doi 101098rsbl20101147 published online 19 January 2011Biol Lett

Simone von Burg Frank J F van Veen Fernando Aacutelvarez-Alfageme and Joumlrg Romeis wheat

parasitoid community structure on genetically modifiedminusAphid

Supplementary data

mlhttprsblroyalsocietypublishingorgcontentsuppl20110118rsbl20101147DC1ht

Data Supplement

Referencesref-list-1httprsblroyalsocietypublishingorgcontentearly20110118rsbl20101147fullhtml

This article cites 19 articles 1 of which can be accessed free

PltP Published online 19 January 2011 in advance of the print journal

Subject collections

(482 articles)environmental science (2078 articles)ecology

(59 articles)biotechnology Articles on similar topics can be found in the following collections

Email alerting service hereright-hand corner of the article or click Receive free email alerts when new articles cite this article - sign up in the box at the top

publication Citations to Advance online articles must include the digital object identifier (DOIs) and date of initial online articles are citable and establish publication priority they are indexed by PubMed from initial publicationthe paper journal (edited typeset versions may be posted when available prior to final publication) Advance Advance online articles have been peer reviewed and accepted for publication but have not yet appeared in

httprsblroyalsocietypublishingorgsubscriptions go to Biol LettTo subscribe to

This journal is copy 2011 The Royal Society

on January 20 2011rsblroyalsocietypublishingorgDownloaded from

on January 20 2011rsblroyalsocietypublishingorgDownloaded from

Biol Lett

doi101098rsbl20101147

Published online

Community ecology

Aphidndashparasitoidcommunity structure ongenetically modified wheatSimone von Burg1 Frank J F van Veen2Fernando Alvarez-Alfageme3 and Jorg Romeis3

1Institute of Evolutionary Biology and Environmental Studies Universityof Zurich Winterthurerstrasse 190 8057 Zurich Switzerland2Centre for Ecology and Conservation College of Life and EnvironmentalSciences University of Exeter Cornwall Campus Penryn CornwallTR10 9EZ UK3Agroscope Reckenholz-Tanikon Research Station ART Reckenholzstrasse191 8046 Zurich SwitzerlandAuthor for correspondence (simonevonburgieuuzhch)

Since the introduction of genetically modified(GM) plants one of the main concerns hasbeen their potential effect on non-target insectsMany studies have looked at GM plant effectson single non-target herbivore species or onsimple herbivorendashnatural enemy food chainsAgro-ecosystems however are characterized bynumerous insect species which are involved incomplex interactions forming food webs In thisstudy we looked at transgenic disease-resistantwheat (Triticum aestivum) and its effect onaphidndashparasitoid food webs We hypothesized thatthe GM of the wheat lines directly or indirectlyaffect aphids and that these effects cascade up tochange the structure of the associated food websOver 2 years we studied different experimentalwheat lines under semi-field conditions We con-structed quantitative food webs to compare theirproperties on GM lines with the properties on cor-responding non-transgenic controls We foundsignificant effects of the different wheat lines oninsect community structure up to the fourthtrophic level However the observed effects wereinconsistent between study years and the variationbetween wheat varieties was as big as between GMplants and their controls This suggests that theimpact of our powdery mildew-resistant GMwheat plants on food web structure may benegligible and potential ecological effects onnon-target insects limited

Keywords insect food web powdery mildewnon-target effect community genetics

1 INTRODUCTIONInsect species that inhabit agricultural ecosystems areinvolved in complex interactions forming so-called foodwebs [1] The diversity and complexity of such foodwebs are considered to be important factors thatdetermine ecosystem function and stability Insect hostndashparasitoid systems are influenced by plant traits whichcan lead to large effects on food web structure [2]Due to their biology interactions between aphids primaryparasitoids and secondary parasitoids can easily be

Electronic supplementary material is available at httpdxdoiorg101098rsbl20101147 or via httprsblroyalsocietypublishingorg

Received 7 December 2010Accepted 22 December 2010

quantifiedandhaveproven tobeauseful systemforexplor-ing multi-trophic interactions

Genetic modification (GM) of plants can causedifferences between the transformed varieties andtheir conventional counterpart For instance changesin phloem sap composition can affect organisms thatfeed upon these plants including sap-feeding aphids[3] In an aphidndashparasitoidndashhyperparasitoid foodweb the effects of host plant traits can cascade up asfar as the fourth trophic level [2] even when thereare no effects on the herbivores [4]

