IOBC-WPRS Bulletin 27(3), 2004

Transcription

IOBC/WPRS
Working group "GMOs in Integrated Plant Production"
Proceedings of the meeting
Ecological Impact of Genetically Modified Organisms
at
Prague (Czech Republic)
26 – 29 November, 2003
Editor:
Jörg Romeis & Franz Bigler
IOBC wprs Bulletin
Bulletin OILB srop
Vol. 27(3), 2004
The IOBC/WPRS Bulletin is published by the International Organization for Biological and Integrated
Control of Noxious Animals and Plants, West Palearctic Regional Section (IOBC/WPRS)
Le Bulletin OILB/SROP est publié par l‘Organisation Internationale de Lutte Biologique et Intégrée
contre les Animaux et les Plantes Nuisibles, section Regionale Ouest Paléarctique (OILB/SROP)
Copyright: IOBC/WPRS 2004
The Publication Commission:
Horst Bathon
Federal Biological Research Center
for Agriculture and Forestry (BBA)
Institute for Biological Control
Heinrichstrasse 243
D-64287 Darmstadt (Germany)
Tel +49 6151 407225, Fax +49 6151 407290
e-mail: [email protected]
Luc Tirry
University of Gent
Laboratory of Agrozoology
Department of Crop Protection
Coupure Links 653
B-9000 Gent (Belgium)
Tel +32 9 2646152, Fax +32 9 2646239
e-mail: [email protected]
Address General Secretariat IOBC/WPRS:
INRA – Centre de Recherches de Dijon
Laboratoire de Recherches sur la Flore Pathogène dans le Sol
17, Rue Sully, BV 1540
F-21034 Dijon Cedex
France
ISBN 92-9067-164-3
web: http://www.iobc-wprs.org
i
Preface
In January 2003, an IOBC/WPRS study group was established to address the ecological
impact of genetically modified organisms (GMOs) within the context of IP. The objectives of
the group were defined as follows:
1.
2.
3.
4.
Exchange and dissemination of scientific knowledge on the ecological impact of
genetically modified organisms
To evaluate the compatibility and integration of genetically modified organisms with
biological control and IPM
Resistance management of the target organisms
Development and harmonisation of methods for testing the ecological impact of
genetically modified organisms
In September 2003, the council of the IOBC/WPRS changed the status of the study group to a
working group. The first full meeting of this new working group was held in Prague, Czech
Republic, from November 26 – 29, 2003. The interest in this first meeting was overwhelming
with more than one hundred participants from 23 countries attending. Participants of the
meeting were retrieved from universities and governmental research institutions, private
industry and regulatory agencies. It appeared that the working group could provide a platform
for discussion among the different parties and to bridge the gap between research, regulation
and application.
During the first two days of the meeting, two keynotes, 32 oral contributions and more
than 40 posters were presented. On the third day, seven half-day long workshops were held.
In total 34 contributions and reports from the different workshops are published in this
volume.
On behalf of all participants, I like to thank all the colleagues that organized a workshop
or acted as session moderators. A special thank goes to the local organizer František Sehnal of
the Entomological Institute in České Budějovice and his team for the excellent work in setting
up the meeting in Prague.
Jörg Romeis
Convenor IOBC/wprs working group
‘GMOs in Integrated Production’
ii
iii
Contents
Preface ………………………………………………………………………………………... i
Contents ……………………………………………………………………………………… iii
I. Key notes
Plant transformation: methodology, applications and the potential for unintended effects.
A.M.R. Gatehouse ……………………………………………………………………………. 1
Molecular solutions for increasing biosafety of transgenic plants.
J. Gressel & H. Al-Ahmad …………………………………………………………………… 7
II. Presentations
Testing rubidium marking for measuring adult dispersal of the corn borer Sesamia
nonagrioides: first results.
R. Albajes, J. Eras, C. López, X. Ferran, J. Vigatà & M. Eizaguirre ……………………….. 15
Analysis of web content of Theridion impressum L. Koch (Araneae: Theridiidae) in
BT (DK 440 BTY, MON 810, Cry1Ab) and isogenic (DK 440) maize.
K. Árpás, F. Tóth & J. Kiss …………………………………………………………………. 23
Impact of transgenic oilseed rape on soil arthropod assemblages.
G. Burgio, F. Ramilli, M.C. Fiore & F. Cellini ……………………………………………... 31
Potential effect of GNA-transgenic potatoes on adult aphid parasitoids.
A. Couty & J. Romeis ………………………………………………………………………. 37
Monitoring of pest and beneficial insect populations in summer sown Bt maize.
G. Delrio, M. Verdinelli & G. Serra ………………………………………………………… 43
Assessing expression of Bt-toxin (Cry1Ab) in transgenic maize under different
environmental conditions.
A. Dutton, M. D’Alessandro, J. Romeis & F. Bigler ……………………………………….. 49
Tracking Bt-toxin in transgenic maize to assess the risks on non-target arthropods.
A. Dutton, L. Obrist, M. D’Alessandro, L. Diener, M. Müller, J. Romeis & F. Bigler .......... 57
Comparison of the nodulation ability and abundance of aerobic bacteria in the
rhizosphere of transgenic and non-transgenic lines of alfalfa.
N. Faragová, J. Faragó & J. Gálová …………………………………………………………. 65
Research programme to monitor corn borer resistance to Bt-maize in Spain.
G.P. Farinós, M. De La Poza, M. González-Núñez, P. Hernández-Crespo, F. Ortego
& P. Castañera ………………………………………………………………………………..73
iv
Results of a 4-year plant survey and pitfall trapping in Bt maize and conventional
maize fields regarding the occurrence of selected arthropod taxa.
B. Freier, M. Schorling, M. Traugott, A. Juen & C. Volkmar ................................................ 79
Population development of some predatory insects on Bt and non-Bt maize hybrids
in Turkey.
M. Güllü, F. Tatli, A.D. Kanat & M. İslamoğlu ..................................................................... 85
The GMO Guidelines Project and a new ecological risk assessment.
A. Hilbeck, D. Andow & E. Underwood …………………………………………………… 93
European corn borer (Ostrinia nubilalis): Studies on proteinase activity and proteolytical
processing of the B.t.-toxin Cry1Ab in transgenic corn.
R. Kaiser-Alexnat, W. Wagner, G.-A. Langenbruch, R.G. Kleespies, B. Keller
& B. Hommel .......................................................................................................................... 97
First investigations on the effects of Bt-transgenic Brassica napus L. on the trophic
structure of the nematofauna.
B. Manachini, S. Landi, M.C. Fiore, M. Festa & S. Arpaia ……………………………….. 103
Studies on the effects of Bt corn expressing CryIAb on two parasitoids of Ostrinia
nubilalis Hb. (Lepidoptera: Crambidae).
B. Manachini & G.C. Lozzia ………………………………………………………………. 109
Implications for the parasitoid Campoletis sonorensis (Hymenoptera: Ichneumonidae)
when developing in Bt maize-fed Spodoptera littoralis larvae (Lepidoptera: Noctuidae).
M. Meissle, E. Vojtech & G.M. Poppy ................................................................................. 117
Production of Cry1Ab toxin in E. coli for standardisation of insect bioassays.
Nguyen Thu Hang, T. Meise, G.-A. Langenbruch & J.A. Jehle ........................................... 125
No effects of Bt maize on the development of Orius majusculus.
X. Pons, B. Lumbierres, C. López & R. Albajes ………………………………………….. 131
Impact of growing Bt-maize on cicadas: Diversity, abundance and methods.
S. Rauschen, J. Eckert, A. Gathmann & I. Schuphan ........................................................... 137
Impact of genetically modified herbicide resistant maize on the arthropod fauna.
I.I. Rosca …………………………………………………………………………………… 143
A biannual study on the environmental impact of Bt maize.
F. Sehnal, O. Habuštová, L. Spitzer, H.M. Hussein & V. Růžička ....................................... 147
Determination of fungi species, relationship between ear infection rates and
fumonisin quantities in Bt maize.
F. Tatli, M. Güllü & F. Ozdemir ........................................................................................... 161
v
Spider communities in Bt maize and conventional maize fields.
C. Volkmar, M. Traugott, A. Juen, M. Schorling & B. Freier .............................................. 165
Larvicidal activities of transgenic Escherichia coli against susceptible and Bacillus
thuringiensis israelensis-resistant strains of Culex quinquefasciatus.
M.C. Wirth, W.E. Walton, R. Manasherob, V. Khasdan, E. Ben-Dov, S. Boussiba
& A. Zaritsky ………………………………………………………………………………. 171
Peculiarities of Cry proteins to be taken into account during their in vivo and
in vitro study.
I.A. Zalunin, L.P. Revina, L.I. Kostina & G.G. Chestukhina ……………………………… 177
III. Workshop reports
Workshop report - Hybridization & Fitness of Hybrids.
D. Bartsch & H. den Nijs ...................................................................................................... 187
Workshop report - Impact of GM crops on pollinators.
D. Babendreier & S. Kühne .................................................................................................. 191
Workshop report - Impact of GM crops on natural enemies.
J. Romeis …………………………………………………………………………………... 193
Workshop report - Biodiversity implications off-crop.
A. Lang …………………………………………………………………………………….. 197
Work shop - Resistance management.
A. Gathmann ………………………………………………………………………………. 203
Workshop report - Monitoring/Bioindicators.
S. Arpaia …………………………………………………………………………………… 205
Workshop report - Soil organisms and functions.
W. Büchs …………………………………………………………………………………… 209
vi
vii
List of Participants
AASEN, Solveig
Agricultural University of Norway, INA-Ur, Postboks
5003, 1432 Ås, Norway
E-mail: [email protected]
ALBAJES, Ramon
Universitat de Lleida, Centre UdL-IRTA, Rovira Roure
191, 25198 Lleida, Spain
AN, Ying
Faculty of Agriculture, Iwate University, Ueda 3-18-8,
Morioka 020-8550, Japan
ARPAIA, Salvatore
Italian National Agency for Energy and Environment, S.S.
106 Jonica Km 419,5, 75023 Rotondella (MT), Italy
ÁRPÁS, Krisztina
Szent István University, Faculty of Agricultural and
Environmental Sciences, Department of Plant Protection,
Páter K. u. 1., 2103 Gödöllő, Hungary
BABENDREIER, Dirk
Agroscope FAL Reckenholz, Swiss Federal Research
Station for Agroecology and Agriculture, Reckenholzstr.
191, 8046 Zurich, Switzerland
BALL, Louise F.
Defra, 3/1410, Ashdown House, 123 Victoria St., London,
UK
BARTSCH, Detlef
Robert Koch Institut - Center for Gene Technology,
Wollankstrasse 15-17, 13187 Berlin, Germany
BIGLER, Franz
BIRCH, A. Nick E.
Host Parasite Coevolution Programme, Scottish Crop
Research Institute, Invergowrie, Dundee DD2 5DA,
Scotland, UK
BRINDZA, Peter
Fakulta Biotechnologie a Potravinárstva SPU, Tr. A.
Hlinku 2, 949 76 Nitra, Slovak Republic
viii
BRØDSGAARD, Henrik
Danish Institute of Agricultural Sciences, Research Centre
Flakkebjerg, 4200 Slagelse, Denmark
BÜCHS, Wolfgang
Federal Biological Research Centre for Agriculture and
Forestry, Institute for Plant Protection in Field Crops and
Grassland, Messeweg 11/12, 38104 Braunschweig,
Germany
BURGIO, Giovanni
DISTA, University of Bologna, via Fanin 42, 40127
Bologna, Italy
BUTTS, Edgar R.
248 Wood House Road, Fairfield, Connecticut, USA
CHÁB, David
Research Institute of Crop Production, Drnovská 507, 161
06 Praha 6 – Ruzyně, Czech Republic
CHAMPION, Gillian
Broom‘s Barn Research Station Higham, Bury St.,
Edmunds, Suffolk IP28 6NP, UK
COUTY, Aude
Laboratoire de Biologie des Entomophages, Université
Picardie - Jules Verne, 33 rue Saint Leu, 80039 Amiens,
France
DĄBROWSKI, Zbigniew T.
Department of Applied Entomology, Warsaw Agricultural
University, Poland
DELRIO, Gavino
Dipartimento Protezione delle Piante, Università di
Sassari, via E. De Nicola, 07100 Sassari, Italy
DEN NIJS, Hans C.M.
Institute for Biodiversity & Ecosystem Dynamics,
University of Amsterdam, Kruislaan 318, Amsterdam
1098 SM, The Netherlands
DEWAR, Alan
Broom’s Barn Research Station Higham, Bury St.
ix
DINTER, Axel
DuPont de Nemours, DuPont Str. 1, 61352 Bad Homburg,
Germany
DOUBKOVÁ, Zuzana
Ministry of the Environment, ČR, Vršovicka 65, 100 10
Praha 10, Czech Republic
DROBNÍK, Jaroslav
Biotrin, Viničná 5, Praha 2, Czech Republic
ECKERT, Jörg
Biology V, RWTH Aachen, Worringerweg 1, 52076
Aachen, Germany
FARAGOVÁ, Natália
Research Institute of Plant Production, Bratislavská cesta
122, 921 68 Piešťany, Slovak Republic
FREIER, Bernd
Federal Biological Research Centre of Agriculture and
Forestry, Institute for Intergrated Plant Protection,
Stahnsdorfer Damm 81, 14532 Kleinmachnow, Germany
GÁLOVÁ, Janka
Department of Botany and Genetics, Constantine the
Philosopher University, Tr. A.Hlinku 1, 949 74 Nitra,
Slovak Republic
E-mail: [email protected], [email protected]
GARCIA-ALONSO, Monica
Syngenta, Jealott s Hill Research Station, Brachnell, RG
42 6EY, UK
GARNER, Beulah H.
Broom's Barn Research Station Higham, Bury St
GATEHOUSE, Angharad M.R.
University of Newcastle (UNEW), School of Biology,
Newcastle upon Tyne, NE1 7RU, UK
GATHMANN, Achim
Biology V, RWTH Aachen, Worringerweg 1, 52076
Aachen, Germany
GRESSEL, Jonathan
Department of Plant Sciences, Weizmann Institute of
Science, Rehovot, 76100, Israel
x
GÜLLÜ, Mustafa
Plant Protection Research Institute, Department of
Entomology and Phytopathology, P.O. Box 21, 01321
Adana, Turkey
HABUŠTOVÁ, Oxana
Institute of Entomology, Acad. Sci., Branišovská 31, 370
05 České Budějovice, Czech Republic
HAYLOCK, Lisa
Broom's Barn Research Station Higham, Bury St.
HEARD, Matthew
Centre for Ecology and Hydrology, Monks Wood, Abbots
Ripton, Huntingdon, Cambridgeshire, PE28 2LS, UK
HILBECK, Angelika
Swiss Federal Institute of Technology, Geobotanical
Institute, Zürichbergstr. 38, 8044 Zurich, Switzerland
HILL, Steven
Defra, 3/ Hio Ashdown house, 123 Victoria St., London,
UK
HUSBY, Jan
Directorate for Nature Management, 7485 Trondheim,
Norway
HRAŠKA, Marek
Univ. South Bohemia, Fac. Biol. Sci. and Institute of Plant
Molecular Biology AS CR, Branišovská 31, 370 05 České
Budějovice, Czech Republic
HUSSEIN, Hany Mohamed
Pests and Plant Protection Dept. National Research
Centre, Al–Tahrir St., Dokki Cairo, Egypt
HÝBLOVÁ, Jana
Česká zemědělská univerzita, Kamýcká 129, 16521 Praha
6, Czech Republic
JACOBS, Erik
Monsanto Services International, Avenue de Tervuren
270-272, 1150 Brussels, Belgium
KAATZ, Hannes
Universtät Halle, Institut für Zoologie, AG Molekulare
Ökologie, Hoher Weg 4, 06099 Halle (Saale), Germany
xi
KALUSHKOV, Plamen K.
Institute of Zoology, Bulgarian Academy of Sciences
Boul. Tzar Osvoboditel 1, 1000 Sofia, Bulgaria
KOCOUREK, František
KRAUS, Pavel
ÚKZÚZ (CISTA), Hroznová 2, 656 06 Brno, Czech
Republic
KUČERA, Ladislav
Research Institute of Crop Protection, Drnovská 507, 161
06 Praha 6 – Ruzyně, Czech Repulic
KÜHNE, Stefan
Federal Biological Research Centre for Agriculture and
Forestry, Institute for Integrated Plant Protection,
LANDI, Simona
Istituto di Entomologia Agraria, Università degli Studi di
Milano, via Celoria 2, 20133, Milano, Italy
LANG, Andreas
Bavarian State Research Center for Agriculture, Institute
of Plant Protection, Lange Point 10, 85354 Freising,
Germany
LUDY, Claudia
Bavarian State Research Centre for Agriculture, Institute
of Plant Protection, Lange Point 10, 85354 Freising,
Germany
LUKÁŠ, Jan
02 Praha 6 - Ruzyně, Czech Republic
LUMBIERRES, Belén
Universitat De Lleida, Centre UDL-IRTA, Rovira Roure
MANACHINI, Barbara
Istituto di Entomologia Agraria, Università degli Studi di
Milano, via Celoria 2, 20133 Milano, Italy
xii
MEISSLE, Michael
NAVRÁTILOVÁ, Miloslava
Státní rostlinolékařská správa, Zemědělská 1A, 613 00
Brno, Czech Republic
NEDVĚD, Oldřich
Faculty of Biological Sciences, Univ. South Bohemia,
Branisovska 31, 370 05 České Budějovice, Czech
Republic
NGUYEN, Thu Hang
DLR- Rheinpfalz, Breitenweg 71, 67435 Neustadt/Wstr.,
Germany
OBRIST, Lena
OELKE, Maren
Centre for Agriculture Landscape and Land Use
Researche, Eberswalder Strasse 84, 15374 Müncheberg,
Germany
ORTEGO, Felix
Centro de Investigaciones Biologicas, CSIC, Ramiro de
Maeztu 9, Madrid 28040, Spain
OVESNÁ, Jaroslava
Research Institute of Crop Protection, Drnovská 507, 161
POHL, Matthias
TÜV Hannover / Sachsen – Anhalt e.V., Am TÜV 1,
30519 Hannover, Germany
POLGÁR, Laszlo A.
Plant Protection Institute, Hungarian Acad. Sci., Herman
O. u. 15, POB. 102, 1525 Budapest, Hungary
PONS, Xavier
Universitat de Lleida Centre UDL– IRTA, Rovira Rovre ,
xiii
PREVOST, Geneviève M.
Lab. de Biologie des Entomophages, Université de
Picardie - Jules Verne, 33 rue Saint Leu, 80039 Amiens,
France
RADU, Constantin Stefan
Monsanto Europe s.a., Na Šafránce 27, 101 00, Praha 10,
Czech Republic
RAKOUSKÝ, Slavomír
Institute of Plant Molecular Biology AS ČR, Branišovská
31, 370 05 České Budějovice, Czech Republic
RAMILLI, Fabio
DiSTA- University of Bologna, via Fanin 42, 40127
Bologna, Italy
REUTER, Hauke
UFT (pept 10), University of Bremen, P.O. Box 330 470,
28334 Bremen, Germany
ŘÍHA, Karel Jr.
Research Instutute of Crop Production, Drnovska 507, 161
ŘÍHA, Karel
CISTA (ÚKZÚZ), Hroznová 2, 65606 Brno, Czech
Republic
ROMEIS, Jörg
ROSCA, Ioan I.
University of Agricultural Sciences and Veterinary
Medicine, Av. Marasti 59, sector 2, 71322 Bucharest,
Romania
SANVIDO, Olivier
SCHLÜNS, Ellen
Universität Halle, Institut für Zoologie, AG Molekulare
Ökologie, Hoher Weg 4, 06099 Halle (Saale), Germany
xiv
SCHORLING, Markus
SEHNAL, František
Institute of Entomology CAS, Branišovská 31, 370 05
České Budějovice, Czech Republic
SICK, Martina
Stahnsdorfer Damm 81,15432 Kleinmachnow, Germany
SINGER, Martin
MONSANTO ČR s.r.o., Rybková 1, 60200 Brno, Czech
Republic
SPAGNOLETTI, Angela
Ministry of the Environment, Via Cristoforo Colombo 44,
00147 Roma, Italy
SPITZER, Lukáš
Faculty of Biological Sciences, Univ. South Bohemia,
Branišovská 31, 370 05 České Budějovice, Czech
Republic
STARK, Christine
Centre for Agricultural Landscape and Land Use
Research, Eberswalder Strasse 84, 15374 Müncheberg,
Germany
SWEET, Jeremy
NIAB, Huntingdon Road, Cambridge, CB3 0LE, UK
SZEKERES, Dóra
Department of Plant Protection, Szent István University,
Páter Károly u.1., 2103, Gödöllő, Hungary
SZÉNÁSI, Ágnes
Szent István University, Department of Plant Protection,
Páter Károly u.1, 2103 Gödöllő, Hungary
SZENTKIRÁLYI, Ferenc
Plant Protection Institute, Hungarian Acad. Sci., Herman
Ottó út 15, 1022 Budapest, Hungary
xv
TATLI, Fahri
Plant Protection Research Institute, Department of
Phytopathology and Entomology, P.O. Box 21, 01321
Adan, Turkey
TENCALLA, Francesca
Monsanto, Services International S.A., Avenue de
Tervuren 270-272, 1150 Brussels, Belgium
TINLAND, Bruno
Monsanto Services International S.A., Avenue de
Tervuren 270-272, 1150 Brussels, Belgium
TORNIER, Ingo
GAB Biotechnologie GmbH, Eutinger Str. 24, 75223
Niefern, Germany
E-mail: Ingo.tornier@gab –biotech.de
UNDERWOOD, Evelyn
Swiss Federal College of Technology, Geobotanical
Institute, Zürichbergstr. 38, Zurich, Switzerland
VOLKMAR, Christa
Institut für Pflanzenzüchtung und Pflanzenschutz, L.Wucherer-Str. 02, 06108 Halle (Saale), Germany
WEHRES, Ute
University of Technology Aachen Department of Ecology,
Worringerweg 1, 52056 Aachen, Germany
WENNSTROM, Anders
Dept. of Ecology & Environmental Science, Umea
University, 901 87 Umea, Sweden
ZALUNIN, Igor A.
The Scientific Research Institute for Genetics and
Selection of Industrial Microorganisms, 1-st Dorozhni
proezd 1, 113545 Moscow, Russia
ZARITSKY, Arieh
Ben-Gurion University of the Negev, Dep. Life Sciences,
POB 653 Be’er- Sheva, Israel
ZEMEK, Rostislav
Institute of Entomology CAS, Branišovská 31, 370 05
České Budějovice, Czech Republic
xvi
I. Key notes
GMOs in Integrated Production
IOBC wprs Bulletin Vol. 27 (3) 2004
pp. 1-5
Plant transformation: methodology, applications and the potential for
unintended effects
Angharad M.R. Gatehouse
University of Newcastle, School of Biology, Newcastle upon Tyne, NE1 7RU, UK (E-mail:
[email protected])
Introduction
Plant breeding has always exploited genetic methods both by using natural genetic variation
combined with artificial selection, and by inducing new variability by artificial means. A
classic example of the latter is the use of potent mutagens such as radiation to generate new
crop varieties, developed in the early part of the 20th century. The International Atomic
Energy Agency lists more than 2000 new crop varieties currently in use which have come
from mutation breeding (Chrispeels & Sadava, 2003). In almost all cases these varieties have
novel characteristics that did not exist in the crop in nature. More recently recombinant DNA
technology, or molecular plant breeding, has been used to generate crops with novel traits
where exogenous DNA is integrated into the host genome using a range of techniques defined
as ‘gene transfer technologies’ (reviewed in Hansen & Wright, 1999). With plants, the two
most commonly used methods for delivering foreign DNA are (i) Agrobacterium-mediated
transformation (Hiei et al., 1994) and (ii) Biolistic (micro-projectile) bombardment (Christou
et al., 1991) (Fig. 1). Major advantages of this technology include the transfer of single genes,
the ability to use genes from different species i.e. thus increasing the available ‘gene-pool’,
the availability of tissue specific transgene expression and the opportunity for protein
engineering. It can thus be considered as a very controlled means of introducing desired traits,
such as resistance to insect pests, into specific crop species.
The concept of utilising a transgenic approach to host plant resistance was realised in the
mid 1990s with the commercial introduction of genetically modified corn (maize), potato and
cotton plants expressing genes encoding the entomocidal -endotoxin from Bacillus
thuringiensis (Bt). More recently this strategy has been extended to include the pyramiding
(stacking) of genes encoding different Bt toxins for greater levels of pest control and field
durability; currently > 10 million hectares are planted to Bt crops globally. Although not as
yet a commercial reality, other strategies based on the use of plant derived genes (enzyme
inhibitors, lectins) and those from animal sources, including insects (biotin-binding proteins,
neurohormones, enzyme inhibitors), are being developed. The use of fusion proteins to
increase the spectrum and durability of resistance is also actively being pursued (review see
Ferry et al., 2004). However, if transgenic insect-resistant crops are to play a useful role in
crop protection, it is apparent that they must be compatible with the other components of
integrated pest management (Gatehouse & Gatehouse, 1999).
1
2
Genetic Modification (GM)
Source organism
Transfer
DNA
Agrobacterium
Tungsten
micro-particles
Gall
formation
Tissue
culture
T
Two routes
to Gene transfer
Final
plant
...
..
..
Micropropagation
Particle gun
Figure 1. Diagrammatic presentation of methodologies for plant transformation.
Plant transformation
In nature, Agrobacterium tumefaciens is able to infect many plant species through wound
sites, where it causes a type of plant tumour known as a crown gall. The pathogenic ability of
these bacteria is associated with a plasmid known as the tumour-inducing (Ti) plasmid; the
tumour is produced by the transfer of a specific piece of DNA, termed the T-DNA, to the
plant cells where it becomes integrated into the plant chromosomal DNA. Agrobacteriummediated transformation thus exploits the biological ability of this bacterium to copy and
transfer the T-DNA by inserting the target genes for transfer into the T-DNA region.
However, since the natural Ti plasmids are very large (>200 kb), it is not possible to clone
target genes directly into these plasmids. In practice a binary vector system is usually
employed whereby the T-DNA is transferred from the Ti plasmid (which is then referred to as
‘Disarmed’ Ti plasmid) to a second plasmid known as the binary vector. This method of
transformation is often the method of choice since it is simple, resulting in permanent genetic
changes i.e. stable transformation. One of the major limitations of Agrobacterium-mediated
transformation was associated with the range of plant species known to be infected by this
bacterium; however, this spectrum of plant species has been extended from the original group
of dicots known to produce galls after infection, to include many other species, including
cereals, that exhibit little or no gall formation.
3
The biolistic method of plant transformation is based on the physical delivery of DNAor RNA-coated gold or tungsten particles into target tissues by acceleration. However, the
mechanism by which these particles are able to deliver DNA into living cells without damage
is still not clear. As expected, both methods have their own relative advantages. Since the
biolistic method relies on physical, rather than genetic, parameters it has a broad application
range and the technique is often genotype-independent; all major crops have been transformed
using particle bombardment. Another major advantage associated with this method is the
ability to simultaneously introduce multiple genes by co-transformation. Furthermore, it
enables linear fragments of DNA to be transformed into host cells without the need for
introducing plasmid backbone sequences i.e. it enables ‘clean’ transgene integration to be
carried out. However, one of the problems associated with this method is the potential transfer
of multiple copies of the gene of interest which can produce problems in gene expression and
instability of the transferred genes in the plant’s genome. Furthermore, the rate of
transformation efficiency is significantly lower compared to that obtained by Agrobacteriummediated transformation.
Unintended effects of plant transformation
Since their first commercialization, concerns have been expressed over the wide-scale
growing of genetically modified (GM) crops. These concerns have addressed both the
environmental safety of such crops, particularly in relation to gene flow and their impact on
non-target herbivores, and their safety in the human food chain.
The potential for unintended effects to occur as a consequence of transformation have
long been recognised. However, it is important to stress that the chances for this to occur are
no greater with recombinant DNA technology, as with any other form of plant breeding or
natural recombination. Furthermore, although such effects are known to occur, they do not
necessarily have consequences for either environmental or food safety, but must be taken into
account when assessing risk. It is thus important to have clear definitions as to what is meant,
for example, by ‘unintended’ effects (see Fig. 2). These were defined by a European Network
on Safety Assessment of Genetically Modified Crops (www.entransfood.com) as:
 Intended effects of genetic engineering are those that are targeted to occur from the
introduction of the gene(s) in question and that fulfil the original objectives of the
genetic transformation process.
 Unintended effects represent a statistically significant difference in the phenotype,
response, or composition of the GM plant compared with the parent from which it is
derived, but taking the expected effect of the target gene into account. Such
comparisons should be made when GM and non-GM counterparts are grown under the
same regimes and environments.
 Predictable unintended effects are those unintended changes that go beyond the
primary expected effect(s) of introducing the target gene(s), but that may be explicable
in terms of our current knowledge of plant biology, and metabolic pathway integration
and interconnections.
 Unpredictable unintended effects are those changes falling outside our present level
of understanding.
4
Figure 2. Diagrammatic presentation of plant genome and consequences for insertion of
foreign genes
Detection of unintended effects
Detection of unintended effects in plants developed by GM technology relies primarily on the
comparative analysis of levels of selected key nutrients and toxic compounds present in the
GM crop with its traditional non-modified counterpart. Identified alterations in composition
may fall within the natural range of variations and thus not be of toxicological concern, or fall
outside these ranges and thus require further toxicological/nutritional evaluation. These types
of analyses are referred to as Targeted Approaches and include:
 Demonstration of ‘substantial equivalence’, i.e. identification of similarities and
differences between the GM crop and a comparator with a history of safe use
(FAO/WHO, 2000)
 Demonstration of unintended effects by phenotypic selection and by investigation of
defined constituents.
Thus a critical component of the Targeted Approach is the selection of compounds to be
analysed. Unfortunately, to date, there is little guidance on which parameters should be
measured for the comparison, which analytical methods should be used and which sampling
procedures should be followed to provide statistically sound analyses.
5
In order to increase the probability of detecting unexpected effects, profiling techniques are
under development. These so called Non-Targeted Approaches are based on modern
genomic, protein and metabolite detection techniques which provides a ‘global’ overview of
gene expression and chemical composition of the GM and non-GM crop.
Conclusion
While the recombination mechanism provides plants with a “natural” means to develop new
genetic variability, such modifications could be a source of unintended effects both in
classical and modern breeding approaches. Since genes are ‘hot spots’ for recombination, the
production of deletions and filler DNA could result in changes to gene sequences, leading to
gene disruption and/or the production of novel proteins in plants (Fig. 2). However, one of the
advantages of recombinant DNA technology versus conventional plant breeding is the
inherently higher accuracy of the technology and the fact that molecular approaches can be
used to provide essential information on what genetic elements have been inserted into the
transgenic plant and the regions into which they have been inserted. This is not possible with
classical breeding approaches.
References
Chrispeels, M.J. & Sadava, D.E. 2003: Plants, Genes and Crop Biotechnology. 2nd Ed, Jones
& Bartlett, Sudbury, USA.
Christou, P., Ford, T.L. & Kofron, M. 1991: Production of transgenic rice (Oryza sativa L.)
plants from agronomically important indica and japonica varieties via electric discharge
particle acceleration of exogenous DNA into immature zygotic embryos. Bio/Technology
9: 957-962.
FAO/WHO 2000: Safety Aspects of genetically Modified Foods of Plant Origin. A report of a
joint FAO/WHO Expert Consultation on Foods Derived from Biotechnology, Geneva,
Switzerland, 29 May - 2 June 2000.
Ferry, N., Edwards, M.G., Mulligan, E.A., Emami, K., Petrova, A.S., Frantescu, M., Davison,
G.M., &Gatehouse, A.M.R. 2004: Engineering resistance to insect pests. In: Handbook
of Plant Biotechnology, Eds in Chief P. Christou and H. Klee. John Wiley and Sons Ltd,
UK: 372-394.
Gatehouse, J.A. & Gatehouse, A.M.R. 1999: Genetic engineering of plants for insect
resistance. In: Biological and Biotechnological Control of Insects Pests, Eds. J.E.
Rechcigl and N.A. Reichcigl, CRC Press LLC: 221-241.
Hansen, G. & Wright, M.S. 1999: Recent advances in the transformation of plants. Trends
Plant Sci. 4: 226-231.
Hiei, Y., Ohta, S., Komari, T. & Kumashiro, T. 1994: Efficient transformation of rice (Oryza
sativa L.) mediated by Agrobacterium and sequence-analysis of the boundaries of the TDNA. Plant J. 6: 271-282.
6
pp. 7-13
Molecular solutions for increasing biosafety of transgenic plants
Jonathan Gressel, Hani Al-Ahmad
Department of Plant Sciences, Weizmann Institute of Science, Rehovot, 76100, Israel
(E-mail: [email protected])
Abstract: There are many ways to prevent transgene introgression from crops to other varieties, or to
related weeds or wild species (containment strategies), as well as to preclude the impact should
containment fail (mitigation strategies). The needs are most acute with rice and sunflowers, which
have con-specific weeds, and with oilseed rape, sorghum, barley, which have closely related weeds.
Containment and mitigation are critical for pharmaceutical crops, where gene flow from the crop to
edible varieties must be precluded. Some gene flow (leakage) is inevitable with all containment
mechanisms and once leaked, could then move throughout populations of undesired species, unless
their spread is mitigated. Leakage even occurs with chloroplast-encoded genes, a >0.03% pollen
transmission was found in the field. We focused on mitigation, which should be coupled with
containment as a last resort. A mechanism for mitigation was proposed where the primary transgene
(herbicide resistance, etc.) is tandemly coupled with flanking genes that could be desirable or neutral
to the crop, but unfit for the rare weed into which the gene introgresses. Mitigator traits include
dwarfing, non-bolting, no secondary dormancy, no seed shattering, and poor seed viability, depending
on the instance. We demonstrated the potential utility of the concept using tobacco as a model, and
dwarfing as the mitigator with herbicide resistance as the primary gene. Hybrids with the tandem
construct were unable to reach maturity when grown interspersed with the wild type. Such mitigation
should greatly decrease risk of transgene movement especially when coupled with containment
technologies, allowing cultivation of transgenic crops having related weeds. As the number of
transgenic plants being released is increasing, and the problems of monitoring such genes increases
geometrically, we suggest that a uniform biobarcodeTM system be used, where a small piece of noncoding DNA carrying an assigned variable region is used to mark transgenic crops, allowing
monitoring.
Key words: gene introgression, transgene containment, tr ansgene mitigation, gene flow
Introduction
Farmers in most of the world have begun to realize the benefits that accrue from cultivating
transgenic crops, whether to prevent soil erosion by using post-emergence herbicides or use
less expensive/toxic insecticides while contributing to farmer and environmental health and
safety. Herbicide resistant crops are especially useful for controlling crop-related weeds where
there had been no herbicide selectivity. Several crops (e.g., wheat, barley, sorghum, rice,
squash, sunflower, sugarbeets, oats, and oilseed rape) can naturally interbreed with closely
related weedy relatives under field conditions, in both directions (Ellstrand et al., 1999;
Gressel, 2002). There is a concern that transgenes may escape from engineered crops into
related weedy species by hybridization and backcrossing. This could potentially result in
hybrids and their progeny with enhanced invasiveness or weediness (Ellstrand et al., 1999).
Many of the engineered genes such as those conferring resistance to herbicides, diseases, and
to stresses may grant a fitness advantage to a weedy species growing in the same agricultural
ecosystem. There is also the rather emotive issue of transgene flow from crops such as maize
bearing transgenes encoding pharmaceuticals to other varieties. Farm produced
7
8
pharmaceuticals especially enzymes and antibodies, can be produced inexpensively in plants,
without the need for animal tissue culture cells grown in a medium of expensive serum
albumin that is all too easily contaminated with pathogenic mycoplasms, prions and viruses.
Still, there is reason not to want the pharmaceutical transgenes in other varieties of the crop.
Two general approaches are discussed below to deal with the problems of transgene flow:
containment of the transgenes within the transgenic crop; transgenic mitigation of the effects
of the primary transgenic trait should it escape and move to an undesired target. While most
containment mechanisms will severely restrict gene flow, some gene flow (leakage) is
inevitable and could then spread through the population of undesired species, unless
mitigated.
Containing transgene flow
Several molecular mechanisms have been suggested to contain the transgene within the crop
(i.e. to prevent outflow to related species), or to mitigate the effects of transgene flow once it
has occurred (Gressel, 1999, 2002; Daniell, 2002). The containment mechanisms include
utilization of partial genome incompatibility with crops such as wheat and oilseed rape having
multiple genomes derived from different progenitors. When only one of these genomes is
compatible for interspecific hybridization with weeds, the risk of introgression could be
reduced if the transgene was inserted into the unshared genome where there is presumed to be
no homeologous introgression between the non-homologous chromosomes. It has not been
reported if this mechanism works in wheat, and it was modeled to be ineffectual for oilseed
rape (Tomiuk et al., 2000) due to considerable recombination between the B and C genomes.
Another containment possibility is to integrate the transgene in the plastid or
mitochondrial genomes (Maliga, 2002). The opportunity of gene outflow is limited due to
maternal inheritance of these genomes. This technology does not preclude the weed from
pollinating the crop, and then acting as the recurrent pollen parent. The claim of no paternal
inheritance of plastome-encoded traits (Bock, 2001; Daniell, 2002), was not substantiated.
Tobacco (Avni & Edelman, 1991) and other species (Darmency, 1994) often have between a
10–3–10–4 frequency of pollen transfer of plastid inherited traits in the laboratory. Pollen
transmission of plastome traits can only be easily detected using both large samples and
selectable genetic markers. A large-scale field experiment utilized a Setaria italica (foxtail or
birdseed millet) with chloroplast-inherited atrazine resistance (bearing a nuclear dominant red
leaf base marker) crossed with five different male sterile yellow- or green-leafed herbicide
susceptible lines. Chloroplast-inherited resistance was pollen transmitted at a 310-4
frequency in >780,000 hybrid offspring (Wang et al., 2004). At this transmission frequency,
the probability of herbicide resistance from plastomic gene flow is orders of magnitude
greater than by spontaneous nuclear genome mutations. Chloroplast transformation is
probably unacceptable for preventing transgene outflow, unless stacked with additional
mechanisms.
A novel additional combination that considerably lowers the risk of plastome gene
outflow within a field (but not gene influx from related strains or species) can come from
utilizing male sterility with transplastomic traits (Wang et al., 2004). Introducing plastomeinherited traits into varieties with complete male sterility would vastly reduce the risk of
transgene flow, except in the small isolated areas required for line maintenance. Such a double
failsafe containment method might be considered sufficient where there are highly stringent
requirements for preventing gene outflow to other varieties (e.g. to organically cultivated
ones), or where pharmaceutical or industrial traits are engineered into a species. Plastomeencoded transgenes for non-selectable traits (e.g. for pharmaceutical production) could be
9
transformed into the chloroplasts together with a trait such as tentoxin or atrazine resistance as
a selectable plastome marker. With such mechanisms to further reduce out-crossing risk,
plastome transformation can possibly meet the initial expectations.
Other molecular approaches suggested for crop transgene containment include: seed
sterility, utilizing the genetic use restriction technologies (GURT) (Oliver et al., 1998), and
recoverable block of function (Kuvshinov et al., 2001). Such proposed technologies control
out-crossing and volunteer seed dispersal, but theoretically if the controlling element of the
transgene is silenced, expression will occur. Another approach includes the insertion of the
transgene behind a chemically-induced promoter so that it will be expressed upon chemical
induction (Jepson, 2002). However, there is a possibility of an inducible promoter mutating to
become constitutive. Schernthaner et al. (2003) proposed an impractical technology using a
“repressible seed-lethal system”. The seed-lethal trait and its repressor must be simultaneously
inserted at the same locus on homologous chromosomes in the hybrid the farmer sows to
prevent recombination (crossing over), technology that is not yet workable in plants. The
hemizygote transgenic seed lethal parent cannot reproduce by itself, as its seeds are not viable.
If the hybrid could be made, half the progeny would not carry the seed lethal trait (or the trait
of interest linked to it) and they will have to be culled, which would not be easy without a
marker gene. The results of selfing or cross pollination within the crop and leading to
volunteer weeds where 100% containment is needed, would leave only 25% dead and 50%
like the hybrid parents and 25% with just the repressor. Thus, the repressor can cross from the
volunteers to related weeds as can the trait of choice linked with the lethal, and viable hybrid
weeds could form. The death of some seed in all future weed generations is inconsequential to
weeds that copiously produce seed, as long as the transgenic trait provides some selective
advantage.
None of the above containment mechanisms is absolute (Figure 1), but risk can be
reduced by stacking containment mechanisms together, compounding the infrequency of gene
introgression. Still, even at very low frequencies of gene transfer, once it occurs, the new
bearer of the transgene can disperse throughout the population if it has just a small fitness
advantage.
Fertilization by pollen
Buffer zone
Field of GM crop
Containment techniques:
• Plastid transformation
Other fields:
• Related crops
• Related weeds
• Male sterility
• Terminator (GURT)
• Repressible seed-lethal
(SL/R)
Transgenic mitigation:
> 10 %
< 10 -4
< 10 -6
Figure 1. Containment systems allowing gene flow in one or more directions.
10
Mitigating further flow of ‘leaked’ transgenes
If the transgene has a small fitness disadvantage, it will remain localized as a very small
proportion of the population. Therefore, gene flow should be mitigated by lowering the fitness
of recipients below the fitness of the wild type so that they will not spread. A concept of
“transgenic mitigation” (TM) was proposed (Gressel, 1999), in which mitigator genes are
added to the desired primary transgene, which would reduce the fitness advantage to hybrids
and their rare progeny, and thus considerably reduce risk. This TM approach is based on the
premises that: 1) tandem constructs act as tightly linked genes, and their segregation from
each other is exceedingly rare; 2) The TM traits chosen are neutral or favorable to crops, but
deleterious to non-crop progeny due to a negative selection pressure; and 3) Individuals
bearing even mildly harmful TM traits will be kept at very low frequencies in weed
populations because weeds typically have a very high seed output and strongly compete
among themselves eliminating even marginally unfit individuals (Gressel, 1999). Thus, it was
predicted that if the primary gene of agricultural advantage being engineered into a crop is
flanked by TM gene(s), such as dwarfing, uniform seed ripening, non-shattering, antisecondary dormancy, or non-bolting genes in a tandem construct, the overall effect would be
deleterious after introgression into weeds, because the TM genes will reduce the competitive
ability of the rare transgenic hybrids so that they cannot compete and persist in easily
noticeable frequencies in agroecosystems (Gressel, 1999). Weeds are usually copious pollen
producers and set large numbers of seeds, many of which germinate during the following
season.
120
WT Alone
1600
TM Alone
WT Alone
TM Alone
2.5
1200
1
cm
80
2.5
1
cm
2.5 cm
800
40
2.5 cm
400
1
1
0
120
WT
Mixed Segregants
5
2.5
1
cm
0
1600
TM
Mixed Segregants
WT
TM
1200
80
800
40
5
cm
1/ 2.5
0
5 cm
400
2.5
1
1/ 2.5 / 5 cm
0
60
80
100
120
140 60
Time (d)
80
100
120
140
100
120
140
100
120
140
Time (d)
Figure 2. Suppression of growth and flowering of TM (transgenic mitigator) bearing tobacco
plants carrying a dwarfing gene in tandem with a herbicide resistance gene (open symbols)
when in competition with the wild type (closed symbols) (right panels), and their normal
growth when cultivated separately without herbicide (left panels). The wild type and
transgenic hemizygous semi-dwarf/herbicide resistant plants were planted at 1, 2.5, and 5 cm
from themselves or each other, in soil. See Al-Ahmad et al. (2004) for further details.
11
We used tobacco (Nicotiana tabacum) as a model plant to test the TM concept: a tandem
construct was made containing an ahasR (acetohydroxy acid synthase) gene for herbicide
resistance as the primary desirable gene, and the dwarfing gai (gibberellic acid-insensitive)
mutant gene as a mitigator (Al-Ahmad et al., 2004). Dwarfing would be disadvantageous to
the rare weeds introgressing the TM construct, as they could no longer compete with other
crops or with fellow weeds, but is desirable in many crops, preventing lodging and producing
less straw with more yield. The dwarf and imazapyr resistant TM transgenic hybrid tobacco
plants (simulating a TM introgressed hybrid) were more productive than the wild type when
cultivated alone. They formed many more flowers than the wild type, which is an indication
of a higher harvest index (Figure 2). Conversely, the TM transgenics were weak competitors
and highly unfit when co-cultivated with the wild type in ecological simulation competition
experiments (Figures 2, 3). The lack of flowers on the TM plants in the competitive situation
(Figure 2) led to a zero reproductive fitness of the TM plants grown in a 1:1 mixture with the
wild type at the spacings used, which are representative of those of weeds in the field (Figure
3). The highest vegetative fitness was less than 30% of the wild type (Figure 3).
Vegetative
Fitness
Reproductive
0.3
0.3
0.2
0.2
0.1
0.1
5 cm
2.5 cm
1 cm
0
60
80
100
120
0
140
100
120
140
Time (d)
Figure 3. Suppressed vegetative and reproductive fitness of TM transgenics in competition
with wild type tobacco. The points represent the calculated ratio of data for TM to wild type
plants in Figure 2.
Thus, it is clear that transgenic mitigation should be advantageous to a crop growing alone,
while disadvantageous to a crop-weed hybrid living in a competitive environment. If a rare
pollen grain bearing tandem transgenic traits bypasses containment, it must compete with
multitudes of wild type pollen to produce a hybrid. Its rare progeny must then compete with
more fit wild type cohorts during self-thinning and establishment. Even a small degree of
unfitness encoded in the TM construct would bring about the elimination of the vast majority
of progeny in all future generations as long as the primary gene provides no selective
advantage while the linked gene confers unfitness. Further large-scale field studies will be
needed with crop/weed pairs to continue to evaluate the positive implications of risk
mitigation. We have inserted the same construct into oilseed rape and are testing the selfed
progeny, as well as hybrids with the weed Brassica campestris=B. rapa. The rare hybrid
offspring from escaped pollen bearing transgenic mitigator genes would not pose a dire threat,
12
especially to wild species outside fields, as the amount of pollen reaching the pristine wild
would be minimal. Pollen flow decreases exponentially with distance, belying the unbased
‘demographic swamping’ by ‘Trojan genes’ giving rise to ‘migrational meltdown’, as
predicted by Haygood et al. (2003). Indeed, their model is based on a series of assumptions
that do not describe plant behaviour. They assume that the wild species will be pollinated
only by the crop, ignore the buffering effect of the soil seedbank, and assume replacement
rates similar to animals having a few progeny per generation, instead of thousands that
compete during self-thinning.
The containment of pharmaceutical transgenes has been physical, and as evidenced by
recent human error that allowed temporary volunteer escape of “Prodigene” maize with such
genes. The biological containment strategies described above may be preferable to depending
on humans, and the mitigation strategies should work as well. Maize pharmaceutical
transgenes are expressed in embryo tissues, and a potential tandem mitigating gene could be
any dominant gene that affects the endosperm, e.g. the various “shrunken seed” loci,
especially those where sugar transformation to starch is inhibited. Such shrunken seeds, with
their high sugar content, are somewhat harder to store than normal maize but are extremely
unfit in the field, and are unlikely to over winter. Because the endosperm of corn is 67%
pollen genes, it is important that expression of pharmaceutical encoding genes be only in the
embryo.
Monitoring transgene movement
Using the various containment and mitigation strategies it should be possible to keep ‘leaks’
below risk thresholds, which should be mandated by science-based regulators on a case-tocase basis. As the numbers of transgenic species being released is increasing, and the
problems of monitoring for such genes increases geometrically, we suggest that a uniform
biobarcodeTM system be used, where a small piece of non-coding DNA with uniform
recognition sites are at the ends (for single PCR primer pair amplification) with an assigned
variable region in between. Thus, PCR-automated sequencing could be used to determine the
origin of ‘leaks’, contamination, liability, as well as intellectual property violations (Gressel &
Ehrlich, 2002).
Acknowledgements
The research on transgenic mitigation was supported by the Levin Foundation, by INCO–DC,
contract no. ERB IC18 CT 98 0391, H. A.-A. by a bequest from Israel and Diana Safer, and
J.G. by the Gilbert de-Botton chair in plant sciences.
References
Al-Ahmad, H.., Galili, S. & Gressel, J. 2004: Tandem constructs to mitigate transgene
persistence: tobacco as a model. Mol. Ecol. 13: 697-710.
Avni, A. & Edelman, M. 1991: Direct selection for paternal inheritance of chloroplasts in
sexual progeny of Nicotiana. Mol. Gen. Genet. 225: 273-277.
Bock, R. 2001: Transgenic plastids in basic research and plant biotechnology. J. Mol. Biol.
312: 425-438.
Daniell, H. 2002: Molecular strategies for gene containment in transgenic crops. Nature
Biotech. 20: 581-586.
13
Darmency, H. 1994: Genetics of herbicide resistance in weeds and crops. In: Herbicide
Resistance in Plants: Biology and Biochemistry, eds. Powles and Holtum. Lewis, BocaRaton: 263-298.
Ellstrand, N.C., Prentice, H.C. & Hancock, J.F. 1999: Gene flow and introgression from
domestic plants into their wild relatives. Ann. Rev. Ecol. System. 30: 539-563.
Gressel, J. 1999: Tandem constructs: preventing the rise of superweeds. Trends Biotech. 17:
361-366.
Gressel, J. 2002: Molecular Biology of Weed Control. Taylor and Francis, London.
Gressel, J. & Ehrlich, G. 2002: Universal inheritable barcodes for identifying organisms.
Trends Plant Sci. 7: 542-544.
Haygood, R., Ives, A.R. & Andow, D.A. 2003: Consequences of recurrent gene flow from
crops to wild relatives. Proc. Royal Soc. B. 270: 1879-1886.
Jepson, I. 2002: Inducible herbicide resistance. United States Patent 6,380,463.
Kuvshinov, V., Koivu, K. & Kanerva, A. & Pehu, E. 2001: Molecular control of transgene
escape from genetically modified plants. Plant Sci. 160: 517-522.
Maliga, P. 2002: Engineering the plastid genome of higher plants. Curr. Opin. Plant Biol. 5:
164-172.
Oliver, M.J., Quisenberry, J.E., Trolinder, N.L.G. & Keim, D.L. 1998: Control of plant gene
expression. United States Patent 5,723,765.
Schernthaner, J.P., Fabijanski, S.F., Arnison, P.G., Racicot, M. & Robert, L.S. 2003: Control
of seed germination in transgenic plants based on the segregation of a two-component
genetic system. Proc. Natl. Acad. Sci. USA 100: 6855-6859.
Tomiuk, J., Hauser, T.P. & Bagger-Jørgensen, R. 2000: A- or C-chromosomes, does it matter
for the transfer of transgenes from Brassica napus. Theor. Appl. Genet. 100: 750-754.
Wang, T., Li, Y., Shi, Y., Reboud, X., Darmency, H. & Gressel, J. 2004: Low frequency
transmission of a plastid-encoded trait in Setaria italica. Theor. Appl. Genet. 108: 315320.
14
II. Presentations
pp. 15-21
Testing rubidium marking for measuring adult dispersal of the corn
borer Sesamia nonagrioides: first results
Ramon Albajes1, Jordi Eras2, Carmen López1, Xavier Ferran2, Josepa Vigatà1, Matilde
Eizaguirre1
1
Universitat de Lleida, Centre UdL-IRTA, Lleida, Catalunya, Spain; 2Universitat de Lleida,
Departament de Química, Lleida, Catalunya, Spain (E-mail: [email protected])
Abstract: Transgenic corn with the insecticidal capacity of Bacillus thuringiensis Berliner has been
deployed in Spain in the last five years, reaching about 32,000 ha in the present 2003 to control the
corn borers Sesamia nonagrioides Lef. and Ostrinia nubilalis Hbn. The former species is more
damaging than the latter in most of Spain. Because it is oligophagous on a narrow range of Gramineae
and it is largely sedentary, there is a serious risk that S. nonagrioides will develop resistance to B.
thuringiensis toxins if no strategies to prevent it are implemented. Insertion of non-Bt corn amongst Bt
fields is one of the strategies for delaying resistance development. For the viability of this strategy, the
dispersal capacity of the target species needs to be known. To this end, a method for measuring the
dispersal capacity of S. nonagrioides based on marking adults with rubidium was developed. A plot of
maize was sprayed with rubidium chloride and then infested with S. nonagrioides eggs. Adults
emerged from larvae developed in treated plants were caught in pheromone and light traps located at
several distances (0-400m) from the source. Catches were collected weekly and analysed in the
laboratory. Moths were dried and then digested by nitric acid treatment, and the rubidium content in
the resulting material was measured by flame atomic emission spectrometry. Control moths consisted
of adults of the same sex developed on untreated maize plants. Almost 3,000 S. nonagrioides adults
were caught in 2002 in light and pheromone traps and analysed for their rubidium contents. Catches
did not statistically vary during the season (from August to October) in relation to both the distance
from the rubidium source and the common direction of wind when pheromone trap catches are
considered whereas results from light traps show that more adults are caught in the traps closer to the
Rb source. It is provisionally concluded that refuges can be located at least 400 m from Bt fields with
no decrease in the probability of a female that has emerged from a non-Bt field mating with a male
that has emerged from a Bt field or vice versa.
Key words: dispersal, Bt maize, rubidium, Sesamia nonagrioides
Introduction
The Mediterranean Corn Borer, Sesamia nonagrioides Lefèbvre, affects maize yield in several
countries of the Mediterranean basin. Overwintering larvae pupate in the spring, adults emerge
from late March to May and mated females lay eggs into the leaf sheath. In the study area, S.
nonagrioides has two complete generations and a third incomplete one. Newly hatched larvae
initiate stem tunnelling into stems or the ear until pupation. In late summer or early autumn
mature larvae move down to the base of the stem where they excavate a cell and overwinter.
Tunnelling limits the intake of water and nutrients by developing plants and reduces plant
yield. Later in the season, yield reductions result from stalk breakage and the feeding activity
of the larvae on kernels. Pathogens that enter through S. nonagrioides feeding wounds in
kernels may additionally reduce grain quality.
Several commercial varieties of genetically modified corn with the insecticidal capacity
of Bacillus thuringiensis Berliner toxins, Bt varieties, have been successfully tested against
15
16
Ostrinia nubilalis Hübner in the U.S.A. (Anonymous, 1998) and some of them have proved to
be also resistant to S. nonagrioides (author’s observations). This kind of transgenic corn has
been deployed in Spain in the last five years, reaching about 32,000 ha in the present 2003.
On the basis of the experience obtained with chemical insecticides in the last decades,
transgenic resistant varieties may loose their efficacy if populations of targeted insects
become resistant to B. thuringiensis toxins (Gould, 1998). The Scientific Committee of Plants
of the European Union recommended in 1999 the inclusion of S. nonagrioides in the Btresistance monitoring as a serious candidate to develop resistance to B. thuringiensis toxins
due to several features of its biology. Among these features the Commission cited the
monophagy of the insect on maize and sorghum and a few wild Graminaceae and Typhaceae,
that it does not disperse through the season, and that the female is rather sedentary
(Anonymous, 2003). Several tactics for delaying the development of resistance in insect
populations exposed to the bacterial toxins have been pointed out. Among these is the
creation of refuges of non-Bt corn that are a reservoir of individuals with susceptibility genes.
One critical assumption of this strategy is that susceptible adults emerging from non-Bt maize
can mate with adults from Bt or non-Bt maize with the same probability (Gould, 1998). If
non-Bt refuges are too far from the Bt field the chances of this assumption being valid are
low.
The dispersal capacity of the targeted insect, an aspect of the S. nonagrioides biology that
is very poorly known, therefore determines the size and placement of non-Bt refuges. Among
adults of the first generation, that is adults emerged from overwintering individuals, long
displacements to colonise maize fields early in the season have been pointed out by Larue
(1984). Consistently with these authors, we have repeatedly observed that maize fields are
colonised by S. nonagrioides adults even in cases in which fields are isolated from other moth
sources. On the other hand, in a previous work in which mating behaviour was studied in the
laboratory (López et al., 2003), the authors observed that S. nonagrioides females rarely move
before mating and, although this feature must be confirmed in more real field conditions, this
would mean that mainly males are responsible for gene flow at long distances.
Complementarily, adult movement between adjacent Bt and non-Bt fields has been studied by
the authors in recent years on the basis that the Bt field is not likely to produce adults during
most of the season and males caught in S. nonagriodies pheromone traps placed in Bt fields
migrated from outside the field (author’s unpublished results). This kind of study, however,
does not usually allow one to measure long-distance movements, so other techniques are
needed.
Insect population movement has been studied with several techniques, including insect
marking and the capture of marked insects at increasing distances from the release point
(Reynolds et al., 1997). Insects can be effectively marked by producing an increased
concentration in their bodies of a normally rare element. In the case of rubidium, it can be
incorporated into the host-plant metabolism by spraying it with rubidium chloride so that it is
naturally assimilated into herbivore body tissues. Rubidium has been used to mark insects for
the purpose of dispersal studies of herbivores and predators (van Steenwyk, 1991; Prasifka et
al., 2001). In order to preliminarily test whether S. nonagrioides adults could be marked with
rubidium by applying it on the foliage of their host plant (maize), and to optimise application
and analytical methods, several previous laboratory and greenhouse trials were carried out in
2001. In this paper the results obtained in 2002 on the applicability of the marking method in
the field are presented.
17
Material and methods
Marking adults in the field
In previous experiments, the optimal doses and frequency of rubidium chloride application for
S. nonagrioides adult marking were determined. In addition to sufficiently increasing the
rubidium concentration in the insect body for its clear identification in the analysis of body
contents, the rubidium dose must be as cheap as possible in field application and avoid
phytotoxical effects on the sprayed maize plants.
One maize field of at least 1 km length was selected in the Lleida area (Catalonia, Spain).
The coordinates of the experimental field were estimated using a GPS (Garmin emap). The
geo-referenced Universal Mercator are utm-31n, X=294,015-294,715; Y=4628243-4628985.
Within the field, a plot of 1,000 sqm (17 x 60) was selected for spraying. The plot was
sprayed with a solution of rubidium chloride (Sigma, >99% purity) at a concentration of 3.5
g/l by means of a Pamany atomiser (3HP Ergolino engine and 25 atm Annovi/reverbero
pump). The treatment was applied twice in a one-week interval (July 4, 10) when maize was
in Hanway’s (1966) stage 6, before the beginning of the second S. nonagrioides adult flight,
in order to cover all the larval development of the second generation with a high amount of
rubidium in the plants. Rubidium chloride solution was applied at a rate of 1000 l/ha early in
the morning to avoid the drift to the neighbouring plots by wind. The plot was infested with a
total of 34 S. nonagrioides egg batches on June 26, July 1 and July 4 to assure a sufficient
infestation.
Adult captures
In an east-west axis with the centre in the rubidium-treated plot, ten pheromone traps were
placed on 4 July, each charged with 100 μg of the S. nonagrioides pheromone blend (Z)-11hexadecen-1-yl acetate; (Z)-11-hexadecen-1-ol; (Z)-11-hexadecenal, and dodecyl acetate in a
ratio of 77:8:10:5 (Sans et al., 1997)]. Two traps were placed in the centre of the treated plot
and four each in the eastern and western orientations at distances increasing by 100 m;
therefore two traps were placed at 0 m, and four traps in each direction at 100, 200, 300 and
400 m. Catches were counted and collected every week until 17 October, and taken to the
laboratory where they were kept frozen until processing and analysis during the winter.
Additionally and mainly to catch females, light traps were also placed (9 July) at increasing
distances from the treated plot in a E-W axis but were separated from the pheromone traps by
at least 50 m. One light trap was placed in the treated plot, and the other four traps at 200 and
400 m in the eastern or western direction respectively. Trapping and recording dates were the
same as those indicated for the pheromone traps. Catches, however, were identified and
recorded in the laboratory. As for the pheromone traps, the catches were kept frozen in the
laboratory until processing and analysis.
Sample preparation and rubidium content analysis
Individual moth samples were dried in an oven maintained at 70±3ºC for 24 hours and then
weighed. After drying, 0.5 ml of concentrated HNO3 (69%) was added to each sample. To
complete moth digestion, 0.4 ml of H202 (30%) was added. Samples were maintained in a
sand bath at 70±5ºC until digestion and then dried. Each digested moth was then diluted to 0.5
ml HNO3 (1+1), 0.5 ml NaCl (1.8 w/v) and 4 ml distilled water.
Samples were analysed for rubidium content by flame atomic emission spectrometry
(FAES) using a SOLAAR 929 UNICAM spectrometer. Samples were drawn into the
nebuliser chamber directly. The rubidium content of the sample was then measured by the
amount of energy emitted at 780.0 nm, the most sensitive wavelength for rubidium detection
18
in AE. Total intensity was the average of three measures over a period of 3 s, with the
absolute rubidium concentration established by calibration of a standard solution of 10, 25,
50, 100, 250 μg/l. The standard solution was prepared by dilution of a purchased standard of
995 μgRb/ml (Sigma).
Untreated adults are needed to compare their rubidium contents with that of field adults.
Adults emerged from the laboratory rearing were discarded as they frequently displayed a
high level of rubidium, probably acquired through the wheat germ or the beer yeast, two
abundant components of the rearing medium. Therefore, adults emerged from larvae
developed on untreated maize plants were collected in the field and used as background
adults. A sample was considered as marked if its rubidium level exceeded the mean
background plus three times the standard deviation. This gives a maximum error of 0.13%.
Statistical analyses
Captures of marked insects in pheromone or light traps at each distance and orientation during
the recording period were analysed with a split-plot design (Gomez & Gomez, 1984). As
orientation was not found to significantly affect (P>0.05) trap catches, ANOVA was
performed with distance as the main plot and the week as the subplot. When needed LSD was
used to compare means.
Results and discussion
A total number of 1833 and 552 adults were caught in pheromone and light traps respectively
and they were processed and analysed for Rb contents. From those 136 and 37 were Rbpositive.
Pheromone traps
Although the predominant wind in the area blows from west to east, pheromone traps placed
on the western vs. eastern there were no significant (P<0.05) differences in the number of
marked insect catches (F=0.15; P=0.70; df= 1,56), and this factor was pooled in the error term
of the main plot. Distance from the marked insect source (the rubidium-treated plot) did not
significantly affect the number of marked males caught in pheromone traps (P>0.05) (Figure
1). This means that S. nonagrioides males can disperse at least 400 m from where they
emerge. As within this distance mating can be assumed to occur at random, a male emerging
in one field has the same probability of mating with a female of the same field as with a
female emerged in a field within a radius of 400 m. Under such a perspective, refuges of nonBt fields can be placed at a distance of 400 m from Bt fields to delay resistance to B.
thuringiensis toxins in S. nonagrioides. Winds in this area do not seem to greatly affect the
direction of male flight to find a mate.
Light traps
There were significant differences in the number of marked males caught between 0 m and
the other two distances (200 and 400 m) at which light traps were placed (P<0.05) (Figure 2).
Only a total of 13 females were caught during the recording period; of those 2, 9, and 2 were
caught at 0, 200, and 400 m from the rubidium-treated plot. In view of these low numbers,
female catches were not statistically analysed.
The use of attractive trapping techniques may misestimate the number of adults moving
out of the field in which they emerge and may explain the differences obtained in this work.
Attraction of light traps is constrained by crop canopy, as the maize was as high as the light
traps, whereas pheromone traps may attract males from longer distances.
19
mean n.marked males/trap
A major interest in the estimation of S. nonagrioides dispersal capacity is its comparison
with the dispersal capacity of O. nubilalis. Both corn borers are present in the study area and
both are targeted by the Bt-maize varieties currently commercialised. A strategy of non-Bt
refuges must take into account the dispersal capacities of the two corn borers. Dispersion of
O. nubilalis males has been stated to be 1-3 km or even more (Showers, 1993, 2001) although
the authors did not study the frequency of such displacements or whether males move after
their first mate in the natal field. On the other hand it has been sawn that dispersal may be
affected by several factors like agronomic practices, host availability, climate or other regionspecific variables as well as their interactions (Hunt et al., 2001).
2,4
2,1
1,8
1,5
1,2
0,9
0,6
0,3
0
0
100
200
300
400
distance (m) from Rb-marked male source
mean n.marked males/trap
Figure 1. Mean number (+S.E.) of rubidium-marked males caught in pheromone traps placed
at increasing distances from the plot sprayed with rubidium. Each value is the average of 2
replications and 15 recording weeks. There were no significant (P>0.05) differences between
distances.
2,1
1,8
1,5
1,2
0,9
0,6
0,3
0
0
200
400
distance (m) from Rb-marked male source
Figure 2. Mean number (+S.E.) of rubidium-marked males caught in light traps placed at
increasing distances from the plot sprayed with rubidium. Each value is the average of 2
replications and 15 recording weeks. There were significant (P<0.05) differences between 0
and both 200 and 400 m.
20
In conclusion, the present results show that the use of rubidium-marked adults allows
one to estimate the dispersal capacity of S. nonagrioides adults. This technique offers several
advantages in comparison with other techniques used to study insect movement: (i) it is easy
to apply; (ii) it does not interfere with natural conditions as insects are marked naturally when
they feed on host plants that have been sprayed with RbCl; and (iii) it does not need a lot of
manpower in the marking and capturing phase. The main disadvantage is that processing and
analysis are rather long and expensive.
Acknowledgements
This research was partially funded by the Spanish National R&D Plan (AGF99-0782). Maria
Casamitjana carried out the sample processing and analyses. The company Sociedad Española
de Desarrollos Químicos (SEDQ) provided the S. nonagriodies pheromone blend to bait
pheromone traps.
References
Anonymous, December 15, 1998. Bt corn for control of European Corn Borer. Insect Pest
Management for Field and Forage Crops.
http://www.aces.uiuc.edu/ipm/field/iapmh/ipmffcbtecb.html
Anonymous, December 2, 2003. Opinion of the Scientific Committee of Plants on Btresistance monitoring. Opinion exposed in December 1999.
http://europa.eu.int/comm/food/fs/sc/out35_en.html
Gomez, K.A. & Gomez, A.A. 1984: Statistical procedures for agricultural research, 2nd ed.
