–Œ™Œ‹
–G—“ˆ•›š
Y_
]
\]
XX
XY
X_
XZY
[]
X]_
Y[
Z]W
Z_
Y[
Z^
[^
YZ
\W
Y`
XX
\W
\
X
XZ
X^
ZY
Z
XY
Z
XZ
Y
Z
ZY
XY
^
\
Z\
]^
\_
Z
\
YW
W
\[
XZW
Y[
X
[`
X\
W
W
X
W
Y
Y
_
Z
^
]
W
]
W
X
X
[
Z
Z
W
X
Y
W
X
\
Z
Y
X
X
W
X
X
[
X
W
Z
[
X^
[
W
X
^
W
Z
^
_
W
^
Y
Y_
]
\\
XX
XW
X]
XY[
[Z
X]X
X_
Z]W
ZY
Y[
Z]
[]
X`
[^
Y]
XX
[`
Z
X
XY
XY
Y`
X
XX
Y
XZ
X
Y
Y_
XX
^
Y
ZX
\W
\[
Z
[
XZ
W
\X
XYZ
X]
X
[Y
XZ
\
\
YX
Z
X
`
\\
X\
ZX
^
X\_
YW
YW
Y^
Y`
X\
YZ
`
_
Z]
Y
X
Y
Z
X\
W
]
Y
XW
X
W
Y]
XW
]
X
Y\
Y`
Z[
Y
X
`
W
XZ
^^
X[
W
X`
_
X_
_Z
Z_
Y^
XW
\]
[[
Z\
X`
Z`
[[
]Z
_Z
^\
]Z
^`
[`
Z\
^Z
^Z
]^
XWW
X^
Y\
\Y
W
\\
XWW
^^
XWW
W
`Z
`X
_]
\W
_X
\_
]Z
]^
Y\
]`
W
Y\
]Z
__
W
[\
]Y
2003 Annual Report
Table 1 (contd)
u–UG–G
LG™ŒŠ–Œ™ G
—“ˆ•›šG
™ŒŠ–Œ™Œ‹
–G—“ˆ•›š
t–›Œ™
mˆ›Œ™
wœ™—–šŒ
u–UG–G
šŒŒ‹šG
™ŒŠŒŒ‹
u–UG–G
šŒŒ‹šG
‹šŠˆ™‹Œ‹
G{–›ˆ“Gu–UG
j–‹Œ
jyTZ^
jyTZ_
jyTZ`
jyT[W
jyT[X
jyT[Y
jyT[Z
jyT[[
jyT[\
jyT[]
jyT[^
jyT[_
jyT[`
jyT\W
jyT\X
jyT\Yh
jyT\Yi
jyT\Z
jyT\[h
jyT\[i
jyT\\
jyT\]
jyT\^
jyT\_
jyT\`
jyT]W
jyT]X
jyT]Y
jT[
jT[
jT[
jTX_
jTX_
jTX_
jTZZ
jTZ`
jTZ`
jTZ`
jTXY^
jTXY^
jTY[Z
jTY[Z
jTY[Z
jTY[Z
ztXYX`T`
jTY[Z
jTY[Z
t{hpG_
tiyhGXh
tjvsGYW\]
tjvsGYYW]
tthsG]]
t{hpGY
t{hpG_
t{hpG_
t{hpG_
jt[\^[T^
v~Y_WTX
ztXYX`T`
jt[\^[T^
tjvsGYW\]
tjvsGYYW]
jt[\^[T^
jtZZW]T[
jt[\^[T^
ztXYX`T`
j~]]T]W
ztX^[XTX
jt[\^[T^
j~]^T[Y
v~Y_WTX
ztXYX`T`
jTY[Z
tjvsGYYW]
t{hpG_
jTY[Z
jTX_
jTXY^
jTX_
jTX_
jTX_
jTX_
jTZZ
jTZ`
hjtkY
hjtkTw{u
hjtkY
hjtkY
hjtkT€y{
hjtkT€y{
hjtkY
hjtkX
hjtkY
hjtkY
hjtkThjy
hjtk[
hjtkY
hjtkThjy
hjtkTw{
hjtkY
hjtkY
hjtk[
hjtkX
hjtkX
hjtkT€y{
hjtkT€y{
hjtkT€y{
hjtkT€y{
hjtkT€y{
hjtkX
hjtkX
hjtkX
X\
ZX
]
Z
XY
ZY
Y_
X\
Y\
\
Y
Z
`
\
Z[
`[
Y
\
`
XXW
^
^
XY
XX
XZ
^
[
X_
W
Y
W
X
W
\
[
X
Z
W
W
X
Y
X
]
XW
W
X
W
YW
W
X
\
^
X
Y
Y
]
X\
Y`
]
Y
XY
Y^
Y[
X[
YY
\
Y
Y
^
[
Y_
_[
Y
[
`
`W
^
]
^
[
XY
\
Y
XY
XY
Y^
[
W
XY
X\
X\
XX
X\
[
Y
X
]
Z
YX
[Z
W
[
^
^W
\
Y
^
[
XW
\
W
XW
_W
`Z
]^
W
XWW
\]
]Z
^`
]_
_W
XWW
\W
_]
^\
^\
\X
WQ
XWW
^_
^_
^X
ZZ
XWW
XWW
_Z
XWW
WQ
_Z
YZX\
YX[
YXWX
XXY_
\[
{v{hs
–GšŒŒ‹š
Other activities in the tissue culture section this year include Embryo rescue of
173 and 123 seeds respectively from 2 S1 families (AM313 and AM320) of
cassava for molecular mapping of cyanogenic potential in collaboration with the
Swedish Agricultural University (SLU), Uppsala,
Conclusions and perspectives
The establishment CMD resistance breeding populations at CIAT for molecular
marker-assisted breeding of resistance has become routine. Although a severe
attack of mites reduced considerable the % recovery of plants, more than 1000
plants have been processed this year, compared to about less than 500 last
year. Future perspectives are to tackle the problems discovered this year with
mite infestation in the growth rooms, as well as the current size of the growth
room and raise this number to at least 5000 plants every year.
Project IP3: improving cassava for the developing world
Output 8-88
References
Okogbenin E. (2003). QTL mapping of early root yield, morphology and root
quality in cassava. Ph.D dissertation submitted to the University of Ibadan,
Ibadan, Nigeria.
Activity 8.24. Dissemination of Improved Cassava Varieties as Tissue
Culture plantlets
Collaborators: Adriana Alzate, Luis Guillermo Santos, Bernardo Ospina,
Hernan Ceballos, Martin Fregene (CIAT).
Funding: CIAT core funds
Important outputs
1) More than 3,500 plantlets of cassava were multiplied in the cassava tissue
culture facility and shipped to Latin America, Asia and Africa. Germplasm
is shipped within 2 months from when the request is received smaller
requests of 5-10 plants per genotype take a month.
Rationale
CIAT cassava project has responsibility to make available improved germplasm
to partners in the NARs in Asia, Latin America and Africa. The cassava tissue
culture facility was therefore used to propagate clean materials of improved
varieties for shipment to NARs in Latin America and Africa on request from
partners. The facility has also been used to clean-up and transfer into in vitro
field plants from the field that have were requested for by partners. An example
is the shipment to Vietnam of 57 genotypes selected from a diallel experiment
for further evaluation in Vietnam. We present a report of shipments of plants in
vitro of cassava made to several countries from February to September of the
2003.
Methodology
In vitro plants of the improved varieties were received from the genetic resources
unit (GRU) and multiplied using 4E media according to standard protocols
established at CIAT (Roca and Mroginski 1991). The method utilized for cleanup and transfer of field plants to tissue culture is that routinely used at CIAT
with some modifications, in summary: stems with apices and nodes are
obtained from plants in the field or green house, the leaves are removed leaving
approximately 5mm of petiole. The stem cuttings are washed with sufficient tap
water and cut into nodes and apices, with one node per piece. In the flow hood,
the fragments are placed into a 250ml sterile flask and washed with 70%
ethanol for 30 seconds with vigorous shaking. The ethanol is discarded and the
washed again with 10% sodium hypochlorite (0.5% v/v sodium hypochlorite)
and one drop of liquid soap for 5 minutes, with vigorous shaking. The
hypochlorite solution is discarded and the cuttings washed 3X with sterile
double distilled water. The cuttings are then transferred to a sterile petri dish
Output 8-89
2003 Annual Report
and, using a sharp sterile scapel, the extremes that have been in contact with
the solutions are eliminated. The stem cuttings are then planted in 4E medium
and placed in a growth room at 28oC with photo-period of 12h light and 12h
hours darkness to develop into full plants.
Results
A total of 3506 plants from a list of 39 improved genotypes were shipped to
collaborators in Haiti, Peru, Cuba, Nigeria, Nicaragua, India and the Dominican
Republic this year (Table 8.23). Aside from this, 5 plants of 44 genotypes were
also shipped to Crop Research Institute (CRI), Kumasi Ghana, for the Ph.D.
study of Ms Elizabeth Okai who is looking for heterotic patterns between
germplasm from Africa and Latin American based on SSR marker clustering.
Ten plants each of a selection of 57 genotypes from a diallel study was also
shipped to Vietnam for the Ph.D. study of Ms Cach who was at CIAT in 2002 for
6 months analyzing the same diallel experiment (Table 8.24). Stakes from the
57 genotypes in the field were planted in the sreen house, disinfected and nodal
cutting established in vitro in 4E media.
Conclusions and perspectives
The cassava tissue culture facility is being used to share valuable germplasm
with collaborators all over the world within a reasonable period of time,
germplasm is shipped within 2 months from when the request is received,
smaller requests of 5-10 plants per genotype take a month while larger ones
take up to 2 months. Future perspectives include an enlargement of thee
growth room to allow for the simultaneous processing of several shipment
requests.
References
Roca WM; Mroginski LA. 1991. Cultivo de tejidos en la agricultura:
Fundamentos y aplicaciones. [Tissue culture in agriculture: principles and
applications.] Centro Internacional de Agricultura Tropical (CIAT), Cali,
Colombia. 970 p.
Project IP3: improving cassava for the developing world
Output 8-90
Table 8.23. Summary of shipments of in vitro plants of cassava (Manihot
esculenta Crantz) made to several countries from February to September
2003.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
COUNTRY/ DATE OF SHIPMENT
GENOTYPE HAITI
PERU
CUBA
NIGERIA
May
May 6/03 May 7/03
23/03
Jun 11/03
# Plants # Plants
# Plants # Plants
BRA 383
30
30
5
CM 1335-4
8
CM 2772-3
15
5
5
8
CM 3306-4
30
30
5
CM 3750-5
CM 4574-7
30
30
5
CM 4843-1
30
30
5
8
CM 489-1
CM 4919-1
30
30
5
8
CM 507-37
30
30
5
8
CM 523-7
30
30
5
CM 5306-8
30
30
5
8
CM 6119-5
30
8
5
8
CM 6438-14 30
30
5
8
CM 6740-7
30
30
5
8
CM 6754-8
CM 6921-3
30
30
5
8
CM 7033-3
30
18
5
8
CM 7073-7
30
30
5
8
CM 7514-8
30
30
5
8
CM 7951-5
30
30
5
8
CM 8027-3
30
30
5
8
HMC-1
MCOL 1468 6
MCOL 1684 MCOL 2063
(Secundina) MCOL 1734 MCOL 2215 8
PER 183
30
30
5
SM 1411-5
8
SM 1433-4
30
30
5
8
SM 1460-1
30
30
5
8
SM 1565-15 30
30
5
8
SM 1741-1
8
SM 1821-7
30
20
5
8
SM 2019-9
30
5
SM 805-15
30
30
5
SM 909-25
30
30
5
8
TAI-8
30
30
5
TOTAL
795
716
130
190
Output 8-91
NICARAGUA INDIA
Rep. DOMINICANA
Jul 18/03
# Plants
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
-
Sep 2/03
# Plants
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
Sep 10/03
# Plants
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
-
30
30
30
30
30
30
30
30
780
10
10
10
10
10
10
10
10
10
10
10
10
370
5
20
20
20
20
20
20
20
20
525
2003 Annual Report
Table 8.23. List of plants rescued from the greenhouse and placed in vitro for
shipment to Vietnam (Plants for Ms Cach)
Genotype
CM 9106-18
CM 9106-7
CM 9148-2
CM 9148-3
CM 9703-17
CM 9703-24
CM 9907-1
CM 9907-3
CM 9921-25
CM 9923-1
Genotype
CM 9923-30
CM 9926-17
CM 9945-22
CM 9945-27
CM 9946-11
CM 9946-12
CM 9949-1
CM 9949-25
CM 9952-1
CM 9952-19
Genotype
CM 9954-18
CM 9954-23
CM 9957-1
CM 9957-21
CM 9958-1
CM 9958-6
CM 9966-27
GM 236-26
GM 236-7
GM 237-13
Genotype
GM 237-22
GM 238-29
GM 246-25
GM 246-3
GM 247-27
GM 248-26
GM 250-24
GM 250-29
GM 251-12
GM 251-9
Project IP3: improving cassava for the developing world
Genotype
GM 255-2
GM 258-2
GM 258-3
GM 259-3
GM 266-24
GM 272-1
GM 272-4
GM 273-13
GM 280-15
GM 280-24
Genotype
GM 281-24
GM 281-28
GM 287-10
GM 287-13
GM 289-9
GM 291-11
GM 291-7
Output 8-92
Activity 8.25. Updating CassavaDB, a ACEDB-type data base for Results
of Genome Mapping
Collaborators: C. Buitrago; F. Rojas; J.Tohme, M. Fregene (CIAT)
Funding: CIAT core funds
Important output
1) The cassava genome mapping database CassavaDB, an AceDB style data
base, is being updated with new information from mapping studies
Rationale
The cassavaDB is an AceDB-type genome database designed specifically for
handling bioinformatic data flexibly. It includes tools designed to manipulate
genomic data that is
graphic, flexible and portable. It can be operated on various Unix workstations
(SUN, DEC, NEXT, SGI ...), The first principle of the program is that any piece
of data stored in AceDB can very easily be exported in flat ascii files to be used
by other programs.
The second principle of the program is that write acces to the database is
organised in macro transactions that is called sessions. All the information is
stored in objects, which fall into a number of classes. The classes are standard
units such as genes, alleles, strains, clones, papers, authors, journals, etc., and
the names are in most cases the standard names. The CassavaDB was initially
hosted at the plant databases at the National Agricultural Library at
http://probe.nal.usda.gov. This was later moved to the ARS Genome Database
Resource (GDR) server formerly at http://ars-genome.cornell.edu which is now
permanently off-line. At the moment it is hosted at http://ukcrop.net/, the
website of the UK Crop Plant Bioinformatics Network (UK CropNet) established
in 1996 as part of the BBSRC's Plant and Animal Genome Analysis special
initiative, the focus is the development, management, and distribution of
information relating to comparative mapping and genome research in crop
plants.
Since CassavaDB was developed in 1998, hundreds of additional markers,
predominantly SSR markers have been developed and mapped in cassava. This
information is currently not in the database. Also not in the database are
additional genetic maps of cassava made since the first map was published in
1997. Data from several QTL mapping studies have also since become available
since then. There is also a need to have the CassavaDB situated on a local
server at CIAT to enable more frequent updates and also make it more visible to
the international cassava community.
Activities to update CassavaDB.
A compilation of all cassava SSR marker mapping data and map location data
from work done by Mba (et al. 2001), Zarate (2002) Garcia (2002) and
unpublished or undocumented work was initiated in August 2003. This
information was formatted according to requirements for AceDB. Similarly,
Output 8-93
2003 Annual Report
graphics of parental surveys, progeny data and maps were developed from data
from the three mapping studies and prepared for CassavDB. All the above
information will be up-loaded into CassavaDB in collaboration with the CIAT
bioinformatics unit and maintained locally at CIAT. A link to the above will be
made on the cassava web page of the CIAT website.
References
Mba, R.E.C., Stephenson, P., Edwards, K., Melzer, S., Mkumbira, J., Gullberg,
U., Apel, K., Gale, M., Tohme, J. and Fregene, M. (2001) Simple Sequence
Repeat (SSR) Markers Survey of the cassava (Manihot esculenta Crantz)
Genome: Towards an SSSR-Bassed Molecular Genetic Map of Cassava.
Theoretical and Applied Genetics. 102: 21-31.
Garcia, T.. 2002. Mapeo genético de una población F1 de yuca cultivada (
Manihot esculenta Crantz) utilizando Microsatélites que proviene de cDNA.
Tesis de pregrado. Universidad del Valle, Cali. Colombia.
Zarate, L.A. 2002. Mapeo genético de una población F1 de yuca cultivada (
Manihot esculenta Crantz) utilizando Microsatélites. Tesis de pregrado.
Universidad del Tolima. Ibagué. Colombia.
Activity 8.26. Associating Horn Worm Resistance in 60444 (MNG11) with
Introgression from Manihot Glaziovii.
Collaborators: A. Bellotti, P. Chavariagga, N. Morante, H.Ceballos, M. Fregene
Funding: CIAT core funds
Important output
1) Development of gene tagging populations to associate hornworm resistance
in the variety 60444 with introgression from Manihot glaziovii
Rationale
The cassava hornworm (Erinnyis ello) is generally considered the most
important pest of cassava inn the Americas (Bellotti 1981). High populations
can defoliate large plantations in a short time. No naturally occurring
resistance to this important pest has been found in the cassava germplasm
collection. Recently, feeding experiments with leaves of the genotype 60444
(MNg11) transformed with the Bacillus thurigensis protein CRY1Ab, and leaves
from non-transformed plants revealed a very high mortality of the hornworm
when fed with non transgenic plants (Chavariagga et al. 2003, unpublished
data). The above observation suggests a naturally occurring resistance to
hornworm in the genotye 60444. This genotype was developed at the Moor
plantation, Ibadan, Nigeria in the 1950s using third back cross derivatives of
the inter-specific cross between cassava and M. glaziovii for the production of
CMD resistant materials. Resistance to the hornworm could have been
inadvertently introgressed from M. glaziovii (Bellotti 2003, personal
communication).
