Detecting unauthorised and unknown GMOs

Detecting unauthorised and unknown GMOs Arne Holst-Jensen, Knut G. Berdal, Yves Bertheau, Marko Bohanec, Jon Bohlin, Maher Chaouachi, Kristina Gruden, Sandrine Hamels, Anja Krech, Esther Kok, et al. To cite this version: Arne Holst-Jensen, Knut G. Berdal, Yves Bertheau, Marko Bohanec, Jon Bohlin, et al.. Detecting unauthorised and unknown GMOs. Co-Extra International Conference, Jun 2009, Paris, France. 150 p. ฀hal-02757129฀ HAL Id: hal-02757129 https://hal.inrae.fr/hal-02757129 Submitted on 3 Jun 2020 HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.  INTERNATIONALCONFERENCE       25June2009 Paris,France           ProgrammeandAbstracts  GMANDNONGMSUPPLYCHAINS:THEIRCOEXISTENCEANDTRACEABILITY  www.coextra.eu   CoExtraConference June24,2009 AgroParisTech 16rueClaudeBernard Paris75005,France   StakeholderWorkshop June5,2009 PalaisduLuxembourg 15ruedeVaugirard Paris75006,France Biochip Systems CoExtraInternationalConference  Content  Programme.....................................................................................................................................5 AbstractsofOralPresentations.......................................................................................................9 Session1:IntroductoryPresentations.............................................................................................9 1.1. Reportonthecoexistenceofgeneticallymodifiedcropswithconventional andorganicfarming...........................................................................................................9 1.2. SIGMEAresultsoncoexistenceatthefarmlevel.............................................................11 1.3. TransContainer:OverviewandProgress..........................................................................13 1.4. CoExtraintroduction.......................................................................................................15 SessionA1:MethodsforManagingGeneFlow..............................................................................17 A1.1. Biologicalmeasuresforgeneflowmitigation..................................................................17 A1.2. Biocontainmentofmaizebycytoplasmicmalesterilityandxenia.................................18 A1.3. Pollencontainmentbycleistogamyinoilseedrape.........................................................20 A1.4. Chloroplasttransformationandtransgenecontainment................................................22 A1.5. Mesoscaledispersalofmaizepollenandimplicationsforgeneflow..............................24 SessionA2:CoexistenceandTraceabilityinAgricultureandFoodProduction...............................25 A2.1. Empiricalanalysisofcoexistenceincommoditysupplychains.......................................25 A2.2. ModellingcoexistencebetweenGMandnonGMwithinsupplychains.........................28 A2.3. CostsandbenefitsofsegregationandtraceabilitybetweenGMand nonGMsupplychainsoffinalfoodproducts..................................................................30 A2.4. Consumers’attitudestotheEUtraceabilityandlabellingregulation.............................33 SessionB1:TechnologiesforManagingtheSupplyChain..............................................................35 B1.1. GMOsamplingstrategiesinthefoodandfeedchain......................................................35 B1.2. RationalizationofGMOtestingbyappropriatesubsamplingandcontrolplans............38 B1.3. ModularApproachImplemented:Pros,ConsandFuturePerspectives..........................41 B1.4. Validationofnovelmethodsandtechnologies................................................................44 B1.5. ReferencematerialsandreferencePCRassaysforGMOquantification.........................46 SessionB2:DetectionofGMingredientsinfoodsandFeeds.........................................................49 B2.1. NewrealtimePCRmethodsavailableforroutineGMOdetectionlabs applicabilityandperformance.........................................................................................49 B2.2. ReliabilityandcostsofGMOdetection............................................................................51 B2.3. NonPCRbasedAlternativeAnalyticalMethods..............................................................53 B2.4. DetectingunauthorisedandunknownGMOs..................................................................55 B2.5. NewmultiplexingtoolsforreliableanalysisofGMOs.....................................................57 2 CoExtraInternationalConference Session3:Legal,Liability&RedressIssues....................................................................................59 3.1. Legal,Liability&RedressIssues.......................................................................................59 3.2. Scientificexpertiseandthejudges...................................................................................60 3.3. Juridicalcostbenefitanalysisofcoexistence:uneasythistask!......................................61 Session4:StakeholderViewsinEU...............................................................................................62 4.1. Stakeholderviewsandinteractions.................................................................................62 Session5:DecisionSupportSystems.............................................................................................63 5.1. TheCoExtraDecisionSupportSystem:AModelBasedIntegration ofProjectResults..............................................................................................................63 5.2. AnalyticalDSSmodule–howtosupportdecisionsintheanalyticallab.........................65 5.3. DSSmodulesontransportation(TMmodule)andonunapproved GMOs(UGMmodule).......................................................................................................67 Session6:Experiencesfromthirdcountries..................................................................................69 6.1. BenefitCostAnalysis,FoodSafety,andTraceability.......................................................69 6.2. SegregationMeasuresfor(Non)GMcropsandtheirImplicationsfor SupplyChainsinJapan.....................................................................................................71 6.3. CoExistenceandtraceability:Costsandbenefitsinfoodandfeedsupplychains.........73 6.4. ACompanyPerspective....................................................................................................74 6.5. ProtectingEuropeanqualityagriculture:NonGMfeedsupplyandproduction.............75 7. IntegrationofCoExtraresultsinEUtoolsforcoexistence&traceability.......................78 8. SummaryofmainCoExtradeliverables&results,perspectives,information dissemination&application.............................................................................................79  PosterAbstracts...........................................................................................................................96 P1. AcosteffectiveP35S/Tnosmultiplexscreeningassaywithinternalpositivecontrol.....96 P2. Theproblemofwhentolabelinpresenceoflowamountsoftransgenic material:thecaseofbotanicalimpurities........................................................................97 P3. NIRimagingandchemometricsinsupporttothedetectionatthe singlekernellevelofGMO...............................................................................................99 P4. PerformanceofTaqMan®,LNA,CyclingProbeTechnology,Luxand PlexorrealtimePCRchemistriesinquantitativeGMOdetection.................................100 P5. GMOanalysis:towardsassuringconfidenceinaresult.................................................101 P6. DetectionofBacillusthuringiensisbyrealtimePCR......................................................103 P7. Developmentofanewprobeforqualitativeidentificationand quantificationofBt11maize.........................................................................................105 P8. DevelopmentofconstructspecificTaqManrealtimePCRfordetectionand quantificationoftransgenicBt11maize(Zeamays)......................................................106 P9. Stateoftheartonsamplepreparationandassessingthevalidity ofproceduresderivingtestportionfromlaboratorysamples.......................................108 P10. DesigningthePCRmarkersAgrobacteriumtumefaciensgallformingstrains..............110 3 CoExtraInternationalConference P11. Arapid&simplepointofusediagnosticforGMOdetectioninplants.........................112 P12. Developmentofanintegratedplatformforthedetectionofmaterials derivedfromgeneticallymodifiedcropsinfoodandfeedproducts.............................113 P13. UseofpJANUS¥02001asCalibratorPlasmidforGTS4032 (RoundupReadySoybean)Detection:AnInterLaboratoryTrialAssessment..............114 P14. Testingthe“ModularApproach”:anexamplewithRoundUpReadySoybean...........115 P15. NAIMA:afastquantitativemethodforhighthroughputGMO diagnosticsinfoodandfeedstuffs.................................................................................116 P16. GMOversusmycotoxinssamplingplan:apragmaticapproach....................................117 P17. Approachestomonitortheadventitiouspresenceoftransgenes inexsitucollectionsofnationalgenebanks..................................................................119 P18. Monitoringtheadventitiouspresenceoftransgenesinexsitucotton collectionsoftheNationalGeneBank...........................................................................120 P19. MoleculardiagnosisofcommercializedorunapprovedBtcropsof IndiausingqualitativeandquantitativePCRassays......................................................121 P20. MultiplexingofSIMQUANT............................................................................................122 P21. Useofcomputationalsubtractiontosearchforunknowngeneticmodifications........124 P22. EffectofdifferentstorageconditionsonPCRamplificabilityof genomicDNAextractedfrompelletscontainingmaizeMON810maize......................125 P23. MultiplexDNADetectionSystemForIdentificationOfGenetically ModifiedOrganisms(GMOs)InFoodAndFeedChains;CoExtraWP6results.............127 P24. TheCoExtrawebsite,akeytoolintheCoExtraexternalcommunicationstrategy.....128 P25. Influenceofthe(nonGM)soybeanpriceoncompoundfeedprice..............................130 P26. ThecosteffectivenessofthecoexistenceofGMHToilseedrapeinIreland: ananalysisofcropmanagementstrategies...................................................................131 P27. ModellingcoexistencebetweenGMandnonGMsupplychains..................................133 P28. Supplychaindescriptionandanalysisformaize,potatoesandfresh tomatoesinSlovenia......................................................................................................135 P29. PreferenceheterogeneityamongGermanconsumersregardingGMrapeseedoil.....137 P30. Costsofcoexistenceandtraceabilitysystemsinthefoodindustry inGermanyandDenmark..............................................................................................138 P31. AnalysisoftheextracostsgeneratedonFrench“LabelRouge” chickensupplychainbynonGMfeedpolicy.................................................................139 P32. Towardsanoptimalmanagementregimetofacilitatethecoexistence ofGMandnonGMoilseedrapeinIreland...................................................................141 P33. BrazilianGMOFreeAreasExperimentandtheReleaseofRRSoybeans.......................143 P34. AbibliometricsapproachonSoybeanResearchinBrazil..............................................144 P35. TheAgroindustrialChainofSoybeaninBrazil:BriefNotesontheContractofSale.....145 P36. TimeRequirementsandFinancialExpendituresforCoexistenceMeasuresand TheirImpacttoProfitabilityofGeneticallyModifiedPlantsinSwitzerland..................146 4 COEXTRAINTERNATIONALCONFERENCE TUESDAY,JUNE2 8:00to18:00 Settingup:ExhibitionandPosters AgroParisTech(APT) 14:00to18:00 Conferenceregistration(alsorequiredforattendingthe receptiononTuesdayevening) APT–mainentrance (16rueClaudeBernard, 75005Paris) 15:00to16:00 PressConference(accessreservedforpress)  Chair:FrançoisHoullier(DSPPV,Scientificdirectorof Plantsandderivedproducts,France)  YvesBertheau(INRA,France),FrédériqueAngevin(INRA, France),CoExtraExecutiveCommitteeMembersand BernhardKoch,Prof.ofTortLaw(InnsbruckUni.) APTsalledesconseils Reception: Welcomecocktail&Welcomeaddress CityHallMairiedeParisVème 21PlaceduPanthéon 75005ParisV Métro:Luxembourg(15minutes walkingdistance) 19:00to20:30 WEDNESDAY,JUNE3 SESSION1:ECResearch 8:0010:00 9:009:10 9:109:50 9:5010:20 10:2010:50 10:5011:30 11:3012:00 12:0013:30 13:3015:00 Welcomeaddress RemiTousain(DirectorofAgroParisTech,France)and YvesBertheau(INRA,France) Chair:YvesBertheau(INRA,France)  1.1Reportonthecoexistenceofgeneticallymodified cropswithconventionalandorganicfarming SigridWeiland(DGAgricultureandRuralDevelopment, EC),AliceStengal(DGEnvironment,EC),CiaranMangan (DGResearch,EC) 1.2SIGMEAresultsoncoexistenceatthefarmlevel JeremySweet(NIAB,UK) MIDDAYBREAK SESSION2:Parallelsessions(detailedprogrammeonpage6) ParallelsessionA.1 ParallelsessionB.1   MethodsforManagingGene TechnologiesforManagingthe Flow SupplyChainandDetectionof GMingredientsinfoodsand Feeds  ParallelsessionA.2 19:00 APTAmphiTisserand& Risler 1.3TransContainer:OverviewandProgress RuudA.deMaagdandKimBoutilier(PlantResearchIn ternationalB.V.,TheNetherlands) BREAKANDVISITOFCONFERENCEEXHIBITION&POST ERS 1.4CoͲExtraintroduction YvesBertheau(INRA,France) Chair:JoachimSchiemann(JKI, Germany) 15:00to18:00 APT–mainentrance Registration CoexistenceandTraceabilityin AgricultureandFood/Feed Production Chairs:FrédériqueAngevin (INRA,France&MortenGylling (FOI,Denmark) Chairs:KristinaGruden(NIB, Slovenia)&RobertaOnori(ISS, Italy) ParallelsessionB.2 SessionA:Tisserand  SessionB:Risler GMOdetection  Chair:ArneHolstJensen(NVI, Norway) ConferenceDinneronRiverSeineboat  5 COEXTRAINTERNATIONALCONFERENCE SessionsAandBDetails WEDNESDAY,JUNE3,AgroParisTech,France,Paris  SessionA1 MethodsforManagingGeneFlow Chair: JoachimSchiemann(JKI,Germany) 13:3013:50 A1.1. Biologicalmeasuresforgeneflowmitigation AlexandraHüsken(JKI,Germany) 13:5014:10 A1.2. Biocontainmentofmaizebycytoplasmicmale sterilityandxenia MariaMunsch(ETH,Switzerland) 14:1014:30 A1.3. PollencontainmentbyCleistogamyinoilseedrape XavierPinochet(Cetiom,France) 14:3014:50 14.5015.10  A1.4. Chloroplasttransformationandtransgenecontain ment A1.5. Mesoscaledispersalofmaizepollenandimplica tionsforgeneflow SessionA.2. CoexistenceandTraceabilityinAgricultureandFood Production 15:1015:40 A2.1. Empiricalanalysisofcoexistenceincommodity supplychains 15.4016:10 BREAK 16.1016:30 16:3016:50 16.5017:10 A2.2. ModellingcoexistencebetweenGMandnonGM withinSupplyChains A2.3. Costsandbenefitsofsegregationandtraceability betweenGMandnonGMsupplychainsoffinal foodproducts A2.4. ConsumersattitudestotheEUtraceabilityand labellingregulation RalfBock(MPI,Germany) S.DupontandY.Brunet(INRA,France) Chairs: FrédériqueAngevin(INRA,France)& MortenGylling(FOI,Denmark) JamesCopeland(FERA,UK)andNicolas Gryson(UniversityCollegeofGhent, Belgium)  LouisGeorgesSoler(INRA,France) KlausMenrad,AndreasGabriel(WZS, Germany) JoséM.Gil&MontserratCostaFont (CREDAUPCIRTA,Spain) 17.1018:00 Questionsanddiscussion   SessionB.1. TechnologiesformanagingtheSupplyChain Chairs: KristinaGruden(NIB,Slovenia)& RobertaOnori(ISS,Italy) 13:3013:50 B1.1. GMOsamplingstrategiesinthefoodandfeed chain MarinaMiraglia(ISS,Italy) 13:5014:10 B1.2. RationalizationofGMOtestingbyappropriatesub samplingandcontrolplans YvesBertheau(INRA,France)andRoy MacArthur(FERA,UK) 14:1014:30 B1.3. Themodularapproachimplemented,pros,cons andfutureperspectives MarkvandenBulcke(IPH,Belgium) 14:3014:50 B1.4. Validationofnovelmethodsandtechnologies MarcoMazzara(JRCIHCP,Italy)  14:5015:10 B1.5. ReferencematerialsandreferencePCRassaysfor GMOquantification IsabelTaverniers(ILVO,Belgium) 15:1015:30 Questionsanddiscussion  15:3016:00 BREAK   SessionB.2. DetectionofGMingredientsinfoods B2.1. NewrealtimePCRmethodsavailableforroutine GMOdetectionlabsapplicabilityandperform ance Chair:ArneHolstJensen(NVI,Norway) 16:2016:40 B2.2. ReliabilityandcostsofGMOdetection KristinaGruden(NIB,Slovenia) 16:4017:00 B2.3. NonPCRbasedalternativeanalyticalmethods GuyKiddle(Lumora,UK) 17:0017:20 B2.4. DetectingunauthorisedandunknownGMOs ArneHolstJensen(NVI,Norway) 17:2017:40 B2.5. Newmultiplexingtoolsforreliableanalysisof GMOs MariaPla(CSIC,Spain) 17:4018:00 Questionsanddiscussion  16:0016:20 DoerteWulf(Genescan,Germany) 6 COEXTRAINTERNATIONALCONFERENCE THURSDAY,JUNE4 SESSION3  9:009:20 9:20–9:40 9.4010.00 Legal,liability&redressissues Chair:BernhardKoch(ECTIL,Austria) 3.1.Legal,liability&redressissues BernhardKoch(ECTIL,Austria)andM.A.Hermitte(CNRS,France) 3.2.ScientificexpertiseandthejudgesC.Noiville(CNRS,France) APTAmphiTisserand 3.3.Juridicalcostbenefitanalysisofcoexistence:uneasythistask! MA.Hermitte,G.Canselier(CNRS,France)&Y.Bertheau(INRA,France) SESSION4 10:0010:30 StakeholderviewsinEU: Chair:KristinaSinemus(Genius,Germany)  4.1.StakeholderopinionsandattitudesoncoexistenceofGMOswith conventionalandorganicsupplychains GeorgeSekallaris(NHRF,Greece)andRenèCusters(VIB,Belgium) APTAmphiTisserand 10:3011:00 BREAKANDVISITOFCONFERENCEEXHIBITION&POSTERS  SESSION5  Dataintegration&DecisionSupportSystems Chair:NevenaAlexandrova(ABI,Bulgaria) 11:0011:20  5.1.TheCoExtraDecisionSupportSystem:Amodelbasedintegration ofprojectresults MarkoBohanec(JSI,Slovenia) 5.2. AnalyticalDSSmodule–howtosupportdecisionsinthe analyticallab KristinaGruden(NIB,Slovenia) 5.3.DSSmodulesontransportation(TMmodule)andonunapproved GMOs(UGMmodule) EstherKok(RIKILT,TheNetherlands) 11:2011:40  11:4012:00 12:0013:30 APTAmphiTisserand  MIDDAYBREAK SESSION6  Experiencesfromthirdcountries Chair:MortenGylling(FOI,Denmark) 13:3014:00  6.1.Benefitcostanalysis,foodsafety,andtraceability JamesHammitt(HarvardUniversityCentreforRiskAnalysis,Boston,USA) 14:0014:30  6.2.Segregationmeasuresfor(non)GMcropsandtheirimplications forsupplychainsinJapan MasashiTachikawa(IbarakiUniversity,Japan) 14:3015:00  6.3.CoExistenceandtraceability:costsandbenefitsinfoodandfeed supplychains BillWilson(NorthDakotaUniversity,USA) 15:0015:30 6.4.CompanyPerspective RandalGiroux(Cargill,USA) 15:3016:00 BREAK 16:0016:30 16:3017:00 17:0017:30 6.5.ProtectingEuropeanqualityagriculture:nonGMfeedsupplyand production RenaudLayadi,(RegionBretagne,France) 7.IntegrationofCoExtraresultsinEUtoolsforcoexistence& traceability GuyvandenEede&EmilioRodriguezCerezo(EuropeanCommission/JRC) 8.SummaryofmainCoExtradeliverables&results,perspectives Informationdissemination&application YvesBertheau(INRA,France) 17.3018:00 9.ConcludingComments fromCoExtra,INRA,ECrepresentatives 19:00 Freeevening APTAmphiTisserand  APTAmphiTisserand 7  COEXTRAINTERNATIONALCONFERENCE Stakeholderworkshop PalaisduLuxembourg,Paris Friday,June5 8:008:30 Registration Entrance Chair(wholeday):YvesBertheau(INRA,France)  Stakeholderpanel:GarlichvonEssen(EuropeanSeedAssociation,ESA),Arnaud Petit(CommitteeofProfessionalAgriculturalOrganisations,GeneralCommit teeforAgriculturalCooperationintheEuropeanUnion,COPACOGECA),Agnès Davi(ConfederationofFoodandDrinkIndustriesoftheEU,CIAA),OlivierAn drault(UFCQueChoisir,FederalUnionofConsumers),MireilleFerri(Vice présidenteRégionIledeFrance),MaaikeRaaijmakers(“PlatformBiologica”) Moderator:OlivierdeLagarde(journalist) 8:309:00 Introductorytalk MarionGuillou(CEOINRA,France)  IntroductiontoCoExtra YvesBertheau(INRA,France) 9:009.20 Fromseedstosilo:agriculturalcoexistenceandtraceabilityissues FrédériqueAngevin(INRA,France) 9:2010:30 Roundtable Panelquestions,thenaudiencequestions 10.3010:50 Legalissues BernhardKoch(ECTIL,Austria) 10:5012:00 Roundtable Panelquestions,thenaudiencequestions 12:0014:00 LUNCHBREAK 14:0014:30 Supplychainmanagementandeconomicissues MortenGylling(FOI,Denmark) 14:3015:30 Roundtable Panelquestions,thenaudiencequestions 15:3016.10 Stakeholderopinionsandattitudes  Somelessonsfromstakeholderinteractionsforthefutureofcoexistence RenèCusters(VIB,Belgium)  Statement PascaleHebel(CREDOC,France) 16.1017:10 Roundtable Panelquestions,thenaudiencequestions 17:1017:40 Finalcomments 17.4017.55 CoExistenceofGMandnonGMsupplychains:thepointofviewoftheCom missionerinchargeofAgriculture. JulienMousnier(MemberofthecabinetofMsFischerBoel,EC,Brussels) 17:5518.10 ConclusionsbyJeanLouisBorloo(FrenchMinisterofEnvironment) SalleMédicis SalleRenéCoty SalleMédicis 8 CoExtraInternationalConference AbstractsofOralPresentations  Session1:IntroductoryPresentations  1.1. Reportonthecoexistenceofgeneticallymodifiedcropswith conventionalandorganicfarming SigridWeilandandAliceStengal, EuropeanCommission,Brussels  Coexistencereferstothechoiceofconsumersandfarmersbetweenconventional,organicandGM cropproductionincompliancewiththelegalobligationsregardingthelabellingofGMOs.GMOsas wellasfoodandfeedcontaining,consistingof,orproducedfromGMOshavetobelabelledinorder to guarantee an informed choice. As this potentially implies economic losses, e.g. where GMOs appear in conventional or organic products, suitable technical measures have to be taken to segregate GM from nonGM production. Whilst environmental and health aspects of GM crop cultivationmustbeexhaustivelyaddressedalreadyduringtheauthorisationprocedure,theyarenot to be considered in the context of coexistence. Coexistence measures have their focus on the economicimpact. Member States may take appropriate national measures on coexistence in order to avoid the unintendedpresenceofGMOsinotherproducts.TheCommissionRecommendationonguidelines for the development of national strategies and best practices on coexistence is intended to help MemberStatesdevelopnationallegislativeorotherstrategiesforcoexistence. The Commission published recently its second report on coexistence providing an update of the stateofnationalcoexistencemeasuresbasedoninformationprovidedbytheMemberStates.The reportalsogivesanoverviewoftheactivitiesundertakeninresponsetothemandateprovidedby theconclusionsoftheAgriculturalCouncilofMay2006. With15MemberStateshavingadoptedlegislationoncoexistence,comparedtofourin2006,there hasbeensignificantprogressinthedevelopmentofcoexistencelegislation.Theapproachesapplied inMemberStatesdifferwithrespecttoadministrativeproceduresandthetechnicalspecificationsof segregation measures. These differences reflect the regional variation of agronomic, climatic and otherfactorsdeterminingthelikelihoodofGMOadmixturetononGMcrops.Astudylaunchedby the Commission shows that all national jurisdictions foresee a minimum protection in cases of economicdamagesresultingfromGMOadmixtureinnonGMcropsunderregularconditionsoftort lawwhichdiffersbetweenMemberStates.Themajorityofthemhasnotadjustedtheconditionsof generaltortlawtothespecificcaseofGMOadmixture. In parallel to the development of national coexistence regulation, there has been a moderate expansion of the cultivation of GM crops. However, commercial experience necessary for the assessmentofthebestwayforwardtoaddresscoexistenceisstilllimited. Research activities concerning various aspects of coexistence are still ongoing in many Member States,illustratingtheneedforfurtherdevelopingtheknowledgebase.Inviewoffurtherassessing and enhancing the efficiency of national coexistence measures, the European Coexistence Bureau (ECoB), created by the Commission, is developing, in collaboration with the Member States, crop specificBestPracticeDocuments. 9 CoExtraInternationalConference From the present report the Commission concludes that there is no need to deviate from the subsidiaritybased approach towards coexistence. The Commission will continue to foster the exchangeofinformationwithMemberStatesregardingcoexistenceandsupportfurthercoexistence relatedresearchbasedonclearlyidentifiedneeds.  