John Wiley and Sons, New York, USA.
Gould, F. 1998: Sustainability of transgenic insecticidal cultivars: integrating pest genetics
and ecology. Annu. Rev. Entomol. 43: 701-726.
Hanway, J.J. 1966: Growth stages of corn. U.S. Department of Agriculture Technical Bulletin
No. 976.
Hunt, T.E., Higley, L.G., Witkowski, J.F., Young, L.J., & Hellmich, R.L. 2001: Dispersal of
adult European corn borer (Lepidoptera: Crambidae) within and proximal to irrigated and
non-irrigated corn. J. Econ. Entomol. 94: 1369-1377.
Larue, P. 1984: La Sésamie du maïs (Sesamia nonagrioides Lef.). Dégâts et actualisation de
lutte. Défense Vegetaux 227: 163-181.
López, C., Eizaguirre, M., & Albajes, R. 2003: Courtship and mating behaviour of the
Mediterranean corn borer, Sesamia nonagrioides (Lepidoptera: Noctuidae). Spanish J.
Agric. Res. 1: 43-51.
Prasifka, J.R., Keinz, K.M., & Sansone, C.G. 2001: Field testing Rb marking for quantifying
intercrop movement of predatory arthropods. Environ. Entomol. 30: 711-719.
Reynolds, D.R., Riley, J.R., Armes, N.J., Cooter, R.J. Tucker, M.R. & Colvin, J. 1997:
Techniques for quantifying insect migration. In: Methods in Ecological and Agricultural
Entomology, eds. Dent and Walton. CAB International, Wallingford, UK: 111-145.
Sans, A., Riba, M., Eizaguirre, M. & López, C. 1997: Electroantennogram, wind tunnel and
field responses of male Mediterranean Corn Borer, Sesamia nonagrioides, to several
blends of its sex pheromone components. Entomol. Exp. Appl. 82: 121-127.
Showers, W.B. 1993: Diversity and variation of European corn borer populations. In:
Evolution of Insect Pests, eds. Kim and McPheron. Wiley, N.Y.: 287-309.
21
Showers, W.B, Hellmich, R.L., Derrick-Robinson, M.E., & Hendrix, W.H. 2001: Aggregation
and dispersal behaviour of marked and released European corn borer (Lepidoptera:
Crambidae). Environ. Entomol. 30: 700-710.
Van Steenwyk, R.A. 1991: The use of elemental marking for insect dispersal and mating
competitiveness studies: from the laboratory to the field. Southw. Entomol. Suppl. 14:
15-23.
22
pp. 23-29
Analysis of web content of Theridion impressum L. Koch (Araneae:
Theridiidae) in BT (DK 440 BTY, MON 810, Cry1Ab) and isogenic
(DK 440) maize
Krisztina Árpás, Ferenc Tóth, József Kiss
Szent István University, Faculty of Agricultural and Environmental Sciences, Department of
Plant Protection, Páter K. u. 1., H-2103 Gödöllő, Hungary (E-mail: [email protected])
Abstract: Bt maize was grown on nearly 10,1 million hectares all over the world in 2002. We have
studied the effect of the Bt transgene on non-target organisms that live and feed within or near maize
fields via trophic interactions. Based on our previous maize field samplings, we decided for the
generalist predator Theridion impresssum as our test organism. Spider webs were collected from Bt
and isogenic maize plots in two successive years. Captured insects in the web were identified to
species level. The most frequent items were aphids, planthoppers and beetles. Although the number of
captured insects in web samples from isogenic maize was notably higher than those from Bt maize in
the case of almost each insect order, except for two samples (Sternorrhyncha, 2001; Neuroptera,
2002), the difference between Bt and isogenic samples was not significant. We could also show that
the easily recognizable „pocket-part” of the spider web properly represents the prey composition of the
complete web.
Key words: Bt maize, Theridion impressum, ecological impact assessment, prey composition
Introduction
Transgenic maize that expresses a gene from Bacillus thuringiensis subsp. kurstaki is widely
grown around the world. This maize expresses a Cry1Ab toxin that targets the European corn
borer (Ostrinia nubilalis Hbn.). The popularity of Bacillus thuringiensis toxins is due to their
high activity against target organisms combined with a minimum effect on non-target
organisms and a negligible human toxicity (Schnepf, 1995).
Until now only one study has reported direct toxic effects of Cry1Ab on a non-target
organisms, i.e. larvae of the green lacewing Chrysoperla carnea (Hilbeck et al., 1998).
However, this finding could not be confirmed in a later study (Romeis et al., 2004).
Phytophagous insects that feed on Bt-transgenic maize contain the Cry1Ab toxin in various
concentrations (Dutton et al., 2002). In a previous study we have examined the effect of Bttransgenic maize on the presence and biodiversity of non-target organisms (Kiss et al., 2002).
One of the aspects of the study was to find out whether the Cry1Ab toxin has any impact
through trophic interactions, i.e. on predators feeding on phytophagous insects that feed on
Bt-maize and on spiders that consume not only phytophagous species but also parasitoids and
predators. We therefore wanted to analyses the food range of a selected spider species that
lives in maize stands. First a suitable model had to be found, a species that frequently occurs
on the area, actively occupies it throughout the vegetation period, has a relatively large prey
range (feeds on target and non-target organisms, phytophagous, parasitoid and predator
species), has an easy to follow reproduction cycle and, finally, is little disturbed by our
investigation. According to our earlier studies, the most abundant spider families in maize
fields in the study area are wolf-spiders (Lycosidae), crab-spiders (Thomisidae), philodromid
23
24
spiders (Philodromidae), linyphiid spiders (Linyphiidae), araneid spiders (Araneidae) and
theridiid spiders (Theridiidae) (Tóth et al., unpublished). Since lycosid, thomisid and
philodromid spiders are not web-builders, and consume only a maximum of 1-2 prey items a
day (Nyffeler and Benz 1988), their feeding is difficult to monitor. Linyphiid and araneid
spiders do build a web, but their prey item is crushed apart during consumption and all
residues are thrown out of the web. Due to this special feeding habit, their prey range is
difficult to assess. In contrast, members of the family Theridiidae leave the cuticle of the prey
items intact and incorporate each and every prey finely into the web, allowing us to monitor
the prey species. The most abundant theridiid spider in our plot was Theridion impressum, a
species used earlier as a test organism in both laboratory and field studies (Pekár, 1999, 2002;
Tóth et al., 2002).
The objectives of our study were to determine the frequency of prey items on the basis of
the examination of the spider web content in both Bt and isogenic maize fields. The
hypothesis was that similar to the detection of small mammals by analyzing the sputum of
owls, comparative analysis of insect-assemblage of maize can be carried out by examining the
web content of a generalist and abundant spider species.
Materials and methods
Transgenic (DK 440 BTY, event MON 810) and isogenic (DK 440) maize plots were
sampled for spiders in 2001 and in 2002. The Bt maize was released under the approval of the
Ministry of Agriculture and Rural Development, Hungary. There were 6 replications,
resulting in a total of 12 complete block plots. Plot size was 30x30m (ca. 40 rows/plot, ca.
100 plants/row). In each plot, 500 individual maize plants were visually checked weekly from
28 June to 29 August in 2001 and from 27 June to 25 September in 2002. In the year 2001,
female spiders with their egg-sacks („cocoons”) and contents of the web were also collected,
whereas in 2002, only uninhabited webs and empty egg-sacks were collected. Using a
microscope, prey items found in the web were determined in both years. Insects that were
captured by the web and not eaten by the spider were carefully taken apart after visual
observation. In 2002 we noticed that there were two parts of a web; a „pocket”-part and the
other parts of the web, outside the pocket. This separation of web was necessary since the
borders of a web are hard to define, whereas the pocket has an obvious border. The pocket is
the part of a web, where egg-sacks are guarded by the spider. When created, pocket walls are
consisted of fine spider threads, but as the season wears away, dead bodies of captured prey
and fallen maize anthers start to dominate. The maize anthers simplify the recognition of the
web, which, being woven of thin threads, otherwise is hardly visible.
Statistical analysis of each insect order, acarids and spiders were carried out using a twosample t-test with Welch’s correction, on the basis of cumulative data of each plot. The order
Coleoptera has an emphasized importance, since being chewing feeders, phytophagous
beetles certainly uptake the Cry1Ab toxin.
Analysing the content of T. impressum webs revealed a large range of prey species, including
aphids, leaf beetles, thysanopterans, phytophagous and predator bugs, click beetles, flea
beetles, coccinellid beetles, lacewings, syrphid flies, ants, vespid wasps, parasitoid wasps,
bees, lepidopteran adults and larvae. Most of lepidopteran larvae were identified as cotton
bollworm Helicoverpa armigera Hbn., but larvae of the target organism of Bt maize, the
European corn borer Ostrinia nubilalis Hbn., were also present.
25
In the year 2001, the most frequent insects in the web samples were aphids
(Sternorrhyncha: Aphidina: Aphididae), bugs (Heteroptera), leafhoppers (Auchenorrhyncha)
and beetles (Coleoptera) (Figure 1). These groups of the most frequent insects were closely
followed by dipterans (Diptera), hymenopterans (Hymenoptera) and lacewings (Neuroptera).
Caught in low numbers were thysanopterans (Thysanoptera), spiders (Araneae) and
lepidopterans (Lepidoptera) in almost equal proportion. The difference between the number of
individuals of captured prey in Bt and isogenic maize stand was significant only in the case of
aphids (P = 0.024). We recorded however, that the average number of prey individuals was
generally higher in isogenic maize stands when compared to Bt-transgenic plots. Flea beetles
(Chrysomelidae: Halticinae) and coccinellid beetles (Coccinellidae) accounted for 75% and
5%, respectively of beetles total, whereas the remaining 20% comprised of various other
beetle families.
Average number of individuals / plot
280,0
240,0
200,0
other
20%
Coccinellidae
5%
Bt average
160,0
Iso average
120,0
Chrysomelidae
75%
80,0
40,0
Di
pt
er
Hy
a
m
en
op
te
ra
Ne
ur
op
te
Th
ra
ys
an
op
te
ra
Ar
an
ea
e
Le
pi
do
pt
er
a
St
er
no
rrh
yn
ch
a
He
te
ro
Au
pt
ch
er
en
a
or
rh
yn
ch
a
C
ole
op
te
ra
0,0
Figure 1. Prey-range of Theridion impressum according to its web-content, in Bt and isogenic
maize stands, in 6 replications (2001, Sóskút, average ± SD). Pie chart represents cumulative
data obtained from adding results of Bt and isogenic maize.
26
Average number of individuals /plot
280
240
other
10%
Coccinellidae
7%
200
160
Bt average
Iso average
120
Chrysomelidae
83%
80
40
et
e
H
Au
St
er
no
rrh
yn
ch
a
ch rop
en
te
r
or
rh a
yn
ch
a
Ac
C
a
ol
eo ri
Th
p
ys ter
a
a
H nop
ym
te
en r a
op
te
r
D a
ip
te
N
ra
eu
ro
pt
e
Ar ra
a
Le ne
a
pi
do e
C pter
ol
le a
m
bo
la
0
Figure 2. Prey-range of Theridion impressum according to its web-content, in Bt and isogenic
maize stands, in 6 replications (2002, Sóskút, average ± SD). Pie chart represents cumulative
data obtained from adding results of Bt and isogenic maize.
In the year 2002, the most commonly found prey species caught in T. impressum webs
again belonged to the aphids, bugs, acarids and leafhoppers (Figure 2). However, overall
catches were lower than in the year 2001. The difference between the number of individuals
of captured prey in web samples from Bt and isogenic maize stands was sigificant only in the
case of lacewings (P = 0.034). However, there was a tendency of isogenic maize having larger
number of individuals in each insect order. Beetles were again emphasized in this year:
chrysomelid beetles and coccinellid beetles accounted for 83% and 7%, respectively of
beetles total, whereas the remaining 10% comprised of various other beetle families.
The comparison of the number of individuals belonging to different insect orders within
the pocket part of the web revealed no significant difference between samples from Bt and
isogenic maize stands. The tendency of isogenic maize having larger number of individuals
was noticeable, though. Apparently, the source of data, i.e. pocket vs. complete web, had no
significant effect on relative frequencies (dominance) of insect orders when data from Bt and
isogenic maize were combined (Figure 3).
Sorted by guilds, prey items of webs revealed only one significant difference, in the case
of phytophagous insects in the year 2001 (P = 0.041, Figure 4), whereas in 2002, no
significant differences could be detected (P > 0.05) (Figure 5).
27
Average relative frequency
0,7
0,6
0,5
0,4
Pocket
0,3
Complete
0,2
0,1
St
er
no
rrh
Au H e y n
c h te c ha
en rop
te
o
Th rrh ra
y
ys nc
an h
a
C op t
ol er
H eo a
ym p
en t era
op
te
D ra
N ipt
eu er
Le ro p a
pi te r
do a
pt
A r e ra
C an
ol ea
le
m e
bo
la
0
Figure 3. Comparison of the content of the „pocket” and the complete web of Theridion
impressum in experimental maize plots (2002, Sóskút) (Combined data of Bt maize and
isogenic maize, average ± SD).
500
450
400
Bt
no. indiv. / plot
350
Is o g e n ic
300
250
200
150
100
50
0
H e rb ivo r
P re d a t o r
P a ra s it o id
O m n ivo r
N o ID
Figure 4. Arthropod guilds found in Theridion impressum webs in Bt and isogenic maize
(2001, Sóskút, average ± SD).
28
350
300
Bt
no. indiv. / plot
250
Is ogenic
200
150
100
50
0
Herbivor
P redator
P aras itoid
O m nivor
No ID
Figure 5. Arthropod guilds found in Theridion impressum webs in Bt and isogenic maize
(2002, Sóskút, average ± SD).
The most frequently captured insects in webs of T. impressum were aphids, bugs,
leafhoppers and beetles. This coincides with the experience of Pekár (2000), who analyzed the
T. impressum web content in various arable and horticultural stands.
A study carried out in Italy throughout 1997-98 found no significant difference between
the number of individuals of insects captured in Bt and isogenic maize (Lozzia, 1999). It is
remarkable however, that according to our analysis, the number of individuals found in web
samples of isogenic maize outnumber that of Bt maize plots in both years and with respect to
almost every order. Does this lead us to suppose that web-samples of Bt maize are differ from
isogenic samples? No, our present analysis does not allow us to do so, because although in
each year, there have been significant differences between insect orders (Sternorrhyncha in
2001; Neuroptera in 2002), these results were not confirmed in the other year. For the
complete analysis of this experience further studies (carrying on field experiments, deeper
taxonomic insight, laboratory consumption experiments) are necessary, some of which are
already being carried out.
Finally, our surveys aimed at improving a new, standardizable collection method. Results
prove that it is suitable to collect only the so-called pockets of the spider web, because it is
easy to recognize and it appropriately represents the species composition of prey items of the
complete web.
Acknowledgements
Authors would like to express warm thanks to dr. István Tóth (Plant Prot. Dept. Gödöllő
University), for his enormous help provided in the identification of web content. The study
was supported by application No. EU-5 QLK3-CT-2000-00547 and by the Bolyai János
29
Research Grant (Tóth F.). The content of this publication is the sole responsibility of its
publisher, and in no way represents the views of the Commission or its services.
References
Dutton, A., Klein, H., Romeis, J. & Bigler, F. 2002: Uptake of Bt-toxin by herbivores feeding
on transgenic maize and consequences for the predator Chrysoperla carnea. Ecol.
Entomol. 27: 441-447.
Hillbeck, A., Moar, W.J., Pusztai-Carey, M., Filippini, A. & Bigler, F. 1998: Toxicity of
Bacillus thuringiensis Cry1Ab toxin to the pretator Chrysoperla carnea (Neuroptera:
Chrysopidae). Environ. Entomol. 27: 1255-1263.
Kiss, J., Szentkirályi, F., Tóth, F., Edwards, C.R., Kádár, F., Kozma, E., Árpás, K., Perczel,
M. & és Dömötör, I. 2002: A Bt-kukorica hatása a nem-célszervezetek biodiverzitására
szabadföldön: célok, módszerek és első évi tapasztalatok. [Effect of Bt-maize on the
biodiversity of the non-target organisms in field conditions: aims, methods and
experiences of the first year.] In: Kuroli, G., Balázs, K. and Szemessy, Á. (Eds): 48. Plant
Protection Days, Budapest, Hungary, 06-07 Mar. Abstracts, p. 46.
Lozzia, G.C. 1999: Biodiversity and structure of ground beetle assemblages (Coleoptera
Carabidae) in Bt corn and its effects on non target insects. Boll. Zool. Agr. Bachic. Ser. II
31: 37-50.
Nyffeler, M. & Benz, G. 1988: Feeding ecology and predatory importance of wolf spiders
(Pardosa spp.) (Araneae, Lycosidae) in winter wheat fields. J. Appl. Ent. 106: 123-134.
Pekár, S. 1999: Foraging mode: a factor affecting the susceptibility of spiders (Araneae) to
insecticide applications. Pestic. Sci. 55: 1077-1082.
Pekár, S. 2000: Webs, diet, and fecundity of Theridion impressum (Araneae: Theridiidae).
European J. Entomol. 91: 47-50.
Pekár, S. 2002: Susceptibility of the spider Theridion impressum to 17 pesticides. J. Pest Sci.
75: 51-55.
Schnepf, H.E. 1995: Bacillus thuringiensis toxins: regulation, activities and structural
diversity. Curr. Opinion Biotech. 6: 305-312.
Romeis, J. Dutton, A. & Bigler, F. 2004: Bacillus thuringiensis toxin (Cry1Ab) has no direct
effect on larvae of the green lacewing Chrysoperla carnea (Stephens) (Neuroptera:
Chrysopidae). J. Insect Physiol. 50: 175-183.
Tóth, F., Árpás, K., Kiss, J. & és Zsák, V. 2002: Egy ízeltlábú csúcsragadozó, a Theridion
impressum L. Koch (Araneae: Theridiidae) kiválasztása modellfajként Bt kukorica nem
célszervezetekre gyakorolt hatásának vizsgálatához. [Choice of an arthropod super
predator Theridion impressum L. Koch (Araneae: Theridiidae) as a model species to
study the effect of Bt maize on non-target organisms.] In: Kuroli, G., Balázs, K. and
Szemessy, Á. (Eds): 48. Plant Protection Days, Budapest, Hungary, 06-07 Mar.
Abstracts: 145.
30
pp. 31-35
Impact of transgenic oilseed rape on soil arthropod assemblages
Giovanni Burgio1, Fabio Ramilli1, Maria Carola Fiore2, Francesco Cellini2
1
DiSTA, Alma Mater Studiorum Università di Bologna, viale Fanin 42,40127 Bologna, Italy
(E-mail: [email protected]); 2Metapontum Agrobios, S.S. Ionica 106 km 448.2 –
75010 - Metaponto (MT), Italy
Abstract: A field research was carried out in Metaponto (Southern Italy) to study the environmental
impact of oilseed Cry1Ac canola on soil arthropod assemblages, and to study the relative efficiency of
different sampling methods of soil fauna. The most abundant group sampled by pitfall traps was
Collembola, followed in decreasing order by Staphylinidae, Carabidae, Araneae, Oniscidea. Results
indicate that the total number of each group does not show any statistical difference between the
arthropod soil fauna in transgenic and isogenic line. Mainly Collembola and to a lesser extent
Oribatidae mites were found in soil samples. In soil arthropod fauna monitoring the sampling methods
are to be selected taking into account also the specific soil structure. These preliminary results seems
to demonstrate that transgenic canola does not have negative short-time effects on non-target soil
arthropods. The identification of Collembola, Carabidae, Oribatidae at species level is still in progress.
Key words: transgenic canola, soil arthropods, Collembola, environmental impact, sampling methods
Introduction
The insecticidal toxin produced by Bacillus thuringiensis subsp. kurstaki (Berliner) can
remain active in the soil, where it binds to clay and humic acids (Tapp et al., 1994; Crecchio
& Stotzky, 1998). Saxena et al. (1999) confirmed the presence of the toxin in the exudates
from Bt corn, demonstrating that it was active in an insecticidal bioassay using larvae of the
tobacco horn-worm, Manduca sexta (L.). Accinelli et al (2004) found similar results on
Ostrinia nubilalis (Hb.), using purified toxin from commercial formulation of Bt. A study of
the degradation of CryIAb within transgenic Bt corn tissue in the field (Zwahlen et al., 2003a)
suggests that extended pre- and post-commercial monitoring are necessary to assess the longterm impact of Bt toxin on soil organisms. On the other side Head et al. (2002) demonstrated
that the amount of CryIAc protein accumulated as a result of continuous use of transgenic Bt
cotton is extremely low and does not result in detectable biological activity.
Soil arthropods are considered useful indicators of environmental conditions and
ecosystem health (Paoletti, 1999). Below-ground groups such as springtails, nematodes and
earthworms should be included in risk assessment studies, but have received little attention
(Groot & Dicke, 2002) probably because of difficulties in the identification by nonspecialists. Data on effects of Bt crops on non-target soil microarthropods and nematodes are
available in Al-Deeb et al. (2003), Manachini & Lozzia (2003), Manachini et al. (2003). A
study on effects of transgenic Bt corn litter on the earthworm Lumbricus terrestris (L.) is
presented by Zwahlen et al. (2003b); the effects of transgenic herbicide-resistant soybean on
Collembola is shown by Bitzer et al. (2002). Data on the Carabidae complex in Bt corn in
Italy are available in Lozzia (1999). Büchs et al. (2003) suggested to take into account, in soil
monitoring, the saprophagous Diptera larvae, that are potentially affected by uptake of Bttoxin from plant parts and exudates of the roots.
31
32
The aims of our field research are to study the environmental impact of transgenic
oilseed canola on soil assemblages, with particular attention to Collembola, and to study the
relative efficiency of different sampling methods of soil arthropods.
Field trials were carried out in Metaponto (Southern Italy) to compare the soil arthropod fauna
on transgenic Cry1Ac-canola (cv. Westar) and its isogenic control line (permit No. B/IT/022,
EU Project BT-Bio.No.Ta. contract No. QLK3-CT-2000-00547). The field experiment
consisted in a randomized block design with three replications (200 m² per plot). Plots were
surrounded by a wheat strip (2 meters wide). Arthropod fauna was monitored every 14 days,
from March to May 2003, by a soil sampler (4 cores per plot) and pitfall traps (3 traps per
plot), for a total of 4 samplings. Vinegar (6% acetic acid) saturated with NaCl was used in
pitfall traps as preservative liquid. Arthropods collected by pitfall traps were stored in ethanol
(80%) for identification. The soil cores were put in a modified Tullgren-Berlese extractor and
the arthropods were collected in Falcon vials containing ethanol (80%).
Statistical analysis
Multivariate analysis of variance (MANOVA) and Principal component analysis (PCA) were
carried out to compare data collected by pitfall traps. A non-parametric two-factor ANOVA,
Friedman test, was employed to compare Collembola collected by soil cores between isogenic
and transgenic plots; a non-parametric approach was chosen because of the variance
heteroscedasticity of data.
The most abundant group sampled by pitfall traps was Collembola, followed in decreasing
order by Staphylinidae, Carabidae, Araneae, and Oniscidea (Figure 1). Results indicate that
the total number of each group does not show any statistical difference between the transgenic
and the isogenic line (MANOVA, P=0.71). Principal Component analysis was carried out on
the matrix of data collected by pitfall traps in each plot. The first two principal components
explained the 68.1% of the variance and the eigenvalues were 2.42 and 0.97, respectively. In
Figure 2 the arthropods sampled (Collembola, Staphylinidae, Carabidae, Araneae, Oniscidea)
were plotted against the first two principal components: the plot shows that Collembola and
Carabidae are grouped together. Figure 3 shows the high correlation (r = 0.76) between these
functional bioindicators. The most abundant Collembola species sampled by pitfall traps was
Entomobrya marginata (Tullberg); the most abundant Carabidae species was Harpalus
distinguendus (Duft.).
33
350
300
ISOGENIC
TRANSGENIC
Total/trap/plot
250
200
150
100
50
0
Collembola
Staphylinidae
Carabidae
Aranea
Oniscidea
Figure 1. Mean number of arthropods sampled per trap and plot; bars indicate standard
deviations.
1.2
Second principal component
1.0
ARANEAE
0.8
0.6
0.4
0.2
ONISCIDEA
0.0
CARABIDAE
STAPHYLINIDAE
-0.2
COLLEMBOLA
-0.4
-0.6
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
First principal component
Figure 2. Arthropods sampled in each trap plotted against the first two principal components.
Soil samples collected mainly Collembola and to a lesser extent Oribatidae mites. The
species identification is in progress; as with pitfall traps, the most abundant species sampled
was E. marginata. Statistical analysis indicates that the total number of Collembola collected
by soil samples does not show any statistical difference between the transgenic and the
isogenic line (non-parametric two-factor ANOVA, Friedman test, P=0.56). The mean number
of Collembola per trap and plot collected by soil sampler was higher in transgenic
(26.314.65) than isogenic (14.32.5) (Table 1), but the coefficient of variation (CV) of the
transgenic was very high (CV=55.70) in comparison with the isogenic (CV=17.5), showing a
considerable relative variability of data. This variability of data collected by soil samplers is
probably due to the low suitability of this sampling method in the very compacted field soil.
As evidenced from our results, the most appropriate sampling methods are to be selected in
the soil monitoring, taking into account also the specific soil structure.
34
80
Carabidae (Total/trap)
70
60
50
40
30
20
10
0
100
150
200
250
300
350
400
Regression
95% confid.
Collembola (Total/Trap)
Figure 3. Linear correlation between Collembola and Carabidae sampled by pitfall traps.
Table 1. Mean number of Collembola sampled by soil sampler per trap and plot ( standard
deviations).
Treatments
Isogenic
Transgenic
m  sd
14,3  2,5
26,3  14,6
In conclusion, the preliminary results of this research seems to demonstrate that
transgenic canola does not have negative short-time effects on non-target soil arthropod
sampled. The identification of Collembola and Oribatidae at species level is still in progress.
By agreement of Dipartimento Biologia Evolutiva, Università di Siena, a detailed faunistic
analysis will be carried out after the species identification. The aim is to compare the species
complex of Collembola and other soil indicators between transgenic vs isogenic canola. As
suggested by Al-Deeb et al. (2003), further studies on soil fauna after repeated growing of Bt
crops in the same field for several years are needed.
Acknowledgements
This Research was funded by the Ministero dell’Ambiente e Tutela del Territorio (Italy). We
thank Micaela Fabbri and Laura Depalo (DiSTA, University of Bologna, Italy), and Salvatore
Arpaia (ENEA, Trisaia, Italy).
References
Accinelli, C., Screpanti, C., Vicari, A. & Catizone, P. 2004: Influence of insecticidal toxins
from Bacillus thuringiensis subsp. kurstaki on the degradation of glyphosate and
glufosinate-ammonium in soil samples. Agric. Ecosyst. Environ.: in press
35
Al-Deeb, M.A., Wilde, G.E., Blair, J.M. & Todd, T.C. 2003: Effect of Bt corn for corn
rootworm control on nontarget soil microarthropods and nematodes. Environ. Entomol.
32: 859-865.
Bitzer, R.J., Buckelew, L.D. & Pedigo L.D. 2002: Effects of transgenic herbicide-resistant
soybean varieties and systems on surface-active springtails (Entognatha: Collembola).
Environ. Entomol. 31: 449-461.
Büchs, W., Prescher, S., Müller, A. & Larink, A. 2003: Effects of Bt-maize on terricole,
saprophagous diptera-larvae. Abstracts of “Biodiversity Implications of Genetically
Modified Plants”, Monte Verità Ascona, Switzerland, September 7-12, 2003: 10.
Crecchio, C. & Stotzky, G. 1998: Insecticidal activity and biodegradation of the toxin from
Bacillus thuringiensis subsp kurstaki bound to humic acids from soil. Soil Biol. Biochem.
30: 463-470.
Groot, A.T. & Dicke, M. 2002: Insect-resistant transgenic plants in a multi-trophic context. Plant
J. 31: 387-406.
Head, G., Surber, J.B., Watson, J.A., Martin, J.W. & Duan, J.J. 2002: No Detection of
Cry1Ac protein in soil after multiple years of transgenic Bt cotton (Bollgard) use.
Carabidae) in Bt corn and its effects on non target insects. Boll. Zool. Agr. Bachic. Ser. II,
31: 37-58.
Manachini, B. & Lozzia, G.C. 2003: Biodiversity and structure of nematofauna in Bt corn.
Abstracts of “Biodiversity Implications of Genetically Modified Plants”, Monte Verità
Ascona, Switzerland, September 7-12, 2003: 32.
Manachini, B., Fiore, M.C., Landi, S. & Arpaia, S. 2003: Nematode species assemblage in Btexpressing transgenic eggplants and their isogenic control. “Abstracts of Biodiversity
Implications of Genetically Modified Plants”, Monte Verità Ascona, Switzerland,
September 7-12, 2003: 31.
Paoletti, M.G. 1999 (Ed): Invertebrate Biodiversity as Bioindicators of Sustainable Landscapes.
Elsevier: 447 pp.
Saxena, D. & Flores, S. 1999: Insecticidal toxin in root exudates from Bt corn. Nature 402: 480.
Tapp, H., & Stotzky, G. 1995: Insecticidal activity of the toxins from Bacillus thuringiensis
subspecies kurstaki and tenebrionis adsorbed and bound on pure and soil clays. Appl.
Environ. Microbiol. 61: 1786-1790.
Tapp, H., Calamai, L. & Stotzky, G. 1994: Adsorption and binding of the insecticidal proteins
from Bacillus thuringiensis subsp kurstaki and subsp tenebrionis on clay-minerals. Soil
Biol. Biochem. 26: 663-679.
Zwahlen, C., Hilbeck, A., Gugerli, P. & Nentwig, W. 2003a: Degradation of the Cry1Ab protein
within transgenic Bacillus thuringiensis corn tissue in the field. Mol. Ecol. 12: 765-775.
Zwahlen, C., Hilbeck, A., Howald, R. & Nentwig, W. 2003b: Effects of transgenic Bt corn litter
on the earthworm Lumbricus terrestris. Mol. Ecol. 12: 1077-1086.
36
pp. 37-42
Potential effect of GNA-transgenic potatoes on adult aphid parasitoids
Aude Couty1, Jörg Romeis2
1
Entomology and Nematology Department, Rothamsted Research, Harpenden, AL5 2JQ,
United Kingdom (present address: Laboratoire de Biologie des Entomophages, Université
Picardie-Jules Verne, 33 rue St Leu, 80039 Amiens, France; E-mail: [email protected]); 2Agroscope FAL Reckenholz, Swiss Federal Research Station for Agroecology
and Agriculture, 8046 Zurich, Switzerland
Abstract: Transgenic potatoes transformed with a gene encoding snowdrop lectin (GNA) have been
shown to be partially resistant to aphids. Aphid parasitoids are important natural biocontrol agents of
aphids. They are endoparasitoids during their larval/pupal stages and free-living insects as adults and
could be directly or indirectly affected by GNA during both these stages. We present here a short
review on the potential impact of GNA on adult aphid parasitoids presenting different lifestyles. Data
are presented for the Aphidiidae Aphidius ervi and Aphidius colemani that are pro-ovigenic species
feeding mainly on honeydew and the Aphelinidae Aphelinus abdominalis, a syn-ovigenic species
feeding mainly on host haemolymph.
Key words: lectin, transgenic plant, risk assessment, honeydew, host-feeding
Introduction
Aphid parasitoids and coccinelids are important natural control agents of aphids and have
been used with success in biological control programmes. It is therefore essential to evaluate
the impact of aphid-resistant transgenic plants on these beneficial insects. Transgenic potatoes
expressing a gene encoding snowdrop lectin (Galanthus nivalis agglutinin, GNA) have been
shown to be partially resistant to several aphid species (Down et al., 1996; Gatehouse et al.,
1996; Sauvion, 1995) and their combined use with biocontrol agents (parasitoids/predators)
could be an interesting alternative to pesticide use. Several studies have investigated the
effects of GNA-expressing transgenic plants on aphid antagonists including the parasitoids
Aphidius ervi Haliday (Couty et al., 2001a) and Aphelinus abdominalis Dalman (Couty et al.,
2001b), and the predator Adalia bipunctata L. (Birch et al., 1999; Down et al., 2000). These
studies focused on exposure to GNA as a result of developing in or preying on GNA-fed
aphids. However, little attention has been paid to alternative routes through which these nontarget insects could be exposed to the transgene product. Adult aphid parasitoids could be
directly exposed to the transgene product when feeding upon contaminated food sources such
as plant secretions (nectar), aphid produced honeydew or aphid haemolymph. In this paper we
will focus on these routes of exposure for three parasitoid species belonging to two aphid
parasitoid families, i.e. two Apidiidae (A. ervi and Aphidius colemani Viereck) and one
Aphelinidae (A. abdominalis). These parasitoids differ in their mode of egg maturation and in
their feeding behaviour. They could thus be exposed to GNA through different routes. The
pro-ovigenic Aphidiidae feed mainly on honeydew (Stary, 1970; Hågvar & Hofsvang, 1991)
and it has been shown that aphids feeding on food sources containing GNA do excrete the
protein in their honeydew (Shi et al., 1994; Couty & Poppy, 2000). To address possible
effects of GNA intake through the consumption of GNA-contaminated sugar sources, this
lectin was fed dissolved in a sucrose solution at different concentrations to the two Aphidiidae
species. In contrast, adult Aphelinidae such as A. abdominalis are syn-ovigenic and mainly
37
38
feed upon their host haemolymph (Viggiani, 1984). Therefore, experiments have been
conducted to evaluate whether A. abdominalis is affected when feeding on hosts that were
reared on GNA containing artificial diet. Since binding of GNA to insect tissue is a
prerequisite for any potential effect on the insect physiology (Czapla, 1997), we studied
whether ingested GNA was binding to gut tissues of adult A. ervi and A. abdominalis. The
presence of GNA-binding proteins in their gut tissues was also investigated.
In this paper, we present a review on published results concerning the potential effects of
GNA on three aphid parasitoid species and we also include complementary unpublished
results.
Binding of GNA to adult parasitoid gut tissues
The fate of GNA ingested by adult parasitoids was investigated. Adult females of A. ervi and
A. abdominalis were fed GNA at 500 μg/ml sucrose solution (20%, w/v) for 3 days and
subsequently fed a pure sucrose solution for another 3 days. Subsamples were collected after
24h, 48h and 72h on the pure sucrose solution and proteins were extracted and run on SDSpage before performing western-blotting. GNA could be detected in parasitoids at the three
collecting dates indicating that GNA was remaining in parasitoid guts at least three days after
ingestion. Figure 1 shows an example of a western blot obtained for A. abdominalis females
fed GNA for 3 days and then fed pure sucrose solution for 24h and 48h. Similar blots were
obtained for A. ervi until 72h after the last feeding with GNA, suggesting that the lectin can
bind in a relatively stable way to gut receptors of both parasitoid species.
GNA standards
Con
24H
48H
50 ng
50 ng
Figure 1. Western-blot showing the presence of GNA in adult A. abdominalis females fed
GNA (500µg/ml) dissolved in a sucrose solution for 3 day and then fed pure sucrose solution
for 24h and 48h. No GNA can be found in females fed with sucrose solution alone (Con).
GNA standards are known GNA quantities (50ng) loaded into the SDS-page gel at the same
time as the parasitoid protein extracts.
In order to detect the presence of GNA-binding proteins in parasitoid gut tissues, guts of
adult A. ervi and A. abdominalis were dissected under a binocular and proteins were extracted
from isolated guts. Lectin-blot was performed on protein gut extracts in order to visualise
proteins capable of binding specifically to GNA (Figure 2). A protein extract of Arabidopsis
thaliana acted as a positive control as it was known that it would show two bands of high
39
molecular weight after lectin blot analysis. One band was identified in A. abdominalis
corresponding to a molecular weight close to 13 kDA. In A. ervi, one band can clearly be seen
in the low molecular weight region (less than 9kDa) and several bands can also be seen in the
region of high molecular weight (60 kDa), but complementary experiments are needed to
identify precisely the number of these bands.
However, the fact that GNA can bind to receptors in the parasitoid guts does not
necessarily mean that GNA will affect the physiology of parasitoid digestion and biology
parameters such as longevity, fecundity, progeny emergence and sex ratio.
61.3
36.4
24.7
13.1
9.3
KDa
A.th Ae
Aa
Aa
Ae
A.th
Figure 2. Lectin-blot showing GNA-binding proteins in A. abdominalis and A. ervi female gut
tissues. A: black arrows show proteins binding specifically to GNA. B: control blot showing
sample proteins binding non-specifically to GNA-antibody. A.th: Arabidopsis thaliana, Ae:
A. ervi, Aa: A. abdominalis.
Effect of GNA dissolved in sucrose solution on two Aphidiidae species, A. ervi and A.
colemani
Honeydew is an important energy source for parasitic wasps including Aphidiidae parasitoids
(Stary, 1970; Hågvar & Hofsvang, 1991). Shi et al. (1994) showed that Myzus persicae reared
on transgenic GNA-expressing tobacco plants excrete GNA in their honeydew. This was
verified for Macrosiphum euphorbiae fed artificial diet containing GNA. The aphids were fed
GNA (0.1%) for 6 days and then control artificial diet for 24h. Considering that the ingestion
rate of an aphid on artificial diet is around 25nl/h (Wright et al., 1985), it means that the latter
aphids would have ingested approximately 3.6µg of GNA within 6 days. By using westernblot analysis, it was estimated that the aphids contained only 200-300 ng of GNA. This
indicates that the aphids had excreted more than 90% of the ingested GNA in their honeydew.
The fact that the GNA concentration in the honeydew is close to that of the food source
(phloem/artificial diet) has earlier been reported for the rice brown planthopper, Nilaparvata
lugens (Hemiptera: Delphacidae), when feeding on artificial diet containing GNA (Powell et
al., 1998). GNA concentration in honeydew could also be increased under dry conditions
because of water evaporation. Adult parasitoids may therefore be directly exposed to the
lectin when feeding on honeydew. To investigate potential effects of this type of exposure on
40
A. ervi and A. colemani, adult females were fed sugar solutions with a range of GNA
concentrations potentially found in honeydew.
A. ervi females were fed sucrose solution containing 0%, 0.01 %, 0.05% or 0.1% GNA
(w/v). Female longevity decreased with increasing GNA concentration (Figure 3). An REML
(Restricted Maximum Likelyhood) analysis was performed and a Wald statistics was
produced (Wald test: χ2 = 17.6; df = 3; p < 0.01). The comparison of the 95% confidence
intervals showed that only the females fed 0.1% GNA had a significantly shorter longevity
than the controls. A. colemani females were fed sucrose solution containing 0%, 0.01 %,
0.05%, 0.1% or 1% GNA (w/v). There was also a concentration dependent effect observed.
Longevity of A. colemani females was significantly decreased at the two highest GNA
concentrations (0.1% and 1%) (Romeis et al., 2003). These two sets of experiments were run
independently in different laboratories and came to the same conclusion that GNA ingestion
affects longevity of adult Aphidiidae females.
However, studies on A. colemani revealed no effect of GNA on the daily fecundity
during the first week of adult life, progeny emergence and sex ratio (Romeis et al., 2003).
This result might be due to the fact that this species is pro-ovigenic and therefore does not
produce eggs during the adult stage.
10
days
8
*
6
4
2
0
0%
0,01%
0,05%
0,10%
GNA concentration
Figure 3. Mean longevity (± SE) of female A. ervi fed with sucrose solutions containing
various GNA concentrations (0%, 0.01%, 0.05%, 0.1%, w/v). * p<0.05%.
In order to see whether the effects on longevity could have been due to the fact that
parasitoids were deterred from feeding on the GNA-sucrose solutions, the gustatory response
of A. colemani females to the GNA solutions was also tested. Parasitoid food acceptance was
not affected when compared to a pure sucrose solution. This indicates that the effect observed
on parasitoid longevity can indeed be attributed to a direct effect of GNA on parasitoid
physiology and not to a lack of food intake (Romeis et al., 2003).
Effect of GNA-contaminated aphids on adults of the Aphelinidae species, A. abdominalis
For the syn-ovigenic Aphelinidae, aphid haemolymph constitutes a protein source necessary
for egg maturation. A modification of the haemolymph quality of GNA-intoxicated aphids,
either due the modification of amino-acids composition or to the presence of GNA, could
therefore lead to a lower parasitoid fecundity. In order to test for this potential effect, A.
41
abdominalis females were presented daily with GNA-fed aphids or control aphids for feeding
and oviposition. Control aphids were 6 days old and aphids kept on a diet containing GNA
(1000μg/ml) were 7-9 days old in order to be of similar size/instar as the controls. No effect
was observed on the longevity or the fecundity of female A. abdominalis that had developed
in GNA fed hosts. This is consistent with the fact that no traces of GNA were detected in the
haemolymph of the GNA-fed aphids. Adult A. abdominalis are thus not affected by the
consumption of GNA intoxicated aphids (Couty & Poppy, 2001).
Conclusions
Our work shows that adult aphid parasitoids can physiologically be affected by the ingestion
of GNA as they possess gut receptors binding specifically to this lectin. A negative effect of
GNA ingestion on parasitoid longevity has been shown for females of A. ervi and A.
colemani. However this work was restricted to the identification of a potential hazard and has
been undertaken under artificial laboratory conditions. Therefore complementary semi-field
or field studies using transgenic plants are needed to confirm these results. In order to fully
assess the potential risk of GNA-expressing crops on aphid parasitoid species, additional
information, such as on the rate at which the parasitoids are exposed to the transgene product,
is required. In the case of honeydew, this includes information on (1) the amount of GNA
contained in the honeydew of aphids colonising GNA-expressing plants in the field, and (2)
the quantity of honeydew ingested by parasitoids in the field (Romeis et al., 2003). While it
was found for A. abdominalis that the parasitoids were not affected by GNA-mediated
changes in the aphid haemolymph (Couty & Poppy, 2001), similar studies have not been
conducted for honeydew. Honeydew has often been reported to be a food source of lower
quality when compared to nectar or pure sucrose solution, possibly due to the presence of
aphid synthesized oligosaccharides or plant secondary metabolites (Wäckers, 2000).
Honeydew unsuitability could therefore “mask” the potential toxic effect of a transgene
product such as GNA. However, Hogervorst et al. (2003) showed that honeydews from three
potato-infesting aphids (Aulacorthum solani, Macrosiphum euphorbiae, Myzus persicae)
appeared highly suitable as a food source for A. ervi, making it likely that, in this instance, the
insecticidal properties of GNA would be fully expressed.
Studies on A. colemani suggest that ingestion of GNA by adult parasitoids might have
little effect on the parasitoids’ population dynamics since fecundity of this pro-ovigenic
species was not found to be affected by GNA ingestion during the first week of adult life
(Romeis et al., 2003). However, this might be very different for parasitoids with a synovigenic mode of egg maturation whose fecundity largely depends on the food uptake during
the adult stage.
References
Birch, A.N.E., Geoghegan, I.E., Majerus, M.E.N., McNicol, J.W., Hackett, C.A., Gatehouse,
A.M.R. & Gatehouse, J.A. 1999: Tri-trophic interactions involving pest aphids, predatory
2-spot ladybirds and transgenic potatoes expressing snowdrop lectin for aphid resistance.
Mol. Breed. 5: 75-83.
Couty A. & Poppy G.M. 2001: Does host-feeding on GNA-intoxicated aphids by Aphelinus
abdominalis affect their longevity/fecundity? Physiol. Entomol. 26: 287-293.
42
Couty A., Down R., Gatehouse A.M.R., Kaiser L., Pham-Delègue M-H. & G.M. Poppy.
2001a: Effects of GNA-containing artificial diet and GNA-expressing potatoes on the
developement of an aphid parasitoid, Aphidius ervi (Aphidiidae: Hymenoptera). J. Insect
Physiol. 47: 1357-1366.
Couty A., Poppy G.M., De la Viña G., Clark, S.J., Kaiser L. & Pham-Delègue M-H. 2001b:
Direct and indirect sublethal effects of snowdrop lectin Galanthus nivalis agglutinin
(GNA) on the development of an aphid parasitoid, Aphelinus abdominalis. J. Insect
Physiol. 47: 553-561.
Czapla T.H. 1997: Plant lectins as insect control proteins in transgenic plants. In: Advances in
Insect Control: The Role of Transgenic Plants, eds. Carozzi & Koziel. Taylor & Francis,
London: 123-138.
Down, R.E., Gatehouse, A.M.R., Hamilton, W.D.O. & Gatehouse, J.A. 1996: Snowdrop
lectin inhibits development and decreases fecundity of the glasshouse potato aphid
(Aulacorthum solani) when administered in vitro and via transgenic plants both in
laboratory and glasshouse trials. J. Insect Physiol. 42: 1035-1045.
Down, R.E., Ford, L., Woodhouse, S.D., Raemaekers, R.J.M., Leitch, B., Gatehouse, J.A. &
Gatehouse, A.M.R. 2000: Snowdrop lectin (GNA) has no acute toxic effects on a
beneficial insect predator, the 2-spot ladybird (Adalia bipunctata L.). J. Insect Physiol.
46: 379-391
Gatehouse, A.M.R., Down, R.E., Powell, K.S., Sauvion, N., Rahbe, Y., Newell, C.A.,
Merryweather, A., Hamilton, W.D.O. & Gatehouse, J.A. 1996: Transgenic potato plants
with enhanced resistance to the peach-potato aphid Myzus persicae. Entomol. Exp. Appl.
79: 295-307.
Hågvar, E.B. & Hofsvang, T. 1991: Aphid parasitoids (Hymenoptera, Aphidiidae): biology,
host selection and use in biological control. Biocontr. News Info. 12: 13-41.
Hogervorst, P.A.M., Romeis, J. & Wäckers, F.L. 2003: Suitability of honeydew from potato
infesting aphids as food source for Aphidius ervi. Proc. Exper. Appl. Entomol. 14: 87-90.
Powell, K.S., Spence, J., Bharathi, M., Gatehouse, J.A. & Gatehouse, A.M.R. 1998:
Immunohistochemical and developmental studies to elucidate the mechanism of action of
the snowdrop lectin on the rice brown planthopper, Nilaparvata lugens (Stal). J. Insect
Physiol. 44: 529-539.
Romeis, J., Babendreier, D. & Wäckers, F.L. 2003: Consumption of snowdrop lectin
(Galanthus nivalis agglutinin) causes direct effects on adult parasitic wasps. Oecologia
134: 528-536.
Sauvion, N. 1995: Effets et modes d’action de deux lectines a mannose sur le puceron du
pois, Acyrthosiphon pisum (Harris). PhD thesis, INSA Lyon.
Shi, Y., Wang, M.B., Powell, K.S., Vandamme, E., Hilder, V.A., Gatehouse, A.M.R., Boulter,
D. & Gatehouse, J.A. 1994: Use of the rice sucrose synthase-1 promoter to direct
phloem-specific expression of beta-glucuronidase and snowdrop lectin genes in
transgenic tobacco plants. J. Exp. Bot. 45: 623-631.
Stary, P. 1970: Biology of aphid parasitoids. Junk, The Hague.
Viggiani, G. 1984: Bionomics of the Aphelinidae. Annu. Rev. Entomol. 29: 257-276.
Wäckers, F.L. 2000: Do oligosaccharides reduce the suitability of honeydew for predators and
parasitoids? Oikos 90: 197-201.
Wright, J.P., Fisher, D.B. & Mittler, T.E. 1985. Measurement of feeding rates on artificial
diets using 3H-inulin. Entomol. Exp. Appl. 37: 9-11.
pp. 43-48
Monitoring of pest and beneficial insect populations in summer sown
Bt maize
Gavino Delrio1, Marcello Verdinelli2, Giuseppe Serra2
1
Dipartimento di Protezione delle Piante, sezione di Entomologia agraria, via E. De Nicola,
07100 Sassari, Italy (E-mail: [email protected]); 2Istituto per lo Studio degli Ecosistemi, CNR,
sezione Ecologia Applicata e Controllo Biologico, via E. De Nicola, 07100 Sassari, Italy
Abstract: During 2002, field studies were conducted in Sardinia in a summer sown cultivation of
transgenic maize, expressing the Cry1Ab toxin from Bacillus thuringiensis Berliner (Bt maize), in
order to evaluate the potential impact on beneficial insects. Crop resistance of the Bt maize hybrid
Compa and its non Bt isoline Dracma to corn borers, Ostrinia nubilalis (Hübner) and Sesamia
nonagrioides (Lefebvre), was observed. The attacks by the two pests were always lower in Bt maize
than in non-Bt maize plots: at the end of October, 5.9 and 19.3 larvae of corn borers per plant were
counted in Bt and non-Bt maize, respectively. The infestation of aphids, mainly Rhopalosiphum padi
(L.), was of the same level in both Bt and non-Bt plots. Similarly, no significant differences were
found between Bt and non-Bt maize for predator counts (mirids, anthocorids, syrphids, chrysopids,
coccinellids and arachnids) on sample plants. However the total number of coccinellids, mainly
Propylea quatuordecimpunctata (L.), captured by yellow traps was significantly lower in Bt plots
(P=0.05).
Key words: Bt maize, pests, beneficial insects
Introduction
The European corn borer Ostrinia nubilalis (Hübner) and the West African pink borer
Sesamia nonagrioides (Lefebvre) are the major economic pests of maize, Zea mays L., in
Mediterranean countries where extensive damages can occur with particular virulence on
maize crops sown in summer. Several measures have been introduced to control the corn
borers including cultural and biological control and the use of insecticides. Recently, maize
hybrids have been genetically altered to express the insecticidal δ-endotoxin from Bacillus
thuringiensis subsp. kurstaki Berliner (Btk) in order to control noxious lepidopterans (Koziel
et al., 1993). Some authors have underscored the economic benefits deriving from the use of
Bt maize hybrids (Burkness et al., 2002) whereas others asserted that Bt maize did not
necessarily translate into higher yields in some of the hybrids tested (Catangui & Berg, 2002).
Concerns have been raised regarding the possibility of a negative impact on beneficial insects.
Some studies, conducted both in the laboratory and in the field, dealt with the potential
adverse effects of Bt-transgenic maize (Pilcher et al., 1997; Orr & Landis, 1997; Lozzia et al.,
1998; Lozzia & Rigamonti, 1998; Hilbeck et al., 1998; Al Deeb et al., 2001; Wold et al.,
2001). The aim of this work was to compare the relative resistance to corn borers of a Bt
maize selection (event 176; Syngenta Seeds, Inc.) versus its non-transgenic isoline in a
summer sown cultivation and to study the effects of Bt maize on non-target insects.
43
44
Experimental design
The study was carried out in 2002 at the experimental farm of the University of Sassari
(Sardinia). Transgenic maize (Compa CB) and its non-Bt isoline (Dracma) were sown on June
21st and the two treatments were replicated in 10 plots following a randomised block design.
Each plot, 8 by 8 m, consisted of 11 rows and the rows were planted with 0.75 m spacing
between rows and 0.25 m spacing between plants within a row.
Sampling
Insect population monitoring was performed using yellow sticky traps and counting all insect
stages observed on plants randomly collected from each plot. For each plot, 3 plants were
collected on five dates (August 30th; September 13th, 20th, 27th; October 4th) and 6 plants at the
last two dates (October 14th and 24th). Predators and parasitoids were observed by externally
examining plants, whereas corn borer larvae were detected after the dissecting of stalks and
ears. The yellow sticky traps were staked in each plot (1 trap per plot) and replaced every
week from August 2nd to October 18th. All Arthropods, collected and/or trapped, were
identified at species or family level except for Thysanoptera, Chalcidoidea and Araneae.
Data analysis
For each taxonomic group, the number of insects captured or collected was compared by type
of maize (Bt and non-Bt) and sample date. Analysis was also conducted on data pooled over
the whole sampling period. Because variances were not homogeneous, a non-parametric
Mann-Whitney U test was used.
At harvest, the yields were estimated from plants of the two central rows of each plot.
Grain yield of hybrids was calculated as dry weight of 10 ears and the data were compared
using the Student’s t-test after log transformation of the data when variances were not
homogeneous. Data expressed as percentages were analysed after arc sine transformation.
Infestation of corn borers occurred in both maize hybrids but it was significantly higher for
non-Bt maize (Tab. 1; Fig. 1). O. nubilalis was the major pest and S. nonagrioides infestation
remained low throughout the growing season. The highest population density of maize borers
was detected on September 27th for Dracma (33.1 larvae/plant) and on October 4th for Compa
(8.6 larvae/plant). At the end of October, 5.9 and 19.3 larvae of corn borers per plant were
counted in Bt and non-Bt maize respectively (Fig. 1). The percentage of plants with truncated
stalk under ear insertion level, a direct measure of damage, was significantly higher in non-Bt
than in Bt plots (Tab. 2). But despite the fact that larval density was significantly different
between hybrids, no differences were found both in yield and percentage of damaged kernels.
No differences were found for other phytophagous insect populations (Cicadellidae,
Pentatomidae) between Bt and non-Bt maize (Tab. 1). The infestation of aphids, mainly
Rhopalosiphum padi (L.), was low and at the same level in the two hybrids.
Data concerning beneficial insect counts on sample plants, show that no differences
between hybrids are noticeable, except for tachinids (Tab. 1). The number of mature larvae
and pupae of the corn borer parasitoid Lydella thompsoni Herting was higher in non-Bt than
Bt plants, probably due to the higher attack of corn borers in the former. Syrphids constituted
the most relevant number of aphid predators, followed by coccinellids, of which Propylea
quatuordecimpunctata (L.) was the dominant species.
45
Table 1. Total number of arthropods collected on sample plants in Bt (Compa) and non-Bt
(Dracma) plots.
N*
Lepidoptera (larvae)
O. nubilalis
S. nonagrioides
Geometridae
Hemiptera (nymphs and adults)
Pentatomidae
Cicadellidae
Miridae
Anthocoridae
Diptera (larvae and pupae)
Syrphidae
Tachinidae
Neuroptera (larvae and adults)
C. carnea
Coleoptera
Staphylinidae (adults)
Coccinellidae (adults)
Coccinellinae
P. quatuordecimpunctata
A. bipunctata
C. septempunctata
Scymninae
Coccinellidae (immatures)
Nitidulidae (adults)
Araneae
Maize hybrids
Dracma Compa
U
P**
7
7
7
2894
172
27
719
46
3
6
11
8
0.02
0.09
0.03
7
7
7
7
15
38
70
31
19
24
61
36
25
23
20.5
24
1
0.90
0.65
1
7
7
104
32
155
4
29.5
5.5
0.56
0.01
7
10
10
25
1
7
118
29
13
0.16
7
7
7
7
7
7
7
31
8
1
11
41
310
181
27
20
2
4
49
86
152
21.5
15
21
15
22.5
17
18.5
0.74
0.18
0.59
0.22
0.85
0.34
0.48
*Number of sample dates
**Mann-Whitney U test
number of larvae per plant
35
Dracma (non-Bt)
Compa (Bt)
O. nubilalis
S. nonagrioides
30
25
20
15
10
5
0
30/8
13/9
20/9
27/9
sample date
Figure 1. Infestation of corn borer larvae.
4/10
14/10
24/10
46
Table 2. Damage by corn borers and yield at harvest ( x  SD ; n = 5).
Hybrid
No. of larvae/plant
Dracma (non-Bt)
Compa (Bt)
19.3±2.23 b
5.9±1.36 a
Truncated
plants (%)
17.9±5.03 b
5.0±1.36 a
No.
ears/sample/plot
39.8±0.84 a
43.0±6.0 a
Damaged
kernels (%)
18.0±5.72 a
14.0±2.24 a
Dry weight
of 10 ears (g)
2038.5±207.75 a
2299.2±143.27 a
Values in column followed by the same letter do not differ significantly at P < 0.05 (Student’s t-test)
Population density of Chrysoperla carnea (Stephens) larvae was generally very low in
both hybrids, averaging < 2% of the total number of predators collected. Spiders were the
most numerous generalist predators.
Statistical analysis of captures by yellow traps confirmed the results obtained by insect
counts on plants. No differences between hybrids were found except for the total numbers of
Coccinellidae and for P. quatuordecimpunctata adults (Tab. 3).
Table 3. Total number of adult insects caught by yellow traps in Bt (Compa) and non-Bt
(Dracma) plots.
N*
Thysanoptera
Hemiptera
Aphidoidea
Anthocoridae
Cicadellidae
Diptera
Syrphidae
Neuroptera
C. carnea
Hymenoptera
Chalcidoidea
Braconidae
Coleoptera
Coccinellidae (total)
Coccinellinae
P. quatuordecimpunctata
Psyllobora vigintiduopunctata (L.)
Scymninae
Other Coccinellidae
12
Maize hybrids
Dracma Compa
2726
2725
U
P**
73
0.98
12
12
12
538
68
313
674
61
305
82
69.5
74.5
0.58
0.91
0.91
12
65
61
74.5
0.91
12
6
19
94.5
0.09
12
12
12
12
7473
197
205
135
7946
223
178
71
74
80.5
54.5
106.5
0.93
0.64
0.32
0.049
12
12
12
12
52
1
71
11
27
6
33
5
106
65.5
96.5
95
0.048
0.55
0.16
0.12
*Number of sample dates
**Mann-Whitney U test
The cultivation of summer sown maize in the South of Italy is difficult due to the attacks
of the corn borers, at least where high populations of O. nubilalis and S. nonagrioides occur
(Spanu et al., 1988). In our experiment, Bt-transgenic maize showed a relevant resistance with
a significant reduction of damages, thus there is potential to virtually eliminate the need for
insecticidal sprays.
No significant differences were detected between Bt and non-Bt maize for immature
beneficial insects. However, the total number of coccinellids, mainly P.
quatuordecimpunctata, captured by yellow traps was significantly lower in Bt plots (Fig. 2).
Captures have probably been biased by various factors, depending on the sampling method,
47
and above all on the high mobility of adult coccinellids which could move from one plot to
another. In similar studies other authors did not find any statistical differences for beneficial
insects monitored in Bt and non-Bt field maize (Pilcher et al., 1997; Wold et al., 2001).
Dracma (non-Bt)
Compa (Bt)
number of adults/5 traps/week
25
20
15
10
5
0
2/8
16/8
30/8
13/9
27/9
11/10
sample date
Figure 2. Captures of adult Coccinellids by yellow traps.
Few differences in predator counts should suggest that Bt maize has no impact on these
non-target arthropods. While field observations can furnish information about population
densities, it is difficult to measure negative effects on particular species/functional groups due
to variation in the data and the lack of statistical power due to restrictions in the sample sizes.
In this study the density of some predators was relatively low and the associated variances
was high. To improve the accuracy of population density estimates, more extensive samplings
are desirable and it would be advisable to repeat the experiment over a longer period and to
conduct more detailed studies on the arthropods of interest in the laboratory prior to field
evaluations.
Acknowledgements
Research work supported by Ministero dell’Ambiente e della Tutela del Territorio.
Programme agreement CNR/Ministero dell’Ambiente “Genetically modified organisms and
biodiversity”.
References
Al Deeb, M.A., Wilde, G.E. & Higgins, R.A. 2001: No effect of Bacillus thuringiensis corn
and Bacillus thuringiensis on the predator Orius insidiosus (Hemiptera: Anthocoridae).
48
Burkness, E.C., Hutchison, W.D., Weinzierl R.A., Wedberg, J.L., Wold, S.J. & Shaw, J.T.
2002: Efficacy and risk efficiency of sweet corn hybrids expressing a Bacillus
thuringiensis toxin for Lepidopteran pest management in the Midwestern US. Crop
Prot. 21: 157-169.
Catangui, M.A. & Berg, R.K. 2002: Comparison of Bacillus thuringiensis corn hybrids and
insecticide-treated isolines exposed to bivoltine European corn borer (Lepidoptera:
Crambidae) in South Dakota. J. Econ. Entomol. 95: 155-166.
Hilbeck, A., Baumgartner, M., Fried, P.M. & Bigler, F. 1998: Effects of transgenic Bacillus
thuringiensis corn-fed prey on mortality and development time of immature Chrysoperla
carnea (Neuroptera: Chrysopidae). Environ. Entomol. 27: 480-487.
Koziel, M.G., Beland, G.L., Bowman, C., Carozzi, N.B., Crenshaw, R., Crossland, L.,
Dawson, J., Desai, N., Hill, M., Kadwell, S., Launis, K., Lewis, K., Maddox, D.,
McPherson, K., Meghji, M.R., Merlin, E., Rhodes, R., Warren, G.W., Wright, M. &
Evola, S.V. 1993: Field performance of elite transgenic maize plants expressing an
insecticidal protein derived from Bacillus thuringiensis. Bio/Technology 11: 194-200.
Lozzia, G.C., Furlanis, C., Manachini, B. & Rigamonti, I.E. 1998: Effects of Bt corn on
Rhopalosiphum padi L. (Rhynchota Aphididae) and on its predator Chrysoperla carnea
Stephen (Neuroptera Chrysopidae). Boll. Zool. Agr. Bachic. 30: 153-164.
Lozzia, G.C. & Rigamonti, I.E. 1998. Prime osservazioni sull’artropodofauna presente in
campi di mais transgenico. Proceedings of the Giornate Fitopatologiche, Scicli e
Ragusa, Italia 3-7 Maggio 1998: 223-228.
Orr, D.B. & Landis, D.L. 1997: Oviposition of European Corn Borer (Lepidoptera: Pyralidae)
and impact of natural enemy populations in transgenic versus isogenic corn. J. Econ.
Entomol. 90: 905-909.
Pilcher, C.D., Obrycki, J.J., Rice, M.E. & Lewis, L.C. 1997: Preimaginal development,
survival and field abundance of insect predators on transgenic Bacillus thuringiensis
corn. Environ. Entomol. 26: 446-454.
Spanu, A., Pruneddu, G., Delrio, G. & Mariani, G. 1989: Experiments on the control of
Sesamia nonagrioides and Ostrinia nubilalis on summer sown maize in Sardinia.
Proceedings of the XVth Symposium of the International Working Group on Ostrinia
(IOBC/IWGO), Varna, Bulgaria 11-17 September 1988 :129-136.
Wold, S.J., Burkness, E.C., Hutchison, W.D. & Venette, R.C. 2001: In-field monitoring of
beneficial insect populations in transgenic corn expressing a Bacillus thuringiensis toxin.
J. Entomol. Sci. 36: 177-187.
pp. 49-55
Assessing expression of Bt-toxin (Cry1Ab) in transgenic maize under
different environmental conditions
Anna Dutton, Marco D’Alessandro, Jörg Romeis, Franz Bigler
Agroscope FAL Reckenholz, Swiss Federal Research Station for Agroecology and
Agriculture, 8046 Zurich, Switzerland (E-mail: [email protected])
Abstract: At Agroscope FAL Reckenholz in Zurich, an open glasshouse (GH) system was established
that enables us to maintain transgenic plants under environmental factors close to a field situation and
simultaneously maintain biological containment comparable to a closed glasshouse. Using this system,
we assessed the expression of a Bt-toxin in transgenic maize (MON810) leaves at two developmental
stages and compared Bt-expression of plants grown in the open GH, the closed GH and in the field.
Immunological (ELISA) and insect bioassays using the European corn borer (Ostrinia nubilalis) were
performed with plant material collected in the three environments. Expression of the Bt-toxin was
shown to be higher in young plants as compared to that obtained form older plants. The environment
in which plants were kept also showed to have an influence on Bt-toxin expression. Results on the
biological activity of Bt-toxin in leaves, using O. nubilalis, were comparable to those obtained with
ELISA. The results confirmed that differences in Bt-toxin expression exist between young and old
plants, and that environmental conditions influence expression of Bt-toxin in transgenic maize.
Key words: ELISA, environmental effects, feeding bioassay, Ostrinia nubilalis, transgene expression,
Zea mays
Introduction
Stability of transgene expression is imperative in order for transgenic crops to become an
integral part in agricultural systems. However, numerous factors contribute to gene expression
and the production of insecticidal proteins in genetically modified (GM) plants. Both genetic
and environmental factors have been proposed to explain variation in protein expression. For
crops expressing genes coding for Bt-toxins, changes in protein production have shown to be
affected by factors including growing plants in different environments, nitrogen fertilization
and/or atmospheric CO2 concentrations (Greenplate, 1999; Coviella et al., 2000), among
others. Amount of Bt-toxin in transgenic plants can also differ according to plant age. An
example is that of Event-176 maize hybrids for which lower levels of Bt Cry1Ab were
observed in older plants when compared to young plants leading to Cry1Ab levels below
those required for controlling the second generation of the target pest, Ostrinia nubilalis
(Hübner) (Lepidoptera: Crambidae) (Onstad & Gould, 1998a; Archer et al., 2000). A similar
response has been shown for Bt-cotton expressing Cry1Ac (Olsen & Daly, 2000; Fitt et al.,
1998).
Different commercial varieties of Bt-maize have shown to express variable amounts of
toxin in different plant parts (Dutton et al., 2003). Studies to assess Bt-toxin concentrations
are primarily conducted using Western immunoblot (Fuchs et al., 1990) or immunological
bioassays (ELISA). Western immunoblots are labor intensive and both methods are based on
an antibody-antigen interaction which does not necessarily measure biological activity. Thus
insect feeding bioassays have been recommended to assess Bt-toxin activity (Sims &
Berberich, 1996; Kranthi & Kranthi, 2000) and also to assess the potential interactions of the
49
50
toxin with other secondary plant compounds that may alter the activity of the Bt-toxin (Olsen
& Daly, 2000).
During the development of new transgenic plants, experiments are required to be
performed in containment (e.g. climatic chamber or glasshouse), resulting in environmental
conditions different to those in the field. However, field experiments pose much work and in
Switzerland where field releases of GM plants for commercial and experimental purposes has
been prohibited, an open glasshouse (GH) system was established which allows to perform
experiments in a setup close to the field situation and simultaneously ensures biological
containment comparable to a closed glasshouse (Widmer et al., submitted). Using the Btmaize variety Monumental (MON810) that has been commercially available in the US since
1996, we here present results on the effects of growing transgenic plants in different
environments (open GH, close GH and field), on Bt-toxin expression in order to evaluate the
use of the open GH for risk assessment studies with transgenic plants.
Plants and insects
Transgenic maize (Zea mays L.) (event MON810, variety Monumental, Monsanto) expressing
a truncated synthetic version of the cry1Ab gene from Bacillus thuringiensis subsp. kurstaki
under the CaMV 35 promoter was used in all environments.
Ostrinia nubilalis eggs were obtained from INRA (France) and kept refrigerated at 4 °C.
For hatching and for experiments, insects were incubated in a climate chamber at 25 ± 1°C,
75 ± 5% RH, and 16:8h L : D photoperiod.
Growing environments
Experiments were conducted during the growing season of 2002 in a close GH and an open
GH at Agroscope FAL Reckenholz in Zurich and in the field in the German Rhine Valley
region of Freiburg.
a) Close glasshouse - Plants were grown individually in 10 l plastic pots in compost soil
(Staudenerde, Obi-Ter AG, Switzerland). The experimental design consisted of a randomised
block design (six replicates). Each replicate consisted of five plants grown in a single row.
The growing conditions were 28 ± 4°C, 70 ± 30 % RH. Plants were watered daily with the
fertilizer Superwux (N10:P10:K8) (the first 40 days at a concentration of 0.05% and
subsequently at 0.1%). Additionally, plants were fertilized twice (initially and after 40 days)
with 39 g/pot of Osmocote Exact (N15:P9:Mg3 and trace elements), and once with
Sequestren (10 grains/pot) (K8:N2.4:F) and Plüssfort Campo (N5:P5:26K:2Mg:0.2B:8S) (30
g/pot after 40 days).
b) Open glasshouse - The open glasshouse has been described by Widmer et al. (submitted)
and is to date the only semi-field system in which transgenic plants can be planted for
experimental purposes in Switzerland. The glasshouse is kept open during daytime and only
under sub-optimal weather conditions (rain, wind) and at night the roof and walls are
automatically closed. Plants were grown in 18 thermally insulated plastic containers (80 cm
wide, 120 cm long, and 80 cm high) in a loamy field soil. Each container was split into two
plots. The experimental design consisted of a randomised block design (six replicates). In
each plot nine plants were grown in three rows. Throughout the growing season the
temperature varied between 6.3°C and 40.1°C (average temperature 20.4°C) with RH ranging
between 17 and 100 %. Fertilizer was applied before planting, 139 g P, 15 g K and 14.5 g N
per container and, after 40 days with 19.2 g N.
51
c) Field - Maize plants were grown in a non-replicated field in the German Rhine Valley near
Freiburg. Transgenic plants were grown on 24 rows (80 cm row space) that were 200 m long.
The field was treated with Spectrum (1.5 l/ha, a. I. = Dimethenamid-P) and Stomp SC (2.5
l/ha, a. I. = Pendimethalin) as pre-emergence herbicides, and with Banvel 4 S (0.5 l/ha, a. I. =
Dicamba) as post-emergent herbicide. Plants were fertilized with 700 kg/ha 20N:10P:12K,
and 100 kg/ha urea. Six leaf samples at each plant growth stages were randomly collected
from different field locations.
Sample collection
Maize leaves were collected at two developmental stages from all environments: young plants
(V6: 5-6 full developed leaves) and at anthesis (VT: pollen shed). At stage V6, leaves five, six
and seven were collected. At stage VT, the 3rd and 4th leaf below the flowering spike were
collected. The mid-vein from all leaves was removed and leaves from each stage were pooled.
Pooled samples were lyophilized and subsequently ground to a homogenous powder and
stored at -20 °C. Six samples form each environment and each developmental stage were used
for analysis (see below).
Cry1Ab protein extraction and quantification