Another progeny of the M. glaziovii back cross derivatives, TMS30572, was used
in constructing the molecular genetic map of cassava. Regions suspected to be
introgression of large chunks of the M. glaziovii genome, as demonstrated by
Project IP3: improving cassava for the developing world
Output 8-94
suppression of recombinantion have been noted (Fregene et al. 1997). One of
these regions on linkage group D has been shown to bear resistance QTLs for
CMD and CBB. A study has therefore been initiated to associate this region of
the genome with resistance to hornworm resistance.
Methodology
The genotype 60444 has not been used in breeding at CIAT and there are no
remnant seeds from previous crosses for genetic studies of hornworm
resistance, segregating populations will therefore be developed. The breeding
program at the Moor plantation, Ibadan utilized principally open pollinated
seeds in their breeding activities and the likelihood that some inbreeding may
have occurred in 60444 cannot be ruled out. The development of segregating
populations for mapping resistance to hornworm resistance has to take into
consideration several possibilities of gene action. In light of this, a single fullsib family, by crossing MNg11 and the hornworm susceptible variety MCol
2215, and an S1 family, by selfing 60444 will be developed.
Results
Ten in vitro plants of the genotype 60444 were hardened in the screen house
and transferred to the field for genetic crosses for the development of mapping
populations. It is expected that seeds from the populations will be ready by
May/June next year and sexual seeds from the families will immediately be
germinated and transferred to the field. Once woody stakes will be harvested
from the progenies once they become available and planted in the green house
for the feeding studies.
Conclusions and perspectives
A project to associate the linkage group D region of the cassava genome in the
genotype 60444 with resistance to hornworm resistance has been initiated. It
involves the generation of a full-sib and S1 families from 60444 and their
evaluation for hornworm resistance and molecular markers from the D region of
the cassava genome.
References
Fregene, M.A., F. Angel, R. Gómez, F. Rodríguez, W. Roca, J. Tohme, and M.
Bonierbale (1997). A molecular genetic map of cassava (Manihot
esculenta Crantz). Theor. and Appl. Genet. TAG 95 (3) 431-441.
Belotti 1981. Cassava hornworm In: Lozano J.C., Bellotti, A., Reyes J.A.,
Howeler R., Leihner D., and Doll J. (eds) Field Problems in Cassava.
CIAT, AA6713, Cali Colombia. Pp84-85
Output 8-95
2003 Annual Report
Activity 8.27. Development of Populations Tolerant to Inbreeding
Depression in Cassava
Collaborators: N. Morante, H. Ceballos, M. Fregene.
Funding: CIAT core funds
Important output
1) Development of S2 populations from S1genotypes that flower profusely
Rationale
A S2 recurrent selection program was set-up for the development of cassava
populations tolerant to inbreeding. The principal reason for the program is the
development of populations tolerant to inbreeding and for the development of
pure cassava lines via the doubled haploid technology. A second reason is the
identification of genetic stocks for gene mapping studies, for example excellent
segregation of beta-carotene content and cyanogenic potential (CNP) was
observed in 2 of the S1 families from MCol 72 (beta-carotene) and MTAI8 (betacarotene and CNP). A preliminary selfing of an initial 14 genotypes produced
300 S1 lines that were evaluated in a clonal observation trial last year
(CIAT2002). More than 30 of those S1 lines that flowered profusely were selfed
to produce S2 families. We describe their production and establishment in the
field. Furthermore additional larger sized S1 families have been developed from
20 varieties and have been established in a seedling trial this year.
Methodology
About 30 S1 progenies were selfed and seeds harvested at 40-60 days after
pollination, the early harvest of the seeds was due to the mandatory removal of
all cassava plants in CIAT before the beginning of the “zero cassava” month at
CIAT, one of the measures adopted to control the incidence of white flies at
CIAT. The S2 families were established from embryo rescue as described in
above in the section on the development of mapping populations in cassava.
After one round of multiplication in vitro they were transferred to the screen
house for hardening and to the field.
Results
An initial 14 cassava genotypes were chosen for the development of populations
tolerant to inbreeding. The genotypes were chosen due to their good general
combining ability performance for yield, dry matter yield or root quality. They
include the following lines: MCOl22, CM523-7, MCOL1684, MBRA12,
MCOL2060, MVEN77, MCOL1522, MTAI1, MPAN51, MECU169, MCOL1468,
MCOL72, CM849-1, HMC1. More than 300 pollinations were made per
genotype and between 30-150 seeds were obtained per genotype. During a
clonal observation of the above families, selfings were carried out and 15 small
sized S2 families, of 2-10 progenies were developed (Table 8.24). These seeds
were germinated from embryo axes. Between 5 and 10 plants per S2 progeny
was hardened in the green house and transferred to the field in July this year.
Project IP3: improving cassava for the developing world
Output 8-96
Table 8.24. Summary of S2 families developed from embryo rescue of selfed S1
individuals
Date of
transplant
Grand
Parent
Parent
Code
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jun-03
Jul-03
Jul-03
Jul-03
Jul-03
Jul-03
Jul-03
Jul-03
Jul-03
Jul-03
Total
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
MCol1505
HMC 1
HMC 1
HMC 1
HMC 1
HMC 1
HMC 1
HMC 1
HMC 1
CM849-1
AM244-10
AM244-16
AM244-16
AM244-31
AM244-31
AM244-31
AM244-35
AM244-35
AM244-35
AM244-35
AM244-38
AM244-40
AM244-101
AM244-109
AM244-109
AM244-109
AM244-109
AM244-109
AM244-109
AM244-109
AM244-109
AM24-135
AM24-135
AM24-135
AM266-21
AM266-41
AM266-41
AM266-41
AM266-45
AM266-50
AM266-76
AM266-76
AM312-42
AM 244-10-5
AM 244-16-1
2 AM 244-16
AM 244-31
1 AM 244-31
2 AM 244-31
AM 244-35
1 AM 244-35
2 AM 244-35
3 AM 244-35
AM 244-38
1 AM 244-40
1 AM 244-101
1 AM 244-109
4 AM 244-109
5 AM 244-109
6 AM 244-109
7 AM 244-109
8 AM 244-109
10 AM 244-109
12 AM 244-109
1 AM 244-135
6 AM 244 135
7 AM 244-135
4 AM 266-21
1 AM 266-41
5 AM 266-41
6 AM 266-41
2 AM 266-45
4 AM 266-50
1 AM 266-76
6 AM 266-76
1 AM 312-42
No. of
Plants
2
1
1
5
5
5
2
4
7
3
1
2
3
2
5
5
2
5
2
1
4
1
2
3
2
1
2
1
7
2
1
1
5
95
At 10 months after planting, the progenies will be evaluated for fresh root yield,
dry matter content, foliage weight, harvest index, culinary quality, starch
content/quality, and frog skin disease according to standard CIAT procedures.
Output 8-97
2003 Annual Report
Conclusions and perspectives
The development of populations tolerant to inbreeding has continued with the
development of 15 small sized S2 families. These families will be evaluated at
10 months after planting.
References
CIAT (International Center for Tropical Agriculture), 2001. Annual Report
Project IP3: Improved cassava for the developing world. Cali, Colombia.
Activity 8.28. A Simple Method for the Rapid Multiplication of Clean
Cassava Planting Material
Collaborators: Dr James George (CTCRI, Trivandrum, India), Elizabeth Okai,
Emmanuel Okogbenin (CRI, Ghana), Chris Okeke (NSM,
Nigeria); Dr (Mrs) Acheampong (University of Legon, Ghana),
Bernardo Ospina (CLAYUCA), Martin Fregene
Funding: CIAT core funds and NSM
Important outputs
1) A simple but effective cassava stake rapid multiplication method, based
on 2-node cuttings, has been implemented for the Nigerian Starch Mill
(NSM) to supply their need to clean planting material
2) Forty four improved varieties from IITA are being multiplied to establish a
seed bank for next season planting
Rationale
Good quality and healthy planting material is crucial for high yields in cassava.
Cuttings obtained from diseased and/or pest infested plants can reduce yield
by 30% to 80% (Guritno1985). Another problem with unhealthy planting
material is sprouting. The ability of stakes to sprout is closely related to their
starch content at planting and growth during the first 20 days after plating is
exclusively at the expense of the nutritive reserve previously accumulated in the
stakes (Molina et. al. 1995). Starch content of stakes is reduced with poor soil
fertility and disease/pest attack. A 20% drop in sprouting and a yield loss of
10% due to the use of poor quality planting material in a 2000ha farm
translates to a loss of 8,400 metric ton of fresh cassava roots, at the national
average yield of 14t/hectare. At a conservative estimate of US$50/ton this is a
loss of US$0.42 million. Such a loss quickly wipes out the profit margin and
puts the entire venture at risk.
As part of its assistance package to the Nigerian commercial cassava production
company, the Nigeria Starch Mills (NSM), CIAT and CLAYUCA agreed to provide
technical know-how on rapid multiplication of clean planting material. In
Nigeria, commercially available cuttings are often taken from plants that are
diseased, damaged by insect and pests and inadequately fertilized with a
consequent reduction in yield (Okeke 1994, Ospina and Fregene, Personal
observation 2003). Tissue culture of cassava meristematic tissue has been
successfully applied to obtain in vitro cassava plants that are disease- and pest-
Project IP3: improving cassava for the developing world
Output 8-98
free (Roca et al. 1991). The technology also permits the mass production of in
vitro plants compared to traditional multiplication methods. A multiplication
ratio of 100 is possible using new efficient multiplication systems, such as the
automated temporary immersion systems (ATIS) or “bioreactors”. Apart from the
production of healthy planting materials, rapid tissue culture multiplication can
be used to mass-produce and deploy a new promising variety over a short
period. However, the initial high start-up costs of the automated temporary
immersion system and support tissue culture facilities increases the costs of
establishing a commercial farming operation and diverts funds from other
much need activities, making it less attractive compared to other multiplication
schemes.
To reduce costs, a cheaper tissue-culture based rapid multiplication scheme
that combines an initial step of tissue culture multiplication, hardening, 4
months of field growth, then a 2-node multiplication scheme in a special
nursery, as practiced at CTCRI, Trivandrum, India, followed by another cycle of
field growth and multiplication in the special nursery was proposed to NSM.
The rapid multiplication scheme is expected to produce 2-3 million plants by
July 2004, with enough cuttings to plant a 200ha “seed bank” to be used as a
source for good quality healthy planting material. The scheme can also become
a commercial source of planting material for sale to other large scale
multiplication projects, for example the pre-emptive management of the cassava
mosaic disease (CMD) projects that is being initiated in Nigeria by the
Government of Nigeria, NDDC, some oil companies, and USAID in collaboration
with IITA, Ibadan..
Specific objectives were:
i)
Tissue culture rapid multiplication of 44 improved IITA varieties to
produce at least 200 plantlets per genotype
ii)
Hardening of the plants in the green house and transfer to the field
iii)
Harvest of 2-node cuttings at 4 months after planting from the field and
transfer to the rapid multiplication nursery; transfer of plants from the
nursery to the field after one month.
iv)
Harvest of 2 node cuttings from the new plants at 4 months after
planting and transfer to nursery; transfer of plants from the nursery to
the field after one month
Methodology
Five stakes from 44 IITA improved cassava genotypes were obtained from Alfred
Dixon, IITA, Ibadan, for tissue culture multiplication. A list of the improved
lines is shown in Table 8.25. A request to multiply the materials at the tissue
culture facility of the University of Legon, Accra was granted and multiplication
began first week of July. The five stakes were planted in plastic bags and after
2 weeks, meristematic or nodal cuttings was harvested, cleaned with sodium
hypochlorite and cultured in 1/2 MS media supplemented with BAP and GA
(4E media). After 4, 8, and 12 weeks after planting the plantlets were sub
cloned to obtain the target of 200-250 plants per genotype.
Output 8-99
2003 Annual Report
Hardening of the plants will be in the NSM screen house (65% shade), currently
under sonstruction, using a soil mixture of 3 parts sand and 1 part top-soil in
black plastic bags. Tissue culture plants received will be gently removed from
the glass tubes and placed in the plastic bags with soil, a fungicide will be
applied to control fungus growth and plants fertilized with a commercial
fertilizer rich in phosphorus. A stryofoam cup with holes will be used to provide
the plantlets with high humidity. After one month in the screen house, the
plants will be transferred to the field and watered regularly. About 500kg of
NPK 15:15:15 fertilizer will be added in split applications at 1 and 2 month after
transplanting to the field in bands.
After 4 months of growth, 2-node cuttings will be obtained from each plant
using a sharp knife and transferred to the special covered nursery. The plants
will be watered regularly and at 1 month after planting they will be transferred
to the field. The special nursery is simply a canopy of a mesh that allows in
only 65% of light while excluding the rest over a nursery bed. The shade of the
canopy, combined with constant watering provides an area of high humidity
adequate for the fast growth of cassava. During this period, special care is
taken to rouge out diseased plants and to keep the area free of pests by
spraying appropriate pesticides.
After the plants multiplied in the special nursery have grown, another round of
multiplication is carried out as described above this is the final round of
multiplication. The plants are then allowed to grow until maturity in the field
and used as a seed bank of planting material for storage root production
Results
Ten to 20 plantlets of 44 improved varieties from IITA were established from
nodal cuttings obtained from 2 week old potted plants in tissue culture (4E
media) at the University of Legon. After 4 weeks, the plantlets were subcloned
to obtain a 3 to 5 time multiplication of the original plantlets. Two further
rounds of multiplication were conducted at 8 and twelve weeks to give between
200 and 250 plants per genotype. The plants are ready for green house
hardening and are awaiting the completion of the NSM green house facility
under construction at Ihiala, Nigeria. In the mean time an import permit to
bring the plants from Ghana into Nigeria has been applied for via the Nigerian
Plant Quarantine office and a phyto-sanitary certificate has been requested
from the Ghanian authorities to ship the plants to Nigeria.
Project IP3: improving cassava for the developing world
Output 8-100
Table 8.25. List of 44 improved new cassava varieties that have been multiplied
by tissue culture
____________________________________________________________________________________________
Cassava mosaic Bacterial
Root
CNP (mg HCN/ Root dry Fresh root
Flesh
Genotype
disease
blight
mealiness 100g fresh root Matter(%) yield (t/ha) color
____________________________________________________________________________________________
1. 92/0057
Resistant
Resistant Mealy
Medium
30
25-30
White
2. 92B/00068 Resistant
Resistant Mealy
Medium
32
30-35
White
3. 92/0326
Resistant
Resistant Mealy
Low
30
30-35
White
4. 93/0098
Resistant
Resistant Mealy
Medium
30
32-35
White
5. 92/0325
Resistant
Resistant Mealy
Low
35
20-25
White
6. 97/0162
Resistant
Resistant Mealy
Low
30
30-35
White
7. 97/4769
Resistant
Resistant Mealy
Low
30
30-35
White
8. M98/0028 Resistant
Resistant Mealy
Medium
30
30-35
White
9. 98/0505
Resistant
Resistant Mealy
Medium
35
30-40
White
10. 98/0510
Resistant
Resistant Mod. mealy
Medium
35
40-45
White
11. 99/1590
Resistant
Resistant Mod. mealy
Low
35
25-32
White
12. 99/6012
Resistant
Resistant Mod. mealy
Low
35
35-40
White
13. M98/0040 Resistant
Resistant Mod. mealy
Low
32
40-45
White
14. 91/02324 Resistant
Resistant Non-mealy
Medium
35
35-45
White
15. 92/0067
Resistant
Resistant Non-mealy
Medium
30
25-30
White
16. 92B/00061 Resistant
Resistant Non-mealy
Medium
30
30-35
White
17. 94/0561
Resistant
Resistant Non-mealy
Medium
30
30-35
Yellow
18. 94/0026
Resistant
Resistant Non-mealy
Medium
32
30-35
White
19. 94/0039
Resistant
Resistant Non-mealy
Medium
32
30-40
White
20. 95/0166
Resistant
Resistant Non-mealy
Medium
30
35-40
White
21. 95/0379
Resistant
Resistant Non-mealy
Medium
30
30-35
Yellow
22. 95/0289
Resistant
Resistant Non-mealy
Medium
32
30-35
White
23. 96/1565
Resistant
Resistant Non-mealy
Medium
30
35-40
White
24. 96/1089A Resistant
Resistant Non-mealy
Medium
32
30-35
White
25. 96/0603
Resistant
Resistant Non-mealy
Medium
30
30-35
White
26. 96/1642
Resistant
Resistant Non-mealy
Medium
30
30-35
White
27. 97/3200
Resistant
Resistant Non-mealy
Medium
32
35-40
White
28. 97/2205
Resistant
Resistant Non-mealy
Medium
30
30-35
White
29. 97/4763
Resistant
Resistant Non-mealy
Medium
32
35-40
White
30. 98/2226
Resistant
Resistant Non-mealy
Medium
30
30-35
White
31. 98/0002
Resistant
Resistant Non-mealy
Low
35
40-45
White
32. 97/0211
Resistant
Resistant Non-mealy
Medium
30
30-35
White
33. 96/1569
Resistant
Resistant Non-mealy
Medium
30
30-35
White
34. 96/1632
Resistant
Resistant Non-mealy
Medium
40
35-45
White
35. 97/4779
Resistant
Resistant Non-mealy
Medium
30
30-35
White
36. Z97/0207 Resistant
Resistant Non-mealy
Low
30
35-40
White
37. 98/0581
Resistant
Resistant Non-mealy
Medium
30
30-32
White
38. 98/2101
Resistant
Resistant Non-mealy
Medium
30
30-35
White
39. M98/0068 Resistant
Resistant Non-mealy
Medium
35
40-45
White
40. 99/1903
Resistant
Resistant Non-mealy
Medium
39
40-45
White
41. 99/2123
Resistant
Resistant Non-mealy
Low
35
30-32
White
42. 99/3073
Resistant
Resistant Non-mealy
Medium
30
30-35
White
43. 96/0523
Resistant
Resistant Non-mealy
Medium
30
35-40
White
44. 96/1317
Resistant
Resistant Non-mealy
Medium
32
30-35
Yellow
____________________________________________________________________________________________
Output 8-101
2003 Annual Report
Conclusion and perspectives
A rapid multiplication of clean improved planting material has been embarked
upon for the Nigeria Starch Mill (NSM), if successful it could be a very rapid
method for the production of clean planting material for both the medium and
large scale cassava production sector in Nigeria as well as small scale rural
farming. Although the scheme is still in progress, expected outputs is a 400X
multiplication within a single growing season compared to the traditional 10X
multiplication or 100X via tissue culture alone or 200X via 2-node cuttings
alone.