10 CoExtraInternationalConference 1.2. SIGMEAresultsoncoexistenceatthefarmlevel AntoineMesséan1&JeremySweet2 1 EcoInnov,INRA,BP1,78850ThivervalGrignon,France 2 TheGreen,Willingham,CambridgeCB245JA,UnitedKingdom  In 2003, the European Commission established the principle of coexistence which refers to “the ability of farmers to make a practical choice between conventional, organic and GMcrop production,incompliancewiththelegalobligationsforlabellingand/orpuritystandards”andlaid downguidelinesdefiningthecontextofthiscoexistence1. What needs to be accounted for if we are to introduce in a sustainable manner GM crops throughoutEuropesothatcoexistenceisfeasible?ThecrossdisciplinaryEuropeanSIGMEAResearch Projectwassetuptoprovidetodecisionmakerssciencebasedinformationabouttheappropriate coexistenceandtraceabilitymeasuresthatwouldbeneeded. To this end, SIGMEA brought together the principal teams and thereby the principal programmes studying gene flow in a large number of countries across Europe, representing a wide range of agriculturalsystemsincludingorganicfarming. Withinthelast5years,SIGMEAhas(i)collatedandanalysedEuropeandataongeneflowandthe environmental impacts of the major crop species which are likely to be transgenic in the future (maize,rapeseed,sugarbeet,rice,andwheat), (ii) designedpredictive modelsofgeneflowatthe landscape level, (iii) analysed the technical feasibility and economic impacts of coexistence in the principal farming regions of Europe, (iv) developed novel GMO detection methods, (v) addressed legalissuesrelatedtocoexistence,and(vi)proposedpublicandfarmscaledecisionmakingtools,as wellasguidelinesregardingmanagementandgovernance. SIGMEAhasproducedapracticaltoolboxforaddressingGMimpactsinagriculture: 1. A unique database including more than 100 data sets on geneflow and ecological impacts whichmayinformdecisionmakersonfactorsdrivinggeneflowatthelandscapeleveland on the variability of such processes across Europe, help regulators to set up coexistence measuresatNationallevelsaswellashelpscientiststoidentifyfurtherresearchprioritiesin thatarea. 2. LandSFACTSisauserfriendlywindowsbasedsoftwaretosimulatecropallocationtofields by integrating typical crop rotations and crop spatiotemporal arrangements within agricultural landscapes and could be used for a practical implementation of coexistence measures 3. The generic gene flow platform LandFlowGene, including validated rapeseed and maize modules, is now available as a prototype. It has been used to assess the feasibility of coexistence at the landscape level under various contexts (climate, landscape, cropping systems,adoptionrate)andtesttheeffectofcoexistencemeasures.Thisplatformcouldbe extended to other crops to provide a general framework for informing coexistence in all croppingsystemsofEurope. 4. Structural and organisational factors affecting coexistence in practice have been identified andstrategiesformanagingcoexistenceattheregionallevelhavebeenproposed;  1 Commissionrecommendationof23July2003 (http://ec.europa.eu/agriculture/publi/reports/coexistence2/guide_en.pdf) 11 CoExtraInternationalConference 5. A userfriendly decisionsupport system (SMACAdvisor) to assess maize coexistence feasibilityatthefieldlevelwasdesigned. 6. A comprehensive overview of monitoring and legal issues has been provided and general recommendationshavebeenmade. Altogether, these tools and outcomes can be combined to assess coexistence at various spatial scales(field,farmorregion)andvariousdecisionmakinglevels(farmers,elevators,memberstates, EU). SIGMEAfindingsmakeitpossibletoaddressissuessuchas"whatwillhappen,intermsofgeneflow, ifaparticularGMorganismisintroducedintoaparticularEuropeanregion?"and"howcancropsbe deployed at the landscape level so as to maintain the adventitious presence of GMOs in conventionalcropswithinthelegalthresholds,oranyspecificmarketdrivenrequirements?". The outcome of both field and modelling studies carried out in SIGMEA is that best practices for coexistencearehighlyvariableanddependonlocalcharacteristics,croppractices,environmentsas wellasfarmerstrategiesandpreferences,andthatthefeasibilityofcoexistencedirectlydependson thetargetedthreshold. Formaize,coexistence(definedascomplyingwiththeofficialthreshold)forhybridvarietiesshould be achievable through the use of high purity seed, the management of cross pollination by using varietiesthatfloweratdifferenttimesand/orspatiallyseparatingfields,ortheinstallationofbuffer zonesorthepracticeofdiscardingwherefieldsareincloseproximity.Forlowthresholds(0.1%)orin regions with high density of maize, requested measures such as isolation distances may be impossible to implement and a geographical separation between GM and conventional crops is a reasonablesolution.Forsupplychains,suchasorganicfarming–whichrequiresatotalabsenceof GMOsintheircrops–coexistenceatalocalscaleistechnicallyimpossible. Basedonregionalcasestudiesfindings,contrastingglobalcoexistencescenariosmaybedefinedby consideringdifferentregulationapproaches: x x x A"bottomup"approach,whichwouldlettheprivateactors(collectors,farmers)freetochoose the best way to achieve coexistence guidelines and to meet regulatory or marketbased thresholdrequirements; A "topdown" approach, based on the strong intervention of public authorities with the implementationofcompulsoryuniformmeasures(e.g.,isolationdistances); and a "third way" approach, which provides a focused response of authorities to lift some constraintsonprivateactors. It has been stressed that a coexistence regime based on “uniform isolation distances”, as implemented so far in several member states, is not optimal, not proportional and may lead to unnecessaryadditionalcostsorrendercoexistenceimpossibleinpractice. SIGMEAthusrecommendsthatcoexistencemeasuresshouldbeasflexibleaspossibleanddepend on local climatic, agronomic and environmental factors. This approach would lead to more cost efficientmeasures.Howeverthecurrentregulatoryframeworktosupportsuchanapproachisstill tobedeveloped. SIGMEA has developed tools to support the definition and implementation of flexible measures. Predictivegeneflowmodelsarenowavailable(currentlyonlyformaizeandoilseedrapebuteasily extendabletoothercrops).Thesecanhelpdecisionmakersassessthefeasibilityofcoexistenceat the field, farm and silo level for the various targeted thresholds under various environmental and agronomicconditions.Inadditionsimpledecisionsupporttools,likeSMACAdvisorcanbeusedby farmersoradvisorswhowouldliketoquicklyassesscoexistencefeasibilityusinglimitedamountsof informationatalocalfieldlevel. 12 CoExtraInternationalConference 1.3. TransContainer:OverviewandProgress RuudA.deMaagdandKimBoutilier PlantResearchInternationalB.V.,WageningenUniversityandResearchCentre,Wageningen,TheNetherlands  Background The spread of transgenes from genetically modified crops to conventional and organic crops or to wild relatives remains a source of public and scientific concern in Europe. While movement of transgenesfromgeneticallymodifiedcropsapprovedforcultivationtoconventionalororganiccrops is strictly speaking not a biosafety issue, the EU policy for GMO crops is one of coexistence and traceability,i.e.theconcurrentexistenceofallthreesystems(GMO,conventional,organic)should be facilitated (1). This has led to the development of countryspecific “coexistence measures” regulatingthegrowing,processing,andtracingproceduresforGMcrops(2).Containmentmeasures may be classified as physical, temporal or biological. Current coexistence measures use physical containment, namely minimal isolation distances and pollen barriers, between GMO and conventional or organic crop fields, as well as measures to prevent adventitious mixing during harvestingandprocessing. Coexistence of GM and nonGM crops may be promoted by the implementation of biological transgene containment strategies, involving modification of the GMO crop in such a way as to minimizethespreadoftransgenesthroughpollen,seedorboth.Thecontainmentmechanismused for a particular crop needs to be carefully chosen for the mode of transgene spread that is most relevant for that crop, and be compatible with the harvested product (vegetative parts, fruits, or seeds).WhilenotthefocusoftheabovementionedEUpolicy,biologicalcontainmentmayalsobe beneficial when the spread of transgenes may be undesirable because of human health risks (pharmaceuticalsorrawindustrialproducts)orwhereoutcrossingtowildrelativesisaconsiderable risk. Depending on the particular application, biological containment strategies need to be proven failsafetovaryingdegrees.  TransContainer The EU FW6 project TransContainer, which is coordinated by the authors, comprises 13 partners fromuniversities,researchandgovernmentinstitutes,SMEsandoneindustrialpartner.Theproject isinvestigatinganddevelopinganumberofstrategiesforbiologicalcontainment: x Plastidtransformationasameanstopreventtransgenespread; x Preventionoffloweringasbiologicalcontainmentstrategy; x Controllingtransgenetransmissionthroughpollenandseed Where necessary, we aim to complement these strategies with tightly controllable switches to restore fertility. The crops used are European crops grown for their seeds (oilseed rape), fruits (tomatoandeggplant),orvegetativeparts(sugarbeet,ryegrass,redfescue,poplarandbirch).For some of these crops, several strategies are being developed. Besides developing biological containmentstrategies,theprojectalso: x Investigates the impact of the implementation of these strategies on environmental and foodsafetyandonthepossibleimprovementofcoexistencerules, x Assesses the agroeconomic effects for European agriculture and compares different scenariosforcoexistence, 13 CoExtraInternationalConference x Invokes stakeholder dialogue on socioeconomic and environmental issues by holding interviewsandworkshopsforstakeholdersandthepublic, x Communicates coexistence issues and results of the project to stakeholders and the generalpublicthroughworkshops,theproject’swebsite,andproductionofaDVD. The first results on the biological containment technologies of the Transcontainer project are beginning to emerge, and will be discussed (3, 4). Stakeholder involvement has proven to be a difficulttask,asalargepartofthepublicandtheusers(farmers)areonlyjustcomingtotermswith the introduction of GM crops and the associated coexistence measures. As a result, many of the stakeholdersarenotawareofthedifferentbiologicalcontainmentoptionsorhavenothadtimeto considerthem.Whenopinionsonthistechnologyhavebeengiven,theyusuallyfollowthelinesof the extremely polarized camps in Europe: proponents welcome the option or think that they are unnecessary,whileopponentsatbestdenouncereleaseofallGMOs,andatworstseeaplottoget GMOcropsacceptedorevenanexcusetodevelopGURTs,theinfamous“Terminator”technology. ThisworkwassupportedbytheEUFramework6Programme(Contractnr.023018).  References: 1. EuropeanCommission.LifesciencesandbiotechnologyaStrategyforEurope.[COM(2002) 27],http://ec.europa.eu/biotechnology/pdf/com200227_en.pdf 2. EuropeanCommission.Guidelinesforthedevelopmentofnationalstrategiesandbest practicestoensurethecoexistenceofgeneticallymodifiedcropswithconventionaland organicfarming.http://ec.europa.eu/agriculture/publi/reports/coexistence2/guide_en.pdf 3. Colombo,M.,S.Masiero,S.Vanzulli,P.Lardelli,M.M.Kater,andL.Colombo.2008.AGL23,a typeIMADSboxgenethatcontrolsfemalegametophyteandembryodevelopmentin Arabidopsis.PlantJ.54:10371048. 4. DeMarchis,F,Wang,Y,Stevanato,P,Arcioni,S,andBellucci,M.Genetictransformationof thesugarbeetplastome.TransgenicRes2008DOI10.1007/s1124800891934 14 CoExtraInternationalConference 1.4. CoExtraintroduction YvesBertheau INRA,RoutedeSaintCyr,78026Versaillescedex,France.  CoExtraisanFP6(contract007158)researchprogramofthepriority5(Foodsafetyandquality)of theEuropeanCommissionwhichstartedinApril2005andfinishesinSeptember2009. Its main aim is to provide practical tools to implement coexistence and traceability for the coexistenceofsupplychainsusingeitherGMO,conventionalproductsororganicagriculturederived products. This integrated project completes the two complementary STREPS: SIGMEA working mostlyonfieldcoexistenceandTranscontainerfocusingonbiocontainmentmethods. The coexistence is understood as the ability to farmers to produce the agricultural products they wish, while still enabling the freedom of choice of consumers. The documentary and analytical traceabilitystudiedinCoExtraaretwotoolsnecessaryforbothmanagingthecoexistenceofsupply chainsandcontrollingtheresultsofthismanagement.Theproductstobemanagedoriginateeither from the European agriculture or from imports from third countries. In several aspects this managementofsupplychainsdoesnotdifferfromsystemsalreadyinplace,suchaswaxymaize,or seedsproductions.ThesegregationofsuchspecialitiesisquitewellknownandcontrolledintheEU andseveralthirdcountries,anddoesnotimpacttoomuchEuropeansupplychainscosts.Themain issueinsegregatingGMandnonGMproductsliesthusinaratherlowlabellingthresholdof0.9% andtheuseoftheDNAunittomeasurethis,asrecommendedbytheEC. CoExtrafirstattempted toaddresscoexistencefromthefarmtotheretailerbystarting empirical studiesandmodellinginfields,andstudyingtheiroutcomesmanagementintheupperpartsofthe supply chains. Gene flow studies on long distance of pollen dispersion on fragmented landscape were undertaken and statistical models were validated for e.g. maize. Biocontainment methods, designedtominimizegeneflow,werealsostudied.Theeffectsofseedsadmixtures,aswellasthose ofstackedgenes,onfieldsoutcomesoncurrentpollenflowmodelsandseedspuritywereassessed. Costsbenefitsanalysesofcoexistenceandtraceabilitywereundertakenwhilelookingforthemost costeffectivedetectionmethodstoreducetheirimpactonthefinalcosts.Thepracticesoftraders andthirdcountriesfarmerswereanalyzedinordertodeterminetrendsthatmaypredictthefuture ofEuropeansupplychains. Asaconsequenceofthe178/02Europeanregulation,documentarytraceabilityisawellknownand implementedpracticeinEuropeancompanies.GMOtraceabilitydiffersfromthisgeneralrequestof traceability by adding a longer period of documents preservation. Studies of documentary traceability, particularly in third countries, were undertaken for its positive impact on cost effectiveness on final prices and its current use in the EU. While the European policy opened the doortoanalytical controls,documentarytraceabilityisaunderestimatedwaytotraceproductsat thelowestcostsinsupplychainsprovidedthecriticalpointsofsupplychainsareclearlyidentified andmasteredafterinitialanalyticalcontrols. As it was exemplified in a previous European study (Kelda / Keste) sampling large batches such as shipments of several thousand tons is not an easy task. The same apply to sampling in fields. As samplingisalsocarriedoutforseveralotherpurposessuchasmycotoxins,pathogens,allergens,a survey of sampling plans was carried out and the interest of combining different sampling plans tested. Thanks to the 1829/03 and 1830/03 regulations, detection methods (currently Quantitative Real Time PCR) of EU approved GMOs are all validated through collaborative trials by the CRL (Community Reference Laboratory of the Joint Research Centre at Ispra). However, the 15 CoExtraInternationalConference implementationofsuchmethodsvalidatedbyusingaparticularchemistryandgenerallyaparticular kind of apparatus may be costly and thus induce inappropriate analytical costs. CoExtra thus decided to compare chemistries and apparatuses to provide an enlarged freedom to laboratories applyingthesetechniques.AlternativedetectionmethodstoPCRwerealsostudiedaswellasfitfor purposeapparatustobeusedinfields.Moregenerallyspeaking,severalwaystoimprovethecost effectivenessofcurrentanalyticalmethodswereassessed. AstheGMOproductionisincreasingworldwide,numerousincidentsofinvoluntaryreleaseofGMO occurred over the last years. GMO approved earlier in third country (e.g. asynchronous approvals betweene.g.USAandtheEU)haveappearedontheEuropeanmarkets.Moreworrying,newcomers in GMO production, such as some emerging countries, have developed unapproved GMO which havenowreachedtheEuropeanmarkets.InresponsetothisarrivalofseveralEUunapprovedGMO, CoExtra launched studies for developing detection methods for detecting EU unapproved GMOs. The same applied to GMO with stacked genes; some being unapproved though their isolated counterpartmaybeapproved,andtodetermineaccuratelythekernelscontentsofsampleshaving GMOmixturesofstackedandnonstackedgenes. Inorder toretrieveinformationfrom stakeholders andshareresultswithstakeholders,a dialogue wasinitiatedthroughthewebsite(www.coextra.eu),newsletters,focusgroups,andaStakeholder Advisory Board. In addition, the interviews carried out for the supply chains management and economicstudies.Thisdialoguewasalsoimprovedduringalargestudyofconsumers’attitudesand opinionsinseveralEuropeancountries.FromsomeattitudesobservedinthefocusgroupsCoExtra started studies on how to solve the issue of “low botanical presence’, where , for example a non GMOcargomaybeadmixedwithverylowlevelsofadifferentGMOcultivar. The coexistence and the impact of traceability are both legal issues, thus several studies were launched on the current status of coexistence and traceability legal frame, liability and redress mechanisms.Asthescientificexpertiseperseisalsopronetolegalcontests,astudywaslaunched onthis,aswellacostbenefitanalysisfromalegalpointofviewonasupplychaincasestudy. AlltheresultstobeissuedfromCoExtraaredifficulttosynthesizeinawaythatmakesthemeasily madeavailableandmastered,particularlybyallstakeholderssuchasSMEs.Thisisalsotrueforthe laboratories analysts who in routinely face several issues difficult to solve (as for instance the detectionofunapprovedGMO).CoExtrathuslaunchedasetofmodulesofaDSS(DecisionSupport System), integrating economic parts, management of supply chains with decision rules, laboratory analytical parts including careful assessment of the need for detecting unapproved GMOs in a sample. Alltogether,the4yearsresearchofCoExtrahasbeenperformedbymorethan200scientists,with theirteamsandhasbeenattemptingtoprovideinsightsofcurrentpracticesandsolutionstoissues aswellasprovidingsolutionsforunpredictablesituations.Forthefirsttime,aEUresearchprogram hasbeenaddressingthewholesupplychains,fromseedstoretailersshelves,theirpractices,their requirementsfortakingintoaccountboththeircurrentsolutionsandprovidingnewones.Theneeds ofthesupplychainsandtheirimpactonproductionofcropsprovidednewquestionsoncoexistence andtraceability,includingcostandtimeeffectivenessofanalyticalmethods. ThepracticalimplementationoftheseveralobservationsandsolutionsdevelopedbyCoExtrawill haveimportanttechnical,scientific,economicandlegalimpacts. 16 CoExtraInternationalConference SessionA1:MethodsforManagingGeneFlow  A1.1. Biologicalmeasuresforgeneflowmitigation AlexandraHüskenandJoachimSchiemann Julius Kuehn Institute, Federal Research Centre for Cultivated Plants (JKI), Institute for Biosafety of Genetically Modified Plants,Messeweg11/12,D38104Braunschweig.[email protected]  WP1 (“Biological measures for gene flow mitigation) of CoExtra is aimed at assessing and developingbiologicaltoolsandmethodstoallowproducerstogrowthekindsofcropstheychoose with minimal levels of admixture between GM, conventional and organic products. Therefore, the generalobjectiveofthisWPistoanalyse,furtherdevelopandvalidatemethodsforrestrictinggene flowduringcultivationbyremovingorreducingthefertilityofpollenorseedsaswellastoidentify the major drivers of pollen flow over fragmented landscapes. It focus on crops for which GM varietiesarealreadyapprovedorclosetoauthorisation(maizeandrapeseed),andoncropswhose authorisationisexpectedduringthenext5years(sunflower,tobacco).ThemainaimofWP1isto testthestabilityandreliabilityofbiologicalcontainmenttoolslikecytoplasmicmalesterilityinmaize and sunflower, cleistogamy in oilseed rape and plastid transformation in tobacco. Therefore, parametersofgeneflowofCMSmaizeandcleistogamousoilseedrapehasbeenstudiedunderfield conditions located at different sites in Europe. Moreover, data mining was performed to gain informationaboutthesuitabilityofchloroplasttransformationasacontainmentstrategy. Toolsmodellingvelocityandpollenconcentrationsoverheterogeneousfieldswerealsodeveloped toassessthecrosspollinationratesbetweenGMandconventionalmaizeoverlargedistancesand fragmented landscapes. Based on gathered data a model of fluid mechanics was successfully validated.Fieldexperimentswerecarriedouttogaininformationaboutthemajordriversofmaize pollen flow over fragmented landscapes. Various factors involved in maize pollen emission and pollenflowwereanalysedthroughexistingdataanalysisandduetofieldexperiments.Seedlotsare starting points in an ever increasing supply food chain; therefore field experiments of maize seed admixture(1%GMseeds)havebeenconductedtoevaluatetheeffectofseedthresholdsonthefinal outcrossingrateintheharvestproduct. In this presentation, certain results will be presented, which have been obtained in the work package. 17 CoExtraInternationalConference A1.2. Biocontainmentofmaizebycytoplasmicmalesterilityandxenia MunschM.1,2,C.Weider1,N.K.Christov3,X.Foueillassar4,A.Hüsken5,K.H.Camp2,andP.Stamp1 1ETHZ,InstituteofPlantScience,Switzerland 2DelleySeedsandPlants,Switzerland 3AgroBioInstitute,Bulgaria 4–Arvalis,InstitutduVégétal,France 5JuliusKuehnInstitute,InstituteforBiosafetyofGeneticallyModifiedPlants,Germany.  While the genetically modified (GM) cultivations are spreading all over the world, the question of coexistence between the different farming systems is a main concern in Europe. For GM maize cultivation, the main issue is the release of GM pollen in the environment and the potential fertilizationofconventionaland/ororganicneighboringfields.Besidestudiesonisolationdistances betweenthefields,anotherapproachforgeneflowmitigationconsistsofthebiologicalcontainment of the transgene in cytoplasmic malesterile (CMS) plants. Cytoplasmic male sterility in maize (Zea maize L.) is a natural trait due to a dysfunction in the mitochondrial DNA affecting sporogenesis. CMS plants do not produce and release functional pollen. Three major types of malesterile cytoplasm (T, C and Stype) has been defined in maize according to the specific nuclear restorer genes(rfgenes)thatareabletocountermandthemalesterilityandrestorefertility.Breedersused thismaternallyinheritedtraitsincethe1950stominimizethecostsinhybridseedproduction.The Plushybridsystem,i.e.growingsuitablemixturesofGMcytoplasmicmalesterileplants(80%)and unrelatednonGMmalefertileplants(20%),thelatteractingaspollendonors,isaninterestingway forcontrollingthereleaseofpollenfromgeneticallymodifiedmaize.ThePlushybridsystemrelies onthefactthatthefemalefertilityofCMSplantsisnotaffected,andseedscanbesetifvitalpollen is provided. One prerequisite is however essential; the malesterile trait must be reliable under variousenvironmentalconditions.  EuropeanCMShybridsarereliablebiocontainmenttools[1] Our hypothesis in this study was that one or more environmental factors may influence the expressionofthemalesterility.Therefore,fieldinvestigationswerecarriedoutin2005and2006in theframeoftheEuropeanprojectCoExtra.TwentymodernCMShybridsfromdifferentEuropean breeding companies representing all three cytoplasm types were tested in 17 environments in Switzerland,Bulgaria,GermanyandinFrance.Stableandunstablemalesterilityoccurredinallthree CMStypes.Thereversiontofertilitywasduetoaninteractionbetweengenetic(presenceofminorrf genes)andclimatic(airtemperature,photoperiodandwatervapor)factors.CMSTwasidentifiedas the most stable type of malesterile cytoplasm; nevertheless, due to its susceptibility to the fungi Bipolaris maydis, its use may be limited to the growth of smallscaled transgenic fields, e.g. molecular farming. While CMSS was often subject to restoration of fertility, the C type of male sterilitywassimilartotheTtypewithregardtomaintainingthemalesterilityandcouldbeapplied inalargerscaleforthegrowthofe.g.Btmaize(inmixturewithnontransgenicmalefertileplants). Even in situations, where the malefertile component of the PlusHybrid needs to be genetically modified too (e.g. herbicide tolerant trait), such a cultivation system can reduce the release of transgenicpollenby80%comparedtoaregularGMmaizestand,where100%ofthehybridsrelease transgenicpollen.    