Cry1Ab protein levels were determined using a double sandwich ELISA kit (EnviroLogix
Inc., Portland, Maine). Cry1Ab standards at concentrations 0, 0.5, 2.5, and 5 ppb were used as
calibrators. Spectrophotometric measurements were conducted with a microtiter plate reader
(Dynatech MR 5000), Dynex Technologies, Ashford, UK) at 450 nm and data were analyzed
using the software package Biolinx 2.0 (Dynatech Laboratories Inc.) and Dynex Revelation G
3.2 (Dynex Technologies, Ashford, UK).
To quantify Bt-toxin in transgenic maize, 20  1 mg lyophilized leaf material was mixed
into 0.5 ml buffer. Samples were homogenized in 2 ml Epidorf tubes into which 500  5 mg
glass pearls (1 mm in diameter, B. Braun Biotech International GmbH, Nr. 31/4) were added.
A mixer mill was used (30 Hz, 2 x 1 min) to macerate and extract Bt-toxin from leaf material.
Subsequently, 0.5 ml extraction buffer was added to each sample and the extract was filtered
through a low protein-binding membrane (Acrodisc Syringe Filter 0.45 m, No. 4184). Bt+
leaf extracts were diluted 1:200 and 1:400 for the immunological assay.
Ostrinia nubilalis feeding bioassay
An agar based diet for O. nubilalis was prepared according to the Ivaldi-Sender generalpurpose diet described by Bathon et al. (1991). Water content of the standard diet was
adjusted in order to compensate for the added leaf extract solution. Leaf extracts were
prepared as described for the ELISA using 30 mg of leaf material that was pooled from the
six replicates from each treatment. After extraction and centrifugation (5 min, 13000 g) the
supernatants were diluted with 6.5 ml of extraction buffer. Five ml of each extract were
incorporated into 45 g of the warm (< 60°C) liquid diet. Diet was plated on Petri dishes and
allowed to cool at room temperature and subsequently cut into 0.5 g pieces. Diet pieces were
filled into bioassay tray cells (C-D International, Pittman, NJ) into which single 12h old
neonate O. nubilalis were placed. Cells were then sealed with a vented acetate cover (C-D
International, Pittman, NJ). A total of 32 larvae were tested per treatment. Larvae were
incubated for seven days, after which weight and mortality were recorded.
Statistical analysis
In order to statistically analyze differences in levels of Cry1Ab in maize leaves among
environments, only the replicated environmental systems (open and close greenhouse) were
52
considered. Since no block effect in neither of the environments was found, data were pooled
and a two-way analysis of variance with the dependent variable Cry1Ab concentration, and
the independent variables development stage (young and old plants) and growing
environment (open and close greenhouse) were conducted. Data were log transformed in
order to homogenize variances. Mean differences among treatments were determined using
Tukeys HSD-Test (P < 0.05). Data form the field could not be included in the statistical
analysis given that six samples were taken form a non-replicated field experimental unit.
Weight differences of O. nubilalis larvae when fed leaf material collected in the different
plant growth stages and different environmental conditions were analysed using MannWhitney U-test, with the sequential Bonferroni correction applied.
Results and Discussion
μg Cry1Ab g-1 dry weight + SE
Significant differences in Bt-toxin concentrations were found between young and old plants
(F1,20 = 354.9, P < 0.0001). Younger plants contained higher amounts of Cry1Ab toxin
compared to older plants (Figure 1). Similar findings have previously been reported from Btcotton (Olsen & Daly, 2000). The authors suggested that this may be due to less total protein
in leaves in older plants, what also seems to hold true for our study. Leaves from old
transgenic plants showed consistently lower nitrogen content (and lower Bt-toxin) as
compared to leaves form young plants (data not shown). Although in our case the difference
of Bt-toxin between old and young plants was only two fold compared to cotton where a fivefold toxin difference was observed, this reduction in Bt-toxin in older plants may have
important implications for resistance management of the target pest (Onstad & Gould, 1998).
80
60
Plant stage
young
old
a
40
b
20
c
c
0
Close GH
Open GH
Field
Figure 1. Mean (+SE) Cry1Ab toxin concentration in leaves form young and old Bt-maize
plants grown in three different environmental conditions (different letters represent
differences among treatments at P = 0.05; Tukeys HSD-Test on log transformed data).
Environmental conditions were also shown to influence Bt-toxin concentrations in transgenic
maize plants (F1,20 = 50.7, P < 0.0001) (Figure 1). These environmental effects were primarily
reflected by a strong difference between the Bt-toxin concentration found in leaf material
from young plants grown in the close GH (mean ± SE; 47.0 ± 1.8 μgCry1Ab/g dry leaf
53
weight), compared to leaf material collected form young plants grown in the open GH (26.0 ±
0.9 μgCry1Ab/g). This difference in Bt-toxin concentrations was not observed for older leaf
material collected form plants at the pollen shed stage. Although data from field samples
could not be included in the statistical analysis, the results from the field show comparable
levels of Bt-toxin to those observed in the close GH. A reason as to why young plants in the
open GH expressed the Cry1Ab toxin at a much lower level compared to that observed in the
other two environments is not known. However, a lower nitrogen content in leaves form
young plants in the open GH (35 mg/g leaf dry weight) was observed compared to nitrogen
content in leaves of plants grown in the closed GH (47 mg/g) and in the field (41 mg/g) (data
not shown). Although soil in the containers maintained in the open GH was fertilized priori to
planting, it may be that the amount of nitrogen supplied was below that provided to plants in
the closed GH and in the field. Plants in the closed GH were provided fertilizer continuously
(at every watering day) and plants in the field were fertilized according to a standard for
maize production regime. That nitrogen is correlated to Bt-toxin expression in transgenic
plants has been demonstrated by Coviella et al. (2000) for cotton plants and might also
explain our results. However, another reason for differences among environments in Bt-toxin
expression in maize could be due to the influence of temperature. To our knowledge no
studies have been conducted to assess how temperature may affect Bt-toxin expression in
plants, however studies with other transgenic plants have indicated that too high a temperature
can lower or silence transgene expression (Neumann et al., 1997). In our experiment higher
temperatures were measured in the open GH (> 40°C), where a lower Bt-toxin was measured,
compared to lower temperatures measured in the close GH (36°C) when a higher toxin
content was observed.
Weight of O. nubilalis larvae after seven days fed with the different leaf treatments are
shown in Figure 2. O. nubilalis larvae kept on the control artificial diet (leaf extract of control
maize plants) for seven days, weight an average (± SE) of 5.7 ± 0.4 mg. The weight reduction
of larvae, which were fed with the different treatments containing transgenic leaf extracts,
corresponded well with the amounts of Bt-toxin measured using the ELISA. A statistically
lower larval weight was obtained for individuals fed treatments containing leaf extracts from
young plants (extracts which were found to contain the highest Bt-toxin content), when
compared to weight of individuals fed leaf extracts from old plants. This was with the
exception of the open GH where incorporating extracts from either old or young leaf material
into the diet did not show a different effect on larval weight, reflecting that no differences in
Bt-toxin between old and young plant material exists. However, results form the ELISA show
differences between toxin levels in old and young plants (Figure 1). Since the trend between
Bt-toxin levels and larval weight corresponds, an explanation to this discrepancy between
using these two methods may be explained by a relatively high variability in larval weights.
54
Plant stage
larva weight (mg + SE)
6
young
old
5
4
3
b
b
b
b
2
a
1
a
0
Close
Open GH
Field
Control
Figure 2. Mean (+SE) weight of neonate Ostrinia nubilalis kept seven days on artificial diet
containing leaf extracts from young and old transgenic Bt-maize plants grown in different
environments, and control leaf extract (different letters represent differences among
treatments; Mann-Whitney U-test, Bonferroni corrected, α = 0.003).
larva weight (mg ±SE)
A 95% correlation between Cry1Ab extracted from leaves and O. nubilalis larval weight
(Figure 3) clearly demonstrates that either of these two methods can be used for assessing
toxin expression and activity in Bt-maize.
3
R 2 = 0.9483
y = -2.294x + 3.0765
F 1,201 = 213.3; P < 0.0001
2
1
0
0.5
0.6
0.7
0.8
0.9
1
1.1
µg Cry1Ab/ml leaf extract
Figure 3. Correlation between Cry1Ab-toxin extracted from leaf samples of young and old Btmaize plants grown in three different environments and mean weights (mg ± SE) of neonate
Ostrinia nubilalis kept seven days on a artificial diet containing the different extracts
(Extracts diluted by a factor 10 in the artificial diet).
55
Acknowledgements
We thank M. Waldburger for technical assistance in the open and close GH at Agroscope
FAL Reckenholz, K. Dannemann and V. Heitz (BaWü) for allowing us to use material form
the experimental field in Germany and Monsanto for providing us with the transgenic maize
seeds.
References
Archer, T.L., Schuster, G., Patrick, C., Cronholm, G., Bynum Jr., E.D. & Morrison, W.P.
2000: Whorl and stalk damage by European and Southwestern corn borers to four events
of Bacillus thuringiensis transgenic maize. Crop Prot. 19: 181-190.
Bathon, H., Singh, P. & Clare, G.K. 1991: Rearing methods In: Torticid Pests their Biology,
Natural Enemies and Control, eds. van der Geest and Evenhuis. Elsevier, Amsterdam,
The Netherlands: 283-293.
Coviella, C.E., Morgan, D.J.W. &. Trumble, J.T. 2000: Interactions of elevated CO2 and
nitrogen fertilization: Effects on production of Bacillus thuringiensis toxins in transgenic
plants. Environ. Microbiol. 29: 781-787.
Dutton, A., Romeis, J. & Bigler, F. 2003: Assessing the risks of insect resistant transgenic
plants on entomophagous arthropods: Bt-maize expressing Cry1Ab as a case study.
BioControl 48: 611-636.
Fitt, G.P. 1998: Efficacy of Ingard cotton – patterns and consequences. In: Proceedings, 9th
Australian Cotton Conference, Broadbeach, Queensland: 233-245.
Fuchs, R.L., MacIntosh, S.C., Dean, D.A., Greenplate, J.T, Perlak, F.J., Pershing, J.C.,
Marrone, P.G. & Fischhoff, D.A. 1990: Quantification of Bacillus thuringiensis insect
control protein as expressed in transgenic plants, in Analytical Chemistry of Bacillus
thuringiensis, ACS Symposion Series No. 432: 105-113.
Greenplate, J.T. 1999: Quantification of Bacillus thuringiensis insect control protein Cry1Ac
over time in Bollgard cotton fruit and terminals. J. Econ. Entomol. 92: 1377-1383.
Kranthi, K.R. & Kranthi, S. 2000: A sensitive bioassay for the detection of Cry1A toxin
expression in transgenic cotton. Biocontrol Sci. Techn. 10: 669-675.
Neumann, K., Dröge-Laser, W., Köhne, S. & Broer, I. 1997: Heat treatment results in a loss
of transgene-encoded activities in several tobacco lines. Plant Physiol. 115: 939-947.
Olsen, K.M. & Daly, J.C. 2000: Plant-toxin interactions in transgenic Bt cotton and their
effect on mortality of Helicoverpa armigera (Lepitoptera: Noctuidae). J. Econ. Entomol.
93: 1293-1299.
Onstad, D.W & Gould, F. 1998a: Do dynamics of crop maturation and herbivorous insect life
cycle influence the risk of adaptation to toxins in transgenic host plants? Environ.
Entomol. 27: 517-522.
Onstad, D. W. & Gould, F. 1998b: Modelling the dynamics of adaptation to transgenic maize
by European corn borer (Lepidoptera: Pyralidae). J. Econ. Entomol. 91: 585-593.
Sims, S.R. & Berberich, S.A. 1996: Bacillus thuringiensis Cry1A protein levels in raw and
processed seed of transgenic cotton: determination using insect bioassay and ELISA. J.
Econ. Entomol. 89: 247-251.
Widmer, F., Romeis, J., Schachermayr, G., Schlaich, T., Sautter, C., Winzeler, M. & Bigler,
F. submitted: Assessing ecological effects of wheat engineered with the KP4 gene to
mediate smut resistance.
56
pp. 57-63
Tracking Bt-toxin in transgenic maize to assess the risks on non-target
arthropods
Anna Dutton1, Lena Obrist1, Marco D’Alessandro1, Liliane Diener2, Martin Müller2,
Jörg Romeis1, Franz Bigler1
1
Agriculture, 8046 (E-mail: [email protected]); 2Swiss Federal Institute of
Technology (ETH), 8092, Zurich, Switzerland
Abstract: As an essential component in the risk assessment of insect resistant transgenic plants on
non-target arthropods it is important to determine the exposure of organisms to the insecticidal protein.
Exposure of piercing-sucking arthropods will depend on whether the cells and/or sap fed upon contain
the insecticidal protein. Using transgenic maize (Bt11) expressing the Bacillus thuringiensis (Bt) gene
that encodes for the Cry1Ab protein under the CaMV 35S promoter, localization of Bt-toxin in leaves
was performed with immuno-histochemical staining. The results demonstrate that Bt-toxin in maize
leaves was primarily located in mesophyll cells. In contrast, toxin was observed in minor amounts in
bundle sheath cells but no staining was observed in epidermis cells. In addition, quantification of Bttoxin in four piercing-sucking maize feeding arthropods, including an aphid (Rhopalosiphum padi), a
leafhopper (Zyginidia scutellaris), a thrips (Frankliniella tenuicornis) and a spider mite (Tetranychus
urticae) was performed. Arthropods were allowed to feed on transgenic plants and immunological
assays (ELISA) were conducted to determine the quantities of Bt-toxin contained in the different
species. ELISA results showed that the amount of toxin found in the different arthropods varied
substantially. Aphids contained only trace amounts of toxin. This is in agreement with previous
studies, which showed that this phloem feeding arthropod does not ingest the toxin, as it is not
transported in the phloem sap. Leafhoppers were found to contain an average of 0.20µg toxin/g fresh
weight, followed by thrips (0.91µg), whereas spider mites contained the highest amount (5.56µg).
Possible explanations as to why differences in the amount of toxin found in the various arthropods
might exist in relation to feeding site, feeding rate, excretion rate and toxin degradation are discussed.
Key words: ELISA, Frankliniella tenuicornis, immunolocalization, Rhopalosiphum padi, Tetranychus
urticae, transgenic plant, Zyginidia scutellaris
Introduction
With the increasing use of genetically modified maize expressing the insecticidal Bacillus
thuringiensis (Bt) gene (cry1Ab), which in 2002 covered almost 10 million ha of agricultural
land (James, 2002), the question on its environmental safety is being widely discussed
(Conner et al., 2003). One of the risks of using transgenic plants is that concerning the
unintended effects on non-target organisms including arthropods.
Since the risk that transgenic plants pose on non-target organisms is a function of
exposure to the insecticidal protein and the toxicity of the substance towards the specific
organism, one important component in the risk assessment is to determine exposure. Exposure
will depend on the feeding behaviour of an organism in question, together with the expression
of the toxin in the plant.
Different phytophagous arthropods are known to feed on different plant parts. Some
species have specialized their feeding on specific plant tissues or even on specific cells
57
58
(Bernays & Chapman, 1994). For example, piercing-sucking arthropods such as various
species of thrips are known to feed on both epidermis and mesophyll cells (Chilers & Anchor,
1995), spider mites and leafhoppers are predominantly mesophyll feeders (Van der Geest,
1985; Waloff, 1980) while aphids feed on phloem sap (Auclair, 1963). The feeding behaviour
of these organisms has extensively been studied since these species are often considered
important virus transmitters in crop plants. However, little information is available on the
ingestion of insecticidal protein by these arthropods when exposed to insect resistant
transgenic crops. Their contact and ingestion of transgene proteins may lead to exposure of
other organisms (i.e. natural enemies) at higher trophic levels in the food web.
Insecticidal expression in transgenic plants is influenced by various factors, including the
promoter gene used, place and number of genes inserted, and abiotic factors (Maessen, 1997).
For the commercialization of insect resistant transgenic plants, companies are presently
required to provide information on the expression of insecticidal proteins at the tissue level
(pollen, leaves, roots, seeds) (EPA, 2000). However, regulatory bodies do not request
companies to provide information on insecticidal protein expression at the cellular level.
Expression of Bt Cry1Ab toxin in different transgenic maize varieties has shown to differ
in the various plant tissues (for review see Dutton et al., 2003). Moreover using micro
capillary extraction of phloem sap and ELISA tests, Raps et al. (2001) demonstrated that the
aphid Rhopalosiphum padi (L.) (Hemiptera: Aphididae) did not ingest Bt-toxin when feeding
on two different maize varieties (Bt11 and event 176), as the toxin is not transported in the
phloem sap. In contrast, subsequent studies with spider mites showed that these arthropods
contain amounts of Bt-toxin approximating those in transgenic maize (Dutton et al., 2002).
Since the maize system is comprised of various piercing-sucking herbivores that feed at the
plant cellular level, knowledge on Bt-toxin expression at the cellular level and Bt-toxin
content in herbivores is an imperative requirement for an efficient risk assessment of
transgenic plants. Using immuno-histochemical staining Bt-toxin expression at the cellular
level was assessed for a commercial maize variety (Bt11). In addition ELISAs were
performed to quantify the amount of toxin in four predominant piercing sucking arthropods,
feeding on Bt-transgenic maize.
Plants
Transgenic maize (Zea mays L) plants (Bt11, N4640, Syngenta, Stein, Switzerland, formerly
Northup King) expressing a truncated synthetic version of the cry1Ab gene from Bacillus
thuringiensis var. kurstaki (HD-1) under the CaMV 35S promoter gene, and the
corresponding non-transformed nearest isoline were grown in the greenhouse (24 ± 4°C, 50 ±
20 % RH) in 1l pots. Plants were weekly fertilized with a 1g/l solution of 16N: 6P: 26K. All
experiments were conducted with five to six-week old plants, except for experiments
conducted with spider mites and aphids, for which plants were 10 to 14 weeks old.
Insects
Tetranychus urticae (Koch) (Acari: Tetranychidae) and R. padi were used from colonies kept
in the greenhouse (24 ± 4°C, 50 ± 20 % RH) at Agroscope FAL Reckenholz on either Btmaize or control plants. Frankliniella tenuicornis (Uzel) (Thysanoptera: Thripidae) were
collected in a maize field during the summer of 2001 at Agroscope FAL Reckenholz and kept
on either whole Bt-maize or whole control maize plants in a growth chamber (24 ± 1°C, 70 ±
10 % RH, 16:8 L:D). Zyginidia scutellaris (H.-S.) (Cicadellidae: Typhlocybinae) were
collected from a field during the summer of 2002. Insects were kept in cellophane bags (20.5
59
x 40 cm), which were clipped onto the 7th or 8th leaf of either a Bt-maize or control plant. All
insects were kept for at least one generation on plants, before being collected for ELISA tests.
ELISA with insects
Cry1Ab protein concentrations in the different herbivores were determined using a double
sandwich ELISA kit (EnviroLogix Inc. Portland, Maine, USA). Protein standards for
calibration were 0, 0.5, 2.5 and 5 ppb. Spectrophotometer measurements were conducted with
a microtiter plate reader (Dynatech MR 5000, Dynex Technologies, Ashford, UK) at 450 nm
and data were analyzed using the software package Biolinx 2.0 (Dynatech Laboratories Inc.)
and Dynex Revelation G 3.2 (Dynex Technologies). To quantify Bt-toxin in the four
herbivore species, different arthropod stages and amounts of material were utilized. Five
aphids samples (136 ± 10 mg/1ml buffer) from a mixture of all nymph stages and adults; 11
spider mites samples (1-2 mg/1ml buffer) from a mixture of all nymph stages and adults, six
thrips samples (2-4 mg/1ml buffer) from a mixture of larvae and adult, and three leaf hopper
samples (2-3 mg/ml buffer) from a mixture of larvae and adults, were used for ELISA tests.
Immunolocalization
The 8th leaf from six-week old maize plants was excised, 5mm long and 1mm broad leaf
pieces were cut and immediately placed in fixation solution (4% formaldehyde in 10mM
potassium hydro phosphate) for 30 min and left in vacuum for 15 min and subsequently left
for 120 min in fixation solution. Leaf material was placed in embedding solution (20%
sucrose in Tissue Tek solution) for 2 h before freezing at –23°C in Tissue Tek for sectioning.
Five μm leaf sections were cut with a cryotome at –23°C. The immuno-histochemical
detection of Bt was carried in a two-step indirect method as described by Harris (1994), using
a polyclonal primary antibody (1:500) in PBG buffer. Leaf sections were incubated 1h in 1st
antibody followed by five washes with buffer and subsequently 1h in secondary florescent
antibody (Fluorochrome conjugation/ dichlorotriazinyl amino fluorescein, ANAWA Trading).
To preserve samples on microscope slides, samples were treated with a Polyvinyl-alcoholglycerin solution. Labelling was visualized in a Photomicroscope (Zeiss Axiophot) equipped
with an epi-florescence filter (450-490/FT 510/LP 520) and a camera (Hamamatsu C5810).
Results
Tetranychus urticae kept on Bt-maize contained the highest amount of Bt-toxin (mean ± SE;
5.6  1.1 g/g fresh weight), followed by F. tenuicornis (0.9  0.1g) and Z. scutellaris (0.2 
0.03g) (Figure 1). In contrast, only trace amounts of toxin were detected in R. padi.
Immunolocalization of Bt in leaf tissue showed Bt-toxin predominately in the mesophyll
cells (Figure 2). Slightly detectable amounts of Bt-toxin were observed in bundle-sheath cells.
However, in epidermis cells Bt-toxin was not detected with the florescent staining. Expression
of toxin in other cells was not determined.
Discussion
Immunolocalization of Bt-toxin proved to be a useful tool for assessing the expression of Bttoxin at the cellular level. This method is often used for analyzing the expression of modified
genes in plants to understand gene regulation. To our knowledge this technique is seldom
used for localizing insecticidal proteins expressed in transgenic plants. The localization of one
other insecticidal protein has been that of snowdrop lectin (GNA) expressed in transgenic rice
60
n=6
Frankliniella tenuicornis
n = 11
Tetranychus urticae
n=3
Zyginidia scutellaris
n=5
Rhopalosiphum padi
0
2
4
6
8
μg Cry1Ab/g fresh weight + SE
Figure 1. Cry1Ab Bt-toxin concentrations (g/g fresh weight) measured in four piercing
sucking herbivore arthropods fed on transgenic Bt-maize.
EC
a
BS
MC
b
Figure 2. Immunohistochemical localization of Bt-toxin (Cry1Ab). Cross-section of leaf blade
of (a) Bt-maize (Bt11) and (b) control plant. EC: epidermis cell; BS: bundle sheath cell; MC:
mesophyll cell.
and tobacco (Shi et al., 1994; Rao et al., 1998). Although it is possible to use immunochemical analysis for detecting Bt-toxin at the cellular level in transgenic maize, this method
alone did not predict the content of Bt-toxin found in the different piercing-sucking herbivores
when feeding on transgenic plants.
The CaMV 35S is frequently used as promoter gene in plant biotechnology, as it is
believed to function constitutively in all plant parts and cells (Maessen, 1997). Based on the
presence of a florescent antibody our results show that expression of Bt-toxin under the
CaMV 35S promoter in transgenic maize occurs predominantly in mesophyll cells. This is in
contrast to results observed in other transgenic plants. For example, in rice and tobacco the glucuronidase (GUS) chimeric gene was mainly found to be expressed in leaf epidermis,
mesophyll and vascular bundle cells, when driven by the CaMV 35S promoter. However, in
61
petunia plants, GUS expression driven by the CAMV 35 promoter was not observed in the
mesophyll cells in some plant tissues (Battraw & Hall, 1990; Benfey & Chua, 1989). As
suggested by Battraw & Hall (1990), differences in staining results can be attributed partly to
the different methods employed for histochemical localization. These include fixation and
microtome sectioning, staining method (i.e. incubation time, antibodies used) and microscope
examination. However, differences in protein expression could be attributed to the position of
the inserted gene in the chromosome (Benfey & Chua, 1989), and/or differences in metabolic
activity with corresponding differences in rates of gene transcription and translation in
different plant species (Jefferson et al., 1987). These variations in results among plant species,
attributed to either differences in staining methods used or actual differences in protein
expression, reflect the need for establishing a standard method and for a proper molecular
characterization of transgenic plants.
In the present study, unanticipated results in the amounts of Bt-toxin measured in the
different herbivores were observed. Considering the expression of Bt-toxin at the cellular
level, it was expected that those organisms feeding primarily on mesophyll cells would
contain comparable amounts of Bt-toxin. However, for example T. urticae (Van der Geest,
1985) and Z. scutellaris (Marion-Poll et al., 1987) which are known to solely feed on
mesophyll cells were found to contain different amounts of toxin. In fact, T. urticae had 33
times higher amounts of toxin when compared to that measured in Z. scutellaris. Several
reasons could be attributed to this result. For example, differences in feeding behaviour
between the two species, such as amount of plant material ingested in a given time period.
Differences in metabolic processes of the toxin in the two herbivores (i.e. excretion and/or
digestion of the toxin) may have also contributed to the differences observed. For two
herbivore species it has been reported that the Bt-toxin content varied among the different
developmental stages, possibly attributed to differences in the feeding behaviour. Obrist et al.
(submitted) found that pre-pupae, pupae and newly emerged adults of F. tenuicornis did not
contain Bt-toxin when kept on transgenic maize as compared to larvae which were found to
contain an average (± SE) of 1.38 ± 0.03 g/g fresh weight. Differences in Bt-toxin content in
different developmental stages of another piercing-sucking herbivore, Athalia rosae (L.)
(Hymenoptera: Tenthredinidae), on oilseed rape expressing the Bt Cry1Ac protein were also
observed. While no toxin was detected in eonymphs, pupae and adults, A. rosae larvae were
found to contain 0.4 ± 0.4 or 1.3 ± 0.4 g/g fresh weight according to the transgenic variety
on which they were reared (Howald et al., 2003). The variable amounts of toxin detected in
the faeces of these two herbivores (Obrist et al., submitted; Howald et al., 2003), in F.
tenuicornis (32.01 ± 14.49 g) and in A. rosae (1.1 ± 0.2 g), also demonstrate that toxin
excretion may differ among species. Degradation of Bt-toxin may also differ among
arthropods, but at present little is known on this aspect. In a recent study, Romeis et al. (2004)
reported that the Bt-toxin could not be detected in larvae of the chrysopid predator
Chrysoperla carnea within six days after having ingested the toxin. Since different arthropods
have different ways to process toxic substances (behavioral and physiological) it would be
relevant to have a better understanding on the ingestion, excretion and degradation of
insecticidal proteins by organisms at the different trophic levels in order to perform a
predictable assessment of the risks that transgenic plants may pose on non-target arthropods.
Acknowledgments
We would like to thank D. H. Dean (University of Iowa) for providing us for the Cry1Ab
polyclonal antibody, G. Moritz (Martin-Luther-University of Halle-Wittenberg, Germany)
and R. Mühlethaler (Uni. Basel, Switzerland) for the identification of Frankliniella
62
tenuicornis and Zyginidia scutellaris, respectively. We also thank H. Klein and M.
Waldburger for technical assistance.
References
Auclair, J.L. 1963: Aphids feeding and nutrition. Annu. Rev. Entomol. 8: 439-490.
Battraw, M.J & Hall, T.C. 1990: Histochemical analysis of CaMV 35S promoter-glucuronidase gene expression in transgenic rice plants. Plant Mol. Biol. 15: 527-538.
Benfey, P.N. & Chua, N-H. 1989: Regulated genes in transgenic plants. Science 244: 174181.
Bernays, E.A. & Chapman, R.F. 1994: Patterns of host-plant use. In: Host-Plant Selection of
Phytophagous Insects. Chapman and Hall GmbH, New York: 4-13.
Chilers, C.C. & Achor, D.S. 1995: Thrips feeding and oviposition injuries to economic plants,
subsequent damage and host responses to infestation. In: Thrips Biology and
Management, eds. Parker, Skinner & Lewis. Plenum Press, London, U.K.: 31-51.
Conner, A.J., Glare, T.R. & Nap J.-P. 2003: The release of genetically modified crops into the
environment. Part II: Overview of ecological risk assessment. Plant J. 33: 19-46.
Dutton A., Romeis, J. & Bigler, F. 2003: Assessing the risks of insect resistant transgenic
BioControl 48: 611-636.
Entomol. 27: 441-447.
EPA (Enviromental Protection Agency) 2000: Bt plant pesticides biopesticides registration
action
document.
II
Science
assessment:
Product
Characterization.
http://www.epa.gov/scipoly sap /2000/october/brad2_scienceassessment.pdf
Harris, N. 1994: Immunocytochemistry for light and electro microscopy. In: Plant Cell
Biology: A Practical Approach, eds. Harris & Oparaka, Oxford University Press, Berlin:
157-176.
Howald, R. C., Zwahlen, W. & Nentwig 2003: Evaluation of Bt oilseed rape on the non-target
herbivore Athalia rosae. Entomol. Exp. Appl. 106: 87- 93.
James, C. 2002. Global status of commercialized transgenic crops: 2002. ISAAA Briefs, No.
27. ISAAA, Ithaca, NY.
Jefferson, R.A., Kavanagh, T.A. & Bevan, M.W. 1987: GUS fusions: -glucuronidase as a
sensitive and versatile gene fusion marker in higher plants. EMBO J. 6: 3901-3907.
Maessen, G.D.F. 1997: Genomic stability and stability of expression in genetically modified
plants. Acta. Bot. Neerl. 46: 3-24.
Marion-Poll, F. W. Della Giustina & Mauchamp, B. 1987: Changes of electric patterns
related to feeling in a mesophyll feeling leafhopper Zyginidial scutellaris. Entomol. Exp.
Appl. 43: 115-124.
Obrist, L. B., Klein, H., Dutton, A. & Bigler, F. submitted: Assessing the potential risks for
arthropod predators when feeding on thrips on Bt maize.
Rao, K.V., Rathore, K.S., Hodges, T.K., Fu, X., Stoger, E., Sudhakar, D., Williams, S.,
Christou, P., Bharathi, M., Bown, D.P., Powell, K.S., Spence, J., Gatehouse, A.M.R. &
Gatehouse, J.A. 1998: Expression of snowdrop lectin (GNA) in transgenic rice plants
confers resistance to rice brown plant hopper. Plant J. 15: 469-477.
63
Raps, A., Kehr, J., Gugerli, P., Moar, W.J., Bigler, F. & Hilbeck A. 2001: Immunological
analysis of phloem sap of Bacillus thuringiensis corn and of the nontarget herbivore
Rhopalosiphum padi (Homoptera: Aphidae) for the presence of Cry1Ab. Mol. Ecol. 10:
525-533.
Romeis, J., Dutton, A. & Bigler, F. 2004: Bacillus thuringiensis toxin (Cry1Ab) has no direct
effects on larvae in the green lacewing Chrysoperla carnea (Stephens) (Neuroptera:
Chrysopidae). J. Insect Physiol. 50: 175-183.
Shi, Y., Wang, M.B., Powell, K.S., van Damme, E., Hilder, V.A., Gatehouse, A.M.R.,
Boulter, D. & Gatehouse, J.A. 1994: Use of the rice synthase-1 promoter to direct
phloem-specific expression of glucuronidase and snowdrop lectin genes in transgenic
tobacco plants. J. Exp. Bot. 45: 623-631.
Van der Geest, L.P.S. 1985: Pathogens of spider mites In: Spider Mites: Their Biology,
Natural Enemies and Control (Part B), eds. Helle & Sabelis. Elsevier, Oxford: 171-184.
Waloff, N. 1980: Studies on grassland leafhoppers (Auchenorrhyncha, Homoptera) and their
natural enemies. Adv. Ecol. Res. 11: 81-215.
64
pp. 65-71
Comparison of the nodulation ability and abundance of aerobic
bacteria in the rhizosphere of transgenic and non-transgenic lines of
alfalfa
Natália Faragová1, Juraj Faragó1, Janka Gálová2
1
Research Institute of Plant Production, Section of Genetic and Breeding Research,
Department of Breeding Methods, Bratislavská 122, 921 68 Piešťany, Slovak Republic (Email: [email protected]); 2Constantine the Philosopher University, Faculty of Natural
Sciences, Dept. of Botany and Genetics, Trieda A. Hlinku 1, 949 01 Nitra, Slovak Republic
Abstract: Two genetically modified lines of alfalfa (Medicago sativa L.) containing the gene Ov from
Japanese quail, coding for a methionine-rich protein ovalbumin, were evaluated for nodulation ability
and densities of aerobic bacteria in the rhizosphere. The transgenic lines SE/22-11-1-1 and SE/22-134-1 were derived from a highly regenerable genotype Rg9/I-14-22 selected from cv. Lucia. Controls
were the non-transgenic isogenic line SE/22-GT2, and random genotypes from cv. Lucia. The
genetically modified line SE/22-13-4-1 showed 4.4 % higher number of nodules, 56.1 % higher
average fresh weight of roots and 26.7 % higher green matter dry weight in comparison with nontransgenic isogenic line SE/22-GT2. Statistically significant differences (p<0.05) between transgenic
and non-transgenic lines were detected in the number of stems per plant, root system-size, root mass
and content of total N in stems and leaves. On the selective media, there was a lower number of
ammonification bacteria and aerobic colony-forming microorganisms, whereas a higher number of
Azotobacter and cellulolytic bacteria isolated from the rhizosphere of transgenic plants in comparison
to non-transgenic ones. In spite of some differences in colony numbers in samples isolated from the
root rhizosphere of transgenic and non-transgenic alfalfa, we could not detect statistically significant
differences between individual lines.
Key words: alfalfa, genetically modified organisms, nodulation, rhizosphere, Sinorhizobium meliloti
Introduction
Genetic engineering of plants has created transgenic crops with increased resistance to pests,
herbicides, pathogens and environmental stress, enhanced qualitative and quantitative traits,
and the ability to produce industrial and pharmaceutical compounds (Fraley, 1992).
Because some recombinant microorganisms and transgenic plants carry genes and
produce compounds foreign to their environment, can grow and establish outside of their
natural habitat, and have enhanced survival, persistence and competitive capabilities, there are
several concerns about their environmental release and potential ecological effects (Seidler &
Levin, 1994). According to O'Connell et al. (1996) some genetically modified plants may
release other exudates from roots, than non-transgenic counterparts, and therefore may affect
the microbial community of the rhizosphere. Differences in the composition of root exudates
may also alter the soil ecosystem.
The composition of microbial communities in the rhizosphere is governed mainly by the
quality and quantity of carbon sources that are released as root exudates (Maloney et al.,
1997). Thus, an altered composition of root exudates may select a different community of
rhizosphere microorganisms. Even small modifications, as may exist between different
65
66
cultivars of the same plant species, can result in the selection of different microbial
communities in the rhizosphere (Schmalenberger & Tebbe, 2002).
Currently, a considerable effort is devoted worldwide to develop methods for assessment
of effects of transgenic plants on microbial communities (Di Giovanni et al., 1999; Donegan
et al., 1999). For example, Di Giovanni et al. (1999) assessed the environmental risk of
transgenic plants of alfalfa (Medicago sativa L.) grown in greenhouse conditions, on
rhizosphere bacterial communities. Their approach is currently used for determination,
whether field grown transgenic plants and their non-transgenic counterparts have unique
rhizosphere bacterial communities. Their study revealed, that transgenic plant genotype may
affect rhizosphere microorganisms and therefore additional studies are needed to determine
whether such changes may or may not adversely affect certain biological processes or the
success of future rotational crops.
The aim of our study was the assessment of nodulation ability, isolation and
determination of densities of selected groups of rhizosphere aerobic bacteria in two lines of
alfalfa transformed with the gene Ov from Japanese quail coding for a high-methionine
protein ovalbumin, in comparison with those in their isogenic parental line.
Plant material and propagation
In the experiment aimed at evaluation of nodule-forming ability and effect of plant genotype
on soil biota, two transgenic lines of alfalfa (Medicago sativa L.) encoded SE/22-11-1-1 and
SE/22-13-4-1, were used. The plants were genetically transformed as published elsewhere
(Farago et al., 2000) using a highly embryogenic line Rg 9/I-14-22 selected from the cultivar
Lucia (Faragó et al., 1997). The transgenic plants contained the gene Ov from Japanese quail
(Coturnix coturnix) encoding the protein ovalbumin (Mucha et al., 1991). The isogenic nontransgenic line Rg9/5-14-22 (SE/22-GT2) was used as a control. Randomly chosen plants
obtained from aseptically germinated seedlings of the cultivar Lucia were also included. The
isogenic parental and transgenic alfalfa lines were micropropagated in vitro on MS0.25
medium containing MS salts (Murashige & Skoog, 1962) and 0.25 mg.l-2 indolebutyric acid
(IBA) using the method of axillary branching. Rooted plantlets were acclimatized ex vitro for
4 weeks in a mixture of soil and perlite (3:1, v/v).
Evaluation of nodulation ability of transgenic and isogenic non-transgenic lines
After acclimatization period, the plants were transferred into plastic pots ( 180 mm) filled
with a 3:1 mix (v/v) of autoclaved sand and perlite. Plants were fertilized with mineral
fertilizer each 7th day using Jensen's solution without nitrogen (Somasegaran & Hoben, 1994).
A single starting dose of nitrogen (0.7 mM) was applied before day 14th after planting
(Faragová & Užík, 1999). The plants, at transfer to sand: perlite mix were inoculated with 3
different inoculation mixtures of nitrogen-fixing bacteria Sinorhizobium meliloti: 1.
suspension containing native strains M from the bacterial collection of Research Institute of
Plant Production (RIPP), Piešťany, Slovak Republic, 2. mix containing native strains K from
the bacterial collection of RIPP Piešťany, and 3. mix containing standard commercial strains
D from the collection of Rhizobia of the Gene Bank of Praha-Ruzyně, Czech Republic.
Inoculation suspension were adjusted to optical density equivalent to 108 CFU.ml-2 before
use. Plants inoculated with a sterile physiological solution were included as controls. At
harvest (12 weeks after the end of acclimatization period), the absolute shoot dry weight (mg),
number of shoots per plant, fresh weight of roots (mg), root system-size (score 1-9, 1 = very
poor, 9 = very large), number of nodules per plant, percent of active nodules according to
67
Rice et al. (1977), morphology of nodules (score: 1 = round-shaped, 2 = rectangular-like, 3 =
nodules in clusters) and total N in shoot dry weight (stems and leaves together) in %, were
evaluated.
Evaluation of the density of different groups of aerobic bacteria in the rhizosphere of
transgenic and isogenic non-transgenic plants
After acclimatization period, the plants were transferred into plastic pots ( 160 mm) filled
with pasteurized or non-pasteurized soil. Microbiological analysis of soil was carried out
before the planting and 12 weeks after the end of acclimatization period. At harvest, soil
samples were taken from the surroundings of alfalfa roots in 0.02 m depth. After the
homogenization and sieving through a 2 mm mesh, part of the samples was dried and the rest
prepared for microbiological analyses. The following analyses were performed: 1. the
presence of Azotobacter in the soil using the aggregate method on Ashby's agar, expressed as
percentage of fertile pieces of soil, 2. total number of aerobic microorganisms forming
colonies on Thornton's agar, expressed as the number of colony forming units (CFU) per 1 g
of soil dry weight, 3. density of cellulolytic bacteria using filter paper on the agar plate as a
carbon source and expressed as the number of CFU/g soil DW, 4. density of ammonification
bacteria cultured on MPA (meat-peptone agar) expressed as the number of CFU/g soil DW.
In both experiments, plants were grown in a growth chamber at relative air humidity
about 90 %, photoperiod of 16 h light/8 h dark and temperature of 23 °C. Pots (4 plants in
each) were arranged in a randomized complete design and each treatment consisted of three
replicates.
Evaluation of nodulation ability of transgenic and isogenic non-transgenic lines
The variability of all evaluated traits was significantly affected by both the plant genotype and
the inoculation suspension type (p<0.01). The highest shoot dry weight was observed in
genotypes of cultivar Lucia (310,82 mg, table 1). The parental non-transgenic line had the
lowest average shoot DW, 21.5 and 26.7 % lower than had both the transgenic lines,
respectively. The highest average number of shoots was observed in transgenic line SE/22-111-1. The transgenic plants had in average 11.8 and 16.8 % higher number of shoots,
respectively, than the overall mean. The line SE/22-11-1-1 was characterized with the highest
content of nitrogen in the green matter (stems plus leaves): 18.7 % higher in comparison to
parental line. Donegan et al. (1999) observed significantly lower shoot DW and higher N and
P contents in transgenic plants of alfalfa expressing lignin peroxidase when compared with
parental plants or transgenic plants containing a gene encoding amylase. In our experiment
the highest fresh weight of roots and the best developed root system were observed in
transgenic line SE/22-13-4-1 (Table 1).
The unique property of leguminous plants is their ability to utilize atmospheric N2
through a symbiotic association with rhizobial bacteria. Rhizobial structural nod genes are
expressed in response to biochemical signals from the plant. Expression of structural nod
genes results in the production of specific extracellular lipo-oligosacharide compounds that
elicit root-hair deformation, cortical cell division, and other responses in the susceptible
legume root. Due to the high economic and ecological importance of symbiotic N2 fixation
(National Academy of Sciences USA, 1994), it is very important to ascertain, whether the
ability of infection of root hairs by rhizobia and the formation of nodules are maintained
unchanged in genetically modified plants of alfalfa and to which extent the nodulation ability
is affected by genetic manipulation.
68
Plants of alfalfa inoculated with the mixture of D strains had the highest values of shoot
dry weight, root fresh weight and level of root development (Table 1). The highest increase of
shoot DW after inoculation with the mixture of D strains was observed in the line SE/22-111-1. Similarly, the mixture of nitrogen-fixing strains M increased the herbage DW by 8.1 %,
root fresh weight by 5.54 % and extent of root system development by 1.5 %, when compared
to the overall mean. Moreover, we also counted the highest number of shoots/plant after the
inoculation with a mixture of M strains. Overall positive effect of inoculation was recorded in
the transgenic line SE/22-11-1-1 (increase of shoot DW by 107.8 mg in comparison with a
non-inoculated control) and SE/22-GT2 (increase by 85.9 %). The highest number of nodules
formed on roots of plants derived from the cultivar Lucia (29.9 nodules). The transgenic line
SE/22-13-4-1 formed on their roots 4.4 % more nodules while that SE/22-11-1-1 18.7 % less
nodules when compared with the non-transgenic isogenic line SE/22-GT2 (Table 1). The
highest number of nodules with the highest activity (90-100 % active nodules relative to the
total number of nodules) was observed in plants inoculated with the mixture of M strains. The
activity of nodules of transgenic lines was about 4 % higher in comparison with nontransgenic line. In all the alfalfa lines the rectangular-like shape of nodules predominated.
In general, we observed statistically significant differences (p< 0.05) between transgenic
and non-transgenic alfalfa lines in the average number of shoots per plant, level of root
system development, root fresh weight and total N content. However, whether these
differences are the consequence of transgene expression itself, is not clear.
Table 1. Mean values for evaluated traits of transgenic and non-transgenic plants of alfalfa.
Factor
Number of
stems
Shoot dry
weight (mg)
Root fresh
weight (mg)
Number of
nodules
Lines
SE/22-11-1-1
3.05 b
269.57 ab
1469.25 a
17.67 a
SE/22-13-4-1
2.92 b
281.10 ab
2130.57 b
22.70 a
SE/22-GT2
2.05 a
221.85 a
1364.42 a
21.75 a
LUCIA
2.45 a
310.82 b
1627.37 a
29.90 b
SE(1)
0.15
24.32
160.80
2.08
Inoculation
Mixture M
2.97 b
292.80 ab
1739.32 ab
31.45 b
Mixture K
2.30 a
226.12 a
1452.97 a
23.62 b
Mixture D
2.92 b
317.65 b
1925.82 b
23.80 b
Non-inoculated
2.27 a
246.77 a
1473.50 a
13.15 a
(1)
SE
0.15
160.80
160.80
2.08
x
2.61
270.83
1647.90
23.00
SE(1)
0.07
12.16
80.40
1.04
(1) – standard error (ANOVA)
Means followed by the same letter are not significantly different at 5 % level.
N content
(%)
3.04 d
2.57 b
2.56 a
2.76 c
0.07
2.72 b
2.62 a
2.77 c
2.62 d
0.07
2.73
0.03
Evaluation of the density of different groups of aerobic bacteria in the rhizosphere of
transgenic and isogenic non-transgenic plants
The variability in density of all studied bacterial groups was statistically significantly (p<0.01;
ANOVA) affected by the substrate type used. When alfalfa plants were grown in autoclaved
69
soil, we observed higher densities of ammonifying (by 28 %) and cellulolytic bacteria (by 63
%) in soil samples in comparison with non-sterile substrate. This may be explained by the
competing ability of other microorganisms (e.g. actinomycetes) in non-sterile soil. The
presence only of Azotobacter was not found in the rhizosphere of plants grown in a
pasteurized soil, presumably because of the destruction of soil aggregates by autoclaving as
well as high sensitivity of Azotobacter to unfavorable physical conditions of environment
(Rosypal et al., 1981). The highest proportion of fertile pieces of soil (74 %) was recorded in
samples taken from the rhizosphere of transgenic line SE/22-11-1-1. This is 26 % more than
the proportion of fertile pieces obtained from the rhizosphere of non-transgenic isogenic line
(Table 2). When comparing the densities of ammonifying bacteria in soil samples taken
before or during the cultivation, respectively, in non-sterile substrate, it was shown an
increase of CFU/g DW by 75 % in transgenic and non-transgenic plants, and by 96 % in the
cultivar Lucia, respectively.
According to Donegan et al. (1999), growing transgenic plants of alfalfa containing the
gene for lignin peroxidase and -amylase production caused detectable changes in some
components of the soil ecosystem. There was a significantly higher population level of
cultivable aerobic, spore-forming and cellulose-utilizing bacteria.
In our case, the rhizosphere of genetically modified alfalfa plants with an inserted gene
from Japanese quail encoding ovalbumin was characterized by lower population level of
cellulolytic bacteria compared to the rhizosphere of control cultivar Lucia. The rhizosphere of
transgenic lines SE/22-11-1-1 and SE/22-13-4-1 contained by 9.16x102 and 16.59x102 more
CFU/g soil DW than the rhizosphere of non-transgenic line SE/22-GT2 regardless of the
substrate type. According to subjective visual evaluation, we observed in samples from the
surroundings of roots of transgenic lines more pale spots as a result of cellulose degradation,
while spots on filter paper of non transgenic plant origin were more intense.
Comparison of the densities of aerobic colony-forming bacteria revealed their decrease
by 24.11x103 and 9.38x103 CFU/g DW in SE/22-11-1-1 and SE/22-13-4-1, respectively,
relative to their densities at the beginning of cultivation. On the other hand, the densities of
aerobic colony forming bacteria increased by 55.66x103 and 23.66x103 CFU/g soil DW in
control isogenic line SE/22-GT2 and cultivar Lucia, respectively (Table 2).
Work dealing with risk assessment of genetically modified cultivars of oilseed rape
showed altered rhizosphere bacteria population structure in comparison with non-transgenic
ones (Siciliano & Germida, 1999). Other studies on transgenic potato suggested a significant
effect of method of genetic transformation on the structure of rhizosphere microbial
communities (Lukow et al., 2000; Lottman & Berg, 2001). These results suggest, that
unintended changes in rhizospheral communities of transgenic plants are possible, however,
the extent of these changes may be affected by the plant species as well as the type of genetic
modification (Schmalenberger & Tebbe, 2002).
In general, we observed lower densities of ammonifying bacteria and aerobic colonyforming microorganisms on the selective media and, contrarily, higher densities of
Azotobacter spp. and cellulolytic bacteria from the rhizosphere of genetically modified lines
of alfalfa in comparison with a non-transgenic isogenic line.
Despite of some differences in the numbers of colonies of bacteria isolated from soil
samples taken from the surroundings of roots of genetically modified and non-transgenic lines
of alfalfa, we could not detect statistically significant differences between the individual lines.
Therefore, to assess the relevance of our results in estimation of environmental risk of
transgenic plants of alfalfa additional studies are needed including the biological evaluation of
the population levels of other groups of microorganisms.
70
Table 2. Population densities of different groups of bacteria in the rhizosphere of transgenic
and non-transgenic alfalfa.
Lines
Ammonifying
bacteria
(CFU(1) x 103/g
soil DW)
Aerobic
colony-forming
microorganisms
(CFU x 103/g soil
DW)
Non-sterile substrate
SE/22-11-1-1
94.44
SE/22-13-4-1
105.37
SE/22GT2
103.40
Lucia
113.33
Sterile substrate
SE/22-11-1-1
212.04
SE/22-13-4-1
101.06
SE/22GT2
160.86
Lucia
69.89
Before cultivation
61.17
(1)
– colony forming units, DW: dry weight
Azotobacter
(%)
Cellulolytic
bacteria
(CFU x 102/g
soil DW)
110.00
124.73
189.77
157.77
74.0
52.0
48.0
66.0
35.55
73.11
59.09
92.22
104.81
74.46
83.69
47.31
134.11
0
0
0
0
58.0
112.04
81.91
79.34
146.23
71.76
References
Di Giovanni, G.D., Watrud, L.S., Seidler, R.J. & Widmer, F. 1999: Comparison of parental
and transgenic alfalfa rhizosphere bacterial communities using Biolog GN metabolic
fingerprinting and enterobacterial repetitive intergenic consensus sequence-PCR (ERICPCR). Microb. Ecol. 37: 129-139.
Donegan, K.K., Seidler, R.J., Doyle, J.D., Porteous, L.A., Di Giovanni, G.D. & Watrud, L.S.
1999: A field study with genetically engineered alfalfa inoculated with recombinant
Sinorhizobium meliloti: effects on the soil ecosystem. J. Appl. Ecol. 36: 920-936.
Faragó, J., Hauptvogel, P. & Kraic, J. 1997: Development of a breeding material of alfalfa
with high regeneration ability by recurent somatic embryogenesis. In: Seed Production of
Lucerne, eds. Chloupek and Simon. Academia Prague: 38-39.
Faragó, J., Kraic, P. & Hauptvogel, P. 2000: Development of highly regenerable germplasm
and genetic transformation of alfalfa. In: Use of Agriculturally Important Genes in
Biotechnology, ed. Hrazdina. IOS Press, Amsterdam-Berlin-Oxford-Tokyo-Washington:
117-123.
Faragová, N. & Užík, M. 1999: Hodnotenie efektivity a nodulačnej schopnosti vybraných
kmeňov hrčkotvorných baktérií Sinorhizobium meliloti pri lucerne v závislosti od
dusíkatej výživy. Zborník z 5. odborného seminára „Nové poznatky z genetiky
a šľachtenia poľnohospodárskych rastlín”, Piešťany: VÚRV, Dec. 8, 1999: 178-180.
Fraley, R. 1992: Sustaining the supply. Biol. Technol. 10: 40-43.
Lottmann, J. & Berg, G. 2001: Phenotypic and genotypic characterization of antagonistic
bacteria associated with roots of transgenic and non-transgenic potato plants. Microbiol.
Res. 56: 75-82.
71
Lukow, T., Dunfield, P.F. & Liesak, W. 2000: Use of the T-RFLP technique to assess spatial
and temporal changes in the bacterial community structure within an agricultural soil
planted with transgenic and non-transgenic potato plants. FEMS Microbiol. Ecol. 32:
241-247.
Maloney, P.E., Van Bruggen, A.H.C. & Hu, S. 1997: Bacterial community structure in
relation to carbon environments in lettuce and tomato rhizospheres and in bulk soil.
Microb. Ecol. 34: 109-117.
Mucha, J., Klaudiny, J., Klaudinyová, V., Hanes, J. & Šimúth, J. 1991: The sequence of
japanese quail ovalbumin cDNA. Nucleic Acids Res. 19: 5553.
Murashige, T. & Skoog, F.: A revised medium for rapid growth and bioassays with tobacco
tissue cultures. Physiol. Plant. 15: 473-497.
National Academy of Sciences, USA 1994: Biological nitrogen fixation. Available at:
http://www.nap.edu/readingroom/books/bnf/index.html (Verified: 21.11.2003)
O’Connell, K.P.; Goodman, R.M. & Handelsman, J. 1996: Engineering the rhizosphere:
expressing a bias. Trends in Biotechnol. 14: 83-88.
Rice, W.A., Penney, D.C. & Nyborg, M. 1977: Effects of soil acidity on rhizobia numbers,
nodulation and nitrogen fixation by alfalfa and red clover. Can. J. Soil Sci. 57: 197-203.
Rosypal, S., Hodak, K., Martinec, T. & Kocur, M. 1981: Obecná bakteriologie. SPN, Prague,
pp. 749.
Schmalenberger, A. & Tebbe, C.C. 2002: Bacterial community in the rhizosphere of
a transgenic, herbicide-resistant maize (Zea mays) and comparison to its non-transgenic
cultivar Bosphore. FEMS Microbiol. Ecol. 40: 29-37.
Seidler, R. & Levin, M. 1994: Potential ecological and nontarged effects of transgenic plant
gene products on agriculture, silviculture, and natural ecosystems: general introduction.
Mol. Ecol. 3: 1-3.
Siciliano, S.D. & Germida, J.J. 1999: Taxonomic diversity of bacteria associated with the
roots of field-grown transgenic Brassica napus cv. Quest, compared to the nontransgenic B. napus cv. Excel and B. napus cv. Parkland. FEMS Microbiol. Ecol. 29:
263-272.
Somasegaran, P. & Hoben, H.J. 1994: Handbook for Rhizobia. Springer-Verlag, New York,
Inc., 450 pp.
72
pp. 73-77
Research programme to monitor corn borer resistance to Bt-maize in
Spain
Gema P. Farinós, Marta De La Poza, Manuel González-Núñez1, Pedro HernándezCrespo, Félix Ortego, Pedro Castañera
Departamento de Biología de Plantas, Centro de Investigaciones Biológicas, CSIC, Ramiro
de Maeztu 9, Madrid 28040, Spain (E-mail: [email protected]); 1Present address:
Departamento de Protección Vegetal, Instituto Nacional de Investigación y Tecnología
Agraria y Alimentaria, Madrid 28040, Spain
Abstract: In Spain, about 22,000 hectares of transgenic maize expressing the Cry1Ab toxin from
Bacillus thuringiensis (Bt maize, event Bt 176, var. Compa CB, Syngenta) were planted for the first
time in 1998. A similar surface has been grown every year during the period 1999-2002, representing
around 5% of the total maize growing area of the country. The commercialisation of Bt-maize
provides a new tool for an effective control of two major lepidopteran pests in the Mediterranean
maize growing area, the Mediterranean corn borer (MCB), Sesamia nonagrioides (Lefebvre)
(Lepidoptera: Noctuidae), and the European corn borer (ECB), Ostrinia nubilalis (Hübner)
(Lepidoptera: Crambidae). A resistance monitoring programme funded by the Spanish Ministry of
Environment has been operating for the last five years to monitor target pest resistance to Bt-maize in
Spain. The main objectives of this programme being: a) to establish the baseline susceptibility to the
insecticidal protein in insects derived from non-transgenic fields and to measure the variability in
susceptibility between ECB and MCB; b) to detect changes over time in susceptibility by regular
monitoring of both species on Bt maize fields; and c) to forecast the development of resistance to the
insecticidal protein by laboratory selection. No shifts in susceptibility have been found for Spanish
populations of MCB and ECB after five years of Bt maize cultivation, but systematic field monitoring
needs to be continued.
Key words: Bacillus thuringiensis, Cry1Ab, European corn borer, Mediterranean corn borer,
transgenic maize, insecticide resistance, resistance management
Introduction
Genetically engineered maize plants expressing the Cry1Ab toxin from Bacillus thuringiensis
(Bt maize) are commercially grown in Spain since 1998. A surface of about 22,000 hectares
of Bt maize (event Bt176, var. Compa CB, Syngenta) has been grown every year during the
period 1998-2002, representing around 5% of the total maize growing area of Spain. Bt maize
provides an effective control of two key lepidopteran pests in most Spanish maize growing
regions, the Mediterranean corn borer (MCB), Sesamia nonagrioides (Lefebvre)
(Lepidoptera: Noctuidae) and the European corn borer (ECB), Ostrinia nubilalis (Hübner)
(Lepidoptera: Crambidae).
The commercial introduction of Bt maize in Spain prompted a monitoring research
project funded by the Spanish Ministry of Environment to enable a more science-based
discussion on the risks and benefits of Bt maize to the environment. Development of
resistance in target pests to Bt plants is considered a major risk for the success of this
powerful control tool. So far, field resistance to Bt maize has not been documented, but
laboratory selection assays have shown that ECB populations can respond quite rapidly to
73
74
intense selection pressure with Bt toxins (Chaufaux et al., 2001). The Spanish research
programme to monitor corn borer resistance is consistent with the issues proposed in the draft
protocol of the Expert Group of the European Commission (Draft Protocol for the monitoring
of the European Corn Borer and Mediterranean Corn Borer Resistance to Bt Maize,
Document XI/157/98, submitted by DGXI 29/04/98 [SCP/GMO/014]). Specifically, we are
addressing the following topics: a) to establish the baseline susceptibility to the insecticidal
protein; b) to detect changes over time in susceptibility by regular monitoring of both species
on Bt maize fields; and c) to forecast the development of resistance to the insecticidal protein
by laboratory selection.
Insect collection and culture
For baseline susceptibility of field populations of MCB and ECB and for monitoring changes in
the susceptibility, we collected every year about 300-500 larvae of MCB and/or ECB at 2-3
commercial maize fields in six Spanish maize growing regions (Figure 1). In addition, we
maintained laboratory cultures of MCB and ECB established in 1998 and 2000, respectively.
The larvae were always collected in fall in non-Bt maize (baseline studies) or Bt maize
(resistance monitoring) by dissecting corn stalks prior to harvesting. The conditions for larval
rearing, adult mating, and egg incubation have been described elsewhere (Farinós et al.,
2004).
Figure 1. Spanish geographical areas where MCB and ECB larvae were collected: Badajoz
(), Madrid (), Ebro (), Albacete (), Andalucía () and Galicia ().
Bioassays
The baseline susceptibility to Cry1Ab has been determined as described by González-Núñez
et al. (2000). The bioassays have been conducted using two different Cry1Ab toxin batches
provided by Syngenta. The lyophilized toxin was resuspended in 0.1% (v/v) Triton X-100 and
applied at 9 different concentrations on the surface of 2 ml of diet dispensed in plastic trays
(Bio-Ba-128, Color-Dec Italy). One neonate larva (<24 h old) was confined in each cell and a
total of 48 larvae were tested at each concentration plus a control. Mortality was assessed
after 7 days at 25  0.3 ºC, 70  5% r.h. and constant dark.
Laboratory selection
Selection assays were performed with laboratory populations of MCB and ECB established in
2000 with field-collected larvae. At F1 generation, the laboratory populations were split in two
strains, one submitted to selection (R-strain) and the other maintained as control (C-strain).
75
Larvae of the R-strains (a minimum of 1300 for MCB and 2500 for ECB) were exposed each
generation to increased concentrations of Cry1Ab during 5-7 days to obtain a mortality of
about 80%.
Coincident with the commercial release of Bt maize in Spain in 1998, we established the
baseline susceptibility to Cry1Ab (toxin batch 1) of several geographically different MCB and
ECB populations (González-Núñez et al., 2000). This study revealed small differences in
susceptibility among populations (Table 1), which can be attributed to natural variation, since
they were collected from non-transgenic fields with no records of Bt products being used.
The susceptibility of populations of MCB and ECB collected from Bt-maize fields was
determined in 1999 using two different toxin batches (Farinós et al., 2001). This study
revealed that the toxin batch 2 was more toxic than the toxin batch 1 when tested on the same
populations (Table 1). Thus, to avoid erroneous conclusions, we decided to restrict our
comparisons to those cases in which the same toxin batch was used.
Table 1. Susceptibility of Sesamia nonagrioides (MCB) and Ostrinia nubilalis (ECB) fieldcollected larvae from different Spanish maize growing areas to Cry1Ab.
LC50 (95% CL)a
Species
MCB
Madrid
Andalucia
Galicia
Ebro
Albacete
Badajoz
Lab-1998
ECB
Ebro
Madrid
Badajoz
Lab-2000
a
Toxin batch-1
1998 (NT b)
1999 (T b)
23 (16-30)
27 (16-39)
55 (19-115)
70 (56-87)
-
32 (19-45)
36 (13-87)
-
109 (77-162)
104 (82-140)
-
81 (51-122)
-
-
-
b
1999 (T )
5 (1-9)
3 (2-4)
23 (14-31)
15 (10-20)
-
4 (2-6)
-
Toxin batch-2
2000 (T b)
2001 (T b)
2002 (T b)
10 (5-15)
20 (15-25)
9 (6-12)*
18 (6-28)
8 (5-11)
19 (13-26)*
34 (27-42)
5 (4-6)
15 (9-23)
11 (7-16)
22 (16-32)
10 (7-12)
3 (2-5)
3 (2-4)
34 (28-41)
4 (3-6)
9 (6-12)*
6 (5-7)
3 (2-4)
Lethal concentrations (LC50 with a 95% confidence interval) expressed in ng Cry1Ab/cm2. Data from
González-Nuñez et al. (2000) and Farinós et al. (2001, 2004).
b
Larvae collected from transgenic (T) and non-transgenic (NT) maize
* Lethal concentrations are significantly different (P <0.05) if 95% confidence intervals of the lethal
concentration ratio (LCR) at the LC50 does not include 1 (Robertson & Preisler, 1992). LCR at the LC50
were calculated with respect to the susceptibility baseline (toxin batch-1) or the first sampling year on Btmaize fields (toxin batch-2).
76
As such, in continuation of our study, we assessed the susceptibility to Cry1Ab (toxin
batch 2) by regular monitoring on commercial Bt maize fields during the period 1999-2002
(Farinós et al., 2004). Our results indicate no increase in tolerance in the MCB populations
from Ebro, Albacete and Badajoz and in the ECB populations from Ebro and Badajoz after
five years of Bt maize cultivation in Spain (Table 1). We found a small (3.7-fold), but
significant, increase in tolerance in the MCB population sampled from Madrid in 2001 with
respect to the same population sampled in 1999. However, a gradual trend towards higher
levels of tolerance was not observed, since the susceptibility of the MCB population from
Madrid fluctuated and was again not significant when sampled in 2002. The susceptibility of
the laboratory strains remained relatively constant during the period 2000-2002, indicating no
changes in the activity of the toxin. Moreover, a significant decrease in tolerance for the
MCB population from Albacete between 1999 and 2000 and for the ECB population from
Ebro between 2000 and 2002 suggest that at least part of the variation in susceptibility found
from year to year reflect non-genetic variation. Similar findings have been obtained for ECB
field populations collected from Bt maize fields in U.S. (Siegfried et al., 2001).
Selection for resistance in ECB and MCB laboratory strains resulted in increased levels
of resistance to Cry1Ab after eight generations (Farinós et al., 2004). The level of resistance
achieved by the ECB population was 10 fold. This is lower than the 14-32 fold resistance
obtained by Chaufaux et al. (2001) with populations from U.S. and Europe, suggesting that
the potential to develop low to moderate levels of tolerance to Cry1Ab may be relatively
common among ECB populations. The level of tolerance obtained for MCB was also
moderate (21 fold), demonstrating that loss of susceptibility to Bt toxins can also develop in
MCB populations after eight generations. In Spain, both, ECB and MCB complete two to
three generations per year depending on the latitude of the area.
In conclusion, our monitoring programme has found no shifts in susceptibility for field
populations of MCB and ECB after five years of Bt maize cultivation in Spain. Nevertheless,
MCB and ECB populations have the potential to develop low to moderate levels of tolerance
to Cry1Ab by laboratory selection. Consequently, the Spanish National programme should be
continued to ensure early detection of resistance and the implementation of appropriate
management decisions in a timely manner.
Acknowledgements
We are grateful to Syngenta for providing the Cry1Ab toxin; to M. Eizaguirre, C. López
(Universidad de Lleida), I. Sánchez, C. Caballero, M. Díaz, C. Magaña, and F. Alvarez
(CSIC, CIB) for their time in collecting corn borers. This work was supported by "Ministerio
de Medio Ambiente, Spain".
References
Chaufaux, J., Segui, M., Swanson, J.J., Bourguet D. & Siegfried, B.D. 2001: Chronic
exposure of the European corn borer (Lepidoptera: Crambidae) to Cry1Ab Bacillus
thuringiensis toxin. J. Econ. Entomol. 94: 1564-1570.
Farinós, G.P., de la Poza, M., Ortego F. & Castañera, P. 2001: Monitoring corn borers
resistance to Bt-maize in Spain. In: Proceedings EU Workshop: Monitoring of
Environmental Impacts of Genetically Modified Plants, Berlin, November 2000. German
Federal Environmental Agency, eds. Miklau, Gaugitsch & Heissenberger: 114-118.
77
Farinós, G.P., de la Poza, M., Hernández-Crespo, P., Ortego, F. & Castañera, P. 2004:
Resistance monitoring of field populations of the corn borers Sesamia nonagrioides and
Ostrinia nubilalis after five years of Bt maize cultivation in Spain. Entomol. Exp. Appl.
110: 23-30.
González-Núñez, M., Ortego, F. & Castañera, P. 2000: Susceptibility of Spanish populations
of the corn borers Sesamia nonagrioides (Lepidoptera: Noctuidae) and Ostrinia nubilalis
(Lepidoptera: Crambidae) to a Bacillus thuringiensis endotoxin. J. Econ. Entomol. 93:
459-463.
Robertson J.L. & Preisler H.K. 1992: Pesticide Bioassays with Arthropods. CRC Press, Boca
Raton, FL.
Siegfried, B.D., Zoerb, A.C. & Spencer, T. 2001: Development of European corn borer larvae
on event 176 Bt corn: Influence on survival and fitness. Entomol. Exp. Appl. 100: 15-20.
78
pp. 79-84
Results of a 4-year plant survey and pitfall trapping in Bt maize and
conventional maize fields regarding the occurrence of selected
arthropod taxa
Bernd Freier1, Markus Schorling1, Michael Traugott2, Anita Juen2, Christa Volkmar3
1
Federal Biological Research Centre of Agriculture and Forestry, Institute for Integrated
Plant Protection, D-14532 Kleinmachnow, Germany (E-mail: [email protected]); 2Centre of
Mountain Agriculture, University Innsbruck, A-6020 Innsbruck, Austria; 3University Halle,
Institute Plant Breeding and Plant Protection, D-06108 Halle/Saale, Germany
Abstract: In the Oderbruch region, an outbreak area of the European corn borer (Ostrinia nubilalis),
we investigated the density and community composition of arthropods in Bacillus thuringiensis (Bt)
and conventional (CV) maize fields from 2000 to 2003. Aphids, thrips, bugs and aphid predators on
the plants were counted to obtain density estimates. Community structure of arthropod taxa on plants
taken from field as well as ground-dwelling carabids caught in pitfall traps was investigated by
ordination analysis. The density of most taxa varied considerably between the years. We detected a
few significant differences between Bt and CV maize, but no specific tendencies. However, the
ordination analyses revealed a small but significant influence of maize variety on the arthropod
community. The percentage of total variance explained by maize variety was only 2.7 % in ground
dwellers and 2.1 % in maize dwellers. The present data show that the influence of Bt maize on nontarget arthropods is small compared to field characteristics and yearly changing environmental
conditions.
Key words: Bt maize, Ostrinia nubilalis, arthropods, bio-indicator, side effects, community
composition
Introduction
When attempting to identify possible ecological side effects and to achieve general
surveillance of Bacillus thuringiensis (Bt) maize, field studies are needed in addition to
laboratory tests and plot trials (Wilhelm et al., 2003). Since 2000, farmers in the Oderbruch
region that is heavily affected by outbreaks of the European corn borer (Ostrinia nubilalis)
have been allowed to grow Cry1Ab expressing Bt maize on a limited area in co-operation
with scientific institutions collecting practical ecological background data from large field
cultures of Bt maize. Within this project, we started a long-term field study in Bt and nontransgenic maize fields in the year 2000. We intend to continue the investigations to
ultimately obtain a 10-year data pool.
The aims of this 4-year investigation were the following:
