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 tG m w 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]^\[T_ jT[ jTX_ jTZZ jTZ` jTY[Z jT[ zt`W`TY\ jTZZ zt`W`TY\ jTZ` jT[XZ jTZZ jT[ jTX_ ztX^[XTX jTZZ jTZ` hs~Thjtk hjtk hjtkTt zi hjtkThjy hjtkThjy hjtkThjy hjtkTy{ hjtkTy{ hjtkTy{ hjtkTy{ hjtkX hjtkX hjtkTy{ hjtkTy{ hjtkTy{ hjtkTy{ hjt kTWX y{Thjt k hjtkY hjtkY hjtkY hjtkY hjtkY hjtkY hjtkX hjtkX hjtkX hjtkX hjtkX hjtkX hjtkX hjtk[ hjtk[ hjtk[ hjtk[ hjtk[ hjtkY hjtkY hjtkY hjtkY hjtkY hjtkY hjtkY hjtk[ hjtk[ hjtk[ hjtk[ hjtk[ Output 8-87 uUGG uUGGG {GuUG uUGwG LG G G G G Y_ ] \] 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) uUGG LG G G G t m w uUGG G uUGG G G{GuUG j 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 hjtkTy{ hjtkTy{ hjtkY hjtkX hjtkY hjtkY hjtkThjy hjtk[ hjtkY hjtkThjy hjtkTw{ hjtkY hjtkY hjtk[ hjtkX hjtkX hjtkTy{ hjtkTy{ hjtkTy{ hjtkTy{ hjtkTy{ 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 G 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