We hypothesized that alterations on the plant levelowing to GM could affect the associated aphidndashparasitoidndashhyperparasitoid food web We studied thison two disease-resistant GM wheat lines (Triticumaestivum) their respective non-transformed controlsand a conventional line

2 MATERIAL AND METHODS(a) Study organisms

Metopolophium dirhodum Rhopalosiphum padi and Sitobion avenae arecommon cereal aphids in Europe [5] They are attacked by primaryparasitoids that oviposit inside the living aphid The parasitoid larvapupates inside the dead aphid forming a lsquomummyrsquo Primary parasi-toids are attacked by two guilds of secondary parasitoidsie hyperparasitoids and mummy parasitoids Hyperparasitoidsattack primary parasitoids before mummification whereas themore generalist mummy parasitoids attack the already mummifiedaphid irrespective of whether it contains a primary parasitoid or ahyperparasitoid

(b) Plant material

We used five wheat lines belonging to three different spring-wheatvarieties Bobwhite Frisal and Rubli For Bobwhite and Frisaleach had a transgenic line and its corresponding control line Thetransgenic line of Bobwhite (Pm3b1) carries the transgene Pm3bof hexaploid wheat which confers race-specific resistance to wheatpowdery mildew [6] The null-segregant line Sb1 was used as thecorresponding control line The second wheat pair consisted ofthe non-transformed variety Frisal as a control and the GM Frisalline ChiGlu(A13) containing the anti-fungal barley seed chitinaseand b-13-glucanase [78] This line however has not been shownto display enhanced resistance to powdery mildew [8] The varietyRubli is commercially grown in Switzerland

(c) The convertible glasshouse

The experiments were carried out in a convertible glasshouse thatapproximates field conditions by exposing the plants to outsideenvironmental temperatures and allowing natural colonization byinsects and pathogens The roof and side walls automatically openunder good weather conditions but close during strong winds andrainfall and during the night [9] Plants were grown in 40 plots(80 60 80 cm) arranged in two rows each plot containing aseparate central cylinder (26 cm diameter) in which 10 experimentalwheat plants were grown Buffer plants (ie non-transformed plantsof the same variety) were grown around the central cylinder simulat-ing a near-field situation Each wheat line was replicated eight timesin blocks of five adjacent plots in a row containing the five wheat linesin randomized order Plants were sown by the end of March and har-vested by the end of July Before sowing basic fertilizer was added tothe soil (per plot 2008 58 g P 72 g K 17 g Mg 87 g N 20091032 g P 8 g K 9 g Mg(NO3)2) Plants were watered as requiredFlowering ears of all wheat lines were covered with pergamynpaper bags to prevent pollen from escaping No pesticides wereapplied

(d) Sampling

Experiments were conducted in 2008 and 2009 Every week wecounted and identified all aphids and collected all mummies on theexperimental plants in the central cylinder from mid May until har-vest Parasitoid mummies were stored separately in gelatin capsulesat room temperature After four months emerged parasitoids wereidentified to species level using the same keys as Muller et al [10]

(e) Quantitative food web metrics and data analysis

Traditional binary measures of food web structure are highlysensitive to sample size [11] Therefore we used measures based on

This journal is q 2011 The Royal Society

Table 1 Aphidndashparasitoid associations across 2 years with respective food web codes in brackets as they appear in figures S1and S2 of the electronic supplementary material Numbers in the table correspond to the individuals found for the specificaphidndashparasitoid association

Metopolophium dirhodum (21) Rhopalosiphum padi (22) Sitobion avenae (23)

primary parasitoids

Aphelinus asychis (31) 1Aphelinus varipes (32) 60Aphidius ervi (33) 45 7 15Aphidius picipes (34) 35 11 7Aphidius rhopalosiphi (35) 479 142 10

Binodoxys sp (38) 2Ephedrus plagiator (36) 7 20 3Praon volucre (37) 309 13 7

secondary parasitoids

hyperparasitoids

Alloxysta brachyptera (47) 1Alloxysta brevis (48) 2Alloxysta f-sp (41) 1 3Alloxysta hepaptoma (42) 1Alloxysta r-sp (43) 4 47 1

Alloxysta victrix (45) 136 12 4Phaenoglyphis villosa (46) 77 16 4

mummy parasitoidsAsaphes suspensus (51) 66 26 6Asaphes vulgaris (52) 290 54 6

Coruna clavata (53) 98 2 3Dendrocerus carpenteri (54) 18 1 2Dendrocerus laticeps (55) 2Pachyneuron sp (56) 34 5 13