References
Cock J.H. 1985. Cassava: New Potential for a Neglected Crop. Westview
Boulder, CO
Guritno B. Influence of planting material on plant performance in cassava.
Ph.D. thesis, University of Brawijaya, Malang, Indonesia, 158pp.
Molina J.L. and El-Sharkawy M.A. 1995. Incresing crop productivity of cassava
by fertilizing production of planting material. Field Crops Research
44:151-157
Okeke J.E. Productivity and yield stability in cassava (Manihot esculenta Crantz)
as affected by stake weight. J. Agric. Sci., Cambridge, 122:61-66
Roca WM; Mroginski LA. 1991. Cultivo de tejidos en la agricultura:
Fundamentos y aplicaciones. [Tissue culture in agriculture: principles and
applications.] Center Internacional de Agricultura Tropical (CIAT), Cali,
Colombia. 970 p.
Activity 8.29 TRIPS
1. March 2003 Trip to CRI, Kumasi, Ghana, on the collaborative project: “
Analysis of genetic diversity of local land races from Africa and Latin
America and the search for heterosis”, being carried out as a Ph.D. study
by Ms Elizabeth Okai.
2. April 2003 Trip to NARO Namulonge, Uganda, on the collaborative
project “ Genetic mapping of cyanogenic potential (CNP) in cassava, being
carried out as a Ph.D. study by Ms Elizabeth Kizito, under the BIOEARN
program (SLU with Sida funding)
3. May 2003 Trip to ARI, Mikocheni, Tanzania, on discussions of the project
“A Molecular marker-assisted, farmer-partipatory breeding project to
improve local cassava varieties in Tanzania with resistance to pest and
diseases”
4. May 2003 Trip to ILRI, Nairobi, to visit the bioscience facility particularly
the molecular marker capacity and ongoing cassava work (by Dr Morag
Ferguson, IITA)
5. May2003 Trip to RF-Nairobi, in company of Dr Ferguson to present to Dr
Joe DeVries, RF Assistant Director of Food Security, outcome of the
meeting in Tanzania
6. June 2003. Visit to Nigeria, in company of Bernardo Ospina and Luis
Fernando Cadavid to assess the farming and seed multiplication
operations of the Nigerian Starch Mills (NSM)
Project IP3: improving cassava for the developing world
Output 8-102
7. August 2003. Visit to the John Innes Center (JIC), Norwich, to attend a
one day conference in honour of Prof Mike Gale on his retirement from
JIC; and to discuss a collaborative project with Prof Allison Smith
8. September 2003. Visit to Swedish Agricultural University (SLU),
Uppsala, to attend the 2nd MOLCAS meeting and discuss with Dr Eva
Ohlsson of Sida a proposed visit in October by Dr Hernan Ceballos and
Dr Joachim Voss.
9. September 2003. Visit to Mukono, Uganda on the work planning
meeting of the cassava biofortification challenge program (BCP)
10.September 2003. Return visit to CRI, Kumasi, to review progress made
with establishing the crossing block in the project “ Analysis of genetic
diversity of local land races from Africa and Latin America and the search
for heterosis”
Activity 3.30 TRAINING
Graduate students
x Ms Elizabeth Okai (Ghana) Ph.D. student, University of the Free
State, Bloemfontein, South Africa (6 months, left January 2003)
x Ms Elizabth Kizito (Uganda) Ph.D. student, University of the Free
State, Bloemfontein, South Africa (June-December 2003)
x Henry Ojulong (Uganda) Ph.D. student University of the Free
State, Bloemfontein, South Africa (April 2002 – April 2005)
x Martha Isabel Moreno (Colombia) M.Sc. student, Universidad de
Valle, Cali (September 2002 – August 2003)
x Mr Lekan Akinbo (Nigeria) Ph.D. student University of the Free
State, Bloemfontein, South Africa (November 2003 – November
2006)
x
Under-graduate students
x Gina Puentes (Colombia), B.Sc. student, Universidad Nacional,
Palmira (completed March 2003)
x Wilson Castel Blanco (Colombia) B.Sc. student, Universidad
Nacional, Bogota (Jan-Dec 2003)
x Liliana Cano (Colombia) B.Sc. student, Universidad de Santander,
Bucaramanga (Jan-Dec 2003)
x Ana-Maria Correa Colombia) B.Sc. student, Universidad de Valle
(June 2003 – May 2004).
Visiting Researchers
x Joel Beovides, INIVIT, Cuba. CBN small grant recipient (June –
December 2003)
x Dr Chiedozie Egesi, NRCRI, Nigeria. One week on breeding methods
at CIAT after the doubled haploid workshop (June 2003).
Activity 3.31 PUBLICATIONS
Fregene M., Suarez M., Mkumbira J. , Kulembeka H. , Ndedya E. , Kulaya A. ,
Mitchel S. Gullberg U. , Rosling H., Dixon A., Kresovich S. (2003) Simple
Sequence Repeat (SSR) Diversity of Cassava (Manihot esculenta Crantz)
Landraces: Genetic Structure in a Predominatly Asexually Propagated
Crop Theor Appl Genetics 107:1083-1093
Output 8-103
2003 Annual Report
Okogbenin E. and Fregene M (2001) Genetic Mapping of QTLs Affecting
Productivity and Plant Architecture in a Full-Sib Cross from Non-Inbred
Parents in Cassava (Manihot esculenta Crantz). Theor Appl Genetics (in
presss)
Anderson J., Delseny M., Fregene M., Jorge V., Mba C., Lopez C., Restrepo C.,
Piegu B., Verdier V., Cooke R., Tohme J., Horvath D. 2003. An EST
Resource for Cassava and Other Species of Euphorbiaceae. Plant
Molecular Biology (submitted).
Fregene M., Mba C., Buitrago C., Zarate A., Garcia T., Tohme J. 2003 A
Predominantly Simple Sequence Repeat (SSR) Marker Map of Cassava
(Manihot esculenta Crantz). Plant Molecular Biology (submitted)
Tomkins J., Fregene M., Main D., Kim H., Wing R., and Tohme J. 2003.
Bacterial Artificial Chromosome (BAC) Library Resources for Positional
Cloning of Pest and Disease Resistance Genes in Cassava (Manihot
esculenta Crantz). Plant Molecular Biology (submitted)
Fregene M., Matsumura H., Akano A., Dixon A., Terauchi R. 2003. Serial
Analysis of Gene Expression (SAGE) of Host Plant Resistance to the
Cassava Mosaic Disease Resistance (CMD) Plant Molecular Biology
(submitted)
Activity 3.32 Projects funded and in review with donors
1. A molecular marker-assisted, farmer-partipatory breeding project to improve local
cassava varieties in Tanzania with resistance to pest and diseases (Rockefeller
Foundation) US$181,000 for three years (2003 – 2005)
2. Genetic Mapping of the Linamarin biosynthetic genes CYPD1 and D2 and the
development of markers for CNP in cassava, in collaboration with Prof Birger
Moller, Royal Agriculture and Veterinary University, Copenhagen (DANIDA),
US$33,000 for three years (2003-2005
The following projects have been submitted to donors
1. The molecular diversity network of cassava (MOLCAS) 2004-2006 (IPICs,
Uppsala, Sweden) US$60,000 for three years
2. Mutagenesis of Cassava (Manihot esculenta Crantz) for the Generation,
Identification and Molecular Analysis of Novel Traits. Research Contract
submitted to the International Atomic Energy Agency (IAEA), Vienna, Austria
with the National University of Colombia, Palmira US$31,500 for three years
(2003-2005)
3. Workplan in sub programs 1, 2, 3 and 4 of the Genetic Resources project for 2004
(A total of US$70,000).
Project IP3: improving cassava for the developing world
Output 8-104
OUTPUT 9
Integrated cassava-based cropping systems in Asia: Widespread adoption of farming
practices that enhance sustainability.
The overall objective of this output is to increase the income and agricultural
sustainability in less favored upland areas by developing, together with farmers, efficient
and effective integrated soil and crop management practices that optimize total farm
productivity and contribute to the long-term sustainability of cassava-based cropping
systems in Asia.
Activity 9.1 Soil fertility maintenance through the application of chemical
fertilizers, or the use of intercropping, green manuring, alley cropping
and crop rotations.
Rationale
Because of the near absence of diseases and pest problems in Asia, cassava is often
grown continuously on the same land for many years. But, most cassava soils have a low
inherent fertility. The opening up of land for cultivation of annual crops leads to
exposure of the soil surface to high temperatures resulting in rapid decomposition of
organic matter, while the direct impact of rainfall on the soil surface may destroy soil
aggregrates and lead to runoff and erosion. Continuous cropping with the removal of
cassava roots (and sometimes stems and leaves as well) will lead to depletion of soil
nutrients. Unless these nutrients are replaced in the form of chemical fertilizers, animal
manures or green manures, soil fertility will decrease and productivity decline.
Specific Objectives
a) To determine the immediate and long-term effect of various combinations of N, P and K
applied annually on cassava yields and starch content, as well as on soil fertility.
b) To determine the long-term effect of various green manures on cassava yields and soil
fertility.
9.1.1 Long-term NPK trials
Materials and Methods
Many long-term NPK trials were initiated in several countries in Asia during the late 80s.
Most have been terminated but four trials continue in their 12th to 14th year of cropping.
These trials are located in Thai Nguyen University in north Vietnam; at Hung Loc Agric.
Research Center in Dongnai, south Vietnam; at CATAS in Hainan province of China; and
in Tamanbogo in Lampung province of Indonesia. Usually, two cassava varieties are
Output 9-1
2003 Annual Report
grown in each plot, which receives the same fertilization year after year. Most trials have
12 treatments with various combinations of four levels of N, P and K in an incomplete
factorial design with four replications. Soil samples are taken after land preparation and
before the next planting; these samples are air dried and analyzed for soil fertility
parameters.
Results
Figure 1 shows the effect of various combinations of N, P and K on cassava yield and leaf
life of two varieties grown for the 13th consecutive year in Thai Nguyen University. Both
varieties responded to the application of all three nutrients, but most dramatically to
application of K. Without K yields had declined to 3.5 t/ha while with K yields were
maintained at 20-25 t/ha, which is only slightly below that obtained in the first year, i.e.
26 t/ha. Leaf life at 3 months after planting (MAP) also increased with applications of all
nutrients, but particularly with the first increment of K. Leaf life could be extended from
about 30 days to 90 days with adequate fertilization. Figure 2 shows similar data for
Hong Loc Center, with again the most marked responses to the application of K. This
nutrient also increased the root starch content from about 19% to 26.5%. Figure 3
shows the absolute yield, the relative yield and the change in soil K and P levels during
the 13 crop cycles in Hung Loc Center. There were no significant responses to any of the
nutrients during the first four years, after which the response to K became more and
more pronounced as the exchangeable K content of the soil gradually decreased, even
with the intermediate annual application of 80 kg K2O/ha, in combination with 80 kg N
and 40 P2O5/ha. While there was no significant response to P during the first ten years,
the response became more pronounced during the subsequent three years; the available
P level in the soils decreased only slowly but remained far above the critical level of 5 ppm
in the soil. Similarly, in Tamanbogo the response was mainly to the application of K,
followed by N and P, with the P response becoming more pronounced over time. At
CATAS the responses were initially mainly to N and K (about equal), but have become
increasingly more pronounced to the application of N, followed by K, followed by P.
In Thailand, similar results have been obtained in trials conducted in three locations by
the Department of Agriculture for the past 26-27 years. In most cases, continuous
cropping resulted in the depletion of soil K, resulting in marked responses to the
application of that nutrient. Even after 27 years of continuous cropping yields could be
maintained at levels of 25-30 t/ha with application of only chemical fertilizers as long as
plant tops were reincorporated into the soil after root harvest.
Project IP3: improving cassava for the developed world.
Output 9-2
9.1.2 Soil Improvement trial in Hung Loc Center
Materials and Methods
This experiment was initiated at Hung Loc Agric. Research Center in South Vietnam in
1992. The trial is now in its 12th year. During the first 7 years fertilizers were applied at
a rate of 80 kg N, 40 P2O5 and 80 K2O/ha. During the 8th, 9th and 10th year no fertilizers
were applied, and as of the 11th year plots were split with subplot treatments of with and
without fertilizers. Peanut and cowpea were intercropped, the grain harvested and crop
residues mulched on the soil surface; the three green manures were intercropped with
cassava, pulled out at 3 MAP and mulched. The two alley crop species are permanent,
but they are pruned before cassava replanting and 1-2 times during the crop cycle; the
prunings are incorporated or mulched.
Results
Table 1 shows the results in the 11th year of continuous cropping. There was a marked
response to application of chemical fertilizers in all treatments and a good response to the
two alley cropping treatments, which increased yields on average 56% in the absence of
fertilizers and 30% in the presence of fertilizers. Mucuna sp. competed strongly with
cassava; pigeon pea had little beneficial effect, while Canavalia, peanut and cowpea had a
small positive effect. The effect was similar on starch content, which generally increased
with the application of fertilizers and also increased slightly with most of the green
manure/alley cropping treatments. It must be noted that the root yield response to these
treatments (in the presence of fertilizers) was either non-significant or negative during the
first seven years, but became significant and positive in case of the two alley cropping
treatments in the 8th and following years when no more chemical fertilizers were applied.
In the 11th year (Table 1) application of chemical fertilizers increased yields 82%, while
the best alley crop treatment increased yields 58%, and the best combination of fertilizers
and alley cropping increased yields 154% as compared to the unfertilized check. Thus, it
is clear that to obtain high and sustainable yields, it is best to combine chemical
fertilizers with green manures/alley cropping. This produced also the highest net
incomes. Without fertilizers net incomes were markedly depressed.
A similar trial was conducted for the first year in Khaw Hin Sorn Experiment Station in
Chachoengsao province of Thailand. Cassava variety KU 50 was planted at a planting
distance of 1.0x1.0 m. Six green manure (GM) species were planted between rows at one
MAP cassava. All plots received 156 kg/ha of 15-7-18 except treatment 8 which received
469 kg/ha without GM.
Table 2 shows that highest yields were obtained with
application of 156 kg/ha of 15-7-18 fertilizers without GM, but that starch content was
highest with the higher rate of fertilizer. Although both cassava and green manures grew
well and in reasonable balance, the GMs all had a negative effect on cassava yield,
Output 9-3
2003 Annual Report
especially Mucuna and Crotalaria juncea due to severe competition for light, water and
nutrients. The highest net income was obtained with 156 kg/ha of fertilizers without
GM. Thus, as was the case in the trial conducted at Hung Loc Center described above,
the short-term effect of intercropped GM was negative. Preferably, GM should be planted
and incorporated or mulched before planting cassava to prevent competition, but this
would delay cassava planting to the late rainy season, reducing the number of months
with adequate rainfall and thus resulting in low cassava yields (CIAT Cassava Program
Annual Reports for 1992-1994). Most cassava farmers in Asia can not afford leaving their
land unproductive for six months or one year to grow a green manure crop, so
intercropping is the preferred option, although with its own drawbacks. The experiment
will need to be continued for many years to see the long-term benefit of the green
manures on soil fertility and yields.
Another long-term fertility maintenance trial conducted by the Thai Dept. of Agric. in
Khon Kaen, now in its 23rd year, indicates a marked response to chemical fertilizers; crop
rotation (cassava alternated yearly with peanut followed by pigeon pea) had a beneficial
effect on cassava yields but only after the first ten years, while intercropping with peanut
(with fertilizers) always resulted in a reduction of cassava yields.
Project IP3: improving cassava for the developed world.
Output 9-4
Cassava root yield (t/ha)
= KM 60
= Vinh Phu
25
25
20
20
15
15
10
10
80 N
80 K2O
40 P2O 5
80 K2O 5
5
80 N
40 P 2O5
5
0
0
40
80
160
0
0
20
40
80
0
40
80
160
0-0-0
80-40-80
160-80-160
100
80
80
60
60
40
40
Leaf life (days)
100
0
80 N
80 K2O
40 P 2O5
80 K 2O5
20
0
40
80
kg N/ha
160
0
20
40
kg P 2O 5/ha
80 N
40 P2O 5
80
0
40
80
20
160
0
0-0-0
kg K2O/ha
80-40-80
160-80-160
kg N-P 2O5-K 2O/ha
Figure 1. Effect of annual applications of N, P and K on the final root yield and on the leaf life at 3 MAP of two cassava
varieties grown at Thai Nguyen University, Thai Nguyen, Vietnam in 2002 (13th year).
Output 9-5
2003 Annual Report
= KM 60
Cassava root yield (t/ha)
40
40
30
30
20
20
10
0
Starch content (%)
= SM 937-26
40 P2O 5
80 K2O
0
40
80
80 N
80 K2O
10
80 N
40 P2O
0
160
0
20
40
80
0
40
80
160
0-0-0
80-40-80
160-80-160
35
35
25
25
15
15
80 N
80 K 2O
40 P 2O5
80 K 2O
0
0
40
80
160
80 N
40 P2O
0
0
20
kg N/ha
40
kg P2O 5/ha
80
0
40
80
160
kg K 2O/ha
0-0-0
80-40-80
160-80-160
kg N-P2O 5-K2O/ha
Figure 2. Effect of annual applications of various levels of N, P and K on the root yield and root starch content of two
cassava varieties grown at Hung Loc Agric. Research Center, Thong Nhat, Dongnai, Vietnam in
2002/03 (13th year).
Project IP3: improving cassava for the developed world.