1) Collection of data on the structure of arthropod communities in Bt maize and conventional
maize fields.
2) Testing of different monitoring methods to obtain representative results on population
density and community structure in Bt maize and conventional maize fields.
3) Identification of particular phenomena and significant differences in occurrence of
arthropods indicating possible ecological side effects of Bt maize growing.
79
80
Together with the farmers, adjacent Bt and conventional (CV) maize fields were selected each
spring. One field of each variety was planted in 2000 and 2001, and two of each in 2002 and
2003.
Field sizes:
2000 – 8 ha Bt (Novelis), 3 ha CV (Coach),
2001 – 17 ha Bt (Novelis), 29 ha CV (Flavio),
2002 – 10 ha Bt (MEB 307-Bt), 11 ha CV (Lenz) (field pair 1, 2002-1); 10 ha Bt (MEB 307Bt), 48 ha CV (Lenz) (field pair 2, 2002-2),
2003 – 5 ha Bt (MEB 307 Bt), 5 ha CV (Monumental) (field pair 1, 2003-1); 8 ha Bt (MEB
307 Bt), 9 ha CV (Monumental) (field pair 2, 2003-2).
For ordination analysis data from four additional fields were incorporated in addition to the
above-mentioned fields for the year 2000: an isogenetic (IS, Nobilis) maize field in the
Oderbruch landscape, and an IS (Nobilis), CV (Birko) and a Bt (Novelis) maize field at an
other site near Halle/Saale.
A programme of standard ecological investigations as well as counting, harvesting of
whole plants and pitfall trapping were carried out in the study fields. Samples were taken
from the Bt and CV maize fields along central sampling lines that ran parallel to the border
between both fields.
Arthropod counting
Three plants were observed at 5 or 10 (2001) sampling sites at the end of maize flowering.
This method allows a limited determination of taxa visible to the naked eye and is suitable for
density comparisons.
Harvesting of whole plants with later determination of arthropods on frozen maize plants in
the laboratory
Three plants each were taken from 5 sampling sites at end of maize flowering and 4 weeks
later. In 2000 additional sampling was carried out in June. Because of the more precise
arthropod determination under the microscope, these data could be used for the analyses of
community structure.
Pitfall trapping with later determination of carabids and spiders species
Ten traps were left in place during 4 (2002) or 6 (2000, 2001) weeks after the beginning of
maize flowering. The results for 2003 are not yet available. Pitfall data were used for studying
community composition.
Only a few taxa occurred on the plants in numbers sufficient for a statistical analysis.
Therefore, the individual taxa had to be pooled in higher taxonomic units. To demonstrate the
occurrence of aphid predators as a community, we used a predator unit calculation system that
considers the different feeding potentials of various predators or predator groups (Freier et al.,
1998).
Statistical analyses
Wilcoxon 2-Sample-Test, run using SAS 8.1 software, was used to test for significant
differences in count-based densities between Bt and CV maize. Using CANOCO 4.5 (ter
Braak & Šmilauer, 2002) we investigated the relationship among the arthropod communities
and the environmental variables study site, year, sampling time and maize variety. Ground
and maize-dwelling arthropods were analysed separately. To prevent taxa caught in high
81
numbers from excessively influencing the ordinations, faunal counts were square-root
transformed. Taxa represented by less than 30 individuals were excluded from the analysis
(49 ground and 22 maize-dwelling arthropod taxa). Correspondence analysis (CA) was used
to illustrate the major patterns. For canonical correspondence analysis (CCA) year, field
location, sampling date and maize variety were used as environmental variables. Partial
ordination was used to determine the influence of the maize variety in relation to the other
environmental variables. To test the statistical significances Monte Carlo permutation tests
were calculated.
Results
Count-based densities
Aphids (in particular Rhopalosiphum padi and Metopolophium dirhodum): Considerable
density differences were observed among the years and between the two maize varieties.
Significantly more aphids were found in the conventional field 2003-1. However, the means
of the 6 fields were almost equivalent (Fig. 1).
Thrips (Franklinella tenuicornis and other species): The densities varied considerably among
the years and also between the varieties. Significant differences between Bt and CV maize
were observed in 2001 and 2003-1. However, no significant difference was found when
considering the averages of the 6 field studies (Fig. 2).
Individuals per plant
200
160
n = 5 (2000, 2002, 2003)
n = 10 (2001)
p=0.00595
120
80
40
0
2000
01
02-1
02-2
03-1
03-2
average
Figure 1. Densities of aphids on maize at the end of flowering (means + S.E.).
Bug community: Summation of bugs (phytophagous, carnivorous and polyphagous taxa)
revealed a picture similar to that observed in aphids and thrips. Differences among the 6 fields
and between the two varieties were observed, but no clear preference for one of the two
varieties could be found [means (individuals per plant): CV = 0.82, S.E. = 0.26; BT = 0.99,
S.E. = 0.25].
Aphid predators: The results did not show statistically significant differences among the 6
fields and the two maize varieties studied. The means of the 6 fields indicated a quite similar
predator level in both maize-growing systems [means (predator units per plant): CV = 0.20,
S.E. = 0.04; BT = 0.22, S.E. = 0.06].
82
Individuals per plant
50
p=0.00595
n = 5 (2000, 2002, 2003)
40
n = 10 (2001)
30
20
p=0.0005
10
0
2000
01
02-1
02-2
03-1
03-2
average
Figure 2. Densities of thrips on maize at the end of flowering (means + S.E.).
Ordination analyses of ground and maize-dwelling arthropod communities
Figure 3 visualises the results of the CA comprising the years 2000, 2001 and 2002. Field
characteristics and yearly changing environmental conditions turned out to be more important
for community composition than maize variety. Year, field location and sampling date (maize
dwellers) accounted for 59 % (p<0.001) and 33.5 % (p<0.001) of the total variance in ground
and maize-dwelling arthropods, respectively.
Partial CCA was performed to demonstrate the effect of maize variety (Figure 4). Only
2.7 % and 2.1 % of total variance were explained by maize variety for ground and maizedwellers, respectively. Although the differences in community composition among the three
maize varieties were small, they proofed to be highly significant (p<0.001).
1.0
1.0
H-2000
8-2001
8-2000
O-2002-1
8-2002
7-2001
O-2000
6-2000
O-2002-2
7-2000
O-2001
total
variance: 0.875
-0.8
-0.8
7-2002
total variance: 1.324
1.3
-0.8
-0.8
1.3
Figure 3. Ordination of sample sites: Pitfall traps for ground dwellers (a) and harvested plants
for maize dwellers (b, one symbol represents the central value of plants taken from one field
and one sampling date). Bt - black triangles, CV - grey circles, IS - light grey squares.
Sample sites of same location (a) and same sampling date (b) for ground and maize-dwellers,
respectively, are encircled (H-Halle/Saale; O-Oderbruch).
83
0.4
CCA-2
-0.4
-0.4
0.35
CCA-1
Variance explained by
maize variety
other envir. variables
0.7
ground dwellers (a)
2.7% (p<0.001)
59.0% (p<0.001)
CCA-2
-0.45
-0.30
CCA-1
maize dwellers (b)
2.1% (p<0.001)
33.5% (p<0.001)
Figure 4. Partial ordination of sample sites to show the influence of maize variety on
ordination of samples based on data from 2000-2002. Pitfall traps for ground dwellers (a) and
harvested plants for maize dwellers (b). For meaning of symbols refer to Fig. 3.
Discussion
Our studies yielded simple field-to-field comparisons under practical conditions in an
Ostrinia nubilalis outbreak area. The density analyses were carried out using data from 2000
to 2003, and the ordination analyses were done using data from 2000 to 2002.
Potentially Bt-toxin-sensitive Lepidoptera larvae other than the European corn borer
were very rare in the investigated maize fields. A few caterpillars colonised weeds or rape
plants arising from the previous year. In 2003 some Autographa gamma larvae and one
Orgyia antiqua larva were observed feeding on green maize leaves. All the other insects and
spiders detected were non-target arthropods. Therefore only a small impact of Bt maize would
be expected.
There were a few significant differences in densities of non-target arthropods between
the two maize varieties (Bt maize and CV maize). However, summation of the 4-year data
revealed no clear preference or avoidance for one of the maize varieties. Further years of
study and larger numbers of samples are needed to better identify statistical significance of
small differences (Wold et al., 2001).
Here, we paid particular attention to the arthropod community structure. Ground and
maize-dwelling arthropods exhibited the same general pattern. The most important
environmental factors shaping the arthropod communities were field and yearly changing
environmental characteristics and not the maize variety. The small but significant influence of
variety observed in our study was not affected by the different occurrence of Ostrinia
nubilalis because this species was excluded from the analysis. A recent study conducted by
Brooks et al. (2003) also showed that most species were unaffected by genetically modified
herbicide-tolerant crops.
84
Acknowledgement
We thank Mrs. Birgit Schlage and all students who assisted in the field studies, data
documentation and statistical analyses. We are also grateful to Mr. Joachim Gruel and Mr.
Andreas Schober for carabid identification. We would also like to thank Ms. Suzyon Wandrey
for editing the manuscript.
References
ter Braak, C.J.F. & Šmilauer, P. 2002: CANOCO Reference manual and CanoDraw for
Windows User’s guide: Software for Canonical Community Ordination (version 4.5).
Microcomputer Power (Ithaca, NY, USA), 500 pp.
Brooks, D.R.; Bohan, D.A.; Champion, G.T.; Haughton, A.J.; Hawes, C.; Heard, M.S.; Clark,
S.J.; Dewar, A.M.; Firbank, L.G.; Perry, J.N.; Rothery, P.; Scott, R.J.; Woiwod, I.P.;
Birchall, C.; Skellern, M.P.; Walker, J.H.; Baker, P.; Bell, D.; Browne, E.L.; Dewar,
A.J.G.; Fairfax, C.M.; Garner, B.H.; Haylock, L.A.; Horne, S.L.; Hulmes, S.E.; Mason,
N.S.; Norton, L.R.; Nuttall, P.; Randle, Z.; Rossall, M.J.; Sands, R.J.N.; Singer, E.J. &
Walker, M.J. 2003: Invertebrate responses to the management of genetically modified
herbicide-tolerant and conventional spring crops. I. Soil-surface-active invertebrates.
Phil. Trans. R. Soc. Lond. B 358: 1847-1862.
Freier, B.; Möwes, M.; Triltsch, H. & Rappaport, V. 1998: Predator units - an approach to
evaluate coccinellids within the aphid predator community in winter wheat. IOBC/wprs
Bull. 21(8): 103-111.
Wilhelm, R.; Beißner, L. & Schiemann, J. 2003: Concept for the realisation of a GMOmonitoring in Germany. Nachrichtenbl. Deut. Pflanzenschutzd. 55: 258-272.
Wold, S. J.; Burkness, E. C.; Hutchison, W. D. & Venette, R. C. 2001: In-field monitoring of
J. Entomol. Science 36: 177-187.
pp. 85-91
Population development of some predatory insects on Bt and non-Bt
maize hybrids in Turkey
Mustafa Güllü, Fahri Tatli, Ali Duran Kanat, Mahmut İslamoğlu
Plant Protection Research Institute Department of Entomology and Phytopathology, P. O.
Box 21, 01321 Adana, Turkey (E-mail: [email protected])
Abstract: The present study was conducted in Çukurova, Adana, Turkey, in the years 2001 and 2002.
The goal of the study was to determine the impact of Bt-transgenic (DK-626 Bt) and non-transgenic
maize hybrids (DK-626 and Pioneer P-3394) on the population development of generalist predatory
insects, i.e. Chrysoperla carnea (Stephens) (Neuroptera: Chrysopidae), Scymnus levaillanti Mulsant
(Coleoptera: Coccinellidae), Stethorus gilvifrons Mulsant (Coleoptera: Coccinellidae), Nabis spp.
(Heteroptera: Nabidae) and Orius spp. (Heteroptera: Anthocoridae) under field conditions.
Experiments were designed as randomized complete blocks with four replications. Periodic samplings
were done once a week, in 5 different points, on 5 adjacent plants at the same row. In total 25 plants
per plots were visually examined to count the predators on all vegetative plant parts above soil. The
results revealed that predator populations varied throughout the maize growing season. With one
exception, predator populations did not differ among the maize hybrids. The only case was the
abundance of C. carnea in the year 2002, which was found to be significantly higher in the Bt-maize
plots compared to the non-Bt hybrids.
Key words: predatory insects, population development, transgenic Bt maize, non-transgenic maize.
Introduction
European Corn Borer, Ostrinia nubilalis Hbn. (Lep.: Crambidae) and Corn Stalk Borer,
Sesamia nonagrioides Lep. (Lep.: Noctuidae) are the two main pests of maize that lead to
economical losses in Turkey. Furthermore, other lepidopteran and sucking insect species
become a problem in different phenological growth stages of the plant (Kornoşor, 1999;
Güllü, 2000). It has been reported from different maize varieties that 7 lepidopteran species,
especially European corn borer and corn stalk borer led to a total average loss of 16% to
30.8% in second crop maize in Cukurova, Turkey (Güllü, 2000).
A number of predatory insects are active in maize that help to supress the pest species by
feeding eggs and young larvae of the European corn borer, corn stalk borer, beet armyworm,
Spodoptera exigua Hbn. (Lepidoptera: Noctuidae), cotton leafworm, Spodoptera littoralis
(Boisd.) (Lepidoptera: Noctuidae) and corn earworm, Heliothis armigera (Hbn.)
(Lepidoptera: Noctuidae) and preying on eggs, nymphs and adults of soft bodied insects such
as aphids (Homoptera: Aphididae), leafhoppers (Homoptera: Cicadellidae), mites (Acarina:
Tetranychidae) and thrips (Thysanoptera: Thripidae) (Kornoşor, 1999; Kornoşor & Sertkaya,
1999). In recent years, in many countries, especially in the USA, transgenic maize varieties
that express Cry1Ab or Cry9C genes derived from Bacillus thuringiensis (Bt) subs. kurstaki
were succesfully used to control the European corn borer (Ostlie et al., 1997). Many
laboratory and field studies have been conducted on side-effects of Bt-transgenic plants on
natural enemies (Sims, 1995; Dogan et al., 1996; Orr & Landis, 1997; Pilcher et al., 1997;
Hilbeck et al., 1998a,b; Pfannenstiel & Yeargan, 1998; Schuler et al., 1999; Dutton et al.,
2002; Ponsard et al., 2002) and many research projects are still ongoing.
85
86
Ongoing field trials on Bt-transgenic maize that started in 1998 in Turkey investigate
effects on target pests, non-target herbivores and beneficial insects, fungi, and mycotoxins. In
the present study, the population development of different predatory insects, i.e. the green
lacewing, Chrysoperla carnea (Stephens) (Neuroptera: Chrysopidae), ladybird beetles,
Scymnus levaillanti Mulsant (Coleoptera: Coccinellidae) and Stethorus gilvifrons Mulsant
(Coleoptera: Coccinellidae), pirate minute bugs, Orius spp. (Heteroptera: Anthocoridae) and
demsel bugs, Nabis spp. (Heteroptera: Nabidae) was studied in 2001 and 2002 in Bttransgenic and non-transgenic maize fields.
Maize varieties used included the Bt-transgenic maize (DK-626 Bt, Monsanto) and its nontransformed near isoline (DK-626) as well as a non-transgenic maize hybrid (P-3394) from
Pioneer. Experiments were designed as randomized complete blocks with four replications.
Each plot was 120 m2 [5.6 m (8 rows) x 20m] in size. Periodic samplings were done for
larvae, nymphs and adults of C. carnea, S. givifrons, S. levaillanti, Orius spp. and Nabis spp.
once a week, in 5 different points, on 5 adjacent plants in the same row. In total 25 plants per
plots were visually examined to count the predators on all vegetative plant parts above soil.
Sampling was started at the 2-3 leaf growth stage and terminated at the maturity stage.
Treatment means were separated using Duncan Test at a significance level of P=0.05.
Chrysoperla carnea: First individuals were recorded at the end of July and early August in
2001 and 2002, respectively (Figures 1 a, c), closely coinciding with the reproductive stages
of maize (including tasseling, silking and milky stages).
Number of individuals
per 25 plants
5
DK-626 Bt
2001
2
DK-626
P-3394
3
2
1
2002
3
1
18.10
10.10
3.10
26.9
19.9
12.9
5.9
29.8
22.8
15.8
8.8
1.8
P-3394
2002
2
25.7
DK-626
Hybrids
Mean number of
individuals
per 25 plants
DK-626 Bt
DK-626 Bt
DK-626
P-3394
Date
0.4
0
0
c)
0.8
b)
Date
4
P > 0.05
1.2
24.9
17.9
10.9
3.9
27.8
20.8
13.8
6.8
30.7
23.7
16.7
9.7
2.7
0
a)
2001
1.6
Mean number of
individuals
4
1.4
1.2
a
1
0.8
b
0.6
0.4
0.2
0
DK-626 Bt
d)
b
DK-626
P-3394
Hybrids
Figure 1. Population development of C. carnea in Bt-transgenic and non-transgenic maize
hybrids. Different letters indicate significant differences (Duncan test, P0.05). Error bars
represent SE.
87
This might be due to the fact that maize pollen is a rich food source for adult green lacewings.
Thereafter, C. carnea population fluctuated at lower densities until mid September or October
when maize plants were at maturity stage. There were no significant differences in the mean
number of C. carnea on Bt-transgenic and both non-Bt maize hybrids in 2001 (Figure 1 b)
while a significant difference was recorded among the hybrids in 2002 (Figure 1 d). Mean
number of C. carnea individuals on the Bt-maize were higher than those recorded on the two
non-Bt maize hybrids.
2001
6
4
2
24.9
DK-626 Bt
DK-626
P-3394
4
3
2
1
3
2
1
0
DK-626 Bt
DK-626
P-3394
Hybrids
2002
P > 0.05
1.2
1
0.8
0.6
0.4
0.2
0
Date
18.10
10.10
3.10
26.9
19.9
12.9
5.9
29.8
22.8
15.8
8.8
1.8
0
25.7
per 25 plants
4
b)
Date
2002
c)
17.9
10.9
3.9
27.8
20.8
13.8
6.8
30.7
23.7
16.7
9.7
0
P > 0.05
5
Mean number of
individuals
8
a)
2001
DK-626 Bt
DK-626
P-3394
Mean number of
individuals
10
2.7
per 25 plants
Scymnus levaillanti: First individuals of S. levaillanti were recorded in mid July in 2001 and
in early August 2002 (Figures 2 a, c). Highest population levels of the predator were observed
in mid August 2001 at the tasseling and silking stage of maize, and then declined until end of
September. In 2002, similar trends were seen and highest population levels occurred at late
whorl stage of all hybrids and fluctuated until beginning of October when maize hybrids were
at maturity stages. There were no significant differences in mean number of S. levaillanti
individuals among the different hybrids in both years (Figures 2 b, d).
DK-626 Bt
d)
DK-626
P-3394
Hybrids
Figure 2. Population development of S. levaillanti in Bt-transgenic and non-transgenic maize
hybrids. Error bars represent SE.
Stethorus gilvifrons: Population dynamics of S. gilvifrons differed strongly between the two
study years (Figures 3 a, c). During both years, highest population levels coincided with the
reproductive stages of the maize hybrids then continued at lower densities. The mean number
of individuals did not differe significantly among the hybrids (Figures 3 b, d).
4
8
Mean number of
individuals
DK-626
Bt
DK-626
2001
10
P-3394
6
4
2
a)
2001
2
1
DK-626 Bt
2002
0.16
Mean number of
individuals
DK-626 Bt
DK-626
0.6
P-3394
0.4
0.2
P > 0.05
0.12
0.08
0.04
0
18.10
10.10
3.10
26.9
19.9
12.9
5.9
29.8
22.8
15.8
8.8
1.8
0
25.7
per 25 plants
P-3394
Hybrids
2002
0.8
c)
DK-626
b)
Date
P > 0.05
3
0
24.9
17.9
10.9
3.9
27.8
20.8
13.8
6.8
30.7
23.7
16.7
9.7
0
2.7
per 25 plants
88
DK-626 Bt
DK-626
d)
Date
P-3394
Hybrids
Figure 3. Population development of S. gilvifrons in Bt-transgenic and non-transgenic maize
Nabis spp.: There were no differences among the Bt and non-Bt hybrids in respect the mean
number of Nabis spp. collected (Figures 4 b, d). However, populations fluctuated strongly
throughout the season (Figures 4 a, c). In 2001, highest population levels occurred on 27
August when maize was at early milky stage and population of Nabis spp. was recorded until
mid September. A similar trend was seen in 2002, with the highest populations occurring in
mid-August when maize was at late whorl stage.
DK-626 Bt
2001
2001
P > 0.05
P-3394
7
6
5
4
3
2
1
24.09.01
17.09.01
Date
2002
2
1.5
1
0.5
0
DK-626 Bt
b)
2
DK-626
2002
P > 0.05
0.5
Mean number of
individuals
P-3394
2
1
1
P-3394
Hybrids
DK-626
0.4
0.3
0.2
0.1
0
18.10
3.10
26.9
19.9
10.10
Date
12.9
5.9
29.8
22.8
15.8
8.8
1.8
25.7
0
c)
2.5
DK-626 Bt
3
per 25 plants
10.09.01
03.09.01
27.08.01
20.08.01
13.08.01
06.08.01
30.07.01
23.07.01
16.07.01
09.07.01
a)
02.07.01
0
Mean number of inviduals
Number of inviduals per25 plants
DK-626
DK-626 Bt
d)
DK-626
P-3394
Hybrids
Figure 4. Population development of Nabis spp. in Bt-transgenic and non-transgenic maize
89
Orius spp.: There were no significant differences in the mean number of Orius spp. among
the Bt and non-Bt hybrids in both years (Figures 5 b, d). However, the number of individuals
caught was again significantly influenced by the plants’ phenological growth stage. Again,
highest population levels occurred when maize was in the reproductive growth stages and
Orius spp. was recorded until mid September when the maize plants were in their late
reproductive growth stages (Figures 5 a, c).
2001
15
per 25 plants
30
20
10
24.9
17.9
10.9
3.9
27.8
20.8
Date
13.8
6.8
30.7
23.7
16.7
2.7
30
5
0
DK-626
P-3394
Hybrids
b)
2002
DK-626 Bt
DK-626
P-3394
20
10
M ean n u m ber of
in d ividu als
2002
18.10
10.10
3.10
26.9
19.9
12.9
29.8
22.8
15.8
8.8
1.8
25.7
5.9
Date
d)
P > 0.05
10
8
6
4
2
0
DK-626 Bt
0
c)
10
DK-626 Bt
9.7
P-3394
0
a)
per 25 plants
M ean num ber of
individuals
DK-626
40
P > 0.05
2001
DK-626 Bt
DK-626
P-3394
Hybrid
Figure 5. Population development of Orius spp. in Bt-transgenic and non-transgenic maize
Although the population densities of C. carnea, S. levaillanti, S. gilvifrons, Nabis spp.
and Orius spp. did not differ among the Bt-transgenic (DK-626 Bt) and the two nontransgenic maize hybrids (DK-626 and P-3394) in both study years, large fluctuations were
observed during the growing seasons. The only significant difference that was detected was
for C. carnea. This predator was more abundant in Bt-maize compared to the non-Bt hybrids
in the year 2002. In general, populations of insect predators increased strongly when maize
was in the 12-16 leaves growth stages (late whorl, tasseling, silking, milky stage). During
these growth stages, aphids and thrips were encountered in the whorl leaves and small mite
colonies were found on the lower surfaces of the bottom leaves of maize plant when they
were feeding. While C. carnea and Nabis spp. were found all over the maize plant, Orius
spp. was mainly found on the whorl leaves where most of the thrips were located. In addition
to this, S. levaillanti individuals were observed preying on aphids and S. gilvifrons
individuals preying on mites. Different predator insects, i.e. C. carnea, S. levaillanti, and S.
gilvifrons, were found intensively on tassels, indicating that they feed on pollen. Some
reserchers recorded that the general predators in maize used pollen as a food source and the
90
plant as an oviposition substrat (Andow, 1990; Coll & Bottrell, 1991, 1992; Pilcher et al.,
1997).
Similarly to our study, Pilcher et al. (1997) reported that there were no differences in the
abundance of predators on Bt and non-Bt hybrids in maize field in one year, while predator
abundance was increased in the Bt-hybrid in a second year. While a number of field studies
revealed no negative effects of Bt-transgenic plants on natural enemy populations (e.g.
Jasinski et al., 2003), some reported positive effects (Hoy et al., 1998). While it has been
reported that larvae of C. carnea are negatively affected when fed with lepidopteran larvae
that had been reared on Bt-maize (Hilbeck et al., 1998a,b; Dutton et al., 2002), Pilcher et al.
(1997), reported that there was no detrimental effect of Bt-toxin expressing pollen against the
ladybird beetle, Coleomegilla maculata DeGeer, Orius insidiosus Say and the green
lacewing, C. carnea. Similarly, Lundgren & Wiedenmann (2002), reported that they did not
dedect any effect on C. maculata DeGeer that fed on transgenic corn pollen.
Overall, our study revealed no significant differences in the population developments of
different important predatory insect species between Bt and non-Bt maize fields in Turkey.
References
Andow, D.A. 1990: Characterization of predation on egg masses of Ostrinia nubilalis
(Lepidoptera: Pyralidae). Ann. Entomol. Soc. Am. 83: 482-486.
Coll, M. & Bottrell, D.G. 1991: Microhabitat and resource selection of the European corn
borer (Lepidoptera: Pyralidae) and its natural enemies in Maryland field corn. Environ.
Entomol. 20: 526-533.
Coll, M. & Bottrell, D.G. 1992: Mortality of European corn borer larvae by natural enemies in
different corn microhabitats. Biol. Contr. 2: 95-103.
Dogan, E.B., Berry, R.E., Reed, G.L. & Rossignol, P.A. 1996: Biological parameters of
convergent lady beetle (Coleoptera: Coccinellidae) feeding on aphids (Homoptera:
Aphididae) on transgenic potato. J. Econ. Entomol. 89: 1105-1108.
Entomol. 27: 441-447.
Güllü, M. 2000: Çukurova’da mısırda zararlı lepidopter türlerinin farklı mısır çeşitlerindeki
populasyon gelişmeleri üzerinde araştırmalar.(Studies on the population developments of
harmful lepidopterous species on differernt corn varieties in Çukurova region.) Çukurova
Ünivrsitesi Fen Bilimleri Enstitüsü,(Basılmamış Doktora Tezi) (Unbuplished PhD
Thesis), 198pp.
Hilbeck, A., Baumgartner, M., Fried, P.M. & Bigler, F. 1998a: Effects of transgenic Bacillus
carnea (Neuroptera: Chrysopidae). Environ. Entomol. 27: 480-487
Hilbeck, A., Moar, W.J., Pusztai-Carey, M., Filippini, A. & Bigler, F. 1998b: Toxicity of
Bacillus thuringiensis Cry1Ab toxin to the pretator Chrysoperla carnea (Neuroptera:
Chrysopidae). Environ. Entomol. 27: 1255-1263.
Jasinski, J.R., Eisley, J.B., Young, C.E., Kovach, J. & Willson, H. 2003: Selected nontarget
arthropod abundance in transgenic and non transgenic field crops in Ohio. Environ.
Entomol. 32: 407-413.
Hoy, C.W., Fledman, J. & Gould, F. 1998: Naturally occurring bilogical controls in
genetically engineered crops. In: Conservation Biological Control, ed. Barbosa, P.,
Academic Press, London: 185-205.
91
Kornoşor, S, 1999: Entomological problems of maize in Turkey. Proceedings of the XX
Conference of the International Working Group on Ostrinia and Other Maize Pests.
Adana (Turkey), 4-10 September 1999: 14-23.
Kornoşor, S. & Sertkaya, E. 1999: Lepidopterous pests and their natural enemies on maize in
Çukurova Region. Proceedings of the XX Conference of the International Working
Group on Ostrinia and Other Maize Pests. Adana (Turkey), 4-10 September 1999: 26-31.
Lundgren, J.G. & Wiedenmann, R.N. 2002: Coleopteran-specific Cry3Bb toxin from
transgenic corn polen does not affectthe fitnessof a nontarget species, Coleomegilla
maculata DeGeer (Coleoptera: Coccinellidae). Environ. Entomol. 31: 1213-1218.
Orr, D.B. & Landis, D.A. 1997: Oviposition of European corn borer (Lepidoptera: Pyralidae)
Entomol. 90: 905-909.
Ostlie, K. R., Hutchinson, W.D. & Hellmich, R.L. 1997: Bt corn and European corn borer.
NCR publication 602 University of Minnesota, St. Paul, MN.
Pfannenstiel, R.S. & Yeargan, K.V. 1998: Partitioning two and tree - trophic – level effects of
resistant plants on the predator, Nabis roseipennis. Entomol. Exp. Appl. 88: 203-209.
Pilcher, C.D., Obrycki, J.J., Rice, M.E. & Levis, L.C. 1997: Preimaginal development,
survival, and field abundance of insect predators on transgenic Bacillus thuringiensis
Ponsard, S., Guterrez, A.P. & Mills, N.J. 2002: Effect of Bt-toxin (Cry1Ac) in transgenic
cotton on the adult longevity of four Heteropteran predators. Environ. Entomol. 31:
1197- 1205.
Schuler, T.H., Poppy, G.M., Kerry, K.B., & Denholm, I. 1999: Potential side effects of insectresistant transgenic plants on arthropod natural enemies. Trends Biotechnol. 17: 210-216.
Sims, S.R. 1995: Bacillus thuringiensis var. kurstaki Cry1A(c) protein expressed in transgenic
cotton: Effects on beneficial an other non-target insects. Southwest. Entomol. 20: 493500.
92
pp. 93-96
The GMO Guidelines Project and a new ecological risk assessment
Angelika Hilbeck, David Andow1, Evelyn Underwood
Geobotanical Institute, Swiss Federal College of Technology (ETHZ), Zurichbergstrasse 38,
CH-8044 Zurich, Switzerland (E-mail: [email protected]); 1University of Minnesota,
St. Paul, MN 55108, USA
Abstract: The “GMO Guidelines Project” represents the efforts of public-sector scientists, under the
patronage of the International Organization for Biological Control (IOBC), to bring the collective
wisdom of the scientific community together to help develop a scientific risk assessment that fits the
conditions of the environment and the aspirations of the Cartagena Protocol on Biosafety. The project
is developing international and scientifically acknowledged methodologies with which to evaluate the
risks posed by the cultivation of genetically modified crops. Public sector scientists are welcome to
join the project via the website at www.gmo-guidelines.info.
Key words: GMO Guidelines Project, Cartagena Protocol, risk assessment
Introduction
In September 2003, the Cartagena Protocol on Biosafety went into force, calling for scientific
risk assessments of genetically modified organisms (GMOs) prior to their introduction into
the environment.1 The use and utility of GMOs has been hotly debated for about 15 years and
one of the purposes of the Protocol is to establish the basis on which these controversial
organisms will be evaluated (CBD, 2000). Now that the Protocol has entered into force, it is
essential that the scientific basis for risk assessment be developed so that Parties to the
Protocol can develop the necessary infrastructure to evaluate GMOs. This project represents
the efforts of public-sector scientists to bring the collective wisdom of the scientific
community together to help develop a scientific risk assessment that fits the conditions of the
environment and the aspirations of the Protocol. A group of public sector scientists from all
over the world has been assembled under the patronage of the International Organization for
Biological Control (IOBC) for this task, and funding was obtained from the Swiss
Development Cooperation Agency (SDC) for the “GMO Guidelines Project”.
The project structure
The “GMO Guidelines Project” aims to develop international and scientifically acknowledged
guidelines and methods with which to evaluate the risks posed by the cultivation of
genetically modified crops. The project has divided the scientific scope of the new ecological
risk assessment of GM crops into five sections: Problem Formulation and Options
Assessment (PFOA), Transgene Expression and Locus Structure, Non-target and Biodiversity
Impacts, Gene Flow and its Consequences, and Resistance Evolution and Management. Each
scientific section described how to stage the scientific risk assessment process in relation to
the development of the GM plant itself. Each of the sections is described below.
1
The Biosafety Protocol addresses “living modified organisms” (LMOs), but for the purpose of the project, our
use of GMO is equivalent to the definition of LMO in the Protocol.
93
94
The “Problem Formulation and Options Assessment” section provides a framework for a
deliberative formulation of the problem situation that the transgenic crop plant is expected to
alleviate and a deliberative assessment of the options that may be or become available for
addressing that problem situation in specific crop production contexts. It sets the context for
the environmental analyses that follow in the other sections. It defines the target agroecosystems for which the GM crop or an alternative solution is proposed, including the crop
system, farming system and ecological and structural context, and the people who will be
affected.
The integrity and stability of the structure of the transgene locus, the patterns of
expression of the transgene, and the consistency of transmission between generations have
significant implications regarding environmental risk. The “Transgene Expression and Locus
Structure” section identifies the genetic issues relevant for risk assessment and proposes
methodologies to address these issues and thereby reduce or identify the hazards. It structures
the characterization of the transgene into four components; transgene design, genotypic
characterization, phenotypic characterization and transmission between generations. Some of
these characteristics should be measured in the field, and others can be measured in the
laboratory. Nutritional characterization of the harvested GM crop is outside the scope and
expertise of this project, but the section includes an assessment of all products of the
transgene itself in the plant, in all its copies, plus any marker gene or other inserted genes or
gene parts. The assessment of phenotype includes possible effects the transgene might have
on traits other than the trait it is targeted for (pleiotropic effect) and its possible interaction
with the regulation and expression of other genes to affect a trait in addition to the trait it is
targeted for (epistatic effect).
The assessment of possible non-target and biodiversity impacts prior to environmental
release of a GM crop includes two steps, for which the project has developed new
methodologies. Firstly, a procedure has been developed to determine the non-target species or
structural characteristics of the biota or ecosystem functions that should be tested for impacts
as a highest priority. In order to focus the selection on ecological functional groups, the
project has identified certain functional categories, including a) herbivores including pests and
potential pests, and their function as disease vectors, b) biological control function, or natural
enemies, c) pollinators and pollen feeders, d) soil ecosystem functions, e) weeds, or crop
associated flora, f) plant pathogens, and g) insect pathogens. The following value related nontarget effects will be looked at similarly; f) species of conservation concern, and g) species of
cultural significance. A remaining challenging question is what is the role of other ‘neutral’ or
‘value unknown’ non-target species? The vast majority of species found in an agricultural
field are ‘neutral’ or ‘value unknown’ species. Secondly, scientific procedures and principles
for testing these identified species / characteristics / functions for likelihood of adverse effects
are described. These procedures are primarily laboratory and greenhouse ones, and some field
procedures are also proposed.
The section on gene flow and its consequences develops methods for establishing the
likelihood of intra- and interspecific gene flow, the possibility of subsequent geographic and
genetic spread of transgenes, and the potential ecological effects resulting from gene flow,
such as invasiveness, weediness, effects on non-target species, agricultural production and
biodiversity, crop land races, and impact on conservation issues. All types of recipient
populations are considered, including non GM varieties, landraces, related cultivated species,
feral populations and wild related species. The section addresses whether transgenes are likely
to increase in frequency due to natural and/or artificial selection and what the ecological
consequences of this process might be. It also plans to look at the effectiveness of sterility
mechanisms, their breakdown and management.
95
The “Resistance Evolution and Management” section identifies procedures to determine
the risk that target pests will evolve resistance to the GM crop or cropping practices
associated with the crop in the proposed environment, and feasible management responses
needed to reduce this risk. It also considers approaches for developing a practical monitoring
and response system to detect resistance and to adapt management appropriately.
The project includes three workshops in Kenya, Brazil and Vietnam, where the draft
methodologies are applied to real case study crops and cropping environments, and tested and
developed. We actively encourage the involvement of local scientists at these workshops,
where they interact with invited external experts from the project group. The final results will
be published and distributed internationally and locally in the case study countries. The
project is coordinated by a steering committee, the 14 members of which are responsible for
coordination of the scientific sections and regional groups. Contributing institutions include
the Kenya Agricultural Research Institute, International Centre of Insect Physiology and
Ecology, Brazilian Agricultural Research Institute (EMBRAPA), Vietnamese Ministry of
Agriculture and Rural Development, Scottish Crop Research Institute, Australian Cotton
Cooperative Research Centre, University of Minnesota, Ohio State University, International
Rice Research Institute, Chinese Academy of Agricultural Sciences, and the Swiss Federal
College of Technology. The guidelines are drafted by the project core group, which in
December 2003 consisted of 179 scientists from 26 countries and around 100 public sector
institutions.
The case studies
The workshop in Kenya was held in November 2002, and culminated in the first publication
of the project (Hilbeck & Andow, 2004). This book provides a detailed examination of Bt
maize, in its proposed application in Kenya. We develop components of a scientific risk
assessment process, which are consistent with that called for by the Protocol, and illustrate
how they can be applied to the case study. In our view, risk assessment is not a decisionmaking process; it is an activity that supports a decision-making process. Indeed, we do not
attempt a full-blown risk assessment of Bt maize in Kenya, but illustrate, through this case
study, the scientific and logical process by which risk assessment can be conducted.
The final aim of the project is to develop ‘tools’ for scientific risk assessment of
transgenic plants, which can be taken up and adapted to the specific needs and situation of
each country or institution investigating these questions. The Brazilian case study of Bt cotton
was particularly instructive for the process. The cotton non-target insect complex is very
diverse. Brazil has wild cotton relatives, feral, and ‘dooryard’ cottons as well as the cultivated
and improved cotton crops. Farming systems and environmental conditions are very different
in each cotton-growing region. This complexity reinforced the necessity of a case-by-case
approach. The risks associated with Bt cotton will vary across the different regions of Brazil.
These two workshops have provided us with very illustrative and contrasting agricultural,
climatic, socio-economic and ecological situations, and Vietnam will be very different again.
Even though the case study crop will be the same as in Brazil – Bt cotton – this time it will be
a different transgene construct, and cotton cultivation in Vietnam is mostly on very small farm
units, as a dry season component in crop rotations with rice and many other crops. An
important test will be to see if the project tools and methods can be applied to these new
situations.
96
Participating in the project
Over the coming year all public sector scientists are invited to participate actively in the
discussion and to present their ideas for developing the guidelines. You can register via the
project website - www.gmo-guidelines.info - expressing interest in the scientific section(s)
you wish to work with. If you are interested in being informed about the progress of the
project, but do not wish to contribute or are not a public sector scientist, please register on the
mailing list.
References
CBD (Secretariat of the Convention on Biological Diversity) 2000: Cartagena Protocol on
Biosafety to the Convention on Biological Diversity: Text and annexes. Montreal:
Secretariat
of
the
Convention
on
Biological
Diversity,
www.biodiv.org/doc/legal/cartagena-protocol-en-pdf (accessed 1 December 2003).
Hilbeck, A. & Andow, D.A. (Eds) 2004: Risk Assessment of Transgenic Crops, Volume 1: a
Case Study of Bt Maize in Kenya. CAB International, Oxford, UK.
pp. 97-102
European corn borer (Ostrinia nubilalis): Studies on proteinase
activity and proteolytical processing of the B.t.-toxin Cry1Ab in
transgenic corn
Renate Kaiser-Alexnat, Wolfgang Wagner, Gustav-Adolf Langenbruch, Regina G.
Kleespies, Brigitte Keller, Bernd Hommel1
Federal Biological Research Centre for Agriculture and Forestry (BBA), Institute for
Biological Control, Heinrichstr. 243, 64287 Darmstadt, Germany (E-mail: [email protected]); 1BBA, Institute for Integrated Plant Protection, Stahnsdorfer Damm 81,
14532 Kleinmachnow, Germany
Abstract: One possibility to control the European corn borer (ECB) is the cultivation of B.t.-corn.
However, this can result in the development of resistant pest populations. To analyse possible mechanisms of resistance, a reference system for the identification and quantification of physiological
changes in the midgut was established. Studies on proteinase activities were conducted with a susceptible German ECB population. The digestive proteinases trypsin, chymotrypsin, elastase, and aminopeptidase were identified in the midgut sap of 5th instar larvae. In whole 1st and 2nd instars as well as
in the midgut epithelium of 5th instar larvae, the proteinase aminopeptidase was provable. Besides,
proteolytical processing of the B.t.-toxin (and protoxin) Cry1Ab as present in transgenic corn is
described.
Key words: European corn borer, Ostrinia nubilalis, B.t.-corn, midgut proteinases, trypsin,
chymotrypsin, elastase, aminopeptidase, B.t.-toxin Cry1Ab, protoxin, proteolytical processing
Introduction
In Europe, the economical most important pest in maize (Zea mays L.) is the European corn
borer (ECB, Ostrinia nubilalis). Thus, transgenic corn (B.t.-corn) highly insecticidal to the
larvae of ECB was developed based on a truncated Cry1Ab toxin of Bacillus thuringiensis.
However, the cultivation of the respective cultivars may result in the development of resistant
pest populations.
Depending on the mode of action of B.t.-toxins the potential of insect resistance to B.t.toxins is generally located at any step of the toxic pathway: ingestion, solubilization,
proteolytic processing, binding to specific receptors, membrane integration, pore formation,
cell lysis, and insect death (Ferré & van Rie, 2002). Two main mechanisms of resistance to
B.t.-toxins have been identified in other pest-B.t.-toxin-systems. One of them is proteinasemediated and the other receptor-mediated (Oppert et al., 1997; McGaughey & Oppert, 1998).
In order to establish preliminary reference systems for the characterization of potential
available resistant individuals, first studies on proteinase activities in the midgut of Cry1Ab
susceptible ECB larvae were carried out. Besides, the proteinases were tested for involvement
in the digestion of the B.t.-toxin Cry1Ab and the respective protoxin.
97
98
Isolation of midgut sap and BBMV
ECB larvae were reared on artificial diet up to the 5th instar. For the extraction of both, the
pure midgut sap and the midgut epithelium, the larvae were calmed on ice and dissected. The
total midguts were isolated and collected on ice. Due to the very small sizes of 1st and 2nd
instar larvae it was not possible to separate their midguts. For the sap extraction the midguts
as well as crushed whole 1st and 2nd instar larvae were centrifuged at 13.000 g for 15 min. The
resuite was stored at -18°C until usage. For preparation of brush border membrane vesicles
(BBMV) the isolated midguts were treated as described by Wolfersberger et al. (1987).
Photometrical tests
For the identification and quantification of proteinases, photometrical studies were conducted
using typical proteinase-indicating chromogenic substrates and specific inhibitors according
to the investigations of Wagner et al. (2002): Trypsin was tested with the substrate Nbenzoyl-L-arg-p-nitroanilide (BApNA) and soybean-trypsin-inhibitor (SBTI). Chymotrypsin
was tested with N-succinyl-ala-ala-phe-p-nitroanilide (SAAFpNA) and the inhibitor N-tosylL-phe chloromethylketone (TPCK). Elastase was tested with N-succinyl-ala-ala-pro-leu-pnitroanilide (SAAPLpNA) and elastatinal. Aminopeptidase was tested with leu-p-nitroanilide
(LpNA) and bestatin. Carboxypeptidase was tested with hippuryl-phe and hippuryl-arg.
Proteolytic assays and SDS-PAGE
The experiments were done with purified Cry1Ab toxin and protoxin which was prepared by
Dr. J.A. Jehle (State Education and Research Center for Agriculture, Viticulture and
Horticulture; SLFA Neustadt; Germany). Model proteinases were obtained from Sigma.
Digestions were performed at 25°C with a final toxin concentration of 1 mg/ml and either a
midgut sap dilution of 1:10 or a model proteinase concentration of 0.25 mg/ml, respectively.
Proteolyses were stopped by heating the samples for 2 min at 95°C.
SDS-PAGE was done according to Laemmli (1970) using the Roti-Load1 no. K929.1
denaturation buffer from Roth and 15% polyacrylamide gels. Fluka standards no. 69810
(indicated as “low”) and 69811 (indicated as “high”) were used as reference standards.
Proteinase activity in the midgut sap of 5th instar larvae
In the midgut of a Canadian ECB population, Houseman & Chin (1995) identified the
digestive proteinases trypsin, chymotrypsin, elastase, and aminopeptidase. To compare their
results with German ECB, a reference system, which is also intended to be used to
characterize potential available resistant ECB´s, was established to identify and to quantify
changes in proteinase-activities in the midgut of the pest insect. Thus, midgut sap of German
susceptible 5th instar larvae was extracted and photometrical studies were carried out. Similar
to the above described results, trypsin, chymotrypsin, elastase, and aminopeptidase were
identified (Kaiser-Alexnat et al., 2003). In additional tests the presence of other potential
activities, e.g. carboxypeptidase, could not be highlighted (data not shown).
Beside the examined serine proteinases and metalloproteinases, other classes of proteolytic activities are unlikely to be present in the midgut sap of ECB due to physiological
reasons. As reviewed by Terra et al. (1996), cysteine proteinases are generally common in the
midgut of hemipteran Heteroptera or in slightly acid media and aspartic proteinases are only
active at very acid pH values. Kaiser-Alexnat et al. (2003) demonstrated that the pH of pure
99
larval midgut sap of ECB 5th instar larvae is lightly basic, ranging between 7.2 and 7.5,
depending on the rearing method before sample preparation.
Proteinase activity in whole 1st and 2nd instar larvae
Generally, early larval stages are known to be more sensitive to B.t.-toxin than late instars.
Unfortunately, it was not possible to separate the midguts of 1st and 2nd instar larvae due to
their very small sizes. Thus, with sap of whole larvae it was examined whether the activity of
the above described proteinases is provable, too. As a control, no proteinase activity could be
demonstrated in the haemolymphe (data not shown). In the sap of whole 1st (Fig. 1) and 2nd
(Fig. 2) instar larvae, aminopeptidase activity was identified and quantified in photometrical
tests. The diagrams show the means and the standard deviation of each experiment which was
done three times. Columns indicated as “0” quantify the prevailing proteolytic activity;
columns indicated with an increasing concentration of inhibitor show the specific inhibition of
the proteolyses which is a tool to identify the type of proteinase.
LpNA-hydrolysis
(nmol/ml/min)
150
100
50
0
0
0.1
1
10
Bestatin (µM)
100
LpNA-hydrolysis
(nmol/ml/min)
Figure 1. Aminopeptidase-activity ( sd) in the sap of whole 1st instar larvae.
300
200
100
0
0
0.1
1
10
Bestatin (µM)
100
Figure 2. Aminopeptidase-activity ( sd) in the sap of whole 2nd instar larvae.
Proteinase activity of BBMV from 5th instar larvae
A membrane-bound aminopeptidase is one possible receptor for B.t.-toxins in the midgut
epithelium (Oppert, 1999). Based on this fact and in context with the above described results,
100
aminopeptidase activity of BBMV isolated from 5th instar larvae was demonstrated using the
established test system (Fig. 3). Binding analyses are presently carried out to show the interaction between Cry1Ab and the receptor.
LpNA-hydrolysis
(nmol/ml/min)
50
40
30
20
10
0
0
0.1
1
10
Bestatin (µM)
100
Figure 3. Aminopeptidase-activity ( sd) in the midgut epithelium (BBMV´s) of 5th instar
larvae.
Proteolytical processing of B.t. toxin and protoxin Cry1Ab
After incubation with midgut sap from 5th instar larvae, the B.t.-toxin Cry1Ab was processed
during the first minute. As a result, the 65 kDa toxin was digested for 2 kDa, resulting in a 63
kDa protein. This protein is stable for at least 60 minutes (Fig. 4). In order to show that this
protein shortening is due to proteolytic activity, a control was performed using midgut sap
that was heated at 95°C for 5 min. Due to this denaturation of the proteinases, no digestion of
the B.t.-toxin took place (see also Fig. 4).
Figure 4. Proteolytical processing of B.t.-toxin Cry1Ab with midgut sap (DS) of 5th instar
larvae.
101
To identify the activity among the midgut proteinases, which are responsible for the proteolytical reaction, available model proteinases were used to simulate the midgut conditions,
i.e. bovine trypsin, bovine chymotrypsin, porcine elastase, and Aeromonas aminopeptidase.
As a result, both the 65 kDa toxin as well as the 135 kDa protoxin were digested to 63 kDa by
all types of proteinases proved in the midgut sap of ECB, except aminopeptidase. The proteolytical processing with trypsin and chymotrypsin is demonstrated in Fig. 5 and the one
with elastase and aminopeptidase in Fig. 6.
Figure 5. Proteolytical processing with trypsin and chymotrypsin.
Figure 6. Proteolytical processing with elastase and aminopeptidase.
102
Acknowledgements
Thanks are due to Dr. J.A. Jehle, State Education and Research Center for Agriculture,
Viticulture and Horticulture (SLFA Neustadt), Biotechnological Crop Protection,
Neustadt/Weinstrasse, Germany for providing us with Cry1Ab toxin and protoxin. Thanks are
also due to the BMBF for supporting the research project.
References
Ferré, J. & van Rie, J. 2002: Biochemistry and genetics of insect resistance to Bacillus
thuringiensis. Annu. Rev. Entomol. 47: 501-533.
Houseman, J.G. & Chin, P.-S. 1995: Distribution of digestive proteinases in the alimentary
tract of the European corn borer Ostrinia nubilalis (Lepidoptera: Pyralidae). Arch. Insect
Biochem. Physiol. 28: 103-111.
Kaiser-Alexnat, R., Wagner, W., Langenbruch, G.A., Kleespies, R.G., Keller, B., Meise, T.,
& Hommel, B. 2003: Selection of resistant European corn borer (Ostrinia nubilalis) to
B.t.-corn and preliminary studies for the biochemical characterization. IOBC wprs
Bulletin. 9th European Meeting of the IOBC wprs Working goup „Insect pathogens and
entomoparasitic nematodes“ Growing biocontrol markets challenge research and
development. May 23-29, 2003, Schloß Salzau, Germany, in press.
Laemmli, U.K. 1970: Cleavage of structural proteins during the assembly of the head of
bacteriophage T4. Nature 227: 680-685.
McGaughey, W.H. & Oppert, B. 1998: Mechanisms of insect resistance to Bacillus
thuringiensis toxins. Israel J. Entomol. 32: 1-14.
Oppert, B., Kramer, K.J., & McGaughey, W.H. 1997: Insect resistance to Bacillus thuringiensis toxins. The Chemist: 7-10.
Oppert, B. 1999: Review: Protease interactions with Bacillus thuringiensis insecticidal toxins.
Arch. Insect Biochem. Physiol. 42: 1-12.
Terra, W.R., Ferreira, C., Joradāo, B.P., & Dillon, R.J. 1996: Digestive enzymes. In: Biology
of the Insect Midgut, eds. Lehane and Billingsley. Chapman and Hall, London: 153-166.
Wagner, W., Möhrlen, F., & Schnetter, W. 2002: Characterization of the proteolytic enzymes
in the midgut of the European Cockchafer, Melolontha melolontha (Coleoptera:
Scarabaeidae). Insect Biochem. Molec. Biol. 32: 803-814.
Wolfersberger, M.G., Luthy, P., Parenti, P., Sacchi, V.F., Giordana, B., & Hanozet, G.M.
1987: Preparation and partial characterization of amino acid transporting brush border
membrane vesicles from the larval midgut of the cabbage butterfly (Pieris brassicae).
Comp. Biochem. Physiol. 86A: 301-308.
pp. 103-108
First investigations on the effects of Bt-transgenic Brassica napus L. on
the trophic structure of the nematofauna
Barbara Manachini1, Simona Landi1, Maria Carola Fiore2, Margherita Festa1, Salvatore
Arpaia3
1
Istituto di Entomologia Agraria, Università degli Studi di Milano, via Caloria 2, 20133
Milano, Italy (E-mail: [email protected]); 2Metapontum Agrobios, Metaponto
Italy; 3ENEA – Italian National Agency for New Tecnology, Energy and Environment,
Rotondella, Italy
Abstract: The research was carried out in experimental fields planted with isogenic and transgenic
canola (Brassica napus L., cv. Westar). Canola plants were transformed with a construct carrying a
Bacillus thuringiensis insecticidal protein gene (Bt Cry1Ac), the green fluorescent protein (GFP) and
the Kanamicin resistance (nptII) marker genes, under the control of the cauliflower mosaic virus 35S
promoter. The objective of the present study was to investigate possible effects of the transgenic
canola on the soil fauna, and in particular on the nematode community. Nematoda is a large phylum,
including species with different dietary behaviours, moderate capacity of active motility and a known
sensitivity to several stresses; all these characteristics make them excellent bio-indicators. The
nematode communities were investigated in respect to their abundance and trophic structure in Bttransgenic and isogenic canola fields. Nematodes were extracted from soil samples using the
Baermann funnel method. In total, 17,846 nematodes were collected from 3600g soil, 9,890 from the
isogenic canola plots and 7,956 from the Bt-transgenic ones. The dominant trophic group was that of
the bacteriophagous nematodes, followed by the fungal feeders and plant parasites. While no
difference in abundance was found for bacteriophagous species between the two treatments, fungal
feeders were more abundant in Bt canola plots compared to the isogenic canola. A difference was also
detected for the phytophagous species which were more abundant in the non-transformed canola.
Indicator Species Analysis highlight the correlation between transgenic canola community and fungal
feeders and the isogenic community with phytophagous. Multiple Range Permutation Procedure
revealed a significant difference between the two communities.
Key words: Nematoda, transgenic canola, Cry1Ac, non target soil organisms.
Introduction
Several crops have been genetically modified to express insecticidal proteins from Bacillus
thuringiensis. These plants represent a great promise for pest control (Delannay et al., 1989),
but their use might cause unintended side effects on non-target organisms including natural
enemies.
Most of the studies conducted to investigate the impact of Bt-transgenic plants on nontarget organisms, analyzed the above ground fauna (e.g. Hilbeck et al., 1998; Lozzia et al.,
1998; Birch et al., 1999; Losey et al., 1999; Bernal et al., 2002; Dutton et al., 2002).
Relatively few data are available regarding the potential effects of Bt proteins directly on the
soil fauna, on the soil trophic chain and on soil functions. For example, Yu et al. (1997)
reported an experiment on two different non-target soil arthropods, the collembolan, Folsomia
candida Willem, and the mite, Oppia nitens Koch. No negative effects were recorded on the
life cycle and mortality of these organisms when fed in laboratory on leaf material from Bt103
104
cotton. Field studies were conducted by Szkeres et al. (2003) and Manachini (2000) who
studied the influence of transgenic corn on soil dwelling Carabids. However, there are many
soil macro- and micro-organisms that are more representative of soil ecosystems, that could
be affected by the presence of the toxins from transgenic plants in the soil (via root exudates,
plant tissues). Tapp & Stotzky (1995) observed that Bt toxin could remain in the soil for a
long period of time, especially if it is bound to clay particles. Moreover, Kostella & Stotzky
(1997), also demonstrated that the toxins tightly bound to clays and other soil particles, are
more resistant to bacterial degradation. Consequently, these toxins released from Bttransgenic plants could directly influence the soil fauna.
Nematodes are the most abundant organisms in the soil after Bacteria and Fungi. They
have a reduced motility, are rather sensitive to stresses and include species with different
dietary habits. They are represented in food webs as primary and intermediate consumers
(Bongers & Ferris, 1999), and could potentially suffer directly and indirectly from Bt toxins.
A few studies were carried out on the impact of GM plants on the nematofauna. Manachini &
Lozzia (2002) investigated the nematode community for its general composition, trophic
structure and biodiversity in Bt and control maize, cultivated in clay soils. Manachini et al.
(2003) evaluated nematode species assemblage on eggplant expressing Cry3Bb toxin and
found that the nematode assemblage was comparable between transgenic and control plots.
Experimental canola fields were cultivated between January and April 2003, with Brassica
napus L. cv Westar and the non-transformed isogenic line. The transgenic lines expressed a
truncated synthetic Bacillus thuringiensis Berliner insecticidal crystal protein gene (Bt
Cry1Ac), plus the green fluorescent protein (GFP) and the Kanamicin resistance (nptII)
marker genes, under the control of the cauliflower mosaic virus 35S promoter. The
experimental field was located in Metaponto, Southern Italy (Permit No. B/IT/022). The
experimental layout was prepared according to a completely randomized design with three
replications, for each of the two treatments (transgenic canola and its isogenic control line).
Plot size was 220 m2. In April 2003, thirty soil samples were collected per plot using a core
(35 cm length and 3 cm diameter) following a random sampling scheme, to account for the
typically clustered distribution of nematodes (Moens, 1993). Samples from a single plot were
pooled, put in plastic bags, and kept cool (+4°C) until they were extracted. For each plot,
nematodes were extracted from four replicates of 150g of sub-sample soil by Baermann
funnel method, counted and then processed as described by De Grisse-Cobb (1969). From
each sub-sample, 100 nematodes were randomly picked and observed with a microscope and
separated in trophic groups according to Yeates et al. (1993).
Nematode abundance
In total, 17,846 nematodes were extracted from 3,600 g of soil. Table 1 shows the total
number of nematodes counted and the average of nematodes per 150g of soil in different
samples. The total abundance was 9,890 nematodes in the isogenic canola crop and 7,956
nematodes in the transgenic one. This difference between the two types of communities was
not statistically significant (ANOVA; F=1.86; d.f.=1, 23; p= 0.19).
105
Table 1. Abundance of nematodes in soil samples taken from plots in which Bt-transgenic or
untransformed (isogenic) canola was grown.
Isogenic
Bt-transgenic
Plot Total nematodes Mean
±SE Plot Total nematodes Mean
±SE
Nem./150g
Nem./150g
1 iso.
3389
874.25 125.6 2 Bt
3667
916.75 112.2
5 iso.
2599
649.75 35.7 3 Bt
2055
513.75 75.9
6 iso.
3902
975.50 75.3 4 Bt
2234
558.50 89.2
Total
9890
824.17 76.35 Total
7956
663.00 92.43
Percentage
Trophic structure
The nematode community was investigated for ecological functions and a total of 1278
nematodes were grouped according to Yeates et al. (1993) in feeding groups. The bacterial
feeders were the most abundant group, accounting for about 54% in the isogenic canola plots
and 55% in the transgenic ones, followed by fungal feeders and phytophagous species (Figure
1). The percentage of fungal feeders was higher in Bt canola (38%) than in the isogenic crop
(29%), while the phytophagous species were more abundant (17%) in isogenic canola than in
the Bt canola (7%). Omnivore and predatory nematodes were almost absent, constituting less
than 1% of the nematode community.
100
90
80
70
60
50
40
30
20
10
0
5 4 .7 %
5 4 .5 %
3 7 .8 %
2 8 .6 %
1 6 .7 %
7 .3 %
C a no la B t+
B a c te rivore s
C a no la B tF unga l fe e de rs
P hytopha gous
Figure 1. Functional groups compiling the nematode community in Bt-transgenic and
untransformed canola fields.
Previous results were also confirmed by the Indicator Species Analysis (Dufrene &
Legendre, 1997), that revealed the association of phytophagous nematodes with the isogenic
canola crop (p=0.015) and the fungivores with the transgenic one (p=0.016) (Table 2). The
Multi Response Permutation Procedure (Zimmerman et al., 1985), detected a significant
difference (p=0.032) in the community structure of trophic groups of nematodes between
isogenic and transgenic canola (Table 3).
106
Table 2. Indicators Species Analysis.
Taxon
Bacterivores
Fungivores
Phytophagous
Omnivores
Predators
Average
49
49
30
1
1
Max
50
58
45
3
2
Bt49
39
45
0
2
Bt+
50
58
16
3
0
p
0,802
0,016
0,015
0,477
0,999
Table 3. Multi Response Permutation Procedure.
MRPP
T
Observed
Expected
R
p
-2,396
7,553
7,668
0,015
0,032
There was no statistically significant difference in the total number of nematodes found
in soil from Bt and non-Bt canola, and this result agrees with findings by Saxena & Stotzky
(2001). Nevertheless, nematodes constitute a rather diversified phylum, then their abundance
alone is not a sensitive enough indicator to account for the effects of the Bt toxin in the soil.
Moreover different research showed that toxins from different strains of Bacillus
thuringiensis have deleterious effects on larvae and eggs of nematodes (Meadows et al., 1990;
Bottjer et al., 1985). Therefore the analysis of trophic groups may represent a more
appropriate method to investigate possible Bt effects as suggested by Saxena & Stotzky
(2001).
It is not clear which factors were responsible for the observed differences between the
trophic composition in the nematode community in Bt canola and isogenic canola fields.
Differences in the nutritional composition of the plants and the root exudates could be one
factor. The decreased abundance of phytophagous nematode species in Bt canola fields could
be due to direct effects of the Bt toxin that is likely to be ingested when the nematodes feed
on plant roots. However, these direct effects still have to be investigated in further studies.
Furthermore, fungivores and bacterivores could ingest micro-organisms that live on root
exudates or on plant tissues in which the toxin is expressed, then ingesting the toxin through
this route. At higher trophic levels indirect effects could also be possible due to changes in the
abundance of soil micro-organisms. The fact that bacterivores were present at the same level
in the two different treatments, suggests that there appear to be no quantitative effects of the
Bt canola. Nevertheless this functional group could be influenced in genus composition.
Therefore, research will continue to define the effects at genus and species level.
The primary result of this study is the confirmation of the relevance of nematodes as
possible bioindicators and their usefulness in risk assessment analysis of soil organisms. More
long term studies are necessary to confirm our results of a possible impact of Bt-transgenic
canola on species composition and on the biodiversity of the nematofauna also with the aim
of clarifying possible mechanisms that may lead to different species assemblage, compared to
isogenic canola fields.
107
Acknowledgements
The authors B. Manachini and S. Landi are grateful to the EU project ProBenBt QLK3-CT2002-01969. M.C. Fiore and S. Arpaia were supported through the EU grant QLK3-CT-200000547 (Project Bt-BioNoTa).
References
Bernal, C.C., Aguda, R.M. & Cohen, M. 2002: Effect of rice lines transformed with Bacillus
thuringiensis toxin genes on the brown planthopper and its predator Cyrtorhinus
lividipennis. Entomol. Exp. Appl. 102: 21-28.
Birch, A.N.E., Geoghegan, I.E, Majerus, M.E.N., McNicol, J.W, Hackett, C.A., Gatehouse,
A.M.R. & Gatehouse, J. 1999: Tritrophic interactions involving pest aphids, predatory 2spot ladybirds and transgenic potatoes expressing snowdrop lectin for aphid resistance.
Mol. Breed. 5: 75-83.
Bongers, T. & Ferris, H. 1999: Nematode community structure as a bioindicator in
environmental monitoring. Tree 14: 224-228.
Bottjer, K.P., Bone, L.W. & Gill, S.S. 1985: Nematoda: susceptibility of the eggs to Bacillus
thuringiensis toxins. Exp. Parasitol. 60: 239-244.
De Grisse, A.T. 1969: Redescription ou modification de quelques tecniques utilisées dans
l’etude des Nématodes phytoparasitaires. Med. Fac. Landbouww. Rijksuniv. Gent 34:
351-369.
Delannay, X.B.J., Proksch, R.K., Fuchs, R.L., Sims, S.R., Greenplate, J.T., Marrone, P.G.,
Dodson, R.R., Augustine, J.J., Lavton, J.G. & Fischhoff, D.A. 1989: Field performance
of transgenic tomato plants expressing the Bacillus thuringiensis var. kurstaki insect
control protein. BioTechnology 7: 1265-1269.
Dufrene, M. & Legendre, P. 1997: Species assemblages and indicator species: the need for a
flexible asymmetrical approach. Ecol. Mono. 67: 345-366.
Entomol. 27: 441-447.
Hilbeck, A., Baumgartener, M., Friend, P.M & Bigler, F. 1998: Effects of transgenic Bacillus
carnea (Neuroptera, Chrysopidae). Environ. Entomol. 27: 480-487.
Kostella, J. & Stotzky, G. 1997: Microbial utilization of free and clay-bound insecticidal
toxins from Bacillus thuringiensis and their retention of insecticidal activity after
incubation with microbes. Appl. Environ. Microbiol. 63: 3561-3568.
Losey, J.E., Rayor, L.S. & Carter, M.E. 1999: Transgenic pollen harms monarch butterflies.
Nature 399: 214.
Lozzia, G.C., Furlanis, C., Manachini, B. & Rigamonti, I. 1998: Effects of Bt corn on
Rophalosiphum padi L. (Rhyncota, Aphidididae) and on its predator Chrysoperla carnea
Stephen (Neuroptera, Chrysopidae). Boll. Zool. Agr. Bachic. Ser. II 30: 153-164.
Manachini, B. 2000: Ground beetle assemblages (Coleoptera, Carabidae) and plant dwelling
non-target arthropods in isogenic and transgenic corn crops. Boll. Zool. Agr. Bachic. Ser.
II 32: 181-198.
Manachini, B. & Lozzia, G.C. 2002: First investigations into the effects of Bt corn crop on
Nematofauna. Boll. Zool. Agr. Bachic. Ser. II 34: 85-96.
108
Manachini, B., Fiore, M.C., Landi, S. & Arpaia, S. 2003: Nematode species assemblage in Btexpressing transgenic eggplant and their isogenic control. In: Biodiversity implications
on genetically modified plants, Monte Verità, Ascona, Switzerland, September 7-12,
2003: 31-32.
Meadows, J. & Bone, L.W. 1990: Bacillus thuringiensis strains affect population growth of
the free-living nematode Turbatrix aceti. Invert. Repro. Develop. 17: 73-76.
Moens, M. 1993: Selected techniques used in Nematology and related exercises. Ed.
International Nematology Course. Gent University.
Saxena, D. & Stotzky, G. 2001: Bacillus thuringiensis (Bt) toxin released from root exudates
and biomass of Bt corn has no apparent effect on earthworm, nematodes, protozoa,
bacteria, and fungi in soil. Soil Biol. Biochem. 33: 1225-1230.
Szkeres, D., Kadar, F., Skentkiralyi, F. & Kiss, J. 2003: Structural characteristics and seasonal
dynamics of ground beetle (Coleoptera, Carabidae) assemblages in experimental Bt and
isogenic maize fields in Hungary. In: Biodiversity implications on genetically modified
plants, Monte Verità, Ascona, Switzerland, September 7-12 2003: 26.
Tapp, H. & Stotzky, G. 1995: Insecticidal activity of the toxins from Bacillus thuringiensis
subspecies kurstaki and tenebrionis adsorbed and bound on pure and soil clays. Appl.
Yeates, G.W., Bongers, T., De Goede, R.G.M., Freckman, D.W. & Geordieva, S.S. 1993:
Feeding habits in soil nematode families and genera – an outline for soil ecologists. J.
Nematol. 25: 315-331.
Yu, L., Berry, R.E. & Croft, B.A. 1997: Effects of Bacillus thuringiensis toxin in transgenic
cotton and potato on Folsomia candida (Collembola: Isotomidae) and Oppia nites
(Acari: Orbatidae). J. Econ. Entomol. 90: 113-118.
Zimmerman, G.M., Goetz, H. & Mielke, P.W.Jr. 1985: Use of an improved statistical method
for group comparisons to study effects of prairie fire. Ecology 66: 606-611.
pp. 109-116
Studies on the effects of Bt corn expressing Cry1Ab on two parasitoids
of Ostrinia nubilalis Hb. (Lepidoptera: Crambidae)
Barbara Manachini, Giuseppe Carlo Lozzia
Istituto di Entomologia Agraria, Università degli Studi di Milano, via Celoria 2, 20133
Milano, Italy (Email: [email protected])
Abstract: The results of a two-year field and laboratory study aimed at a better understanding of the
influence of transgenic corn on two parasitoids of Ostrinia nubilalis Hübner (ECB) are reported. In
order to detect the effects on parasitism rate by Lydella thompsoni Herting on ECB, mature larvae of
O. nubilalis were collected from Bt corn (event 176) and from its isogenic line at nine geographical
sites. Considering the entire amount of the larvae, the ECB from transgenic Bt corn displayed a lower
level of parasitism both in percentage and in absolute numbers of parasitoids.Trichogramma brassicae
Bezdenko was the main egg parasitoid species of O. nubilalis recorded in corn fields in Northern
Italy. The levels of parasitism, the vitality of embryos and adults and the sex ratio of T. brassicae were
investigated. For this goal, egg masses of O. nubilalis, collected from isogenic and transgenic corn
fields were used. Statistical analyses revealed no significant differences in the percentage of
parasitism, number, longevity and mortality of adults of T. brassicae emerging from ECB eggs