18 CoExtraInternationalConference MaizePlusHybridsincreasegrainyield[2,3] Beside their potential as a biocontainment tool, maize PlusHybrids combine benefits of male sterility (CMS effect) and allopollination (xenia effect) regarding the grain yield. They often outperform the corresponding malefertile sibpollinated hybrids. The potential gain in yield afforded by modern European PlusHybrid was investigated in a preliminary field trial in 2005 (3 locationsinSwitzerland)andinaEuropeanringtrialin2006and2007(12locationsinSwitzerland, Bulgaria,Germanyandin France). ManyPlusHybridsincreased grainyield,onaverage,by 10%or more and by up to 20% in specific environments. The PlusHybrid effect affected both yield components, CMS leading mainly to a higher number of kernels and the xenia effect mainly to an increase in the thousand kernel weight. While the CMS effect depended strongly on the environment,thexeniawasconsistentinallenvironmentsbutitsextentvaried. CytoplasmicmalesterilityisanelegantwaytominimizeoreveneliminatetheproblemofGMpollen flowofadjacentconventionalororganicfieldsifstableTandCcytoplasmisused.ThePlusHybrid systemwouldbeausefultooltoachieveanagriculturalbiocontainmentsystem.Forthissystem,a high level of male sterility must be maintained, as shown by this study. Furthermore, appropriate combinationsofCMShybridsandfertilepollinatorscanleadtoasignificantgaininyieldthatwould definitelyboosttheacceptanceofabiocontainmentsystemwithcytoplasmicmalesterility.  References: 1. C.Weider,P.Stamp,N.Christov,A.Husken,X.Foueillassar,K.H.CampandM.Munsch,Crop Sci49,7784(2009). 2. M.Munsch,K.H.Camp,P.StampandC.Weider,Maydica,inpress(2009). 3. M.Munsch,P.Stamp,N.Christov,X.Foueillassar,A.Hüsken,K.H.CampandC.Weider, CropSci,inreview(2009). 19 CoExtraInternationalConference A1.3. Pollencontainmentbycleistogamyinoilseedrape XavierPinochet2,AlexandraHuesken1,CharlesNjontie1,MartineLeflon2,DonPendergast3,Simon Kightley3 1 JKI,InstituteforBiosafetyofGeneticallyModifiedPlants,Messeweg11/12,D38104Braunschweig,Germany. 2 CETIOM,CentredeGrignon,BPno.4,F78850ThivervalGrignon,France. 3 NIAB,HuntingdonRoad,Cambridge,CB03OLE,UK  [email protected]or[email protected]  Thediversificationoffarmingproductionsystemswiththeapparitionoftransgeniccrops,aswellas thespecializationofcropscultivarsfordifferentmarkets,requiremeasurestopreventadventitious presence in productions at the field, storage and refinement level. For instance, in oilseed rape crops, such means are necessary to allow the coexistence of productions requiring different fatty acid composition. In fields, adventitious presence in adjacent fields is mainly due to pollen flow, whichhavetobereducedtomakepossiblethecoexistenceofdifferentcrops.Pollenflowbetween adjacent fields may be reduced by physical ways: by putting separation distances between fields grown with the same crops or by surrounding the crop of which the pollen is considered as contaminant with a buffer crop strips. Biological ways of containment, such as male sterility or cleistogamy, may also be used depending on the species. One simple way to prevent pollen flow betweenoilseedrapeistoensurethattheirflowersdonotopen.Cleistogamousplantsdonotexist naturally among the genetic resources of the oilseed rape species, but different lines of cleistogamousoilseedrapewereobtainedbychemicalinducedmutagenesisatINRARennes(Patent FR 97 15768). The cleistogamous trait is controlled by one gene (Renard and Tanguy 1997) and would be a good way of securing biocontainment, on condition that this trait is stable during the floweringperiodandundervariousenvironmentalandagriculturalconditions.Oneaimofourstudy was to test the stability of the cleistogamous trait in the field under several environmental conditions. In this goal, the flower opening level was observed at different dates during the floweringperiod,ontwocleistogamousgenotypestestedinthreelocations,duringtwosuccessive years and under two treatments (with or without the application of a growth regulator at the vegetative restarting at the end of winter). The second aim was estimate the rate of adventitious presenceofcleistogamouslinesbyallopollenunderseveralenvironmentalconditions.Inthisgoal, the allopollination in seed sets collected on Clg1 plants was tested in three locations using erucic acidasamarkerduringtwosuccessiveyears.  MaterialandMethods Thestabilityofthecleistogamoustraitwasassessedfortworapeseedcleistogamouslines,Clg1and Clg2,correspondingtothelines17046and16960,respectively,providedbyINRARennes(PatentFR 97 15768). Control cultivars were used in each site. In each site, a splitsplot field design, using a randomized block design, in four replications, was carried out, with elementary plot having areas between22.5m²and47,5m².Thedevelopmentofthecropwascharacterizedbynotationsofthe dates when key development stages were reached and the plant height at maturity. During the floweringperiod,thestabilityofthecleistogamoustraitwasassessedvisuallybyscoringofopening levelonmatureflowersoftheinflorescencewithathreelevelscale:thefirstclasscorrespondedto thefullopenedflowers,thesecondclassofthetotallyclosedflowersthatappearedlikeabigyellow bud, and the last class of the partially opened flowers. Ten plants were scored per plot, with notationofatleastfiveflowersonthemainstemandononesecondarystem. The allopollination was assessed for one rapeseed cleistogamous line (Clg1, corresponding to the lines 17046 provided by INRARennes (Patent FR 97 15768)). As a pollinator cultivar a high erucic 20 CoExtraInternationalConference acid rapeseed line (Markant) was used in each site. The trial was isolated by at least 500m from otherrapeseedfields.Thetrialwascomposedof2neighbouringplots:Thefirstplotwassownwitha mixtureof99%ofMarcant(erucicline)seedsand1%ofClg1(cleistogamousline)seeds.Thesecond plotwassownwiththecleistogamouslineCleisto1.Eachplotwas50mlongand50mlargeandthe sowingrowshadthesamedirectionasthelimitbetweenthetwoplots,andasthedominantwind. Correlations between rates of seeds derived from crosses with the erucic line and the erucic acid content in seed sets were established in each site according to the erucic acid content of seeds producedbymanualcrossesbetweenClg1andtheerucicline.  Results The first experiment showed that flowers of cleistogamous lines are mostly totally closed, but a variable proportion of flowers were observed as partially open. The average percentage of totally closed flowers (Clg1 and Clg2) reached 72.03% at location 1 (2007), 80.91% at location 2 (2007), 85.05% at location 3 (2007), 86.96% at location 2 (2006), 88.91% in at location 1 and 89.69% at location3(2006),withstandarddeviationsof26.6,24.3,19.3,9.54,7.9and6.6,respectivelyineach site x year. Global analyses of all the data from the six site x year combinations revealed that the environment (site x year) had an effect on the stability of the cleistogamous trait, as differences amongsitesandyearswereobserved.Themaineffectofgenotype(Clg1orClg2)explained33%of thevariabilityofthepercentageoftotallyclosedflowers.Thisstatisticalresultreflectsthedifference ofmeanandofvariance showedby thetwo genotypes:ineachenvironment,Clg1showedahigh stabilitylevelforthecleistogamoustrait,whereasClg2showedahigherandmorevariablerateof partially open flowers. Finally, a low but significant difference was also observed between the notationsdoneontheprimaryoronsecondarystems,andtheapplicationofgrowthregulatorhad nosignificanteffect. The second multisite experiment showed that the environment (site x year) had an effect on the allopollination,asdifferencesamongsitesandyearswereobserved.AllogamyratesofClg1undera highpressureofallopollen(Clg1sampledinerucicblock)varyinthreelocationsbetween4.4%and 16.2%.Thesamplescollectedonopenpollinatedcleistogamousplants(Clg1sampledinClg1block) at different distances from the erucic plots showed that the percentage of allogamy rates rapidly dropped over the initial meters around the pollen source and decreased as the distance from the pollensourceincreased.Insamples(location2)collectedonplantsat0mfromtheerucicplot,the erucic acid content reached at mean 1.64%, but at 6m, we observed only 0.26% of erucic acid. However,erucicacidwasalsodetectedinsamplescollectedat48mfromtheerucicplot,showing thatadventitiouspresence,atlowrates(lessthan0.2%)mayoccuratlargedistances.  Conclusions The main result from our various studies is that cleistogamy has a major potential for limiting crosspollinationduetothestrongreductionofthepollencloud.Wesuggestthatisolationdistances implementedforoilseedrapecouldbedramaticallyreducedwhenusingcleistogamicoilseedrapeas acontainmentstrategy.  References: 1. RenardM.andX.Tanguy(1997):Obtentiondemutantscléistogamesdecrucifères.Brevet FR971576. 2. LeflonM,HueskenA.,NjontieC.,KightleyS.,PendergastD.,PierreJ.,RenardM.,PinochetX. (2009)Stabilityofthecleistogamoustraitduringthefolloweringperiodofoilseedrape AcceptedinPlantBreeding. 21 CoExtraInternationalConference A1.4. Chloroplasttransformationandtransgenecontainment RalphBock MaxPlanckInstitut für Molekulare Pflanzenphysiologie, Am Mühlenberg 1, D14476 PotsdamGolm, Germany; Tel.: +49(0)3315678700,Fax:+49(0)3315678701,Email:rbock@mpimpgolm.mpg.de  Plantswithtransgenicplastidgenomes(referredtoas„transplastomic“plants)provideanattractive alternative to conventional transgenic plants (Ruf et al., 2001; Bock and Khan, 2004) and are increasingly used in metabolic engineering, resistance engineering and molecular farming (Bock 2007a).Theplastidtransformationtechnologyoffersseveraltechnicalattractions,suchas,highlevel transgeneexpression(reachingforeignproteinaccumulationlevelsofupto>70%oftheplant’stotal solubleprotein;Oeyetal.,2009),convenienttransgenestackinginoperons,absenceofepigenetic transgene instability (no gene silencing and position effects) and precise transgene integration by homologous recombination (Bock, 2001; Bock 2007a). Furthermore, the increased biosafety provided by transplastomic plants is of particular relevance to future applications of genetic engineeringinagricultureandbiotechnology.Plastids(chloroplasts)arematernallyinheritedinmost crops. Maternal inheritance excludes plastid genes and transgenes from pollen transmission (Bock 2007b).Therefore,plastidtransformationisconsideredtoprovideasuperbtooltoensuretransgene containmentandimprovethebiosafetyoftransgenicplants.Inalargescalestudy,wehaverecently assessedhowstrictmaternalinheritanceisandhowmuchincreaseintransgeneconfinementplastid transformationtechnologyconfers.Wehavedevelopedanexperimentalsystemfacilitatingstringent selectionforoccasionalpaternalplastidtransmission(Rufetal.,2007).Inalargegeneticscreen,we detectedlowlevelpaternalinheritanceoftransgenicplastidsintobacco(Nicotianatabacum),oneof the currently most preferred species in molecular farming (i. e., the highyield production of pharmaceuticalsinplants).WhilethefrequencyoftransmissionintothecotyledonsofF1seedlings was approximately 1.58 x 105 (upon 100% crossfertilization), transmission into the shoot apical meristemwassignificantlylower(2.86x106).Astheseexperimentsaddresstheworstcasescenario (100% crossfertilization, strong selection for the transgenic plastids), our data demonstrate that plastidtransformationprovidesahighlyeffectivetooltoincreasethebiosafetyoftransgenicplants (Ruf et al., 2007). However, in cases where pollen transmission must be prevented altogether, stacking with other containment methods will be necessary to eliminate the residual outcrossing risk.  References: 1. Bock,R.(2001).Transgenicchloroplastsinbasicresearchandplantbiotechnology.J.Mol. Biol.,312,425438. 2. Bock,R.andKhan,M.S.(2004).Tamingplastidsforagreenfuture.TrendsBiotechnol.,22, 311318. 3. Bock,R.(2007a).Plastidbiotechnology:prospectsforherbicideandinsectresistance, metabolicengineeringandmolecularfarming.Curr.Op.Biotechnol.,18,100106. 4. Bock,R.(2007b).Structure,functionandinheritanceofplastidgenomes.TopicsCurr.Genet., 19,2963. 5. Oey,M.,Lohse,M.,Kreikemeyer,B.andBock,R.(2009).Exhaustionofthechloroplast proteinsynthesiscapacitybymassiveexpressionofahighlystableproteinantibiotic.Plant J.,57,436445. 22 CoExtraInternationalConference 6. Ruf,S.,Hermann,M.,Berger,I.J.,Carrer,H.andBock,R.(2001).Stablegenetic transformationoftomatoplastidsandexpressionofaforeignproteininfruit.Nature Biotechnol.,19,870875. 7. Ruf,S.,Karcher,D.andBock,R.(2007).Determiningthetransgenecontainmentlevel providedbychloroplasttransformation   23 CoExtraInternationalConference A1.5. Mesoscaledispersalofmaizepollenandimplicationsforgeneflow SylvainDUPONT,YvesBRUNET INRA,Ephyse,Bordeaux,France  The growing introduction of genetically modified (GM) crops has generated a host of research efforts aimed at investigating the possibilities for coexistence between GM, conventional and organic farming systems. Published experimental and modelling studies aimed at characterizing pollen dispersal have shown that most pollen emitted by a source field deposits within a short distancefromthelatter,butalsothattheobserveddispersalfunctionshavelongfattails,makingit possibleforpollentocontaminateplantsatratherlongdistances. Such possibility has been recently confirmed from (i) a series of airborne measurements of maize andoilseedrapepollenconcentrationandviabilityintheatmosphericboundarylayer,(ii)chamber measurements of pollen viability in a large range of temperature and humidity conditions and (iii) observationsoffecundationsinisolatedplotsofwhitekernelmaize,atseveralkmfromanymaize field. Inordertobetterunderstandlongrangedispersalofmaizepollenanapproachhasbeendeveloped tosimulatethetrajectoriesanddehydrationofpollengrainsintheatmosphereatregionalscale.To thispurposethenonhydrostaticmesoscaleMesoNHmodelhasbeenmodifiedsoastointroduce sourcetermsforpollenemission,conservationequationsforpollenconcentrationandmoisture,and a deposition velocity. Simulations have been performed over SouthWest France on several days during the maize pollination period. MesoNH is run in a twoway nested configuration including threenestedcomputationaldomainsdowntoa2kmhorizontalresolution.GISbasedlandusemaps are used for the surface conditions, featuring all the maize fields of the region, as previously identified from satellite data. Considering several days during which airborne measurements were performed,observedandsimulatedconcentrationprofilesarefoundtoagreewellthroughoutthe atmospheric boundary layer. The simulations allow the pollen plume to be characterized through each day and deposition maps of viable pollen to be produced. The calculated deposition rates at remotedistancesfromthemaizefieldsareinthesamerangeasthoseobservedinsitu.Theresults provide evidence that background fortitious contamination is unavoidable at regional scale. Additionaltestsimulationswillbeperformedusingspecificlandusepatternsinordertoquantifythe impactoflandscapestructureonregionalpollendeposition. 24 CoExtraInternationalConference Session A2: Coexistence and Traceability in Agriculture and Food Production  A2.1. Empiricalanalysisofcoexistenceincommoditysupplychains JuergenBez1,RomainBourgier2,JamesCopeland3,MiaEeckhout4,ChemaGil5,NicolasGryson4, MortenGylling6,MarianneLeBail2,7,BaptisteLécroart2,MariuszMaciejczak8,VladimirMeglic9,Klaus Menrad10,AntoineMesséan2,LouisGeorgesSoler2,MattthiasStolze11,CiroTapia12,Aurélie Trouillier2 1 FhGIVV,FraunhoferGesellschaftzurAngewandtenForschung,Germany 2 INRA,InstitutNationaldelaRechercheAgronomique,France 3 CSL,CentralScienceLaboratory,UK 4 HogeschoolGent,Belgium 5 CREDA,CentrodeInvestigacionenEconomiaYDesarrolloAgroalimentariosUPCIRTA,Spain 6 FOI,TheDanishResearchInstituteofFoodEconomics,Denmark 7 AgroParisTech,France 8 WarsawUniversityofLifeSciencesSGGW,Poland 9 KIS,AgriculturalInstituteofSlovenia,Slovenia 10 FW,UniversityofAppliedSciencesofWeihenstephan,Germany 11 FiBL,ForschungsinstitutfürBiologischenLandbau,Switzerland 12 INTA,InstitutoNacionaldeTecnologaAgropecuaria,Argentina  Introduction Coexistence refers to the ability of farmers and consumers to make a practical choice between conventional,organic,andgeneticallymodified(GM)products,basedoncompliancewiththelegal obligationforlabellingand/orpuritystandards.AdventitiousmixingofGMmaterialwithanonGM productcanoccuratvariousstagesalongtheproductsupplychain,fromthefieldwherethecropis grown to the handling and processing plant. In the framework of CoExtra, the organization of different supply chains were  analysed and sensitive points and processes were identified with respect to GM and nonGM admixture and traceability. Seven commodity supply chains were investigated in various countries: soybean, maize, sugar beet, rapeseed, wheat, fresh tomato and potatoes. Methodologyused This empirical analysis of coexistence was based upon supply chain analysis and stakeholders’ interviews.Interviewsfocuseduponageneraldescriptionofcompaniesandprocesses,andonthe solutions currently adopted to deal with coexistence between GM and nonGM products. Supply chainshavenotbeenfacedtothecoexistenceissuewiththesamedegree,especiallyduetothefact that only a few GM varieties have been authorized in Europe. Thus, questionnaires also included questions about existing specialties supply chain (such as waxy maize, upper standard rapeseed, erucic rapeseed, etc) to gain an insight into how some stakeholders cope with the coexistence betweendifferenttypesofconventionalproducts.   25 CoExtraInternationalConference Resultsanddiscussion The study of different commodity supply chains enabled the identification of critical points from seed production to retail. Furthermore, different strategies stakeholders may choose were identified,aswellastheprerequisites,strengthsandweaknessesofdifferentstrategies.  Whatarethecriticalpointswithinthesupplychains? Oneofthemostcrucialpointsinthesupplychainiscropproduction.Admixtureatthisstageofthe supply chain may be spread over many different endproduct batches and should be thoroughly managed. Admixtures at crop production level may be due to seed impurities, volunteers, cross pollination between GM and nonGM crops, and insufficient cleaning of sowing and harvesting machinery in case a nonGM crop is sown and/or harvested after a GM crop. The level of risk associatedwithvolunteers,seedimpurities,andcrosspollinationishighlydependentuponthecrop biology. Elevators are identified as one of the main sources of unintended admixture, as in the wheat, soybean,rapeseed,andmaizesupplychains.Therearealsorisksofadmixturetheprocessinglevel whencrushingrapeseed,wetmillingmaizeandprocessingpotatoes.Therisksofadmixtureincrease with the number of operators in the supply chain and product flows. Within the processes of storage, processing and trading various critical points were identified. Therefore, the ex ante perception of coexistence feasibility differs from one commodity to another, and from one stakeholdertoanother.Ontheonehand,somestakeholdersconsiderthatGMOandnonGMOare justdifferentcommodityqualitiesandcanbeprocessedlikeanyotherqualityasfarasanadapted qualityassurancesystemisundertaken.Ontheotherhand,severalstakeholdersfinditdifficultto cope with coexistence and consider that ensuring coexistence between GM and nonGM commoditiesrequiresrestructuringoftheirprocessandadditionalinvestments.  Whatkindofstrategiescanbeadoptedtohandleissuesarisingatthesecriticalpoints? Atthemoment,thereislittleexperienceoncoexistencebetweenGMandnonGMproducts(mainly soybean and to a lesser extent, maize). Interviews showed that downstream stakeholders require conventional(nonGM)products tobecompliantwithalowerthreshold (0.1%or0.01%) than the 0.9% regulatory threshold. In situations where GM and nonGM coexist, strategies adopted to handle coexistence are different between food and feed supply chains. In fact, no labelling rules applyforproductsderivedfromanimalsfedwith(non)GMfeed.Ondemandoftheretailers,food processors have replaced GM soy ingredients with alternative ingredients derived from none GM criticalcropssuchassunflowers.Forthesoybeanfeedsupplychain,systemsofidentitypreservation of nonGM products have been introduced in order to guarantee a purity threshold of 0.1%. As a result, an increased level of contract detail  and some vertical integration of activities have been observedinthesoybeanchain.Furthermore,severalstakeholdershaveintroducedbooksofcharge, whichdescribetheconditionsofproductionanddeliveryofspecificproducts,inordertoensurethe segregation of GM and nonGM flows. Next to production requirements, these books include requirements with respect to sampling plans, GMO detection, registration of activities and management.Allactivitiesareinspectedbyindependentthirdparties.However,asanimalproduct labellingisnotpossible,thefeedindustryhastroubleinassigningavaluetotheeffortsmadebythe manufacturers. At the elevator and processing level, several scenarios for coexistence were identified: (i) spatial isolation(dedicatedplants),(ii)lineisolation(useofseparatedproductionlines),and(iii)temporal specialisationwithalternationofproducts.ThededicationofcompaniesorplantstoeitherGMor nonGM production offers the lowest risk of admixture. In the line isolation strategy, dedicated 26 CoExtraInternationalConference productionlinesareusedin thesame plant,which increasestheriskforadmixtureanddecreases theoverallflexibilityinthecompany.Bothstrategieshowevermaysufferfromundercapacityuse because of changing demands. The temporal specialization of process lines is more flexible but requiresregularcleaningofequipmentsordowngradingofnonGMbatches.Duetodifferencesin sizeandstructure,thechoiceforaspecificstrategyshouldbetakenonacasebycasebasisandis likelytobedrivenbymarketdemand. 27 CoExtraInternationalConference A2.2. ModellingcoexistencebetweenGMandnonGMwithinsupplychains BaptisteLECROART1,AntoineMESSEAN1,LouisGeorgesSOLER2 1 2 INRAEcoInnov,ThivervalGrignon,France, INRAALISS,Ivry,France  Introduction Coexistence is an approach allowing farmers to choose between conventional, organic and geneticallymodified(GM)cropsandallowingconsumerstochoosebetweendifferentfoodproducts subjecttoobligationsregardinglabellingandpurity.CoexistencebetweenGMandnonGMsupply chains is a complex issue, because adventitious mixing of GM material with nonGM product can occuratany oneofthestagesofproductionandanywherealong thesupplychain,fromthefield where the crop is grown to its handling and processing. Another major facet of GM and noGM coexistenceisthefactthattheGMcontentofaproductisnotavisibleattribute.Meantobridgethe gap in information do exist (product testing, using model), but they are subject to error. In this paper,wepresentasimulationmodelofthecoexistencebetweenGMandnonGMproductsalong supplychains.Morespecifically,theframeworkofthemodelisinspiredbythestarchmaizesupply chain. The aim of this model is to assess the ability of the supply chain to provide final nonGM product compliant with a required threshold (0.9% labelling threshold for example) and to discuss theimpactofthemeanstobridgetheinformationgaponthisprobabilityofcompliance.  MaterielandMethods The model simulates GM and nonGM flows, and takes into account admixture and dilution functionsbetweenGMandnonGMbatchesalongthesupplychain.Inspiredontheexampleofthe starchmaizesupplychain,threekeystagesofthesupplychainareconsidered:grainproductionat field level, grain collection (including drying), and processing. Firstly, the MAPOD geneflow model (Angevin et al., 2008) is used to simulate GM adventitious presence in nonGM harvests due to crosspollination between GM and nonGM maize. Within the downstream supply chain, there is onlyonedryerandoneprocessingplant.Hence,GMandnonGMmaterialaresuccessivelyhandled in the same equipments. On the contrary, storage capacities are considered nonlimiting in the model and admixture due to storage equipments is considered negligible. At the maize collection level,themodelsimulatesontheonehandadmixturebetweenseveralbatchesblendedinasame bin, and on the other hand admixture between succeeding batches during drying process. Finally, the model simulates admixture between succeeding batches at processing. We have adopted a compartmentalmodellingapproachoftheprocesstoquantifyrisksofadmixture. StakeholdersdefinethefrequencyatwhichGMandnonGMflowalternateatdryingandprocessing levels (scheduling parameters). GM and nonGM batches are then randomly ordered according to thesevariables. Oncesequencesofbatcheshavebeenscheduled,uncertaintyremainsabouttheGMcontentofthe batches,allthemorethatitisnotavisibleattribute.Threekindsofcontrolsystemmightbesetup inthemodel: 1. Simpletraceability:thissystemallowsstakeholderstoidentifywhetherthebatchescomes fromeitherGMornonGMvarieties. 2. Automaticdowngrading:thesimpletraceabilitysystemissupplementedbyruleson automaticdowngradingofnonGMbatchesdriedand/orprocessedafterGMbatches. 