2 S von Burg et al Aphidndashparasitoid food webs on GM wheat

on January 20 2011rsblroyalsocietypublishingorgDownloaded from

Shannon information theory [12] as described in Bersier et al [13]and adapted for hostndashparasitoid webs [2] which use the densities ofspecies and the frequency of interactions (i) parasitoid diversitywhich is equal to species richness when all species are equally abun-dant but takes on smaller values when abundances are uneven (ii)link evenness which equals one when all trophic links have equal fre-quencies and asymptotically approaches zero for highly unevenfrequencies and (iii) quantitative realized connectance which is theobserved link diversity as a proportion of maximum possible linkdiversity and is a measure of the complexity of the network Since sec-ondary parasitoids cannot be unambiguously linked to primaryparasitoids [10] we calculated all the food web metrics separatelyfor the aphidndashprimary parasitoid and the aphidndashsecondary parasitoidmatrices

We used linear models (LMs) to analyse all the food web metricsWe conducted analyses across both years and for each year separatelyParasitoid diversity was square root-transformed link evenness andrealized connectance were arcsine-transformed Further we analysedcumulative aphid and mummy abundance parasitoid hatching suc-cess and parasitism rates Aphid and mummy abundance weresquare root-transformed parasitism was arcsine-transformed andboth analysed using LM Hatching success was analysed with a gen-eralized linear model with binomial errors All analyses were donewith the statistical software R (R Development Core Team)

3 RESULTSThree cereal aphid species were recorded in the twostudy years M dirhodum (mean seasonal densityper cylinder 2008 886 2009 2484) R padi(2008 663 2009 766) and S avenae (2008 242009 172) Overall aphid abundance was significantlyhigher in 2009 (F166 frac14 16071 p 0001) anddepended on variety (F266 frac14 526 p frac14 0007) and onthe variety GM interaction (F166 frac14 408 p frac140047) which was caused by a higher total aphid abun-dance on Pm3b1 when compared with its control line

Biol Lett

Sb1 whereas abundance was the same on ChiGlu(A13) and Frisal The number of mummies wassignificantly higher in 2009 when compared with2008 (F166 frac14 20323 p 0001) and was positivelyaphid density-dependent (p 0001 R frac14 0888)Parasitism rate was low at 2 per cent in both yearsand negatively correlated with aphid density (p frac140023 r frac14 20253) Neither wheat variety nor GMor their interaction had an effect on parasitism (varietyF266 frac14 210 p frac14 0130 GM F166 frac14 270 p frac14 0105variety GM F166 frac14 204 p frac14 0158)

From a total of 3492 mummies collected in both years2108 hatched (606) Hatching success was influencedby host aphid (p 0001 M dirhodum R padi

S avenae) and by year (p 0001 2009 2008) Weidentified 21 parasitoid species of which eight wereprimary parasitoid species seven hyperparasitoid speciesand six species were mummy parasitoids (table 1) In aparallel field study using the same wheat lines we foundthe same aphid and parasitoid species (electronicsupplementary material)

The analysis across both years revealed significantannual variation for all the food web metrics exceptrealized connectance of secondary parasitoids (figures 1and 2) The variety GMO interaction was significantfor link evenness of primary parasitoids (F166 frac14 8312p frac14 0005) No other differences were found

Separate analysis of the 2 years revealed severaldifferent inconsistent effects In 2009 we onlyfound significant effects on the primary parasitoidweb Primary parasitoid diversity significantly differedbetween the three varieties (figure 1b) and was

10

15

20

25

30

35

40

2008 2009

para

sito

id d

iver

sity

(a) (b)

variety F228 = 1101 p lt 0001

05

06

07

08

09

10

link

even

ness

(c) (d)

variety times GM F128 = 1830 p lt 0001

02

03

04

05

06

Pm

3b

1

Sb

1

Chi

Glu

(A13

)

Fris

al

Rub

li

Pm

3b

1

Sb

1

Chi

Glu

(A13

)

Fris

al

Rub

li

real

ized

con

nect

ance

(e) ( f )

Bobwhite Frisal Rubli Bobwhite Frisal Rubli

Figure 1 Means (+sem) of the calculated food web metrics of the primary parasitoid web per wheat line and yearPm3b1 and ChiGlu(A13) are the GM lines Sb1 and Frisal are their respective non-GM controls Rubli is the conventionalline (ab) Parasitoid diversity (cd) link evenness (ef ) realized connectance