oviposited on Bt or isogenic corn leaves. No statistical differences were recorded for all parameters
analysed for T. brassicae emerging from eggs of ECB surviving on transgenic corn or collected from a
conventional corn crop.
Key words: Transgenic plant, Bt corn, non-target organisms, Ostrinia nubilalis, Lydella thompsoni,
Trichogramma brassicae
Introduction
The European corn borer (ECB), Ostrinia nubilalis Hb., is a very serious pest of corn in
Europe and in the US causing losses of $12/acre with estimated losses approaching 1.5 billion
dollars per year in the USA (USA-EPA, 2003). Researchers are continuously developing
environmentally safe ways of managing O. nubilalis and other corn borers. Current control
methods, i.e., treatments with chemical insecticides, require precise timing of applications for
maximum control, and in addition, insecticides contribute to environmental contamination.
Biological control was also performed but with contrasting results depending on the countries
and other factors as e.g. climate conditions and the biology of the pest (Maini & Burgio,
1991; Hassan, 1994; Bigler et al., 1997; Manachini, 2002a).
Despite consistent losses due to the European corn borer, an effective solution is far from
being found and many growers are reluctant to use current integrated pest management (IPM)
methods to control ECB. This is due to several factors such as: larval damage is often hidden,
heavy infestations are unpredictable, scouting multiple times each summer takes time and
requires skill, insecticides are expensive and raise health or environmental concerns and
benefits of ECB management are uncertain. For these reasons, corn, Zea mays L., was
modified with a gene from Bacillus thuringiensis subsp. kurstaki Berliner that produces a
protein (the Cry1Ab) able to kill the ECB larvae. However, the use of Bt transgenic plants
raises some questions on the possible direct and indirect effects on the environment, and in
particular on non-target organisms. Results are contrasting: some authors (Sims, 1995; Pilcher
109
110
et al. 1996; Orr & Landis, 1997) have reported that the Bt plants they used do not create
problems for the tested insects, on the other hand other authors (Hilbeck et al., 1998, 1999;
Schuler et al., 1999) detected some negative effects. However, it is unlikely that these effects,
found under laboratory conditions, will have an ecological impact in the field (Dutton et al.,
2003).
In an effort to preserve the natural enemies of the ECB, and more general, of non-target
arthropods, it is still a major scientific goal to investigate the use of transgenic corn (Orr &
Landis, 1997; Gould F., 1998; Hilbeck et al., 1998; Manachini, 2003). Natural enemies in
agro-ecosystems are beneficial organisms that help control pest herbivores at or below cropdamaging levels. Since the introduction of pesticides into agro-ecosystems approximately half
a century ago, there have been drastic reductions in the numbers of natural enemies in crop
fields; as a result their functional services, or better the lack thereof, became much more
apparent to farmers and ecologists. Natural enemies are very much reduced in current
intensive agricultural systems and must be substituted by external means such as increasing
use of pesticides. From the problems arising from the misuse of pesticides, it was learned that
modern, innovative sustainable agricultural production methods must continue to conserve
natural enemies as a free or low-cost pest control (Gould, 1998). At the same time all new
pest control substances or technologies must be tested for their side effects on natural
enemies, prior to their large-scale release. Due to all these considerations it is important to
investigate the impact of Bt corn on non-target organisms with particular attention to natural
enemies of O. nubilalis. The most important ECB parasitoid species recorded in the Italian
corn fields are Lydella thompsoni Herting and Trichogramma brassicae Bezdenko
(Manachini, 2002b). This last one is often used in IPM or in biological programs. Bt corn
could affect these natural enemies both directly and indirectly.
The study has been carried out to assess the influence of the transgenic corn crop on the
possible side effects on the two major parasitoids of ECB L. thompsoni and T. brassicae.
Materials and Methods
Collection and insect culture of Ostrinia nubilalis Hb.
The study was carried out in 1999 and 2000 on different strains of ECB collected from
different geographical areas in northern Italy (Fig. 1) and by laboratory bioassays. All
collection sites, except 3 and 3’, were identical in the two years. For these two sites, the
environmental conditions were similar and no statistical differences among the percent
parasitism or ECB attack between the years was recorded (Manachini, 2003). For this reason
they were considered in the further analysis as a single site.
Three hundred 5th instar larvae were randomly collected at each locality in a transgenic
cornfield (event BT 176) and in the isogenic control field and the larvae were brought to the
laboratory. Diapause was broken in a climate chamber at 25  1 °C, 18:6 hours L:D and 75 
5 % RH. Chrysalids were put in cylindrical cages with a white paper on top on which the
adults oviposited the egg masses. Adults were fed with water and honey. ECB egg masses
oviposited on the paper and on transgenic and isogenic corn leaves were collected for rearing
purposes and for the experiments with the egg parasitoid T. brassicae. Egg masses used for
the continuous rearing of ECB were transferred in a plastic box containing a meridic diet. A
corrugated cardboard was inserted in the box as a site of pupation. Ostrinia nubilalis was
reared following the protocol described previously (Manachini, 2002b; Lozzia & Manachini,
2003).
111
. .
Figure 1. Sampling sites: 1. Santhià (TO); 2. Sillavengo (NO); 3. Cornegliano (LO) only for
collection in 1999; 3’. Brignano (BG) only for collection in 2000; 4. Grumello (CR); 5.
Cologna (RO); 6. Montebello (VI); 7. Minerbe (VR); 8. Vedelago (TV); 9. Morsano (PN).
Lydella thompsoni Herting
In order to break diapause of the field collected larvae of O. nubilalis, they were overwintered in the laboratory at 4°C (RH 85%) for 4 months. Afterward, they were placed in a
climatic chamber with 18 hours light at 25°C, 6 hours dark at 18°C and a humidity of 75 ±
5%. Each day the number of L. thompsoni pupae was checked and recorded. The time elapsed
from the pupa formation to the emergence of the adult and adult life span was recorded.. The
adults kept in the above-mentioned conditions were fed a 10 % honey solution.
Analysis of the data
The percent parasitism was calculated on the basis of the number of L. thompsoni pupae
emerging from the ECB larvae after the over-wintering period plus the few L. thompsoni
pupae found directly in the field. Data regarding the biology of L. thompsoni was analysed
using the two-way ANOVA test with the program SPSS 9.0.
Trichogramma brassicae Bezdenko
A Trichogramma brassicae population was established by collecting the parasitized eggs of
O. nubilalis directly in the field. The colony was maintained on eggs of the ECB. Adult wasps
that emerged from the hosts were maintained at 25  1 °C and 18:6 hours L:D and 75  5 %
RH in a glass tube. They were fed with a 10% solution of honey. The parameters and the
methodologies used are derived from quality control standards for Trichogramma and other
natural enemies developed by the Association of Natural Biocontrol Producers and the
International Organization of Biological Control Subcommittee on Quality Control (Bigler et
al., 1997). ECB egg masses oviposited on transgenic and isogenic maize leaves and on
different types of paper as control were presented to 200 mated female wasps for each test.
One egg mass, less than 24 hours old, was used for each female and placed in a glass tube for
30 minutes. The eggs were incubated at the above conditions and parasitsm was determined
by counting the number of black eggs containing T. brassicae pupae. Egg masses oviposited
by the ECB that developed on transgenic or on isogenic corn were used for the bioassay.
Adult ECB, subjected to the two treatments, oviposited on Bt and isogenic corn leaves.
Parasitism rate, emergence of the wasp and development time were considered to evaluate the
potential effect of Bt corn on the performance of T. brassicae.
112
Analysis of the data
Data from all experiments were subjected to ANOVA with the program SPSS 9.0.
Lydella thompsoni Herting
The data of the two year study are reported in Table 1. It shows the percent parasitism of O.
nubilalis larvae by L. thompsoni in transgenic and isogenic corn fields at 9 different localities
in Northern Italy. The number of L. thompsoni pupae collected in the fields just before harvest
(pupae autumn) are also reported. The level of attack of the ECB as an average of the number
of galleries per corn stalk is indicated. The average plant attack by O. nubilalis was higher in
the isogenic corn (2.7 ECB galleries per stalk) compared to the transgenic corn (0.85 ECB
galleries per stalk). It is important to remember that we used Bt 176 which has a reduced
resistance against ECB after flowering. These data confirm what was found by Lozzia &
Rigamonti (1996) showing that the second generation of ECB can attack the transgenic corn.
Parasitism by L. thompsoni was higher for the ECB in the isogenic corn crop. In fact, the
average percent of ECB parasitism by L. thompsoni was 7.18% in the isogenic corn and
5.84% in the transgenic corn.
Table 1. Parasitism of Ostrinia nubilalis by Lydella thompsoni at nine collection sites in the
years 1999 and 2000.
Locality
1
2
3 / 3’
4
5
6
7
8
9
Average
Average
ECB
galleries
/stalk
2.65
3.28
2.24
2.05
2.85
2.00
3.70
2.56
2.97
2.70
Isogenic
Pupae
Autumn
Pupae
Spring
%
Parasitism
0.50
2.00
0.00
0.50
1.50
0.00
0.50
0.50
1.50
0.78
10.00
20.50
2.50
5.00
21.50
3.50
4.50
5.50
13.50
9.61
10.40
13.73
2.70
4.09
12.06
2.63
4.94
5.04
9.06
7.18
Average
ECB
galleries
/stalk
0.88
0.97
0.63
0.72
0.95
0.43
1.08
0.93
1.08
0.85
Transgenic
Pupae
Pupae
Autumn Spring
0.50
1.50
0.50
0.00
1.50
0.00
0.00
0.00
1.00
0.56
5.00
8.50
2.00
7.50
12.50
4.50
2.00
2.00
12.50
6.28
%
Parasitism
4.72
9.16
2.63
6.58
7.42
4.52
2.60
2.46
8.56
5.40
Differences in the parasitism rate were found also among the fields but no differences
were recorded in the two different years (Manachini, 2003). Table 1 shows that in Grumello
(4) and Montebello (6), parasitism was higher in the transgenic fields compared to the
isogenic ones. The highest level of parasitism was found in the isogenic fields in Sillavengo
(2) followed by the ones in Cologna (5); while the lowest level of parasitism was recorded in
the transgenic field of Vedelago (8). These results could be explained in part by the presence
of Bt corn that decreases the population of ECB but also considering that the relationship
between the parasitoid and its host has dramatic fluctuations in parasitism between sites,
alternative hosts and seasons (Grenier et al., 1990; Manachini, 2000). Meta-population studies
focus on the dynamics of populations that occupy discrete patches in fragmented habitats
(Kuske, 2003).
113
Table 2 reports the results regarding the biology parameters of L. thompsoni. On average,
the slowest pupation of L. thompsoni was recorded when the parasitoids developed in ECB
larvae fed transgenic corn in Minerbe (7) (Av=24.00 days), while the fastest pupation was
recorded for L. thompsoni that developed in ECB larvae collected in the isogenic corn field of
Sillavengo (2) (Av=6.97 days). On average, the pupation of L. thompsoni is not later from the
ECB collected in transgenic fields (15.42 days) compared to those from isogenic corn fields
(13.97 days) (F=0.25; df=1,35; p=0.47). Moreover, the time from pupation to emergence of
the adult (Ae) was not different for L. thompsoni developed in ECB on Bt corn (Ae=9.13
days) compared to ECB on isogenic corn (Ae=10.64 days) (F=0.28; df=1,35; p=0.45).
Table 2. Development of Lydella thompsoni emerged from 300 larvae of Ostrinia nubilalis
collected in isogenic and Bt transgenic corn fields during 1999 and 2000.
Isogenic
Transgenic
L
NP Mx Av Mn NE Ae
L
Locality NP Mx Av Mn NE Ae
1
20
19 11.32 6
0
9.69 13.30 14
17 14.05 1
0
8.70 11.70
2
41
31 6.97 1
4 10.42 18.15 29
17 10.32 3
1 11.45 16.90
3/3’
5
26 19.75 8
0 12.25 13.60
6
43 18.00 1
1
9.75 12.50
4
10
15 13.98 13
0 9.955 13.20 12
35 13.35 1
1
9.68 13.80
5
43
32 11.49 2
6
9.10 15.90 32
36 13.50 3
1
6.77 15.70
6
9
39 21.95 3
0
9.94 13.40
9
25 17.25 8
1
9.55 11.85
7
7
42 9.71 3
0
8.80 11.50
4
42 24.00 3
1
9.50 10.55
8
11
20 11.25 6
0
9.88 14.45
9
18 14.00 10
0
7.50 12.85
9
27
38 16.56 7
4 12.98 14.10 24
25 13.35 5
0
8.44 14.65
13.97
1.56 10.64 13.98 15.44
15.42
0.66 9.13 13.49
Average 19.22
NP: number of pupae, Mx (maximum), Av (average) and Mn (minimum) number of days between termination
of diapause of Ostrinia nubilalis and pupation of Lydella thompsoni NE = number of L. thompsoni pupae from
which no adults emerged. Ae= average number of days from pupation of L. thompsoni to emergence of the
adults, L = average longevity of L. thompsoni adults.
The longevity (L) of the adults of L. thompsoni was not different between transgenic and
isogenic corn (F= 1.24; df=1,35; p=0.27). The longest lifespan was noticed for adults emerged
from ECB fed on isogenic corn of the corn field of Sillavengo (2), while the adults of L.
thompsoni developed in ECB fed Bt corn in Minerbe (7) showed the shortest longevity.
Trichogramma brassicae Bezdenko
The direct and indirect effects of Bt corn on parasitism by T. brassicae were investigated
using ECB egg masses oviposited on transgenic or on isogenic corn leaves (for details see
Material and Methods) and O. nubilalis collected in fields with transgenic and isogenic corn.
Figure 2 shows the parasitism for the four possible combinations (A to D). No statistical
difference of parasitism was recorded; considering as variable the ECB populations (from Bt
corn or isogenic corn) (F=0.37; df=1,119; p=0.51) or the type of corn leaves (F=0.56;
df=1,207; p=0.27).
In Figure 3 the percent emergence of adult wasps is reported. The lowest emergence of T.
brassicae was recorded from wasps developed in eggs of the ECB whose larvae fed on Bt
corn and the resulting adults oviposited on Bt leaves (70.59%). The statistical analysis shows
a significant difference to wasps that developed in eggs of the ECB fed as larvae on isogenic
corn and the adults oviposited on isogenic corn as well (92.68%), (F=0.95; df= 1,197;
p=0.05). This result shows the importance of investigating the situation after a number of
generations of both, the egg parasitoid and the ECB.
114
On the whole, no detrimental effects can be detected from the use of Bt corn on the two
parasitoids considered.
100
90
80
71.00
% parasitism
70
71.84
67.00
66.35
B
C
60
50
40
30
20
10
0
A
D
Figure 2. Percent parasitism of Ostrinia nubilalis eggs by Trichogramma brassicae when the
eggs were laid by different O. nubilalis populations on Bt and isogenic corn leaves
% emergence of wasps
A = ECB egg masses oviposited on transgenic corn leaves by O. nubilalis collected in Bt fields;
B = ECB egg masses oviposited on transgenic corn leaves by O. nubilalis collected in isogenic fields;
C = ECB egg masses oviposited on isogenic corn leaves by O. nubilalis collected in transgenic fields;
D = ECB egg masses oviposited on isogenic corn leaves by O. nubilalis collected in isogenic fields
100
90
80
70
60
50
40
30
20
10
0
87.67
92.15
92.68
C
D
70.59
A
B
ECB populations - type of corn leaves
Figure 3. Percent emergence of adults of first generation of T. brassicae. A = ECB egg masses
oviposited on transgenic corn leaves by O. nubilalis collected in Bt fields; B = ECB egg masses
oviposited on transgenic corn leaves by O. nubilalis collected in isogenic fields, C = ECB egg masses
oviposited on isogenic corn leaves by O. nubilalis collected in transgenic fields, D = ECB egg masses
oviposited on isogenic corn leaves by O. nubilalis collected in isogenic fields
115
Acknowledgements
These studies have been supported by MURST 40% 1999 (Ministero Università e Ricerca
Scientifica e Tecnologica) and FIRST 60% 2000 ( Fondo Interno per la Ricerca Scientifica e
Tecnologica).
References
Bigler, F., Suverkropp, B.P. & Cerutti, F. 1997: Host-searching by Trichogramma and its
implications for quality control and release techniques. In: Ecological Interactions and
Biological Control, eds. Andow, Ragsdale and Nyvall. Westview Press: 71-86.
BioControl. 48: 611-636.
Gould, F. 1998: Sustainability of transgenic insecticidal cultivars: integrating pest genetics
and ecology. Ann. Rev. Entomol. 43: 701-726.
Grenier, S., Anglade, P., Naibo, B., Galichet, P.F. & Hawlitzky N. 1990: Distribution of
tachinids (Diptera: Tachinidae), parasitoids of the European corn borer, Ostrinia
nubilalis (Lepidoptera: Pyralidae) in France (1985-87). Entomophaga. 35: 485-492.
Hassan, S.A. 1994: Strategies to select Trichogramma species for use in biological control.
In: Biological Control With Egg Parasitoids, eds. Wajnberg and Hassan. CAB
International, Oxon: 55-71
Hilbeck, A., Baumgartner, M., Fried, P.M. & Bigler F. 1998: Effects of transgenic Bacillus
carnea (Neuroptera, Chrysopidae). Environ. Entomol. 27: 480-487.
Hilbeck, A., Moar, W., Pusztai-Carey, J., Filippini, M. & Bigler, A. 1999: Prey-mediated
effects of Cry1Ab toxin and protoxin and Cry2A protoxin on the predator Chrysoperla
carnea. Entomol. Exp. Appl. 91: 305-316.
Kuske, S.D. 2003: Non target effect of mass releasing Trichogramma brassicae (Hym.,
Trichogrammatidae) against the European corn borer. PhD thesis, Swiss Federal Institute
of Technology Zurich: 71 pp.
Lozzia, G.C. & Manachini, B. 2003: Suscettibilità al Bacillus thuringiensis Berliner in
relazione allo sviluppo di Ostrinia nubilalis Hübner (Lepidoptera: Crambidae). Atti XIX
Congr. Naz. It., Catania, June 10-15, 2002 (in press).
Lozzia, G.C. & Rigamonti, I.E. 1996: Behaviour of the European corn borer, Ostrinia
nubilalis Hb., on transgenic corn. Boll. Zool. agr. Bachic. Ser. II. 28: 51-69.
Maini, S. & Burgio, G. 1991: Biological control of the European corn borer in protected
pepper by Trichogramma maidis Pint & Voeg. and Bacillus thuringiensis Berl. subsp.
kurstaki. Colloques de l'INRA 5: 213-215.
Manachini, B. 2000: Lydella thompsoni Herting (Diptera, Tachinidae), parasite of Ostrinia
nubilalis Hübner (Lepidoptera, Pyralidae) in North Italy. Phytophaga. 10: 69-78.
Manachini, B. 2002a: Compatibility of Chemical and Biological Pesticides. Encyclopedia of
Pest Management, ed. Pimentel. Marcel Dekker, New York: 134-137.
Manachini, B. 2002b: Valutazione dello sviluppo di forme di resistenza in Ostrinia nubilalis
Hb. alla proteina CryIAb espressa nel mais transgenico e dei suoi effetti
sull’entomofauna non target. PhD thesis, Università degli Studi di Bologna, Bologna.
Italy: pp. 200.
116
Manachini, B. 2003: Effects of transgenic corn on Lydella thompsoni Herting (Diptera:
Tachinidae) parasitoid of Ostrinia nubilalis Hb. (Lepidoptera: Crambidae). Boll. Zool.
agr. Bachic. Ser II. 35: 111-125.
Orr, D.B. & Landis, D.A. 1997: Ovoposition of European corn borer (Lepidoptera: Pyralidae)
and impact of natural enemy population in transgenic versus isogeni corn. Ann. Entomol.
Soc. Amer. 90: 905-909.
Pilcher, C.D., Obrycki, J.J., Rici, M.E. & Lewis, L.C. 1996: Pre-imaginal development,
Schuler, T.H., Potting, R.P.J., Denholm, I. & Poppy, G.M. 1999: Parasitoid behaviour and Bt
plants. Nature 401: 825-826.
Sims, S.R. 1995: Bacillus thuringiensis var. kurstaki protein expressed in transgenic cotton:
effects on beneficial and other non-target insects. Southwest. Entomol. 20: 493-500.
USA-EPA, United States Environmental Protection Agency, 2003. - Meeting summary: EPAUSDA Bt crop insect resistance management workshop.
http://www.epa.gov/oppbppd1/biopesticides/summary826.htm.
pp.117-123
Implications for the parasitoid Campoletis sonorensis (Hymenoptera:
Ichneumonidae) when developing in Bt maize-fed Spodoptera littoralis
larvae (Lepidoptera: Noctuidae)
Michael Meissle*, Eva Vojtech, Guy M. Poppy
University of Southampton, School of Biological Sciences, Biodiversity and Ecology Division,
Bassett Crescent East, SO16 7PX, Southampton, UK (*present address: Agroscope FAL
Reckenholz, Swiss Federal Research Station for Agroecology and Agriculture, Reckenholzstr.
191, 8046 Zurich, Switzerland, E-mail: [email protected])
Abstract: In this preliminary study, we examined the effects of Cry1A(b) expressing Bt maize on the
endoparasitoid Campoletis sonorensis (Cameron) developing in Spodoptera littoralis (Boisduval)
larvae. Caterpillars were reared on Bt maize (event Mon 810) and on the non-transformed near-isoline
since hatching. Second instar larvae were parasitised and kept individually in small plastic cups
thereafter until emergence of a parasitoid or moth. C. sonorensis larvae needed significantly longer to
develop in Bt fed host larvae. Cocoon weight, time from pupation to emergence, sex ratio and total
survival was not significantly different from the control. Cry1A(b) levels in plant material, caterpillars
and parasitoid cocoons were analysed by ELISA. S. littoralis ingested the toxin, which was also
detected in the third trophic level (parasitoid cocoons). Thus C. sonorensis larvae were exposed to
Cry1A(b) during development. We present a classical first tier study, which indicates a possible
hazard of Bt maize for C. sonorensis. We could not differentiate whether the observed effects were
caused directly or indirectly by the Bt toxin. During a comprehensive tiered risk assessment scheme,
behaviour as well as population dynamics of hosts, parasitoids and also predators have to be taken into
account, which allows the risk to be quantified. Detailed conclusions about possible ecological
consequences of Bt maize for C. sonorensis and other parasitoids can only be drawn once the entire
risk assessment has been completed.
Keywords: Bacillus thuringiensis, Cry1A(b), ELISA, first tier study, non-target effects, risk
assessment, transgenic maize, tritrophic interactions
Introduction
Parasitic wasps play an important role in biological control of insect pests, and thus they
should be studied as part of a risk assessment of transgenic crops. Parasitoid larvae usually
complete their whole development on a single host individual and are therefore closely
connected with their hosts. Insecticidal proteins like Bt toxins produced in GMOs are ingested
by herbivores feeding on leave tissue (Dutton et al., 2002) and passed on to the next trophic
level. Thus parasitoids developing in herbivores that consume leaf tissue (e.g. caterpillars) are
likely to be exposed to the Bt toxin. Even if there are no direct effects on parasitoids or other
species of higher trophic levels, there may be indirect effects due to altered host or prey
quality (Bourguet et al., 2002; Dutton et al., 2002) or behavioural changes (Schuler et al.,
1999).
Schuler et al. (2000) and Dutton et al. (2003) proposed a 3-tiered risk assessment scheme
consisting of laboratory studies (first tier) to quantify potential toxic effects, semi-field
(second tier) experiments to allow more natural but still controllable conditions and finally
field trials (third tier) that provide information of effects on a large scale. Case-by-case studies
117
118
for important key species are necessary as generalisation is not always possible. In this paper
we present a first tier (laboratory-scale) study to investigate
1)
2)
potential lethal and sublethal effects of transgenic Bt maize on the endoparasitoid
Campoletis sonorensis (Cameron) (Hymenoptera: Ichneumonidae) developing in
Spodoptera littoralis (Boisduval) (Lepidoptera: Noctuidae) larvae.
transfer of Cry1A(b) through the different trophic levels and thus exposure of the
parasitoid to the toxin.
S. littoralis is a pest of various crops, including maize (Hill, 1983) and the species is easy to
culture in the lab. Larvae ingest Bt toxin when feeding on Bt maize and are partly susceptible
to Cry1A(b) (Dutton et al., 2002). Thus S. littorlis is a suitable model organism to study both
direct and indirect effects on higher trophic levels. S. littoralis larvae were raised on maize
leave tissue since hatching. A mixture of different leaf and stem parts from 2–4-week-old
plants of the cultivar MEB307Bt (event Mon 810, Monsanto, “BT”) expressing Cry1A(b) and
the conventional, non-transformed near-isoline Monumental (“CONV”) were fed to the
caterpillars. Bioassays and cultures were maintained at 25 ± 3 °C, 60 ± 10 % RH and a
photoperiod of L:D 14:10 h.
The new world parasitoid C. sonorensis is a generalist that attacks a wide range of
lepidopteran pest larvae and is potentially important for biological control in cotton, maize
and other crops (Lindgren et al., 1970; Lindgren & Noble, 1972). Although S. littoralis is not
a natural host of C. sonorensis, development and survival was reported to be the same as on
Spodoptera frugiperda, a serious pest and natural host of the parasitoid in the new world
(Fritzsche Hoballah & Turlings, 2001). Thus for practical reasons, we have chosen
S. littoralis as a model herbivore. Parasitoid cocoons were kindly provided by the University
of Neuchâtel, Switzerland. Emerged parasitoids were kept with honey and mineral water until
needed.
For the experiments, 5-day-old S. littoralis larvae (second instar) weighing between
0.4 mg and 1.7 mg (initial weight) were parasitised individually in 30 ml plastic cups by
mated C. sonorensis females. The parasitoid was removed immediately after an attack was
observed. Caterpillars from both treatments were offered alternately to a single parasitoid to
make sure that each female attacked the same number of “BT” and “CONV” larvae.
Parasitoid females were replaced when their willingness to attack decreased, resulting in 2 to
39 larvae parasitised per female. The parasitised larvae were kept individually in the cups
thereafter. New plant material was provided daily and cups were cleaned when necessary.
Survival was recorded daily and larvae were weighed 6 days after parasitisation. Time from
parasitisation to cocoon formation was recorded as well as cocoon weight 1 day after
formation to prevent the fragile new cocoons from damage. Time from cocoon formation to
emergence and sex of the emerged adult parasitoid were also noted.
Five samples of maize tissue, each consisting of leaf and stem parts from about 5 plants
were taken during the experimental period from both Bt and conventional maize for Cry1A(b)
toxin level analysis using ELISA. Two samples (5 and 3 individuals) of 11-day-old
unparasitised S. littoralis larvae as well as 1 sample (4 individuals) of freshly formed
C. sonorensis cocoons were analysed from the “BT” treatment together with adequate
“CONV” controls. All samples were stored at –20°C. Toxin levels were analysed with the BtCry1Ab/1AC ELISA PathoScreen kit from Agdia, USA. Protein was extracted from
homogenized material over night at 4°C (ratio of sample fresh weight in mg to added buffer in
119
l was 1:1.5). After centrifugation total protein content was determined in Bradford assays.
After diluting the extracts appropriately (total soluble protein added to test well: maize 0.5 g,
S. littoralis 5 g, C. sonorensis 10 g) to be within the optimal range of sensitivity, the
provided standard test procedure was followed. Absorbance at 620 nm was read with a
microplate reader (Anthos reader 2001, Anthos labtech instruments) and calibrated with
Cry1A(b) standards from 0.0625 ng Cry1A(b)/ml to 16 ng/ml. To compensate possible bias
due to cross-reactions with other proteins, readings for the Bt treatment were corrected by
subtracting the tissue blanks (absorbance of “CONV” controls).
All data were analysed with SPSS 11.0 for Windows (SPSS Inc, Chicago, USA).
Survival probabilities were calculated using the life table procedure. In cases where no
parasitoid larva emerged and S. littoralis pupated, it was assumed that either deposition of the
parasitoid egg was unsuccessful or the egg was encapsulated. Individuals removed for ELISA
as well as disturbed larvae due to handling were considered in the survival probabilities as
censored cases. Survival of both treatments was compared with Wilcoxon (Gehan) statistic.
Since data were not normally distributed, differences between treatments were tested with
nonparametric Mann–Whitney U-tests, using exact significance levels. Correlations were
analysed with the nonparametric Spearman´s rho correlation coefficient.
ELISA tests showed that leaf material from Bt maize contained around 1600 ng Cry1A(b) per
gram fresh tissue (Table 1). S. littoralis larvae contained 645 ng/g fresh weight (40 % of the
concentration in maize), which indicates clearly that they ingested the toxin. As ELISA data
of maize tissue and caterpillars originated from a different, parallel running experiment, they
will be published and discussed in more detail elsewhere (Vojtech et al., submitted). We
measured toxin levels in C. sonorensis cocoons with 110 ng/g fresh weight. As this analysis
could not be replicated, the value is to be judged with caution, but nevertheless it gives an
impression of the order of magnitude. Our analysis showed that the parasitoid was directly
exposed to the toxin. Cry1A(b) is binding to lepidopteran midgut membrane cells (Schnepf et
al., 1998). As C. sonorensis consumes the entire host during its development including the gut
(Wilson & Ridgway, 1975), the toxin was transferred to the parasitoid and was detectable in
the cocoons. Although Cry1A(b) was transferred up to the third trophic level, it apparently did
not accumulate in the food chain since the parasitoids contained much less than their hosts.
Table 1. ELISA of leaf material, S. littoralis larvae and C. sonorensis cocoons. Given is the
number of replicates per sample and calculated concentration of Cry1A(b).
Sample
Number of replicates
Bt maize (event Mon 810)
S. littoralis larvae (day 11)
C. sonorensis (cocoons)
5
2
1
Cry1A(b) concentration
[ng/g fresh weight  S.E.]
1597 ± 438
645 ± 36
110
S. littoralis larvae used for the experiment were all 5 days old, but caterpillars feeding on
Bt maize were lighter than those from the conventional treatment (Mann–Whitney U-test,
p = 0.038). This result confirms that the species is partly susceptible to Bt maize (Dutton et
120
al., 2002). To make sure that observed differences for parasitoid development between the 2
maize treatments in our experiment were not only because of the weight difference
beforehand, initial host weight was checked for correlations with other parameters. There was
no significant correlation with time from parasitisation to pupation, weight of parasitoid
cocoon, time from pupation to emergence and parasitoid sex (Mann–Whitney U-test, p > 0.1).
Thus observed effects can be linked to Bt treatment and are most likely not due to different
weight conditions before the experiment.
C. sonorensis larvae developing in Bt maize feeding hosts needed 2.85 days longer than
those in the conventional treatment (Mann–Whitney U-test, p < 0.001, Figure 1), but
parasitoid cocoons showed no significant difference in weight (Mann–Whitney U-test,
p > 0.2). Apparently, C. sonorensis larvae could compensate the disadvantages of hosts
suffering from Cry1A(b) toxin with longer developmental time. Neither time from pupation to
emergence nor sex ratio was significantly different for the 2 treatments (Mann–Whitney Utest, p > 0.05). However, 6 days after parasitisation, host larvae from the Bt treatment were
still significantly lighter (Mann–Whitney U-test, p<0.001).
15
***
***
BT
CONV
20
10
5
10
Time in days
Weight in mg
30
0
0
Host weight
6 days after
parasitisation
Weight of
parasitoid
cocoon
Time from
parasitisation
to pupation
Time from
pupation to
emergence
Figure 1. Sublethal effects on C. sonorensis developing in Bt and conventional maize fed
S. littoralis hosts. Mean values ± SE, *** indicates significant differences at p < 0.001 level.
This brings up the issue of the nature of the observed effects. As shown by ELISA,
C. sonorensis pupae contained a considerable amount of Cry1A(b). Thus parasitoid larvae
were directly exposed to the toxin. One possible explanation of the longer developmental time
is direct toxicity of Cry1A(b) to the parasitoid larvae, even if this is unlikely as Cry1A(b) is
reported to bind to lepidopteran specific receptors (Schnepf et al., 1998). More probably, the
observed effect was indirect. Caterpillars suffering from Bt maize (“sick prey”) are likely to
provide a suboptimal amount and/or composition of nutrients for parasitoids and predators
(Bourguet et al., 2002; Dutton et al., 2002). In fact, Salama et al. (1983) showed changes of
S. littoralis haemolymph composition when larvae were treated with Bt toxin. Such changes
might result in a lower nutritional host quality for the parasitoid. Further more, a slower
growth rate of Bt maize fed caterpillars (Dutton et al., 2002) may delay parasitoid
development, as fewer nutrients are available in smaller hosts.
Survival over the whole developmental period was not significantly different for the
treatments “CONV” and “BT” (Chi2 = 1.489, df = 1, p > 0.05). Nevertheless, adult parasitoids
121
emerged in only 40 % of parasitised „BT“ larvae whereas 58 % parasitisations of “CONV”
larvae were successful (Figure 2). In 8 cases of the “CONV” treatment, parasitisation was
unsuccessful and S. littoralis developed normally whereas this was the case for only 2 “BT”
larvae. S. littoralis feeding on conventional maize might be able to better defeat the parasitoid
(e.g. encapsulate parasitoid egg). In the following development, mortality for host “BT”
larvae was higher than for their “CONV” counterparts (although this could not been shown
statistically). As larvae feeding on Bt maize were generally weaker and more vulnerable, Bt
toxin may have acted synergistically with the burden of the parasitoid to increase mortality
more than in the conventional treatment, where S. littoralis larvae had to cope with the
parasitoid alone.
Survival [%]
100
0
observed
parasitisation
CONV
BT
successful
eclosion
leaving host
cocoon
emergence
of adult
Developmental stage
Figure 2. Survival of C. sonorensis in its different developmental stages from parasitisation to
emergence of adult, calculated with SPSS Life table procedure. Number of parasitised larvae:
“BT” 45, “CONV” 48
The presented results show that developmental time of C. sonorensis in caterpillars
feeding on Bt maize was negatively affected. ELISA revealed that there was exposure to the
toxin and thus the nature of the effect may be direct. However, it is more likely that they are
indirect effects due to changes in host quality. Still, longer developmental time might
influence parasitoid population dynamics either positively or negatively in the field. Host
larvae are exposed to parasitoids for a longer time, which might increase parasitisation rates,
but the risk of predation might increase as well. However, it is important to notice that apart
from Bt toxins, a number of other factors, for example different weather conditions,
conventional crop varieties with different levels of resistance and the application of
insecticides, are likely to change the nutritional quality, developmental rates, survival and
abundance of hosts with consequences for associated parasitoids like C. sonorensis (e.g.
Isenhour & Wiseman, 1989; Pair et al., 1986). Thus adverse effects have to be put in context
with the natural variation in the field and the influence of common practice, before a
conclusion about safety of Bt maize can be drawn.
The results presented here are a first step in the risk assessment of Bt maize for
C. sonorensis developing in hosts suffering from Bt toxin. This study shows a possible hazard
identified in a first tier experiment. The next stage would be to conduct higher tier tests
(including parameters like behaviour and population dynamics of parasitoids and hosts, but
also predators), which allow a more accurate quantification of risk. Such higher tier testing
122
would also allow trigger values and endpoints to be determined for future first tier tests
(Poppy, 2003). If we know at what point a change for example in development time
influences the population dynamics of parasitoid/host interactions at a more complex spatial
scale, then first tier studies like the one we present here will deliver crucial information for
the development of conceptual models in GM risk assessment.
Acknowledgements
We are grateful to the European Science Foundation for funding our research within the
AIGM programme. We also thank the University of Southampton for providing all materials
and locations, Monsanto for maize seeds, Syngenta UK for initial S. littoralis culture, Cristina
Tamo and Ted Turlings for parasitoid cocoons and Natalie Ferry as well as Andreas Lang for
valuable advice.
References
Bourguet, D., Chaufaux, J., Micoud, A., Delos, M., Naibo, B., Bombarde, F., Marque, G.,
Eychenne, N. & Pagliari, C. 2002: Ostrinia nubilalis parasitism and the field abundance
of non-target insects in transgenic Bacillus thuringiensis corn (Zea mays). Environ.
Biosafety Res. 1: 49–60.
Entomol. 27: 441–447.
BioControl 48: 611–636.
Fritzsche Hoballah, M.E. & Turlings, T.C.J. 2001: Experimental evidence that plants under
caterpillar attack may benefit from attracting parasitoids. Evol. Ecol. Res. 3: 553–565.
Hill, D.S. 1983: Spodoptera littoralis. In: Agricultural Insect Pests of the Tropics and their
Control, ed. Hill, D.S., Cambridge University Press, Cambridge: 377.
Isenhour, D.J. & Wiseman, B.R. 1989: Parasitism of the fall armyworm (Lepidoptera:
Noctuidae) by Campoletis sonorensis (Hymenoptera: Ichneumonidae) as affected by host
feeding on silks of Zea mays L. cv. Zapalote Chico. Environ. Entomol. 18: 394–397.
Lindgren, P.D. & Noble, L.W. 1972: Preference of Campoletis perdistinctus for certain
noctuid larvae. J. Econ. Entomol. 65: 104–107.
Lindgren, P.D., Guerra, R.J., Nickelsen, J.W. & White, C. 1970: Hosts and host-age
preference of Campoletis perdistinctus. J. Econ. Entomol. 63: 518–522.
Pair, S.D., Wiseman, B.R. & Sparks, A.N. 1986: Influence of four corn cultivars on fall
armyworm (Lepidoptera: Noctuidae) establishment and parasitization. Fla. Entomol. 69:
566–570.
Poppy, G.M. 2003: The use of ecological endpoints and other tools from ecological risk
assessment to create a more conceptual framework for assessing the environmental risk
of GM plants. Proceedings of the BCPC International Congress – Crop Science &
Technology, 10–12 November, Glasgow, UK, Vol. 2: 1159–1166.
Salama, H.S., Sharaby, A. & Ragaei, M. 1983: Chemical changes in the haemolymph of
Spodoptera littoralis (Lepidoptera: Noctuidae) as affected by Bacillus thuringiensis.
Entomophaga 28: 331–337.
123
Schnepf, E., Crickmore, N., van Rie, J., Lereclus, D., Baum, J., Feitelson, J., Zeigler, D.R. &
Dean, D.H. 1998: Bacillus thuringiensis and its pesticidal crystal proteins. Microbiol.
Mol. Biol. Rev. 62: 775–806.
Schuler, T.H., Potting, R.P.J., Denholm, I. & Poppy, G.M. 1999: Parasitoid behaviour and Bt
plants. Nature 400: 825–826.
Schuler, T.H., Poppy, G.M. & Denholm, I. 2000: Recommendations for assessing effects of
GM crops on non-target organisms. Proceedings of the BCPC conference – pests &
diseases, 13–16 November, Brighton, UK: 1221–1228.
Vojtech, E., Meissle, M. & Poppy, G.M. (submitted): Effects of Bt maize on the herbivore
Spodoptera littoralis (Lepidoptera: Noctuidae) and the parasitoid Cotesia marginiventris
(Hymenoptera: Braconidae).
Wilson, D.D. & Ridgway, R.L. 1975: Morphology, development, and behavior of the
immature stages of the parasitoid Campoletis sonorensis (Hymenoptera:
Ichneumonidae). Ann. Entomol. Soc. Am. 68: 191–196.
124
pp. 125-130
Production of Cry1Ab toxin in E. coli for standardisation of insect
bioassays
Nguyen Thu Hang1, Thomas Meise2, Gustav-Adolf Langenbruch2, Johannes A. Jehle1
1
Agricultural Service Center Palatinate, Department of Phytopathology, Laboratory for
Biotechnological Crop Protection, Breitenweg 71, D-67435 Neustadt/Weinstr., Germany (Email: [email protected]); 2Federal Biological Research Centre for Agriculture and
Forestry, Institute for Biological Control, Heinrichstr. 243, D-64287 Darmstadt, Germany
Abstract: Standardized bioassays are a prerequisite for any monitoring programme of the effect of
transgenic Bt corn on target and non-target insects. A key aspect of these investigations is the quality
of Cry1Ab toxin used in the bioassays. In order to provide a uniform quality of Cry1Ab toxin, we
expressed and purified the cry1Ab gene in E. coli and analysed the expression product by SDS gel
electrophoresis and Western blot. Two forms of Cry1Ab were produced, the Cry1Ab protoxin (~130
kDa) and its trypsinized form (~ 62-67 kDa). Different batches of proteins were produced and tested.
In bioassays using the European corn borer (Ostrinia nubilalis) a batch-to-batch difference in the
specific activities of different toxin production charges was observed.
Key words: Cry1Ab toxin, heterologous expression, toxin activity, bioassays, O. nubilalis
Introduction
The use of Bt corn expressing the Cry1Ab toxin of Bacillus thuringiensis has raised many
concerns on the selection of Bt resistance in insect populations. Therefore, susceptibility
testing of field collected populations of the target pest will be an important issue for
monitoring. A key aspect of the monitoring will be the availability of standardized qualities of
Cry1Ab toxins for bioassays. We expressed and purified the cry1Ab gene in Echerichia coli
and analysed the expression product by SDS gel electrophoresis and Western blot. The
trypsinized Cry1Ab protein (~64 - 67 kDa) was produced from the solubilized Cry1Ab
protoxin (~130 kDa). The production of different toxin batches and their testing in bioassays
using the European corn borer (Ostrinia nubilalis) is described in this contribution.
Bacterial strain and plasmid
The plasmid pMP encoding the open reading frame of Cry1Ab of Bacillus thuringiensis HD-1
was cloned in E. coli HB101 as host bacteria (Masson et al., 1998).
Bacterial expression and purification of Cry1Ab protoxin
E. coli cells containing the pMP plasmid were grown at 37°C in 500 ml Terrific Broth (TB)
medium with ampicillin (100 µg/ml) using a 2-L Erlenmeyer flask with shaking at 220 rpm
(Edmund Bühler SM-30 CONTROL, TH30). After 2-3 days, the cells produced protein
inclusion bodies. The cells were harvested by centrifugation in a GSA-Rotor (Sorvall) at 7000
rpm for 10 min at 4 °C. The cell pellet was homogenised using a French press cell (batch J1).
Alternatively, the cells enzymatically lysed using lysozym (batch J3 and J4). For enzymatic
125
126
lysis, the cell pellets were weighted and frozen at –20 °C for at least 30 minutes. After resuspension in cell lysis buffer (50 mM Tris/HCl, 5 mM EDTA, 100 mM NaCl) (3 ml per
gram E. coli), 800 µg lysozym per gram E. coli was added to the bacterial suspension and
incubated at room temperature for 20 min. Then, 4 mg of deoxycholic acid (sodium salt) were
added per gram E. coli for cell disintegration. After 10 minutes incubation at 37 °C a DNase I
treatment (200 µg DNase/1 g E. coli) followed for 30 minutes at 37 °C. Finally, the
suspension was cooled down on ice and sonicated for 30 sec at full power (Sonorex RK
255H).
The suspension of protein inclusion bodies was centrifuged at 8000 rpm in the GSARotor for 20 minutes at 4°C. The pellet was resuspended using a tick homogenizer and
washed 3 times with washing buffer (20 mM TrisHCl, pH 7,5; 1M NaCl: 1% Triton X-100),
PBS (10mM Na2HPO4/KH2PO4 pH 7.4; 0.8% (w/v) NaCl). The white inclusion body pellet
was harvested by centrifugation at 8000 rpm for 10 minutes at 4°C.
The insoluble protoxin inclusion bodies were resuspended in solubilization buffer (50
mM CAPS (cyclohexylaminopropane sulfonic acid) pH 10.5 and 0.25% -mercaptoethanol)
and incubated at 37°C for 2 hours. The 130 kDa protoxin Cry1Ab was obtained in supernatant
after centrifugation at 8000 rpm (GSA-Rotor) for 20 minutes, at 4°C. The soluble protoxin
was ultra-filtrated by using the Millipore regenerated cellulose membrane (NMWL 100,000).
Trypsinisation of Cry1Ab
The solubilized Cry1Ab protoxin was treated with 1 mg/ml trypsin (TPCK treated from
Bovine Pancreas) over night at room temperature and then ultra-filtrated by using Millipore
polyethersulfone membrane (NMWL: 50,000).
SDS gel electrophoresis and Western blotting
Protoxin and trypsinized Cry1Ab toxin were analysed by SDS/PAGE. Probes were dissolved
in 4x concentrated SDS-sample buffer and heat to 100°C for 10 min before being load on
10% (w/v) polyacrylamide gels and were subsequently stained with Coomassie Brilliant Blue.
Polyclonal antiserum against Cry1Ab protein was raised in chicken egg yolk. Immunoblotting
on nitrocellulose membrane was performed using chicken antiserum (1:6000) and peroxidaseconjugated AffiniPure Donkey Anti-Chicken IgY++ (Jackson ImmunoResearch) (1:5000).
Insect Bioassays
The stability of the biological activity of the proteins at storage temperatures was tested in
insect bioassays during 4-7 months. Batch J1 was stored at +4°C and – 20°C, respectively.
The batches J3 and J4 were stored at –20°C. Bioassays were performed with first instar
Ostrinia nubilalis at an age of 3 - 24h after hatching. 1 ml of artificial diet was dispersed in
every well of the bioassay tray (Bio-ba-128, Color-Dec Italy) resulting in a surface area of
1,77  0,08 cm2/well. Trypsinized Cry1Ab toxins were diluted to 7 different concentrations.
The dilutions used for bioassays were stored at 4°C and were pipetted on the surface of the
diet. After the diet had dried, one larvae was placed in each well. 32 larvae were used per
dilution, 64 larvae were used as control. The assays were conducted at 25°C and after 7 days
the mortality was assessed. Larvae were considered as dead when they did not move anymore
after stimulation with a hair brush. The statistics were carried out with SAS Procedure
NLMIXED. The LC50 and the 95% confidence interval were calculated.
127
Cry1Ab expression in E. coli
Recombinant Cry1Ab protoxin was expressed from plasmid pMP (Fig. 1) in E. coli HB101.
After 48 hours of incubation protein expression became visible in light microscopy by the
formation of protein inclusion bodies (Fig. 2).
cry1Ab
NdeI
NdeI
NdeI
cry1Ab
pMP
AmpR
NdeI
pMP
ori
AmpR
ori
Figure 1. Expression plasmid pMP (Masson
et al., 1989).
Figure 2. E. coli HB101 expressing
Cry1Ab protein inclusion bodies (black
dots)
Protein characterization using SDS PAGE and Western blotting
Protein size and purity of the prepared Cry1Ab preparations were analyzed using SDS
polyacrylamid gel electrophoresis and Western blot analysis (Fig. 3a and 3b). It was observed
that all different preparation methods resulted in a protein band of initially 67 kDa. Storage of
preparation J1 at 4°C resulted in a further degradation to a final size of about 62 kDa
(compare Fig. 3a, lane J1). Further comparative analyses indicated, however, that this
degradation was not a result of the preparation method but that it was most likely caused by a
residual trypsin activity in the Cry1Ab preparations. This result is conform with the
observation that excessive trypsin addition during toxin activation only results in the smaller
band of 62 kDa, but not in the 67 kDa or intermediate band (data not shown).
J4
J3
a
)
J4
J1
J3
J1
b
)
6
7
k
D
a
J
4 J
3 J
1
J
4 J
3J
1
J
1
Figure 3. (a) SDS PAGE and (b) Western blot with Anti-Cry1Ab. Lanes J1, J3, J4 represent
different production batches, where different cell disintegration methods were used. J1 = French
Press, J3/J4 = Enzymatic method. J3 and J4 were stored at –20°C, J1 was stored at +4°C.
128
Figure 4. LC50 determination and storage stability of the dilutions (stored at +4°C) of Cry1Ab
preparations J1 and J3 in bioassays using neonate O. nubilalis. Vertical bars indicate 95%
confidence limit.
Testing the biological activity and stability of the expressed Cry1Ab
The activity and stability of the Cry1Ab preparations were tested in bioassays using neonate
O. nubilalis. For the bioassays the larvae were kept on artificial media, which was overlaid
with different concentrations of the protein preparations. As shown in Fig. 4, the initial LC50
of preparation J1 and J3 were 53 ng/cm2 and 14 ng/cm2, respectively, and thus showed
significant differences in their biological activities. The activity of batch J4 did not differ from
batch J3 (data not shown). The biochemical reason of this difference could not be figured out,
but could be due to the differing cell preparation procedures. Though the bioassays revealed
some variation, no significant change in toxin activity could be observed during 142 days of
storage at 4°C. Storage at -20°C also did not reveal a decline in activity for at least 214 days
(Fig. 5).
The observed LC50 values are in the same order of magnitude as susceptibility tests
revealed in other labs (Gonzales-Nunes et al., 2000; Chaufaux et al., 2001). However, an
accurate comparison between these bioassays is difficult, because methods of protein
preparation and quantification, bioassay conditions and overall vitality of larvae vary from lab
to lab. This underscores that for a long-term susceptibility monitoring a standardized
methodology using standardized toxins is essential.
129
Figure 5. LC50 determination and storage stability (stored at -20°C) of Cry1Ab preparations J1
in bioassays using neonate O. nubilalis. Vertical bars indicate 95% confidence limit.
Conclusions
Expression of Cry1Ab in E. coli revealed a batch-to-batch difference in its biological activity.
The reason for this difference might be caused by different preparation methods. The biological
activity of the toxins was stable for at least 4 ½ months at +4°C and 7 months at –20°C. The
variability between different bioassays contributed more to the variation of the LC50
determinations than the possible decline of toxin activity.
Acknowledgements
We are grateful to William Moar (Auburn University, USA) for providing plasmid pMP. The
help of Ruud de Maagd (Wageningen University, Nl) and Manuela Berlinghof (DLR
Rheinpfalz) for Cry1Ab protein preparation from E. coli is acknowledged. This study was
supported by the German Federal Ministry of Education and Research (BMBF).
References
Masson, L., Prefontaine, G., Peloquin, L., Lau, P.C. & Brousseau, R. 1989: Comperative
analysis of the individual protoxin components in P1 crystals of Bacillus thuringiensis
subsp. kurstaki isolates NRD-12 and HD-1. Biochem. J. 269: 507-512.
Chaufaux, J., Seguin, M., Swanson J.J., Bourguet, D. & Siegfried, B.D. 2001: Chronic exposure
of the European corn borer (Lepidoptera: Crambidae) to Cry1Ab Bacillus thuringiensis
toxin. J. Econ. Entomol. 94: 1564-1570.
130
Gonzáles-Núnez, M., Ortego, F. & Castanera, P. 2000: Susceptibility of Spanish populations of
the corn borers Sesamia nonagrioides (Lepidoptera: Noctuidae) and Ostrinia nubilalis
(Lepidoptera: Crambidae) to a Bacillus thuringiensis endotoxin. J. Econ. Entomol. 93: 459463.
pp.131-136
No effects of Bt maize on the development of Orius majusculus
Xavier Pons, Belén Lumbierres, Carmen López, Ramon Albajes
Universitat de Lleida, ntre UdL-IRTA, Rovira Roure 191, 25198 Lleida, Spain (E-mail:
[email protected])
Abstract: Orius spp. are the most abundant predators in maize fields, including Bt-transgenic ones, in
the north-east of the Iberian Peninsula. The effect of Bt-transgenic maize (cv. Compa, Event 176), that
expresses the Cry1Ab toxin of Bacillus thuringiensis, is being studied on Orius majusculus in the
laboratory. Nymphal mortality, developmental time, sex ratio and the weight and size of the teneral
adults were measured when O. majusculus nymphs were fed with several combinations of leaf and
pollen of transgenic Bt maize and the isogenic maize hybrids (cv. Compa and Dracma respectively).
No significant differences between treatments were found for any of the biological parameters
measured.
Key words: Bt maize, non-target insects, predators, Orius majusculus
Introduction
Genetically modified maize containing δ-endotoxins of Bacillus thuringiensis Berliner var.
kurstaki (Bt maize) is one of the most advanced methods for controlling the corn borers
Ostrinia nubilalis Hübner (Lep., Crambidae) and Sesamia nonagrioides Lèfebvre (Lep.,
Noctuidae). As it has been shown for other mechanisms that confer resistance to
phytophagous insect in plants (Price, 1986), the possible side-effects of Bt maize on pest
natural enemies need to be known in order to evaluate the potential of this method in the
framework of IPM.
In 2003, a total of 32,000 ha of Bt maize were grown in Spain, most of them located in
Catalonia and Aragon (NE Iberian Peninsula). Since it was registered for the first time in
1998 the Bt maize surface has increased slowly but continuously. Up to 2003 only cultivars
containing the Event 176 were authorised and only one was commercially sown (cv. Compa,
Syngenta Seeds). However, in 2003, five new cultivars (four containing the Event MON810
and one containing the Event 176) have been authorised and the area devoted to Bt maize is
expected to increase in the coming years. All these cultivars express the Cry1Ab toxin, which
is considered very selective in its action, so the effects on non-target arthropods, including
predators, are expected to be minimal. However, the continuous expression of the toxins in
some plant tissues throughout the growing season is a new form of exposure to predators that
did not arise with the traditional use of B. thuringiensis insecticides.
Orius sp. is the main predator occurring in maize fields of the Iberian Peninsula,
including Bt-transgenic ones (Albajes et al., 2003; unpublished data). Orius sp. is also very
frequent in other crops determining the agricultural regional landscape (alfalfa, winter cereals,
sorghum, orchards, etc.). Therefore, possible ecological effects on Orius sp. due to the
cultivation of transgenic maize may have consequences not only in maize fields. Orius spp.
are polyphagous predators in maize fields that consume aphids, leafhoppers, thrips, mites and
eggs and young larvae of Lepidoptera (Lattin, 1999). They are also pollen consumers (Lattin,
1999) and can take supplemental substances by sucking on the maize tissues (Armer et al.,
1998). Therefore, Bt toxin can be ingested by Orius spp. through any of these three ways. The
131
132
effects of transgenic Bt maize on Orius spp. have been previously investigated in the
laboratory by feeding the predator with pollen or prey fed with transgenic or non-transgenic
plant or artificial diet (Pilcher et al., 1997; Zwahlen et al., 2000; Al-Deeb et al., 2001).
However, no studies included the green plant tissue directly as a source of food.
The aim of the present work was to evaluate the effects of Bt plants on the development,
adult size and sex ratio of Orius majusculus (Reuter), the most abundant predator species on
maize (Albajes et al., 2003). Insects reared with combinations of leaf and pollen from Bttransgenic and non-transgenic maize were compared.
Material and Methods
The experiments were conducted in a climatic chamber at 25±1 ºC constant temperature, 35
% of RH, 2000 lux and a photoperiod of 16:8 h (L:D). Two commercial cultivars of maize
were used in the experiments, the transgenic one (Compa CB®, Syngenta Seeds; Bt+
hereinafter) derived from the Event 176, and the corresponding isogenic cultivar (Dracma®,
Syngenta Seeds; Bt– hereinafter). The two maize cultivars were grown in the glasshouse in
clay pots (16 cm diameter and 15 cm height) until the plants had 6-8 completely expanded
leaves. Pollen was collected during anthesis from the Bt+ and Bt– plants in commercial fields
sown in Almenar (30 km to the west of Lleida) by shaking tassels into paper bags. Just after
collection, pollen was carried to the laboratory, meshed to remove anthers and contaminants,
and then poured into plastic vials and frozen at -20 ºC.
The insects for the experiment were obtained from a culture of O. majusculus maintained
in the UdL-IRTA laboratory at the same environmental conditions as the experiment. The
culture was started from a population collected in non-transgenic maize fields in 2001 and
renewed each growing season with new individuals collected in non-transgenic maize fields.
O. majusculus is reared in square glass candy containers of two litres capacity, with their
mouth covered with a mesh to prevent insects escaping or entering. Inside one container only
insects of the same age were maintained in order to avoid cannibalism. Bugs were fed with
eggs of Ephestia kuehniella Zeller as a prey. In each container, a glass vial of 5 ml capacity
filled with water and topped with a dental wick to increase the RH and as a source of water
and two pods of fresh green beans (Phaseolus vulgaris L.) as an alternative food resource and
as an egg-laying substrate were included. The green beans and water were replaced and E.
kuehniella eggs were added twice a week.
Developmental time and nymphal mortality
Green beans with O. majusculus eggs were isolated and, when hatched, the nymphs were used
for the experiment. Less than 24-hour-old nymphs were placed individually in cylindrical
plastic containers (53 mm diameter, 32 mm height) with a piece of maize leaf (about 4 cm
long; only the fifth or sixth leaves were used) and pollen. In each container, only a small
quantity of E. kuehniella eggs were added, in order to ensure the development of the bugs but
to force them to feed on the leaf and pollen. The experiment consisted in feeding 10
individuals of O. majusculus, from the day of egg hatching to adult emergence, with the 4
combinations of leaf and pollen from the Bt+ and Bt– cultivars. The containers were monitored
twice a week until the nymphs reached the fourth instar, and then inspected every day until
moulting to adult. The number of days between egg hatching and adult occurrence was
considered the nymphal developmental time. Nymphal mortality during the nymphal
development was also recorded. Four sets of this experiment were done.
133
Adult size and sex ratio
Candy containers like those described above were used in the experiment. Each container had
one maize plant (2-leaf stage), pollen, some E. kuehniella eggs and a water glass vial topped
with a dental wick. Each container was prepared with the four possible combinations of leaf
material and pollen Bt+ and Bt–. In each container, two or three pods of fresh green beans with
eggs of O. majusculus were included, and nymphs hatched from eggs allowed to develop to
adults in the container. Maize plants were replaced once a week and pollen, E. kuehniella
eggs and water were supplied twice a week. When the first adults occurred the containers
were monitored every day and adults were collected and frozen at -20 ºC. Three days later,
they were sexed and weighted, and the body length and maximum pronotum width were
measured. Weight was determined with a Mettler M3® balance (precision 1 μg) (Mettler
Instruments Ag, Greifensee, Switzerland), and body length and pronotum width were
measured using a Leica Qwin V 2.1 image analyser (Leica Imaging Systems, Cambridge,
UK) connected to a Leica MS5 stereoscopic microscope (Leica Microsystems, Heerbrugg,
Switzerland).
Data analysis
Data were analysed using the SAS statistical package (SAS Institute, 2000). The data of
nymphal mortality and developmental time were analysed by a three-way (leaf, pollen and
sex, and using the sets as blocks) ANOVA using the GLM procedure. Percentage values were
transformed to arcsin (x/100)1/2 before analysis. The data on sex ratio were analysed using a χ2
test. The data of weight and size of teneral adults were analysed by a three-way (leaf, pollen
and sex) ANOVA using the GLM procedure.
Developmental time
The time that O. majusculus required to reach the adult stage (around 14 days. Figure 1) was
longer than those reported in other experiments in which the food was only E. kuehniella eggs
(Fisher et al., 1992; Alauzet et al., 1990). This could suggest that in our experiment the main
sources of food were the provided maize leaf and the pollen. It has been reported that when
Orius insidiosus (Say) and Orius tristicolor (White) are fed arthropods they develop faster
than when they are provided with plant tissue (Naranjo & Gibson, 1996). However, the
developmental time was very similar to that reported by Zwahlen et al. (2000) when O.
majusculus was fed with Anaphothrips obscurus (Müller) (Thysanoptera: Thripidae) in a
tritrophic experiment with Bt maize. The reason for this similarity could be the lower mean
temperature they used (22 ºC) and the supply of eggs of E. kuehniella in our studies.
Values of developmental time were very similar for all plant and pollen combinations
and no effects of the plant (F = 1.72, P = 0.19), the pollen (F = 0.66, P = 0.42) or the
plant*pollen interaction (F = 0.55, P = 0.46) were found (Figure 1). This lack of effect also
occurred when the sex was considered and plant*pollen*sex interaction was non-significant
(F = 0.16, P = 0.69). Our results agree with those of Pilcher et al. (1997), who did not find an
acute toxic effect of transgenic Bt pollen on O. insidiosus. Zwahlen et al. (2000) also found
no differences in developmental time of O. majusculus when A. obscurus fed on Bt and nonBt maize was offered as prey. Al-Deeb et al. (2001) compared the number of days to
adulthood of males and females of O. insidiosus feeding on O. nubilalis fed on Bt and non-Bt
diet and reported that no significant differences occurred for either sex.
134
Developmental time
16
days
14
12
10
Bt+ Bt+
Bt+ Bt–
Bt– Bt+
Bt– Bt–
Leaf x Pollen
Figure 1. Developmental time (± se) required by O. majusculus fed with different
combinations of leaf and pollen from transgenic (Bt+, cv. Compa) and non-transgenic (Bt–, cv.
Dracma) maize cultivars.
Nymphal mortality
Although the percentages of nymphal mortality varied among the different plant and pollen
combinations, no effects of the plant (F = 0.03, P = 0.87), the pollen (F = 0.13, P = 0.72) or
the plant*pollen interaction (F = 0.62, P = 0.45) were found (Figure 2).
Survival of O. insidiosus remained unchanged when it was fed on transgenic Bt or nontransgenic pollen (Pilcher et al., 1997). Zwahlen et al. (2000) found no significant differences
between the mortality of O. majusculus fed on thrips reared on Bt transgenic or non-Bt plants.
Furthermore, Al-Deeb et al. (2001) found no significant differences in mortality when
nymphs of O. insidiosus were fed on Bt or non-Bt reared European corn borer (O. nubilalis)
larvae.
Nymphal mortality
%
40
35
30
25
20
15
10
5
0
Bt+ Bt+
Bt+ Bt–
Bt– Bt+
Bt– Bt–
Leaf x Pollen
Figure 2. Nymphal mortality (± se) of O. majusculus fed with different combinations of leaf
and pollen from transgenic (Bt+, cv. Compa) and non-transgenic (Bt–, cv. Dracma) maize
cultivars.
135
Sex ratio
The proportion of females and males did not change statistically (χ2 = 6.82, P = 0.08) when
O. majusculus was fed on Bt or non-Bt leaves and on Bt and non-Bt pollen, and sex ratio can
be considered as 1:1.
Weight and size of teneral adults
Mean values of weight, body length and pronotum width of O. majusculus obtained in the
experimental conditions can be seen in Table 1. There were significant differences between
the weight (F = 233.08, P < 0.0001), length (F = 99.19, P < 0.0001) and width (F = 332.88, P
< 0.0001) of females and males. However, no effects of the plant (weight: F = 0.01, P = 0.95;
length: F = 0.00, P = 0.98; width: F = 1.35, P = 0.25), the pollen (weight: F = 1.46, P = 0.23;
length: F = 0.73, P = 0.39; width: F = 0.07, P = 0.79) or the plant*pollen interaction (weight:
F = 3.34, P = 0.07; length: F = 0.75, P = 0.39; width: F = 0.00, P = 0.96) were found. The
triple plant*pollen*sex interaction was also non-significant.
Our results agree with those of Al-Deeb et al. (2001), who reported that despite the
differences in weight and length between males and females, there was no effect of the Bt on
either sex when O. insidiosus was fed with larvae of the European corn borer reared on a Bt
diet.
Table 1. Mean values (± se) of weight, body length and pronotum width of O. majusculus
females and males reared on combinations of leaf and pollen from transgenic (Bt+, cv.
Compa) and non-transgenic (Bt–, cv. Dracma) maize cultivars.
Leaf x pollen
Sex
n
Weight (μg)
Length (μm)
Width (μm)
Bt+ x Bt+
Females
Males
27
44
466.3 ± 12.2
365.1 ± 6.2
2314.3 ± 20.9
2139.4 ± 16.1
893.8 ± 13.0
804.7 ± 5.5
–
Females
Males
31
32
489.0 ± 13.2
388.6 ± 10.1
2322.7 ± 20.2
2206.8 ± 38.5
886.2 ± 7.3
822.1 ± 6.3
Females
Males
48
41
489.1 ± 11.6
371.3 ± 8.2
2325.8 ± 34.0
2160.0 ± 30.7
883.1 ± 7.7
804.0 ± 5.3
Females
Males
21
38
496.6 ± 14.0
354.5 ± 8.3
2363.4 ± 25.1
2122.0 ± 19.4
895.7 ± 7.4
794.5 ± 6.5
Bt+ x Bt
–
Bt x Bt+
–
Bt x Bt
–
Summarising, leaf material and pollen from transgenic Bt maize does not seem to cause
any negative effect on the nymphal mortality, developmental time, sex ratio and size of the
adults of O. majusculus. Despite the lack of negative effects in the parameters studied before
reproduction, additional studies are being carried out to determine possible effects on
fecundity.
Acknowledgments
We thank the CICYT and MCYT who has funded this research, project numbers AGF990782 and AGL2002-204.
136
References
Alauzet, C., Bouyjou, B., Dargagnon, D. & Hatte, M. 1990: Mise au point d'un elevage de
masse d'Orius majusculus Rt. (Het: Anthocoridae). Bull. SROP/WPRS XIII/2: 118-122.
Albajes, R., López, C. & Pons, X. 2003: Predatory fauna in corn fields and response to
imidacloprid seed-treatment. J. Econ. Entomol. 96: 1805-1813.
Al-Deeb, M.A., Wilde, G.E. & Higgins, R.A. 2001: No effect of Bacillus thuringiensis corn
Armer, C.A., Wiedenmann, R.N. & Bush, D.R. 1998: Plant feeding site selection on soybean
by the facultative phytophagous predator Orius insidiosus. Entomol. Exp. Appl. 86: 109118.
Fisher, S., Linder, C.H. & Freuler, J. 1992: Biologie et utilisation de la punaise Orius
majusculus Rt. (Het: Anthocoridae) dans la lutte contre les thrips Frankliniella
occidentalis Perg. et Yhrips tabaci Lind., en serre. Revue. Suisse Vitic. Arboric. Hortic.
24: 119-127.
Lattin, J.D. 1999: Bionomics of the Anthocoridae. Annu. Rev. Entomol. 44: 207-231.
Naranjo, S.E. & Gibson, R.L. 1996: Phytophagy in predaceous Heteroptera: effects on life
history and population dynamics. In: Zoophytophagous Heteroptera: Implications for
Life History and Integrated Pest Management, eds. Alomar and Wiedenmann. Thomas
Say Publications in Entomology, Entomological Society of America: 57-93.
Price, P.W. 1986: Ecological aspects of host plant resistance and biological control.
Interactions among three trophic levels. In: Interactions of plant resistance and
parasitoids and predators of insects, eds. Boethel and Eikenbary. Ellis Horwood,
Chichester, UK: 11-30.
SAS Institute. 2000: SAS/STAT User’s Guide, Version 8. SAS Institute, Cary, NC.
Zwahlen, C., Nentwig, W., Bigler, F. & Hilbeck, A. 2000: Tritrophic interactions of
transgenic Bacillus thuringiensis corn, Anaphothrips obscurus (Thysanoptera:
Thripidae), and the predator Orius majusculus (Heteroptera: Anthocoridae). Environ.
Entomol. 29: 846-850.
pp. 137-142
Impact of growing Bt-maize on cicadas: Diversity, abundance and
methods
Stefan Rauschen, Jörg Eckert, Achim Gathmann, Ingolf Schuphan
RWTH Aachen University, Chair of Biology V (Ecology, Ecotoxicology, Ecochemistry),
Worringerweg 1, D-52056 Aachen, Germany (E-mail: [email protected])
Abstract: One concern of growing transgenic maize expressing Bacillus thuringiens (Bt) toxin is the
possible negative impact on the diversity and abundance of non-target organisms. The presented study
focuses on the potential side effects of genetically modified Bt-corn on cicadas. Feeding on the
contents of plant cells they can be assumed to ingest Bt-toxin when their diet encompasses Bt-plants.
Therefore they may be used as indicator organisms in assessing the side effects of Bt-maize on
herbivores. The field experiment comprised three treatments in a randomized block design with 8
replications each: (i) Bt-maize (MON 810, Novelis) expressing the recombinant Cry1Ab toxin, (ii) the
isogenic variety (Nobilis) with chemical insecticide application (Baythroid) and (iii) the isogenic
variety without insecticide treatment as a control. Several methods were used to find the best means of
monitoring the diversity and abundance of cicadas. In total, we found and identified 5 different cicada
species. The most abundant species was Zygidinia scutellaris. We found no significant differences in
the numbers of Z. scutellaris feeding on Bt-maize and those on the control field. In contrast, the
number of cicadas was decreased significantly due to insecticide treatment.
Key words: Bt, herbivores, monitoring, cicadas, non-target organisms, insecticide effect
Introduction
Genetically modified maize expressing a toxin gene derived from Bacillus thuringiensis
yields a high level of resistance against the European Corn Borer Ostrinia nubilalis (Hübner).
Other herbivores have been shown to ingest the Bt-toxin in varying amounts when feeding on
Bt-plants (Head et al., 2001). Consequential effects of this on predators have also been
examined (Dutton et al., 2002).
Cicadas have been shown to be common in maize fields (Schmitz & Bartsch, 2001). As
they are feeding on the cell sap of different host plants, they can be assumed to ingest Bt-toxin
when their diet encompasses Bt-plants. This assumption is supported by the results of a
preliminary ELISA test (unpublished data). Therefore, this work focuses on cicadas as
potential monitoring organisms. Two questions are adressed in this study:
(i) Which is the best method of assessing the abundance of cicadas? To answer this