3. PCRTesting:inadditiontothesimpletraceabilitysystem,testingisusedtogather informationonthenonGMbatches.Themodeltakesintoaccountthefactthattestingcan 28 CoExtraInternationalConference beinaccurate(Starbird,2007).Weassumedaproportionalerrorbysimulatingmeasurement uncertaintywithalognormaldistribution.Testingcanbecarriedoutbeforeand/orafter processing. Two contrasted sets of admixture parameters (at drying and processing levels) were taken into accountforthesimulations,correspondingtolowandhighlevelofadmixturebetweensucceeding lots.Inaddition,previousstudieshavehighlightedthatthedistributionofGMadventitiouspresence innonGMharvestsisquitevariableamongregions(LeBailetal.,submitted).Thus,threecontrasted distributionsoftheGMadventitiouspresenceinnonGMharvestweretakenintoaccount,inorder to assess the effect of the input purity rate on the output purity rate. As far as the scheduling scenarioswereconcerned,twovaluesoftheschedulingparametersweretakenintoaccount:10and 100. Forscenarios2(automaticdowngrading)and3(PCRtesting),themodelidentifiesthestrategythat maximisestheprofit.Profitdependsonthenumberofbatchesofeachtype(GMandnonGM),on the testing cost and, on the probability that nonGM batches are compliant with the required threshold,accordingtoclientstesting.Clienttestingisperformedseveraltimesandthemeanvalue isconsideredfortheprofitcalculation.  Resultsanddiscussion Workonthesimulationmodelisstillongoing.Nonetheless,firstsimulationresultsshowthatchain organization,fromtheupstreamproducerstothedownstreamstakeholders,playsacrucialrolein maintainingorimprovingthenonGMproductcompliancewiththelabellingthreshold.Inaddition, modelshouldallowcomparingvariousstrategies.  References: 1. AngevinF.,KleinE.K.,ChoimetC.,GauffreteauA.,LavigneC.,MesséanA.,MeynardJ.M., 2008.Modellingimpactsofcroppingsystemsandclimateonmaizecrosspollinationin agriculturallandscapes:TheMAPODModel.EuropeanJournalofAgronomy,28(3):471484. 2. LeBailM.,LécroartB.,GauffreteauG.,AngevinF.,MesséanA.,2009.Effectofthestructural variablesoflandscapesontherisksofspatialdisseminationbetweenGMandnonGM maize.submittedtoEuropeanJournalofAgronomy. 3. StarbirdS.A.,2007.Testingerrors,suppliersegregation,andfoodsafety.Agricultural economics,36,325334.   29 CoExtraInternationalConference A2.3. CostsandbenefitsofsegregationandtraceabilitybetweenGMand nonGMsupplychainsoffinalfoodproducts KlausMENRAD1,AndreasGABRIEL1, 1 ScienceCentreStraubing,UniversityofAppliedSciencesWeihenstephan,Straubing,Germany  Aims Thebasicintentionofthispartoftheprojectwastoquantifythecostsandtoidentifythebenefitsof traceabilityandcoexistencesystemsforGMfood(andfeed)fromtheseedtothefinalproductat theretailstageinseveralcountriesandsupplychainsrespectingthe0.9%thresholdforlabellingof GM food. Thereby the production and processing stages of eligible GM crops like wheat, sugar, rapeseed, soya and maize are analyzed with respect to cost structures originating from efforts to organize coexistence and segregation. The multinational analysis of several supply chains with partly different end products allows meaningful comparison of economic and technical consequences of coexistence measures on the different stakeholders along the supply chains. As someoftheanalyzedfoodstuffslikesoylecithin,sugarandstarchderivatesareusedasingredients forcomplexfood(andfeed)products(likee.g.chocolate,frozenpizza,compoundfeed),theimpacts ofGMandnonGMcoexistenceonthevaluechainsofsuchcomplexproductsareanalyzedaswell. Asthesecomplexproductsarecomposedofseveralcriticalingredientsandthistypeofproductis closer to the food retailer and consumer, the compliance with coexistence regulations and thresholdshastoberealizedinanevenmorecomplexenvironment.Anothertargetoftheproject was to detect benefits emerging by the implementation of product differentiation systems and assesstheirimpactsonthedifferentlevelsofthevaluechainfromtheseedproducertotheprivate consumeroffoods.  Methodology For calculating the coexistence and segregation costs, an Excelbased simulation model has been developedwhichincludespotentialcostcategoriesoneachlevelofthevaluechains.Thetotalcosts ateachlevelfollowstheprincipletoaggregateallincurredcostsforcultivating,transportationand processing of the raw material crops on the different levels and to increase the price of the final product at each level. The resulting price for the secured nonGM crop or product represents automaticallythenonGMcommoditypriceonthenextlevelofthevaluechain,whilethepriceof GM commodity is assumed as the current price level without any coexistence and traceability measures.Thisprincipleisusedatallstagesofthesupplychainthusaggregatingtheadditionalcosts forrespectingthe0.9%thresholdofadventitiouspresenceonalllevels(attheseedlevelthe0.5% thresholdismainlyrespected)andsettingthepriceforthenonGMproductattheendofthevalue chain. This conceptual approach is also perpetuated when identifying the costs in the processing of complexfoodandfeedproductslikechocolate,frozenpizzaandcompoundsoyandwheatfeed.The model allows for an isolated view on every single ingredient that carries potential risk of GM contaminationandtheemergingcosttypescanbecalculatedseparately. Subsequently, benefits of introduced coexistence and product differentiation systems (IP, segregation or traceability systems) are analyzed within a literature research and finally the emergingbenefitsofsuchsystemsareconfrontedwiththeoriginatingcosts.   30 CoExtraInternationalConference Results The generated cost calculation model was applied on the food and feed value chains of wheat (starch,flour,feedadditive),sugar(sugarbeetasrawmaterial),rapeseedoil,soy(feedadditive)and maize (starch, feed additive) in the participating countries Germany, Denmark, Poland, UK and Switzerland.Basically,thecoststructuresandtheresultsofthecostcalculationsbetweenthesingle countries do not only differ because of national differences in implementing the existing co existence regulations of the EU, divergent farming or industry structure, but also due to the information given in the conducted interviews and available data e.g. concerning costs of specific activities.WhileforaSwissoilmillcompanythecommoditydeliveringsystemisquitemanageable and the input testing of elevated rapeseed is negligible, the bigger companies in Germany and Poland, with several processing sites cannot manage threats of admixture without monitoring systems at the entry gates. Another example for differing cost structures is the impact of field structuresoncoexistenceschemesinfarminginthedifferentcountries.Severalpossiblestrategies ofmaintainingisolationdistancesbetweenGMandnonGMfieldscanbeapplieddependingonthe regional field distribution and national regulation of liability. While for the German farmers it is assumed that the GM rapeseed farmer has to compensate the loss of gross margin by cultivating alternative crops on a certain discard width by the nonGM farmer, the conditions in Polish agriculture determine buffer zones on GM fields as additional effort to maintain coexistence in rapeseedproduction.Thus,theindividualcombinationofcosttypesandtheparticularoriginofdata havetoberespectedbythecomparisonofthecountryspecificresultsofthecostcalculation. Theprojectteamfacedthemostuncertainfiguresattheproducerlevels(seedandcropproduction) oftheregardedvaluechains.DuetothestilllackingthresholdonGMadventitiouspresenceformost cropsintheEUonlyveryfewinformationexistconcerningthenecessarymeasuresandadditional costsofcoexistenceincertifiedseedproduction.Additionalcostsofcoexistenceandsegregation efforts are calculated with 38 or 86 € per ha respectively for the Danish and German wheat crop production.Forrapeseed,acropwithaquitehighriskofreceptivenessofpollenfromotherplants andvarieties,thetotaladditionalcostsarestatedfrom40€perhainDenmarkupto74€perhain Germany and Poland. At elevator level, within its functions of storage, drying and distribution the riskofadmixtureisdeterminedasquitehigh.Dependingonthesizeoftheelevatorcompanyandits capacities,theadditionalcostsvaryfrom7to16€pertonwheat(Denmark,Germany),18to64€ pertonrapeseed(Germany,Denmark,Switzerland)andareestimatedwitharound30€pertonfor the elevating of maize in Germany. The high ranges in the cost figures can be explained with the different possibilities of the company to apply certain segregation strategies. Transfering these additional costs to the final processing level, the mills, crushers and processors, the increased expensesfortheusedrawmaterials causedby thecoexistenceactivitiesin the previous levelsof the value chain result in the highest costs for commodities and transport when implementing co existence and traceability management systems. Over all regarded chains these commodity costs together with costs for required monitoring systems form more than 90% of all costs for implementingcoexistencesystems.Thetotaladditionalcoexistencecostsattheendofthevalue chain,whichhadtobeaddedonthegeneralproducerprice,arecalculatedwithatleast25€perton wheat starch in Germany, 11 € per ton wheat flour in Denmark or 22 € per ton rapeseed oil in Poland.Theadditionalcoexistencecostsinthecaseofsugarshowthelowestexaltations,asbeet productionandprocessingimplylowerriskandbetterconditionstoavoidadmixtureandmaintain thresholds.Theidentifiedadditionalprizeloadingsofmonofoodproductsareincludedinthecost calculationofcompoundfeedandmultiingredientfoodproductsinordertoanalyzetheeconomic impactofcoexistencesystemswhenhandlingseveralrawmaterialsinonefinalconsumerproduct.    31 CoExtraInternationalConference Conclusion Accordingtotheresultsoftheanalysedfoodsupplychains,significantadditionalcostsareexpected by organising coexistence between genetically modified and nonGM products in the value chain fromproductionoffarmcropsuptotheproduction/processinglevelsofthesinglesupplychainsand bymaintainingmandatory(orvoluntary)thresholdsandregulations.Dependingonfactorslikecrop requirements, farming, storage and elevating systems, processing strategies, monitoring managements etc, the total additional costs of coexistence and product segregation systems can raise up to 13% of the total product turnover at the gates of rapeseed oil mills or starch industry processing wheat and maize. However, as in most value chains the question of coexistence currentlyisatheoreticaloneintheEU,theimplementationandpermanentrunningofcoexistence andsegregationsystemsinthefoodindustrycandecreasetheadditionalcostsduetosavingse.g.in thetestingrequirementsofrawmaterialsorroutineproceduresduringthedocumentationprocess.     32 CoExtraInternationalConference A2.4. Consumers’attitudestotheEUtraceabilityandlabellingregulation. M.CostaFont1andJ.M.Gil1,M.Gylling2,A.Gabriel3,K.Menrad4,P.J.Jones5,W.B.Traill5,R.B. Tranter5,M.Sajdakowska6andM.S.RakowskaBiemans6 1 CREDAUPCIRTA,Barcelona,Spain. 2 InstituteofFoodandResourceEconomics,UniversityofCopenhagen,Denmark. 3 CentreofCompetenceforBiogeneticResources,Straubing, 4 UniversityofAppliedSciences,Weihenstephan,Germany. 5 SchoolofAgriculture,PolicyandDevelopment,UniversityofReading,UK. 6 WarsawUniversityofLifeSciences,Warsaw,Poland.  The introduction of new technologies in the food industries have revolutionized the efficiency of foodproduction,buthasalsoexertedimportantdemandsideeffectsthatcannotbedismissed.This is because new technologies are associated with scientific uncertainty given that not all the social andindividualconsequencesoftheirinceptionarefullyknown.AsMoschini(2008)argued,basedon Gaskelletal.(2006),generalpublicoppositionorreticencetowardsgeneticallymodified(GM)agro food applications responds to: i) human health and environmental concerns, ii) ethical considerations and iii) the role of patents and property rights of multinational corporations. This varietyofreasonsagainstGMagrofoodproductionrevealsacomplexformationprocessofpublic opinion towards GM agrofood production and therefore a complex process for understanding consumers’finaldecision andintentionsregarding GMfood.Themainobjectiveofourstudyisto investigate consumers’ general attitude towards GM food and their willingness to pay (wtp) a premiumforconventionallyproducednonGMfoodandorganicfood. To do that we have first performed a literature review in order to bring together the published evidence on the behavioural frameworks and evidence on the process leading to the public acceptanceofGMfood.Indoingso,weemployasetofclearlydefinedsearchtoolsandalimited number of comprehensive key words. This review concluded: first, that the population can be segregatedinthreemaingroupsregardingattitudestowardGMfood,namely:(i)antiGMfoodor pessimistic, (ii) risktolerant or information searchers and finally (iii) GMaccepters or optimistic. Second,thatconsumerattitudestowardsGMfoodaredrivenbythreemaindimensions,i)risksand benefit perceptions associated to GM food; ii) individual values and attributes and finally iii) knowledgeanditsrelationwithvalues. From the previous review it was also concluded that consumer behaviour towards GM agrofood productionhasmanyanalogieswithotherbehavioursanalysedinthepast.Thisisthecaseofother risky technologies such as pesticide risk exposure, hormonetreated meat, atomic energy and so forth. For instance previous studies based on the Fishbein Multiattribute Model (Fishbein, 1963) revealedthatanattitudeorintentiontowardsaproductorbehaviourisbasedonknowledgeabout theproductorbehaviouritself(Bredahl,1998);thatis,ontheattributesthatpeopleassociatetothe productorbehaviour(Freweretal.,1998).Followingthistheory,wehaveconsideredthatthebest way to study consumers’ final intentions towards GM and nonGM agrofood products entail the applicationofchoiceexperiments.Withinthechoiceexperimentframeworkindividualsareallowed to select among different alternative options, where each option is characterised by a number of attributeswithdifferentlevels(Burtonetal.,2001).Thereforeindividualswillchooseanalternative, amongasetofalternativesthatgeneratestothemthehighestutility. Following consultation with stakeholders, a number of food commodities for study were to be chosen. On the one hand fresh food, e.g. fresh tomatoes, on the other longstored processed commodities, e.g. oil seed or cornflakes. The analysis was performed by means of a multicountry survey (Denmark, Germany, Spain, GB and Poland). The main results of the survey can be 33 CoExtraInternationalConference summarizedasfollows.Freshnessandflavourcanbeconsideredasthemostimportantelementfor food purchasing. However, in GB, Poland and Spain price is also considered. There is a general negative attitude towards GM food in all countries. University scientists and consumer groups are the more trusted sources of information, and Denmark and Germany responders feel themselves more informed that the rest. Regarding to organic food, only German and Danish consumers do spend on organic food. Moreover there is an agreement among countries regarding positive attitudes towards organic food. The study also revealed that GM technology is not considered by respondents as very risky compared with pesticides, artificial hormones or irradiation. Finally, respondents in all study countries prefer conventional food over GM food. However, Spanish respondentsmadeaslightexceptionsincetheywerepreparedtopayapremiumforGMfoodwith healthbenefits.Moreover,allstudycountryrespondentsexceptPolishones,assignedahigherutility fororganicfoodinrelationtoconventionalcounterpart. 34 CoExtraInternationalConference SessionB1:TechnologiesforManagingtheSupplyChain  B1.1. GMOsamplingstrategiesinthefoodandfeedchain ValérieAncel1,GianniBellocchi2,GilbertBerben3,YvesBertheau1,CarloBrera4,MarziaDeGiacomo4, Eric Janssen3, Andre Kobilinsky1, Petra Kozjak5, Roy Macarthur6, Marina Miraglia4, Roberta Onori4, MariaPla7,NinaPapazova8,RomanaRutar5,IsabelTaverniers8,JelkaŠuštarVozlic5. 1 INRA,InstitutNationaldelaRechercheAgronomique,France JRCIHCP,JointResearchCentre,InstituteforHealthandConsumerProtection,EuropeanCommission,Italy 3 CRAW,Belgium 4 ISS,ItalianNationalInstituteofHealth,DepartmentofVeterinaryPublicHealthandFoodSafety,Italy 5 KIS,AgriculturalInstitute,Slovenia 6 CSL,CentralScienceLaboratory,UnitedKingdom 7 CSICIRTA,ConsejoSuperiordeInvestigacionesCient´ficas–InstitutdeRecercaiTecnologiaAgroalimentaries,Spain 8 ILVO,InstituteforAgriculturalandFisheries,Belgium,  Thesamplingplanistheprocedureoftakingasample,fromalot,foranalysisandisthemostcrucial stepinthe“analyticalchain”whenevertheanalyteisnothomogeneouslydistributedinthelot.Itis imperativethatthesamplingstepisperformedasaccuratelyaspossiblesothatthesamplecollected isrepresentativeofthebatchoffoodorfeedunderinvestigationandtogetthemostaccurate“true value”.Withouttheimplementationofagoodsamplingplan,misclassificationofthelotcouldeasily occur, negatively impacting sampling objectives: undesirable economic and legal impacts in trade and inaccurate information being provided to risk assessors/managers. Accordingly, the GMO samplingshouldbeconsideredwiththeothersamplingissuesfacedinthedomainoffoodandfeed safetyandquality. Among the steps usually employed in the evaluation of the GMO level in a lot (sampling, sample preparationandanalysis),thesamplingstepisthemajortotalerrorcontributorandisdependenton theGMOlevel.DuetothevarianceassociatedwitheachstepoftheGMOevaluation,a100%level of certainty is unachievable; resulting in overestimation and underestimation with inaccurate decisionsproducingadversefinancialorsocialimplications. SamplingofGMOinfoodandfeedcommoditiesisperformedbydifferentstakeholderswithawide spectrumofgoals,allofthemimplyingdifferentscenariosandconsequentlyoftenneedingdifferent methodologies. CommissionRecommendation2004/787/ECof4October2004providesdetailedtechnicalguidance forsamplinganddetectionofGMOsrelatedtoRegulation(EC)No.1830/2003inbulkandverylittle anduncleardetailsforpackedproducts.However,amongthegeneralprinciplesforGMOsampling, the Recommendation states that the Member States should take into account the point in the supply chain in which testing is being performed and the degree of heterogeneity, therefore indicatingthatsamplingcanbemodulateddependingonthesituation.Inrespecttotraceabilityand internalqualitycontrolpurposesoperatorsalsoneedanarrayofsamplingprocedurethroughoutthe foodandfeedchain.TheRecommendationalsounderlinestheneedtousesamplingproportionate to the desired specific objectives and the possibility to use sampling strategies other than those indicated in the Recommendation. In other words, it would be appropriate to, in addition to the guidelinessuggestedbytheRecommendation,developfurther“fitforpurpose”samplingstrategies.  35 CoExtraInternationalConference Development of sampling methods has been an important goal within the CoExtra project. In addition, a Modular Decision Support System (DSS) has been developed for producers and official control authorities, to support decisions related to the selection of “fit for purpose” sampling methods. CoExtra goals include: (i) development of “as simplest as possible” sampling methodologies compatible with reliable results; (ii) meeting the needs of different food and feed operators; (iii) optimization of sampling in the different steps of food and feed chain, of the associatedcostanddegreeofaccuracy.Thispresentationwillgiveanoverviewofthedevelopments from CoExtra project relevant to the sampling in different scenarios (field, bulk, processing and retail). IntheCommissionRecommendation2004/787/ECsamplinginthefieldisnotspecificallyaddressed whileincertaincasesofcoexistenceitisimportanttodetermine,beforeharvestinginthefield,the levelofadventitiouspresenceofGMOsinanonGMOfield.IntheCoExtracontext,fieldtrialswere conducted in two successive years aimed at developing a reliable sampling procedure for maize plants in the field (model for fragmented landscape with very small field sizes). Every year 3600 samples were collected to determine outcrossing rate in the field, using data mining techniques. Based on the predictions of spatial variability of outcrossing rate, various possible sampling procedures were tested using the statistical Programme R and different sampling schemes were thendevelopedandvalidated. Asforthefoodprocessingchaintherearethreetypesofmaterialwhichmayneedtobeanalyzedfor the presence of GMOs: raw materials, primary ingredients and final food products. The soybean processingchain(fromgraintolecithin)hasbeenchosenas“casestudy”withrespecttoitsuseina wide range of foodstuff ingredients and additives. The study also provided a simple framework to assistinthedecisionmakingtoallocateresources(broadly“sampling”and,“analysis”costs)andto balancethecostofcontrolversustheriskassociatedwithincorrectdecisionsbasedontestresults. Applying control plans that have been optimised for efficiency between sampling and analysis for thesoyabeanscenario,fitnessforpurposeparameterscanbeattainedmoreeasilywhensampling forsoybeanflourisundertaken. Due to labelling requirements, packaged products are expected to be one of the main targets for controlbodies.Theproblemismultifaceteddependingonmanyfactorsincludingparticletypeand size of different products. Experimental studies on GM soybean packed products were performed and data were processed via distributionfree statistical procedures supplied by software SISSI (Shortcut In Sample Size), to estimate sampling errors associated with number of incremental samples. Dedicated software tools to support sampling and subsampling plans aimed at GM detection through the food and feed chain were also developed: SISSI a novel approach to estimate the optimalsamplesizeinexperimentaldatacollectionandOPACSA(OPtimalACceptanceSamplingby Attributes)anewstatisticaloptimisationsoftwareincludingacostfunctiontofindthecheapestand mostreliablemodeofanalysisbysubsampling. Finally,considerationofgeneralcontrolplansshouldbeundertakenwhereseveralanalytescouldbe sampled,withlowcostsamplingmethodologies.Inthisregard,anongoingexperimentalstudyfor validatingsamplingmethodologiesformycotoxins(Reg.401/2006andfollowing)fitforpurposefor GMOs is in progress. The aim of this study if to verify if the current sampling methodologies for mycotoxins (the more heterogeneously distributed analyte in a lot) could fulfill the requisite of a representative sampling also for GMOs and derived products. Initial results of this study are presented. CommissionRecommendationof4October2004ontechnicalguidanceforsamplinganddetection ofgeneticallymodifiedorganismsandmaterialsproducedfromgeneticallymodifiedorganismsasor inproductsinthecontextofRegulation(EC)No1830/2003.(2004).OfficialJournaloftheEuropean UnionL348:1826. 36 CoExtraInternationalConference References: 1. BellocchiG.,ConfalonieriR.,AcutisM.,GenoveseG.2006.SISSI:aresamplingbased softwareforsamplesizedetermination.Proc.9thEuropeanSocietyforAgronomyCongress, 46September,Warsaw,Poland,741742 2. ThenewsoftwareOPACSA:maximumcontrolminimalcost http://www.coextra.eu/researchlive/reportage851.html. 3. Kobilinsky,A.,Bertheau,Y.,.2005.Minimumcostacceptancesamplingplansforgrain control,withapplicationtoGMOdetection.ChemometricsandIntelligentLaboratory Systems75:189–200 4. COMMISSIONREGULATION(EC)No401/2006of23February2006,layingdownthe methodsofsamplingandanalysisfortheofficialcontrolofthelevelsofmycotoxinsin foodstuffs.OfficialJournaloftheEuropeanUnionL70:1234 37 CoExtraInternationalConference B1.2. RationalizationofGMOtestingbyappropriatesubsamplingand controlplans YvesBertheau1andRoyMacArthur2 1 INRA,Versailles,France 2 FERA,York,UnitedKingdom  Samplingisageneralissueforalldetectionpurposes,beitinfield,insilos,bargesorshipments,with packedorunpackedproducts,incompanies,orontheshelvesofretailers.Therepresentativenessof abulksamplefromalotisanimportantissueforensuringtheaccuracyofmeasurementsmadeon the lot that is sampled, in particular where the target analytes may be homogeneously or heterogeneouslydistributed.Samplinguncertaintyalsoplaysamajorroleintheuncertaintyofthe measurementmadeonlaboratorysampleforwhichgenerallyasmallsubsample,supposedtobe representativeofthelaboratorysampleiseffectivelyanalyzed. NumerousCEN,ISOandprivatesamplingplansareavailable,forgeneralapplication.Inthecaseof GMOs,theECreleasedarecommendationonsamplingthatisnotappliedbyMemberStatesdueto itshighcosts. It is important to remember that a single lot may be sampled in parallel for to detect several analytes:GMOs,pathogens,allergensandmycotoxinsororganismsproducingthem,and analysed with immunological, physical, chemical or DNA based methods. The first question is thus, are samplestakenforonepurposeusableforanotherpurpose?Twoissuesthenarise: 1)Doesthevariousdifferentriskstatusesthatalotmayhavefordifferentanalytesmakeacommon samplingplanasensibleoption? 