Output 9-6
Cassava root yield (t /ha)
Relative yield (%)
= N0P0 K0
= N0P2 K2
= N2P0 K2
= N2P2 K0
= N2P2 K2
Hung Loc
30
40
30
20
20
10
10
0
0
120
120
100
100
80
80
60
60
40
40
20
20
0
0
0.25
0.25
0.20
0.20
0.15
0.15
Critical K-level
0.10
0.10
0.05
0.05
0
0
Soil P (ppm)
Soil K (me/100g)
40
40
40
30
30
20
20
10
10
Critical P-level
0
0
1
Output 9-7
2
3
4
5
6
7
Crop cycle
8
9
10
11
12
13
Figure 3. Effect o f annual applications of N, P and K on cassava root yield, relative
yield (yield without the nutrient over the highest yield with the nutrient) and
the exchangeable K and available P (Bray 2) content of the soil during
thirteen years of continuous cropping in Hung Loc Agric. Research Center,
Dong Nai, Vietnam.
2003 Annual Report
Table 1. Effect of green manures, intercropping and alley cropping systems, with and without chemical fertilizers on
cassava and intercrop yields as well as on gross and net income when cassava, KM60, was grown for the 11th
consecutive year at Hung Loc Agric. Research Center in Thong Nhat district, Dongnai, Vietnam in 2002/03.
Treatments1)
1.
2.
3.
4.
5.
6.
7.
8.
1)
2)
3)
4)
C monoculture
C+pigeon pea GM
C+Mucuna GM
C+Canavalia GM
C+peanut IC
C+cowpea IC
C+Leucaena AC
C+Gliricidia AC
Root yield
(t/ha)
Biomass
intercrops(t/
ha)
+F2)
+F
20.12
20.06
18.56
21.48
22.44
23.91
24.31
28.04
-F2)
11.04 14.06 11.34 2.46
14.75 15.50 5.33
14.63 3.30
17.40 15.79
17.09 9.39
Intercrop
yield
(t/ha)
Starch
content
(%)
-F
+F
-F
+F
2.19
2.61
2.04
14.80
8.78
0.108
0
-
0
0
-
24.8
25.4
25.4
26.1
26.0
25.5
26.2
25.9
-F
23.9
24.6
23.5
23.7
24.8
24.3
24.7
25.0
Product.
costs4)
(‘000d/ha)
Gross
income3)
(‘000d/ha)
+F1)
-F
+F
6,640
6,620
6,125
7,088
7,945
7,890
8,022
9,253
3,643
4,640
3,742
4,867
5,115
4,828
5,742
5,640
4,601
5,201
5,201
5,201
5,201
5,201
4,901
4,901
-F
3,525
4,125
4,125
4,125
4,125
4,125
3,825
3,825
C = cassava; GM = green manure; IC = intercrop; AC = alley crop
+F = 80 kg N+ 40 P2O5 + 80 K2O/ha; -F = no fertilizers
Prices:
cassava
dong
330/kg fresh roots
peanut
5,000/kg dry pods
Costs:
labor
25,000/mday
fertilizers
0.951 mil. dong/ha
fertilizer application
0.125 mil. dong/ha
cassava cultivation without intercrops or fert.
3.525 mil. dong/ha
intercrop/GM seed
0.200 mil. dong/ha
intercrop/GM/AC maintenance/harvesting
0.300 mil. dong/ha
intercrop planting
0.200 mil. dong/ha
weeding with intercrop or GM
0.100 mil. dong/ha less than without
Project IP3: improving cassava for the developed world.
Output 9-8
Net
income
(‘000d/ha)
+F
-F
2,039
118
1,419
515
924 -383
1,887
742
2,744
990
2,689
703
3,121 1,917
4,352 1,815
Table 2. Effect of greenmanures and/or chemical fertilizers on the root yield and starch
content of cassava, KU 50, as well as the gross and net income when grown at
Khaw Hin Sorn research station in Khaw Hin Sorn, Chachoengsao, Thailand in
2002/03.
Treatments
1. Check without GM; 156
kg/ha 15-7-18
2. Crotalaria juncea; 156 kg/ha
15-7-18
3. Canavalia ensiformis; 156
kg/ha 15-7-18
4. Pigeon pea ICPL 304; 156
kg/ha 15-7-18
5. Cowpea CP 4-2-3-1; 156
kg/ha 15-7-18
6. Mucuna; 156 kg/ha 15-7-18
7. Mungbean; 156 kg/ha 15-718
8. Check without GM; 469
kg/ha 15-7-18
Average
GM
Root
fresh
yield
(t/ha)
yield
(t/ha)
-
46.45
24.6
41.43
19.07
22.36
17.79
36.58
24.3
32.41
16.16
16.25
6.92
40.35
24.9
36.23
17.17
19.06
4.45
38.23
23.9
33.57
16.61
16.96
18.63
38.54
23.7
33.68
16.81
16.87
11.78
6.54
36.73
40.07
24.9
24.2
32.98
35.42
16.20
16.88
16.78
18.54
43.44
25.5
39.53
20.76
18.77
40.05
24.5
35.66
17.46
18.20
11.03
Starch
Gross Product
Net
.
conten income1) costs income
t
 (baht/ha) 
(%)
1)Prices:
cassava baht 1.0 /kg fresh roots at 30% starch; 0.02 baht reduction per 1%
starch reduction.
Activity 9.2 Development of efficient and economical soil preparation practices.
Rationale
When land is prepared by tractor for cassava production, the most common practice is to
plow first with a 3-disk plow, followed by a 7-disk harrow, which in some cases is
followed by a ridger. With this practice all crop residues and weeds are incorporated, the
land is free of weeds and the soil is loose, which makes it easy to push in cassava stakes
by hand. However, after many years of using this practice, soil structure has degraded,
soil organic matter has decreased and a very compact plow sole has formed at about 20
cm depth which impedes drainage. The poor drainage causes not only water logging and
root rots, but it may also cause severe erosion as the top soil becomes quickly saturated
Output 9-9
2003 Annual Report
with water after a heavy rainstorm, resulting in excessive runoff and gulley erosion along
natural drainage ways. Thus, alternative methods of land preparation, including zero- or
conservation tillage, were investigated.
Specific Objectives
a) To determine the immediate and long-term effects of various methods of mechanical land
preparation (sometimes in combination with herbicides) on cassava yield and starch
content, as well as their economic feasibility.
Materials and Methods
Three identical experiments were conducted in Thailand at TTDI’s Research Center in
Huay Bong, in Kasetsart University’s Experiment Station in Khaw Hin Sorn, and in a
farmer’s field near Rayong Field Crops Research Center (FCRC). Ten soil preparation
treatments were established in main plots using various combinations of a chisel plow,
subsoiler, cassava harvester, 3-disk plow, 7-disk plow and ridger. Four varieties were
planted in subplots in each main plot. Each trial had either three or four replications.
Results
Table 3 shows the results at TTDI and in Khaw Hin Sorn. In TTDI the highest root yield
and starch content were obtained when weeds were killed with Glyphosate and the land
prepared with a subsoiler followed by chisel plow. This treatment produced the highest
gross and net income but also had the highest production costs, partially because the
cost of harvest and transport is directly related to root yield. Zero tillage produced similar
yields and net income as complete tillage (3-disk and 7-disk plow followed by ridger).
Project IP3: improving cassava for the developed world.
Output 9-10
Table 3. Effect of various methods of land preparation on the average root yield and
starch content as well as the production costs, gross and net income obtained with
four cassava varieties planted at TTDI, Huay Bong, Nakhon Ratchasima, and in
Khaw Hin Sorn, Chachoengsao, Thailand in 2002/03.
TTDI
Treatment
Khaw Hin Sorn
Root Starch Gross Product Net
Root Starch Gross Product Net
yield content income1 costs income yield content income2 costs income
(t/ha) (%) (‘000 B/ha) (t/ha) (%) (‘000 B/ha)
1. No tillage; Glyphosate
2. Chisel plow; Glyphosate
3. Subsoiler; Glyphosate
4. Subsoiler+chisel;
Glyphosate
5. Subsoiler+7 disk;
Glyphosate
6. Cassava harvester;
Glyphosate
7. Subsoiler+7 disk plow
8. Subsoiler+3 disk+7 disk
9. 3 disk+7disk plow
10. 3 disk+7disk+ridging
26.07
25.10
24.32
28.71
21.9
25.2
25.1
25.5
27.06
27.71
26.80
31.87
13.10
13.77
13.76
15.87
13.96
13.94
13.04
16.00
32.71
34.18
33.01
37.65
23.5
23.6
25.7
22.9
28.46
29.80
30.17
32.30
15.58
16.73
16.79
18.79
12.88
13.07
13.38
13.51
25.35
23.8
27.28
14.97
12.31
36.24
22.4
30.73
18.60
12.13
25.52
23.6
27.36
14.20
13.16
39.50
26.3
36.58
18.66
17.92
24.90
26.40
23.31
26.57
24.3
24.7
23.9
24.7
27.04
28.88
25.13
29.07
14.01
15.53
13.57
15.08
13.03
13.35
11.56
13.99
28.65
38.95
41.99
46.35
22.9
24.4
22.3
22.1
24.58
34.59
35.52
39.03
15.52
19.43
19.12
20.92
9.06
15.16
16.40
18.11
Average
25.63
24.3
27.82
14.39
13.43
36.92
23.6
32.18
18.01
14.16
Price: cassava bath 1.2/kg fresh roots at 30% starch;
0.02 baht reduction per 1% starch reduction.
2) Price: cassava bath 1.0/kg fresh roots at 30% starch;
0.02 baht reduction per 1% starch reduction.
1)
In Khaw Hin Sorn cassava yields were considerably higher than at TTDI and highest
yields were obtained with complete land preparation. A rather high yield and highest
starch content were also obtained using the cassava harvester and Glyphosate; highest
net income was obtained with complete land preparation of 3-disk and 7-disk plow
followed by ridging. This treatment also had the highest production costs.
It appears that in areas with a compacted subsoil or hard pan the use of a subsoiler and
chisel plow may have a distinct advantage of improving internal drainage; this will also
reduce runoff and erosion.
Activity 9.3 Determination of the response to micronutrients and Mg, and the most
effective way of applying Zn in calcareous soils.
Rationale
Cassava may suffer from Mg deficiency in acid soils, while micronutrient deficiencies are
often observed in calcareous or high pH soils. In the calcareous soils at TTDI in Huay
Output 9-11
2003 Annual Report
Bong, Nakhon Ratchasima, Thailand, many cassava plants show severe symptoms of Zn
deficiency.
Specific Objectives
a) To determine the most limiting micronutrients and Mg in several soils of Thailand.
b) To determine the most efficient way of applying Zn to cassava grown in calcareous soils.
Results
A preliminary trial conducted last year on the application of Fe, Zn, Cu and B in
calcareous soils of TTDI indicated a slight positive effect of soil application of Zn as well
as to a stake treatment for 15-minutes in 2% ZnSO4.7H2O. Similar treatments with Fe
were not effective, while application of B and Cu actually decreased yields.
A similar trial with Fe, Zn, B and Mg conducted this year in acid soils of Kalasin and
Ubon Ratchathani Field Crop Research Stations indicated no response to any of these
nutrients in Kalasin, but a good response to application of 5-10 kg Zn and 5 kg Fe/ha to
the soil in Ubon Ratchathani. Stake treatments were less effective. A detailed trial on
several methods and rates of application of Zn in TTDI (Table 4) indicates that best
results were obtained with application of three foliar sprays with 4% ZnSO4 at 1, 2 and 3
MAP; with a soil application of 30 kg Zn/ha; or a combination of stake treatment, 5 kg
Zn/ha to the soil and three foliar sprays with 1% ZnSO4. The foliar application is the
cheapest of these three treatments, but will leave no residual effect in the soil. A stake
treatment in 2% ZnSO4 was only intermediately effective, but this is by far the cheapest
method of application.
Activity 9.4 Evaluation of the use of plastic mulch for weed and erosion control.
Rationale
Planting high value crops such as watermelon, strawberries or pineapple on plastic
mulch is a common practice in both developed and developing countries. A few years ago
Chinese farmers in Guangxi province started planting watermelon on plastic mulch as an
intercrop in cassava; they are now planting even monocropped cassava on plastic,
usually alternating 1 m wide strips covered with plastic with 1 m wide uncovered strips.
Cassava stakes are planted slanted along the edge of the plastic. Good quality plastic
may be used 2-3 years, while the thinner plastic (not biodegradable) degrades in about
one year. In Taiping town of Wuming county about 25% of the cassava area was planted
on plastic in 2002.
Project IP3: improving cassava for the developed world.
Output 9-12
Table 4. Effect of methods and levels of application of Zn on the root yield and starch
content of two cassava varieties, as well as the cost of application when grown at
TTDI Research and Development Center at Huay Bong, Daan Khun Thot, Nakhon
Ratchasima, Thailand in 2002/03.
Root yield (t/ha)
Treatment
1.
2.
3.
4.
5.
6.
7.
8.
9.
10
11
12
13
14
15
16
1)
2)
3)
Check, no Zn
Stake dip, 2% ZnSO4
Stake dip, 4% ZnSO4
Stake dip, 6% ZnSO4
Stake dip, 8% ZnSO4
Soil application, 5kg Zn/ha
Soil application, 10kg Zn/ha
Soil application, 20kg Zn/ha
Soil application, 30kg Zn/ha
Foliar application 1% ZnSO4
Foliar application 2% ZnSO4
Foliar application 3% ZnSO4
Foliar application 4% ZnSO4
Stake 2%+5kg Zn+1% foliar
Stake 2%+5kg Zn+2% foliar
Stake 2%+5kg Zn+4% foliar
Rayong
72
KU-50 2 var.Av.
Starch content (%)
Rayong
72
KU-50
Cost (baht/ha)
ZnSO41) labor Total
13.6
16.9
15.3
22.0
23.2
0
18.6
17.1
17.9
21.6
22.6
128
250
378
13.3
17.2
15.3
21.1
21.4
256
250
506
15.9
14.7
15.3
22.6
22.1
384
250
634
15.8
14.0
14.9
21.5
21.4
512
250
762
15.4
16.6
16.0
20.3
22.8
694
250
944
16.4
19.6
18.0
22.0
22.4
1,389
250
1,639
16.6
18.9
17.8
21.6
23.0
2,778
250
3,028
19.3
20.2
19.8
22.0
21.9
4,166
250
4,416
13.5
14.6
14.1
19.6
21.1
1923)
750
942
17.0
14.6
15.8
21.9
22.5
384
750
1,134
18.3
17.1
17.7
21.4
21.9
576
750
1,326
22.8
19.4
21.1
21.0
21.7
768
750
1,518
21.9
17.6
19.8
22.8
23.1
1,014 1,250 2,264
19.0
16.3
17.6
22.3
21.5
1,206 1,250 2,456
18.8
19.6
19.2
23.9
24.7
1,590 1,250 2,840
17.3
17.2
17.2
21.7
Average
1 kg ZnSO4.7H2O= 32 baht = US$ 0.75
assuming usage of 200 l/ha for stake treatment
assuming application of 200 l/ha, 3 times at 1, 2 and 3 MAP
2)
0
0
22.3
Specific Objectives
a) To determine the effect of plastic mulch on cassava yield, and on gross and net income
as an alternative to hand weeding or chemical weed control.
b) To determine the effect of plastic mulch on cassava yield and soil erosion.
Materials and Methods
An experiment was conducted in Hung Loc Center on the use of plastic mulch, applied at
either 20 or 30 days after planting (DAP), as an alternative to hand weeding, while three
FPR on-farm trials were conducted in An Vien commune of Dongnai province to compare
the use of 100% plastic cover with either hand weeding or application of pre-emergence
herbicides.
In CATAS, Danzhou, Hainan, China, cassava was planted with 100% plastic mulch on
about 5% slope as one of seven alternative practices to control erosion.
Output 9-13
2003 Annual Report
Results
Table 5 shows the results of the weed control trial at Hung Loc Center in south Vietnam.
Planting cassava on 100% plastic mulch significantly increased cassava yields, from 23.8
to 34.6 t/ha. However, due to the relatively high cost of plastic in Vietnam (about
$300/ha) the net income for hand weeding was slightly higher than for the use of plastic
mulch. In the three FPR trials, average cassava yields on plastic were 40.6 t/ha versus
31.7 t/ha for hand weeding and 24.5 t/ha for the use of Dual as a pre-emergence
herbicide (Table 6). However, the use of plastic mulch produced the lowest net income
and 60% of farmers still preferred hand weeding, while only 20% preferred the use of
either plastic or chemical weed control.
Table 5. Effect of the use of plastic mulch in comparison with hand weeding on cassava
yield and gross andnet income at Hung Loc Agric. Research Center in Thong Nhat
district, Dongnai, Vietnam in 2002/03.
Treatments1)
No weeding
Weeding by hand
Cover with plastic at 20 DAP
Cover with plastic at 30 DAP
Weeding by hand + plastic at 30 DAP
1)
2)
Root
yield
(t/ha)
0.20
23.80
34.58
31.25
34.60
c
b
a
a
a
Gross
Product.
Net
2)
2)
income
costs
income
 (‘000 dong/ha) 
66
7,854
11,411
10,312
11,418
2,386
5,366
9,205
9,038
9,506
-2,320
2,488
2,206
1,274
1,912
B/C
0.03
1.46
1.24
1.14
1.20
Prices:cassava
dong 330/kg fresh roots
Costs:
labor for harvest
50,000/tonne
labor for weeding by hand (3x)
1.80 mil. dong/ha
cassava cultivation without weeding, fertilization or harvest: 1.30 mil. dong/ha
cost of plastic
4.80 mil. dong/ha
labor for weeding by hand (1x)
0.30 mil. dong/ha
fertilizer cost
0.951 mil. dong/ha
fertilizer application costs
0.125 mil. dong/ha
cost of covering soil with plastic
0.300 mil. dong/ha
Project IP3: improving cassava for the developed world.
Output 9-14
Table 6. Average results of three on-farms weed control trials conducted in An Vien
commune, Thong Nhat, Dongnai, Vietnam in 2002/03.
Treatments1)
Root
yield
(t/ha)
1 Weeding by hand
2. Dual applic. 2 l/ha
3. Cover soil with plastic
31.75
24.55
40.60
1)
2)
3)
Top
yield
(t/ha)
22.78
19.55
25.75
Gross
Weeding Product.