question, 4 different methods of quantitating insects are evaluated in their applicability to
investigations in species abundance and diversity of cicadas: visual assessments, yellow traps,
sticky traps and sweep net catchings. This leads to a recommendation of means for monitoring
this arthropod group.
(ii) Does the growing of Bt-maize or the use of insecticides affect the diversity and
abundance of cicadas?
137
138
Experimental design
The experimental site consisted of two maize fields approximately 500m apart, situated near
Bonn (Germany). One field measured 182m by 248m and was divided into 15 plots, the other
measured 178m by 186m and contained another 9 plots. The size of each plot was 0.25 ha.
Three treatments were tested in a randomized block design with 8 replications each: (i) Btmaize (MON 810, Novelis) expressing the recombinant Cry1Ab toxin, (ii) isogenic variety
(Nobilis) with chemical insecticide application on 11th of July 2003 (Baythroid 50, 750ml/ha
in a water amount of 200 l/ha) and (iii) the isogenic variety without insecticide as a control.
The abundance of cicadas was measured using 4 different methods, between 4th and 13th of
August 2003.
Visual assessment
Four plants per plot were assessed on the 4th of August 2003 . Different parts of the plants –
leaves, stipe, cobs and panicle – were scored individually. Plants were not damaged or altered
in any way, nor were animals removed from them during the assessment.
Yellow traps
One yellow trap (340mm x 260mm) was placed in the middle of each plot at a height of
approximately 1.2 metres. Traps were filled with water and a small amount of detergent.
Insects were sampled in yellow traps between 8th and 11th of August 2003.
Sticky traps
Sticky traps were constructed from bamboo sticks, a rectangular wire frame and a clear plastic
folder (300mm x 240mm) that was slipped over the frame. The surfaces of the folder were
covered with insect glue (Fa. Temmen, Hattersheim, Germany). One glue trap was placed
within 5 metres distance from each yellow trap, reaching a height of approximately 1 metre
above ground. Traps were used from 8th to 11th of August 2003.
Sweep nets
Catchings with a sweep net (diameter: 40cm, mesh width: 1,5mm) were carried out four times
in each plot from 6th to 13th of August 2003. Beginning approximately 5m into a given row of
maize plants, thirty steps were taken while holding the sweep net at a height of about 90cm
above ground.
Statistics
As the data achieved are not Gaussian (normal) variables, a Kruskal Wallis Test was
performed. For this, the program SPSS 11 for Windows (SPPSS Inc 2001) was used.
Species diversity and abundance
Five taxa of cicadas were identified: Zyginidia scutellaris (H.-S.), Macrosteles spec.,
Psammotettix alienus (Dahlb.), Empoasca pteridis (Dahlb.) and Laodelphax striatella
(Fallen). A total of 2023 individuals were caught with different methods, with Z. scutellaris
accounting for over 94% (Tab. 1).
139
Table 1. Species and number of individuals caught with different methods (total numbers for
each trap type and species as well as the percentage of all individuals are shown).
species
yellow traps
400
Zyginidia scutellaris
2
Macrosteles spec.
1
Psammotettix alienus
26
Empoasca pteridis
1
Laodelphax striatella
430
total
sweep nets
427
3
5
5
1
441
sticky traps total % of individuals
1078
1905
94,17
24
29
1,43
39
45
2,22
11
42
2,08
0
2
0,1
1152
2023
100
Effects of Bt maize and insecticide
Only data of Z. scutellaris were sufficient for statistical analysis, because all others species
were extremly rare. All methods used showed similar abundance patterns for Z. scutellaris
(Fig. 1). A significant difference between Bt-treatment and the isogenic variety was not
detected, but the abundance of Z. scutellaris was significantly affected by the insecticide
treatment. The results show that Bt maize seems to pose no environmental threat to Z.
scutellaris, in contrast to the application of insecticides. This may hold true for the other
cicada species found in this experiment, but is inconclusive due to their very low abundance.
Our general observations concur with the results of other studies that investigated the effects
of Bt-maize on non target arthropods in comparison to the effects of insecticide treatments under
field conditions: Bourguet et al. (2002) enumerated several non-target species, including the
aphid Metapolophium dirhodum (Walker), the predator bug Orius insidiosus (Say), the syrphid
Syrphus corollae (Meigen), the ladybird Coccinella septempunctata (L.) and the lacewing
Chrysoperla carnea (Stephens) at four different locations in France. These researchers found no
significant differences in the abundance of insects on Bt-maize fields and plots with the isogenic
variety used.
Dively & Rose (2003) examined the effects of Bt transgenic and conventional insecticide
control on non-target enemy communities in sweet corn. The authors discerned 177 taxonomic
groups and 101 recognized families. Only in a few they found some changes in abundance and
distribution pattern, which they attribute to plant-mediated factors and the absence of feeding
injury by target pest species. For all other insect groups and communities, no effect of Bt maize
was found.
The effects of Bt-maize on thrips and their antagonists have been studied by Eckert et al.
(2004). At their research site near Bonn, they found several thysanopteran species, such as
Limothrips cerealium (L.) and Frankliniella intonsa (Tryb.), as well as some predatory bug
species, among these Orius majusculus (Reut.). No effect of the cultivation of Bt-maize was
found, but the reduction of the insect abundance in the insecticide treated parts of the maize
fields was significant.
140
v is u a l a s s e s s m e n ts
25
A
y e llo w tra p s
60
a
a
50
individuals/trap
individuals/plot
20
15
10
5
b
0
a
B
a
40
30
20
b
10
0
Bt
IN S
IS O
Bt
s tic k y tra p s
180
160
IS O
s w e e p n e ts
a
70
a
C
IN S
60
D
120
individuals/plot
individuals/trap
140
a
100
80
60
40
b
20
50
40
a
30
20
b
10
0
0
Bt
IN S
IS O
Bt
IN S
IS O
Figure 1. Number of Z. scutellaris individuals caught with the different methods and cicadas on
group level for visual assessments. Given are box plots for the three treatments: the horizontal
line gives the median, the box represents 75%, upper and lower dashes 90% of all values,
respectively. The filled circles mark min. and max. values. A: Kruskal-Wallis-test visual
assessments: χ2=16.42, df=2, n=24, p<0.001; B: Yellow traps: χ2=15.64, df=2, n=24, p<0.001; C:
sticky traps χ2=15.71, df=2, n=24, p<0.001; D: sweep nets: χ2=13.96, df=2, n=24, p=0.01;
significantly different groups are indicated with different letters (Dunn’s Test p<0.05). (Bt: Btmaize MON 810, Novelis; INS: isogenic variety with Baythroid treatment; ISO: isogenic
variety without insecticide treatment).
Advantages and drawbacks of the applied methods
Although the results achieved with the different methods are basically the same, different
advantages and drawbacks of each method and its limitations have to be considered.
Preparing, installing and dismantling yellow traps are labour intensive. Sticky traps can be
constructed and installed quickly, but desiccated insects or insects remaining stained with glue
can complicate identification. Sweep net catching and visual assessment can be handled more
easily.
Unwanted species are often caught with yellow traps in fairly high numbers, which requires
time-consuming sorting of individuals. Sticky traps do also catch undesired insect groups, but
cicadas can separately be pealed off. Visual assessment and sweep net catching can be
conveniently adjusted to catch and quantify specifically the cicada species only.
The taxonomic level on which cicadas can be identified differs between the methods. With
all methods except the visual assessments, individuals can be determined down to the species
level. The visual examination of plants yields abundances only on a higher taxonomic level,
because adults and nymph stadiums cannot be easily identified under field conditions.
141
In terms of time required for each method, a first set of results can be obtained from yellow
traps after three to four hours, approximately 2 hours for the sticky traps and roughly one hour
for the sweep net catching and the visual assessment.
Conclusions
Although Bt-toxins have no observable adverse effects on cicadas, these arthropods may
operate as mediators through which the Bt-toxin might influence higher trophic levels. This
influence might not be directly linked to the Cry1Ab toxin, but rather be grounded on a low
prey quality that reduces the fitness of the predators (Groot & Dicke, 2002; Romeis et al.,
2004). Kiss et al. (2002) emphasize the role of leafhoppers in the context of mediation. We
agree with these researchers that the influence of Bt-toxin taken up by cicadas on the
abundance of predators should be the object of further research.
All methods applied in this study can be used for monitoring the species composition and
abundance of cicadas, but a combination of sweep net catchings and sticky traps is recommended
for investigating their diversity and abundance.
With the sweep nets, a first overview of the species that can be found is quickly and easily
obtained. Such preliminary data is useful, as it guides the decision at which species to look in
particular, which in the following will be quantified efficiently with a more robust method, such
as the sticky traps.
Acknowledgements
We would like to thank Herbert Nickel (Department of Ecology of the Institute of Zoology and
Anthropology, University of Göttingen) for kindly identifying the cicada species and the
BMBF for financial support.
References
Bourguet, D.,Chaufaux, J., Micoud, A., Delos, M., Naibo, B., Bombarde, F., Marque, G.,
Eychenne, N. & Pagliari, C. 2002: Ostrinia nubilalis parasitism and the field abundance of
non-target insects in transgenic Bacillus thuringiensis corn (Zea mays). Environ. Biosafety
Res. 1: 49-60
Dively, G.P. & Rose, R. 2003: Effects of Bt transgenic and conventional insecticide control
on the non-target natural enemy community in sweet corn. Proceedings of the 1st
International Symposium on Biological Control of Arthropods. Honolulu, Hawaii, USA.
2002. USDA-Forest Service FHTET-03-05: 265-274.
Dutton, A., Klein, H., Romeis, J. & Bigler, F. 2002: Uptake of Bt-toxin by herbivores feeding on
transgenic maize and consequnces for the predator Chrysoperla carnea. Ecol. Entomol. 27:
441-447.
Eckert, J., Gathmann, A. & Schuphan, I. 2004: Auswirkungen des Anbaus von Bt-Mais auf
Nichtzielorganismen: Thripse und ihre Gegenspieler. Mitt. Dtsch. Ges. Allg. Ent. 14: 439442.
Groot, A.T. & Dicke, M. 2002: Insect-resistant plants in a multi-trophic context. Plant J. 31: 387406
Head, G., Brown, C.R., Groth, M.E. & Duan, J.J. 2001: Cry1Ab protein levels in phytophagous
insects feeding on transgenic corn: implications for secondary exposure risk assessment.
Entomol. Exp. Appl. 99: 37-45.
142
Kiss, J., Szentkirályi, F., Tóth, F. Szénási, Á., Kádár, F., Árpás, K., Szekeres, D. & Edwards,
C.R. 2002: Bt-Corn: Impact on non-targets and adjusting to local IPM systems. In:
Ecological Impact of GMO Dissemination in Agro-Ecosystems (Lelley, Balazs and Tepfer,
eds.), Facultas, Wien: 157-172.
Romeis, J., Dutton, A. & Bigler, F. 2004: Bacillus thuringiensis toxin (Cry1Ab) has no direct
effect on larvae of the green lacewing Chrysoperla carnea (Stephens) (Neuroptera:
Chrysopidae). J. Insect. Physiol. 50: 175-183.
Schmitz, G. & Bartsch, D. 2001: Biozoenotische Untersuchungen in Maisfeldern bei Bonn und
Aachen. Mitt. Dtsch. Ges. Allg. Ent. 13: 615-618.
SPSS Inc 2001: SPSS for Windows 11, Chicago.
pp. 143-146
Impact of genetically modified herbicide resistant maize on the
arthropod fauna
Ioan I. Rosca
University of Agricultural Sciences and Veterinary Medicine-Bucharest, Av. Marasti 59,
71322 Bucharest, Romania (E-mail: [email protected])
Abstract: Agricultural research is engaged increasingly to manipulate genes, among other biological
methods. The evaluation of hazards connected with the release of genetically modified organisms
should consider, among other things, environmental impacts. Possible impact of Roundup Ready
maize on the trophic chains in agrocenosis is of concern to farmers who cannot any longer afford
expensive mechanical or chemical control of weeds, to policy makers who recommend expansion of
maize and to organizations and societies interested in environment conservation. The new technology
of maize cultivation, based on using genetically modified herbicide resistant hybrids has offered
the opportunity to assess possible environmental effects. We have investigated if the herbicide
resistant hybrids have an influence on epigeic fauna captured in pitfall traps or on the fauna present
on maize plants, knowing that the use of pesticides could affect both, the composition and the
abundance of organisms in the agro-ecosystem. The “Sörensen” index of similarity was used to
evaluate differences between maize fields. In a system where we compared the herbicide resistant
maize (Event NK 603) with its isoline as control, we found no differences between the treatments,
concluding that the herbicide treatment, associated with the cultivation of Roundup Ready maize
affects neither key species captured in the pitfall traps nor natural enemies living on the plants.
Key words: Roundup Ready maize, pitfall traps, natural enemies, epigeic fauna,
Introduction
In the USA, but also in the other parts of the world, genetically modified (GM) plants are
authorized to be cultivated on large scale, especially maize, cotton and soybean. Needless to
underline the advantages of cultivating such plants that have resulted in a sheer revolution in
agriculture of many countries. Romania tries to follow the world tendencies respecting the
legal regulations for the cultivation of GM plants.
The new technology of maize cultivation, based on herbicide (glyphosate) resistant GM
hybrids has offered the possibility to assess possible impacts of this new technology on
biodiversity, on arthropod fauna captured in pitfall traps and on natural enemies living on
maize plants (Rosca et al., 2001, 2002).
In order to respond to the concern of people regarding the possible influence of GM
plants on the environment, we have evaluated the potential impact of Roundup® resistant
maize (Event NK 603) on epigeic and plant dwelling arthropods in maize fields.
Experiments were carried-out on the Experimental Didactic Station “Moara Domneasca”,
near Bucharest, on two 1200 m2 fields under conventional farm practices. Roundup Ready®
(Event NK 603) maize was grown on one field and its isoline as a control on the other field.
These two plots were 200 m apart.
143
144
Treatments with the herbicide Roundup® were done in the GM maize plot on May 8,
2003 (2 l/ha, in 150 liters of water per hectare, corn stage 1-3 leaves) and May 22, 2003 (3
l/ha, in 150 liters of water per hectare, corn stage 6-8 leaves).
The epigeic fauna was collected bi-weekly from May till September 2003 in 10 pitfall
traps per field. The traps were filled with 4% formaldehyde and opened for 24 hours. At the
dates of trap setting, visual counts were used to register natural enemies on 4 x 10 maize
plants per field..
Animals collected in pitfall traps and the unidentified specimens on maize plants were
preserved in 70 % alcohol and determined in the laboratory.
The structures of animal communities of the two fields was compared by the Sörensen
index of similarity Is=2c x 100/a+b.
Where: Is = Sörensen index; c = number of common species existing in two animal
communities; a = number of species existing only in the first community; b = number of
species existing only in the second community
Results and discussions
Epigeic fauna in pitfall traps
We have identified a total of 8816 animal specimens. Insects accounted for 7834 specimens
(3854 in Event NK 603 and 3980 in the isoline field). The most common insect species
belong to coleopterans with a total of 4986 specimens (2555/2431). Ground beetles were the
most abundant predators in pitfall traps. The most abundant species were Harpalus pubescens
Mull., Pterostichus vulgaris L., Harpalus griseus Panz., Carabus coriaceus L. and Carabus
cancelatus Illig.(Table 1). There was no statistical difference between fields.
The second abundant group of insect species belong to hymenopterans, including the
common ant species, Formica fusca (173/217), Formica rufa (77/63), Lasius flavus (241/230)
and Lasius niger (39/28). The third abundant species belong to orthopterous insects, most
common being Gryllus desertus (401/422) and Gryllus campestris (32/55). The fourth
abundant group of species belong to heteropterous insects, most common being Pyrrchocoris
apterus (202/277).
There are 832 identified specimens of spiders (441/391). The most common species
belong to the genus Philodromus (98/111). The second abundant group of species belongs to
the Opiliones , followed by Salticus sp., Pardosa sp., Tetragnatha sp. and Maripissa sp.
We identified a total of 137 specimens of Myriapoda, 28 belonging to the species Iulus
terrestris and 109 to Lithobius forficatus.
Main species of natural enemies living on maize plants
The most abundant group of species which were found on maize plants belong to the
Heteroptera (Nabis spp.), Coleoptera (Coccinelidae), Neuroptera (Chrysopa spp.) and spiders.
The average number of natural enemies counted on 40 plants per field during nine
observation dates is presented in Table 2.
145
Table 1. Structure of the main coleopteran insect fauna captured in pitfall traps.
Species
Harpalus pubescens Mull.
H. griseus Panz.
H. zabroides Dej.
H. aeneus F.
H. distinguendus Duft.
Pterostichus vulgaris L.
P. cupreus L.
P. melas Creutz.
Cicindella soluta Dej.
Carabus coriaceus L.
C. cancelatus Illig.
Drasterius bimaculatus Ross.
Amara aenea Deg.
Tanymecus dilaticollis Gyll.
Bothynoderes punctiventris Germ.
Total
Event NK 603
1657
175
43
15
10
189
29
37
12
101
46
5
17
16
4
2356
Isoline
1702
243
32
10
17
231
42
44
12
84
56
15
15
9
7
2519
Total
3359
418
75
25
27
420
71
81
24
185
102
20
32
25
11
4875
Table 2. Main species of natural enemies observed on maize plants. Mean of 40 plants per
field and nine observation dates from May to September 2003.
Event NK
Event
Species
603
Isoline
Species
NK 603
Nabis pseudoferus
1.75
1.5 Coccinella 7 punctata
0.1
Nabis feroides
0.1
0.1 Chrysopa sp.
0.5
Nabis ferus
1.25
1.25 Tetragnatha sp.
0.1
Nabis rugosus
0.25
0.25 Meta segmentata
0.1
Anthocoris sp.
0.1
0.1 Araneus diadematus
0.2
Orius sp.
1.25
1.5
Isoline
0.1
0.5
0.1
0.1
0.2
Similarity of communities
The Sörensen index of 98.15 (a=51; b=57 and c=53) shows that the two animal communities
are not different (100% means no difference between faunas and 1% signifies two complete
different faunas).
This indicates that the cultivation of Roundup Ready ® (Event NK 603) maize has no
influence on the biodiversity of the epigeic fauna captured in pitfall traps and on the main
species of natural enemies living on maize plants.
146
References
Rosca I. & Sabau I. 2001: Researches regarding influence of Roundup treatment and Roundup
Ready corn cultivation on usual fauna. In: Proceedings of IWGO Conference, Legnano,
Italy, 27 October-3 November 2001: 387-392.
Rosca I. 2002: Influence of Bt. Corn against useful entomofauna. The Bt. Maize Forum
26 th and 27 th September 2002 – Peyrehorade, France.
Rosca I. 2003: Biodiversity in Roundup Ready corn culture. Abstracts, XXXIII Meeting ESNA,
27-31 August, 2003, Viterbo-Italy: 95.
pp. 147-160
A biannual study on the environmental impact of Bt maize
František Sehnal1,2, Oxana Habuštová1, Lukáš Spitzer2, Hany M. Hussein1, Vlastimil
Růžička1
1
Entomological Institute, Academy of Sciences, and 2Faculty of Biological Sciences,
University of South Bohemia, Branišovská 31, 370 05 České Budějovice, Czech Republic (Email: [email protected])
Abstract: The cultivar MON810 (Bt maize), which expresses Bacillus thuringiensis toxin Cry1Ab,
and its parental non-transgenic cultivar were each grown on five 0.5 ha plots in a 7.6 ha field in 2002
and in a 14 ha field in 2003. Toxin expression in the Bt maize prevented plant damage by the
European corn borer, Ostrinia nubilalis, without any detectable effect on other arthropods. No
significant differences between the Bt and the control maize were found in the occurrence of aphids
Rhopalosiphum padi, R. maidis, and Metopolophium dirhodum, the thrip Franklinella occidentalis, the
eggs of lacewings, and the nymphs and adults of the predatory bug Orius. Similarly, neither the
biodiversity nor the abundance of carabid beetles, spiders and staphylinid beetles, the three groups
dominating the epigeic fauna, differed between the Bt and the non-Bt plots.
Key words: Bt maize, GM crops, European corn borer, thrips, aphids, Orius, Carabidae, Staphylinidae,
spiders
Introduction
Bacillus thuringiensis is the oldest and the most successful microbial agent used in the control
of insect pests. This soil bacterium produces several toxins, of which the crystalline proteins
(Cry) specify the activity of each bacterial strain on certain species of Lepidoptera, Diptera,
and Coleoptera (MacIntosh et al., 1990). The toxin Cry1Ab, which was introduced into the
maize genome, provides excellent protection against the stem borers such as the European
corn borer, Ostrinia nubilalis (Clark et al., 2000). Economic advantages of the transgenic Bt
maize led to its rapid spread since it had been commercialised in 1996. In 2002, Bt maize was
grown on 12.4 million hectares (includes 2.5 mil. ha of the double-transgenic herbicidetolerant Bt maize) in different parts of the world, which accounts for 9% of the total maize
acreage (James, 2002). In Spain, which is the only European country with commercial Bt
maize plantations, the planted area increased from 5% of total maize acreage in 2001 to 10%
in 2003, indicating the preferences of farmers when given a free choice. Elsewhere in Europe,
however, part of the public and some of the political leaders resist the use of Bt maize and
other genetically modified crops. Some of their arguments are unfounded but the possibility
of unwanted environmental side effects justifies certain caution (Wolfenbarger & Phifer,
2000). To probe justification of this concern in the case of Bt maize, we initiated a
comparative study on the composition of insect communities in plots of the Bt and the non-Bt
maize. This article is a brief summary of the results of a two-year study.
147
148
Crop planting
Field trials were performed in Southern Bohemia (altitude 300 m) in 2002 and 2003. Maize
cultivar MON 810 expressing the Cry1Ab protein, and the parental cultivar without the
Cry1Ab gene (near isogenic cultivar) were each planted on five 0.5 ha plots staggered in the
fields of 7.6 ha (2002) and 14 ha (2003). Stripes of bare land 10 m wide separated the plots.
Field margins were seeded with the near isogenic cultivar. In 2002, the field was treated with
the herbicide Guardian EC on May 14 and 28, and the maize was sown on May 15. In the
next year, herbicide treatment was done simultaneously with sowing on April 23. No other
treatments were applied during the vegetation period. The trials were terminated at the time of
seed ripening on September 17 in 2002 and August 22 in 2003. The crop was then shredded to
pieces smaller than 1 cm and next day ploughed about 25 cm deep into the soil.
Toxin measurements and insect studies
The amounts of Cry1Ab were measured in fresh plant tissues with enzyme-linked
immunosorbent assay (ELISA). Commercial kit (Linaris, Germany) specific for Cry1Ab was
used in 2002 and a kit detecting Cry1Ab and Cry1Ab/1Ac in 2003. Sensitivity threshold was
0.025 μg Cry1Ab in one gram of fresh plant tissue. Measurements were done six times during
the season when plants were collected for the analysis of associated entomofauna. To this end,
10 plants from each plot were transported in plastic bags to the laboratory to be taken apart.
Individuals of selected insect species were counted either on the whole plant or on its portions
and then recalculated for the whole plant. The epigeic insects were collected in 5 plastic
pitfall traps on each plot. The traps, 10 cm in diameter and with a layer of 3-4% formaldehyde
at the bottom (Pekar, 2002), were usually left in the field for two weeks in 2002 and one week
in 2003. Since in 2002 as many as 19% of all traps in the Bt and 13% in the non-Bt plots were
lost due to heavy rains, only four traps per plot were analysed in that year. Most collected
animals were identified to the species level. The catches were analysed separately for each
plant and trap and summed up for the respective plot. Final statistical analysis was performed
with different plot combinations to evaluate the effect of plot position as well as the effect of
the maize type.
Results
Toxin expression and the target pest
The measurements of toxin expression in maize plant leaves revealed a relatively steady level
throughout the season but the values measured in 2002 were several times higher than in 2003
(Table 1). Since we could not verify the difference with an independent test it cannot be
excluded that it was due to the use of different ELISA kits (Cry1Ab standards were provided
with the kits). The values established at any given time were fairly uniform both within and
between the plots. In July and August 2003 we compared Cry1Ab content in different parts of
the plants. While the leaves contained 1 to 2.5 ppm toxin, in the stems we found only 0.06, in
roots 0.075, in flowers 0.05, and in pollen 0.08 ppm. Cry1Ab content in the grain varied
between 0.025 (milk-stage ripening) and 0.088 ppm (ripe grain stored for six months).
Toxin presence in the leaves prevented attack of maize by the European corn borer
(ECB) Ostrinia nubilalis. Adults and eggs of ECB were found on the Bt and Non-Bt plants
with similar frequency but boring caterpillars occurred only in the latter. Those hatching on
the Bt-plants apparently died before they could cause visible boring tunnels. The distribution
of ECB on plots with the non-Bt maize seemed to be incidental but the numbers of insects
149
were too small to prove it statistically. At the end of summer of 2002, ECB interrupted
development as diapausing prepupae, which is typical for the region, whereas in 2003 with
unusually hot summer, ECB ecdysed to pupae at the end of August and apparently emerged as
adults in September. Since they had little chance to accomplish another generation before the
onset of winter, it is expected that the population density of ECB will be reduced in 2004.
Table 1. Expression of the Bt toxin in maize leaves and occurrence of the European corn borer
caterpillars in the Bt and non-Bt plants at distinct vegetation stages.
Plant stage and date
(year, month, day)
4-6 leaves, 020620,
4-6 leaves, 030611
Flowering, 020711,
Flowering, 030709
Husk formation
Husk formation, 030723
Milk-stage, 020820
Milk-stage, 030806
Waxy stage, 030820
μg Cry1A(b) per g of leaf
Bt-maize
Non-Bt maize
6.6 ± 0.7
0
1.05 ± 0.19
0
6.4 ± 1.4
0
1.36 ± 0.18
0
?
?
2.47 ± 0.17
0
7.5 ± 1.3
0
1.07 ± 0.47
0
0.90 ± 0.09
0
Ostrinia nubilalis per 50 plants
Bt-maize
Non-Bt maize
0
0
0
0
0
17
0
10
0
12
0
26
0
30
0
26
0
26 (pupae)
Variations in plot quality
Our experimental fields were relatively large and appeared uniform. The expression of
Cry1Ab in the Bt and the infestation by ECB of the non-Bt plants were alike in the compared
plots. However, the distribution of insects revealed considerable differences related to the plot
position within the field. The variations in moisture and soil quality across the fields and the
character of adjacent areas seemed to play important roles that were accentuated by unusual
weather conditions (excessive rain and floods in 2002 and heavy spring rains followed to a
long period of heat and draught in 2003). In an evaluation of the ground beetles and spiders
we proved that the four upper and the four bottom plots in the slightly slanted 2002 field were
significantly different (Spitzer et al., 2003). The existence of plot differences independent of
the maize cultivar is also documented on the example of two aphid species and the total
number of ground beetles and spiders collected at the peak season of 2002 (Table 2). It can be
seen that variations in insect abundance often depended on the plot position rather than on the
maize cultivar (Bt or non-Bt). Significant difference between the Bt and the non-Bt maize was
found on June 20 in case of the aphid Metopolophium dirhodum. However, when averaged for
the season, differences in the aphid occurrence on the Bt and non-Bt plants were insignificant.
150
Table 2. Numbers of aphids Rhopalosiphum padi and Metopolophium dirhodum found on ten
randomly chosen plants per plot on June 20, 2002, and numbers of ground beetles and spiders
caught in five pitfall traps per plot in July 12-27, 2002. Plot numbers are shown in brackets;
plots 1 to 5 were arranged in one, and plots 6 to 10 in another, parallel row. The last line
presents average values ± standard deviations.
R. padi
M. dirhodum
Ground beetles
Spiders
Bt maize Non-Bt
Bt maize Non-Bt
Bt maize Non-Bt
Bt maize Non-Bt
433 [1]
307 [2]
230 [1]
95 [2]
262 [1]
232 [2]
139 [1]
172 [2]
534 [3]
538 [4]
469 [3]
154 [4]
290 [3]
369 [4]
155 [3]
113 [4]
324 [5]
412 [6]
87 [5]
252 [6]
400 [5]
240 [6]
158 [5]
168 [6]
266 [7]
728 [8]
196 [7]
96 [8]
264 [7]
264 [8]
174 [7]
179 [8]
894 [9] 261 [10] 447 [9] 115 [10]
236 [9]
313 [10]
233 [9]
167 [10]
490 ± 248 449 ± 188 286 ± 166 142 ± 66 290 ± 64 284 ± 57 172 ± 36 160 ± 27
Non-target insects on the maize plants
On visual inspections we detected no differences in insect communities on the Bt and non-Bt
maize plants. Some insects, such as adult Diptera and Hymenoptera, were only accidental
visitors not associated with maize. Aphids and thrips and their natural enemies occurred on
the plants regularly but only the predatory Orius bugs and in 2003 also the lacewing eggs
were collected in sufficient numbers for statistical evaluation.
The thrips Franklinella occidentalis was the most abundant species in both years of study
(Fig. 1). Thrips appeared on the plants at the stage of about 4 leaves and the numbers of
nymphs reached maximum at the time of flowering at the end of June and beginning of July.
The peak of nymph abundance preceded that of adults by nearly three weeks (Habuštová et
al., 2004). In the course of July, nearly 40 thrips were found per one plant in 2002 and nearly
60 in 2003. Thrips persisted on the plants until our last collections at the end of August.
Certain differences in the fluctuations of thrip numbers on the Bt and non-Bt plants in course
of the seasons were due to differences in the plot position; when averaged for the whole year,
thrip infestations of the Bt and the non-Bt maize were statistically indistinguishable.
60
50
50
40
40
Number
Number
2002
60
30
20
10
2003
30
20
10
Non Bt
0
22.6.
11.7.
Bt
30.7.
20.8.
0
11.6. 25.6.
9.7. 23.7.
6.8. 20.8.
Figure 1. Average numbers of the thrips Franklinella occidentalis per one maize plant.
Bt
151
Two aphid species, Rhopalosiphum padi and Metopolophium dirhodum, were found on
the maize in both experimental years. The initial aphid abundance was much higher in the wet
year 2002 than in the hot and dry 2003 (Fig. 2). The distribution of aphids across the fields
was much affected by the plot position. At certain times, one or the other aphid species
seemingly preferred either the Bt or the non-Bt maize, but statistical evaluation of the data
collected over the season revealed that these differences were insignificant. At the beginning
of July 2003, we also detected an apparently incidental and very low maize infestation with
Rhopalosiphum maidis; only 11 specimens were found on fifty Bt, and 47 specimens on fifty
non-Bt plants.
Predatory bugs of the genus Orius occurred in low but consistent numbers throughout the
growing season (Fig. 3). Changes in bug abundance on the Bt and the non-Bt plants in course
of both years were virtually identical. Summation of the values from the 2002 collections
yielded a higher number for the Bt than for the non-Bt maize, while the relationship was
reversed in 2003. Neither difference was statistically significant.
2002
50
2003
50
40
Number
Number
40
30
20
30
20
10
10
0
0
22.6.
11.7.
11.6.
30.7.
9.7.
20.8.
20
2002
20
25.6.
23.7.
Non Bt
Bt
6.8.
20.8.
2003
15
Number
Number
15
10
10
5
5
Non Bt
0
22.6.
11.7.
Bt
30.7.
20.8.
0
11.6. 25.6.
9.7.
23.7.
Non Bt
Bt
6.8.
20.8.
Figure 2. Numbers of the aphids Rhopalosiphum padi (top graphs) and Metopolophium
dirhodum (bottom) per plant of the Bt and non-Bt maize.
152
2002
2,5
2,5
2
Number
Number
2003
3
3
1,5
2
1,5
1
1
0,5
0,5
Bt
Non Bt
0
22.6.
11.7.
30.7.
0
11.6. 25.6.
20.8.
9.7.
23.7.
Non Bt
Bt
6.8.
20.8.
Figure 3. Numbers of Orius bugs per one plant of the Bt and non-Bt maize.
The eggs of lacewings were found on the maize plants rarely in 2002 and frequently in
2003. The numbers found on the samples of ten plants greatly varied (Table 3) but the
difference in the values summed for the whole season proved insignificant with the Student
test. In spite of the relatively high numbers of the eggs, lacewing larvae were found only
exceptionally, probably because they dropped to the ground when the plants were cut and
packed for examination.
Table 3. Numbers (average per plot ± standard error) of lacewing eggs on ten plants in 2003.
Date
June 6
June 25
July 9
July 23
August 6
August 20
Total
Non-Bt maize
0.4
2.8 ± 1.6
16.8 ± 3.0
16.8 ± 6.1
10.2 ± 4.4
1.2 ± 0.4
48.2 ± 7.62
Bt maize
0.2
1.8 ± 1.3
14.2 ± 4.2
9.8 ± 5.0
28.0 ± 8.2
8.6 ± 2.2
62.6 ± 10.05
Epigeic beetles and spiders
Adult ground beetles (Carabidae), rove beetles (Staphylinidae), and spiders (Araneae)
dominated in pitfall catches. Occasionally we found considerable numbers of springtails
(Collembola) and adult Diptera that apparently emerged from the soil in the vicinity of the
respective trap. These and other arthropods whose catching was occasional were not
evaluated. The species numbers and abundance of carabids, staphylinids, and spiders were
higher in 2002 than in 2003 (Table 4), probably due to differences in the field character,
diverse weather patterns, and different lengths of trap exposure. Variations in soil properties
and adjacent habitats were apparently responsible for differences among the plots within each
field. However, the most abundant species were identical in virtually all plots in both years.
Pterostichus melanarius represented more than one quarter of all carabid beetles in 2002
and more than one third in 2003 (Table 5). The highly abundant carabids further included
153
Poecilus cupreus, Bembidion quadrimaculatum, and Calathus fuscipes in both 2002 and
2003. However, representation of some of the less abundant species was distinctly different in
the two years of study. For example, Bembidion lampros was much more common in the field
of 2002 than in that of 2003, and the representations of Calathus granulatus and Pterostichus
niger were reversed.
Table 4. Total numbers of collected species and specimens of the epigeic beetles and spiders
(ca 2% of the staphylinid beetles could not be identified to the species level). The 2002 data
represent catches in four traps per plot with a total exposure time 93 days (6 collections, each
for about a fortnight). In 2003, carabids and spiders were collected 6 times during the season,
each time in five traps per plot for about one week, totalling 43 exposure days. However, the
species of staphylinids were determined only in two spring catches (14 days exposure).
Taxon
Total numbers
Carabidae
Species
Specimen
Species
Specimen
Species
Specimen
Staphylinidae
Spiders
Non-Bt
30
3137
>44
869
38
2731
2002
Bt
32
3012
>42
880
36
2664
Total
40
6149
>55
1749
45
5395
Non-Bt
25
1787
>12
255
23
2124
2003
Bt
21
1694
>11
186
14
2114
Total
31
3481
>14
441
24
4238
The abundance of rove beetles in the 2003 field was much lower than in the 2002 field. A
biannual comparison is aggravated by the fact that only a part of the material collected in
2003 has been determined (Table 4). Most of the highly abundant species were nevertheless
found in both years. However, in contrast to the ground beetles and spiders, no clearly
dominating species can be identified among the rove beetles (Table 6). For example, in 2002 a
non-identified Atheta species that ranked seventh in abundance was only about three times
less common than the most frequent species Aleochara bipustulata.
The analysis of spiders revealed enormous differences between the dominating and the
rare species (Table 7). Oedothorax apicatus represented 90% of all spider species in 2002 and
94% in 2003. The second most common spiders, Erigone dentipalpis in 2002 and Trochosa
ruricola in 2003, made up only 2.6% and 2% of the total spider catches, respectively.
Pardosa agrestis ranked third in the relative spider abundance in both years but reached
maximally 1.7% of the total spider counts. Relative representation of most remaining species
was below 1%. Substantial differences in the relative abundance in the 2002 and 2003 fields
were found in a few species. For example, Erigone dentipalpis and Pardosa palustris were
common in 2002 but rare in 2003.
154
Table 5. Total numbers of carabid beetles caught on five non-Bt and five Bt plots in 2002 (4
traps per plot, total exposure time 93 days) and 2003 (5 traps per plot, exposure time 43 days).
2002
Species
Pterostichus melanarius
Poecilus cupreus
Calathus fuscipes
Bembidion quadrimaculatum
Pseudoophonus rufipes
Trechus quadristriatus
Bembidion lampros
Clivina fossor
Harpalus affinis
Carabus granulatus
Poecilus versicolor
Calathus melanocephalus
Pterostichus niger
Agonum muelleri
Loricera pillicornis
Carabus scheidleri
Bembidion properans
Bembidion femoratum
Nebria brevicollis
Pterostichus vernalis
Calathus ambiguus
Pterostichus strenuus
Carabus violaceus
Notiophilus palustris
Amara aulica
Amara consularis
Anchomenus dorsalis
Microlestes maurus
Microlestes minutulus
Agonum sexpunctatum
Amara ovata
Harpalus smaragdinus
Notiophilus aquaticus
Amara eurynota
Amara aenea
Amara littorea
Chlaenius nitidulus
Badister lacertosus
Badister unipustulatus
Lasiotrechus discus
Amara similata & Dyschirius globossus
Amara cursitans & Anisodactylus signatus
Harpalus rubripes & Carabus hortensis
Platynus assimilis & Pterostichus nigrita
2003
Non-Bt
Bt
Non-Bt
Bt
956
640
346
424
129
147
121
89
50
11
37
39
2
31
24
17
21
6
13
5
7
5
0
3
2
2
1
1
1
2
1
1
1
0
0
1
0
0
0
0
1+1
0
0
0
725
642
482
415
106
158
146
84
59
2
34
25
0
30
16
18
13
14
5
9
5
5
0
1
2
2
2
2
2
1
1
1
1
1
1
0
0
0
0
1
0
1+1
0
0
612
375
328
114
178
23
3
36
20
46
0
2
27
0
1
0
2
0
1
0
0
2
8
0
0
1
0
0
0
0
0
0
0
0
1
0
2
0
1
0
0
0
1+1
1+1
576
451
215
123
154
17
11
37
11
40
0
2
42
1
0
0
0
0
0
2
0
3
2
0
1
0
0
0
0
0
0
0
0
1
0
1
0
2
1
0
0
0
0
0
155
Table 6. Identified species of the staphylinid beetles that were collected on five non-Bt and
five Bt plots in 2002 (4 traps per plot from spring to autumn, total exposure 93 days) and
2003 (5 traps per plot, total exposure only 14 days in spring).
Species
Aleochara bipustulata
unidentified Aleocharinae
Omalium caesum
Oxytelus rugosus
Xantholinus linearis
Quedius boops
Atheta spp.
Xantholinus jarrigei
Lestera longelytrata
Aleochara bilineata
Tachinus fimetarius
Oxypoda lividipennis
Xantholinus longiventris
Gabrius nigritulus
Tachyporus pusillus
Gyrohypnus angustatus
Tachyporus hypnorum
Atheta fungi
Oxytelus tetracarinatus
Xantholinus tricolor
Carpelimus corticinus
Gabrius pennatus
Stenus clavicornis
Quedius cinctus
Mycetoporus erichsonanus
Omalium oxyacanthae
Xantholinus spp.
Heterothops quadripunctulus
Philonthus concinnus
Lathrobium fulvipenne
Arpedium quadrum & Gyrohypnus punctulatus
Philonthus cognatus & Philonthus atratus
Stenus (Nestus) spp.
Dinaraea angustula
Dinaraea linearis & Gabrius subnigritulus
Quedius nitipennis & Quedius fulgidus
Leptacinus sulcifrons & Ocypus nero
Omalium rivulare & Paederus schönherri
Quedius molochinus & Scopaeus laevigatus
Stenus flavipalpis & Tachinus corticinus
Tachyporus chrysomelinus
Astenus pulchellus & Lathrobium geminum
Mycetoporus longulus & Othius melanocephalus
Philonthus debilis & Philonthus varians
Drusilla canariculata & Aploderus caelatus
2002
Non-Bt
134
57
45
40
40
42
29
20
24
17
6
11
8
9
5
6
4
5
3
5
3
4
1
2
3
5
0
4
3
0
1+1
1+1
1
0
2+2
2+2
0
0
0
0
0
1+1
1+1
1+1
0
Bt
75
53
48
49
42
28
36
29
18
22
13
6
6
5
9
8
6
5
6
4
5
3
5
4
3
1
4
0
0
2
1+1
1+1
1
0
0
0
1+1
1+1
1+1
1+1
1
0
0
0
0
2003
Non-Bt
1
2
0
19
1
0
9
0
0
0
0
0
0
0
0
0
0
8
0
2
18
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1+1
Bt
0
7
1
17
1
0
10
2
0
0
0
0
0
0
0
0
0
2
0
7
7
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
156
Table 7. Collections of spider species on five non-Bt and five Bt plots in 2002 (4 traps per
plot, total exposure time 93 days) and 2003 (5 traps per plot, exposure time 43 days).
Species
Oedothorax apicatus
Erigone dentipalpis
Erigone atra
Pardosa agrestis
Trochosa ruricola
Bathyphantes gracilis
Pachygnatha degeeri
Porrhomma microphthalmum
Meioneta rurestris
Pardosa palustris
Walckenaeria vigilax
Robertus arundineti
Diplostyla concolor
Pardosa paludicola
Lepthyphantes tenuis
Trochosa terricola
Pachygnatha clercki
Pardosa pullata
Lepthyphantes flavipes
Dicymbium nigrum
Micrargus herbigradus
Pardosa prativaga
Araeoncus humilis
Xerolycosa nemoralis
Tallusia experta
Hahnia nava
Robertus lividus
Pelecopsis parallela
Xysticus kochi
Alopecosa cuneata
Arctosa leopardus
Micrargus subaequalis
Gongylidiellum vivum & Oedothorax fuscus
Tiso vagans & Walckenaeria nudipalpis
Microneta viaria & Microlinyphia pusilla
Pisaura mirabilis & Oedothorax retusus
Argenna subnigra & Pardosa riparia
Mangora acalypha & Theridion impressum
Centromerita bicolor & Centromerus pabulator
Porrhomma sp.
2002
Non-Bt
Bt
2309
103
78
27
23
24
45
25
26
11
9
8
8
5
1
2
2
1
1
4
1
1
1
2
1
0
1
1
1
1
0
0
1+1
1+1
1+1
1+1
0
0
0
0
2290
89
52
61
16
11
38
30
12
12
6
6
3
5
4
2
1
2
3
1
3
1
0
1
2
3
1
1
1
1
0
0
0
0
0
0
1+1
1+1
1+1
0
2003
Non-Bt
Bt
1982
1
7
36
48
17
2
3
3
1
3
2
2
0
1
1
2
1
0
0
0
2
3
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
1
1987
0
5
14
52
30
2
7
2
0
1
2
4
0
0
0
4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
2
Discussion
Ostrinia nubilalis (ECB) has not been regarded a serious pest of maize in Southern Bohemia
where our experiments were performed. Until several years ago, ECB was found rarely and
157
was taken for an occasional invader carried in by the wind. The relatively high incidence of
ECB occurrence in our plots reflects a general population rise of this species in Central
Europe. The incidence of maize attack by ECB in 2002 and 2003 was similar, in spite of
considerable difference in the weather pattern. In 2003, all ECB individuals found in our field
at the end of August were in the pupal stage. This was probably a general phenomenon in the
region that suffered from an exceptionally hot summer. It can be assumed that ECB adults
emerged sometimes at the beginning of September, too late to produce another generation.
Since the adults cannot overwinter, we presume that the incidence of maize attack by ECB
will be reduced in 2004. The reduction, if it occurs, will be temporal. ECB has been
established in the region and the damage of maize caused by its larvae is likely to rise. Our
data confirm that Bt maize is fully resistant to this pest (Clerk et al., 2000).
ELISA measurements showed that the expression of Cry1Ab toxin in the MON810
maize cultivar begins long before the ECB begins to fly and deposit the eggs. The amount of
toxin present in the leaves in 2002 was about five times higher than in 2003 but we cannot
exclude that the difference was unreal, being due to the use of different ELISA kits. Our field
observations confirmed that the egg laying ECB adults do not distinguish between the Bt and
the non-Bt plants (Orr & Landis, 1997). In both 2002 and 2003, the toxin seemed to kill the
hatched ECB larvae efficiently before they bored into the plant tissues.
We found no organism other than ECB to be affected by toxin expression. The lack of
effect on aphids is not surprising in view of toxin absence in the phloem sap of Bt maize
(Raps et al., 2001). Experiments with R. padi and R. maidis proved that they do not take-up
toxin with their food and that their population performance on the Bt maize is identical with
that on the non-transgenic maize (Head et al., 2001; Dutton et al., 2002). We do not know if
thrips feeding on Bt maize are also devoid of the toxin. They may be similar to the spider
mites Tetranychus urticae, which ingest but are not affected by the toxin (Dutton et al., 2002).
In some cases, toxin accumulated in the herbivores can affect predators that have no choice of
other diet. This was reported for the larvae of the lacewing Chrysoperla carnea that were fed
tiny caterpillars grown on the Bt maize (Hilbeck et al. 1998). Reduced numbers of
chrysopids, but not of ladybirds (Coccinelidae), were recorded also in the field experiments
with Bt maize (Wold et al., 2001). Our data did not permit statistical evaluation of the
abundance of these predators. The numbers of lacewing eggs on the Bt and the non-Bt plants
were virtually identical but we do not know if the parental adults had developed on the plot
where the eggs were collected. The lack of difference in the abundance of the Orius bugs on
the Bt and non-Bt maize is consistent with the report showing that feeding Orius insidiosus
with caterpillars that had ingested Cry1Ab did not affect the bug performance (Al-Deeb et al.,
2001). These authors also found no difference in bug abundance on the Bt and non-Bt maize
fields.
A few very abundant species referred to as agrobionts typically dominate agricultural
habitats (Luczak, 1979). This is clearly seen in the species composition and population
density of the epigeic predators. The results of our collections are typical for the fields of
Central Europe. Pterostichus melanarius and Poecilus cupreus dominated carabid catches in
our fields (Table 5), in the south-eastern part of the Czech Republic (Šťastná & Bezděk,
2002), and together with Pseudophonus rufipes in southern Poland (Olbrycht, 2002) and a
region south of Moscow in Russia (Timraleev et al., 2002). Luka et al. (2000) examined
seasonal population dynamics of these species in Switzerland and concluded, consistently
with our experience, that Poecilus cupreus is most common in spring, being replaced by
Pterostichus melanarius at the beginning of summer. In northern Germany, the domination of
Pterostichus melanarius and Bembidion lampros was associated with high abundance of
158
Anchomenus dorsalis a Poecilus versicolor (Imler, 2003), two species that were rare in our
collections.
The rove beetles identified in our study are also typical for the fields of Central Europe
(Bohac, 1999). Aleochara bipustulata, a species that feeds on dipteran larvae attacking plants
of the Brassicacae family, dominated in 2002 probably due to the presence of oilseed rape left
behind in the field from the cultivation in 2001. The other rove beetles common in our
collections, the saprophagous species Omalium caesum and Oxytelus rugosus, and the
predatory genus Xantholinus, are regular components of the epigeic field founa. Only the
relatively high abundance of the hydrophilic Quedius boops in 2002 was rather exceptional,
being associated with unusually moist weather in that year.
Our data on spiders are consistent with the conclusions of recent reviews (Hänggi et al.,
1995; Samu & Szinetár, 2002). It is known that Oedothorax apicatus colonizes the fields
preferentially, Trochosa ruricola and Pardosa agrestis are also recognized agrobionts, and
Erigone dentipalpis, Erigone atra, Porrhomma microphthalmum, and Meioneta rurestris are
typical inhabitants of open habitats, cultivated fields included (Buchar & Růžička 2002). A
few specimens of Arctosa leopardus and Pardosa prativaga caught in our traps came
probably from moist stripes of land adjacent to our fields.
It should be stressed that epigeic predators are an important component of agricultural
ecosystems because they keep many potential pests below the threshold of economic damage.
Their prey includes insects feeding on both the aerial and the subterranean parts of the maize
plants and potentially transmitting the Cry1Ab toxin. In spite of this expected exposure to the
toxin, we found no differences between the Bt and the non-Bt plots either in the beetle or in
the spider communities. This is in sharp contrast to the use of insecticides that often kill
indiscriminately and reduce populations of the beneficial insects such as carabid beetles (Lee
et al., 2001).
Acknowledgements
This work was performed as part of the research project AV0Z5007907 and was supported by
grant Z5007907-I011 from the Grant Agency of the Academy of Sciences. Identification of
the rove beetles (Staphylinidae) by Dr. Matúš Kocián is gratefully acknowledged.
References
Al-Deeb, M.A., Wilde, G.E. & Higgins, R.A. 2001: No effect of Bacillus thuringiensis corn
Bohac, J. 1999: Styphylinid beetles as bioindicators. Agric. Ecosyst. Environ. 74: 357-372.
Buchar, J. & Růžička, V. 2002: Catalogue of Spiders of the Czech Republic. Peres, Praha :
351 pp.
Clark, T.L., Foster, J.E., Kamble, S.T. & Heinrichs, E.A. 2000: Comparison of Bt (Bacillus
thuringiensis Berliner) maize and conventional measures for control of the European
corn borer (Lepidoptera: Crambidae). J. Entomol. Sci. 35: 118-128.
Entomol. 27: 441-447.
Habuštová, O., Hussein, H.M. & Sehnal, F. 2003: Insect communities on maize expressing a
Bt-toxin. Acta Phytotech. Zootech., in press.
159
Hänggi, A., Stöckli E. & Nentwig W. 1995: Habitats of Central European spiders.
Miscellanea Faunistica Helvetica 4: 1-460.
Head, G., Brown, C.R., Groth, M.E. & Duan, J.J. 2001: Cry1Ab protein levels in
phytophagous insects feeding on transgenic corn: implications for secondary exposure
risk assessment. Entomol. Exp. Appl. 99: 37-45.
Hilbeck, A., Baumgartner, M., Fried, P.M. & Bigler, F. 1998: Effects of transgenic Bacillus
carnea (Neuroptera: Chrysopidae). Environ. Entomol. 27: 480-487.
Irmler, U. 2003: The spatial and temporal pattern of carabid beetles on arable fields in
northern Germany (Schleswig-Holstein) and their value as ecological indicators. Agric.
Ecosyst. Environ. 98: 141-151.
James, C. 2002. Preview: Global status of commercialized transgenic crops: 2002. ISAAA
Briefs No. 27. ISAAA, Ithaca, N.Y.
Lee, J.C., Menalled, F.B. & Landis, D.A. 2001: Refuge habitats modify impact of insecticide
disturbance on carabid beetle communities. J. Appl. Ecol. 38: 472-483.
Luczak, J. 1975: Spider communities in the crop fields. Polish Ecol. Studies 5: 151-200.
Luka, H., Pfiffner, L. & Niggli U. 2000: Impact of various crops on ground beetles
(Coleoptera, Carabidae). Mittl. Entomol. Gesellsch. Basel 50: 18-38.
Macintosh, S.C., Stone, T.B., Sims, S.R., Hunst, P., Greenplate, J.T., Marrone, P.G., Perlak,
F.J., Fischhoff, D.A. & Fuchs, R.L. 1990: Specificity and efficacy of purified Bacillus
thuringiensis proteins against agronomically important insects. J. Insect Pathol. 56: 258266.
Olbrycht, T & Czerniakowski, Z.W. 2002: Species diversity of Carabidae beetles on fodder
beet cultivation and waste lands. Progr. Plant Protect. 42: 415-416.
Entomol. 90: 905-909.
Pekar, S. 2002: Differential effects of formaldehyde concentration and detergent on the
catching efficiency of surface active arthropods by pitfall traps. Pedobiologia 46: 539547.
Raps, A., Kehr, J., Gugerli, P., Moar, W.J., Bigler, F. & Hilbeck, A. 2001: Immunological
analysis of phloem sap of Bacillus thuringiensis corn and of the nontarget herbivore
Rhopalosiphum padi (Homoptera: Aphididae) for the presence of Cry1Ab. Mol. Ecol. 10:
525-533.
Saeglitz, C. & Bartsch, D. 2003: Regulatory and associated political issues with respect to Bt
transgenic maize in the European Union. J. Invertebr. Pathol. 83: 107-109.
Samu, F. & Szinetár, C. 2002. On the nature of agrobiont spiders. J. Arachnol. 30: 389-402.
Schuler, T.H., Poppy, G.M., Kerry, B.R. & Denholm, I. 1998: Insect-resistant transgenic
plants. Trends Biotechnol. 16: 168-175.
Spitzer, L., Růžička, V., Hussein, H.M, Habuštová, O. & Sehnal, F. 2004: Expression of a
Bacillus thuringiensis toxin in maize does not affect epigeic communities of carabid
beetles and spiders. Acta Phytotech. Zootech., in press.
Šťastná, P. & Bezděk, J. 2002: Ecotonal effect in species representation and quantity of
Carabidae (Coleoptera) in agrocenoses of the Kromeriz District. Acta Univ. Agricult.
Silvicult. Mendelianae Brunensis 50: 103-109.
160
Timraleev, Z.A., Aryukov, V.A. & Bardin, O.D. 2002: Comparative analysis of ground beetle
(Coleoptera, Carabidae) fauna and population in agroecosystems and meadows of
Mordovia. Zool. J. (in Russian) 81: 1517-1522.
Wold, S.J., Burkness, E.C., Hutchison, W.D. & Venette, R.C. 2001: In-field monitoring of
J. Entomol. Sci. 36: 177-187.
Wolfenbarger, L.L. & Phifer, P.R. 2000: The ecological risks and benefits of genetically
engineered plants. Science 290: 2088-2093.
pp. 161-164
Determination of fungi species, relationships between ear infection
rates and fumonisin quantities in Bt maize
Fahri Tatli, Mustafa Güllü, Fatih Ozdemir
Plant Protection Research Institute, Department of Phytopathology and Entomology, Adana,
Turkey (E-mail: [email protected])
Abstract: Field trials were conducted in Cukurova (East Mediterranean region of Turkey) during 2001
and 2002 to analyse fungal species as well as the relationship between maize ear fungal infection rates
and fumonisin quantities in conventional hybrids (DK-626 and P-3394) and insect protected (DK-626
Bt) Bt maize. Each year, the trial included 8 blocks, 4 of them receiving 3 insecticide treatments and
the other remaining untreated. The lowest fungal infection index was obtained for the insect protected
DK-626 Bt maize, with average values equal or less than 1.50 in both 2001 and 2002, compared to
values equal to or above 2.41 for both control groups over the two years. Fusarium moniliforme was
found to be the most prevalent fungal species, present in 69% or more of the infested grain analysed in
the study. Over the duration of the study, mean fumonisin concentrations in the conventional maize
hybrids DK-626 and P-3394 ranged from 15.6 – 18.1 ppm and 12.7 – 17.7 ppm, for 2001 and 2002
respectively, whereas values ranged from 2.5 - 2.6 and 0.63 – 0.78 ppm in DK-626 Bt maize for those
two years. Overall, mean fumonisin concentrations in Bt maize were therefore 6 to 7 times lower than
in conventional maize in 2001 and 15 to 20 fold lower in 2002.
Key words: maize, fumonisin, Fusarium moniliforme, ELISA
Introduction
In Turkey, maize is the third most important agricultural crop after wheat and barley, with
planted areas equivalent to 550.000 ha and an annual domestic maize production of
approximately 2.3 million tons. The East Mediterranean Cukurova region, which includes the
Adana, Mersin and Osmaniye provinces, has approximately 45% of its total production and
25% of its total area devoted to maize. In this region, maize is grown as the 1st crop; a 2nd crop
is also sown from the end of May to early June after wheat. The most damaging insect pests
are Sesamia nonagrioides Lef. followed by Ostrinia nubilais Hbn. Infestation levels and
damage are more severe in 2nd crop maize grown after wheat. In addition to reducing the grain
yield, insect damage creates ports of entry for fungal infections. The main fungal infection
observed during maize production in the Cukurova region include: Common Smut, Ustilago
maydis "DC" Corda, Leaf Spot, Drechslera maydis "Nisik" Subram & Jain, Root Rot,
Pythium sp., Fusarium sp., Root Rot, Pythium sp. and Fusarium sp. (maize stalk and leave).
In the grain, Fusarium moniliforme Sheld, Aspergillus sp. and Penicillium sp. are frequently
found.
This project, funded by Monsanto Europe S.A., was designed to investigate and compare
fungal infections and to measure the levels of fumonisins, produced by Fusarium
moniliforme, in the grain of insect protected and conventional maize grown as second crop in
the Cukurova region.
161
162
The maize hybrids used for the study were: the insect protected Bt maize DK626 Bt
(including event MON 810 from Monsanto), the near-isogenic conventional maize hybrid
DK626 and P3394, a classical non-transgenic reference hybrid that is grown in the region.
Sowing and other agronomic practices were done according to regional requirements for a
second maize crop. Trials in 2000 and 2001 were set up using a randomized split plot design
with 4 replicates. Each of the 4 blocks was divided in 2 sub blocks. Four sub-blocks received
3 applications of the insecticide lamba-cyhalotrine at 2-3 week intervals, starting the third
week of July. The 4 other sub-blocks were not treated with insecticide. Test plots consisted of
8 maize rows by 20 m. Ten days after silk emergence, all maize ears of the first block (treated
as well as untreated with insecticide) were wounded at 3 different places with a nail punch of
1-2 mm. This technique is classically used to create additional ports of entry (like the holes
and tunnels created by the insects) for fungal infection. At harvest, the degree of fungal
infection, main fungi species present and mycotoxin concentrations were compared across
treatments. Statistical analyses were carried out using MSTAT-C software.
Determination of fungal infestation
At harvest, 50 plants were selected randomly from each plot. The primary ear of each plant
was harvested and disease presence was recorded. Fungal infection of each ear was ranked
from 1 to 5 according to the following rating scale:
1: there is no fungal infection observed on grains,
2: fungal infection observed on 1-10% of grains,
3: fungal infection observed on 11-20 % of grains,
4: fungal infection observed on 21-30% of grains,
5: fungal infection observed more than 31 % of grains
Disease index for each plot was calculated by summing the individual ratings and then
dividing the total by the number of ears (50).
Identification of main fungal species present
After observation of fungal infection, the 50 ears from each plot were shelled by hand and
kernels were carefully mixed. Fifty kernels were randomly-selected from each replicate and
hybrid/treatment combination. These kernels were surface disinfected with 1% NaOCI for 2
min., rinsed two times with sterile water, and plated out on PDA (Potato Dextrose Agar) with
antibiotics. After 7 days of incubation in constant daylight at 25 C, Fusarium species were
identified on the base of conidial morphology (Booth, 1977; Nelson et al., 1983; Rossman et
al., 1987).
Mycotoxin analysis
The analysis of fungal mycotoxins focused exclusively on fumonisins (B1, B2 and B3). Fifty
kernels were randomly-selected from the samples used for identification of main fungal
species present in maize grain. CD-ELISA test kits (quantitative kit for fumonisin, Veratox,
Neogen Corp., Lansing, MI) were used to determine fumonisin levels in accordance with the
kit instructions. Ground maize samples (25 g) were extracted with 125 ml of a 70:30 (vol/vol)
mixture of methanol and water by blending for 2-3 min. Whatman No.1 filter paper was used
to filter the extract. A 100-µl portion of filtered sample was diluted (1:80) in a 10% aqueous
methanol solution.
Fumonisin standards (quantitative kit for fumonisin, Veratox, Neogen Corp., Lansing,
MI) or diluted sample extracts (100-µl) were added to mixing wells containing 100-µl of
163
fumonisin horseradish peroxidase conjugate. The sample and conjugate solution were mixed
with a multichanel pipettor, and 100-µl of this mixture was transferred to the antibody-coated
wells and incubated for 15 min at room temperature. Reagents were washed from antibody
wells with distilled water. Then 100-µl of an enzyme substrate was added and incubated for
15 min. The development of the color was stopped by the addition of a stopping reagent.
Fumonisin concentrations were assessed by recording optical density readings at 650 nm by
using a Bio-tek EL301 microwell strip reader.
Fungal infestation
As shown in Table 1, the lowest average fungal infestation index values were obtained for the
insect protected DK-626 Bt maize, with values of 1.54, 1.77 and 1.50, 1.50 in insecticide free
and insecticide applied plots in 2001 and 2002, respectively. For each hybrid, no significant
difference was found with or without insecticide application.
Table 1. Fungal infection observed on cobs at harvest, results for 2001 and 2002 [Insecticideuntreated plots (-), Insecticide-treated plots (+)].
Variety
DK-626 Bt
(-)
DK-626 Bt
(+)
DK-626
(-)
DK-626
(+)
P-3394
(-)
P-3394
(+)
Replicate 1
2001
2002
1.98
1.86
Fungal infestation index
Replicate 2
Replicate 3
2001
2002
2001
2002
1.46
1.85
1.44
1.07
Replicate 4
2001
2002
1.26
1.26
Average
index
2001
2002
1.54 a
1.50 c
2.0
1.85
1.63
1.38
1.68
1.46
1.75
1.34
1.77 a
1.50 c
2.56
2.96
2.50
2.89
2.62
2.78
3.26
3.31
2.74 bc
2.98 ab
2.52
2.67
2.36
2.28
2.60
2.98
2.16
2.77
2.41 b
2.67 ab
3.12
2.83
3.22
3.36
2.58
3.14
3.26
4.00
3.05 c
3.33 a
2.82
2.4
2.22
2.10
3.28
3.44
2.89
2.48
2.80 bc
2.60 a
Main fungal species present
The colonies that developed from grain shelled from infected maize ears were Fusarium
moniliforme - 70% (2001) and 69 % (2002), Penicillium sp. - 15% (2001) and 13.3% (2002),
Fusarium oxysporum - 7.5% (2001) and 6.2 % (2002) and Aspergillus spp. - 7.5% (2001) and
6.2 % (2002). For both years, Fusarium moniliforme, which causes ear rot and produces
fumonisin mycotoxins, represented most of the fungal biomass.
Mycotoxin concentrations
Fumonisin (B1, B2 and B3) analysis by ELISA indicated that total mean fumonisin
concentrations in the grain of conventional maize hybrids DK-626 and P-3394 ranged from
15.6 -18.1 ppm and 12.7 – 17.7 ppm, for 2001 and 2002 respectively, whereas values ranged
from 2.5-2.6 and 0.63 – 0.78 in the insect protected DK-626 Bt maize for those two years.
Mean fumonisin concentrations in Bt maize were therefore 6 to 7 times lower in 2001 and 15
to 20-fold lower than in conventional maize in 2002.
164
Table 2. Fumonisin concentrations determined by ELISA [Insecticide untreated plots (-),
Insecticide-treated plots (+)].
Variety
DK-626 Bt
(-)
DK-626 Bt
(+)
DK-626
(-)
DK-626
(+)
P-3394
(-)
P-3394
(+)
Replicate 1
2001
2002
3.5
0.8
Fumonisin concentration (ppm)
Replicate 2
Replicate 3
2001
2002
2001
2002
2.0
1.0
2.5
0.4
Replicate 4
2001
2002
2.0
0.8
Average
concentratrion (ppm)
2001
2002
2.5 a
0.78 c
3.7
0.5
2.3
0.4
1.8
0.8
2.8
0.7
2.6 a
0.63 c
16.8
17.3
15.1
18.7
18.3
18.2
19.9
12.8
17.5 c
16.75 ab
14.1
5.5
15.6
10.6
15.4
18.0
17.4
16.7
15.6 b
12.70 b
18.8
18.7
19.7
17.5
18.0
16.6
18.8
18.0
18.1 c
17.7 a
17.2
12.8
16.4
12.2
15.0
15.6
17.9
17.8
16.6 b
14.6 ab
Based on two years of results (2001 and 2002), the present study indicates that the level of
fungal infection, in general, was lower in Bt maize than in conventional non-Bt lines under
the local conditions in the Cukurova region. Using different Bt maize lines (CBH351, MON
810, and Bt 11 hybrids), Munkvold & Hellmich (1999) also found a lower levels of fungal
infection than in isogenic and control lines. Furthermore, across all hybrids, there were highly
significant correlation between insect damage, Fusarium ear rot severity and fumonisin
concentrations. The results of this study also show the challenges in a country like Turkey
where quality problems in maize have been faced for many years.
References
Booth, C. 1977: Fusarium: Laboratory Guide to the Identification of the Major Species.
C.M.I. Kew Surrey, England.
Nelson, P.E., Tounssoun, T.A. & Marasas, W.F.O. 1983: Fusarium species, an illustrated
Manual for Identification; The Pennsylvania State University Press: University, PA.
Rossman, A., Palp, M. & Speilman, L. 1987: A Literature Guide for the Identification of Plant
Pathogenic Fungi. American Phytopathology Society St. Paul Minnesota 252.
Miller J.D. 2000: Factors that affect the occurrence of fumonisin. Environ. Health Perspect.
109: 321-324.
Munkvold, G.P. & Hellmich, R.L. 1999: Genetically modified, insect resistant maize:
Implications for disease management. (http:// www.scisoc.org/feature/Bt
Maize/Top.html)
Munkvold, G.P. & Hellmich, R.L. 2003: Transgenic control of insect pests in maize reduces
mycotoxin concentrations in grain. (http:// www.scisoc.org/feature/Bt Maize/Top.html).
pp. 165-170
Spider communities in Bt maize and conventional maize fields
Christa Volkmar1, Michael Traugott2, Anita Juen2, Markus Schorling3, Bernd Freier3
1
Martin-Luther-University, Halle-Wittenberg, Institute for Plant Breeding and Plant
Protection, Ludwig-Wucherer-Str. 2, D-06108 Halle/Saale, Germany (E-mail:
[email protected]); 2Centre of Mountain Agriculture, University of Innsbruck,
Techniker-Str. 13, A-6020 Innsbruck, Austria; 3Federal Biological Research Centre for
Agriculture and Forestry, Institute for Integrated Plant Protection, Stahnsdorfer Damm 81,
D-14532 Kleinmachnow, Germany
Abstract: Investigations on spider communities in maize fields were conducted in Oderbruch
(Brandenburg, Germany) from 2000 to 2002 and near Halle/Saale (Saxony-Anhalt, Germany) in 2000.
Pitfall traps (10 per maize variety) were used to compare spider communities in Bacillus thuringiensis
(Bt), isogenic and conventional maize. The collected data were analysed using canonical community
ordination (CANOCO). In addition, species were classified into wandering and web-spinning spiders,
because the latter are thought to be more exposed to Bt pollen. In all fields the spider coenosis was
composed of species typical for open-land habitats (e.g. Oedothorax apicatus, Erigone atra, Pardosa
agrestis). Web-spinning spiders, dominated by Linyphiidae, accounted for more than 90 per cent of
the catches. No negative effect of Bt maize was observed on web-spinning spiders, as their abundance
was similar in all three maize varieties. Ordination analysis showed that maize variety had no impact
on the spider community while other factors such as field characteristics and yearly changing
environmental conditions were of great importance for community composition.
Key words: Bt maize, spider community, pitfall trap, CANOCO
Introduction
For the commercialization of Bacillus thuringiensis (Bt)-expressing maize in Germany,
monitoring of ecological side-effects is required. One essential topic is the evaluation of
possible effects of the Bt toxin (Cry1Ab) on non-target-organisms. Recent laboratory-based
studies (Hilbeck et al., 1998; Dutton et al., 2002) reported detrimental effects of Bt maize on
an important predatory insects, larvae of the green lacewing Chrysoperla carnea. However, a
drawback of many laboratory-based studies is the limited comparability to the practical field
situation. Field studies on conventional fields, however, offer the advantage to conduct
studies under realistic farming conditions.
Spiders constitute an important element of invertebrates in arable land (Nyffeler &
Sunderland, 2003). Since Bt maize expresses the toxin Cry1Ab also in its pollen (Fearing et
al., 1997), spiders are potentially exposed to the Bt toxin by feeding on pollen-contaminated
prey and spider-webs that are loaded with maize pollen (Ludy & Lang, 2003).
The present three-year study was conducted to investigate the influence of Bt maize on
the spider coenosis in large-scaled fields under conventional farming conditions. For the
analysis of the species community two approaches were used: Ordination analysis (ter Braak
& Šmilauer, 2002) and comparison of functional groups (Wilson, 1999).
Functional groups are defined as groups of species with similar ecological preferences.
Thus, ecological profiles that are valid for a number of species can be developed and used for
165
166
scientific generalization. This balanced approach offers an alternative to the analysis of single
species on one side and the whole spectrum on the other side (Petchey & Gaston, 2002).
The following field studies were conducted by the Institute for Integrated Plant Protection of
the Federal Biological Research Centre for Agriculture and Forestry.
In 2000, two sites situated in Oderbruch (Brandenburg, Germany) and in the Central
German Dry Region near Halle/Saale (Saxony-Anhalt, Germany) were investigated. The
Oderbruch area is permanently infested with the European corn borer, Ostrinia nubilalis (the
target species of the Bt maize), while infestation with this pest at the site near Halle/S. is
generally moderate. Field size was 7 and 15 ha in Oderbruch (location Neulewin, O-NE) and
Halle/Saale (location Spickendorf, H-SP), respectively. Cultivation methods were similar at
both locations with the exception of ploughing (Oderbruch - ploughed, Halle/Saale - not
ploughed). The fields were divided in 3 parts and planted with either Bt maize or one of two
non-transformed maize varieties (see Table 1).
In 2001, studies were conducted in Oderbruch (location Seelow, O-SE). Two
neighbouring fields with identical crop rotation and cultivation were selected. One field (17
ha) was planted with Bt maize, the other one (29 ha) with conventional maize.
In 2002, studies were conducted again in Oderbruch at two locations (Altreetz, O-AR,
Neureetz, O-NR). Each of the two fields (9 ha and 12 ha) was divided in two parts, one
planted with Bt maize, the other one with a conventional variety (Table 1). In each field part,
10 pitfall traps were set up along a central line at a distance of 20 m. Pitfall traps (10 cm
opening, 2 % formalin solution) were emptied weekly. Each year catches of six weeks
between the beginning of June and mid-August were selected for analysis.
Table 1. The following maize varieties were cultivated.
Bt maize (Bt)
2000 Mon810, variety ‘Novelis 270’
2001 Mon810, variety ‘Novelis 270’
2002 variety ‘MEB307’
isogenic maize (Iso)
variety ‘Nobilis 270’