2)Canwehave,inthesamelocation,analyticallaboratoriesworkingindifferentfieldsofanalysisin order they can share the samples and define common analytical procedures for subsampling and extraction;particularlyiftheanalytes,e.g.proteinsandDNA,differamongtheanalyses? In the absence of a common sampling plan, the aim of the analysis should be used to define sampling plan: for instance sampling plans for environmental purposes differfrom seeds analyses, samplingplansusedseedsofcommoditiesdifferfromthoseusedforcostlyvegetableseeds,suchas salads.  Another issue is how analytical performance impacts on and combines with the uncertainty associated with sampling. For example, the uncertainty associated with a typical quantitative PCR based method might be a factor of 2. Hence, the amount of effort put into sampling for GMOs shouldbejustenoughtonotincreasethatlevelofanalyticaluncertainty.Oncethatgoalisachieved, addingadditionalefforttowardssamplingdoesnotimprovethedecisionsmadeaboutlots.Another issue is the need for of more sensitive techniques, be it quantitative or qualitative detection methods.TheuseofqualitativemethodsisincreasingrapidlyinGMOdetectionforthepurposeof detectingunapprovedGMOs,forinstancewiththe‘MatrixApproach’usingthevalidatedDualChip® microarray developed during the frame of CoExtra. Such need to use qualitative methods is also amplified by the lack of CRMs2 below the LOQ3 of the current quantitative methods for approved GMOs and for unapproved GMOs forwhich reference materials can be either missing orprovided only as DNA extracts. The issue of LLP4 may also require either more sensitive fully quantitative methodsorwaystoassesstheGMOcontentofalotwhennoquantitativevalidatedmethodsare  2 CertifiedReferenceMaterial 3 LimitofQuantification 4 LowLevelPresence 38 CoExtraInternationalConference available.Inthatcontext,themandateoftheCRL5shouldberapidlyextendedtothevalidationof qualitative methods, particularly those used in the ‘Matrix Approach’ and their related controls of donororganismswhenneeded.  Two statistics based methods can then be used to interpret the results of qualitative analytical methods,togiveanestimateofthequantityofGMOspresentinalotofasampletakenfromalot. TheSIMQUANTmethod,basedonMPN6atthelevelofDNAcopiesinsolutionprovidesamethodfor measuringthequantityofGMOsatlowconcentrationsusingqualitativedetection.Asubsampling strategy,alsocalledcontrolplansbymultiattributes,isanothermethodformeasuringthequantity of GMOs using qualitative test results. It provides an estimate with a confidence interval for assessingtheGMOcontentofasamplerelativetoathreshold.Subsamplingisadetectionstrategy particularly used in quality controls (automobiles, manufactured pieces, seeds) which has not, till now,beensufficientlyconsideredintheGMOdetectionarea.  Bytakingintoaccountacostfunction,theOPACSAsoftwarealsoenablestheanalysttodetermine themostcosteffectivewaytodetectanyanalyte,beitbyusingsingleordoublestagessystems.It alsoenablesthesellerandbuyertocalculateand negotiate theirrisksbydefiningtwovalues,the AQL7 and the LQL8 (see figure). The analytical and sampling uncertainty associated with the measurement of GMOs in lots have led stakeholders to apply practical contractual limits of approximately0.1%estimatedGMOcontentwhentestinglotsagainstthe0.9%labellingthreshold. The practical contractual limits have been freely negotiated by stakeholders based on their assessments of risk and LQL AQL values. As, the EC recommendation is also based on a sub samplingstrategy,morecosteffectivesamplingplanscouldbeusedbyMemberStates. D E   5 CommunityReferenceLaboratory 6 MostProbableNumber 7 AcceptableQualityLevel 8 LowQualityLevel 39 CoExtraInternationalConference Inconclusion,rationalizationofthesamplingissueisstillamatterofworkwhosesolutiondepends on the ability of scientists and stakeholders of different analytical methods to work together for developing common and harmonized, more detailed, sampling plans. In the case of GMO, the mycotoxinsampling plan would be an effective alternative to the EC recommendation while the OPACSA software of CoExtra could be used for increasing the costeffectiveness of the detection protocolsofEuropeanstates.Qualitativemethodsarebeingmoreandmoreused,anewparadigm can be distinguished in the future of analysts training. The experience of ISTA9 in diffusing such a detectionandchoicesupportingmethodologywouldbeveryusefulfortraininganalystsoftheGMO area.Theeffectivecosteffectivenessofsuchmethodsshouldbeappreciatedmoreindepthwhen quantitative methods are available. However, in all cases where qualitative or very sensitive methodshavetobeused,forinstanceincaseofharmfulproducts,thesubsamplingstrategy,with single or double stages, is the method of choice for its ability to both partners of a transaction to negotiatetheirrisks.   9 InternationalSeedTestingAssociation 40 CoExtraInternationalConference B1.3. ModularApproachImplemented:Pros,ConsandFuturePerspectives MarcVandenBulcke1,GianniBellocchi2,GilbertBerben3,MalcolmBurns4,KatarinaCankar5,Marzia De Giacomo6, Kristina Gruden5, Arne HolstJensen7, Aliosha Malcewsky8, Marco Mazzara2, Roberta Onori6,NinaPapazova9,EdwigeParlouer10,IsabelTaverniers9,DörteWulff11,DavidZhang12. 1 ScientificInstituteofPublicHealth,J.Wytsmanstraat14,B1050,Brussels,Belgiumemail:[email protected] 2 JRCIHCP,JointResearchCentre,InstituteforHealthandConsumerProtection,EuropeanCommission,Italy 3 CRAW,WalloonAgriculturalResearchCentre,Belgium 4 LGC,AnalyticalTechnology,LGCLimited 5 NIB,NationalInstituteofBiology,Ljubljana,Slovenia 6 ISS,ItalianNationalInstituteofHealth,Italy 7 NVI,NationalVeterinaryInstitute,Norway 8 UPAR,UniversityofParma,Italy 9 ILVO,InstituteforAgriculturalandFisheries,Belgium 10 SCL67,ServiceCommundesLaboratoires,France 11 EUROFINS,GeneScan,Germany 12 GIPGEVES,GEVESDomaineduMagneraud,LaboratoireBIOGEVES,France  TheuseofGeneticallyModifiedOrganisms(GMO)issubjectedtolegalconstraints,eitherwithina deregulating (e.g. USA) or an authorizing (e.g. EC) framework. In either case, compliance with the legal framework is mandatory. Validated methods (and reference materials) represent essential componentswithintheenforcementcompliancewiththelegislationbothbytheproducersandthe officials. Compliance measures invoke investments in quality assurance/quality control (QA/QC) in all enforcement activities of food safety and quality. View the increasing number of GMO, their diversity and their complexity, the high standards set in general for the validation of detection methodsattheEClevelstressesevenstrongertheneedforanefficientcontrolmanagementinthe GMO area. A centralized enforcement organisation at the EC (through Community and National Reference Laboratories) increases transparency for the stakeholders and the consumers. QA/QC processesinproduction,manufacturing,distributionandsalescaninsuchorganisationmoreeasily bestreamlinedandharmonized.FlexibilityandcompatibilityofQA/QCprocessesalongthischainis animportantadvantageforincreasingtransparencyandreducingcost. Withinmethodvalidationtwobasicconceptsareprevalent:ontheonehand,the"globalapproach", prevalentintheUSAandinotherdetectionareas,inwhichthewholeprocessfromtheproductto thefinalmeasurementoutcomeistobevalidatedasawhole.Ontheotherhand,withina"modular approach", the different steps in the analytical process of a food or feed matrix ((sub) sampling, homogenization, extraction, etc.) are considered as separate entities and each of these can be validatedonitsown. HolstJensen and Berdal (2004) have proposed to introduce the concept of ‘modularity’ for the analytical proceduresandvalidation ofmethodsinGMOanalysis.Thebasicideaisthatingeneral aftersamplingfrombulklots,GMOanalysisconsistsofalimitedsetofdistinctstepsthatrepresenta certainelementaryunitwiththeprocess,thesocalled'module'.InGMOanalysisstrictosensu,after sampling,alaboratoryanalyticalsampleissubjecttothefollowingstepsofmodularanalysis:sub sampling, sample homogenization, analyte extraction, target detection and finally target quantification. 41 CoExtraInternationalConference Within the CoExtra project, a number of aspects directly concerning the validity of the modular approach have been assessed, especially technology equivalence, the lack of biasintroduction by moduleinterchangeandthedeterminationofthemeasurementuncertainty,ifnecessary,andthe mathematicalexpressionofmoduleinteractionsandinterdependencies. Technology equivalence for various steps in the GMO analysis both in a global as in a modular approachhasbeenassessed:differentDNAquantificationmethods,extractionmethods,qualitative andquantitativePCRmethodsandthecommutableuseofdifferenttypesofreferencecalibrators. WhileproducttypedependencyastotheabsolutequantitiesofmeasuredtotalDNAlevelscouldbe demonstrated,theinfluenceatthePCRlevelwasnotsignificant.InthecaseofDNAextraction,for onemethodconsiderableaberrantdifferenceswereobservedfromtheothermethodsatthefinal% GMmeasured.AconsiderabledeviationwasobservedwithvariousPCRmethodsatdifferentlevels, with serious errors at higher GMO % levels. Finally, the exchangeable use of different types of reference calibrators (plasmid and genomic DNA) in quantitative GMO analysis was documented bothformaize(commutabilitystudy)andsoy(interlaboratorytrial). Bias introduced by varying the methods applied within a particular module (e.g. changing for instance extraction methods within the extraction module) is a second point that was addressed. Here, it was shown that different extraction methods could be used for various products without introducinglargevariationinthefinalGMcontentmeasurement(exceptforoneparticularmethod, seeabove).ItwasalsodemonstratedthatthequalityoftheDNAanalytecouldhighlyinfluencethe GM% measurement due to PCR inhibition. Criteria and methods assessing the PCR inhibition rate havebeendevelopedwithintheproject.Finally,thetargetintegrityisanotherimportantfactor,and somegeneraldestabilizingfeaturesoftargetmoleculeshavebeenidentified(esp.neighbouringTT sequencesintheDNAtargetamplicon). These results were analyzed by conventional statistical approaches (such as ttest, ANOVA,) to express the interdependency between modules and the difference between extraction methods and food and feed products. Within CoExtra, these statistical approaches were successfully combinedwiththe"vagueset"or"fuzzylogic"mathematics,implementedbythesoftwareAMPE.In thisway,an overallassessmentof "fittopurpose" betweenmethodproduct combinationscanbe expressed in an alternative mathematical way. As such, any subjectivity or individual preferences about the choice of methods foranalysis can be documented. Also, the need for defining transfer criteriabetweenmoduleswasinvokedofwhichanalytepurityandintegritywereconsideredasthe mostimportantones. According to the obtained results within the Coextra project, the "modular approach" can be considered as a useful approach in GMO analysis. Coextra documents valid modularity for: DNA content determination, for DNA extraction, for the reliable use of PCR in a wide range of % GM contentandfortheapplicationofdifferentcalibratorsinGMOquantificationbyrealtimePCR. As such, the "modular approach" provides a good basis for developing a costeffective validation process by the stakeholder. Such approach requires however generally accepted performance criteria for the different types of detection methods (e.g. ENGL criteria), accepted statistical evaluation tools (such as AMPE, SeedCalc, etc.) and appropriate reference materials (such as the IRMMCRMs).Furthereffortswillhavetobemadetointegratetheuseofalltheseelementsinthe future(e.g.CRMscertifiedfortargetcopynumbers). To evaluate whether a certain/new method can be applied within a particular module, requires a reliablereferenceframework(e.g.aGMOanalysisDbase)supportedbya'DecisionSupportSystem' (DSS).SuchDSScouldthenhandlealsoanyexceptionstothegeneralapplicabilityofsomemethods, by applying "fuzzy logics" analysis. An optimal path for GMO analysis (in terms of performance, applicability, cost, etc) can be determined taking into account the experience with food and feed productsandthetraceabilityinformation. 42 CoExtraInternationalConference  However,noranaccepted"GMOanalysisDbase"oranoperational"DSS"arecurrentlyavailableto the stakeholders, leaving the application of the "modular approach" still open for discussion. Establishing a fully operational DSS, that is constantly updated, is thus a major goal for the near future,ifthe"modularapproach"istobesuccessfullyappliedbythestakeholders.Thedevelopment ofaDSSfordetectingGMOinthebroadrangeofapplications,shouldgreatlybenefitfromthevast experienceobtainedwithintheCoExtraproject.IntheEC,theENGLandtheECNRLGMOnetworks couldplayakeyroleherein.  References: 1. HolstJensenA,BerdalKG.(2004):Themodularanalyticalprocedureandvalidation approachandtheunitsofmeasurementforgeneticallymodifiedmaterialsinfoodsand feeds.JournalofAOACInternational,87(4):92736. 43 CoExtraInternationalConference B1.4. Validationofnovelmethodsandtechnologies M. Mazzara1, G. Bellocchi1, C. Savini1, N. Foti1, R. Onori2, M. de Giacomo2, M. Van den Bulcke3, A. Lievens3,S.Hamels4,S.Leimanis4andG.VandenEede1 1 JointResearchCentre,InstituteforHealthandConsumerProtection,EuropeanCommission.ViaFermi2749,21027Ispra (VA),Italy.Contactauthor:[email protected]. 2 ItalianNationalInstituteofHealth,DepartmentofVeterinaryPublicHealthandFoodSafety,GMOandMycotoxinsUnit, VialeReginaElena299,00161Rome,Italy. 3 ScientificInstituteofPublicHealth,DivisionofBiosafetyandBiotechnology,J.Wytsmanstreet14,1050Brussels,Belgium). 4 EppendorfArrayTechnologies,20Rueduséminaire,5000Namur,Belgium  Thereisacontinuousandincreasingneedforreliableanalyticalmethodstoassesscompliancewith national and international requirements in all areas of analysis. The reliability of a method is determinedbythesocalledvalidation,whichistheprocedureprovidingevidenceofsuitabilityofan analyticalmethodforitsintendedpurpose.Basedontheresultsofavalidationstudy,amethodis considered or not as fit for the intended purposes. In most cases, formal validation requires the assessmentoftheperformanceoftheproposedmethodbymeansofaninterlaboratorystudy,also knownascollaborativestudyorringtrial. While well established validation key parameters and data analysis procedures are available for singletarget qualitative and quantitative methods (see for example ISO 5725 and http://gmo crl.jrc.ec.europa.eu/doc/Min_Perf_Requir_Analyt_methods_131008.pdf), the validation of novel methods and technologies developed by the CoExtra project required the development of novel approaches to summarise the information provided by individual validation indices and tests statisticsintocomprehensiveindicatorsofmethodperformance. With the aim of providing easy access to statistical and numerical tools for analytical method validation, the freely available Analytical Method Performance Evaluation (AMPE) software was created. Through AMPE, a variety of validation metrics (indices and test statistics) is provided for comparing measurementsfromalaboratoryanalysisandreferencevaluesfromstandardsamples. Provisions are also provided for analyses based on blank samples. In its innovative part, AMPE suppliesprovisionsforfuzzybasedaggregationofvalidationmetrics. Through the application of fuzzy logic, aggregated indicators are proposed as suitable tools for overall evaluation of analytical methods, allowing also objective comparison across different methods. Fuzzylogic based indicators were developed that allow summarising the information obtained by independent validation statistics into one synthetic index of overall method performance. The possibility of having a comprehensive indicator of method performance has the advantage of permitting direct method comparison, facilitating the evaluation of many individual, possiblycontradictorymetrics.Initsoriginaldevelopment,thefuzzybasedexpertsystemwasused to validate novel methods developed by the project, the DualChip® GMO microarray and the so calledpJANUSplasmids,andtotestthe“modularapproach”tomethodvalidation. The DualChip® GMO is a novel multiplex screening method for the detection and identification of GMO, based on the use of multiplex PCR followed by microarray. The technology is based on the “Matrix Approach” i.e. on the identification of quite ubiquitous GMO genetic target elements first amplifiedbyPCR,followedbydirecthybridisationoftheampliconsonapredefinedmicroarray.The validation was performed within the framework of CoExtra, in collaboration with twelve laboratories. The method was evaluated with predefined performance criteria with respect to the JRCIHCP CRLGMFF method acceptance criteria. Data were processed according to ISO 5725 standard and the overall method performance met the acceptance criteria. However, creating reproducibledatawithahighlevelofconsistencyacrossarrayexperimentsandvariousplatformsis 44 CoExtraInternationalConference widelyacceptedbythescientificcommunityasamajorissue.Thecomplexnatureofamicroarray experiment results in many potential sources of variability, which can affect performance. In responsetothischallenge,thefuzzylogic basedapproachwassuccessfullyappliedto theanalysis anddatainterpretationofthechipvalidationexercise. Asecondmethodwasdeveloped,basedonrealtimePCRandtheuseofnovelcalibrationmolecules (plasmidpJANUS)incomparisontogenomicDNAcalibrant(theclassicalapproach).Themethodwas validated among various laboratories and theevaluation of the interlaboratory study performed indicatedthatthedualtargetplasmidpJANUSTM02001canbeusedascalibrantindeterminingthe amountofRoundupReady®soybeanlineGTS4032.TheequivalenceofplasmidandgenomicDNA templates as calibrants for the quantification of the GM soybean event GTS4032 was demonstrated:thisequivalencewasassessedthroughconventionalstatisticalanalysis(ANOVA)and with the application of the fuzzylogic based approach, that resulted remarkably in line with the expert interpretation of statistical results; this can be seen an a further validation of the suitable applicationofthenovelfuzzylogicanalysisinGMOdetectionmethodsevaluation. Astudywasperformedtoprovideaproofofconceptfortheapplicationofthemodularapproachto analyticalmethodsinthefieldofGMOtestingbasedontheuseoftheRealtimePCR.Forgeneral informationonthemodularapproachrefertoM.VandenBulckepresentation,“ModularApproach Implemented:Pros,ConsandFuturePerspectives”. Modularity implies independency and thus flexibility of combining modules on the one hand, and uniformity and harmonisation on the other hand. If modular validation is to be applied, fit for purposeproceduresandgeneralacceptanceofminimumrequirementsforeachmoduleareneeded in order to evaluate the uncertainties associated with each module. In order to provide scientific evidence on the applicability of this approach, the experiments carried out focused on the interactions between DNA extraction methods and PCR analysis. It was found that for the correct application of the modular approach appropriate performance criteria should be met by DNA extracts(assessmentofqualitycharacteristicsofDNA)sothattheycanbefitforthepurposeofthe followinganalyticalmodule,independentlyfromtheprecedingmatrixDNAextractioncombination. WiththeexceptionofoneDNAextraction/matrixcombination,thestudyprovidedgoodevidenceof independencyoftheanalyticalmodulestested,suggestingthatamodularapproachcanbecorrectly employedinmethodvalidationandanalyticalcontrol. 45 CoExtraInternationalConference B1.5. ReferencematerialsandreferencePCRassaysforGMOquantification Taverniers1I,VandenBulcke2M,Trapmann3S,Brodmann4P,Gruden5K,HolstJensen6A,Roth7L, Allnutt8T,Freyer9R,Pla10M,Zhang11D,Kuznetzov12B,Bertheau13Y. 1 InstituteforAgriculturalandFisheriesResearch(EVILVO),TechnologyandFoodSciencesUnit,Burg.VanGansberghelaan 115,B9820Merelbeke,Belgium 2 ScientificInstituteofPublicHealth(IPH),J.Wytsmanstraat14,B1050,Brussels,Belgium 3 ECJRC,InstituteforReferenceMaterialsandMeasurements(IRMM),Retieseweg14,111,2440Geel,Belgium 4 BiolytixAG,Witterswil,Switzerland 5 NationalInstituteofBiology(NIB),Ljubljana,Slovenia 6 NationalVeterinaryInstitute(NVI),Oslo,Norway 7 BundesinstitutfürRisikobewertung(BfR),Berlin,Germany 8 FERA,UK 9 EurofinsGeneScan,Engesserstr.4,D79108Freiburg,Germany 10 CSICIRTA,Girona,Spain 11 GIPGEVES,France 12 BioEng,Russia 13 INRA,routedeSaintCyr,78026Versaillescedex,France  The EU legislation regulating the approval and environmental release of any new GMO includes several core elements: (1) a safety assessment carried out by the European Food Safety Authority (EFSA), (2) availability of reference materials and eventspecific PCR methods for detection, (3) mandatorylabellingofGMOderivedorcontainingfood/feedproductsaboveathresholdof0,9%, and(4)postmarketmonitoringandtraceability(EC2001,EC2003a,EC2003b,EC2004a,EC2004b, EC2004c). An applicant submitting a dossier for authorization of a new GM event, must provide an event specificPCRmethod,whichisthenevaluatedandvalidatedbytheCommunityReferenceLaboratory for GMOs (CRLGMFF). Upon validation and acceptance, the protocol describing the method is published at the CRL website and considered as a reference method for the member states’ laboratories in charge of GMO testing for compliance with the EU legislation. Commission recommendation 787/2004 provides technical guidelines on sampling and detection of GMOs and materialscontainingGMOsinthecontextofregulation1830/2003/EC.TheGMOimpuritiesmustbe determined on a single ingredient level, i.e. as a percentage GMDNA, defined as the relative proportion of the GMDNA sequence copy number in relation to the plant taxon specific DNA sequencecopynumber.