Net
Farmers’
Harvest income2) costs3)
costs3) income preference
index
(%)
 (‘000 dong/ha) 
0.58
0.56
0.61
8,573
6,629
10,962
Fertilized with 80 kg N+ 40 P2O5 + 80 K2O/ha
Prices: cassava
Costs: harvest cassava
plastic
Dual
hand weeding
herbicide application
fertilizers + application
cassava cultivation without weeding
covering soil with plastic
1,600
520
5,000
5,163
3,723
9,006
dong
3,410
2,906
1,956
60
20
20
270/kg fresh roots
50,000d/tonne
470/m2
160,000/l
1.600 mil. dong/ha
0.200 mil. dong/ha
1.076 mil. dong/ha
1.976 mil. dong/ha
0.300 mil. dong/ha
Table 7 shows the results of the erosion control trial at CATAS. Cassava planted on
plastic produced the highest yield of 35.7 t/ha compared with 31.4 t/ha for the control
treatment. However, dry soil losses by erosion in the plastic treatment (average of 2 Reps)
was only 8.5 t/ha as compared to 80.0 t/ha in the control treatment. The use of plastic
mulch was the most effective way to reduce erosion, followed by the planting of contour
hedgerows of closely spaced cassava stems, used for cassava leaf production for animal
feeding.
Table 7. Effect of several crop management practices on dry soil loss by erosion as well as
on the root yield and root DM content of cassava, cv. ZM 8803 planted in FPR
demonstration plots at CATAS, Danzhou, Hainan, China in 2002/03.
Treatments
1. Cassava monoculture
2. C+peanut
3. C+Panicum maximum hedgerows
4. C+Brachiaria decumbens
hedgerows
5. C+vetiver grass hedgerows
6. C+closely spaced cassava
hedgerows1)
7. C with plastic mulch
1)
Dry soil loss (t/ha)
Root yield (t/ha)
80.01
50.55
31.58
39.44
31.37
28.17
21.29
30.54
31.43
9.39
30.12
29.34
8.52
35.67
2 rows
Output 9-15
2003 Annual Report
Activity 9.5 Evaluation of cassava varieties and determination of optimum NPK
fertilization, plant spacing, cutting height and frequency for cassava
leaf production.
Rationale
Cassava root pellets are widely used in Europe for animal feeding. In Thailand this is not
yet widely practiced due to the availability of other cheap raw materials for the production
of animal feed, such as broken rice and maize. The low protein content of cassava roots
and the inadequate local supply of soybean limits the local use of cassava in animal feed
rations. However, cassava leaves are known to contain high levels of crude protein with a
good amino acid spectrum. Recent research indicate that the low-medium tannin content
of cassava leaves actually improves protein digestibility. Thus, intensive research was
initiated to identify the best varieties for leaf production and to determine the most
economic way of producing high yields of leaves as well as roots.
Specific Objective
a) To determine the best varieties and cultural practices for obtaining high leaf and root
yields and maximize net farm income.
Materials and Methods
a) Varietal evaluation: 25 cassava varieties, selected for abundant top growth, were
evaluated in Rayong and Khon Kaen FCR Centers in east and northeast Thailand
as well as at TTDI in Nakhon Ratchasima, while 14 varieties were evaluated in
Songklaa FCR Center in south Thailand, where rainfall is higher and better
distributed. Cassava stakes were planted at 30x30 cm and plant tops were cut off
at 15-20 cm above the soil at 2½ -3 MAP, and thereafter at 2-3 months intervals
depending on rainfall and rate of regrowth. Plants were fertilized with a total of
1500 kg/ha of 15-15-15, applied at planting and after the 2nd and 4th cut, and with
200 kg N/ha as urea applied after the 1st and 3rd cut. In Rayong and Khon Kaen
plant tops were cut five times during a 1-year growth cycle, with the 5th cut
coinciding with the final root harvest. In Songklaa plants were cut three times in
the first year and allowed to regrow again without root harvest for a second year.
b) NPK application: In both Rayong and Khon Kaen two varieties were planted in 12
plots receiving various combinations of N, P and K. They were planted at 30x30
cm and plant tops were cut off five times during a 1-year growth cycle before
harvesting the roots. Half of P and K were applied at planting and after the 2nd
cut, while one quarter of N was applied at planting and after the 1st, 2nd and 3rd
cuts.
c) Variety x plant spacing: Three varieties were planted at 30x30, 40x40, 50x50 and
60x60 cm. They were fertilized with 15-15-15 and urea as described above under
Project IP3: improving cassava for the developed world.
Output 9-16
a) and plant tops were cut five times during a 1-year cycle in both Rayong and
Khon Kaen.
d) Cutting height x frequency: Rayong 72 was planted at 30x30 cm at Rayong and
Khon Kaen FCR Centers. Plants were fertilized alternately with 15-15-15 and urea
as described above. In main plots plant tops were cut at either 15, 20 or 25 cm
above the ground; in subplots the frequency of cutting varied, either at 1½, 2, 2½
or 3 month intervals after the first cut at 2½ MAP. In this case, plants were cut
either 4, 5 or 6 times during a 1-year cycle, before roots were harvested at 12 MAP.
Results
a) Varietal evaluation: Table 8 shows the results of the evaluation of 25 varieties for
leaf production at Rayong FCRC. The total production of dry forage (including
leaves, petioles and green stems), the sum of five cuts during a 12-month growth
cycle, varied from 9.9 to 19.6 t/ha. This is considerably higher than the yields
reported in Santander de Quilichao in 1986 during a 14-month growth cycle
(Cassava Annual Report for 1986), where the highest dry forage yield was 13.9
t/ha. It is lower, however, than the highest yield of 26.0 t/ha obtained during a
13-month period in Ayapel, Cordoba, Colombia (Rosero Valencia, 2002). In spite of
frequent cutting of tops at 2-3 month intervals, many varieties still produced high
root yields; these varied from 10.1 to 35.3 t/ha. Starch contents of roots were
rather low probably due to frequent cutting and very close plant spacing, but they
still averaged at about 21%.
The highest yield of dry leaves was obtained with the breeding line CMR 42-07-9,
which also produced the highest net income; the highest root yield was obtained
with the high-yield and high-starch variety Rayong 90, which produced the second
highest net income. On average, the value of harvested roots accounted for 22% of
total gross income, while that of leaves accounted for the remaining 78%. Thus,
root yields are still an important source of income, even when the main objective is
to produce leaves.
In Khon Kaen, the total dry leaf yield (sum of five cuts) of the same 25 varieties
varied from 9.1 to 16.6 t/ha, while root yields varied from 1.4 to 26.1 t/ha. Thus,
in Khon Kaen the yields of both roots and leaves were considerably lower than in
Rayong, probably due to a much poorer soil and longer dry season. The highest
leaf yield, root yield and net income were obtained with the breeding line CMR 41111-129. No data are yet available for total protein yields, but crude protein
contents for the first cut in Rayong varied from 15.4 to 21.7% (average 20.0%)
while in Khon Kaen this varied from 13.7 to 22.6% (average 18.8%).
Output 9-17
2003 Annual Report
Table 8. Dry leaf yield from five cuts of tops, final root yield and root starch content as
well as gross and net income obtained in the the cassava variety trial for leaf
production conducted in Rayong FCRC in 2002/03.
Fresh
Dry leaf yield (t/ha)

Variety
Starch
Gross
root yield content income
1)
Product.
Net
costs
income
1st
2nd
3rd
4th
5th
(%)
cut
cut
cut
cut
cut
Tota (t/ha)
l
 (‘000 baht/ha
1. Rayong 1
2. Rayong 5
3. Rayong 60
4. Rayong 90
5. Rayong 72
6. KU 50
7. OMR 41-23-41
8. CMR 41-42-3
9. CMR 41-60-24
10. CMR 41-61-59
11. CMR 41-111-129
12. CMR 41-114-125
13. CMR 35-22-196
14. CMR 41-20-58
15. CMR 41-96-2
16. OMR 41-33-34
17. CMR 42-01-2
18. CMR 42-07-9
19. CMR 42-21-59
20. CMR 42-54-53
21. CMR 42-59-173
22. CMR 42-61-108
23. CMR 42-87-318
24. CMR 42-90-338
25. HB-60
2.401
2.433
1.814
2.393
2.313
2.817
1.645
2.621
2.596
2.280
2.154
1.994
1.476
3.169
1.775
2.217
3.114
1.948
2.125
0.951
3.002
1.502
2.911
3.073
2.223
4.489
4.025
3.523
3.958
3.172
4.852
3.365
4.661
4.433
4.325
4.206
4.324
2.109
4.409
4.314
5.498
4.888
5.753
4.619
3.369
5.549
3.870
4.720
4.861
3.423
3.057
3.086
2.135
2.927
2.130
3.285
2.571
2.830
2.868
3.050
2.694
2.921
2.281
3.485
2.515
3.472
2.537
3.530
2.601
2.535
3.012
2.635
2.356
2.883
3.417
0.414
0.767
0.403
0.404
0.356
0.392
0.479
0.578
0.504
0.471
0.509
0.554
0.383
0.552
0.483
0.512
0.444
0.866
0.543
0.546
0.586
0.569
0.518
0.387
0.498
6.322
4.973
4.286
4.961
4.140
5.420
6.538
5.138
5.554
5.772
5.446
5.809
3.692
4.985
5.270
6.881
5.522
7.521
5.410
5.654
5.198
5.405
5.629
5.579
3.972
16.68
15.28
12.16
14.64
12.11
16.77
14.60
15.83
15.95
15.90
15.01
15.60
9.94
16.60
14.36
18.58
16.51
19.62
15.30
13.05
17.35
13.98
16.13
16.78
13.53
16.71
32.59
27.61
35.35
30.20
23.24
16.30
21.16
22.69
20.79
21.54
25.39
16.96
15.02
17.07
12.45
16.13
21.40
20.85
22.30
10.14
13.98
22.24
15.80
23.97
14.5
22.9
17.0
23.7
21.0
24.2
20.1
24.0
22.0
16.2
22.2
22.3
23.6
21.3
20.5
20.4
24.3
17.5
19.0
20.9
19.7
20.0
18.5
25.2
21.6
78.25
89.08
69.07
89.46
73.20
87.62
71.47
81.94
82.86
78.65
78.22
83.88
54.55
78.81
71.27
84.38
80.33
94.53
77.46
70.44
77.45
67.10
81.65
81.40
74.06
66.58
69.33
64.55
69.36
65.19
68.44
64.18
66.84
67.39
66.82
66.04
67.74
59.23
66.04
64.12
67.52
66.24
71.07
66.18
64.09
65.54
62.87
67.47
66.44
65.07
11.67
19.75
4.52
20.10
8.01
19.18
7.29
15.10
15.47
11.83
12.18
16.14
-4.68
12.77
7.15
16.86
14.09
23.46
11.28
6.35
11.91
4.23
14.18
14.96
8.99
Average
%DM
2.278 4.269 2.833 0.509 5.403 15.29
20.03 22.7
21.4 28.76 24.51
20.87
20.9
78.29
66.17
12.11
1)
Prices: cassava roots: baht 1.00/kg fresh roots at 30% starch; baht 0.02 reduction for
every 1% starch reduction
cassava leaves: 4.00/kg dry leaves 1 US$ = 42 baht
b) NPK fertilization: Table 9 shows the results of the NPK fertilization trial for leaf
production conducted with two varieties in Rayong FCRC. Highest yields of dry leaves
were 14.7 t/ha for Rayong 72 and 15.3 t/ha for Rayong 5, both obtained at the
highest rate of fertilizer application, i.e. 600 kg N, 300 P2O5 and 300 K2O/ha. In case
of Rayong 72 the highest root yield of 36.6 t/ha was obtained with 600 kg N combined
with 150 kg P2O5 and 150 K2O/ha, but with Rayong 5 the highest root yield of 35.6
t/ha was obtained with a lower rate of N, i.e. 300 kg N, 150 P2O5 and 150 K2O/ha.
Without fertilizer or without N application, leaf yields were only 4-6 t/ha while root
yields were also markedly reduced; but this treatment produced the highest starch
content in both varieties. Lowest root starch contents were obtained at the highest
rates of fertilization; this could be due to the very high rates of N application, the very
Project IP3: improving cassava for the developed world.
Output 9-18
close spacing or the frequent cutting of tops. Highest net incomes were obtained with
application of 150 kg P2O5 and 150 K2O/ha combined with either 300 or 600 kg N/ha.
At these intermediate levels of application, the value of roots accounted for about 40%
of the total gross income. Thus, to optimize income, farmers should not concentrate
only on leaf production while neglecting root production.
In Khon Kaen the leaf and root yields as well as the root starch contents were
considerably lower than in Rayong resulting in mostly negative net incomes. Highest
leaf yields of both varieties were obtained with 600 kg N, 150 kg P2O5 and 150 K2O/ha.
Without fertilizer or N application, both dry leaf and fresh root yields were again very
low (about 4 and 8 t/ha, respectively) but with the highest root starch content. The
highest net income was achieved in both varieties when no P was applied, i.e.
application of only 300 kg N and 150 K2O/ha. Thus, it is clear that for high leaf yields
and high net income, relatively high rates of fertilizers should be applied and in a ratio
of N, P2O5 and K2O of about 4:1:2.
c) Variety x plant spacing: Table 10 shows the results of the variety x plant spacing trial
conducted in Khon Kaen FCRC. Increasing plant density (i.e. decreasing plant spacing
from 60x60 to 30x30 cm) tended to increase leaf production in Rayong but had little
effect in Khon Kaen. However, root production in both locations decreased with
increasing plant density, while starch contents were little affected. Leaf yields varied
from 8.9 to 15.6 t/ha in Rayong and from 8.9 to 11.8 t/ha in Khon Kaen with the
three varieties producing similar leaf yields. However, root yields of Rayong 72 were
considerably higher than those of Rayong 5, which again were higher than those of
CMR 41-60-24, in both locations. Table 11 shows the interaction between plant
spacing and variety in terms of total dry leaf yield, fresh root yield and net income in
Khon Kaen. It is clear that Rayong 72 was the most productive variety in terms of
both root and leaf production, resulting in the highest net income. While in Rayong
the highest plant density produced the highest leaf yields but the lowest root yields, in
Khon Kaen the lowest plant density (60x60 cm spacing) produced both the highest leaf
and root yields, while in both locations the wider spacing produced the highest net
income. Thus, even for leaf production cassava should not be planted at a very close
spacing (30x30 cm) as this reduces root production. At this spacing, gross income
may be higher (Table 10) but net income tends to be lower due to higher production
costs. At the wider spacing (60x60 cm) the value of roots accounted for about 40% of
total gross income.
A similar trial of four varieties x two plant spacings was conducted in Hong Loc Center
in south Vietnam. Highest leaf yields were obtained at the closer spacing of 35x35 cm,
but higher root yields were obtained at the wider spacing of 70x70 cm; the latter also
produced the highest average net income. The highest leaf yield was obtained with KM
94 (= KU 50), while the highest root yield and average net income were obtained with
KM 140.
Output 9-19
2003 Annual Report
Table 9. Dry leaf yields from five cuts, root yields and starch content, as well as gross
and net income in the NPK fertilizer trial for cassava leaf production in Rayong
FCRC, Rayong, Thailand in 2002/03.
V1 = Rayong 72
Fresh
Treatments1)
Dry leaf yield (t/ha)

1st
2nd
3rd
4th
5th cut Total
cut
cut
cut
cut
1. N0P0K0
2. N0P2K2
3. N1P2K2
4. N2P2K2
5. N3P2K2
6. N2P0K2
7. N2P1K2
8. N2P3K2
9. N2P2K0
10. N2P2K1
11. N2P2K3
12. N3P3K3
0.726
0.922
0.804
1.410
2.676
1.121
1.160
2.030
1.053
1.768
1.594
2.566
1.273
1.622
1.860
2.701
3.945
2.835
2.969
3.236
2.575
3.094
2.771
3.729
0.846
1.113
1.252
1.933
2.783
2.440
2.112
2.232
1.753
2.220
2.036
2.651
Av.
1.486 2.717 1.948
0.339
0.494
0.456
0.711
0.480
0.654
0.532
0.457
0.531
0.355
0.422
0.508
root
yield
Starch Gross Product
Net
content income2) costs
income
(t/ha)
(%)
 (‘000 baht/ha) 
1.278
1.786
2.263
3.465
4.321
3.732
3.965
3.618
3.636
3.226
3.247
5.271
4.46
5.94
6.63
10.22
14.21
10.78
10.74
11.57
9.55
10.66
10.07
14.73
20.21
22.67
20.91
34.92
36.62
30.92
33.63
32.95
29.40
33.54
33.01
35.15
25.6
24.7
22.0
21.3
20.1
21.0
24.0
21.8
21.0
22.8
21.2
16.1
36.27
44.03
44.08
69.72
86.21
68.47
72.55
73.83
62.31
71.35
67.48
84.30
34.30
44.56
47.00
56.88
66.04
51.39
54.59
62.86
52.71
56.02
58.15
73.17
1.97
-0.53
-2.92
12.84
20.17
17.08
17.96
10.97
9.60
15.33
9.33
11.13
0.495 3.317
9.96
30.33
21.8
65.05 54.81
10.24
Fresh
root
Starch
yield
content income2)
V2 = Rayong 5
Dry leaf yield (t/ha)
Treatments1)
1. N0P0K0
2. N0P2K2
3. N1P2K2
4. N2P2K2
5. N3P2K2
6. N2P0K2
7. N2P1K2
8. N2P3K2
9. N2P2K0
10. N2P2K1
11. N2P2K3
12. N3P3K3

1st
2nd
3rd
4th
5th cut Total
cut
cut
cut
cut
0.948
0.910
1.117
1.600
2.391
1.387
1.310
2.127
1.153
1.931
1.451
2.723
1.441
1.514
2.053
3.405
3.551
3.057
2.978
3.218
2.474
3.305
2.707
3.692
0.943
1.119
1.395
2.627
2.362
2.250
2.243
2.427
1.661
2.064
2.089
3.019
0.481
0.667
0.513
1.100
0.621
0.605
0.861
0.710
0.533
0.434
0.549
0.826
1.424
1.781
2.489
4.337
4.617
3.606
3.985
3.852
3.040
3.813
3.385
5.037
5.24
5.99
7.57
13.07
13.54
10.91
11.38
12.33
8.86
11.55
10.18
15.30
(t/ha)
(%)
19.81
21.31
22.10
35.61
28.99
27.81
34.69
31.72
26.41
31.17
23.88
27.29
23.6
21.0
20.1
22.0
18.3
20.7
21.6
22.5
20.0
21.5
19.1
15.3
Gross
Product
Net
.
costs
income
 (‘000 baht/ha) 
38.23
41.43
48.00
82.19
76.37
66.28
74.38
76.28
56.57
72.07
59.39
80.47
35.05
44.24
48.36
60.20
63.24
50.70
55.59
63.37
51.14
56.36
55.81
71.69
3.18
-2.81
-0.36
21.99
13.13
15.58
18.79
12.91
5.43
15.71
3.58
8.78
Av.