Aphidndashparasitoid food webs on GM wheat S von Burg et al 3

on January 20 2011rsblroyalsocietypublishingorgDownloaded from

highest on Rubli The variety GM interaction wassignificant for primary parasitoid link evenness(figure 1d) By contrast in 2008 we only foundeffects on the secondary parasitoid web a significantvariety effect on link evenness (figure 2c) was foundand significantly higher realized connectance on theGM plant lines (figure 2e) The aphidndashparasitoidfood webs are shown in the electronic supplementarymaterial

4 DISCUSSIONWhile other studies have assessed the effects of GMplants on aphids and their parasitoids [1415] this isthe first to look at quantitative food web metrics ofan aphidndashparasitoid community

Aphid abundance on the GM line Pm3b1 washigher when compared with the control line Sb1

Biol Lett

This could directly be due to genetic differences inthe plants or to indirect effects caused by the differencein mildew infection As expected the Pm3b1 plantsshowed enhanced powdery mildew-resistance whencompared with the control Sb1 whereas ChiGlu(A13) was equally susceptible to powdery mildewas its control line Frisal In an earlier study weshowed that the Pm3b1 plants did not affect the per-formance of M dirhodum in the absence of powderymildew [16] We therefore suspect that an indirecteffect through mildew causes this difference in aphidabundance

Parasitism rate was negatively host density-dependentindicating that parasitoids keep up with the aphidpopulation growth only to a certain point The earlypresence of parasitoids in the field therefore seems cru-cial for keeping aphid populations under control as hasbeen suggested before [17] Parasitoid diversity remained

2

3

4

5

6

para

sito

id d

iver

sity

Pm

3b

1

Sb

1

Chi

Glu

(A13

)

Fris

al

Rub

li

Pm

3b

1

Sb

1

Chi

Glu

(A13

)

Fris

al

Rub

li02

03

04

05

06

real

ized

con

nect

ance

GM F127 = 737 p = 0011

06

07

08

09

10

link

even

ness

variety F227 = 358 p = 0042

2008 2009

(a) (b)

(c) (d)

(e) ( f )

Bobwhite Frisal Rubli Bobwhite Frisal Rubli

Figure 2 Means (+sem) of the calculated food web metrics of the secondary parasitoid web per wheat line and year Pm3b1and ChiGlu(A13) are the GM lines Sb1 and Frisal are their respective non-GM controls Rubli is the conventional line(ab) Parasitoid diversity (cd) link evenness (ef ) realized connectance

4 S von Burg et al Aphidndashparasitoid food webs on GM wheat

on January 20 2011rsblroyalsocietypublishingorgDownloaded from

generally unaffected by the wheat lines except forprimary parasitoid diversity in 2009 which was higheron Rubli Higher parasitoid diversity has been linked tohigher parasitism rates [18] an effect that was notobserved here

Link evenness was affected in the primary parasitoidand in the secondary parasitoid webs but differently soVariety affected the secondary web in 2008 whereasGM within the varieties affected link evenness of theprimary parasitoid web in different directions in2009 Food web structure on the transgenic Pm3b1line was more dominated by a few stronger linkswhereas on the control line Sb1 they were moreevenly distributed For the GM line ChiGlu(A13)and its control this was the other way aroundIn 2008 connectance of secondary parasitoid foodwebs was increased on the GM lines when comparedwith the control lines with potential implications forfood web stability [19]

Biol Lett

Quantitative food webs allow comparison of thediversity and complexity of species assemblageswhich are considered to be the major factors determin-ing ecosystem functioning and stability Despite theobserved changes in community structure among thedifferent wheat lines studied it is difficult to draw con-clusions about the ecological implications of theseeffects Observed differences in community structurewere inconsistent between the two study years andthere was as much variation between wheat varietiesas there was between GM and control lines within var-ieties This phenomenon has been observed before inother studies dealing with non-target effects of GMcrops [20] Plant characteristics seem to be more dis-tinct between varieties than between GM plants andthe corresponding controls Therefore we believethat the observed GM effects are of little ecologicalsignificance and fall within the natural variationobserved among cultivars

Aphidndashparasitoid food webs on GM wheat S von Burg et al 5

on January 20 2011rsblroyalsocietypublishingorgDownloaded from

Mario Waldburger helped with the sampling Beat Keller andChristof Sautter provided the plant material Food webgraphs were drawn using a Mathematica code provided byH C J Godfray This project was supported by the SwissNational Science Foundation (SNF grant 405940-115604)and is part of the National Research Programme NRP 59(wwwNRP59ch)