conventional maize (Con)
variety ‘Aqui’*
variety ‘Flario’
variety ‘Lenz’
*with insecticide application (Ambush, 9 g a.i./ha) on 03/06/2000.
Using CANOCO 4.5 (ter Braak & Šmilauer, 2002), the relationship among the spider
communities and environmental variables (study site: O-NE, H-SP, O-SE, O-AR, O-NR;
year: 2000, 2001, 2002; maize strain: Bt, Iso, Con) was investigated. To prevent taxa caught
in high numbers from excessively influencing the ordinations, faunal counts were square-root
transformed. Only taxa represented by at least 30 individuals were used for ordination
analysis (Table 2). Canonical Correspondence Analysis (CCA) was used to illustrate the
major patterns in relation to year and study site. These environmental variables were then
used as covariables and their effect partialled out from the ordination to determine the relation
between the spider community and maize strain. To test the statistical significances, Monte
Carlo permutation tests were calculated.
In addition, the spider coenosis was analysed by employing the concept of functional
groups. Spiders were classified in two prey-capturing guilds: web-spinning spiders, like
167
Tetragnathidae, Theridiidae and Linyphiidae and wandering adult spiders, like Thomisidae,
Clubionidae, Pisauridae and Lycosidae.
Table 2. Spider-taxa used for ordination analysis. n = total catches in three years.
web-spinning spiders
wandering spiders
Linyphiidae
Abbreviation
Abbreviation
n
oed.api
26254 Opiliones
op
Oedothorax apicatus
eri.atr
3294 Lycosidae
lyco
Erigone atra
lini
1218 Pardosa agrestis
par.agr
Linyphiidae
eri.den
1144 Pardosa spec.
par.sp
Erigone dentipalpis
634 Trochosa ruricola tro.rur
Porrh. microphthalmum por.mic
mei.rur
516 Trochosa spec.
tro.sp
Meioneta rurestris
403 Pardosa prativaga par.pra
Lepthyph. tenuis-group lep.ten
bat.gra
329 Thomisidae
thom
Bathyphantes gracilis
ara.hum
125 Gnaphosidae
Araeoncus humilis
pac.deg
115 Zelotes spec.
zel.sp
Pachygnatha degeeri
oed.fus
41
Oedothorax fuscus
n
2115
1540
526
171
151
70
32
31
30
In order to achieve a realistic reproduction of commercial farming conditions, studies were
conducted using field-by-field comparison of large-scale maize fields. Data analysis focused
on two aspects. In a first approach, the spider coenosis was analysed by employing the
concept of functional groups.
In all years, at all locations and in all maize varieties more than 90 per cent of the
individuals were web-spinning spiders. This guild was dominated by the family Linyphiidae
and was more abundant than wandering spiders. The web-spinning Linyphiidae Oedothorax
apicatus and Erigone atra were dominant in all fields, while E. dentipalpis was dominant in
Halle/Saale in 2000, Meioneta rurestris in Oderbruch in 2000 and Porrhomma
microphthalmum in Halle/Saale 2000 and Oderbruch 2001. Lycosidae Pardosa agrestis was
the dominant wandering species in all three seasons in Oderbruch.
It is supposed that web-spinning spiders are exposed to a higher level of Bt maize pollen,
due to the pollen load in their nets. Furthermore, web-spinning spiders potentially accumulate
the Bt toxin since they ingest it by preying on herbivorous arthropods that had themselves fed
on Bt maize (Ludy & Lang, 2003). Nevertheless, catches of web-spinning spiders were not
decreased in Bt maize compared to non-transformed maize varieties.
In a second approach, the composition of the spider coenosis was analysed, using the
canonical correspondence-analysis in order to investigate the effects of maize variants and
environmental conditions on the spider community.
Canonical correspondence analysis showed that yearly changing environmental
conditions and field characteristics (environmental variables: year and study site) strongly
influenced the spider community (Figure 1). These environmental variables explained 44.6
per cent of the total variance (total inertia: 0.54) in community composition.
168
Only 2.4 per cent of the variance can be explained by the maize varieties, showing their
negligible influence on the spider community (Fig. 2). Nevertheless, not only the
environmental variables years and study sites but also the maize varieties proved to be
significant (p<0.05).
Figure 1. Canonical correspondence analysis (first and second axis): biplot of sampling points
and environmental variables. Samples of Bt, isogenic and conventional maize crops are
indicated by bright triangles, grey squares and dark circles, respectively.
With one exception, for none of the species difference in abundance was found among
the different maize varieties (Figure 3). Only Bathyphantes gracilis was found in higher
numbers in isogenic maize compared to Bt maize. No other effect of Bt maize on webspinning or wandering spiders could be detected.
A recent study by Brooks et al. (2003) and Volkmar et al. (2003) on arthropods in
genetically modified herbicide-tolerant crops (GMHT) revealed similar results, as most
species were unaffected by the GMHT crops. Some spider species significantly preferred or
avoided the GMHT crops, which was supposed to reflect differences in herbicide regime,
tillage system, crop rotation, weed density and prey availability.
In conclusion this three-year study showed that the influence of Bt maize on spider
communities is negligible compared to that of field characteristics and yearly changing
environmental conditions. Nevertheless, effects of transgenic maize on spider communities
may be possible in the long term and should be considered in future monitoring programs.
169
Figure 2. Partial ordination analysis after
excluding effects caused by year and study
site: biplot of sampling points and maize
varieties. Samples of Bt, isogenic and
conventional maize crops are indicated by
bright triangles, grey squares and dark circles,
respectively.
Figure 3. Partial ordination analysis after
excluding effects caused by year and study
site: biplot of species and maize varieties.
Web-spinning and wandering spiders are
indicated by white and black circles,
respectively.
References
ter Braak, C.J.F. & Šmilauer, P. 2002: CANOCO Reference manual and CanoDraw for
Windows User’s guide: Software for Canonical Community Ordination (version 4.5).
Microcomputer Power (Ithaca, NY, USA): 500 pp.
Brooks, D.R., Bohan, D.A., Champion, G.T., Haughton, A.J., Hawes, C., Heard, M.S., Clark,
S.J., Dewar, A.M,; Firbank, L.G., Perry, J.N., Rothery, P., Scott, R.J., Woiwod, I.P.,
Birchall, C., Skellern, M.P., Walker, J.H., Baker, P., Bell, D., Browne, E.L., Dewar,
A.J.G., Fairfax, C.M., Garner, B.H., Haylock, L.A., Horne, S.L., Hulmes, S.E., Mason,
N.S., Norton, L.R., Nuttall, P., Randle, Z., Rossall, M.J., Sands, R.J.N., Singer, E.J. &
herbicide-tolerant and conventional spring crops. I. Soil-surface-active invertebrates. Phil.
Trans. R. Soc. Lond. B 358: 1847-1862.
Dutton, A., Klein, H., Romeis, J. & Bigler, F. 2002: Uptake of Bt-toxin by herbivores
feeding on transgenic maize and consequences for the predator Chrysoperla carnea.
Ecol. Entomol. 27: 441–447.
170
Fearing, P.L., Brown, D., Vlachos, D., Meghji, M. & Privalle, L. 1997: Quantitative analysis
of Cry 1 A (b) expression in Bt-maize plants, tissues and silage and stability of
expression over successive generations. Mol. Breed. 3: 169–176.
Hilbeck, A., Baumgaertner, M., Fried, P.M. & Bigler, F. 1998: Effects of transgenic Bacillus
carnea (Neuroptera: Chrysopidae). Environ. Entomol. 27: 480–487.
Carabidae) in Bt corn and its effects on non-target insects. Boll. Zool. Agr. Bachic Ser. II
31: 37–58.
Ludy, C. & Lang, A. 2003: Trapped in a sticky web: Bt maize and food web context of webbuilding spiders. Abstracts of the IOBC/wprs working group meeting "Ecolocical impact
of genetically modified organism", Nov. 26-29 2003, Prague: 35.
Nyffeler, M. & Sunderland, K. D. 2003: Composition, abundance and pest control potential of
spider communities in agroecosystems: a comparison of European and US studies. Agric.,
Ecosyst. & Environ. 95: 579-612.
Entomol. 90: 905–909.
Petchey, O.L. & Gaston, K.J. 2002: Functional diversity (FD), species richness and
community composition. Ecol. Lett. 5: 402–411.
Volkmar, C., Lübke-Al Hussein, M., Jany, D., Hunold, I., Richter, L., Kreuter, T. & Wetzel,
T. 2003: Ecological studies on epigeous arthropod populations of transgenic sugar beet at
Friemar (Thuringia, Germany). Agric., Ecosyst. & Environ. 95: 37-47.
Wilson, J.B. 1999: Guilds, functional types and ecological groups. OIKOS 86: 507–522.
pp. 171-175
Larvicidal activities of transgenic Escherichia coli against susceptible
and Bacillus thuringiensis israelensis-resistant strains of Culex
quinquefasciatus
Margaret C. Wirth1, William E. Walton1, Robert Manasherob2, Vadim Khasdan2, Eitan
Ben-Dov2,4, Sammy Boussiba3,4, Arieh Zaritsky2,4
1
Department of Entomology, University of California, Riverside, CA 92521, USA; 2Life
Sciences Department, Ben-Gurion University, POB 653, Be’er-Sheva 84105, Israel;
3
Microalgal Biotechnology Lab, Blaustein Instit Desert Res, Sde Boker 84990, Israel; 4Bio
San Ltd., POB 3, Ariel 44837, Israel (E-mail: [email protected])
Abstract: Bacillus thuringiensis subsp. israelensis (Bti) is an efficient agent to control mosquitoes,
being specific and avoiding resistance development in the targets. Its larvicidal activity is contained in
a crystal composed of four major proteins, Cry4Aa, Cry4Ba, Cry11Aa and Cyt1Aa. The genes for
three (excluding Cry4Ba) were expressed with p20 in Escherichia coli in all 15 possible combinations.
The two E. coli clones producing Cyt1Aa, P20 and Cry4Aa, with and without Cry11Aa respectively,
display the highest toxicity against Aedes aegypti larvae ever reached in transgenic bacteria. Cyt1Aa
synergizes the other toxins, which is why it dramatically reduces the likelihood of resistant
development in the targets. Six of the clones were bioassayed against five strains of Culex
quinquefasciatus, four of which had been selected for resistance against various toxin combinations.
The susceptible strain was > 5- and 7-fold more sensitive to the combinations with all 4 genes than to
those without cyt1Aa or p20, respectively. The same combination was only 2.5-3-fold less toxic to
larvae of the resistant strains. The clone not producing Cyt1Aa was thousands-fold less toxic to the
resistant strains, except the strain selected with Cyt1Aa. Recombinant bacteria expressing all larvicidal
toxins of Bti have the potential to be as active as current Bti products but with additional advantages
for mosquito control.
Key words: Bacillus thuringiensis subsp. israelensis, toxin genes, expression in recombinant bacteria,
synergism, Culex quinquefasciatus, Aedes aegypti
Introduction
Bacillus thuringiensis subsp. israelensis (Bti) is the best biocontrol agent against larvae of
mosquitoes and black flies, vectors of many infectious diseases, owing to four major insecticidal crystal proteins (ICPs) (of 134, 128, 72, and 27 kDa) organized in a proteinacious crystal
(Margalith & Ben-Dov, 2000). The encoding genes (cry4Aa, cry4Ba, cry11Aa, and cyt1Aa)
reside on a 128 kb plasmid pBtoxis (Berry et al., 2002). The proteins differ qualitatively and
quantitatively, in their toxicity levels and against different species of mosquitoes. The crystal
is much more toxic than each of the ICPs alone. Cyt1Aa is the least toxic but very synergistic
(Crickmore et al., 1995), most likely due to its different mode of action (Butko, 2003).
No resistance has been detected toward Bti in field populations of mosquitoes despite 20
years of extensive usage (Margalith & Ben-Dov, 2000). Selection in the laboratory with Bti
has produced negligible resistance in Aedes aegypti (Goldman et al., 1986) or in Culex
quinquefasciatus. Resistance of the latter was obtained by selection to Cry4Aa, Cry4Ba and
Cry11Aa, alone or in combinations (Georghiou & Wirth, 1997), but the strains thus obtained
retained their original sensitivity levels in the presence Cyt1Aa (Wirth et al., 1997).
171
172
Resistance was thus completely suppressed by Cyt1Aa although its mode of action is yet to be
fully understood (Butko, 2003). Cross-resistance was observed, for example, between strains
resistant to Cry11Aa and Cry4A+Cry4B, and vice-versa (Wirth & Georghiou, 1997), but all
of the selected strains remained sensitive to Cry toxin mixtures plus Cyt1Aa (Georghiou &
Wirth, 1997). Cyt1Aa was found to suppress the expression of resistance to heterologous Cry
toxins in agricultural pests as well (Federici & Bauer, 1998; Wirth & Georghiou, 1997) hence
might be useful in managing resistance to bacterial insecticides (Wirth & Georghiou, 1997;
Wirth et al., 1998).
Despite its high efficacy against mosquito larvae, endurance of Bti’s activity is shortlived (Margalith & Ben-Dov, 2000). One way to overcome its limitations is to clone ICP gene
combinations for expression in other bacteria (Boussiba et al., 2000; Manasherob et al.,
2003a). The accessory protein P20, also encoded by a gene on pBtoxis, raises the levels of
Cyt1Aa, Cry11Aa and Cry4Aa in Escherichia coli (Khasdan et al., 2001) and protects
transgenic bacteria from the lethal effect of Cyt1Aa (Manasherob et al., 2003b).
All 15 possible combinations of 4 genes (cry4Aa, cry11Aa, cyt1Aa, p20) were expressed
in E. coli, two of which, producing Cyt1Aa, P20 and Cry4Aa (with or without Cry11Aa),
displayed the highest toxicity against A. aegypti larvae ever reached in transgenic bacteria
(Khasdan et al., 2001; 2003). Five out of the 6 clones containing cry4Aa or cry11Aa (with or
without p20) displayed varying levels of synergism with cyt1Aa: they are 1.5- to 34-fold more
toxic than the respective clones without cyt1Aa against exposed larvae. Here, we determine
their toxicities against resistant lines of C. quinquefasciatus and demonstrate that those clones
expressing combinations with cyt1Aa effectively suppress resistance.
Bacterial strains and plasmids
E. coli XL-Blue MRF' (Stratagene, La Jolla, Calif.) was used as a host. All 15 combinations
of 4 B. thuringiensis subsp. isralensis genes were cloned as described previously (Khasdan et
al., 2001).
Preparing lyophilised powder
The recombinant E. coli were grown overnight after induction by IPTG (0.5 mM). Cells were
harvested by centrifugation, washed in sterile distilled water and lyophilised. Recombinant E.
coli pellets were gently crushed to fine powder between weighing paper, weighed, and used to
prepare stock suspensions.
Bioassays of toxicities against and cross-resistance of mosquito larvae
(a) C. quinquefasciatus: Five colonies of C. quinquefasciatus were reared under standard
laboratory conditions and tested for susceptibility to the six most toxic E. coli recombinants;
four of them were reared under long-term selection pressure with B. thuringiensis
recombinants expressing various combinations of Bti toxins (Georghiou & Wirth, 1997). The
current level of resistance to their respective selection toxin(s) is found in Table 1.
(b) Stock suspensions were prepared in deionised water in flasks containing glass beads and
agitated on a vortex mixer until homogeneous. Ten-fold serial dilutions and stock suspensions
were kept frozen at 4oC when not in use. Assays exposed groups of 20 early-fourth instars in
100 ml deionised water in 8 oz. plastic cups. Ten different concentrations, from 0.2 to 200 g
ml-1, plus an untreated control, were used for each test, and tests were replicated 5 times on 5
different days. Mortality was evaluated after 24 hours. The susceptible colony was tested
concurrently with the 4 resistant colonies. Normally distributed data was analysed using a
173
Probit program for the PC (Raymond et al., 1993). Non-linear results were reported as
average mortality for those concentrations exhibiting toxicity. Resistance ratios were
calculated by dividing the LC value of the selected colony by that of the susceptible colony.
For non-linear data, resistance levels were estimated from the mortality observed at the
highest concentration tested (20 g ml-1).
Unselected, susceptible C. quinquefasciatus (Syn-P) were highly susceptible to all but one
(pHE4-AR) of the six E. coli recombinants tested (Table 1). Toxicity, in order of highest to
lowest activity, was pVE4-ADRC > pVE4-ARC > pHE4-AD > pHE4-ADR > pHE4-ADC. Only
the two most toxic clones, pVE4-ADRC and pVE4-ARC, demonstrated high toxicities against
the selected colonies as well. The selected colonies displayed high levels of resistance and crossresistance against clones expressing various combinations of Cry4Aa, Cry11Aa and P20, but
when presented with the Cyt1Aa-producing clones, toxicities were greatly enhanced and
resistance values were reduced to 1.9 – 3.3-fold. The exception was pVE4-ADC (expressing
cry4Aa + cry11Aa + cyt1Aa) that had the lowest activity against Syn-P among the 3 clones
expressing Cyt1Aa. This apparent discrepancy is readily explained by the lethal effect of cyt1Aa
expressed in E. coli, cytotoxicity that is relieved by co-expression with p20 (Mansaherob et al.,
2003b). The need for both P20 and Cyt1Aa for high toxicity is supported by the relatively low
toxicity of pHE4-ADR (expressing p20 but not cyt1Aa). The fact that toxicity of the clone with
pHE4-ADR is half than that with pHE4-AD can simply be explained by the additional gene in
the former, expressing the non-toxic P20 at the expense of the two toxins, exhausting the
bacterial apparent limited resources. However, toxicity order of the same clones seems to be
reversed when presented to susceptible colony of A. aegypti (Khasdan et al., 2001), a
discrepancy that should be resolved.
The most active recombinant E. coli strain, with pVE4-ADRC (expressing all four
genes), displayed an LC50 of 0.593 g ml-1, 30-fold higher than native Bti (Pasteur Institute
standard powder IPS80, with LC50 of 0.02 g ml-1 in the same test procedure) against Syn-P.
The recombinants are thus not as active against Culex as Bti per weight of technical powder,
but they lack another major toxin gene of Bti, cry4Ba, the contribution of which when
expressed in E. coli has not yet been determined. Furthermore, expression of three of these
genes in the cyanobacterium Anabaena PCC 7120 prolongs the exposure of targeted
mosquitoes and hence increases the recombinants’ effective activities (Manasherob et al.,
2003a). Regardless, these data show that Bti toxins can be effectively expressed in a bacterial
host other than B. thuringiensis, and express similar synergy and high activity against
resistant mosquitoes.
174
Table 1. Toxicities (LC50, in g ml-1) and Resistance Ratios (RR; in brackets) of E. coli clones
producing toxins from Bti tested against susceptible and Bti-resistant strains of C.
quinquefasciatus.
Clone of
Escherichia coli
(genes)
pHE4-AD (cry4Aa+
cry11Aa)
pHE4-ADR (cry4Aa+
cry11Aa+p20)
pVE4-ADRC (cry4Aa+
cry11Aa+cyt1Aa+p20)
pVE4-ARC (cry4Aa+
cyt1Aa+p20)
pVE4-ADC (cry4Aa+
cry11Aa+cyt1Aa)
pHE4-AR (cry4Aa+
p20)
Selection of Culex quinquefasciatus
and resistance ratios (RR) with
Cry11A
Cry4A +
Cry4A +
Syn-P
Cry4A+
(wild-type)
Cry4B +
Cry4B +
Cry4B
RR
=200
50
Cry11A +
RR50=1.0
RR50=678 Cry11A
CytA
RR50=74
RR50=6
1.51
3.10
0.593
0.928
4.24
>200
>>200
(>>130)
122,226
(~39,000)
1.93
(3.3)
1.77
(1.9)
>200
(>50)
>>200
>>200
(>>130)
167,680
(~54,000)
1.47
(2.5)
1.75
(1.9)
>200
(>50)
>>200
>>200
(>>130)
8,614
(~2,800)
1.46
(2.5)
2.70
(2.9)
>200
(>50)
>>200
~200
(~130)
29.5
(9.5)
1.52
(2.6)
1.96
(2.1)
>>200
(>50)
>>200
Acknowledgements
Mark Itsko is gratefully acknowledged for help in preparing the powders.
References
Berry, C., O’Neil, S., Ben-Dov, E., Jones, A.F., Murphy, L., Quail, M.A., Harris, D., Zaritsky,
A. & Parkhill, J. 2002: The complete sequence and organization of pBtoxis, the toxincoding plasmid of Bacillus thuringiensis subsp. israelensis. Appl. Environ. Microbiol.
68: 5082-5095.
Boussiba, S., Wu, X.-Q., Ben-Dov, E., Zarka, A. & Zaritsky, A. 2000: Nitrogen-fixing
cyanobacteria as gene delivery system for expressing mosquitocidal toxins of Bacillus
thuringiensis subsp. israelensis. J. Appl. Phycol. 12: 461-467.
Butko, P. 2003: Cytolytic toxin Cyt1A and its mechanism of membrane damage: data and
hypotheses. Appl Environ Microbiol. 69: 2415-2422.
Crickmore N., Bone, E.J., Williams, J.A. & Ellar, D.J. 1995: Contribution of the individual
components of the δ-endotoxin crystal to the mosquitocidal activity of Bacillus
thuringiensis subsp. israelensis. FEMS Microbiol. Lett. 131: 249-254.
Federici, B.A. & Bauer, L.S. 1998: Cyt1Aa protein of Bacillus thuringiensis is toxic to the
cottonwood leaf beetle, Crysomela scripta, and suppresses high levels of resistance to
Cry3Aa. Appl. Environ. Microbiol. 64: 4368-4371.
175
Georghiou, G.P. & Wirth, M.C. 1997: Influence of exposure to single versus multiple toxins
of Bacillus thuringiensis subsp. israelensis on development of resistance in the mosquito
Culex quinquefasciatus (Diptera: Culicidae). Appl. Environ. Microbiol. 63: 1095-1101.
Goldman, I.F., Arnold, J. & Carlton, B.C. 1986: Selection for resistance to Bacillus thuringiensis
subsp. israelensis in field and laboratory populations of the mosquito Aedes aegypti. J.
Invertebr. Pathol. 47: 317-324.
Khasdan, V., Ben-Dov, E., Manasherob, R., Boussiba, S. & Zaritsky, A. 2001: Toxicity and
synergism in transgenic Escherichia coli expressing four genes from Bacillus
thuringiensis subsp. israelensis. Environ. Microbiol. 3: 798-806.
Khasdan, V., Ben-Dov, E., Manasherob, R., Boussiba, S. & Zaritsky, A. 2003: Mosquito
larvicidal activity of transgenic Anabaena PCC 7120 expressing toxin genes from
Bacillus thuringiensis subsp. israelensis. FEMS Microbiol. Letts. 227: 189-195.
Manasherob, R., Otieno-Ayayo, Z.N., Ben-Dov, E., Miaskovsky, R., Boussiba, S. & Zaritsky,
A. 2003a: Enduring toxicity of transgenic Anabaena PCC 7120 expressing mosquito
larvicidal genes from Bacillus thuringiensis subsp. israelensis. Environ. Microbiol. 5:
997-1001.
Manasherob, R., Zaritsky, A., Metzler, Y., Ben-Dov, E., Itsko, M. & Fishov, I. 2003b:
Compaction of the Escherichia coli nucleoid caused by Cyt1Aa from Bacillus
thuringiensis subsp. israelensis. Microbiology 149: 3553-3564.
Margalith, Y. & Ben-Dov, E. 2000: Biological Control by Bacillus thuringiensis subsp.
israelensis. In: Insect Pest Management: Techniques for Environmental Protection, eds.
Rechcigl and Rechcigl. CRC Press LLC, Boca Raton, FL: 243-301.
Raymond, M., Prato, G. & Ratsira, D. 1993: Probability analysis of mortality assays
demonstrating quantal response, version 3.3. License L93019. Praxeme, Saint Georges
d’Orques, France.
Sayyed, A.H. Crickmore, N. & Wright, D.J. 2001: Cyt1Aa from Bacillus thuringiensis subsp.
israelensis is toxic to the Diamondback Moth, Plutella xylostella, and synergizes the
activity of Cry1Ac towards a resistant strain. Appl. Environ. Microbiol. 67: 5859-5861.
Wirth, M.C. & Georghiou, G.P. 1997: Cross-resistance among CryIV toxins of Bacillus
thuringiensis subsp. israelensis in Culex quinquefasciatus (Diptera: Culicidae). J. Econ.
Entomol. 90: 1471-1477.
Wirth, M.C., Georghiou, G.P. & Federici, B.A. 1997: CytA enables CryIV endotoxins of
Bacillus thuringiensis to overcome high levels of CryIV resistance in the mosquito, Culex
quinquefasciatus. Proc. Natl. Acad. Sci. USA. 94: 10536-10540.
Wirth, M.C., Delécluse, A., Federici, B.A. & Walton, W.E. 1998: Variable cross-resistance to
Cry11B from Bacillus thuringiensis subsp. jegathesan in Culex quinquefasciatus (Diptera:
Culicidae) resistant to single or multiple toxins of Bacillus thuringiensis subsp. israelensis.
Appl. Environ. Microbiol. 64: 4174-4179.
pp. 177-185
Peculiarities of Cry proteins to be taken into account during their in
vivo and in vitro study
Igor A. Zalunin, Lyudmila P. Revina, Lyubov I. Kostina, Galina G. Chestukhina
Scientific Research Institute for Genetics and Selection of Industrial Microorganisms, 1-st
Dorozhny pr. 1, 117545, Moscow, Russia (E-mail: [email protected])
Abstract: Classification of the Cry toxins of Bacillus thuringiensis is explained, and the toxin
properties, which should be considered in research on their action, are reviewed. Particular attention is
paid to toxin solubilisation and the role of proteolysis in their activation and inactivation. Possible
interference of proteases coating B. thuringiensis crystals with toxin action is emphasised. Attention is
also drawn to the antibacterial properties to some toxins and to the combinations of several Cry types
within a single crystal that has evolved in some strains of B. thuringiensis. It is suggested that such
combinations be considered in the development of future Bt crops. Virtually all literature pertinent to
the subject is summarised in this review.
Key words: Bacillus thuringiensis, Cry toxins, insecticidal proteins, insect digestion
Introduction
Cry proteins produced by various Bacillus thuringiensis (Bt) strains are involved in very
efficient biological control of insects. An extensive study of Cry proteins that began in the
50th of the last century (Angus, 1956a,b; Fitz-James et al., 1958) permitted to obtain a vast
knowledge about their structure, mechanism of action, natural variability, and evolution.
Comprehensive data are represented in numerous reviews (Hoefte & Whiteley, 1989; Bravo,
1997; Crickmore et al., 1998; Schnepf et al., 1998). Specific goal of this paper is to describe
physico-chemical characteristics of entomocidal (Bt) proteins such as solubility, stability, and
sensitivity to proteolysis. This information is needed for biological experiments to provide
reliable and interpretable results.
General characterisation of the Cry protein group
The term Cry proteins refers to a large class of proteins that are synthesised by different
strains of B. thuringiensis during sporulation as a form of crystal-like inclusions (crystals) in
the sporangia (Schnepf et al., 1998; de Maagd et al., 2001). Some other microorganisms have
recently been reported to produce Cry proteins (Barloy et al., 1996, 1998; Zhang et al., 1997).
Despite differences in their primary structure, Cry proteins have similar secondary and
tertiary structures (Hoefte & Whiteley, 1989; Hodgman & Ellar, 1990; Li et al., 1991;
Grochulski et al., 1995). The modern nomenclature of Cry proteins is based on the degree of
amino acid sequence identity (Crickmore et al., 1998). The formal name of each protein
consists of four hierarchical ranks successively designated by a number, capital letter, lower
case letter and another number. Cry proteins with less then 45% sequence identity are
designated different numbers of the upper rank. Within each of such subclasses (from Cry1 to
Cry43), proteins with less than 78% and 95% identity are distinguished with different letters
of the secondary and tertiary ranks, respectively.
177
178
Cry proteins are toxic only to certain species of Lepidoptera, Diptera, Coleoptera and
Hymenoptera (Schnepf et al., 1998). Activity of most of them is confined to one order, but
some are active on both Lepidoptera and Diptera (Cry2A, Cry1Ab7) (Nicholls et al., 1989;
Haider et al., 1989), and some on Lepidoptera, Coleoptera and Diptera (Cry 1Ba) (Zhong et
al., 2000). The representatives of subclasses Cry5, Cry6, Cry12, Cry13 and Cry21 are toxic to
nematodes (Crickmore et al., 1998; Schnepf et al., 1998). Cry proteins active against the same
order of insects may differ in the spectrum of sensitive species.
Two groups of Cry proteins can be distinguished on the basis of the molecular mass:
endotoxins Cry1, Cry4, Cru5, Cry7, Cry9, Cry14, Cry21, Cry26, and Cry 28 have masses
130140 kDa, while Cry2, Cry3, and Cry11A have masses 7075 kDa (Bulla et al., 1977;
Lilley et al., 1980; Chestukhina et al., 1980, 1982; Hoefte & Whiteley, 1989). Intermediate
values of molecular mass of 8090 kDa were lately observed for Cry1I, Cry11B, Cry13,
Cry18, Cry20, Cry22, and Cry27 subclasses (Crickmore, 2003 ).
Both 130-140 kDa and 70-75 kDa proteins are assigned to the Cry proteins in spite of the
two-fold difference in molecular mass because they have similar organisation of the active
domain. Endotoxins of the mass 130140 kDa are protoxins and each of them consists of an
N-terminal superdomain and a C-terminal moiety that is readily degradable (Chestukhina et
al., 1982). The N-terminal superdomain is a true toxin (Lecadet & Martouret, 1967; Schnepf
et al., 1998). Endotoxins with molecular mass of 70 kDa are analogues of the N-terminal
superdomains of the 130140 kDa endotoxins (Li et al, 1991; Grochulski et al., 1995).
Dissolving of Cry proteins
Entomocidal crystals are soluble neither in water and buffers with physiological pH values,
nor in organic solvents (Hannay, 1953; Angus, 1956b; Glatron et al., 1972). Individual
molecules of the 130140 kDa endotoxins are connected in the crystals by disulfide bonds
and non-covalent interactions. Hence, the best solvents for Cry proteins are agents reducing
the disulfide bonds (e.g. 2-mercaptoethanol and dithiothreitol) at pH values 9.510.5 (Glatron
et al., 1972; Huber et al., 1981). Practically, no inactivation is observed during this procedure
(Nishiitsutsuji-Uwo et al., 1977). Any of these factors (reducing agents and alkaline pH
values) taken separately cannot provide for crystal dissolving (Glatron et al., 1972). Very high
pH values (12) can only cause dissolving of crystals formed by 130140 kDa endotoxins in
the absence of a reducing agents (Glatron et al., 1972). However, a partial denaturation and
inactivation of entomocidal proteins occur under these conditions (Nishiitsutsuji-Uwo et al.,
1977). The dissolved proteins can be transferred into milder conditions by decreasing pH
value to 9.0 – 9.5 and removal of the reducing agent, for example by dialysis. However, the
disulfide bonds are likely to restore under storage leading to the formation of oligomers.
Cysteine residues involved in the intermolecular disulfide bond formation are apparently
localised in the C-terminal half of the protoxin molecule. The 70 kDa endotoxins are bound in
the crystals by non-covalent interactions only, and no reducing agent is required for their
solubilisation. These crystals are readily dissolved at highly alkaline pH values (10.012.0).
Since these conditions can cause a partial denaturation of the toxin, the lowest possible pH
values should be used to ensure quantitative dissolving of active protein. Suitable pH value
depends on the toxin; for example, Cry3A crystal is dissolved at pH 10.0 – 10.5 (McPherson
et al., 1988; Koller et al., 1992), whereas Cry11A endotoxin is only extracted from crystals
formed by ssp. isralensis at pH values close to 12.0 (Chestukhina et al., 1985). After
dissolving, pH value can be reduced to 8.5 – 10.0, but it is necessary to take into account that
aggregation of protein molecules can occur under storage. Addition of 2 M urea is
recommended to prevent aggregation (Chestukhina et al., 1994).
179
Cry protein proteolysis
The B. thuringiensis endotoxins are functioning in the insect gut and are strictly controlled by
the digestive proteolytic enzymes. The Cry protein structure has several features that
implicate maintenance of the biological activity in the aggressive gut environment and use of
proteolysis to elevate the activity. As said above, 130140 kDa proteins are protoxins whose
activation is accompanied by degradation of the C-terminal half of the molecule (Chestukhina
et al., 1982). The structure of this area is sensitive to proteolytic degradation and the progress
of proteolysis is independent of protease specificity. C-terminus seems to consist of short and
readily cleaved domains with molecular mass of about 15 kDa that are promptly subjected to
further degradation (Chestukhina et al., 1982). A short 2030 amino acid N-terminal fragment
is cleaved from the molecule of protoxin in the course of activation (Nagamatsu et al., 1984).
The active toxin molecules with molecular masses of 6570 kDa, which are produced by
protoxin proteolysis, are in most cases resistant to further degradation. They consist of three
domains with the N-terminal one formed by a bunch of 7 -helices, whereas the remaining
two domains contain -strands (Li et al., 1991; Grochulski et al., 1995). The domains are
tightly packed due to numerous non-covalent interactions. As a rule, the loops connecting the
domains and various structure elements (-helices and -strands) inside the domains are
strongly protected from the contact with proteolytic enzymes.
However, some toxins are subjected to further proteolysis that occurs in two principally
different ways. The first is represented by proteolytic attack of a loop binding -helices in the
N-terminal domain. For example, trypsin hydrolyses activated Cry4A and Cry4B toxins and
chymotrypsin and subtilisin attack activated Cry9A toxin (Angsuthanasombat et al., 1991,
1992, 1993; Zalunin et al., 1998). In both cases, a 50 kDa fragment is produced by hydrolysis
of C-terminal region of the loop that binds helices 5 and 6. The loop is prone to protease
action because the initial site of the helix 6 includes an amino acid residue that is
unfavourable for -helix formation. This is Gly206 in Cry4B (Chungjatupornchai et al.,
1988) and Pro222 in Cry9A (Smulevitch et al., 1991). Endotoxins of the subclasses Cry3A
(Caroll et al., 1989) and Cry2A (Nicholls et al., 1989) are hydrolyzed at the loop binding the
3 and 4 helices. In case of Cry1Ab, the midgut proteases of lepidopteran larvae split off
helices 1 and 2a (Miranda et al., 2001).
The scheme of proteolysis in the Cry11A endotoxin is quite different. It is attacked at the
loop binding the 4 and 5 strands of domain II and split into two fragments of molecular
masses 30 – 35 kDa (Dai & Gill, 1993; Yamagiwa et al., 2002; Revina et al., 2004).
Proteolysis at some loops inside domain II was also shown in Cry1Ab (Chestukhina et al.,
1990; Convents et al., 1991) and Cry1Ac (Choma et al., 1990). However, the loops in the
Cry1A domain II are better protected from proteolysis than in Cry11A. Proteolysis in this
region of Cry1A requires either partial denaturation of the protein (Choma et al., 1990) or
relatively large amount of enzymes and long time incubation (Pang et al., 1999; Lightwood et
al., 2000; Miranda et al., 2001).
It seems that the limited proteolysis is related to the mechanism of the entomocidal
proteins interaction with the membranes of the potential host cells. If true, the peculiarities in
the spatial structure of various toxins seem to determine both their toxicity and the process of
their activation in the gut of susceptible insects (Zalunin et al., 1998).
In many reported cases, the processing of activated toxins is not accompanied with their
significant inactivation (Caroll et al., 1989; Zalunin et al., 1998; Revina et al., 2004). This is
often related to the fact that after cleavage of a specific peptide bond, the products of such
limited proteolysis remain bound by noncovalent interactions in an active complex. For
180
example, the products of Cry11A hydrolysis by subtilisin retain the initial biological activity.
The toxicity is lost after the separation of the hydrolysis products by chromatography, but it is
readily restored after the fractions are combined again (Revina et al., 2004).
However, the 50 kDa fragment of Cry4B toxin retains its biological activity after
purification by chromatography (Zalunin et al., 1998), indicating that this fragment alone is
sufficient. In Cry1Ab, the loss of helix 1 by limited proteolysis appears necessary for
functionally significant oligomerisation of molecules before pore formation (Miranda et al.,
2001; Gomez et al., 2002). Sometimes, additional proteolysis of a superdomain leads to
modification in the spectrum of the biological activity. For example, the 51 kDa product of
Cry2A proteolysis by chymotrypsin possesses cytolytic activity against an insect cell line that
resists action of the original toxin and its 58 kDa fragment (Nicholls et al., 1989). In
Cry1Ab7, the conversion of 55 kDa toxin to a 53 kDa fragment after the treatment with
mosquito gut juice caused the loss of activity against lepidopteran cell line and appearance of
activity against the dipteran cells (Haider & Ellar, 1987).
On the other hand, there are examples of the decrease in biological activity of toxins to
the appropriate insect after additional proteolysis. For example, the 48 and 49 kDa fragments
obtained after the treatment of Cry4B with Aedes aegypti gut juice are significantly less toxic
to this insect than a 50 kDa fragment obtained by trypsin digestion Cry4B (Zalunin et al.,
1998). In contrast to Cry11A, the hydrolysis of Cry1A proteins in the region of domain II
leads to inactivation of these proteins (Miranda et al., 2001; Pang et al., 2002).
Sometimes the conditions in larva gut are unfavourable for the activation of Cry proteins.
For instance, the Cry1B and Cry7A proteins are active against some coleopteran species only
after preliminary in vitro dissolution and activation (Lambert et al., 1992; Bradley et al.,
1995).
In general, insect gut juice can elicit a different course of proteolysis than observed in
vitro with the use of purified proteases. The binding of Cry toxins to cell membranes also can
affect the course of proteolysis (Gomez et al., 2002). An alteration of the gut proteolytic
potency can be one of the ways of arriving of insect resistant forms (Oppert et al., 1997).
It is necessary to focus on another factor that affects proteolysis of the Cry endotoxins.
The entomocidal crystals isolated from bacterial cells contain a bulk of adsorbed bacterial
proteases (Chestukhina et al., 1980). Since crystal dissolving occurs under rather harsh
conditions, these enzymes can cause proteolysis of endotoxins during this procedure.
Therefore the adsorbed enzymes should first be washed off the crystals, for example with 1 M
NaCl, and inhibitors of proteolytic enzymes (ethylenediaminotetraacetic acid and
phenylmethylsulfonyl fluoride) should be added to toxin solutions. We suppose that similar
problems can arise when the contents of endotoxins in plants and animals are measured.
Proteolysis by concomitant proteases comprises the greatest danger when Cry protein
samples are prepared for the denaturing electrophoresis. Preliminary protein precipitation by
trichloroacetic acid is the most helpful procedure allowing both to concentrate the sample and
to efficiently inhibit proteases.
Multiple endotoxin production by single B. thuringiensis strains
Crystals of many B. thuringiensis strains are characterised by simultaneous presence of
several different endotoxins (Pfannenstiel et al., 1984; Yamamoto et al., 1988; Lecadet et al.,
1988; Granum et al., 1988; Chestukhina et al., 1994). The similarity of their physical and
chemical properties can hide such multiplicity. Ion-exchange chromatography can be used for
an efficient separation of endotoxins, whereas immunological methods and determination of
181
the N-terminal amino acid sequences are helpful for their identification (Chestukhina et al.,
1994).
The production of multiple entomocidal proteins broadens the spectrum of target insects,
enhances the efficiency due to synergism among the toxins (Chilcott & Ellar, 1988; Khasdan
et al., 2001), and aggravates the development of resistance (Tabashnik et al., 1996).
Encloning of individual genes of even very efficient toxins in plants will probably lead to
toxin resistance. Simultaneous cloning of several toxins can hardly be avoided in the future.
Toxin assemblies that evolved by natural evolution are worth studying as a source of material
for combined cloning.
Antibacterial effect of B. thuringiensis entomocidal proteins
Several entomocidal proteins have recently been reported to have antibacterial activity. For
example, the majority of endotoxins produced by the B. t. israelensis ssp. inhibit growth of
several Micrococcus and Streptomyces species and some other bacteria (Yudina et al., 2003).
In this connection, it is very important to investigate the toxic effect of Cry proteins on the
symbiotic microorganisms of insects.
Conclusions
Unfortunately, the physical and chemical state, and the fate of Cry proteins inside the insect
organism and even in the model cell culture systems are still obscure. The experiments with
cell cultures and membrane preparations are often conducted at the pH values that limit toxin
solubility. In the experiments on insects, we are often deprived of information about such
important parameters as is the degree of Cry proteins dissolving in the gut contents, the
progress and completeness of their proteolysis, toxin losses due to inefficient binding (for
example, to the peritrophic membrane). Toxin proteolysis in the gut depends on enzyme
composition, presence of endogenous and exogenous protease inhibitors, temperature, and
rate of gut emptying. These physiological parameters are rarely followed and our knowledge
of Cry protein toxicity and specificity is consequently not precise.
Toxin groups simultaneously produced by different B. thuringiensis strains are of great
interest. It is likely that most efficient combinations of entomocidal proteins, which enhance
the biological effect and prevent the development of resistance, have been selected in the
course of evolution and should be considered as template in prospective toxin combinations in
the genetically modified crops.
References
Angsuthanasombat, C., Crickmore, N. & Ellar, D.J. 1991: Cytotoxicity of a cloned Bacillus
thuringiensis subsp. israelensis CryIVB toxin to an Aedes aegypti cell line. FEMS
Microbiol. Lett. 83: 273-276.
Angsuthanasombat, C., Crickmore, N. & Ellar, D.J. 1992: Comparison of Bacillus
thuringiensis subsp. israelensis Cry IVA and CryIVB cloned toxins reveals synergism in
vivo. FEMS Microbiol. Lett. 94: 63-68.
Angsuthanasombat, C., Crickmore, N. & Ellar, D.J. 1993: Effects on toxicity of eliminating a
cleavage site in a predicted interhelical loop in Bacillus thuringiensis CryIVB δendotoxin. FEMS Microbiol. Lett. 111: 255-262.
Angus, T.A. 1956a: Association of toxicity with protein-crystalline inclusions of Bacillus
sotto Ishiwata. Can. J. Microbiol. 2: 122-131.
182
Angus, T.A. 1956b: Extraction, purification and properties of Bacillus sotto toxin. Can. J.
Microbiol. 2: 416-426.
Barloy, F., Delecluse, A., Nicolas, L. & Lecadet, M.M. 1996: Cloning and expression of the
first anaerobic toxin gene from Clostridium bifermentans subsp. malaysia, encoding a
new mosquitocidal protein with homologies to Bacillus thuringiensis delta-endotoxins. J.
Bacteriol. 178: 3099-3105.
Barloy, F., Lecadet, M.M. & Delecluse, A. 1998: Cloning and sequencing of three new
putative toxin genes from Clostridium bifermentans CH18. Gene 211: 293-299.
Bradley, D., Harkey, M.A., Kim, M.K. Biever, K.D. & Bauer, L.S. 1995: The insecticidal
CryIB crystal protein of Bacillus thuringiensis ssp. thuringiensis has dual specificity to
coleopteran and lepidopteran larvae. J. Invertebr. Pathol. 65: 162-173.
Bravo, A. 1997: Phylogenetic relationships of Bacillus thuringiensis delta-endotoxin family
proteins and their functional domains. J. Bacteriol. 179: 2793-2801
Bulla, L.A., Kramer, K.J. & Davidson, L.I. 1977: Characterization of the entomocidal
parasporal crystal of Bacillus thuringiensis. J. Bacteriol. 130: 375-383.
Carroll, J., Li, J. & Ellar, D.J. 1989: Proteolytic processing of a coleopteran-specific δendotoxin produced by Bacillus thuringiensis var tenebrionis. Biochem. J. 261: 99-105.
Chestukhina, G.G., Zalunin, I.A., Kostina, L.I., Kotova, T.S., Katrukha, S.P. & Stepanov,
V.M. 1980: Crystal- forming proteins of Bacillus thuringiensis. The limited hydrolysis
by endogenous proteases as a cause of their apparent multiplicity. Biochem. J. 187: 457465.
Chestukhina, G.G., Kostina, L.I., Mikhailova, A.L., Tyurin, S.A., Klepikova, F.S. &
Stepanov, V.M. 1982: The main features of Bacillus thuringiensis δ-endotoxin molecular
structure. Arch. Microbiol. 132: 159-162.
Chestukhina, G.G., Zalunin, I.A., Kostina, L.I., Bormatova, M.E., Klepikova, F.S., Khodova,
O.M. & Stepanov, V.M. 1985: Structural features of crystal-forming proteins, produced
by Bacillus thuringiensis subspecies israelensis. FEBS Lett. 190: 345-348.
Chestukhina, G.G., Tyurin, S.A., Kostina, L.I., Osterman, A.L., Zalunin, I.A., Khodova, O.M.
& Stepanov, V.M. 1990: Subdomain organization of Bacillus thuringiensis entomocidal
protein’s N-terminal domains. J. Protein Chem. 9: 501-507.
Chestukhina, G.G., Kostina, L.I., Zalunin, I.A., Revina, L.P., Mikhailova, A.L. & Stepanov,
V.M. 1994: Production of multiple δ-endotoxins by Bacillus thuringiensis: δ-endotoxins
produced by strains of the subspecies galleriae and wuhanensis. Can. J. Microbiol. 40:
1026-1034.
Chilcott, C.N. & Ellar, D.J., 1988: Comparative toxicity of Bacillus thuringiensis var
israelensis crystal proteins in vivo and in vitro. J. Gen. Microbiol. 134: 2551-2558.
Choma, C.T., Surewicz, W.K., Carey, P.R., Pozsgay, M., Raynor, T. & Kaplan, H. 1990:
Unusual proteolysis of the protoxin and toxin from Bacillus thuringiensis structural
implications. Eur. J. Biochem. 189: 523-527.
Chungjatupornchai, W.P., Hoefte, H., Seurinck, J., Angsuthanasombat, C. & Vaeck, M. 1988:
Common features of Bacillus thuringiensis toxins specific for Diptera and Lepidoptera.
Eur. J. Biochem. 173: 9-16.
Convents, D., Cherlet, M., Van Damme, J., Lasters, I. & Lauwereys, M. 1991: Two structural
domains as a general fold of the toxic fragment of the Bacillus thuringiensis δendotoxins. Eur. J. Biochem. 195: 631-635.
Crickmore, N. 2003: Full list of delta-endotoxins. http://www.biols.susx.uk/Home/Neil
Crickmore/Bt/
183
Crickmore, N., Zeigler, D.R., Feitelson, J., Schnepf, E., Van Rie, J., Lereclus, D., Baum, J. &
Dean, D.H. 1998: Revision of the nomenclature for the Bacillus thuringiensis pesticidal
crystal proteins. Microbiol. Mol. Biol. Rev. 62: 807-813.
Dai, S.M. & Gill, S.S. 1993: In vitro and in vivo proteolysis of the Bacillus thuringiensis
subsp. israelensis CryIVD protein by Culex quinquefasciatus larval midgut proteases.
Insect Biochem. Mol. Biol. 23: 273-283.
Fitz-James, P.C., Toumanoff, C. & Yong, E. 1958: Localization of a toxicity for silkworm
larvae in the parasporal inclusion of Bacillus cereus var alesti. Can. J. Microbiol. 4: 385396.
Glatron, M.-F., Lecadet, M.-M. & Dedonder, R. 1972: Structure of the parasporal inclusion of
Bacillus thuringiensis Berliner: Characterization of a repetitive subunit. Eur. J. Biochem.
30: 330-338.
Gleave, A.P., Williams R. & Hedges, R.J. 1993: Screening by polymerase chain reaction of
Bacillus thuringiensis serotypes for the presence of cryV-like insecticidal protein genes
and characterization of a cryV gene cloned from B. thuringiensis subsp. kurstaki. Appl.
Gomez, I., Sanches, J., Miranda, R., Bravo, A. & Soberon, M. 2002: Cadherin-like receptor
binding facilitates proteolytic cleavage of helix alpha-1 in domain I and oligomer prepore formation of Bacillus thuringiensis Cry1Ab toxin. FEBS Lett. 513: 242-246.
Granum, P.E., Pinnavaia, S.M. & Ellar, D.J. 1988: Comparison of the in vivo and in vitro
activity of the δ-endotoxin of Bacillus thuringiensis var morrisoni (HD-12) and two of its
constituent proteins after cloning and expression in Escherichia coli. Eur. J. Biochem.
172: 731-738.
Grochulski, P., Masson, L., Borisova, S., Pusztai-Carey, M., Schwartz, J.-L., Brousseau, R. &
Cygler, M. 1995: Bacillus thuringiensis Cry1A(a) insecticidal toxin – crystal structure
and channel formation. J. Mol. Biol. 254: 447-464.
Haider, M.Z. & Ellar, D.J. 1987: Characterization of the toxicity and cytopathic specificity of
a cloned Bacillus thuringiensis crystal protein using insect cell cultures. Mol. Microbiol.
1: 59-66.
Haider, M.Z., Smith, G.P. & Ellar, D.J. 1989: Delineation of the toxin coding fragments and
an insect – specificity region of a dual toxicity Bacillus thuringiensis crystal protein
gene. FEMS Microbiol. Lett. 58: 157-164.
Hannay, C.L. 1953: Crystalline inclusions in aerobic sporeforming bacteria. Nature 172:
1004.
Hodgman, T.C. & Ellar, D.J. 1990: Models for the structure and function of the Bacillus
thuringiensis δ-endotoxins determined by compilation analysis. DNA Seq. 1: 97-106.
Hoefte, H. & Whiteley, H.R. 1989: Insecticidal crystal proteins of Bacillus thuringiensis.
Microbiol. Rev. 53: 242-255.
Huber, H.E., Luthy, P., Ebersold, H.-R. & Cordier, J.-L. 1981: The subunits of the parasporal
crystal of Bacillus thuringiensis size, linkage and toxicity. Arch. Microbiol. 129: 14-18.
Khasdan, V., Ben-Dov, E., Manasherob, R., Boussiba, S. & Zaritsky, A. 2001: Toxicity and
synergism in transgenic Escherichia coli expressing four genes from Bacillus
thuringiensis subsp. israelensis. Environ. Microbiol. 3: 798-806.
Koller, C.N., Bauer, L.S. & Hollingworth, R.M. 1992: Characterization of the pH-mediated
solubility of Bacillus thuringiensis var. san diego native delta-endotoxin crystals.
Biochem. Biophys. Res. Commun. 184: 692-699.
Lambert, B., Hofte, H., Annys, K., Jansens, S., Soetraert, P. & Peferoen, M. 1992: Novel
Bacillus thuringiensis insecticidal crystal protein with a silent activity against
coleopteran larvae. Appl. Environ. Microbiol. 58: 2536-2542.
184
Lecadet, M.-M. & Martouret, D. 1967: Enzymatic hydrolysis of the crystals of Bacillus
thuringiensis by the proteases of Pieris brassicae. II. Toxicity of the different fractions of
the hydrolysate for larvae of Pieris brassicae. J. Invertebr. Pathol. 9: 322-330.
Lecadet, M.-M., Sanchis, V., Menou, G., Rabot, P., Lereclus, D., Chaufaux, J. & Martouret,
D. 1988: Identification of a δ-endotoxin gene product specifically active against
Spodoptera littoralis Bdv. among proteolysed fractions of the insecticidal crystals of
Bacillus thuringiensis subsp. aizawai 7.29. Appl. Environ. Microbiol. 54: 2689-2698.
Li, J., Carroll, J. & Ellar, D.J. 1991: Crystal structure of an insecticidal protein the δendotoxin from Bacillus thuringiensis subsp. tenebrionis at 2.5 Å resolution. Nature 353:
815-821.
Lightwood, D.J., Ellar, D.J. & Jarrett, P. 2000: Role of proteolysis in determining potency of
Bacillus thuringiensis Cry1Ac delta-endotoxin. Appl. Environ. Microbiol. 66: 51745181.
Lilley, M., Ruffell, R.N. & Somerville, H.J. 1980: Purification of the insecticidal toxin in
crystals of Bacillus thuringiensis. J. Gen. Microbiol. 118: 1-12.
Maagd de, R.A., Bravo, A. & Crickmore, N. 2001: How Bacillus thuringiensis has evolved
specific toxins to colonize the insect world. Trends Genetics 17: 13-199.
McPherson, S.A., Perlak, F.J., Fuchs, R.L., Marrone, P.G., Larvik, P.B. & Fischhoff, D.A.
1988: Characterization of the coleopteran-specific protein gene of Bacillus thuringiensis
var tenebrionis. Bio/Technology 6: 61- 66.
Miranda, R., Zamudio, F.Z. & Bravo, A. 2001: Processing of Cry1Ab delta-endotoxin from
Bacillus thuringiensis by Manduca sexta and Spodoptera frugiperda midgut proteases:
role in protoxin activation and toxin inactivation. Insect Biochem. Mol. Biol. 31: 11551163.
Nagamatsu, Y., Itai, Y., Hatanaka, G., Funatsu,G. & Hayashi K. 1984: A toxic fragment from
the entomocidal crystal protein of Bacillus thuringiensis. Agric. Biol. Chem. 48: 611619.
Nicholls, C.N., Ahmad, W. & Ellar, D.J. 1989: Evidence for two different types of
insecticidal P2 toxins with dual specificity in Bacillus thuringiensis subspecies. J.
Bacteriol. 171: 5141-5147.
Nishiitsutsuji-Uwo, J., Ohsawa, A. & Nishimura, H.S. 1977: Factors affecting the insecticidal
activity of δ-endotoxin of Bacillus thuringiensis. J. Invertebr. Pathol. 29: 162-169.
Oppert, B., Kramer, K.J., Beeman, R.W., Johnson, D. & McGaughey, W.H. 1997: Propteasemediated insect resistance to Bacillus thuringiensis toxins. J. Biol. Chem. 272: 2347323476.
Pang, A.S., Gringorten, J.L. & Bai, C. 1999: Activation and fragmentation of Bacillus
thuringiensis delta-endotoxin by high concentrations of proteolytic enzymes. Can. J.
Microbiol. 45: 816-825.
Pfannenstiel, M.A., Ross, E.J., Kramer, V.C. & Nickerson, K.W. 1984: Toxicity and
composition of protease-inhibited Bacillus thuringiensis var. israelensis crystals. FEMS
Microbiol. Letts. 21: 39-42.
Revina, L.P., Kostina, L.I., Ganushkina, L.A., Mikhailova, A.L., Zalunin, I.A & Chestukhina,
G.G. 2004: Reconstruction of Cry11A Bacillus thuringiensis ssp. israelensis endotoxin
from fragments, corresponding to N- and C- moieties of its molecule restores original
biological activity. Biochemistry (Moscow) 69: 225-233.
Schnepf, E., Crickmore, N., Van Rie, J., Lereclus, D., Baum, J., Feitelson, J., Zeigler, D.R. &
Dean, D.H. 1998: Bacillus thuringiensis and its pesticidal crystal proteins. Microbiol.
Mol. Biol. Rev. 62: 775-806.
185
Tabashnik, B.E., Malvar, T., Lin Y.-B., Finson, N., Birthakur, D, Shin B.-S. Park, S.-H,
Masson, L., de Maagd, R.A. & Bosch, D. 1996: Cross-resistance of the diamondback
moth indicates altered interactions with domain II of Bacillus thuringiensis toxins. Appl.
Smulevitch, S.V., Osterman, A.L., Shevelev, A.B., Kaluger, S.V., Karasin, A.I., Kadyrov,
R.M., Zagnitko, O.P., Chestukhina, G.G. & Stepanov, V.M. 1991: Nucleotide sequence
of a novel delta-endotoxin gene cryIg of Bacillus thuringiensis ssp. galleriae. FEBS Lett.
293: 25-28.
Yamagiwa, M., Ogawa, R., Yasuda, K., Natsuyama, H., Sen, K. & Sakai, H. 2002: Active
form of dipteran-specific insecticidal protein cryllA produced by Bacillus thuringiensis
subsp. israelensis. Biosci. Biotechnol. Biochem. 66: 516-522.
Yamamoto, T., Ehmann, A., Gonzalez, J.M., & Carlton, B.C. 1988: Expression of three genes
coding for 135 kilodalton entomocidal proteins in Bacillus thuringiensis kurstaki. Curr.
Microbiol. 17: 5-12.
Yudina, T.G., Konukhova, A.V., Revina, L.P., Kostina, L.I., Zalunin, I.A. & Chestukhina,
G.G. 2003: Antibacterial activity of Cry- and Cyt-proteins from Bacillus thuringiensis
ssp. israelensis. Can. J. Microbiol. 49: 37-44.
Zalunin, I.A., Revina, L.P., Kostina, L.I., Chestukhina, G.G. & Stepanov, V.M. 1998: Limited
proteolysis of Bacillus thuringiensis CryIG and CryIVB δ-endotoxins leads to formation
of active fragments that do not coincide with the structural domains. J. Protein. Chem.
17: 463-471.
Zhang, J., Hodgman, T.C., Krieger, L., Schnetter, W. & Schairer, H.U. 1997: Cloning and
analysis of the first cry gene from Bacillus popilliae. J. Bacteriol. 179: 4336-4341.
Zhong, C., Ellar, D.J., Bishop, A., Johnson, C., Lin, S. & Hart, E.R. 2000: Characterization of
a Bacillus thuringiensis delta-endotoxin which is toxic to insects in three orders. J.
Invertebr. Pathol. 76: 131-139.
186
III. Workshop reports
pp. 187-189
Workshop report - Hybridization & Fitness of Hybrids
Detlef Bartsch1, Hans den Nijs2
1
Robert Koch Institut - Center for Gene Technology, Wollankstrasse 15-17, 13187 Berlin,
Germany (E-mail: [email protected]); 2Institute for Biodiversity & Ecosystem Dynamics,
University of Amsterdam, Kruislaan 318, Amsterdam 1098 SM, The Netherlands (E-mail:
[email protected])
This workshop was attended by 12 participants who discussed the phenomena of
hybridization, the consecutive introgression, and aspects of hybrid fitness, and its
consequences. The main focus was on plants, the relevant taxa in European context appeared
to be Beta, Brassica, Helianthus, Maize, Olive, Orange, Oryza, Solanum, Triticum, etc.
The workshop participants realised that recently, a massive body of data has become
available with respect to gene flow from crops into their wild relatives. Large majority of crop
species are now known to have exchanged genes with their closest wild relatives, given the
fulfillment of a range of conditions, like sympatric occurrence, synchronous flowering, and
cross compatibility. There is full reason to assume that these processes will continue to
happen. Important reviews are by Eastham & Sweet (2002) for European crops and Ellstrand
(2003), who presents a world-wide overview. It is therefore acknowledged that gene transfer
from GM crops to wild relatives will sooner or later, inevitably take place, so it are the
eventual effects that matter.
Because traits that confer fitness loss to the wild relative will be negatively selected
against, it will be neutral traits, or traits that are under positive selection pressure that will be
subject to stable introgression into the receiving wild genome. The workshop consecutively
discussed which traits will have greatest change to establish in the wild genome. In general,
traits conferring tolerance or resistance to biotic or abiotic stress could bring such advantages:
resistance to insects, nematodes, virus, and fungi. Among abiotic stress traits, tolerance to
temperature changes, salinity levels, moisture content or drought, and also changes in edaphic
factors (aluminium content, pH-level, nutrient level, etc.) will be important. Bt GM crops are
seen as a first test case in this respect, better than the first generation of GM crops that only
contains herbicide resistance traits.
The workshop reached the conclusion that, among other things, population sizes and
ecological relations of wild relatives may change due to introgression of crop traits, but there
was also the common opinion that the outcome of these processes will be difficult to predict,
if possible at all, given, for instance the current lack of insight in and understanding of
invasion processes and the traits that confer invasivity.
General conclusion of this part of the workshop: Gene flow is common, but not a hazard per
se. In the Environmental Risk Assessment formula it rather represents the exposure term, not
effect (=hazard). Similarly, the effect of gene flow leading to any change in fitness of a wild
relative is not by definition a hazard.
Whether or not gene flow and its consecutive processes lead to concern in the context of ERA
will be determined by the answer to the question: what is the damage?
187
188
The workshop extensively discussed the problem how to define damage: Here, value
judgements need to be done based on a broader than only scientific base.
Like many presentations during the IOBC meeting showed, certain changes in arable field
fauna and their trophic relations are frequently found when Bt crops are grown. The workshop
was of the opinion that many interesting studies have been done, many other are obviously
underway, but changes reported so far, seem to be far from dramatic, and in many cases from
the statistics point of view only marginally significant.
Further to this, the workshop participants agreed that in many cases, base line reference
data are missing, so that it proofs to be difficult to weigh the changes found to the ecosystem
dynamics as they fluctuate under current agricultural practice, using conventional crop races.
The participants expressed the need for explicit base line data to refer to, and also explicit
criteria for assessing damage.
The participants distinguished several levels/area’s of potential impact effects:
 In-crop matters
 Do we need to worry about the within-agricultural field biota?
 How will upscaling to the landscape level affect impact?
 Is policy directed to conserve historical romantic (?) patterns?
 Off-crop biodiversity, surrounding areas, (nearby) nature reserves (Ecological relationships
and structure on landscape scale)
 Trophic levels, ecosystem consequences
 Effects in time/space including changing baselines, independent from GM crop
developments (climatic changes, urban developments, changes in agricultural patterns –
bio-agriculture)
 Agronomic vs. ecological impacts
What are key issues? Will the upcoming EU Directive for Environmental Liability give
sufficient criteria?
The participants realised that the above considerations were well beyond the very workshop
theme of hybridization, and finalised by putting together a set of suggestions to IOBC, that
could help our organization to improve integration of studies and results, and give society the
barely needed clear answers as soon as possible.
Recommendations to IOBC
1. IOBC could further stimulate networking among the risk assessment workers, and organise
well targeted workshops, on which basis researchers could better co-ordinate their studies.
2. Developing data bases and meta-analyses over the wide array of publications could help to
reach integrated conclusion. There appears to be a strong need for integration of the vast
body of studies done so far.
3. The organisation should be broader, to include more students from disciplines like botany.
4. IOBC should seek possibilities to convince regulators that evaluations of GM organisms
should explicitly include a weighing of costs (the possible risks) and the (hopeful) benefits!
189
References
Eastham, K. & Sweet, J. 2002: Genetically modified organisms (GMOs): The significance of
gene flow through pollen transfer. European Environment Agency, Kopenhagen,
Environmental issue report No 28: 75 pp.
(http://reports.eea.eu.int/environmental_issue_report_2002_28/en/GMOs%20for%20ww
w.pdf)
Ellstrand, N.C. 2003: Dangerous Liaisons?: When Cultivated Plants Mate with Their Wild
Relatives. Series: Syntheses in Ecology and Evolution. The Johns Hopkins University
Press, Baltimore: 268 pp.
190
pp. 191-192
Workshop report - Impact of GM crops on pollinators
Dirk Babendreier1, Stefan Kühne2
Agriculture,
Reckenholzstr.
191,
8046
Zurich,
Switzerland
(E-mail:
2
[email protected]); Federal Biological Research Centre for Agriculture and
Forestry (BBA), Institute for Integrated Plant Protection, Stahnsdorfer Damm 81, 14532
Kleinmachnow, Germany (E-mail: [email protected])
1
The workshop on the impact of GM-crops on pollinators was attended by a small group of 6
scientists from Italy, Denmark, Switzerland and Germany. The participants focused on the
influence of Bt-toxins on bees, methodological aspects of bee experiments, difficulties in the
statistical analysis of experiments with social insects, the importance of bees for gene flow
and the role of solitary bees as model organisms for a risk assessment.
As a first step, the participants agreed that this IOBC/wprs group should focus on bees
rather than on all pollinators. First, bees are altogether the most important group of
pollinators. Second, some other important pollinators, such as syrphids, are already being
covered by the group working on natural enemies. Finally the group decided that other
pollinators that may also be important in certain crops such as Pollen beetles should not be
covered by the group. Though the opinion was raised by a participant of the conference that
such herbivores might be included in a risk assessment of transgenic plants, the group agreed
that it is not useful to include pest species that may cause significant damage in agriculture.
As a first topic the group discussed the impact of Bt crops on bees. The aim was to figure
out whether we could conclude on the risks for bees based on available literature and
unpublished data. Such a conclusion should be interesting because Bt crops are already grown
on a large scale worldwide. The group noticed that there is a lot of information available on
the effects of Bt toxins and Bt crops on bees but that these mainly focused on honey bees and,
to a lesser degree, on bumble bees. It was agreed that for this group of eusocial bees there is
no evidence for negative effects due to the growing of Bt crops (see also Malone & PhamDèlegue, 2001).
The group than discussed what gaps could be determined in the risk assessment for bees.
It was concluded that a very large group of bees was virtually neglected, i.e. the solitary bees.
This group is species-rich and often abundant and important for pollination of crop and wild
plants. Moreover, a comparison of the life history of solitary bees and social bees reveal some
large differences that may be important to determine the risk of growing GM plants. For
instance, honey bee larvae are fed with very small amounts of pollen (which may contain the
transgene compound) (Babendreier et al., 2004) while the offspring of solitary bees in general
do feed virtually only on pollen.
As another problem it was stated that very few long-term studies have been carried out so
far. In this context, it was acknowledged by the group that studies on the colony level of
social bee species are difficult and laborious. For this reason, often only few replicates have