1.587 2.783 2.017 0.658 3.447 10.49 27.56 20.5 64.31 54.65
9.66
1) N0 = 0 N
P0 = 0 P
K0 = 0 K
N1 = 150 kg N/ha
P1 = 75 kg P2O5/ha
K1 = 75 kg K2O/ha
N2 = 300 kg N/ha
P2 = 150 kg P2O5/ha
K2 = 150 kg K2O/ha
N3 = 600 kg N/ha
P3 = 300 kg P2O5/ha
P3 = 300 kg K2O/ha
2)Prices: cassava roots: baht 1.00/kg fresh roots at 30% starch; 0.02 baht reduction for every 1% starch
reduction, cassava leaves: 4.00/kg dry leaves
Project IP3: improving cassava for the developed world.
Output 9-20
Table 10. Dry leaf yields from five cuts, fresh root yields and starch content, as well as
gross and net income obtained in the variety x plant spacing trial for leaf
production in Khon Kaen FCRC in 2002/03.
2nd cut
3rd cut
4th cut
5th cut Total
Fresh
root
Starch
Gross
Product.
Net
costs
income
yield content income2)
(t/ha)
(%)
(‘000 baht/ha)
Dry leaf yield (t/ha)
Treatments1)
1st cut
A-1
-2
-3
-4
2.098
2.439
2.425
2.737
4.183
3.923
3.705
3.414
1.544
1.213
1.488
1.339
0.922
0.729
0.695
0.733
2.902
2.549
3.202
3.553
11.65
10.85
11.51
11.78
23.53
22.77
22.66
15.28
13.9
13.0
13.6
13.1
62.55
58.43
61.27
57.24
48.82
49.63
54.06
60.80
13.73
8.80
7.21
-3.56
B-1
-2
-3
-4
2.995
2.373
3.491
3.819
4.511
2.807
2.978
2.479
1.447
1.047
1.058
0.841
0.828
0.640
0.605
0.424
2.523
2.022
2.409
2.354
12.30
8.89
10.54
9.92
8.05
6.31
5.31
3.87
12.3
11.1
10.4
10.3
54.40
39.48
45.39
42.03
45.36
43.03
48.31
55.67
9.04
-3.55
-2.92
-13.64
C-1
-2
-3
-4
1.872
2.359
3.018
2.751
3.184
3.415
3.650
3.194
1.232
1.184
1.374
1.145
0.945
0.720
0.636
0.852
2.325
2.131
2.375
2.167
9.56
9.81
11.05
10.11
15.41
10.23
12.76
4.74
11.7
10.3
12.3
7.6
48.01
45.44
52.44
43.06
44.34
45.10
50.88
56.12
3.67
0.34
1.56
-13.06
Av.
2.698
3.453
1.243
0.727
2.543
10.66
12.58
11.6
50.81
50.18
0.63
1)
A = Rayong 72
B = CMR 41-60-24
C = Rayong 5
1=
2=
3=
4=
baht
60x60 cm
50x50 cm
40x40 cm
30x30 cm
1.00/kg fresh roots at 30% starch; 0.02 baht reduction per 1%
Prices: cassava
starch reduction
cassava leaves 4.0/kg dry leaves + petioles + green stem
2)
Table 11. Effect of plant spacing on the total dry leaf yield, fresh root yield and net income
obtained with three cassava varieties planted in Khon Kaen FCRC in 2002/03.
Plant
spacing
Total dry leaf yield (t/ha)

Rayong
CM
Rayong
72
41-60-24
5
Av.
Fresh root yield (t/ha)

CM Rayong
Rayong
5
Av.
72 41-60-24
60
50
40
30
11.65
10.85
11.51
11.78
12.30
8.89
10.54
9.92
9.56
9.81
11.05
10.11
11.17
9.85
11.03
10.60
23.53
22.77
22.66
15.28
8.05
6.31
5.31
3.87
15.41
10.23
12.76
4.76
15.66
13.10
13.58
7.97
13.73
8.80
7.21
-3.56
9.04
-3.55
-2.92
-13.64
3.67
0.34
1.56
-13.06
8.81
1.86
1.95
-10.09
11.45
10.41
10.13
10.66
21.06
5.89
10.79
12.58
6.55
-2.77
-1.87
0.64
x
x
x
x
Av.
60
50
40
30
Net income (‘000 baht/ha)

Rayong
CM
Rayong
72 41-60-24
5
Av.
d) Cutting height x frequency: Tables 12 and 13 show the results of the cutting height
x frequency trial conducted in Rayong FCRC. An intermediate frequency of cutting
plant tops at 2½ month intervals (five cuts in a 1-year crop cycle) produced the
Output 9-21
2003 Annual Report
highest leaf yields in Rayong, while a slightly less frequent cutting regime (every 3
months) produced highest leaf yields in Khon Kaen. However, in both locations
highest root yields and net income were obtained at the lowest frequency of
cutting, i.e. at 3 month intervals. Highest leaf yields were obtained with a cutting
height of 15 cm in Rayong, while cutting height had little effect on leaf yield in
Khon Kaen; root yields, however, were considerably higher at 25 as compared to
15 cm in Khon Kaen while cutting height had little effect on root yield in Rayong.
It can be concluded that plant tops should not be cut too frequently; an interval of
2½-3 months between cuttings produced higher leaf yields, higher root yields and
starch contents and higher net income. At this lower frequency the value of roots
accounted for about 40% of total gross income.
Table 12. Dry cassava leaf yields from 4-6 cuts, root yields and starch content, as well as
gross and net income in the cutting height x frequency trial conducted in Rayong
FCRC in 2002/03.
Fresh
root Starch
Dry leaf yield (t/ha)
Treatment1
Gross Product
Net
.
yield conten income2) costs income

t
1st
(%)
2nd
3rd
4th
5th
6th cut Total (t/ha)
 (‘000 baht/ha) 
cut
cut
cut
cut
cut
A-1
-2
-3
-4
2.536
2.535
2.620
2.947
3.536
3.826
4.197
5.387
1.920
2.366
1.954
1.097
1.154 1.909 1.794
0.598 4.473 3.460 2.903 4.630 -
12.85
13.80
15.13
14.06
19.56
34.84
29.17
43.29
16.43
21.78
17.08
23.70
65.65
84.31
82.15
94.07
64.86
68.31
68.24
65.43
0.79
16.00
13.91
28.64
B-1
-2
-3
-4
2.961
2.476
2.380
2.649
3.779
3.716
4.079
4.412
1.979
2.417
1.750
0.743
0.948 1.769 1.615
0.575 4.522 3.344 2.768 4.304 -
13.05
13.71
14.32
12.11
21.18
32.17
31.25
35.18
17.83
21.90
17.70
23.10
68.23
81.80
80.84
78.77
65.51
67.49
67.91
61.09
2.72
14.31
12.93
17.68
C-1
-2
-3
-4
2.420
2.382
2.289
2.380
3.005
3.355
4.137
4.623
1.973
2.116
1.544
0.917
0.928 1.659 1.662
0.568 3.709 3.466 2.745 3.668 -
11.65
12.13
14.18
11.59
20.83
32.64
29.74
37.61
16.53
21.88
19.20
24.05
61.82
75.86
80.04
79.49
63.87
65.87
67.35
61.17
-2.05
9.99
12.69
18.32
Av.
2.548 4.004 1.731 2.304 2.940 1.7690 13.21 30.62
20.10
77.75
65.59
12.16
Prices: cassava roots: baht 1.00/kg fresh roots at 30% starch; baht 0.02 reduction for
every 1% starch reduction
cassava leaves
4.00/kg dry leaves
2) Cutting height Cutting frequency
A = 15 cm
1 = 1½ months intervals after 1st cut
B = 20 cm
2 = 2 months intervals after 1st cut
C = 25 cm
3 = 2½ months intervals after 1st cut
4 = 3 months intervals after 1st cut
1)
Project IP3: improving cassava for the developed world.
Output 9-22
Table 13. Effect of cutting height and frequency on total dry leaf yield, fresh root yield and
net income in the cutting height x frequency trial for cassava leaf production conducted
in Rayong FCRC in 2002/03.
Cutting
frequency
(months)
Dry leaf yield(t/ha)
Height (cm)
15
20
25
Av.
11.65
12.13
14.18
11.59
12.52
13.21
14.54
12.59
Fresh root yield (t/ha)
Height (cm)
15
20
25
Av.
19.56
34.84
29.17
43.29
21.18
32.17
31.25
35.18
20.83
32.64
29.74
37.61
Net income (‘000 baht/ha)
Height (cm)
15
20
25
Av.
1½
2
2½
3
12.85
13.80
15.13
14.06
13.05
13.71
14.32
12.11
20.52
0.79
33.22 16.00
30.05 13.91
38.69 28.64
Av.
13.96
13.30 12.39 13.21 31.71 29.95 30.20 30.62 14.83
2.72 2.05
0.49
14.31 9.99 13.43
12.93 12.69 13.18
17.68 18.32 21.55
11.91
9.74 12.16
It may be concluded from these four types of experiments that special varieties can be
selected for optimizing cassava leaf production, but that high root production should not
be neglected as roots may account for 30-40% of the total value of a 1-year crop. Stakes
should be planted at intermediate spacing of 60x60 cm, while tops should be cut at 2½-3
month intervals in order to maximize both leaf and root production. High applications of
300-600 kg N/ha should be applied, in combinations with 0-150 kg P2O5 and about 150
K2O/ha. Part of the nitrogen should probably be applied after every cutting to stimulate
new shoot formation, while P and K should be applied at planting or fractionated at
planting and about 5-6 MAP. Complete data on protein yields are not yet available, but
average yields of 450-760 kg/ha were obtained in the first cut.
Activity 9.6 Conducting FPR trials on varieties, fertilization, weed control, green
manures, intercropping, erosion control and pig feeding in Thailand,
Vietnam and China.
Rationale
New high-yielding cassava varieties and improved production practices, including
effective ways to reduce erosion in cassava fields, have been developed in collaboration
with national cassava programs in Asia over the past 20-25 years. While new varieties
have spread almost spontaneously (but with large investments for multiplication and
distribution of planting material by governments of Thailand, Vietnam and China),
improved cultural practices, and especially soil conservation practices, have not been
widely or spontaneously adopted. This is mainly because farmers are either not aware of
these new technologies or they consider them too difficult or too expensive. About ten
years ago, a new project funded by the Nippon Foundation in Japan, aimed to enhance
the adoption of soil conservation and other improved practices by the use of farmer
participatory research (FPR) methodologies.
Output 9-23
2003 Annual Report
Specific Objectives
a) To develop efficient and locally adapted cultural practices to reduce erosion and
increase farm income by the conducting of FPR trials by farmers on their own fields.
b) To identify the most suitable cassava varieties for a particular location through FPR.
c) To enhance the adoption of new varieties and improved cultural practices.
Materials and Methods
The second phase (1999-2004) of the Nippon Foundation project is being conducted in
Thailand, Vietnam and China in collaboration with numerous research and extension
institutions in these countries (see list of collaborators). In 2003 the project further
expanded to include 32 sites in China, 33 sites in Thailand and 34 sites in Vietnam for a
total of 99 sites. In Thailand and Vietnam one “site” is usually a village with many
farmers conducting FPR trials, while in China one “site” is often a single farmer or a few
farmers conducting trials or testing new varieties. Figure 4 shows the location of those
sites. In 2003 the project expanded to nine new sites in China and eight new sites each
in Thailand and Vietnam. Tables 14 and 15 show the number and types of FPR trials
being conducted in Thailand and Vietnam in 2003/04. In Thailand this includes trials
on new varieties, erosion control, chemical fertilizers and organic manures, green
manures, weed control and plant spacing, for a total of 91 FPR trials. In Vietnam farmers
are conducting trials on new varieties, erosion control, chemical fertilizers and organic
manures, intercropping, plant spacing, weed control, varieties for leaf production, and pig
feeding with cassava root and leaf silage, for a total of 133 trials in 2003. In China
farmers evaluated mainly new varieties.
Before conducting FPR trials, farmers of a new site had generally visited FPR
demonstration plots to see a wide range of new technologies, or they had visited an old
site where some new technologies had already been tested or adopted. After these visits,
farmers in the new sites discussed among each other and selected the type of trials to be
conducted and the treatments/varieties to be tested. Project staff helped farmers set out
the trials, establish the selected treatments and provided planting material of new
varieties, green manures, new hedgerow species etc. to be tested. Project staff visited the
sites regularly to take data and to discuss/rectify any problems. At time of harvest all
participating farmers in the village helped each other to harvest about 20-100 m2 in each
plot. The harvested roots were piled up and a sign with the treatment, the yield of
cassava and intercrops obtained, and the soil loss measured, was placed next to the pile
of harvested roots. Later the same day or the following day a field day was organized with
participation of provincial, district and subdistrict officials, extension staff, and farmers
from the village and surrounding areas. They all received sheets showing the various
treatments in each trial, so they could evaluate each treatment and note down the yield
and other pertinent data when they visited each trial in the field. Finally, all data were
Project IP3: improving cassava for the developed world.
Output 9-24
tabulated, averaged over different sites (if the same trial was conducted by different
farmers) and gross income, production costs and net income were calculated for each
treatment. These results were discussed with all farmers participating in the field day,
after which farmers raised their hands to show their preferences for the various
treatments. The most preferred treatments would often be tested again the following
planting season. In China and north Vietnam most trials were planted in early spring,
i.e. Feb or March; in Central Vietnam they were planted in the late rainy season of DecMarch; in South Vietnam and Thailand they were mostly planted in the early rainy
season of May-June, but also in some areas in the late rainy season of Sept-Oct.
•
32
28
27
26
19
18
25
20 22
21
23
17
24
16 15
•
31
6
12
5 7 119
3 4
8 10
2 1
13 14
15
16
17
18
19
20
21
23 22
24
30
29 •
•
3 2 1
4 13 9 7 8 14
512
6 10
11
•
•
11
33
32
31
12 14 17
13 15 16
•
6 57 8
4
3 9
12
10
26
2529
2728
•
18
21 20 19
23 22
24
25
26
28
29 27
30
•
32 31
•
30
33
34
Figure 4. Location of FPR pilot sites in China, Thailand and Vietnam in
the Nippon Foundation cassava project in 2003.
Output 9-25
2003 Annual Report
Table 14. Number and type of FPR trials conducted by farmers in various sites in
Thailand in 2003/04.
Province
8 Kalasin
10 Roy Et
Subdistrict
Huay
Phueng
Phoo Chai
Nikhom
2
-
2
-
2
2
2
Khamphaung
Tabaekbaan
8
3
5
-
4
-
-
2
2
2
2
2
-
-
2
2
4
5
3
-
-
4
2
-
4
4
4
-
7
2
4
-
-
4
-
25
11
17
11
15
10
2
20 Nakhon
Ratchasima
28 Chonburi
Khonburi
31 Ratchaburi
Baan
Poong
Sai Yook
33 Kanchanaburi
Total
Erosion Chemical Chemical
Green
Weed
Plant
Varietes control fertilizers + org. fert. manures control spacing
District
Bo Thong
Total no. of FPR trials: 91
Kaset
Sawan
Khaw
Khalung
Sai Yook
Project IP3: improving cassava for the developed world.
Output 9-26
Table 15. Number and type of FPR trials conducted by farmers in various sites in Vietnam in 2003.
Province
1 Thai
Nguyen
2
3
4
5
6 Bac Can
7 Tuyen
Quang
8
District
Tien Phong
-
-
“
“
Pho Yen
Phu Luong
Na Ri
Son Duong
Dac Son
Minh Duc
Hong Tien
Yen Do*
Hao Nghia
Thuong Am
1
3
2
-
1
1
-
3
“
Van Yen
Yen Chau
Thanh Ba
15
16
17 Ha Tay
Phu Ninh
“
Thach
That
Chuong
My
Luong Son
Lac Son*
Nhu Xuan
A Luoi
19 Hoa Binh
20
22 Thanh Hoa
26 Thua
Thien-Hue
27
28
29
30 Dong Nai
31 Binh
Phuoc
32
33 Baria
Vungtau
34
Village Varietes Erosion Fertili- Inter- Plant Weed
control zation cropping spacing control
Pho Yen
11 Yen Bai
13 Son La
14 Phu Tho
18
Commune
Nam Dong
“
Huong Tra
Thong
Nhat
Dong Phu
-
-
-
-
-
3
1
3
1
-
-
-
-
-
3
2
1
-
-
-
1
1
-
1
1
-
1
3
-
-
-
-
-
1
2
2
1
1
3
3
3
-
-
-
-
Tran Phu
5
-
-
3
-
-
-
-
Dong Rang
4
1
1
-
4
3
-
5
-
-
-
Yen Cat
Hong Ha
1
2
1
3
-
-
5
Thuong Long
Huong Hoa
Huong Van
An Vien
2
1
1
2
-
3
-
-
-
1
-
8
-
Dong Tam
1
1
2
-
-
-
-
-
3
2
2
2
2
2
-
3
-
-
-
2
2
2
2
-
-
-
-
35
23
24
26
8
3
1
13
“
“
Am
Thang
Hong
Tien
Mau A
Phuong Linh Kieu
Tung
Thong Nhat Phu Ho
Bao Thanh
Thach Hoa
Chan Than Minh Lap
Chau Duc Suoi Rao
“
Pig
Varieties
for leaf feeding
production
Son Binh
Total
-
Total no. of FPR trials = 133
Results
Results of the 169 FPR trials conducted in Vietnam in 2002/03 have been “summarized”
in 49 Tables. Tables 6 and 16 to 20 are a few examples of results obtained in each type
of trial. Most of these trials show very convincingly that: 1) some new varieties, especially
KM 94 and a few new breeding lines, can markedly increase yields; 2) contour hedgerows
of vetiver grass, Paspalum atratum or Tephrosia candida are very effective in reducing
erosion, especially when combined with intercrops and adequate fertilization; 3)
Output 9-27
2003 Annual Report
applications of well-balanced fertilizers, especially when combined with animal manure,
will markedly increase yields and net income; 4) intercropping with one or two rows of
peanut between cassava rows is highly profitable; 5) cassava should be planted at about
80x80 or 100x80 cm; 6) weeding by hand is still more economic than using preemergence herbicides or planting on plastic mulch; and 7) pig feed rations containing
60% ensiled cassava roots is about 20% cheaper per kg weight gain than conventional
maize-based diets.