‘Copynumber’referstothenumberof‘haploidgenomeequivalents’(HGE), theEUrecommendedunitforexpressingGMOcontents(EC2004c). Implementation of the legislation on food and feed labelling and traceability brings up two main requirements: (1) the availability of control materials and calibrants, and (2) the availability of specific and accurate analytical methods, both being crucial aspects for reliable detection and quantificationofGMOsinawiderangeofagriculturalproducts.Thispresentationisasummaryon bothtopics. First,anoverviewispresentedonthedifferenttypesof(certified)referencematerials(RMs,CRMs) and calibrants for GMOs and their specific objectives. Two main types of control materials exist: matrixmaterials,andcalibrantmaterials.WhileearlymaterialsforGMOswerepowders,obtained from raw materials and certified for their mass ratio of GM powder relative to the total species powder, now matrix CRMs and calibrant CRMs exist with certified copy number ratios (haploid 46 CoExtraInternationalConference genomeequivalents,HGEpercentage).Anoverviewisgivenonthesematerialsandwhereandhow tousethemintheanalyticalprocedureforGMOdetectionandquantification.Mainoutcomesfrom the experimental work done within CoExtra are presented. An interlaboratory comparative study wasperformedforevaluationoftheapplicabilityofdifferenttypesofDNAcalibrants:plasmidDNA (pDNA), extracted genomic DNA (gDNA), multiple strand displacement DNA (mdaDNA). The study also aimed at evaluating the best conditioning of DNA for longterm preservation and stability of DNA over long periods of time. Both gDNA and pDNA proved to be suitable calibrants under the particular conditions of this study. While mdaDNA can only be used for qualitative measurements but not as calibrants for quantification, it was considered as an interesting alternative to produce positive or negative samples when control material is only available in limited amounts. This Co Extra study contributed to the further development of calibrants suitable for the expression of measurement results in copy number ratios. It is a milestone towards the improved availability of CRMsforcalibrationpurposes. Second,theimportanceofreferencegenesystemsofreferenceassays(RAs)forGMOquantification isextensivelydiscussed.ControllingtheadventitiouspresenceofGMOsinvolvesdetectionofasmall proportion of GMO (typically 05%) in a background (95100%) of nonGMO species. For this purpose, an eventspecific PCR method is combined with a species or target taxonspecific PCR assay. The method, submitted by the applicant for international validation and acceptance by the CRLGMFF, has to fulfill minimum performance requirements which are defined by the European Network of GMO Laboratories (ENGL) and published at the CRL webpage (http://gmo crl.jrc.ec.europa.eu).Thenewmethodrequirements,whichwillbeappliedafter13thofApril2009, defineseveralguidelinesregardingthetargettaxonspecificsequence:(1)Theabsenceofallelicand copynumbervariationhastobedemonstratedacrossagloballyrepresentativeanddiversespecies’ varietieslist;(2)Allelicand/orcopynumbervariationshallbereportedbytheapplicantifknown;(3) Thespecificityofthetargetsequenceshallbeinsilicovalidatedagainstpubliclyavailablesequences databasesandexperimentallydemonstratedbyabsenceofamplificationproducts,whenthetarget sequence specific assay is applied to individual PCRs on pure genomic DNA of a representative sampleoftheclosestrelativestothetargettaxa,aswellasofthemostimportantfoodcrops(ENGL 2008). Briefly,areliablereferencePCRassaymustbespecificforthetaxoninquestioni.e.shouldnotgive anysignalwithothercloselyrelatedtaxaontheonehand,andshouldgiveauniformpositivesignal amongdifferentvarietieswithinthetaxonontheotherhand.Lowspecificityanduniformitycould lead to under or overestimations of the GM content and thus render the methods unfit for their purpose. Within CoExtra, existing reference assays for the main GM crops have been extensively evaluated for their copy number, sequence stability, specificity and uniformity and new improved systemshavebeendeveloped.Severalissueswillbepresentedhereconcerningtheformulationofa corecollectionofspeciesandvarieties,tobetestedwhendevelopingandvalidatinganewreference assay. How should the borders of a GM ‘target taxon’ be defined? What are the criteria for the selectionofspeciesandvarietiestobeincludedinsuchacorecollection?Whatarethefactorstobe consideredinspecificityanduniformitytestingofanyreferencetargettaxonspecificassay? Thepurposeofestablishingcorecollectionsforspecificity/uniformitytestingistoensurethatthese parametersareincludedinthevalidationofnewlydesignedreferenceassaysandtoassurethatthe assaywillamplifyefficientlyinalltypesofplantmaterialssubjectedtoGMOtesting.Ideally,acore collectionofspeciesandvarietieshastobe(1)representative–tocoverrelevantplantspeciesor other botanical taxa and the existing variation within this/these taxon/taxa; (2) dynamic – to be updatedpermanentlywithnewvarieties/lines;(3)accessibletoallpartiesinvolvedinGMOtesting: biotech companies developing PCR detection systems and enforcement labs involved in official controlofGMOs. AreferencePCRassayshouldonlybeacceptedprovidedthatitistargettaxonspecificanduniformly amplifies within the whole (market) gene pool. A decision support system is presented for the 47 CoExtraInternationalConference selection procedure of botanical taxa and varieties/lines, based on specificity and uniformity. Specificity here concerns the risk for cooccurrence of the nontarget species/variety under considerationwiththetargettaxon,andthecorrespondingriskofcoamplification.Uniformityhere concernstherepresentativityofthespecies/varietyconsideredrelativetothemarketedvarietiesof thespecies,andwhatthemarketshareisofthespecies/varietyconsidered.Thisisaprocesswhich hastotakeintoaccounttaxonomic,phylogenetic,breedingandagronomicdata.Inthisprocessof selection, reference is made to existing collections of varieties and species, such as the CPVO (CommunityPlantVarietyOffice)listofEUregisteredplantvarieties. 48 CoExtraInternationalConference SessionB2:DetectionofGMingredientsinfoodsandFeeds  B2.1. NewrealtimePCRmethodsavailableforroutineGMOdetectionlabs applicabilityandperformance Theo R. Allnutt2, Stefan Baeumler1, Christoph Bahrdt1, Gilbert Berben3, Yves Bertheau7, Mark van den Bulcke4, Adrian Härri5, Boris Kuznetsov6, Valérie Laval7, Ralf Seyfarth5, Marina Sukhacheva6, DoerteWulff1 1 EurofinsGeneScan,Germany 2 FoodandEnvironmentResearchAgency,UK; 3 CRAW,Belgium;4ScientificInstituteofPublicHealth,Belgium; 5 Biolytix,Switzerland; 6 BioEng,Russia; 7 INRA,routedeSaintCyr,78026Versaillescedex,France  Reliable and costeffective methods for GMO detection are essential for establishing an efficient system for traceability as well as for monitoring different aspects of GMO coexistence with conventionalcrops.AfterseveralyearsofprimarilygelbasedPCRanalyses,realtimePCRmethods have become stateoftheart for qualitative detection as well as for quantification of genetically modified components in food and feed. Within the framework of CoExtra, several new realtime PCRbased methods have been developed in order to enhance efficiency and costeffectiveness of GMO analysis, further improve reliability of GMO analysis, overcome certain limitations of current realtime PCR methods and finally complement the existing method portfolio with methods for identification and quantification of nonEU approved GMOs formerly not detectable or not quantifiable. Enhancing efficiency and costeffectiveness is of major importance as GMO analysis is getting increasinglycomplexduetothegrowingnumberofapprovedandcommercializedGMOs.Whereas inthebeginningofGMOanalysisascreeningfor35Spromotersequenceswascommonpractice,the situationhasbecomemuchmoredifficult.ThisisalsoduetothefactthatmoreandmoreGMOsare lackingthetypicalscreeningelements.Consequentlythereisanurgentneedformultiplexscreening andidentificationmethodsinordertoavoidanincreaseincostfortraceabilitytoaneconomically unbearable level. Within CoExtra several multiplex realtime PCR assays ranging from duplex to pentaplex format have been developed and thoroughly validated providing improved tools for screeningfortraditionalscreeningelementslike35Spromoterandnosterminatoraswellastools targeting additional screening targets. Some of the newly developed assays will be presented. Furthermoremultiplexspecificrequirementsformethodvalidationwillbeaddressed. AsecondobjectiveofPCRmethoddevelopmentwithinCoExtrawasfurtherimprovingreliabilityof GMOanalysis.InthiscontextrealtimePCRcontrolassayshavebeendevelopeddetectingimportant donororganismsofbuildingblocksfrequentlyusedinGMplants.Incaseofpositivetestingresults for screening targets originally derived from Agrobacterium, Bacillus and figwort mosaic virus the newly developed control assays detecting these donororganisms can be used in order to confirm thatthepositivescreeningresultsaretrueindicatorsofGMOsandarenotduetothepresenceof bacterial(Agrobacterium,Bacillus)orviralDNA(FMV)inthefoodorfeedsample.Anothermeansof enhancing reliability of GMO analysis was the development and validation of an improved IPC (InternalPositiveControl)whichcanbeusedforcostefficientandsensitiveverificationofabsence ofPCRinhibition.ExamplesofassaysincludingthisIPCwillbepresented. 49 CoExtraInternationalConference Another objective of the developmental work within CoExtra was to overcome limitations of currentrealtimePCRmethodssuchastoenableGMOquantificationinsampleswithverylowDNA contentortomakeGMOanalysisportableforonsitetesting.Asanexamplethemethodforonsite realtimePCRquantificationofGT73Brassicanapuswillbepresented. Finally realtime PCR assays detecting two transgenic potato events which are certified for food consumption in Russia will be highlighted which complement the portfolio of eventspecific detectionmethodsavailableforEUapprovedGMOs. 50 CoExtraInternationalConference B2.2. ReliabilityandcostsofGMOdetection Theo R. Allnutt4, Mira Ayad3i, Stefan Baeumler10, Gilbert Berben8, Knut Berdal2, Charlotte Boydler Andersen2, Yves Bertheau3, Peter Brodmann6, Meti Buh Gašpari1, Katarina Cankar1, Frédéric Debode8, Teresa Esteve5, Arne HolstJensen2, Eric Janssen8, André Kobilinsky3, Jose La Paz5, David Lee7, Valérie Laval3, Maria Pla5, Torstein Tengs2, Doerte Wulf10, David Zhang9, Jana Žel1, Kristina Gruden1 1 NationalInstituteofBiology,Slovenia 2 NationalVeterinaryInstitute,Norway 3 INRA,France 4 FoodandEnvironmentResearchAgency,UK; 5 CSIC,Spain; 6 Biolytix,Switzerland; 7 NIAB,UK; 8 CRAW,Belgium; 9 GIPGEVES,France;  10 GeneScan,Germany  The analytical procedure used for GM detection and quantification at the laboratory level is composedofdifferentmodules.Eachofthosemodulescanimpacttheaccuracyofanalyticalresult qualitatively and/or quantitatively. Within the CoExtra project we have investigated different aspects of the reliability of GMO detection. A system for quantification of GM presence in the samples with low DNA content was established, for example. Additionally, effort was put into improvementsrelatedtotradeoffbetweenreliabilityandcostofanalysis. DifferentDNAextractionsfromhighlyrefinedmaterials(lecithinandoil)havebeeninvestigatedin detail. The sample preparation steps as well as the DNA extraction and purification steps were optimized. Both types of samples benefited from a hexane extraction step followed by DNA extractionandpurification.ThemethodshavebeenoptimizedformaximumDNArecoveryandlow handson time. Standard operating procedures are available for analytical labs that would like to evaluate the procedure or for further validation studies. The procedure for DNA extraction from lecithin has already been inhouse validated and the protocol successfully transferred to different labs.Toimprovequantificationperformance,theSIMQUANTapproachwasdeveloped.Theideaisto perform a series of PCR reactions and quantify the target numbers in the sample using the distributionofpositive/negativeresultsandmostprobablenumberstatistics.Oneoftheadvantages ofSIMQUANTisalsolesssensitivityofqualitativePCRtoinhibitorsinreactionswhencomparedto quantitativePCR.TheSIMQUANTwasadditionallyupgradedtomultiplexaversion,thusincreasing itsapplicability. ThequalityofextractedDNAisknowntoinfluencesignificantlythefinalresultofGMdetectionand quantification. One approach to control this step is “matrix by matrix” validation of the DNA extractionmethod.Forvalidation,thequalityofDNAsolutionsshouldbecontrolledbytestingthe presence or not of statistically significant inhibitory effects. This is usually done by adding an exogenousDNA(othertaxongenomicDNAthanthattestedorplasmid,providedtheyareinhibitor free)containingaspecificPCRtargetintotheDNAsolutionsunderstudyataconcentrationcloseto thelimitofquantificationorlimitofsensitivity(forexample,50100copies)andthenbyamplifying thespecificPCRtargetcontainedbytheexogenousDNA.Themostconvenientandcheapestwayis tousetheDNAofothertaxonasexogenousDNA.Theprobleminmostroutinedetectionlabsisthat the matrix is not well defined. Composition of feed and food samples can vary from supplier to 51 CoExtraInternationalConference supplier and even from batch to batch, making ‘matrix by matrix’ DNA extraction validation not feasible.Modularvalidationcanthenbeperformed,providingappropriatecontrolsofPCRinhibition areappliedwitheverysample. In quantitative analysis, the target number quantification also introduces a bias. Two calculation methods can be used. The Ct method relies on both amplicons having similar efficiencies of amplificationforaccuratequantification.Thereforethebiasofthemethodwouldonlybeacceptable if working with well established matrixes (e.g. with raw materials) if properly validated in combinationwithDNAextractionmethodorifthecalibrationstandardsareofthesamematrixtype. Routine detection of genetically modified (GM) organisms is most often performed on Applied Biosystemsmachines(ABI7700andABI7900),usingtheirprominentchemistry–TaqMan®andtheir Mastermix. With new developments in this area many different apparatuses and chemistries are availableonthemarketthatcouldpotentiallyoutperformtheprevioussystems.Withintheproject thedifferentapparatusesandperformanceofalternativechemistrieswasevaluated,thusextending options in the methodology used in routine practice to recent technological advances. For comparison of apparatuses a small ring trial was organized within which 8 different realtime PCR modelswereincludedwhichwereavailableinthelabsofdifferentWP5 partners.Some machines werealsoreplicatedindifferentlabstoevaluateinterlabvariability.ApplyingCRLvalidationmethod acceptancecriteria(25%RSDr,50%Bias)resultssuggestthatthetypeofmachineusedisnotcritical inGMquantification,atleastforthemethodsexaminedhere. Similarly,thecomparisonofdifferentavailablechemistrieswasorganizedtotestthosemostwidely used(MGB®,SYBR®GreenandMolecularBeacons)indifferentlaboratoriesandtargetingdifferent genes, while the more recently introduced were tested less extensively (Plexor, LNA, lux). The conclusionwasthatTaqMan®,MGB®,LNAareequalinperformancecharacteristicsandtheycanbe used whenever they are better suited for the particular application, e.g. if there is special needs regarding specificity or target regions are problematic for design of longer TaqMan® probes. Molecular Beacons systems were more difficult to design to achieve a robust assay. SYBR® Green chemistry performed well, its drawback being slightly lower sensitivity when compared to probe basedassays.TheotherprimeronlybasedsystemthatperformedwellwereAmpliflourandPlexor, while some of the more exotic systems performed significantly below specifications given by the manufacturer. 52 CoExtraInternationalConference B2.3. NonPCRbasedAlternativeAnalyticalMethods JimMurray LumoraLtd,UK  Althoughpolymerasechainreaction(PCR)hastodatebeentheoverwhelmingmethodofchoicefor thelaboratorybaseddetectionofGMOsbecauseofitssensitivity,familiarityofmethodology,well developed standard operating procedures and availability of suitable equipment in testing laboratories, PCR nevertheless suffers from a number of distinct disadvantages. These include the relativelyhighcostofequipmentandoftheassaysthemselves,potentialforcontaminationandthe sensitivity to certain classes of contaminants and inhibitors, leading to a requirement for reliable DNApurificationstrategies.Asaresult,assaysneedtobecarriedoutinalaboratoryandtheneedto accumulate batches of samples further slows the total time required for the assay chain from sampling to the eventual result in the laboratory. These issues, together with the difficulty in designingcosteffectiveportabledevicesforPCR,havedriventhesearchforalternativestoPCR,a number of which are now becoming available that seek to overcome some of the perceived limitationsofthePCRapproach. WithinWP5oftheCoExtraproject,anumberofnewalternativestoPCRbasedmethodshavebeen evaluated, which offer potential advantages over PCR for speed, cost, scale or portability. The purpose of this presentation will be to review these methods and report on their suitability and potential for GMO screening applications. For more details on nonPCR nucleic acid amplification methods,thepublicationbypartnerNIBmaybeconsulted(1),andforareviewonnovelanalytical approachestoGMOtestingsee(2).Themethodsreportedcomprisetwomainclasses.Thefirstare truealternativestoPCR,i.e.molecularteststhat,likePCR,alsotestforthepresenceofspecificDNA sequences, but which employ an alternative nonPCR method of nucleic acid amplification. These includetheuseofstranddisplacingpolymerasesataconstanttemperature(e.g.LAMPandRDC),or the use of transcriptionmediated amplification (e.g. NASBA). All these methods do not require temperature cycling, operate at a constant temperature, and offer potential advantages including cost, speed, portability and reduced sensitivity to inhibitorsoverPCR. Thetechnicaladvantagesof theseapproachesincludethepossibilitytocombinetheirusewithnovelreportersystems,andthe useofanewbioluminescentoutputknownasBARThasbeenevaluatedinconjunctionwithLAMP and RDC. The second type of method does not seek to detect DNA sequences but employs spectroscopic techniques to distinguish GM and nonGMO material. Method evaluation for their suitabilityforGMOdetectionwascarriedoutbythepartnersNIB,INRA,CSL,Lumora,NIABandCRA WwithinWP5oftheEUCoExtraProject. In the first class, techniques known as NASBA, LAMP and RDC were evaluated, either alone or in conjunction with the BART bioluminescent reporter system. In the second, nearinfrared spectroscopywasevaluated.Themaincharacteristicsofthesetechniquesissummarised. Loopmediated Isothermal Amplification (LAMP), developed by the Eiken Chemical Company is a simple,rapid,specificandcosteffectivenucleicacidamplificationtechnology.Detailsaredescribed on http://loopamp.eiken.co.jp/e/lamp/index.html. It is characterized by the use of 4 different primers, specifically designed to recognize 6 distinct regions on the target DNA template, in a process that proceeds at a constant temperature driven by a strand displacement reaction. Amplification and detection of target genes can be completed in a single step, by incubating the mixture of DNA template, primers and a strand displacement DNA polymerase, at a constant temperature.Itprovideshighamplificationefficiency,withreplicationoftheoriginaltemplatecopy, occurring1091010timesduringa1560minreaction.Theprimerpairsusedinthisamplificationcan bedesignedusingawebtoolathttp://primerexplorer.jp/e/. 53 CoExtraInternationalConference RDC (Reaction déplacement chimeric) is an isothermal DNA amplification procedure developed by Biomerieux, and is based on the use of chimeric primers consisting of an RNA stretch embedded within flanking DNA sequences. Cleavage of the hybrid duplex between the RNA region formed when the primer is hybridized to its DNA target provides the initiation for a stranddisplacing polymerase. For details see US Patent 5824517 (Cleuziat and Mandrand; (http://www.patentstorm.us/patents/5824517.html). BothRDCandLAMPareamongisothermalamplificationtechnologiesthatcanbeinterfacedwitha unique reporter system known as BART (bioluminescent assay for realtime). BART is a bioluminescencerealtimeassaydevelopedbyLumora[http://www.lumora.co.uk/]thatallowsthe quantitative analysis of DNA amplification in real time. In BART, PPi produced during DNA amplification is converted to ATP by the action of ATP sulphurylase. This ATP is then used in a coupled simultaneous reaction by thermsotable firefly luciferase and luciferin to produce a light outputpermittingrealtimeanalysisofamplificationkinetics.AuniquefeatureofBARTisaninitial burstoflight,associatedwiththeonsetofexponentialamplification,followedbyarapiddecrease, as pyrophosphate reaches a critical threshold. The time to reach this light peak is therefore a function of the amount of target DNA in the sample at the beginning of the reaction (time to maximum;Tmax),andauniquefeatureoftheBARTreporter.QuantificationofBARTisbasedontime to peak and not absolute light intensity, making it less prone to inhibition simplifying data interpretation and the hardware requirements. LAMP in conjunction with BART provides a robust, sensitiveandreliablemethodforqualitativedetectionofGMOsatlowlevelsofpresence(0.1%)and has the potential for quantitative or semiquantitative manifestations. It is also suitable and demonstratedinsmall,lowcostdevicesthatcanbeusedinthefieldorotherlowresourcesettings, bothbecauseoftheequipmentrequirementsanditsabilitytofunctionwithverysimpleandrapid DNApreparations,evenfromfreshleaftissue. NASBAisanisothermalnucleicacidamplificationmethodthatmimicsretroviralreplicationandwas originallyappliedtodetectionandquantificationofRNAtargets,buthasalsobeenadaptedforDNA detection, and it was evaluated in this manifestation. Amplification occurs because the target is transcribed into RNA, which is then reversetranscribed back into DNA, thereby providing more template copies for RNA transcription. The transcription is carried out by T7 RNA polymerase and requires the incorporation of the appropriate promoter sequence onto the template, which is achievedbyappropriateprimerdesign.ThismethodwasmodifiedtoallowDNAamplificationusing atwostepprocedure:firststepwithtailedprimers,secondstepwithuniversalprimers.NASBAwas developed well with performance characteristics similar to PCR, and adaptation to realtime detectionusingMolecularBeaconshasbeenreported.HoweverintegrationwiththeBARTsystemis notstraightforwardduetothehighconcentrationofATPpresentintheNASBAreaction. Near Infrared spectroscopy uses spectral properties of sample in IR to detect GMO’s. The method was developed for Roundup Ready soybean (GTS4032) due to its specific characteristics. The method is noninvasive and can be applied onsite, therefore suitable for analysis of large sample lotsofmoreexpensivematerial,e.g.seeds.  References: 1. Morisset,D.,etal.(2008).EurFoodResTechnol227:12871297. 2. Michelini,E.,etal.(2008)AnalBioanalChem392:355–367. 54 CoExtraInternationalConference B2.4. DetectingunauthorisedandunknownGMOs. ArneHolstJensen1,KnutG.Berdal1,YvesBertheau2,MarkoBohanec8,JonBohlin1,Maher Chaouachi2,KristinaGruden3,SandrineHamels6,AnjaKrech5,EstherKok4,AnjaB.Kristoffersen1, ValerieLaval2,SergeLeimanis6,MarieLøvoll1,DanyMorriset3,AnneNemeth5,NinaPapazova7,Theo Prins4,PetraRichl5,JoseRemacle6,TomRuttink7,IsabelTaverniers7,TorsteinTengs1,Jeroenvan Dijk4,DoerteWulff5,HaiboZhang1,JanaZel3,MartinZnidarsic8 1 NationalVeterinaryInstitute,Norway 2 INRA,France 3 NationalInstituteofBiology,Slovenia 4 RIKILT,Netherlands 5 EurofinsGeneScan,Germany 6 EAT,Belgium 7 ILVO,Belgium 8 JožefStefanInstitute,Slovenia  The global trade and spread of technological competence and capacity to develop genetically modified (GM) organisms (GMOs) such as plants, in combination with the cultural and regional differences in terms of suitability, need and acceptance of GMOs is a potential cause of disputes. WithintheEuropeanUnion(EU)aswellasseveralotherjurisdictions,noimport,useorreleaseof GMOderivedmaterialislegalwithoutpriorauthorisation.Amongtherequirementsthatneedtobe met prior to authorisation of a GMO in the EU is the availability of a validated and specific, quantitativedetectionmethodandcorrespondingreferencematerialfortheGMOinquestion. FieldtrialsareapartoftheperformanceassessmentofGMOs.Thesefieldtrialspotentiallyleadto low level contamination of neighbouring fields. Birds or rodents may spread grains or seeds and incomplete sanitation or other human error may lead to unintended spread of viable material. Finally, intended distribution into the environment or food/feed chain can not be complete ruled out. Validated detection methods as well as reference materials are usually not made available in thesecases. Most of the GMOs commercialised in the world at present are herbicide resistant and/or insect tolerant. The trait genes inserted into these GMOs are usually well known and belong to a few groups: pat/bar, epsps and various crygenes. Genetic elements associated with the genes to facilitateandregulatetheirexpressionarealsowithfewexceptionswellknownandbelongtoafew groups,e.g. thecauliflowermosaicvirus(CaMV)35Spromoterandterminator(P35Sand T35S)or the Agrobacterium tumefaciens nos terminator (3’nos). The availability of other trait genes and regulatory elements is, however, increasing rapidly. Commercial or other interests may prevent relevant information from being disseminated to stakeholders such as competent authorities and laboratoriesperformingGMOdetection. Thereisnosinglegeneticmarkerthatcanbetracedasa“GMOlabel”.Instead,itisnecessarytouse methodsspecifictoparticulargeneticmarkerssuchasregulatoryelementsortraitgenes(screening methods),fusionmotifsbetweenregulatoryelementsandtraitgenes(constructspecificmethods) orfusionmotifsbetweeninsertedDNAandtherecipientDNA(eventspecificmarkers).Thenumber ofGMOsthatmaybedetectedwiththemethodsdependonthetargetedgeneticmarker,andthe analyst may need to balance broadspectered screening against ability to specifically identify the GMOs that may be detected. Detection of unauthorised GMOs may often be achievable by using screening methods. However, with these methods it may be quite difficult to determine that the detectedGMOmaterialiscomingfromanunauthorisedGMO. Ontheotherhand,theabsenceof specificdetectionmethodsandreferencematerialsformostunauthorisedGMOsleavesfewoptions 55 CoExtraInternationalConference for identification. GMOs that can not be detected with the commonly used screening methods becausetheintroducedDNAsequencesorgeneticelementsareunusual(novel)maybeclassifiedas unknown GMOs. These are of course particularly difficult to detect and identify. Presence of non declaredingredients(e.g.