been carried out with the result that such studies may lack the statistical power to detect
significant effects. Though this issue was not discussed in depth it was agreed that this
problem should not be ignored in future studies. One possible solution to this problem could
be to work on microcolonies (small groups of workers with or without a queen) as has been
done already in studies dealing with the risk assessment of pesticides.
191
192
Finally, the workshop discussed the importance of bees for gene flow and whether we do have
methods to guarantee the thresholds levels of < 0.9% or even less as requested for
coexistence. Recently, a report based on a large scale study in the UK (Ramsay et al., 2003)
supported the view that bees are very important for gene flow, especially when distances of
more that 10 m are considered. However, this study was conducted on oil seed rape and it is
obvious that the impact of bees for gene flow is negligible in some other crops which are not
or only rarely visited by bees.
Regarding the potential cooperation with the Global IOBC group it was agreed that
Salvatore Arpaia who attended this workshop, but also is leading the global activities of the
IOBC on the effects of GM crops on pollinators, should try to find overlap and connections
between the two groups.
In addition to these interesting discussions and statements, the workshop was very
productive in stimulating new cooperations between the workshop participants.
Questions/topics for further discussion and activities
 It is being proposed that methods have to be adapted in order to test for effects on solitary
bees. Solitary bees may be tested in addition to honey bees for effects of upcoming
transgene products.
 Furthermore, there is a need for more long term studies.
 It should be clarified what studies should be done during pre-release risk assessment vs.
post release studies.
 How much should we consider the pollinators themselves and how much the pollination
service as an ecological function?
References
Babendreier, D., Kalberer, N., Romeis, J., Fluri, P. & Bigler, F. 2004: Pollen consumption in
honey bee larvae: a step forward in the risk assessment of transgenic plants. Apidologie
35: in press.
Malone, L.A. & Pham-Dèlegue, M.H. 2001: Effects of transgene products on honey bees
(Apis mellifera) and bumblebees (Bombus Sp.). Apidologie 32: 287-304.
Ramsay, G., Thompson, C. & Squire, G. 2003: Quantifying landscape-scale gene flow in
oilseed rape Final report of DEFRA Project RG0216.
http://www.defra.gov.uk/environment/gm/research/reports.htm
pp. 193-195
Workshop report - Impact of GM crops on natural enemies
Jörg Romeis
Agriculture, Reckenholzstr. 191, 8046 Zurich, Switzerland
(E-mail: [email protected])
During the first meeting of the IOBC/wprs working group entitled ‘Ecological impact of
genetically modified organisms’ from November 26-29 in Prague, a workshop was hold that
dealt with the possible impact of genetically modified plants on entomophagous arthropods.
More than 30 participants from universities and governmental research institutions, private
industry and regulatory agencies attended the workshop.
Due to the fact that most participants had a background in entomology, the focus of the
discussion was on insect-resistant GM crops. In addition to commercialised Bt-transgenic
crops (maize, cotton, potato), lectin-expressing plants were also discussed.
The possible effects of GM crops on entomophagous arthropods is a major concern, since
these organisms play an important role in natural pest regulation and may affect the
development of resistance towards the transgene product in the target pest. Thus, good levels
of compatibility between GM-based strategies with biological control is necessary for a
sustainable deployment of a GM crop (i.e. within an IPM framework).
The group agreed on the fact that research should focus on selected species or functional
groups rather than doing full faunistic evaluations, since the latter often lack the statistical
power to detect significant effects largely due to the limitations in sample size. It was
suggested that all functional groups relevant to a particular crop should be assessed taking
into account the information available for the trait (mode of action of the insecticidal protein,
history of use, expression pattern,…) but only certain groups would need testing. Testing
would be necessary for those groups where there was a potential risk identified, i.e. exposure
could not be ruled out or hazard could be possible due to the mode of action of the trait.
The group discussed the criteria on how to select species to be tested in risk-assessment
studies of GM crops that have been proposed by Dutton et al. (2003; BioControl 48: 611636):
i) Economic/ecological importance in the crop
It was agreed that species should be selected that play an important role in the
agroecosystem. However, we were aware of the fact that we often lack the information on
the importance of a specific entomophagous arthropod in respect to its role in regulating a
herbivore population.
ii) Likelihood of exposure to the transgene product (toxin)
Potential exposure was regarded as an important criterion for test species selection. To
make a decision, information must be available on where and when a transgene product is
expressed in the plant and on the feeding behaviour of both the herbivorous and
entomophagous arthropods. Since a detailed crop characterization is necessary for the
application for consent to cultivate a GM crop according to Directive 2001/18/EC,
protein expression data are typically generated by the notifier.
There was a strong feeling expressed by some members of the group, that, where
appropriate, the risk assessment also may include trophic systems that are of importance
193
194
for the agroecosystem, even though the transgene product may not be transported among
certain species groups and/or trophic levels (e.g. aphids and their antagonists in Bt-maize
events which do not express Bt-toxins in phloem sap). The basis for this is the fact that
risk assessment should also evaluate potential ‘unintended adverse effects’ of the genetic
modification, if they occurred, and ensure a sustainable use of the GM technology in IPM
systems.
iii) A diversity of species should be considered
The participants agreed that, in general, the species selected should be selected from
different taxonomic or functional groups. However, there might be reasons to test more
species from a certain taxonomic order (see next criteria).
iv) Knowledge on the toxic specificity of the insecticidal protein
When selecting the test species, one should take into account the known specificity of the
transgene product. For example, plant lectins in general are known to have a broader
range of activity when compared to Bt-toxins. Therefore more species have to be tested
for a broad spectrum toxin (e.g. lectin) than a more specific toxin (e.g. Bt-toxin). Also, in
cases where the GM plant expresses a Bt-toxin with a known specificity on a particular
insect order, more attention should be given to assessment of the non-target species from
this particular order (e.g. predatory Coleoptera in the case of Bt-potato expressing the
Cry3A toxin).
v) Amenability and availability of the species
While we agreed that this is not a purely scientific criterion, it should not be neglected.
Amenability and availability should not be the driving force for the selection of a
particular test organisms. However, to use well-understood and easy to rear species will
for example increase the reproducibility of the tests in different labs.
Compromises have to be taken. For example in the case of soil organisms, most
predatory arthropods in the soil are difficult to rear and therefore difficult or impossible
to test. Therefore, soil studies might have to focus more on soil functions rather than on
individual species or their interactions. This could, for example, be done in mesocosm
studies.
The second phase of the workshop dealt with the question on when and why are field
evaluations required in risk-assessment studies. Three areas were identified:
i) Assessment of ‘unintended effects’
Where potential ‘unintended effects’ of the genetic modification occurred, they should
also be considered in the risk assessment, rather than the novel protein only. To study
effects that result for example from interactions between the transgene product and
secondary plant compounds, it would be important to grow the GM crop in a realistic
environment (temperature, photoperiod, planting density, exposure to normal biotic and
abiotic stress factors). Crop characterization that is conducted prior to registration of a
new GM crop includes plant composition analysis. This knowledge might enable us to
detect whether or not any ‘unintended effects’ of the genetic modification have occurred,
to provide key elements for consideration in the risk assessment, or, where applicable, for
case-specific monitoring of certain ‘unintended effects’ (e.g. plant metabolites with
known ecological function in defence, attraction of pollinators etc.).
ii) Detection of long term effects
Pre-release risk assessment studies might reveal effects that need to be further evaluated
in a case specific monitoring to detect possible long-term effects.
iii) Compatibility of a GM crop with other IPM measures
195
The group considered that compatibility of a GM crop with IPM practices is important
for sustainable pest control. However, studies on the compatibility are more likely a postrelease (after commercialization) type of work.
Questions that were only raised, but that need further discussion include:
 What are the requirements for pre-release risk-assessment (EU directive, EPA guidelines)?
What could/should be done post-release?
 We agreed that a clear structure for a risk-assessment is required. However, there was no
agreement on whether it should follow a clear tiered system approach, like the pesticide
testing. Furthermore, decision points (trigger values) have to be identified and agreed.
 There was a strong opinion that risk-assessment of GM crops should be put into
perspective, i.e. in the larger context of agricultural production, in particular with
conventional pest control practices such as the use of insecticides or existing IPM systems.
But the question remained on when this should happen in the course of the risk assessment.
The conventional practice should be considered in a needs analysis, prior to the
production/release of a GM crop. Conventional practise should also be considered later in
the decision making process, when deciding whether to field release a GM crop or not.
 A discussion is needed on why possible ‘unintended effects’ are assessed for GM crops but
not for all conventionally bred varieties with similar traits. It was argued that unintended
effects affecting human/animal nutrition or ecological services (pollinators, natural
enemies) should be checked for in both conventional and GM crops.
196
pp. 197-201
Workshop report - Biodiversity implications off-crop
Andreas Lang
Bavarian State Research Center for Agriculture, Institute of Plant Protection, Lange Point
10, 85354 Freising, Germany (E-mail: [email protected])
Summary of the workshop compiled in collaboration with F. Bigler, A.N.E. Birch, M. GarciaAlonso, M. Navrátilova and F. Szentkirályi
Introduction
The workshop “Biodiversity implications off-crop” took place on Friday, 28th November
2003, between 14:00 and 17:00 h. Altogether 16 people from seven nations participated (Tab.
1). The main topics of the workshop were (i) effects of transgenic crops on non-target
organisms in off-crop habitats, and (ii) monitoring the possible GMO effects in off-crop areas.
First, Andreas Lang gave a short general introduction on biodiversity, stating that in general
biodiversity is considered as species richness for reasons of convenience, though it comprises
the variety of life at all levels (e.g. genetic, species, functions, ecosystems, biomes).
Biodiversity is recognized as a value by itself, and is therefore subject to protection. Often
there are ethical reasons to protect biodiversity, e.g. for species protection. Frequently unsaid
assumptions imply that higher biodiversity causes a higher productivity, a higher stability and
a higher sustainability of agro-ecosystems (which may not be true in all cases, and may often
depend on the specific circumstances).
Biodiversity can be considered within the transgenic crop fields themselves as well as in
other fields, and in off-crop areas, the latter consisting of field margins, neighbouring (semi-)
natural habitats or water bodies. In the context of risk assessment of GMOs, biodiversity is
mainly a post-release and a field research issue, and is also closely related to monitoring
aspects such as selection of bio-indicators, monitoring design, definition of effects and
ecological hazard, and threshold values. This includes Flora and Fauna, and all possible
trophic levels (e.g. predators, parasitoids, pollinators, detritivores, plants). However, the
workshop was constrained to invertebrates, according to the expertise of the participants.
Further, the discussions were limited to Bt maize and Lepidoptera as a case study (i.e. case
specific monitoring), because Bt maize is most likely the first transgenic crop to be cultivated
across Europe. (It should be noted, however, that there was no general consensus that
Lepidoptera are the most important taxon for monitoring Bt maize effects, i.e. should they
only be considered for those GM maize products expressing toxin in the pollen?)
The discussions during the workshop focused on the following main points: Indicator
species, Bt toxin susceptible species, monitoring issues, information already available, other
examples of risk assessments. At the end of the day it was concluded to take future activities
and special task subgroups were established.
Selection of indicator species
There was a general agreement that a step-by-step process is a sensible approach to select
indicator species, both for pre-release risk assessment and post-release monitoring. In general,
the selection should follow an exclusion procedure, i.e. only species sensitive to Bt, occurring
close to maize fields, and potentially exposed (feeding on maize parts expressing the toxin)
should be considered. In essence, this brings about the need to develop a decision tree.
197
198
Possible criteria of such a decision tree could include: The susceptibility of lepidopteran
larvae to Bt, an overlap in time of the occurrence of larvae and pollen shedding, a spatial
overlap of Bt maize field margins and larval habitats, the occurrence of larval host plants near
maize fields within potential pollen deposition area, the utilization of host plants by larvae,
exposure to the toxin (like feeding on maize leaves expressing Bt or food plants dusted with
GM pollen,…), and others. It was agreed that a certain risk or a direct effect by the transgenic
crop must be proven, anticipated or at least plausible in order to justify the implementation of
a monitoring and subsequent selection of indicator species. However, in some cases potential
indirect effects may also validate a field survey if identified in pre-release risk assessment.
(During the workshop discussion we did not consider possible indirect effects of Bt maize)
Susceptibility of larvae to Bt toxin
As it is known that different butterfly/moth species differ in their susceptibility to the Bt
toxin, information about the vulnerability of the species is clearly needed, especially with
regard to the selection of bio-indicators. However, the workshop group members recognised
that only few papers are published on the effect of Bt maize pollen on lepidopteran larvae
apart from the publications about the monarch butterfly. However, several participants of the
workshop had data or publications in preparation (and not published yet) or knew of people
who had useful data. It therefore seems that there is much more data available than known so
far, and it was concluded to try to collect this information (see section Future Tasks).
Monitoring issues
In general, the main topics discussed with regard to monitoring issues were that in a (casespecific) monitoring only assumptions derived from the risk assessment should be followed
up, that a monitoring should be based on a decision tree, that a correct comparison to the
transgenic crops should be considered, and that the variance of the field data must be
accounted for somehow. In particular, as a baseline for comparison, the current management
practice for conventional maize should provide the comparison for a control against future
effects of Bt maize. Another possibility (or additional necessity) is the comparison with prerelease data and the analysis of time-series. It has to be noted, that the latter would require a
monitoring period of at least one year before the cultivation of transgenic crops, in order to
have information about an unaffected “nil situation”.
It was realised that the high variance of monitored field data will possibly obscure any
potential effects of GMOs. This variance, e.g. in butterfly/moth numbers, can be caused by
regional differences, by farming systems, by management practices (pesticides!), by field
margin types (food plant composition) and other nearby crops, by the population dynamics
and metapopulation structure of Lepidoptera (sink and source habitat types causing variable
immigration/emigration rates), and many others. This problem can be counteracted either by
taking into account the variables causing the variance or by increasing sample sizes; however,
both approaches will result in increased efforts in terms of time and money needed. Likewise,
the area to be monitored will determine the necessary effort and expenses (how far is offcrop?). In the case of Bt maize and Lepidoptera, it was agreed that only habitats and field
margins very close to the fields would need to be considered, because the majority of the Bt
maize pollen is not dispersed very far. However, for endangered or rare butterfly and moth
species, one would also need information on a landscape level, about existing and utilized
habitat patches, migration data, and naturally occurring population dynamics. There was some
discussion about the number of indicator species to be monitored (5-6 species? 10 species?
199
More?), which did not come to a common conclusion. However, it was felt that the number of
indicator species will strongly be influenced by the decision tree, which will identify the
susceptible and exposed species. Another point raised but not answered was whether
monitoring adult butterflies and moths is the right stage to be counted, because it is the larvae
which are exposed to Bt pollen, and adult counts, which mainly reflect foraging activity, may
not necessarily be correlated to numbers of larvae. The concepts of “ecological damage” and
“threshold values” were considered to be very crucial and urgent. When designing monitoring
schemes, we will need a good understanding of which effects are to be considered as being
ecologically relevant, and would therefore potentially justify the termination of a given GMO
cultivation. However, it was also acknowledged that we know close to nothing about how
certain effects can be translated to such threshold values, e.g. what a detected effect such as a
10% reduction in population size of a certain species would mean ecologically. Similarly, we
were uncertain how to translate the size of an effect measured under laboratory conditions to
what might happen under realistic field conditions (e.g. is < 50% effect in the lab okay, or
< 30% in the field, compare the pesticide experience?). A further question was posed: Would
it constitute no hazard if the Lepidoptera population recovers by the next year? Following on
from this point, how many years should effects be measured and how accurately can we
measure an effect?
Information already available
We discussed how much information is already available about the distribution and
occurrence of butterflies and moths in different European countries. Such information would
help in the selection of the possible indicator species, would allow to adjust for regional
differences, and would surely reduce the need for some additional studies in general. Ferenc
Szentkirályi reported about an existing long-term data set of Macrolepidoptera in Hungary,
which is based on catches of a light trap network from over 35-40 years from more than ten
different sites. Franz Bigler reported the Swiss biodiversity monitoring programme which
includes the counting of all butterfly species in a transect of 2.5 km seven times per year. The
total number of species observed ranges from 10 to 70 depending on the habitats represented
in the transects. The programme is run since 2002 on a total of 100 transects per year
throughout Switzerland. The overall costs for the butterfly programme alone amounts to 0.3
million Euro per annum, and includes a special training of the monitoring persons (40
butterfly specialists). Andreas Lang mentioned the existence of a database in Bavaria called
“Artenschutzkartierung” (= species protection monitoring) of the Bavarian Agency for the
Protection of the Environment (Bayerisches Landesamt für Umweltschutz, http://
www.bayern.de/lfu/natur/arten_und_biotopschutz/ask/index.html), which exists since 1980.
Achim Gathmann told about a database called Lepidat provided by the Federal Agency of
Nature Conservation (Bundesamt für Naturschutz, www.bfn.de). Jeremy Sweet mentioned the
UK butterfly monitoring scheme (www.bms.ceh.ac.uk), which was set up in 1976 to provide
information on changes in the abundance of butterflies. Currently this scheme receives data
from about 130 sites annually, and the butterflies are recorded weekly from April to
September by walking a fixed transect at each site. Alan Dewar also mentioned the
Rothamsted light trap database (26 traps), which has been extended to cover 100 sites
throughout the UK (www.rothamsted.bbsrc.ac.uk/pie/IanGrp/IanLight.html). This network
started in 1965 but some data exist from Rothamsted farm back to 1933. Within this network
all larger moths are identified daily throughout the year. Further light traps are operated in
Denmark,
Sweden,
Finland,
the
Baltic
states
and
other
countries
(www.rothamsted.bbsrc.ac.uk/pie/Graphics/LightTraps2002.gif). The impression of all
200
members of the workshop was that most probably such information is also available from
other countries but is not currently known to us. The collection and compilation of such
information was considered to be an important task (see section Future Tasks).
Other examples of risk assessments
The group wondered if this kind of workshop or the IOBC/wprs in general can have any
impact on the development of regulatory guidelines. In response, Franz Bigler and Ramon
Albajes reported their 25 years experience with the IOBC/wprs working group “Pesticides
and beneficial organisms”. In this working group scientists together with regulators and
industry representatives developed guidelines for the risk assessment of pesticides, which are
now implemented in EU regulations or followed by other countries. Although there was a
general agreement that the pesticide situation is clearly different from the GMO situation,
nevertheless we may learn some lessons from the pesticide risk assessment example.
Therefore, the members of the workshop would appreciate a talk about this subject at the next
meeting, and Franz Bigler agreed to look for a potential speaker. Alternatively, or
additionally, people from other research areas of risk assessments may be invited, e.g. from
the Boxworth experience (UK) with farming systems, from monitoring programmes in the
United States (natural enemies, pesticides, GMOs), or from biological control with parasitoids
and side effects on non-target arthtopods including butterflies (FAL Zürich-Reckenholz, CH).
Future tasks
1.) Decision Tree. A “sub-group” was founded with the aim to develop a decision tree for the
selection of butterfly species indicative of Bt maize effects in the field (members: J. Sweet,
M. Garcia-Alonso, N. Birch, A. Lang). It was agreed that the first task would be to check
whether other groups have already developed such a decision tree (in order to prevent double
work), and to find and examine already published decision trees (e.g. Schmitz et al. 2003).
Monicá Garcia-Alonso and Jeremy Sweet agreed to check concerning activities in the
industry and elsewhere, and to report on it before the group starts to work out a final decision
tree.
2.) Susceptibility of lepidopteran larvae to Bt toxin. Michael Meissle volunteered to compile
data on the susceptibility of larvae to Bt (but would nevertheless greatly appreciate every
person joining and supporting him). It was agreed that the list should only comprise results of
studies with GMOs (i.e. excluding Bt sprays), and that the list should cover both butterflies
and moths (and their susceptibility to pollen, green tissue, artificial diets). The first activity
will be to identify and contact all potential information sources (e.g. data from US regulators,
IOBC/nrs, INRA France, industry, biopesticide act in Hungary, etc.). A possible strategy to
get information may be to distribute a questionnaire among the attendants of the Prague
conference, or within the IOBC bulletin/newsletter. The next step would be to identify the
information to be included in such a compilation (for instance, possible columns could be the
species tested, developmental stage tested, country tested, specific Bt toxin type tested,
methods, lab/semi-field/field, effects, reference, contact persons, and others), and to insert the
collected data (and possibly rank the species according to their susceptibility). Eventually, a
validated and compiled list can then be published in the IOBC bulletin or newsletter.
3.) Lepidoptera databases in Europe. It was concluded that there is a need for an overview on
existing information about field data of butterflies and moths. It was anticipated that a
comprehensive list would be very valuable, including such information as source of the
database, contact persons, region covered, habitats, phenology, accuracy and recentness of the
201
data. Eventually, a list would be compiled where this information is summarized for as many
countries in Europe as possible, which would enable any person/institution looking for such
information to get a quick overview and a signpost. Andreas Lang volunteered as a contact
person for this task, and he would appreciate receiving any information concerning existing
databases of Lepidoptera in Europe. Additionally, promising institutions or authorative bodies
as well as participants of the Prague conference will be addressed directly. Further, it was
suggested to distribute a concerning questionnaire in the IOBC Bulletin and newsletter.
4.) Commercial lepidopteran breeders. The question was raised whether a list of commercial
breeders would be helpful once lepidopteran indicators for a Bt risk assessment have been
affirmed. This was identified as a relevant task, but postponed, because the species required
clearly depend on the results of the above points (susceptibility of different species, bioindicator selection, the implementation of a specific decision tree, monitoring decisions).
5.) The meeting of the workshop “Biodiversity implications – off crop” was generally
appreciated, and it was decided to gather again at the meeting in Montpellier in September
2004 at a next full meeting of the working group “Ecological Impact of Genetically Modified
Organisms’. Possibly, results of the above constituted task groups can be presented as posters
or talks on this occasion. A forthcoming topic of the next meeting should also be monitoring
methods and design, an issue not touched in the discussion of the Prague workshop. Also, the
subgroup “Biodiversity” would like to invite and to bring in experience from outside in terms
of talks by researchers from other but similar fields of risk assessment research.
Table 1. Participants of the workshop “Biodiversity implications off-crop”.
Name
Albajes, Ramon
Bigler, Franz
E-Mail
[email protected]
[email protected]
Birch, Nick
[email protected]
Garcia-Alonso,
Monica
Gathmann, Achim
Kaatz, Hannes
Landi, Simona
Lang, Andreas
Ludy, Claudia
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
Meissle, Michael
[email protected]
Area of interest/expertise
Monitoring, non-target
Monitoring, non-targets
Biosafety, chemical ecology, nontarget impacts
Environmental risk assessments of
gmos
Monitoring (butterflies)
Monitoring, non-target effects, bees
Butterflies, monitoring
Butterflies, monitoring
Non-target effects, spiders
Plant-insect interactions, non-target
effects
Navrátilova,
Miloslava
Pons, Xavier
Reuter, Hauke
Sanvido, Olivier
[email protected]
Monitoring, non-target effects
[email protected]
[email protected]
[email protected]
Sweet, Jeremy
[email protected]
Szentkirályi, Ferenc
[email protected]
Non-target effects
Modelling, monitoring
Design of monitoring
Environmental risk assessments and
monitoring of GMOs
Light-trapping, monitoring
p. 203
Workshop report - Resistance Management
Achim Gathmann
RWTH Aachen, Biology V, Worringerweg 1, 52056 Aachen, Germany (E-mail:
[email protected])
During the first IOBC/wprs meeting “Ecological impact of GMOs”, a resistance management
workshop took place. Altogether 12 people from 7 countries participated. At the beginning of the
workshop two questions were addressed (i) Is there an interest for a subgroup on resistance
management within the IOBC/wprs working group? This question needed to be answered in respect to
the low number of talks and posters about resistance management at the meeting in Prague. (ii) What
will be future activities of a subgroup on resistance management within the IOBC/wprs working
group?
All participants unanimously agreed to the first question. Transgenic plants are an upcoming
topic within the EU. A resistance management is essential to protect transgenic traits like insect or
herbicide resistance. The subgroup could be an open platform where scientist, regulators and people
from the industry could discuss issues regarding resistance management. Besides all participants
agreed that activities should focus on insect resistant Bt-maize in the near future because it is and will
be the forthcoming transgenic crop in Europe. That doesn’t mean the subgroup is not open for other
upcoming new issues or could learn from other projects in resistance research.
In the second part of the workshop we discussed about the future activities of the
subgroup and special objectives were identified. First, a network of scientist, regulators and
people from industry should be established. We identified a number of different national and
international research programs. We will try to initiate more exchange of knowledge and
hopefully will be able to coordinate future research activities. In particular, contact with
working groups in eastern parts of Europe, the Middle East and North Africa is desirable for
covering all areas of the IOBC/wprs. In a first step a database should be built up with data
from the working groups and their research fields. Therefore a questionnaire will be created
and distributed. All known working groups and stake holders should than be invited to
participate in future workshops. Secondly we addressed the completion of baseline data as the
most important point at the moment. Basis susceptibility, genetic diversity of European (ECB)
and Mediterranean corn borer (MCB) populations and frequency of resistance alleles are of
interest. Identified problems are among others the availability of a toxin standard, reference
strains for Europe, standard methods for testing and how many and how often ECB/MCB
populations should be tested. To go one step forward, we attempt to develop a guideline for
basis susceptibility tests.
In summary further tasks of the subgroup will be (i) to develop a questionnaire to get
information about research activities of different working groups (ii) to summarize and
discuss results from different working groups/projects (iii) to circulate this information and
protocols (iv) to attempt to develop a guideline for basis susceptibility tests in Europe.
203
204
pp. 205-208
Workshop report - Monitoring/Bioindicators
Salvatore Arpaia
ENEA, Italian National Agency for New Technology, Energy and Environment, S.S. 106
Jonica Km 419.5 I-75023 Rotondella (MT), Italy (E-mail: [email protected])
Introduction
As a consequence of the increasing concern about the biosafety of genetically modified plants
(GMP), more ecological studies are being developed in this field. As early studies were
mainly devoted to highlight possible pathways of transgenic toxic exposure to higher trophic
levels (e.g. Hilbeck et al., 1999; Birch et al., 1999; Losey et al., 1999), presently more efforts
are being addressed to analyse several components of the arthropod fauna along the food web
under natural conditions (e.g. Arpaia & Fiore 2003; Brooks et al., 2003; Haughton et al.,
2003; Szekeres et al., 2003; Wan Xue et al., 2003). Moreover, meaningful studies on soil
food webs have been produced recently (e.g. Gupta et al., 2003; Manachini et al., 2003;
Zwahlen et al., 2003), adding new knowledge on transgenic agro-ecosystem functioning.
A major impulse to consider monitoring as a key topic to be developed by field scientists
in the West Paleartic Region, comes from the new EU legislation on GMOs that endorsed a
compulsory post-release monitoring for GMPs to be released in the environment for
commercial purposes.
The workshop
Seventeen people (public scientists, industry representatives and regulators) attended the
workshop session during the IOBC/WPRS Meeting in Prague. Due to time constraint, some
apriori choices were made to restrict the focus of the discussion. First of all, a brief outline of
the legal framework was given and the participants agreed to concentrate on “case specific
monitoring” as opposed to “general surveillance” plans that also are considered by EU
legislation but that would deserve a wider, landscape approach that would require the joint
efforts of different expertise not available within this IOBC/WPRS working group. As a
consequence, it was also decided that the discussion would be concentrated on arthropods and
soil macro-organisms.
The “case specific monitoring” is planned to identify hypothetical adverse effects related
to a specific GM crop in comparison to non-GM farming systems as a benchmark. Insights
gained into environmental effects and risks through pre-release biosafety research will be
included in the planning of monitoring programmes with the goal to narrow down the subjects
to be monitored. It is assumed that GM crops are registered for commercial release only if the
pre-release biosafety data show that environmental risks are acceptable. Nevertheless,
questions related to long-term effects of GM crops will remain and monitoring has to be
carried out for several growing seasons to produce a scientific feedback that may also have
regulatory and commercial consequences.
205
206
Criteria for GMP monitoring
According to basic ecological concepts, two general criteria were put forward to guide the
preparation of any meaningful monitoring plan. First of all, there is the issue of selecting an
appropriate baseline level of diversity. We therefore suggest that comparative observations
should always be carried out, and the “current agricultural practices”
for the specific crop in the area of study being the most appropriate controls.
The second very broad rule is based on the consideration that agro-ecosystems in the
West Paleartic Region are indeed very different. Food webs and/or trophic interactions may
therefore be different in different areas; consequently monitoring programs should be
structured region wise. The subsequent logical step involves decisions about “what to
monitor” in a case specific monitoring program. As a guideline, a list of priorities was
proposed: ecological functions, biodiversity and anthropocentric values. The first leading
criterion is the study of ecological functions through the analysis of arthropod functional
groups. Generally speaking, there is very limited information about the actual ecological role
of most organisms in agro ecosystems, as only for some major pests and a few of their natural
enemies specific studies have been conducted at field level. It is therefore clear that an
ecologically meaningful analysis in agro-ecosystems will generally involve several species.
Five groups of organisms with different roles in food webs are suggested for field studies:
herbivores (target and non target), predators, parasitoids, pollinators and saprovores. Each
study should be conducted including trophic relationships between these functional groups,
and therefore food web dynamics should also be addressed.
A second, very important criterion is constituted by the estimate of agro-ecosystem
biodiversity. While we recognised that species based biodiversity is not the only possible way
to estimate animal diversity, since its use is widely understood and largely adopted even for
regulatory purposes (e.g. Rio Biodiversity Convention) we agreed that it represents a
meaningful way to evaluate biodiversity. There is a large array of numerical indices that have
been proposed for measuring biodiversity. The discussion in the workshop was not aimed at
analysing pros and cons of each existing index, nevertheless we agreed on the opinion that
simple  diversity indices (e.g. Shannon-Weaver) are not adequate enough for comparative
studies. Therefore the combined use of several indexes (e.g. profile analyses, multivariate
indices) is suggested for analysing, comparatively, data matrices obtained from the
observations of species assemblages in transgenic and control fields and field margins.
Another major advantage of some multivariate indices is that their estimate does not change
when Organismal Taxonomic Units (OTUs) are used, provided that the ecological functions
are correctly interpreted. The limited taxonomic expertise that might be experienced during
monitoring plans can then be overtaken with this approach.
Finally, a third aspect may be necessary to build specific monitoring plans:
anthropocentric values need to be also considered. Independently from their ecological role,
some species may represent special interests for humans. For example, endangered species or
species of cultural significance (e.g. the Monarch butterfly in the U.S.A.) might be studied in
order to assess any possible harm due to transgenic plants. Some other species might
constitute a direct or indirect source of income for growers and therefore a special attention
for them is justified by economical considerations.
207
This report represents only a first approach to this very important issue. The working group
should further stimulate networking with the aim of preparation of more detailed guidelines,
taking also into account similar attempts that are currently being made in different geographic
areas. These guidelines might be useful at various levels:
- at a scientific level, to help scientists to produce comparable studies at the highest possible
scientific standard, relying on the largest possible agreement upon experimental design
and data analysis;
- at regulatory level, as these guidelines could be offered to EU, national agencies and
companies involved in GMP post-release monitoring
References
Arpaia, S. & Fiore, M.C. 2003: The study of species assemblage as an estimate of insect
biodiversity in experimental crop fields. Proceedings of the Workshop on Biodiversity
Implications of Genetically Modified Plants, September 7-12, 2003 Monte Verità,
Ascona, CH: 6-7
Birch, A.N.E., Geoghegan, I.E., Majerus, M.E.N., McNicol, J.W., Hackett, C.A., Gatehouse,
A.M.R. & Gatehouse, J. 1999: Tritrophic interactions involving pest aphids, predatory 2spot ladybirds and transgenic potatoes expressing snowdrop lectin for aphid resistance.
Mol. Breed. 5: 75-83.
Brooks, D.R., Bohan, D.A., Champion, G.T., Haughton, A.J., Hawes, C., Heard, M.S., Clark,
S.J., Dewar, A.M., Firbank, L.G., Perry, J.N., Rothery, P., Scott, R.J., Woiwod, I.P.,
Birchall, C., Skellern, M.P., Walker, J.H., Baker, P., Bell, D., Browne, E.L., Dewar,
A.J.G., Fairfax, C.M., Garner, B.H., Haylock, L.A., Horne, S.L., Hulmes, S.E., Mason,
N.S., Norton, L.R., Nuttall, P., Randle, Z., Rossall, M.J., Sands, R.J.N., Singer, E.J., &
herbicide-tolerant and conventional spring crops. I. Soil surface-active invertebrates.
Phil. Trans. R. Soc. Lond. B 358: 1847-1862
Gupta, V.V.S.R., Watson, S. & Roberts, G. 2003: Impacts of cultivation of genetically
modified plants on soil biota communities and biological processes. Proceedings of the
Workshop on Biodiversity Implications of Genetically Modified Plants, September 7-12,
2003 Monte Verità, Ascona, CH: 20-21
Haughton A.J., Champion, G.T., Hawes, C., Heard, M.S., Brooks, D.R., Bohan, D.A., Clark,
S.J., Dewar, A.M., Firbank, L.G., Osborne, J.L., Perry, J.N., Rothery, P., Roy, D.B.,
Scott, R.J., Woiwod, I.P., Birchall, C., Skellern, M.P., Walker, J.H., Baker, P., Browne,
E.L., Dewar, A.J.G., Garner, B.H., Haylock, L.A., Horne, S.L., Mason, N.S., Sands,
R.J.N., & Walker, M.J. 2003: Invertebrate responses to the management of genetically
modified herbicide-tolerant and conventional spring crops. II. Within-field epigeal and
aerial arthropods. Phil. Trans. R. Soc. Lond. B 358: 1863-1877
Hilbeck, A., Moar, W.J., Pusztai-Carey, M., Filippini, A & Bigler, F. 1999: Prey-mediated
effects of Cry1Ab toxin and protoxin and Cry2A protoxin on the predator Chrysoperla
carnea. Ent. Exp. Appl. 91: 305-316
Losey, J.E., Rayor, L.S. & Carter, M.E. 1999: Transgenic pollen harms monarch butterflies.
Nature 3999: 214
Liu, W., Wan, F., Guo, J.& Lövei, G.L. 2003: Spider diversity and seasonal dynamics in
transgenic Bt- vs. conventionally managed cotton fields in china. Proceedings of the
Workshop on Biodiversity Implications of Genetically Modified Plants, September 7-12,
2003 Monte Verità, Ascona, CH: 28
208
Manachini, B., Fiore, M.C., Landi, S. & Arpaia, S. 2003: Nematode species assemblage in Btexpressing transgenic eggplants and their isogenic control. Proceedings of the Workshop
on Biodiversity Implications of Genetically Modified Plants, September 7-12, 2003
Monte Verità, Ascona, CH: 231-232.
Paine, T. 1969: A note on trophic complexity and community stability. Am. Nat. 103: 91-93.
Szekeres, D., Kádár, F., Szentkirályi, F. & Kiss, J. 2003: Structural characteristics and
seasonal dynamics of ground beetle (Coleoptera, Carabidae) assemblages collected in
experimental Bt- and isogenic maize fields in Hungary. Proceedings of the Workshop on
Biodiversity Implications of Genetically Modified Plants, September 7-12, 2003 Monte
Verità, Ascona, CH: 26
Zwahlen, C., Hilbeck, A., Howald, R. & Nentwig, W. 2003: Effects of transgenic Bt corn
litter on the earthworm Lumbricus terrestris. Mol. Ecol. 12: 1077-1086.
pp. 209-215
Workshop report - Soil Organisms and Functions
Wolfgang Büchs*
BBA, Institute for Plant Protection in Field Crops and Grassland, Messeweg 11/12, 38104
Braunschweig (E-mail: [email protected])
Introduction
Soil organisms and functions usually form a minor part of biosafety research programmes.
Soil organisms include a wide range of taxa such as bacteria, other microorganisms, plant
seeds, earthworms, collembola, diptera-larvae and other decomposers, carnivorous ground
beetles, soil-related stages of herbivorous pests, and ending up with functions as toxin
adsorption to soil particles, accumulation in the soil fluid, and persistance in plant residues
which are crucial soil functions to be considered in biosafety protocols.
To promote soil research, a workshop on „Soil organisms and functions“ was held during the
IOBC/wprs working group meeting „Ecological Impact of Genetically Modified Organisms“
in Prague on 26 November 2003. The workshop was joined by 13 colleagues representing
eight countries (Belgium, Czech Republic, Germany, Hungary, Israel, Italy, Switzerland and
the United States of America). The workshop discussion touched the following questions:
- What is currently going on and who is active in research on the GMO impact on soil
organisms and functions?
- Which soil taxa or topics are not well represented in today‘s biosafety research?
- Which are the major gaps in biosafety research regarding soil taxa and functions with
special reference to the GMO impact?
- Can a database of active research groups help to enhance interdisciplinary
communication and what is the appropriate form of such a database?
- Some research tasks (e.g. soil microorganisms and GMO´s) are well represented in
biosafety research but not in the IOBC/WPRS working group. How can they be
integrated?
- Inter- and transdiciplinary cooperation is of major importance in research on soil
organisms and functions related to GMO´s: How can it be promoted?
- Is there an interest to establish a group dealing with soil organisms in relation to GMO
environmental impacts and how should this group define itself and organise its work
within the IOBC working group.?
- If and how can/should research interests of the working group members meet political
requirements (e.g. features of reliable monitoring procedures)?
- What are the topics a subgroup intends to focus on in the future and which kind of
work (philosophy) should be appreciated?
It was the aim of the workshop to get an overview on the demands and needs of soil biologists
whose work is related to the GMO impact and to set a seed for future joint activities.
*
It should be considered that the majority of statements within this paper represent the results of a discussion of the
workshop participants. Thus, they do not reflect in all aspects the personal intention of the author nor do they represent the
view of the institution the author belongs to.
209
210
Database
As introduction, a short overview on the activities of research on soil organisms and functions
in connection to GMO impacts including an evaluation regarding potential gaps as there are
neglected taxa or research areas was presented by Wolfgang Büchs, who provided also a draft
proposal of a database on research groups active in soil ecology in relation to GMO impacts.
An example for the use of the database is shown in Table 1. It consists of informations on
- the research group: namely the person who is responsable for the research ideas or who
supervises the research conducted. It includes also informations on the nation the
research group is situated. An e-mail-address can be added.
- the GMO crops and the taxa considered as it can be derived from papers, abstracts etc.
- tasks, that means the main topic(s) a research group deals with or has a focus upon.
- references which are listed as far as they are easy to collate from reference databases or
provided directly by the author.
- Finally it was tried to order the work of a research group into more general categories, so
that the user of the database is able to find research groups active in the area he/she is
interested rather simple by using a keyword. Table 2 shows exemplarly which categories
were chosen and how they were used.
For clarity reasons the research tasks were summarized and ordered into six categories
(monitoring, toxin behaviour and persistance, horizontal gene transfer, functional aspects,
„single species tests“ and community related research) which seemed to be key components
of the research conducted.
Because in many cases informations from more than one reference/source of the same
research group were collated, the informations sampled in the database are not completely
consistant. That means, e.g. not every paper of a research group deals with all taxa or GMO´s
listed and not all research of one group was conducted only in the field or in the laboratory
respectively. However, this loss of detailed information is an inavoidable effect if information
is concentrated.
A first analysis of about 50 preliminary entries in the draft database showed that
particularly work on soil microorganisms (bacteria, fungi) is most well represented in
connection with GMO´s and their impacts. Furthermore, Collembola, Coleoptera-larvae
(herbivorous pests as well as carnivorous predators) are considered followed by Oligochaeta,
protozoa, Tracheophyta (seed banks in the soil), Nematoda and soil functions. All other taxa,
among them particularly soil Acarina or soil dwelling larvae of Diptera and Lepidoptera show
only single entries. A complete lack of research with regard GMO impacts was recorded for
some soil bound taxa particularly important in forest ecosystems (e.g. Protura, Diplura,
Thysanura, Symphyla, Pauropoda, Diplopoda, Chilopoda, Gastropoda, Rotatoria and
Lichens), which is likely to be caused by the fact that there are recently no GMO applications
in forests that are ready for commercial use in the near future. In some taxa, research is
currently focused mainly on more or less toxicological assessments of single species (e.g.
Oligochaeta, Collembola, herbivorous Coleoptera-larvae [mostly pests]) whereas research on
carnivourous Coleoptera-larvae, Nematoda, Protozoa, Tracheophyta and to a certain extent on
microorganisms has more a focus on community issues. There is a special task on horizontal
gene transfer which is nearly exclusively restricted to microorganisms.
The draft database was discussed and it was agreed to start with a database including
research groups that are active in soil ecology and GMO impacts. But also soil ecologist, not
dealing with GMO´s are invited to participate. It was realised that the best way to ensure that
211
entries in the database are correct would be if the input of informations would be done
directly by the research groups themselves. Therefore, the workshop participants agreed to
contribute inputs or manage contacts to relevant research groups in their home country and –
as far as possible – neighboured regions or countries.
Organising special sessions on „Soil organisms and functions“
In order to promote research on soil organisms in connection with GMO effects and due to the
large range of taxa and subjects and the overall complexity of soil ecolsystems it was strongly
recommended to continue the work started with this workshop by organising special sessions
on topics of interest in this area aiming to establish a subgroup on „Soil organisms and
Functions“ within the IOBC/WPRS working group.
It was noted that the largest share of research groups dealing with GMO´s is covered by
groups that work on microorganisms. These research groups meet usually in their own circles
and are currently not well represented in the IOBC/WPRS working group. However, a nearly
complete lack of interdisciplinary work and information exchange between microbiologists
and research groups dealing with other taxa in respect to GMO was identified. It was
therefore decided to encourage microbiologists to join the IOBC/WPRS working group which
particularly deal with microbial communities and their ecological functions and which are
interested to interlink their work with other soil taxa.
Self-understanding of the group „Soil organisms and functions“
It was agreed that the group on „Soil organisms and Functions“ is to be an open forum for all
activities in soil ecology in relation to GMO´s. This implies to address the complex
interactions between several soil organisms and their functions in respect to their ecological
consequences. That means their effects on ecosystem functions and on interrelations (e.g.
food webs, dominance structures, fitness) between populations of different species. Thus,
interdisciplinary work that interlinks for example microorganisms with higher taxa or abiotic
soil characteristics to soil organism communities under influence of GMO´s shall be a focus.
The group emphasizes its independency of any interests related to GMO releases and
registration tasks. It feels only obliged to enhance scientific knowledge, to foster
collaboration between scientists, to share experience and - as far as possible - facilities and to
implement joint research activities.
Future activities
Methodological approaches and protocols
Regarding the assessment of GMO impacts it was agreed that methodology has to be adopted
to each crop and soil type. It was recognized that the soil ecosystem has its very special
complexity and reacts very conservative with the consequence that effects are longlasting. It
has to be considered that GMO´s features (e.g. a herbicide tolerance or an insecticidal effects)
are not restricted to a limited time period after the application as it is for pesticides, but can
have an impact during the whole life time of the crop plant and beyond. This results in
demands for a complex and precautionary risk assessment of GMO effects on the soil
ecosystem. It was stated that in particular the pre-marketing risk assessment of a GMO crop
variety has to be conducted very carefully and should comprise obligatory field research on
all potential effects that are based on hypothesis independant from the likelihood they are
estimated to occur. Thus, a tier-based pre-marketing risk assessment starting with testing
212
methods which are restricted to functional endpoints as for instance C/N-degradation or other
summarising parameters, and continuing with testing procedures which approach from tier to
tier more and more the field situation (only if effects are detected) was considered by the
majority of the group as not appropriate to exclude the occurrence of detrimental effects once
the crop is commercially released. However, this point needs further discussion since the
group has recognized that the use of highly sophisticated combinations of several methods for
each soil taxon which lead to the most reliable statements will be not always feasible within
the registration procedure of GMO crops. In consequence, there is a need for simplified
methodological approaches and/or indicators to assess GMO effects. Therefore, one focus of
future activities will be the development of methodological protocols and their validation.
Methodological approaches to assess soil taxa shall be classified into two categories
A) An elementary methodological approach which is relatively easy to conduct regarding
input of time, labour or money and which demands only a fair or moderate skill
level. It is assumed that these methods provide only very basic informations on GMO
impacts with a very limited power of statement.
B) An advanced methodological approach oriented at scientific demands which needs
much more efforts regarding the methods applied, the experimental design, the
amount of time and labour and demands a high skill level for instance in taxonomy or
in the application of modern techniques (e.g. PCR and other biochemical methods). It
is assumed that the advanced methodological approach provides much more precise
informations of GMO effect. The power of statements is more reliable.
As elementary methods will be mostly demanded due to limitations in time and money within
the GMO registration procedure one focus within this methodological work should be the
validation and evaluation of these methods i.e.
- how limited or reliable are the statements derived from the results carried out with these
elementary methods;
- whether e.g. certain species or parameters easy to assess are appropriate as (surrogate)
indicators to assess for instance GMO effects on functions, biodiversity, abundance,
fitness etc.
Further tasks and focal points of discussion
Furthermore the discussion touched further gaps in research in respect to GMO´effects on
„Soil organisms and Functions“ which are designed to be in focus of the activities of the
group:
-
-
Assessment criteria (baselines, damage thresholds etc.)
Currently research is too much taxon related, i.e approaches are lacking that interlink
more than one functional group of soil organisms (e.g. interactions between
microorganisms and decomposers and/or interactions between decomposers and
predators) and thus, cover food chain effects.
Integration of conventional cultivars into GMO assessment: currently research and
assessments in most cases are restricted to the comparison of the GM cultivar and its
isogenic pendant. However, first results with decomposers including other conventional
cultivars show on the one hand that even conventional cultivars have the potential to
affect decomposer communities in a similiar range as GM cultivars do, on the other hand
they give indications that a genetical modification is not restricted to the feature which is
intended to be changed by the genetically modifiaction but changes the features of a crop
213
-
plant as a whole. This is particularly important regarding soil organisms as decomposers
or herbivores (e.g root feeders). In consequence the GMO crop plant has to be considered
as a complex ecological system which demands holistic approaches particularly to assess
impacts on soil organisms. The fact that conventional cultivars affect decomposers in a
similar extent as GM crops leads to questions whether registration of cultivars (either GM
or non GM) should consider effects on ecological functions as it is already established for
instance within the pesticide registration? Another question raises regarding risk
assessment of the effects of GM cultivars: Is it really tenable not to look for effects of
GM cultivars on soil organisms if already one GM cultivar carrying the same construct
has been registered?
Tests with isolated and/or synthesized toxins (test routines similar to that within
pesticide registration procedures) are extemly limited regarding their power to assess
GMO impacts, because GMO is a „living“ system permanently active and subject to
changes which is likely to cause unpredictable reactions and effects. This leads to
questions as: are test routines derived from ecotoxicology appropriate to detect GMO
effects? Do reflect so called „worst-case“-laboratory tests in any way the range of (partly
very subtle) effects that potentially can occur in field soils? Are statements derived from
such tests strong enough to decide whether a GM crop is „harmless“ or not and whether
further research is necessary? Or should test routines from the start be more oriented on
the real situation in the field (e.g. not using isolated/synthezised toxins but for instance
Bt-plant parts that are to be decomposed)? If actually „classic“ toxicological effects are
considered should research be more advanced, i.e. that not mortality but mechanisms of
the toxin activity and effects during digestion have to be put in focus? Is it sufficient to
test functional endpoint as indicators of the processes running beforehand? Do we need
more polyphasic monitoring and assessment approaches?
Once again the preliminary state of discussion has to be emphasized. The intention was
mainly to set seeds for further discussions. It is one future task of the group „Soil organisms
and Functions“ to discuss aspects mentioned above more deeply and comprehensive under
involvement of further research groups. Soil ecologists are invited to join the database and to
participate in future activities.
Acknowledgements
Thanks to all participants for the intensive and engaged discussions. Especially I thank Dr.
Barbara Manachini (Milano, Italy) and Dr. Sabine Prescher (Braunschweig, Germany) for
their assistance during the session and Prof. Dr. Frantisek Sehnal (Prague, Czech Republic) as
well as Dr. Joerg Romeis (Zurich, Switzerland) for helpful comments and support within the
elaboration of the workshop report.
214
Table 1. Featured example of a draft database on research groups working on soil organisms
and soil functions related to GMO impacts
Group: Stotzky et al., Laboratory of Microbial Ecology, New York University, New York, USA
GMO: Bt-corn, Bt-rice (Cry1Ab), Bt-Potato (Cry3 A), Bt- cotton, Bt-canola, Bt-tobacco (Cry1Ac)
Taxon: Microorganisms (?), earthworms, nematodes, protozoa, bacteria, fungi, (field?), bacteria,
fungi, algae (in vitro = Lab)
Task: Behaviour and persistance of Bt-Toxin in soil; effects of the toxin on several taxa
Level: Lab
Ref.:
Koskella J; Stotzky G (1997). Microbial utilization of free and clay-bound insecticidal toxins from
Bacillus thuringiensis and their retention of insecticidal activity after incubation with microbes.
Applied and Environmental Microbiology 63(9), 3561-3568.
Saxena D; Flores S; Stotzky G (1999). Insecticidal toxin in root exudates from Bt corn. Nature
402(6761), 480.
Tapp H; Stotzky G (1995). Dot blot enzyme linked immunosorbent assay for monitoring the fate of
insecticidal toxins from Bacillus thuringiensis in soil. Applied and Environmental Microbiology 61(2),
602-609.
Tapp H; Stotzky G (1998). Persistence of the insecticidal toxin from Bacillus thuringiensis subsp.
kurstaki in soil. Soil Biology and Biochemistry 30(4), 471-476.
Venkateswerlu G; Stotzky G (1992). Binding of the protoxin and toxin proteins of Bacillus
thuringiensis subsp. kurstaki on clay minerals. Current Microbiology 25(4), 225-233.
Crecchio C; Stotzky G (1998). Insecticidal activity and biodegredation of the toxin from Bacillus
thuringiensis ssp. kurstaki bound to humic acids from soil. Soil Biology and Biochemistry 30(4), 463470.
Tapp H; Calamai L; Stotzky G (1994). Adsorption and binding of the insecticidal proteins from
Bacillus thuringiensis ssp kurstaki and tenebrionis on clay minerals. Soil Biology and Biochemistry
26(6), 663-679.
Tapp H; Stotzky G (1995). Insecticidal activity of the toxins from Bacillus thuringiensis ssp. kurstaki
and tenebrionis adsorbed and bound on pure and clay soils. Applied and Environmental Microbiology
61(5), 1786-1790.
Saxena D; Stotzky G (2001). Bt toxin uptake from soil by plants. Nature Biotechnology 19(3), 199200.
Saxena D; Stotzky G (2000). Insecticidal toxin from Bacillus thuringiensis is released from roots of
transgenic Bt corn in vitro and in situ. FEMS Microbiology Ecology 33(1), 35-39.
Stotzky G (2000). Persistence and biological activity in soil of insecticidal proteins from Bacillus
thuringiensis and of bacterial DNA bound on clays and humic acids. Journal of Environmental
Quality 29(3), 691-705.
Crecchio C; Stotzky G (2001). Biodegradation and insecticidal activity of the toxin from Bacillus
thuringiensis subsp. kurstaki bound on complexes of montmorillonite-humic acids-Al
hydroxypolymers. Soil Biology & Biochemistry 33(4/5), 573-581.
Saxena D; Stotzky G (2001). Bacillus thuringiensis (Bt) toxin released from root exudates and
biomass of Bt corn has no apparent effect on earthworms, nematodes, protozoa, bacteria, and fungi in
soil. Soil Biology & Biochemistry 33(9), 1225-1230.
Saxena D; Flores S; Stotzky G (2002). Bt toxin is released in root exudates from 12 transgenic corn
hybrids representing three transformation events. Soil Biology & Biochemistry 34(1), 133-137.
Category: pure research; behaviour of Bt-Toxin; soil functions (decomposition); effects on species
and/or communities of soil animals
215
Table 2. Database: Research groups working on effects of GMO’s on soil organisms and
functions – Examples for categorisation
L = Laboratory; S = „Semi-field“ (e.g. microcosms etc.); F = Field
X = focus; (X) = weak focus; X = main focus

IOBC-WPRS Bulletin 27(3), 2004

Transcription

Similar documents

embassy of the republic of zambia

PROJECT DIRECTOR Pat Badani “Al Grano” (grain) evaluates the

9 December 2015, PE Express Indaba, Elzea Snacks celebrate 40

The Evolution of Plant Breeding: RTDS™ versus other

this PDF file

TECH TYKES TEAM NEWS

Do you know JOSERA?

Bioreactors and transgenic animals

Keep your silage cool with Pioneer brand inoculants.

The Newsletter of The Nature Institute