Tables 21 to 23 show results of similar FPR trials conducted in Thailand, while Table 24
is a summary table of six FPR variety trials conducted in Hainan, China.
Activity 9.7 Enhancing adoption of new varieties and improved management
practices through farmer participatory research (FPR) and extension
(FPE) activities.
Rationale
The target of the 5-year second phase of the Nippon Foundation project is to benefit at
least 8000 farmers by improving their livelihoods and protecting the soil from further
degradation. In order to reach this many farmers it was necessary to keep expanding the
number of pilot sites, to involve more and more farmers in FPR and to develop additional
FPE methodologies to reach other farmers and other communities that were not directly
involved in FPR trials.
Specific Objectives
a) To enhance the adoption by as many farmers as possible of new varieties and improved
practices that increase farmers’ income and reduce soil erosion and nutrient depletion in
cassava-based cropping systems.
Project IP3: improving cassava for the developed world.
Output 9-28
Table 16. Average results of two FPR variety trials conducted by farmers in Hong Tien
commune, Son Duong district, Tuyen Quang, Vietnam in 2002.
Varieties
1.
2.
3.
4.
5.
6.
7.
8.
9.
Vinh Phu
La Tre (SC205)
KM60
KM94
KM95-3
KM98-7
OMR38-71-12
OMR37-52-6
OMR37-52-8
1)
out of 38 farmers
Cassava
yield
(t/ha)
28.70
32.00
35.70
39.50
32.00
32.60
38.00
55.70
27.50
Gross
Product.
Net
income
costs
income
 (‘000 dong/ha) 
B/C
14,350
16,000
17,850
19,750
16,000
16,300
19,000
27,850
13,750
3.31
3.70
4.12
4.56
3.70
3.76
4.39
6.43
3.18
4,330
4,330
4,330
4,330
4,330
4,330
4,330
4,330
4,330
10,020
11,670
13,500
15,420
11,670
11,970
14,670
23,520
9,420
Farmers’
preference1
 (%)
5
5
18
84
3
13
29
100
0
Table 17. Results of three FPR erosion control trials conducted by farmers in Dong Rang
village, Dong Xuan commune, Luong Son district, Hoa Binh Vietnam in 2002.
1. Nguyen Van Tho; 16% slope
Treatments1)
1. C+T; no NPK; no hedgerows (TP)
2. C+T; with NPK; vetiver hedgerows
3. C+T; with NPK; Tephrosia hedgerows
4. C+P; with NPK; vetiver hedgerows
5. C+P; with NPK; Tephrosia hedgerows
Dry
Yield (t/ha)
Gross Product.
Net
soil loss
income2) costs2)
income
(t/ha) cassava taro peanut
 (‘000 dong/ha) 
12.4
8.75
2.60
6,798
4,780
2,018
0
16.87
2.60
10,452
5,732
4,720
0
15.30
3.00
10,185
5,732
4,453
0
15.30
0.51
9,690
6,242
3,448
0
14.63
0.60
9,884
6,242
3,642
B/C
1.42
1.82
1.78
1.55
1.58
2. Mr. Bui Thanh Mai; 12% slope
Treatments1)
1. C+P; no NPK; no hedgerows (TP)
2. C+P; with NPK; no hedgerows
3. C+P; with NPK; Tephrosia hedgerows
4. C+P; with NPK; Flemingia
hedgerows
5. C+P; with NPK; vetiver grass
hedgerows
Output 9-29
Dry
Yield (t/ha)
Gross
Product.
Net
soil loss
income2)
costs2)
income
(t/ha) cassava peanut
 (‘000 dong/ha) 
8.80
10.00
0.53
7,415
5,290
2,125
2.60
14.60
0.48
9,210
6,192
3,018
0.25
14.40
0.45
8,955
6,242
2,713
0.25
15.60
0.40
9,220
6,242
2,978
0
15.60
0.40
9,220
6,242
2,978
B/C
1.40
1.49
1.43
1.48
1.48
2003 Annual Report
3. Mr. Bui Thi Bam; 16% slope
Treatments1)
1. C; no NPK; no hedgerows (TP)
2. C+P; with NPK; vetiver grass hedgerows
Dry
Yield (t/ha)
Gross
Product.
Net
soil loss
income2)
costs2)
income
(t/ha) cassava peanut
 (‘000 dong/ha) 
26.00
6.50
2,925
3,000
-75
0
13.75
0.60
9,487
6,242
3,245
B/C
0.98
1.52
1) C = cassava, T = taro, P = peanut; NPK = 40 kg N+ 40 P O + 80 K O/ha; TP =
2 5
2
traditional practice
2) prices: cassava
dong
450/kg fresh roots
taro
1,100/kg fresh corms
peanut
5,500/kg dry pods
urea (45% N)
2,500/kg
1,000/kg
fused Mg phosphate (15% P2O5)
2,500/kg
KCl (60% K2O)
peanut seed (84 kg/ha)
10,000/kg dry pods
taro corms (300 kg/ha)
1,100/kg
labor
10,000/manday
labor for monoculture without fert. (300 md/ha)
= 3.0 mil. dong/ha
labor for intercropping without fert. (445 md)
= 4.45 mil dong/ha
labor for fertilizer applic. (8 md/ha)
= 0.08 mil. dong/ha
labor for hedgerow planting and maintenance
= 0.05 mil. dong/ha
cost of NPK fertilizers
= 0,822 mil. dong/ha
Table 18. Average results of three FPR fertilizer trials conducted by farmers in Suoi Rao
and Son Binh villages, Chau Duc district, Baria-Vungtau, Vietnam in 2002/03.
Treatments
Cassava
yield
(t/ha)
Gross
Fertil.
Product.
Net
income2)
costs
costs
income
 (‘000 dong/ha) 
B/C
1.
2.
3.
4.
25.88c
29.23bc
46.93a
42.73ab
12,422
14,030
22,526
20,510
2.20
2.07
3.44
2.78
0N+0P+0K
80N+40P2O5+80K2O
40N+40P2O5+80K2O
40N+40P2O5+40K2O
+5t/ha FYM
CV (%)
1)
0
1,037
814
1,648
5,640
6,777
6,554
7,388
6,782
7,253
15,972
13,122
Farmers’
preference
(%)
0
0
100
0
14.49
Prices:
cassava dong
480/kg fresh roots
urea (45% N)
2,500/kg
1,100/kg
SSP (17% P2O5)
2,500/kg
KCl (60% K2O)
FYM
200/kg
labor for cassava monoculture without fertilizer:
5.64 mil dong/ha
labor for fertilizer application:
0.1 mil dong/ha
Project IP3: improving cassava for the developed world.
Output 9-30
Table 19. Average results of three FPR intercropping trials conducted by farmers in Tran
Phu commune, Chuong My district, Ha Tay, Vietnam in 2002.
Yield (t/ha)
Gross
Seed Product. Net
Farmers’
preference
Cassava Intercrop income1) costs
costs income
(%)
 (‘000 dong/ha)  B/C
Treatments
1.
2.
3.
4.
5.
1)
Cassava monoculture
C+1 row peanut
C+2 rows peanut
C+mungbean
C+soybean
22.2
25.0
24.0
22.9
25.7
0.884
1.916
0
0.400
7,770
13,170
17,980
8,015
10,995
0
480
960
500
500
4,330
5,810
6,290
5,830
5,830
3,440
7,360
11,690
2,185
5,165
Prices: cassava
dong 350/kg fresh roots
peanut
5,000/kg dry pods
peanut seed
12,000/kg dry pods
mungbean seed
25,000/kg dry grain
labor for cassava monoculture without fertilizers:
labor for intercropped cassava without fertilizers:
cost of fertilizers
labor for fertilizer application:
1.79
2.27
2.86
1.37
1.89
0
35
85
10
10
2.8 mil. dong/ha
3.8 mil. dong/ha
1.43 mil. dong/ha
0.1 mil. dong/ha
Table 20. Effect of increasing levels of ensiled cassava roots in pig feed on the average
growth of nine pigs, on feed conversion ratio and feed costs in Huong Van
commune, Huong Tra, Thua Thien Hue, Vietnam in 2002/03.
Treatments1)
No.
pigs
Control diet
45% ECR
60% ECR
9
9
9
1)
2)
Life weight (kg)
initial 3 months
27.72
27.56
28.44
75.94
77.95
80.32
LWG2)
(g/day)
535.8
559.9
576.5
DFI2)
(kg DM)
1.58
1.55
1.63
FCR2)
(kg DM/kg
gain)
2.89
2.73
2.76
Feed cost
(VND/kg gain)
7057
5960
5763
%
100
84.5
81.7
ECR = ensiled cassava roots
LWG = life weight gain; DFI = daily feed intake; FCR = feed conversion ratio
Output 9-31
2003 Annual Report
Table 21. Results of an FPR fertilizer and manure trial conducted in Khut Dook village,
Baan Kaw, Daan Khun Thot, Nakhon Ratchasima, Thailand in 2001/02.
Root Starch Gross Fertilizer Production Net
yield content income2
cost3)
costs3)
income
(t/ha)
(%)
(‘000 baht/ha)
Treatments1)
1. No fertilizers or manure
2. Chicken manure+rice hulls, 400
kg/rai
3. Pelleted chicken manure, 100
kg/rai
4. 15-7-18 fertilizer, 50kg/rai
5. 13-13-21 fertilizer, 50kg/rai
6. 16-20-0 fertilizer, 50kg/rai
7. 15-15-15 fertilizer, 50kg/rai
18.75
30.42
25.0
26.2
21.56
34.98
0
2.50
10.87
17.15
10.69
17.83
26.70
21.1
30.71
2.00
15.39
15.32
29.68
32.22
26.08
30.36
24.1
27.4
25.9
26.9
34.13
37.05
29.99
34.91
2.66
3.13
2.50
2.81
16.73
17.89
15.61
17.07
17.40
19.16
14.38
17.84
1)1ha
= 6.25 rai
cassava
baht 1.15/tonne irrespective of starch content
3)Costs: chicken manure
1.0/kg
pelleted chicken manure
3.20/kg
15-7-18
8.50/kg
13-13-21
10.0/kg
16-20-0
8.0/kg
15-15-15
9.0/kg
harvest + transport roots
270/tonne
cassava production without fertilizer or harvest 5,812/ha
2)Prices:
Table 22. Results of an FPR intercropping trial conducted by a farmer in Thung Krabam
subdistrict, Law Khwan, Kanchanaburi, Thailand in 2002/03.
Treatments1)
C+monoculture
C+peanut
C+melon
C+pumpkin
C+sweet corn
Cassava Starch Intercrop Gross income2) Production
Net
yield
content
yield
cassava intercrop
costs
income
(t/ha)
(%)
(t/ha)
 (‘000 baht/ha) 
30.00
30.75
24.00
32.13
31.00
23.0
23.5
23.2
23.8
24.5
0.562
0.250
1.250
1.250
22.90
23.37
18.10
24.61
24.18
8.43
1.25
12.50
3.13
16.97
20.49
16.11
18.30
17.87
5.93
11.31
3.24
18.81
9.94
Farmers’
preference
(%)
0
0
10
5
C = cassava
Prices: cassava bath 0.89/kg fresh roots at 30% starch; 0.02 baht reduction per 1%
starch reduction
peanut
15/kg
melon
5/kg
pumpkin
10/kg
sweet corn
2.5/kg
1)
2)
Table 23. Results of an FPR green manure trial conducted by a farmer in Huay Faa
village, Nikhom, Huay Phueng, Kalasin, Thailand in 2002/03.
Project IP3: improving cassava for the developed world.
Output 9-32
Treatments1)
Root
yield
(t/ha)
Starch
content
(%)
1.
2.
3.
4.
24.63
26.00
32.00
20.253)
26.5
27.8
27.5
23.6
no green manure
C+mungbean
C+Crotalaria juncea
C+Canavalia ensiformis
Gross Product.
Net
income2)
costs
income
 (‘000 baht/ha) 
21.06
23.24
28.32
25.16
14.12
14.43
16.04
12.87
Farmers’
preference
(%)
6.94
8.81
12.28
12.29
0
12
0
78
1)Green
manures planted 1½ months after cassava, pulled up 1½ months later, except for
Canavalia which
was left to produce seed
2)Prices:
cassava
bath 0.96/kg fresh roots at 30% starch
0.03 baht reduction per1% starch reduction
Canavalia 10/kg dry grain
3)Also 250 kg/ha of Canavalia seed
Table 24. Results of FPR cassava variety trials conducted by farmers in six sites in
Hainan province, China in 2002/03.
Cassava root yield (t/ha)
Variety
SC 205
SC 124
SC 8002
SC 52)
SC 62)
ZM 8229
ZM 8316
ZM 8639
ZM 8641
ZM 8803
MBra 900
OMR 36-409
CMR 36-4012
1)
2)
A1)
B
23.25
27.50
33.25
21.00
21.25
27.00
21.25
21.50
24.00
24.06
24.69
31.25
34.38
-
-
-
C
81.25
-
D
E
F
57.81
-
12.50
18.87
12.75
14.62
10.75
18.62
18.50
13.87
22.37
32.87
31.87
31.25
52.60
26.25
29.37
32.37
34.37
-
15.62
Average Ranking
A-E-F
19.37
7
28.33
21.87
1
5
21.33
23.08
24.12
24.08
6
4
2
3
-
A = old Songtao village, Qiongzhong county
B = Nanlao village, Nankun town, Tunchang county
C = Maling village, Nankun town, Tunchang county
D = Qiaozhi farm, Danzhou city
E = October field farm, Changjiang county (average 2 Reps, based on 5 plants/plot)
F = Qifang town, Baisha county (average 2 Reps, based on 5 plants/plot)
SC 5 = ZM 9057; SC 6 = OMR 33-10-4
Output 9-33
2003 Annual Report
Materials and Methods
To reach as many farmers as possible and to show them the benefits of new cassava
varieties and improved practices developed specifically for their region, the following FPE
methodologies were used:
a) Cross-visits: Farmers from “new” sites visit farmers from “old” sites where new
practices have already been tested and adopted. This is an effective way of farmer-tofarmer extension.
b) Field days: At time of harvest, with participation of local officials, extension workers
and farmers of neighboring communities, to visit the FPR trials and discuss and
evaluate the results.
c) Large-scale field days: organized at provincial level with participation of national,
provincial and local government officials, school children, hundreds of farmers and
representatives of TV and newspapers.
d) Training courses: For researchers and extension staff or for local extension workers
and key farmers from each pilot site.
e) Setting up of community-based self-help groups: these are called “Cassava
Development Villages” in Thailand.
f) Production of pamphlets, booklets, posters and a project website: to distribute
information quickly and efficiently.
Results
Most of these FPE activities have been described in detail before (2002 CIAT Annual
Report, PE-3); they were continued in 2003 except that no large-scale field days were
organized this year. But smaller field days with participation of 100-200 people were
usually organized at each site when the cassava trials were harvested.
In 2002/03 there were three training events: in Dec 2002 there was an FPR training
course for farmers and local extensionist with 31 participants held in Khon Kaen,
Thailand; in Jan 2003 there was an FPR training course for 100 farmers and local
officials held in Yongning county, Guangxi, China; and in Aug 2003 there was an FPR
training course for 66 farmers and local officials in Nhu Xuan district of Thanh Hoa
province of Vietnam. During the Thai training course, farmers and extension workers
from each site were asked to bring some photos of project activities in their village in
order to make a poster. A farmer or extensionist from each site then explanined the
posters to the other course participants, so they could share and exchange experiences.
These hand-made posters were then transferred on to the computer by scanning the
photos and typing in the text, after which they were printed and distributed to each of the
district or subdistrict extension offices. Well-illustrated extension booklets about new
varieties and production practices were also produced in Vietnam and China in the local
languages.
Project IP3: improving cassava for the developed world.
Output 9-34
These activities led to the widespread adoption of new cassava varieties and cultural
practices, including erosion control practices. In China the testing of new varieties in
many FPR trials in Hainan and Guangxi provinces has resulted in the adoption of new
varieties – mainly SC 5 in Hainan, and GR 911 and Nanzhi 199 in Guangxi – in an
estimated 4,600 ha, corresponding to 1.6% of the total cassava area in these two
provinces.
Table 25 shows data for the adoption of vetiver grass contour hedgerows to control
erosion and the planting of green manures in various sites in Thailand in 2003. While an
additional 320,000 vetiver plants were planted in 2003, it is not sure to what extent this
increased the total length of hedgerows as some new vetiver plants were used to repair
old hedgerows. Farmers in several sites also planted green manures, either before
cassava or as an intercrop in cassava. There was a clear preference for Canavalia
ensiformis (jack bean) over other green manures as Canavalia is highly tolerant of poor
soils and drought, and does not compete too much when intercropped between cassava
rows; also, farmers can sell Canavalia seed back to the Land Development Department.
In Thailand, the Dept. of Agric. Extension continued to expand the number of
community-based self-help groups called “Cassava Development Villages” (see 2002 CIAT
Annual Report, PE3). In 2003 a total of 21 such groups had been initiated, each led by a
committee of five elected officers and each managing a rotating fund from which members
could borrow in case of emergencies or to buy production inputs. These 21 “villages” had
a total of 865 members and an average rotating fund of 70,000 baht ($1,750) each. Most
members had adopted new varieties - mainly KU 50, Rayong 90, Rayong 72 or Rayong 5 –
and most were applying chemical fertilizers, while some had planted vetiver grass
hedgerows or green manures (Table 25).