“botanicalimpurity”)mayfurthercomplicatetheanalyticalwork. AnypresenceofunauthorisedorunknownGMOorderivedmaterialinthefood/feedsupplychainin theEUisbydefinitionillegal,andmayposearisktosocietyandeconomy,theenvironmentand/or human and animal health. Socioeconomic risk is exemplified by the restrictions on import of rice fromtheUSAtotheEUasaconsequenceofcontaminationofAmericanricewiththeLibertyLink 601ricein2006.Riskto theenvironmentisexemplifiedby potentialspread ofthetraitgeneto a wildrelativeoftheGMO.Introducingforexampleanherbicideresistanceorinsecttolerancegeneto awildplantspeciesmayimprovethefitnessofthewildplantrelativetocompetingplantspeciesin itsenvironment,oritmayaffectthediversityand/orabundanceofinsectsthatbirdsdependupon forfeedingtheirbrood.Risktohumanand/oranimalhealthisexemplifiedbythepossibilitythata food/feedplantusedasabiogeneratorforpharmaceuticalsisintroducedintothefood/feedchain. Abilitytodetect,identifyandcharacteriseunauthorisedorunknownGMOsisthereforenecessaryto beabletodefine,delimit,preventandremoveproblems. Traceabilityfacilitatestheidentificationoftheoriginofmaterial,andglobalinformationnetworks, databases,etc.mayprovideinformationaboutdevelopmentsofnewGMOs,novelgeneticelements that are potentially exploitable and authorisations outside the stakeholder’s own jurisdiction. This type of information can be used by stakeholders to improve their ability to detect, identify and characteriseunauthorisedorunknownGMOs,aswellastoprioritisedevelopmentsandapplications ofparticularanalyticalmethods. Developmentofanalyticalmethodsandstrategiesfordetection,identificationandcharacterisation of unauthorised and unknown GMOs has been a major priority within the CoExtra project. In parallelamodulardecisionsupportsystem(DSS)hasbeendevelopedinwhichtraceabilityandother information can also be taken into consideration. These developments together are expected to significantlyreducethechallengesposedbyunauthorisedandunknownGMOs. Thispresentationwillgiveanoverviewofthestateofthearttechnologiesanddevelopmentsfrom the CoExtra project relevant to the detection, identification and characterisation of unauthorised andunknownGMOs,andwillalsopointoutsomeoftheremainingandpossiblefuturechallengesof relevance. 56 CoExtraInternationalConference B2.5. NewmultiplexingtoolsforreliableanalysisofGMOs MariaPla CSIC,Barcelona,Spain  To enforce the existing regulations on commercialisation and environmental release of genetically modified organisms (GMO), adequate tools for its detection, identification and quantification are required. The most accepted GMO detection methods are based on specific DNA sequence detection by means of polymerase chain reaction (PCR) techniques, able to detect even small amountsoftransgenesequencesinrawmaterialsandprocessedfoods.PCRassayscanbeusedfor screening purposes (e.g. targeting transgenic elements commonly used in GMOs), to detect junctionsofcontiguoustransgenicelements,andtoidentifyaGMOevent(bytargetingthejunction regions between the insert and recipient plant genomic DNA or eventspecific rearrangements). Additionalamplificationofaplantspeciesspecificgeneisnecessaryascontrol.Anumberofthese methods(includingquantitativeassays)areavailablethathavebeenvalidatedbyofficialbodiesor referencelaboratoriese.g.theEUJointResearchLaboratory. ThepresenceofGMOmaterialonthemarketisincreasing;andsoisthenumberofGMOsapproved worldwide (including stacked events) and in the pipeline. At the same time, the genetic elements introducedintonewGMOsandthehostplantspeciesarebecomingmorediverse.Thisincreasesthe cost and working power required for GMO analysis. In this context, the widely used singletarget detectionmethodsarenotconsideredsufficienttofulfilthecurrentandenvisagedneedforanalysis. Consequently it is necessary to introduce new analytical technologies for reliable, low cost, high throughput,standardisedGMOanalysis. Thedevelopmentofanalyticalmethodsandstrategiesformultiplexdetection,identificationand/or quantificationofGMOhasbeenamajorprioritywithintheCoExtraproject.Thecombinationoftwo or more PCR assays in one single reaction (multiplexing) is not an easy strategy due to the interactionandcompetitionbetweenthereactioncomponentsandproducts;andthecombination of high numbers of reactions is at the expense of the sensitivity and uniform amplification of the differenttargets.Numerousduplexreactions–oftentargetingthetransgenicsequenceandacontrol or two major screening elements are available, and so are oligoplex PCRs targeting a limited number of sequences. Above a certain degree of multiplexing, novel strategies (as compared to agarosegelsandrealtimePCRchemistries)arerequiredtoidentifythereactionproducts.Examples arecapillarygelelectrophoresis(CGE)basedandhybridizationinarrayformattechnologies.These approaches can allow simultaneous detection of the products of a number of oligoplex PCRs performedinparallel,resultinginhighermultiplexinglevel,throughputandlowercost. Theuseofoligoplex(andmultiplex)PCRassaysisforeseenasafirstanalysisthatallowsqualitative detection of GMO(s) in a sample. It can be then complemented with singleplex, validated (if possible), specific realtime PCR assays for GMO quantification when required. However, some oligoplexapproachesincorporatespecialadaptationstoachieve(semi)quantitativeresults,suchas quantitativecompetitivePCRortheuseofbipartiteprimers. ThelimitationsofPCRforachievinghighgrademultiplexingareoneofthereasonsthatprompted the study of alternative, non PCRbased approaches that could potentially allow multiplexing. Examples are the loopmediated isothermal amplification (LAMP) strategy coupled to bioluminescent assay for realtime (BART) detection system; and the NASBA (nucleic acids based amplification) implemented microarray analysis (NAIMA). Near infrared (NIR) spectra of individual kernelscanallowGMOdetectionbycomparisontopredefinedpatterns. New multiplex approaches have recently been designed for simultaneous detection of very high numbersoftargetsequences:thesecanbeconsideredashighgrademultiplexapproaches.Someof 57 CoExtraInternationalConference them include a first ligation step that is dependent upon hybridization of two oligonucleotide sequences to the target, subsequent amplification (with universal primers) and detection by hybridizationonarraysupport.ExamplesareaSNPlexmethod(forsinglenucleotidepolymorphisms detection)designedtoidentifyGMOtargets;andasystembasedonpadlockprobes(circularizable probes). In a very different approach, a whole genome amplification (WGA) technique can allow producinglargeamountsofgenomicDNAofthesamplethatarethenhybridizedtospecialprobesin microtiterplatesormicroarraystodetectGMOtargets(e.g.highdensitytillingmicroarray). This presentation will give an overview of the new technologies for multiplex analysis of GMO developed within the CoExtra project; and will also discuss on aspects such as the need and problems of validation of multiplex methods; or the difficulties in coupling a high level of multiplexingwithcosteffectiveness(includingthedevicesrequired)andsimplicityofthemethod. 58 CoExtraInternationalConference Session3:Legal,Liability&RedressIssues  3.1. Legal,Liability&RedressIssues BernhardKoch1andMAHermitte2 1 ECTIL,Vienna,Austria 2 CNRS,France.  Whilethelegalandregulatoryframeworkgoverningcoexistenceandtraceabilityhasbeenassessed more generally within the CoExtra project, the prime focus of this conference session will be on liabilityissuesthatmayariseinthecontextofGMOproductioninthefoodandfeedsupplychain. ThesemattershavebeenaddressedfromtwoangleswithintheframeworkofCoExtra: Oneteamofresearchersexaminedcontractualrelationshipsandtheobligationsarisingtherefrom, startingfromtheproductiontotheultimatedistribution.Thechainofcontractslinkingthevarious actorsmayincludedutiesexceedingtherequirementsoflocal,national,orinternationallaw.Itmay well happen that despite compliance with the latter the more stringent contractual rules are infringedupon.Theconsequencesofsuchbreacheshavebeenanalysed,withaparticulareyeonthe roleofgraintradersandtheirshareofresponsibility. AnotherresearchgroupexploredpotentialdelictualliabilitiesthatmayariseinthecourseofGMO production, analysing how the current liability regimes in Europe and selected nonEuropean jurisdictions wouldrespondtoharmcausedtothirdpartiessuchasconsumersorbystanders,and howdamagetotheenvironmentwouldbeaddressed.Specificaspectsofcrossborderclaimswere highlighted. Alternative options for compensation such as fund or other redress schemes were considered and compared to more traditional ways of indemnification. International liability regimes,inparticularthosepossiblybuildingupontheCartagenaProtocolonBiosafetyinthefuture, werealsotakenintoaccount.        59 CoExtraInternationalConference 3.2. Scientificexpertiseandthejudges C.Noiville CNRS–CRDSTParis1,France  Assciencehasbecomeagrowingfoundationofdecisionmaking,disputesmoreandmoreariseon the scientific basis of such decisions, at least when they deal with environmental or health issues. Whatisthequalityofthescientificreportsonwhichthedisputeddecisionrests? Doesthepresentstateofscientificknowledgejustifythisdecision? Haveallrelevantscientificdatabeentakenintoaccount? Wasn’tthepreviousscientificassessmenttooabbreviated? Wegiveelementsinordertobetterunderstandandmanagethesenewanddecisiveaspectsofrisk decisionmaking.    60 CoExtraInternationalConference 3.3. Juridicalcostbenefitanalysisofcoexistence:uneasythistask! 1 MA.Hermitte ,G.Canselier1andY.Bertheau2 1 CNRS,UMR8103,21bisruePasteur,92240Malakoff,France 2 INRA,Versailles,France  The various analysis costs/benefits applied to the transgenic plants issue mainly concern their advantages and drawbacks in comparison with conventional plants. We also find analysis of additional costs, attributable to nonGMOs supply chains. But the existing studies do not take in considerationalloftheactorsfromgeneticresourcesadministratorstothesocietyingeneralnor theensembleofparametersfromthecostsoftraceabilitytothebenefitsresultingfromsubsiding conflicts.Furthermore,overcostslinkedtocoexistenceareingeneralmixedwithcostsuninvolved with coexistence. Finally, a global societal analysis has up to now never been made. These are methodsdifficultieswhichwehavetriedtofigureout.    61 CoExtraInternationalConference Session4:StakeholderViewsinEU  4.1. Stakeholderviewsandinteractions René Custers1, George Sakelaris2, Jeremy Sweet3, Klaus Minol4, Kristina Sinemus4, Mojca Milavec5, AnkJansen6andMarieCheneval7 1 FlandersInstituteforBiotechnology(VIB),Gent,Belgium 2 NationalHellenicResearchFoundation(NHRF),Athens,Greece 3 TheGreen,Willingham,CambridgeCB245JA,UnitedKingdom 4 GeniusGmbH,Darmstadt,Germany 5 NationalInstituteofBiology(NIB),Ljubljana,Slovenia 6 Schuttelaar&Partners,DenHaag,TheNetherlands 7 Adriant,Nantes,France  The activities described in this presentation are part of the CoExtra project on “GM and nonGM supply chains; their Coexistence and Traceability” (www.coextra.eu). Part of this project is organisinginteractionwithrelevantstakeholdersaffectedbytheissueofcoexistencewiththegoal to: x Maptheopinionsandattitudesofrelevantstakeholderswithregardtocoexistence. x Createinteractionbetweenstakeholdersassuch.  WithintheCoExtraprojectstakeholderinteractionwereorganisedondifferentlevels: x InteractionwithagroupofstakeholdersonaEuropeanlevelthroughaStakeholderAdvisory Board. x RegionalstakeholderworkshopsinsevenEuropeancountries. x Anonlinequestionnaire The presentation describes the outcome of regional stakeholder workshops in several European MemberStatesandoftheonlinequestionnaire.WheretheCoExtraquestionnairetriedtosurvey general opinions and attitudes towards coexistence Europewide, the regional stakeholder workshopsweredesignedtogomoreindepthandgetmoredetailedinformationonmoretechnical topics.Boththeworkshopsandthequestionnairehavehadoutcomesofaqualitativenature.Byin largetheseoutcomesarenotcontradictory.Therearedifferencesbetweencountries,andthepolicy contexts in those countries play an important role, but comparing the results of one stakeholder categoryindifferentcountries,theytendtohavesimilaropinions. Some of the main results are concerning seed thresholds, costs of coexistence and traceability as wellastheharmonizationofcoexistenceruleswithintheEuropeanUnion.Thereisanoverwhelming wish to have GM labelling thresholds for seeds regulated and a general conviction and concern about the costs that coexistence regimes will entail in practice. Additionally, stakeholders are concerned about the practicalities of sampling and testing strategies. Guidance may be necessary here,andperhapsalsoadiscussiononwhethertestingisnecessaryinallsituations.Harmonization is seen as advantageous, but especially with the aim to prevent any unfair competition between differentEUcountries.Moststakeholdersarenotadvocatesofahybridregulatorymodelwithrules bothontheEuropeanandthecountrylevel,butsomemaystresstheneedforflexibility,especially onthepracticallevel. 62 CoExtraInternationalConference Session5:DecisionSupportSystems  5.1. TheCoExtraDecisionSupportSystem:AModelBasedIntegrationof ProjectResults MarkoBohanec1,7,YvesBertheau6,CarloBrera3,KristinaGruden2,ArneHolstJensen4,EstherJ.Kok5, Baptiste Lécroart6, Antoine Messéan6, Marina Miraglia3, Roberta Onori3, Theo W. Prins5, Louis GeorgesSoler6,MartinŽnidarši1 1 IJS,JožefStefanInstitute,DepartmentofKnowledgeTechnologies, Jamovacesta39,SI1000Ljubljana,Slovenia 2 NBI,NationalInstituteofBiology,Ljubljana,Slovenia 3 ISS,NationalInstituteofHealth,Rome,Italy 4 NVI,NationalVeterinaryInstitute,Oslo,Norway 5 RIKILT,InstituteofFoodSafety,Wageningen,TheNetherlands 6 INRA,InstitutNationaldelaRechercheAgronomique,France 7 UniversityofNovaGorica,NovaGorica,Slovenia  CoExtra (2005–2009) is an EU research programme on coexistence and traceability, which has involved almost 200 scientists from 54 partner organisations and has produced an extensive collectionofresults,suchasdata,scientificfindings,obtainedknowledgeandexpertise,formulated recommendations,developedmethodsandmodels,scientificpublications,etc.OneofthemainCo ExtraaimsisalsotoproduceaDecisionSupportSystem(DSS).TheideaoftheDSSistointegrateCo Extraprojectresultsinaformofacomputerbasedinformationsystemthatwouldbeoperational, easily accessible for various categories of users, and could provide data and advice for various decision problems that occur in supply chains involving genetically modified organisms (GMOs). In principle,theDSSisaimedatprovidingsupportfordecisionmakingandnottomakedecisionson behalf of stakeholders. It provides tools, and methods to assess various ‘decision alternatives’, to changevariousdecisionrelatedparametersandinvestigatetheireffects,tovisualisetheresultsof evaluationsandanalyses,andtomaintaindatarelatedtothedecisionsinvolved. Inparticular,theCoExtraDSSaddressesthefollowingdecisionquestions: 1. Whichmethodsperformbestorcanbeusedatallforagivenanalyticalorsamplingpurpose? 2. Will my (intermediary) product, given a current set of used procedures and materials, contain GMOsbelowaspecifiedthresholdlevel? 3. Isthereanypossibilitythatmy(intermediary)productcontainsunapprovedGMOs? 4. WhatarethecostsassociatedwithmaintainingGMOcontentbelowsomespecifiedthreshold? These questions are general and thus interesting for various potential users of the DSS: EU policy makers, farmers, importers, transporters, feed/food producers, retailers, consumers, analytical laboratories, users of test reports from analytical laboratories, and operators and managers of officialcontrol.AlthoughfirstdevotedtoGMOandnonGMOsupplychainsmanagement,itsquite genericstructuremaybeadaptedtothemanagementofsupplychainsfacingotherqualityorsafety issues,suchaspathogens,allergensandmycotoxinsproducingorganisms. Approach.WeareusingtheapproachofmodelbasedDSS,whichwecombinewithdatabasedDSS andelementsofsimulation,dataanalysisandcommunicationbasedDSS.ModelbasedDSS(Power, 2002) emphasize access to and manipulation of a model, for example, statistical, financial, 63 CoExtraInternationalConference optimization and/or simulation models. In CoExtra, we have primarily used two types of models: qualitative multiattribute models (Žnidarši, et al., 2008) and decision trees (De Ville, 2006). The modelshavebeendevelopedinanexpertmodellingway,thatis,incollaborationbetweendecision analystsandexpertsforagivendecisionproblem,andusingthemodellingsoftwareDEXi(Bohanec, 2008). Ingeneral,thepurposeofallthedevelopedmodelsisthreefold: x tocaptureandrepresentexpertknowledgeintheformofhierarchicallystructuredvariablesand decision rules, which can be reviewed, published, discussed, disputed and communicated betweenexperts,stakeholdersandotherinterestedgroups; x toactivelyassessandevaluatedecisionalternatives; x to analyze these alternatives using decisionanalysis tools, for instance, to find the advantages anddisadvantagesofalternativesandanalysetheeffectsofchangesby“whatif”andsensitivity analysis. Currently, the CoExtra DSS consists of six decision models and a database. The models are the following: x Analytical Model: aimed at the assessment of analytical methods, in particular DNA extraction andDNAanalysismethods; x SamplingModel:assessmentofsamplingplans; x Unapproved GMO Model: assessing the risk of contamination with unauthorized GMO varieties based on traceability data about the product (for instance, type of product, country of origin, typeandmodeoftransportation); x Transportation Model: assessing potential GMO presence due to transportation and handling basedonproducttraceabilitydata; x Dryer and Starch Models: assessing the effect of control parameters (such as using different strategiesforhandlingGMOandnonGMObatches)tothecollectionandprocessingofmaize. The CoExtra database stores data on food/feed products, sampling and analytical methods, and operational taxonomic units, which include GMO and taxa. This data is used as input into the models,butisalsosuitableforbrowsing,searchingandcreatingcomplexqueriesandreports.  References: 1. Bohanec,M.(2008).DEXi:Programformultiattributedecisionmaking,User'smanual, Version3.00.IJSReportDP9989,JožefStefanInstitute,Ljubljana.Availableat: http://kt.ijs.si/MarkoBohanec/pub/DEXiManual30r.pdf 2. CoExtra(2005–2009):GMandnonGMsupplychains:theirCOEXistenceandTRAceability. EUprojectFP6FOOD20057158.http://www.coextra.eu/ 3. Power,D.J.(2002).Decisionsupportsystems:conceptsandresourcesformanagers. Westport,Conn.,QuorumBooks. 4. DeVille,B.(2006).DecisionTreesforBusinessIntelligenceandDataMining.SASPublishing. 5. Žnidarši,M.,Bohanec,M.,Zupan,B.(2008)EuropeanJournalofOperationalResearch,189, 594608.   64 CoExtraInternationalConference 5.2. AnalyticalDSSmodule–howtosupportdecisionsintheanalyticallab KristinaGruden1,ArneHolstJensen3,YvesBertheau4,MartinŽnidarši2,MarkoBohanec2,5 1 NIB,NationalInstituteofBiology,Ljubljana,Slovenia 2 IJS,JožefStefanInstitute,Ljubljana,Slovenia 3 NVI,NationalVeterinaryInstitute,Oslo,Norway 4 INRA,InstitutNationaldelaRechercheAgronomique,France 5 UniversityofNovaGorica,NovaGorica,Slovenia  The analytical DSS module (AM) addresses the common situation faced by analytical laboratories involvedinthefood/feedproductionchain:givensomekindoffoodorfeed“product”,theyhaveto analyse it for different “purposes”. Common “purposes” related to GMO are the detection and quantificationofGMOsintheproduct.ThisisdoneinaseriesofactivitiesthattypicallyinvolveDNA extractionfollowedbyDNAdetectionmethods. Theselectionofeachmethodinthesequencedependsonanumberoffactors.Generally,weshould consider: x x x x propertiesoftheproduct(producttype,ingredients,...), purposeoftheanalysis(detection,quantification,...), propertiesofmethods(limitofdetection,applicabilityinthesituation,compatibilitywithother methodsinthesequence,...), capabilitiesofthelaboratory(availableequipment,skills,...). Thus,thiscanbeadifficultdecisionproblemthatrequiresextensiveknowledgeandskills.Thegoal oftheAMmoduleoftheCoExtraDSSistoprovidesupportforthefollowingdecisionquestions: 1. Isamethod“fitforpurpose”inthegivensituation? 2. Whichmethodis“bestforpurpose”? 3. Whichnewmethodis“bestforfurtherdevelopment”? Answers to these questions are provided by means of a qualitative multiattribute model. This model, which was developed in collaboration between domain experts and decision analysts, consists of hierarchically structured variables (attributes) and decision rules. Attributes represent properties of analytical methods, the hierarchy defines dependency relations between them, and decision rules determine the aggregation of attributes. For instance, Figure 1 shows a part of the attribute structure that is used to assess the “fitness for purpose” of analytical methods. This is assessedonthebasisof“purposefitness”and“sitefitness”,andeachoftheseisdeterminedfurther onabasisofmoreandmoredetailedpropertiesofanalyticalmethods. 65 CoExtraInternationalConference LOD QuantitativePerformance FitForQuantification FitForScreening PurposeFitness Specificity Purpose FitForOnSite FitForPurpose FitForOnSiteScreening FitForOnSiteExtraction SiteFitness OnSiteDetection Robustness OnSiteExtraction FitForExtraction Site Yield Purity  Figure1:ApartoftheAMmultiattrubutemodeladdressingmethods'“fitnessforpurpose”.  The AM module of the DSS is prepared to help stakeholders in optimising the DNAbased GMO detection.Foreachanalyticalsituationtheuserneedstodefineattributesrelatedtotheproperties ofsampletobeanalysedandthemethodsavailableinthelab(oravailableforimplementationin thelab).Thedataareenteredintoadatabasethatwillallowexchangeofentriesbetweendifferent users.DependingonthepurposeoftheanalysistheDSSwillsuggestwhichmethod(combinationof methods)isbesttobeappliedinthelab.Thusitisabletohelpalsotothelabsindecisionsrelatedto developmentofnewmethodsandtheiractualimplementationinthelab. Thesystemwasevaluatedusingmethodscurrentlyusedindetectionlabs,e.g.columnbasedDNA extraction and simplex realtime PCR detection, and the methods newly developed within the traceabilityworkpackagesofCoExtra.    66 CoExtraInternationalConference 5.3. DSSmodulesontransportation(TMmodule)andonunapproved GMOs(UGMmodule) EstherJKok1,TheoWPrins1,MartinŽnidarši2,MarkoBohanec2,3 1 2 RIKILT,InstituteofFoodSafety,Wageningen,TheNetherlands IJS,JožefStefanInstitute,Ljubljana,Slovenia 3 UniversityofNovaGorica,NovaGorica,Slovenia  As part of the CoExtra DSS two modules have been developed that aim to guide producers and other users to assess 1) whether their product is correctly labelled in terms of GMO (genetically modified organism) regulations and 2) whether their product might contain any materials derived from unapproved GMOs. In both cases the producer is guided to analyse the documentary data relatingtotheoriginandthetransportationoftheproduct,ortheunderlyingrawmaterials,inorder to estimate the chances of the unintended presence of GMOs or the potential presence of unapprovedGMOs,respectively. The CoExtra DSS module on transportation (TM module) aims to help the producer, or other Co ExtraDSSuser,toestimatewhethertheproductshouldbelabelledasaGMOproductaccordingto ECRegulation1829/2003,ornot.ThisRegulationongeneticallymodifiedfoodandfeedallowsthe presenceofGMOderivedmaterialsinGMOfreebatchesuptothelevelof0.9%,ifthispresenceis adventitiousandtechnicallyunavoidable.Inthatcasetheproductdoesnothavetobelabelledasa GMOproduct.Inallothercasestheproductingredient(s)thatareGMOderivedwillhavetolabelled assuch.TheTMmodule,whosestructureofattributesisshowninFigure1,assessesavailabledata on origin and logistics of the different (raw) materials that constitute a final product in order to determine whether the product should probably be labelled or not. This assessment relates primarilytodocumentarydata,butmayalsocompriseanalyticaldataonindividualcomponents,if available.Theanalyticaldataofindividualcomponentsassuchwillnotalwaysbesufficientasmore ingredientsderivedfromthesamecropspeciesmaybeincludedinthefinalproduct.Inthatcaseall thesecomponentsshouldbeincludedintheassessmentforcorrectlabelling.Relevantdocumentary datawillinclude,amongstothers,dataonthecountryoforigin,thenumberofharboursandbulk carriers included in the transportation system, and information on whether coexistence measures havebeenimplementedinthecountriesinvolved. The CoExtra DSS module on the potential presence of unapproved GMOs (UGM module) aims to helptheproducer,orotherCoExtraDSSuser,todeterminethechanceofunintendedcommingling of any (EU)unapproved GMO in (one of the raw materials constituting) the final product to be marketed. The basis for this assessment is the same EC Regulation 1829/2003 that stipulates that unapprovedGMOsarenotallowedinfoodandfeedproductsthatarebroughtontotheEuropean market.IfunapprovedGMOsaredetectedthismaythereforehavesevereeconomicimplicationsfor the producer or importer of a particular product. The UGM module (Figure2) assesses available logisticaldocumentationontheindividualingredientsofthefinalproducttodeterminethechance that any (unintended) commingling with unapproved GMOs has occurred during growth, harvest, transportation, storage and processing of the final product. The UGM module also assesses both documentary as well as analytical data that is available on the final product and the constituting ingredients.Relevantdocumentarydatawillinclude,amongstothers,dataonthecountryoforigin, the production area of approved GMOs in this country, the transportation system used for the productorunderlyingrawmaterials,thelogisticalrouteandnumberofharboursandsilosinvolved inthesupplychain.  