1 Price P W Bouton C E Gross P McPheron B AThompson J N amp Weis A E 1980 Interactions among3 trophic levels influence of plants on interactionsbetween insect herbivores and natural enemies AnnRev Ecol Syst 11 41ndash65 (doi101146annureves11110180000353)

2 Bukovinszky T van Veen F J F Jongema Y amp Dicke M2008 Direct and indirect effects of resource quality onfood web structure Science 319 804ndash807 (doi101126science1148310)

3 Hunt E J Pritchard J Bennett M J Zhu X BarrettD A Allen T Bale J S amp Newbury H J 2006 TheArabidopsis thaliana Myzus persicae model system demon-strates that a single gene can influence the interaction

between a plant and a sap-feeding insect Mol Ecol 154203ndash4213 (doi101111j1365-294X200603090x)

4 Harri S A Krauss J amp Muller C B 2008 Trophic cas-cades initiated by fungal plant endosymbionts impairreproductive performance of parasitoids in the second

generation Oecologia 157 399ndash407 (doi101007s00442-008-1094-y)

5 Vickerman G P amp Wratten S D 1979 Biology and peststatus of cereal aphids (Hemiptera Aphididae) in

Europe review Bull Entomol Res 69 1ndash32 (doi101017S0007485300017855)

6 Zeller S L Kalinina O Brunner S Keller B ampSchmid B 2010 Transgene environment interactionsin genetically modified wheat PLoS ONE 5 e11405

(doi101371journalpone0011405)7 Zhu Q Maher E A Masoud S Dixon R A amp Lamb

C J 1994 Enhanced protection against fungal attack byconstitutive coexpression of chitinase and glucanase genesin transgenic tobacco Bio-Technology 12 807ndash812

8 Bieri S Potrykus I amp Futterer J 2003 Effects ofcombined expression of antifungal barley seed proteins intransgenic wheat on powdery mildew infection MolBreed 11 37ndash48 (doi101023A1022145915983)

9 Romeis J Waldburger M Streckeisen P Hogervorst

P A M Keller B Winzeler M amp Bigler F 2007

Biol Lett

Performance of transgenic spring wheat plants and effectson non-target organisms under glasshouse and semi-field

conditions J Appl Entomol 131 593ndash602 (doi101111j1439-0418200701231x)

10 Muller C B Adriaanse I C T Belshaw R ampGodfray H C J 1999 The structure of an aphid-parasitoid community J Anim Ecol 68 346ndash370

(doi101046j1365-2656199900288x)11 Banasek-Richter C Cattin M F amp Bersier L F 2004

Sampling effects and the robustness of quantitative andqualitative food-web descriptors J Theor Biol 22623ndash32 (doi101016S0022-5193(03)00305-9)

12 Shannon C E 1948 A mathematical theory ofcommunication Bell Syst Tech J 27 379ndash423

13 Bersier L F Banasek-Richter C amp Cattin M F 2002Quantitative descriptors of food-web matrices Ecology83 2394ndash2407 (doi1018900012-9658(2002)083[2394QDOFWM]20CO2)

14 Lumbierres B Stary P amp Pons X 2010 Effect of Btmaize on the plantndashaphidndashparasitoid tritrophic relation-ships BioControl (doi101007s10526-10010-19317-

10527)15 Hogervorst P A M Wackers F L Woodring J amp

Romeis J 2009 Snowdrop lectin (Galanthus nivalisagglutinin) in aphid honeydew negatively affects survivalof a honeydew-consuming parasitoid Agric ForestEntomol 11 161ndash173 (doi101111j1461-9563200800412x)

16 von Burg S Muller C B amp Romeis J 2010 Transgenicdisease-resistant wheat does not affect the clonalperformance of the aphid Metopolophium dirhodumWalker Basic Appl Ecol 11 257ndash263 (doi101016jbaae201002003)

17 Langer A Stilmant D Verbois D amp Hance T 1997Seasonal activity and distribution of cereal aphid parasi-

toids in Belgium Entomophaga 42 185ndash191 (doi101007BF02769896)

18 Tylianakis J M Tscharntke T amp Klein A 2006 Diver-sity ecosystem function and stability of parasitoidndashhostinteractions across a tropical habitat gradient Ecology87 3047ndash3057 (doi1018900012-9658(2006)87[3047DEFASO]20CO2)

19 May R M 1973 Stability and complexity in model ecosys-tems Princeton NJ Princeton University Press

20 Sanvido O Romeis J amp Bigler F 2007 Ecological

impacts of genetically modified crops ten years of fieldresearch and commercial cultivation Adv BiochemEng Biotechnol 107 235ndash278