A recent survey (2002/03) indicate
Output 9-35
2003 Annual Report
Table 25. Extent of adoption of vetiver grass contour hedgerows for erosion control, and planting of green manures for soil
improvement in various FPR pilot sites in Thailand in 2003.
Adoption of vetiver grass
Cassava
Vetiver
Vetiver
1.
2.
3.
4.
5.
7.
8.
9.
10.
11.
12.
14.
15.
16.
17.
18.
19.
20.
21.
23.
24.
27.
28.
31.
32.
33.
Province
District
Subdistrict
Kalasin
Kalasin
Kalasin
Kalasin
Kalasin
Mueang
Mueang
Nong Kungsri
Sahatsakhan
Sahatsakhan
Kalasin
Kalasin
Kalasin
Roy Et
Kamphaengphet
Chayaphum
Nakhon Ratchasima
Nakhon Ratchasima
Nakhon Ratchasima
Nakhon Ratchasima
Nakhon Ratchasima
Nakhon Ratchasima
Nakhon Ratchasima
Prachinburi
Chachoengsao
Chachoengsao
Sra Kaew
Chonburi
Ratchaburi
Kanchanaburi
Kanchanaburi
Naamon
Huay Phueng
Don Chaan
Phoo Chai
Khanuwaralakburi
Thep Sathit
Thepharak
Thepharak
Sri Khiiw
Daan Khun Thot
Soeng Saang
Soeng Saaang
Khonburi
Naadii
SanaamChaikhet
Thaa Takiap
Wang Sombuun
Bo Thong
Baan Poong
Law Khwan
Sai Yook
Phuu Po
Khamin
Nong Bua
Noonburi
Noon Nam
Kliang
Naamon
Nikhom
Dong Phayung
Khampha-ung
Bo Tham
Naayaang Klak
Bueng Prue
Bueng Prue
Paang Lako
Baan Kaw
Noon Sombuun
Sratakhian
Tabaekbaan
Kaeng Dinso
Thung Prayaa
Khlong Takraw
Wang Sombuun
Kaset Suwan
Khaw Khalung
Thung Krabam
Sai Yook
Total 11
22
25
No. of
area with
farmers vetiver(ha)
(no. of
plants)
hedgerows
(km)1)
61
67
63
47
49.0
110.4
59.2
40.6
85,500
111,600
86,170
128,330
8.6
11.2
8.6
12.8
50
50
50
42
42
26
53
62
27
34
32
42
75
42
-
24.0
24.0
24.0
27.2
27.2
34.2
49.4
132.5
4.8
24.0
27.2
10.4
27.2
220.8
27.2
-
56,000
216,000
28,500
4,000
68,000
83,000
80,000
130,000
80,000
20,000
100,000
60,000
50,000
100,000
90,000
80,000
-
4.0
20.0
2.2
0.4
3.0
5.5
11.0
15.0
20.0
2.0
5.0
4.5
2.0
5.3
9.0
3.0
-
865
943.3
1,657,100
153.1
Adoption of green manures2)
Canava- Crotal.
Mung
Cowpe
a
lia
juncea
bean
(kg)
(kg)
(kg)
(kg)
73
300
3,000
6
300
-
6
-
6
-
1,000
-
-
-
306
6
6
1,500
5,873
Cassava area with hedgerows and hedgerow length are approximate, as some hedgerows were damaged by tractor while
others needed to be partially replanted
because of poor establishment due to drought.
2) Amount of seed provided to farmers in 2003.
1)
Project IP3: improving cassava for the developed world.
Output 9-36
that new cassava varieties were planted in about 1 million ha in Thailand, corresponding
to about 98% of the total cassava area. This, and the use of more fertilizers and other
improved practices has resulted in a gradual increase in cassava yields in Thailand from
13.8 t/ha in 1994 to 16.4 t/ha in 2002. This yield increase is valued at about US$50./ha. With an average cassava planted area of about 3 ha per household, each family
could receive an average $150.- extra gross income per year due to the combined use of
new varieties and improved cultural practices. For the whole country this may add up to
about 50 million more US dollars in cassava farmers’ pockets every year, while both
farmers and the cassava processing industry have also benefited from the higher starch
content of these new varieties.
Table 26 shows a detailed record of adoption of new technologies in six communes in Pho
Yen district and three communes in Son Duong districts in north Vietnam during
2002/03. The greatest extent of adoption occurred in Tien Phong and Dac Son
communes where the project has worked longest, starting in 1994. In 2002 most farmers
in these communities had adopted new varieties, mainly KM 98-7 and KM 95-3 in Pho
Yen and KM 94 in Son Duong district, intercropping with peanut or black bean, and
improved fertilization by combining chemical fertilizers high in K with pig manure. In Pho
Yen only a few households adopted erosion control practices since slopes in the area are
not very steep, but in Son Duong district 43 farmers planted hedgerows of vetiver or
Paspalum atratum to control erosion in 18 ha of cassava fields.
In Van Yen district of Yen Bai province in north Vietnam the completion of a new cassava
starch factory in early 2003 was accompanied by a rapid adoption of new high-yielding
varieties and improved cultural practices in order to increase cassava yields and produce
enough raw material for the factory. Thus, in early 2003, the area planted to new
varieties increased to 3000 ha in Van Yen (from 1000 ha in 2002) and 500 ha in
neighboring Yen Binh district, the application of 60 kg N, 40 P2O5 and 80 K2O/ha, the
intercropping of peanut in 55 ha and the planting of 500 km of double-row contour
hedgerows of Tephrosia candida or Paspalum atratum to control erosion, improve soil
fertility and supply cut-and-carry grass for cattle and buffaloes. In 2002 these improved
practices applied in 1000 ha had resulted in average yields of 30 t/ha compared with 10
t/ha for the traditional varieties and practices. At least 3000-5000 farmers in Yen Bai
province have benefited economically from the adoption of new technologies.
Over the past ten years a total of 24 medium- to large-scale cassava starch factories have
been built in Vietnam while another 18 factories are in the planning stage. In each case
the local government wants to ensure the availability of enough raw material by testing
new varieties and production practices and facilitating the adoption of new technologies
by the importation of planting material of new varieties, distribution of the the right type
of fertilizers, and seed of suitable intercrops or hedgerow species. It was estimated that
Output 9-37
2003 Annual Report
in 2001/02, 92,000 ha of cassava in Vietnam (35% of total cassava area) had been
planted with new varieties (mostly KM 94). This must have increased to far over 100,000
ha in 2002/03 and 2003/04. If we assume that the average cassava area in Vietnam is
still 0.27 ha per household, as it was in 1991/92 (Vietnam Cassava Production Survey),
then we can estimate that at least 350,000 farmers have adopted new varieties (and
probably also improved production practices) and have benefited economically from the
cassava research and FPR project conducted in Vietnam over the past ten years. Cassava
yields in Vietnam have gradually increased from 8.4 t/ha in 1994 to 12.6 t/ha in 2002
(FAO, 2003); the value of this additional production is about US$100/ha or 33 million US
dollars for the whole country. Cassava is now the third most important food crop (after
rice and maize) in Vietnam and is recognized as a very important vehicle for rural
development and the alleviation of poverty.
Project IP3: improving cassava for the developed world.
Output 9-38
Table 26. Extent of the dissemination of new cassava varieties in six communes of Pho
Yen district, Thai Nguyen, and three communes in Son Duong district of Tuyen
Quang province, Vietnam in 2002/03 and their effect on yield and gross income.
District
Commune
Variety
Pho Yen
1.
Tien Phong
2.
Dac Son
3.
Minh Duc
4.
Hong Tien
5.
Nam Tien
6.
Van Phai
Vinh Phu
KM 95-3
KM 98-7
Vinh Phu
KM 95-3
KM 98-7
Vinh Phu
KM 95-3
KM 98-7
Vinh Phu
KM 95-3
KM 98-7
Vinh Phu
KM 95-3
KM 98-7
Vinh Phu
KM 95-3
KM 98-7
Son Duong
1.
Am Thang
2.
3.
Hong Tien
Cap Tien
(local)
(local)
(local)
(local)
(local)
(local)
Vinh Phu
(local)
La Tre (local)2)
KM 603)
KM 943)
KM 95-3
KM 98-7
Vinh Phu
(local)
La Tre (local)
KM 94
KM 98-7
La Tre (local)
KM 60
KM 94
KM 98-7
No. of
farmers
Area
(ha)
Cassava
yield
(t/ha)
Gross
income1)
(‘000
d/ha)
18
50
50
6
5
11
34
34
34
12
16
28
9
9
9
9
17
25
0.64
1.60
3.56
0.27
0.22
0.49
1.62
0.65
0.86
0.61
0.58
1.42
0.40
0.29
0.36
0.32
0.70
1.46
25.4
32.3
34.3
25.1
28.3
30.5
23.5
27.4
28.1
25.4
31.1
32.3
23.5
28.1
30.2
23.8
29.2
31.3
12,700
16,150
17,150
12,550
14,065
15,250
11,750
13,700
14,050
12,700
15,550
16,150
11,750
14,050
15,100
11,900
14,600
15,650
14
2.57
19.8
9,900
10
9
23
13
3
2
3.65
0.42
5.06
1.05
0.14
0.25
16.1
28.3
36.3
27.1
30.0
17.0
8,050
14,150
18,150
13,550
15,000
8,500
26
30
1
20
20
20
20
11.39
4.29
0.14
12.25
0.20
0.28
0.20
15.4
36.1
28.0
19.6
29.4
35.7
30.0
7,700
18,050
14,000
9,800
14,700
17,850
15,000
1)Price:
cassava
dong 500/kg fresh roots.
Tre = SC 205 introduced from China in 1967-1972.
3)KM 60 = Rayong 60; KM 94 = KU 50, both introduced from Thailand in 1989-1991.
2)La
Output 9-39
2003 Annual Report
Collaborators:
Within CIAT:
Rod Lefroy, Coordinator in Asia, Vientiane, Laos
Peter Horne, FLSP Project, Vientiane, Laos
Hernan Ceballos, Project Manager, IP3
Outside CIAT:
Mr. Watana Watananonta, Project Coordinator for Thailand, DOA, Bangkok, Thailand
Dr. Surapong Charoenrath, Field Crops Research Inst., DOA, Bangkok, Thailand
Mrs. Atchara Limsila, Rayong Field Crops Research Center, DOA, Thailand
Mr. Danai Suparhan, Rayong Field Crops Research Center, DOA, Thailand
Mr. Somphong Katong, Rayong Field Crops Research Center, DOA, Thailand
Dr. Peaingpen Sarawat, Khon Kaen Field Crops Research Center, DOA, Thailand
Dr. Chairoj Wongwiwatchai, Khon Kaen Field Crops Research Center, DOA, Thailand
Dr. Chalaem Maswanna, Khon Kaen Field Crops Research Center, DOA, Thailand
Mrs. Saowari Tangsakul, Banmai Samrong Field Crops Res. Station, DOA, Thailand
Mr. Samnong Nual-on, Kalasin, Field Crops Res. Station, DOA, Thailand
Mr. Suttipun Brohmsubha, Bureau Agric. Com. Prom. and Mgmt., DOAE, Bangkok,
Thailand
Mr. Kaival Klakhaeng, Bureau Agric. Com. Prom. and Mgmt., DOAE, Bangkok,
Thailand
Mrs. Wilawan Vongkasem, Bureau Agric. Com. Prom. and Mgmt., DOAE, Bangkok,
Thailand
Mr. Suwit Phomnum, Bureau Agric. Com. Prom. and Mgmt., DOAE, Bangkok,
Thailand
Mr. Apichart Chamroenphat, Bureau Agric. Com. Prom. and Mgmt., DOAE, Bangkok,
Thailand
Mr. Chanchai Wiboonkul, Bureau Agric. Com. Prom. and Mgmt., DOAE, Bangkok,
Thailand
Ms. Sunan Muuming, Bureau Agric. Com. Prom. and Mgmt., DOAE, Bangkok,
Thailand
Ms. Naruemol Pukpikul, Bureau Agric. Com. Prom. and Mgmt., DOAE, Bangkok,
Thailand
Mrs. Chonnikarn Jantakul, Bureau Agric. Com. Prom. and Mgmt., DOAE, Bangkok,
Thailand
Ms. Methinee Keerakiat, Bureau Agric. Com. Prom. and Mgmt., DOAE, Bangkok,
Thailand
Mr. Somchit Chinno, Provincial Ext. Office, Kalasin, DOAE, Thailand
Mrs. Nuttaporn Jaruenjit, District Ext. Office, Naamon, Kalasin, DOAE, Thailand
Mr. Chaipipop Yotachai, District Ext. Office, Huay Phueng, Kalasin, DOAE
Mr. Thinnakorn Withayakorn, District Ext. Office, Sahatsakhan, Kalasin, DOAE,
Thailand
Mr. Nava Takraiklaang, District Ext. Office, Donchaan, Kalasin, DOAE, Thailand
Mrs. Anurat Srisura, Provincial Ext. Office, Nakhon Ratchasima, DOAE, Thailand
Mr. Choosak Aksonvongsin, District Ext. Office, Daan Khun Thot, DOAE, Thailand
Mr. Wirawudhi Kaewpreechaa, District Ext. Office, Daan Khun Thot, DOAE, Thailand
Mr. Chatphon Wongkaow, District Ext. Office,Khanuworalakburi, Kamphaengphet,
DOAE, Thailand
Mr. Sanga Saengsuk, Provincial Ext. Office, Kanchanaburi, DOAE, Thailand
Mr. Parinya Phaithuun, District Ext. Office, Lawkhwan, Kanchanaburi, DOAE,
Thailand
Mr. Sonsing Srisuwan, District Ext. Office, Thepsathit, Chayaphum, DOAE, Thailand
Project IP3: improving cassava for the developed world.
Output 9-40
Mr. Numchai Phonchua, District Ext. Office, Thatakiap, Chachoengsao, DOAE,
Thailand
Mr. Prayoon Kaewplod, Provincial Ext. Office, Chachoengsao, DOAE, Thailand
Mr. Sanit Taptanee, District Ext. Office, Nadi, Prachinburi, DOAE, Thailand
Mr. Sanit Phuumphithayanon, Provincial Ext. Office, Chayaphum, DOAE, Thailand
Mr. Banyat Vankaew, TTDI, Huay Bong, Nakhon Ratchasima, Thailand
Mr. Preecha Petpraphai, TTDI, Huay Bong, Nakhon Ratchasima, Thailand
Mrs. Supha Randaway, Land Development Dept. Bangkok, Thailand
Mrs. Kittiporn Srisawadee, Land Development Dept. Bangkok, Thailand
Mr. Decha Yuphakdee, Land Development Dept., Nakhon Ratchasima, Thailand
Dr. Somjat Jantawat, Kasetsart University, Bangkok, Thailand
Dr. Tran Ngoc Ngoan, Project Coordinator for Vietnam, Thai Nguyen Univ., Vietnam
Dr. Nguyen The Dang, Thai Nguyen University, Thai Nguyen, Vietnam
Mr. Nguyen Viet Hung, Thai Nguyen University, Thai Nguyen, Vietnam
Mr. Nguyen The Nhuan, Thai Nguyen University, Thai Nguyen, Vietnam
Dr. Thai Phien, National Institute of Soils and Fertilizers, Hanoi, Vietnam
Mr. Tran Minh Tien, National Institute of Soils and Fertilizers, Hanoi, Vietnam
Mr. Nguyen Hue, National Institute of Soils and Fertilizers, Hanoi, Vietnam
Mrs. Trinh Thi Phuong Loan, Root Crops Research Center, VASI, Hanoi, Vietnam
Mr. Hoang Van Tat, Root Crops Research Center, VASI, Hanoi, Vietnam
Mrs. Nguyen Thi Cach, Hue University of Agriculture and Forestry, Hue, Vietnam
Mrs. Nguyen Thi Hoa Ly, Hue University of Agriculture and Forestry, Hue, Vietnam
Dr. Hoang Kim, Hung Loc Agric. Research Center, IAS, Dong Nai, Vietnam
Mr. Nguyen Huu Hy, Hung Loc Agric. Research Center, IAS, Dong Nai, Vietnam
Mr. Vo Van Tuan, Hung Loc Agric. Research Center, IAS, Dong Nai, Vietnam
Mr. Tong Quoc An, Hung Loc Agric. Research Center, IAS, Dong Nai, Vietnam
Mr. Tran Cong Khanh, Hung Loc Agric. Research Center, IAS, Dong Nai, Vietnam
Dr. Tran Thi Dung, Thu Duc University of Agric. and Forestry, HCM, Vietnam
Mrs. Nguyen Thi Sam, Thu Duc University of Agric. and Forestry, HCM, Vietnam
Mr. Li Kaimian, Chinese Academy Tropical Agric. Sciences, Hainan, China
Mr. Huang Jie, Chinese Academy Tropical Agric. Sciences, Hainan, China
Mr. Ye Jianqiu, Chinese Academy Tropical Agric. Sciences, Hainan, China
Mr. Tian Yinong, Guangxi Subtrop. Crops Res. Inst., Nanning Guangxi, China
Mr. Li Jun, Guangxi Subtrop. Crops Res. Inst., Nanning Guangxi, China
Mr. Ma Chongxi, Guangxi Subtrop. Crops Res. Inst., Nanning Guangxi, China
Mrs. Chen Xian Xiang, Guangxi Subtrop. Crops Res. Inst., Nanning Guangxi, China
Mrs. Pan Huan, Guangxi Subtrop. Crops Res. Inst., Nanning Guangxi, China
Mr. Liu Jian Ping, Honghe Animal Husbandry Station, Mengzhe, Yunnan, China
Mr. J. Wargiono, Central Institute for Food Crops, Bogor, Indonesia
Dr. Koeshartojo, Research Inst. for Legumes and Tuber Crops, Malang, Indonesia
Budget $342,508 from Nippon Foundation, Tokyo, Japan
Output 9-41
2003 Annual Report