67 CoExtraInternationalConference Both modules, that complement each other, will be explained and illustrated on the basis of a numberofexampleproducts.  Attribute GM_Presence TraceabilityData Products CropRisk ProductType CropSpecies ProcessingLevel Countries NumberOfCountries CountriesAtRisk CoexistenceMeasures Transportation Storage DedicatedSilos Carriers NumberOfHarbours DedicatedCarriers AnalyticalData AnalyticalDataAvailable AnalyticalData Description Transportation Module: Assessment of GM presence due to transportaiton Risk due to traceability data Risk due to product characteristics Crop/product type Product type Crop species Processing level Risk due to the properties of countries and regions of origin Number of countries involved in storage Are there countries at risk involved? Are coexistence measures in place in countries? Risk due to transportation route Risk due to storage Dedicated silos used for non-GMO? Risk due to carriers Number of harbours involved Dedicated carriers used for non-GMO? Analytical data available about unintended admixture Is analytical data available? Analytical data, if available  Figure2:AttributestructureoftheTMmodule.    Attribute UGM GeographicalOrigin EU GM_Region SystemsUsed TraceabilitySystemInPlace IP_GMO IP_Other AnalCtrl_Systems PrivateContracts Logistics Log_Complexity Interactions Companies Log_Storage Harbour Silo MethodsUsed AppropriateMethods AppropriateSampling AppropriateAnalysis AnalyticalResults ResultsAvailable Results Certificate Description Detection of Unapproved GM using Traceability Data only UGM risk related to the geographical origin of the product Does the product originate in an EU country? Does the product originate in a region of large GMO production? UGM risk due to used traceability systems Is a traceability system in place? Are IP systems for GMO being used? Are other IP systems being used? Are there systems used that include analytical control? Are there any private contracts? UGM risk originating in logistics UGM risk due to logistics complexity Number of interactions in the supply path Number of companies involved in logistics UGM risk due to storage used Has the product been shipped through harbor(s)? Has the product been stored in siloses? UGM risks based on the appropriateness of used methods and available results Have appropriate methods been used? Have appropriate sampling methods been used? Have appropriate analysis methods been used? Risks according to analytical results Are analytical results available? Analytical results, if available Regulation-based certificate with relation to GMO;s under emergency measure  Figure3:AttributestructureoftheUGMmodule.   68 CoExtraInternationalConference Session6:Experiencesfromthirdcountries  6.1. BenefitCostAnalysis,FoodSafety,andTraceability JamesK.Hammitt HarvardUniversity,CenterforRiskAnalysis,USA  Quantitativemethodsforanalysisofenvironmentalpolicyandotherdecisionsarepredicatedonthe need to make tradeoffs among valued outcomes. In general, no policy is best for everyone in the population, and no policy is best for all attributes of concern (e.g., risks of multiple health and environmental consequences, resources devoted to reducing risks). In determining whether one policyisbetterthananother,itisnecessarytoevaluatewhethertheharmsimposedonsomepeople (e.g.,costsofcompliance)areoffsetbythebenefitsconferredonothers(e.g.,reducedhealthrisk). Similarly,onemustdeterminewhetherthelossesonsomeattributes(e.g.,theresourcesdevotedto compliance that cannot be used for other social purposes) are offset by the gains on others (e.g., reducedhealthrisk). Benefitcostanalysis(BCA)isintendedtopredictwhetherasocietywouldjudgeitselfbetteroffwith apolicychange,inthesensethatindividualswhobenefitfromthepolicycouldcompensatethose whoareharmed(withmoney)sothateveryonewouldpreferthepolicychangewithcompensation tothestatusquo.BCArequiresthatall(significant)effectsonallaffectedindividualsbequantifiedin monetaryterms.Thevalueofabenefittoanindividual(e.g.,reducedhealthrisk)isdefinedasthe maximumamountofmoneyshecouldpaytoreceivethatbenefitandstilljudgeherselfbetteroff thanifshedidnotreceivethebenefitanddidnothavetopay.Analogously,thevalueofaharmto an individual is the minimum amount of monetary compensation he would require in order to accept the harm and judge himself better off than if he did not suffer the harm and receive the compensation.Bysummingthesemonetaryvaluesacrossthepopulation,onecancalculatethenet benefits and so determine whether there is a surplus of the value of benefits over the value of harms. If so, then in principle compensation could be paid in such a way that everyone in the populationjudgeshimorherselfbetteroff. Themonetaryvalueofachangeinhealthriskmaydependoncharacteristicsoftheriskinaddition totheprobabilityandseverityofhealtheffect.Peopletendtobemorefearfulof,anddemandmore government regulation of, risks that are viewed as dreaded or uncertain. Dreaded risks are those perceivedtobeuncontrollableandinvoluntary(totheindividual),potentiallycatastrophic,affecting future generations, and where the potential harms are not distributed equitably in relation to the benefits of the riskproducing activity. Uncertain risks are those that are unobservable, newly recognized,havedelayedconsequences,orarenotwellunderstoodbyscience. Monetary values of health risk can be estimated using revealed and statedpreference methods. Revealedpreference methods are based on observing the choices people make in which they implicitly or explicitly trade changes in health risk against money. People who choose a more expensive,saferfoodareassumedtovaluetheincreasedsafetyatmorethantheincrementalcost, andthosewhochoosethelessexpensive,lesssafefoodareassumedtovaluetheriskreductionless than the incremental cost. A critical assumption of revealedpreference studies is that consumers understandthedifferencesinrisk,cost,andanyotherimportantattributesamongalternativefood types.Studieshavebeenconductedcomparingorganicallyandconventionallyproducedfoodsand experimentalstudieshavevariedtheriskassociatedwithmicrobialcontamination.  69 CoExtraInternationalConference Statedpreferencestudiesarebasedonsurveys,inwhichrespondentsareaskedwhatchoicesthey would make in a hypothetical setting. Statedpreference studies have addressed risks associated withavarietyoffoodbornerisksincludingpesticidesandmicrobialcontamination.Comparedwith revealedpreference studies, a strength of these studies is that respondents can be asked about hypothetical foods that are not yet available and can be informed about the risks and other characteristics. A weakness is that survey respondents have less incentive to consider the choice carefully,astheydonothavetopaythecostsandfacetherisksastheydoinarevealedpreference study.     70 CoExtraInternationalConference 6.2. SegregationMeasuresfor(Non)GMcropsandtheirImplicationsfor SupplyChainsinJapan MasashiTachikawa IbarakiUniversity,Japan  In Japan, the decision to introduce mandatory labelling in August 1999 became the watershed of nonGMO identity preservation business, although implementation of labelling scheme was designated to start April 2001. Just after the decision, all kinds of food item under mandatory labelling system were changed to be manufactured using nonGM soybean and maize. Japanese GMOlabellingsystemisdifferentfromthatofEuropeanandonlyfooditemswhichmaintainrDNA or protein intact and detectable are subject of mandatory labelling. Therefore, food items such as refinedoils,sweetenerandsoysauceareexemptedfromlabelling.IncontrasttoEUsystem,feedis totallyexcludedfromlabelling. The original intention of introduction of mandatory labelling is to enable consumers to allow choosingtheirproductbasedonlabels.However,theresultisthatcompletereplacementofGMO withnonGMOwashappenedforfooditemswhicharesubjecttomandatorylabellingontheone handandcontinuousdependencyonGMOforthosewhichareexemptfromlabellingontheother. HereJapaneseconsumerscannotpracticetheirrighttochooseforGMOproduct,andGMOsbecame invisiblefromconsumers’viewpointwhilesomeitems,typicallysoyoil,arestillheavilydependent on GMOs. This invisibility of GMOs on our table keeps consumers’ uncertainty to linger, since no material experiences have been accumulated regarding this issue among consumers. This is an unintendedparadoxicalconsequenceofintroductionofmandatorylabellinginJapan. The cost of identity preservation of nonGMOs and their product occurs at various stages of food chain, and it depends on case by case (roughly speaking, they sum up about 15% to 20% of total cost).Coststructureisverydifficulttoassess,andmightbedifferentfromcompanytocompany.In this paper, I would like to elucidate how these cost are being managed for each food item by manufacturers,suchascornstarch,soyfoodetc.Basically,theseIPcostsofnonGMproductshas not been transferred to the price of final products, but rather absorbed by manufacturers. This inabilityofmanufacturerstransfercosttofinalproductcouldbeexplainedastheincreasingmarket power of retailers and deflational Japanese economy almost over a decade. The exception is non GMO dairy products manufactured using nonGMO feed. As the dairy product is exempted from mandatory labelling, this nonGMO labelling is totally voluntary, and manufacturers try to appeal thiskindofproductasspecialdifferentiatedproduct. Severalyearshavepassedsince2001andthereseemednoparticularfactortochangethissituation, twoseparatedmarketsofGMOandnonGMO,exceptincreasingpercentageofGMadoptionwithin USfarmers.However,USbiofueldemandandrelatedsubsidieshascompletelytransformedabove picture for nonGM market, in particular for maize. The market situation when the Japanese governmentdecidedtointroduceGMmandatorylabellingin1999wasthatthemaizepricewasso low that farmers were willing to make every effort to get additional premium, such as IP for non GMO. This situation has completely changed because of soaring market price for commodities. Without any additional effort, US farmers are now enjoying high market prices. It is widely recognized that nonGMO procurement is very difficult to sustain for a long term. Soaring grain prices give a large cross pressure upon food manufacturers from both ends of food chain, and eventuallyresultinfurthercostpricesqueeze. Another important issue of segregation today is low level presence and commingling of unauthorizedeventsingrainsforfood,feedandprocessing.Basedontheseriesofcontamination,it iswidelysharedthatsomekindofriskmanagementmeasuresneedtobetakentoavoidthiskindof contaminations and following disruption in food chain. In the similar vein, US government has 71 CoExtraInternationalConference proposed policy on low level presence of GMOs and early notification of risk information to government agencies. US biotech industry organization has also begun its initiative on quality control and risk management system called “Excellence Through Stewardship (ETS)”. Along with theseinitiatives,JapanesegovernmentnowtakesonestepfurthertofacilitatecoordinationwithUS government,biotechindustryandtraderstoelucidateeachroletominimizethiskindofdisruption. However, this initiative of Japanese government is limited to “feed” only. In this paper, I also describethisactivitiesandimplicationsonsegregationandtraceability.                 72 CoExtraInternationalConference 6.3. CoExistenceandtraceability:Costsandbenefitsinfoodandfeed supplychains WilliamW.Wilson DepartmentofAgribusiness,NorthDakotaStateUniversity,Fargo,ND58105,USA [email protected]  European Union (EU) traceability requirements impose added costs and risks on suppliers. This is trueforprospectivederegulatedtraitsgrowninothercountries.Inthispaperwedescribethecosts and benefits of traceability, as well as the operational implications of conforming to these requirements.Wedrawonseveralsetsofresultsthatanalyzetraceabilitycosts,strategiesandrisks respectively. IN particular, we use results from stochastic simulation models to determine optimal testingstrategiesandmarginalcoststoconformtoEUtraceabilityrequirementsforexportsofnon genetically modified (nonGM) wheat from the United States. The optimal strategy is chosen to maximizeanintegrator=sutility.Costcomponentsincludecertifiedseed,certificationandauditing, testing,traceability,qualityloss,andapremiumfortheaddedriskofadualtraceabilitysystemover a single nontraceability system. Adventitious commingling risks are defined stochastically. Results indicatethattraceabilityrequirementscanbeconformedtowithreasonablebuyerandsellerriskat atotalcostof$18/nonGMmt.    73 CoExtraInternationalConference 6.4. ACompanyPerspective RandalGiroux, CargillIncorporated,USA  Effectivecoexistenceisonemeasuretoassuretheproductionandprocessingofabundantsupplies ofsafeandnutritiousfoodsonasustainablebasis,whileallowingcustomersandsupplierstobefree to choose whether to use conventional, organic, or agricultural biotechnology products consistent with underlying consumer preferences and choices. To enable effective commercial coexistence, there must be a recognition that this is a dynamic, evolving, and complex marketplace involving diverseagriculturalsystems. Several countries have adopted all three agricultural systems and each has evolved towards some form of coexistence, that is commerciallyrelevant and effective for their needs. It must be recognizedthatthe concurrentuseofdifferentproductionsystemscansometimeslimitindividual choices of both farmers and retail consumers. While true consumer demand eventually influences what farmers grow, sometimes there are temporary market failures in meeting emerging demand foraparticularcroporproduct. In developing coexistence, some markets have discovered commercial realities that can either enableorinhibitcoexistence.Ifthedesiretodevelopcoexistencecontinues,marketswillcontinue to evolve towards enabling key features and addressing the inhibitors. Key features enabling coexistenceinclude: x x x x x x x x The development and availability of identitypreservation (IdP) systems and test methods to meet marketbased thresholds for adventitious presence that are appropriate to specific applicationsandneeds; The willingness of customers and retail consumers to pay a premium for differentiated food products,e.g.organic,nonGM,andotherspecialtyproducts;and, Commercialagreements(contracts)basedonclear,verifiableandachievablespecificationswith limitedgovernmentmandate; Severalkeyfactorsalsoposeachallenegecoexistenceincluding: Failure to adequately contain regulated GM events or products being developed through breedingprogramsand/orfieldtrialsassociatedwithproductdevelopment. Adventitious Presence (AP) policies for LLP and food/feed labelling that are not commercially achievable. Asynchronous approvals and zero tolerance policies can have significant upstream and downstreameffects.Exportsofanentirecrop(GM,organic,andnonGM)canbeplacedatrisk. Complianceandenforcementprotocolsthatarenotbasedonconsistentstandardsandhavenot beensuitablyvalidatedordemonstratedtheyarefitforpurposeastheproductmovesthrough thefoodandfeedsupplychains This presentation will provide a general overview of commercial perspectives on managing coexistenceandsomelessonswehavelearnedinimplementingthesesystemsintobothdomestic andglobalfoodandfeedsupplychains.  74 CoExtraInternationalConference 6.5. ProtectingEuropeanqualityagriculture:NonGMfeedsupplyand production TheactionoftheEuropeanGMFreeRegions2005–2010 RenaudLayadi InternationalNetworkprojectManager ConseilRégionaldeBretagne–KuzulRannvroBreizh.France.renaud.layadi@regionbretagne.fr  Europe’sagriculturalpolicyatacrossroads Since the 1958 Stresa conference, Europe has built a unique tool to promote its agriculture: the CommonAgriculturalpolicy.Duringthe70iesand80ies,Europebecameselfsufficientandamajor player on world food markets: the European budget being largely directed towards the CAP with morethan50%ofthetotalspending. Inthe90ies,thecontextbegantochange: x x x The historical disagreement between USA and Europe about subsidies changed into a North – South conflict, the Cairns Group of 18 major exporting countries opposing to the USAEurope subsidy policy. The GATT and later the WTO supplied its members with a efficient action and negotiationplatform. The extension of the EU with the integration of new members questioned the opportunity to havehalfoftheEuropeanbudgetdirectedtowardsCAP. European agriculture realised, under the pressure of Mediterranean countries, that quality agriculture products had to be promoted and issued the AOPIGP regulation including a “cultural” content to foodstuff. Northern countries concentrated on organic farming and also pushedtowardsacommonpolicyonthesubject.  QualityagricultureinEuropeandinournetwork On of the main aspect of the GMFree regions network (44 regions in Austria, Belgium, France, Greece,Italy,UK,Spain)istheverystrongpresenceofqualityagricultures: x x x To keep a high quality standard products enabling the sector to employ numerous farmers at workandalivelyruralsector. To manage landscape and environment either because there are important needs and/or becauseit’sacomponentoftheregion’simage. To be able to face a perspective of progressive deregulation of world food markets and competitionwithglobalizedfoodstuff.  CAPReformoutlook:towardsastrategicalternative Although the CAP has been frozen for the next few years, after 2010 one can expect important changes.FollowingtheintenselobbyingbytheCairnsGroupbutalsoothersectorsinvolvedinthe internationalnegotiations,manyproductionswillhaveaneasieraccesstoEuropeanmarkets:dairy products, poultry, beef, most of which are produced with GM feedstuff. Direct subsidies will be questioned and most probably reduced unless they show unquestionable advantages on environmental(andmaybesocial)issues. Regional agricultural networks in all the European regions will be faced with stronger competition andonlyveryfewalternatives. 75 CoExtraInternationalConference  Either: x Trytocompetewithlowpriceproductscomingfromabroad.Thisfirststrategycanonlybemet withindustrialisationofEuropeanfarming:lowemploymentrate,largeproductionunitsinwell organised territories close to logistic facilities and probable use of GMOs in feedstuff. If this model is chosen, our regions’ agriculture will not be able to sustain environmental, social or culturalgoals.Theirspecificitywilldisappearaswellasasignificantpartofruralemployment. OR x Promote specific agricultures to avoid direct confrontation with globalized foodstuff: This secondstrategyreliesonatotalqualitymanagementschemeincludingastrongterritoriallink with social, landscape and environment management issues; it also means that no genetically modifiedmaterialshouldbeusedthroughouttheproductionprocess.AOPIGP,organicandalso someprivatelabelsareconcernedbythismarketshareoftheEuropeanfoodmarket.  Feedstuffissueandspecificagricultures–step1accomplished Although some countries of the network have taken steps towards self reliance in feedstuff, the abilitytokeepanimportednonGMfeedsupplyiscrucialforqualityfoodstuffproduction.Todayall theregionsofthenetworkimportsoytoincreaseproteinrateintofeedstuffandifEuropeimports morethan37millionmetrictonsofequivalentsoycakeonecanguessthatourregions’potential, due to their strong agricultural sector, is at least 10 million metric tons; part of which is being directedtoqualityagricultures. Thisisthereasonwhyournetworkhaslaunchedaspecificactiononfeed: x x x x October 2005, mission of 11 regions to Brazil, meeting with Federal Ministers of Environment (Mrs Silva) and Foreign Trade (M Furlão), letters of intentions with the State of Paranà, stakeholdersinParanà,SantaCatarina,Goias,visittoproducers... December 2007, 1st global nonGM feed global meeting organised by our network at the CommitteeoftheRegionsinBrusselsgatheringsoyproducersfromBrazil,Canada,China,India, USA as well as with 117 European businesses collecting meat & milk from more than 680000 farms,48regions,21membersoftheEuropeanParliament... October 2008, 2nd global nonGM feed global meeting organised by the nonGM Trade in Brussels.StrongattendancefromNorthernEuropetradersofficialannouncementofthecreation oftheBraziliannonGMsoyproducers’association(ABRANGE)... January–February2010,3rdnonGMfeedglobalmeetingtobeorganisedbyournetworkatthe CommitteeoftheRegionsfocussedontheuseonnonGManditslinkwithqualityagriculture. PartnershipwithAREPO(AssociationofQualityproductionRegionsofEurope). In just 3 years our network has registered valuable successes. Besides raising awareness in the European regions and creating a market intelligence network worldwide, our message expressed rightfrom2005aboutthenecessityforBrazilianproductiontoorganisehasgonethrough(creation oftheABRANGE).Indiansoyproduction(100%GMFree)isnowsupplyingtheEUmarket,andthe perspectiveofacompletedroughtinnonGMsoysupplytoEuropeseemstohavebeenavoided.  Thenextsteps–therelationshipwithconsumers The issue for our network, now that the nonGM soy supply seems secured on the long run, is to value the efforts of nonGM soy producers (Brazil, India, Canada, USA, China) and of nonGM soy users(Europeanqualityproductions). 76 CoExtraInternationalConference ThenetworkhasobservedwiththeupmostattentiontheGermannonGMfeedlabel,issupporting it’s own producers to follow suit and will lobby the European Commission to facilitate the implementationofthiskindoftool.ThatiswhynonGMfeedsupplyinthenextnetwork’sbusiness meetingwillnotbethemainsubjectandregionswillprefertofocusonhowtovaluenonGMfeed use through regional strategies on feed autonomy or success stories of productions that have bannedGMfeedcomponentimportsorproductions.    77 CoExtraInternationalConference 7. IntegrationofCoExtraresultsinEUtoolsforcoexistence&traceability GuyvandenEede EuropeanCommission–JointResearchCentre,Italy  The European Network of GMO Laboratories (ENGL), chaired by the Joint Research Centre of the EuropeanCommission,isapanEUNetworkofenforcementand/or(national)referencelaboratories thatdealswithalltechnicalissuesrelatedtotheenforcementoftheGMOregulationsinEurope.Its activities have had a significant impact on the technical capacities, not only of the participating laboratories,butalsoonthecapacitiesofGMOtestfacilitiesingeneral. ThisleadingrolehasbeenacknowledgedbytheEuropeanCommissionandtheEuropeanParliament and has nominated the JRC as Community Reference Laboratory for GM Food and Feed and has entrustedasupportiveroletothe ENGL.Thankstothe JRC,the EUregulationshaveinscribed the obligation for notifiers to provide detailed information as well control samples and reference materials, which has had a major impact on testing. Ten years ago it was unforeseeable that all information about eventspecific methods is published on the Internet even before the GMOs are approved. The discussions within ENGL are of the highest possible scientific quality: indeed the Commission truststhattheJRCandENGLbasetheirsolutionsandproposalsforharmonisedtestingontechnical solutionsbasedonthebestknowledgeavailableandonscienceofthehighestquality.Thereforeitis imperativethatitsmembers,aswellastheJRC,arenotonlyawareofthescientificdevelopments butarepartofscienceprojectsofexcellencethemselves. ENGL has been already well represented in FP5 (QPCRGMFOOD) and it has been a substantial associate in launching CoExtra and has been a privileged partner in having access to the information,basedonanagreementbetweenCoExtraandENGL. Many of the advancements of CoExtra have already been discussed in ENGL and will certainly impacttheactivitiesofthenetwork. Todayweseealsoachangeofpracticemovingforwardfromvalidationofmethodsfornotification purposes towards validation of approaches for control purposes, for instance by looking at matrix approaches and concomitant decision trees and by applying new readytouse qualitative tools, furtherimprovingtheharmonisationthroughouttheEuropeanUnionandbeyonditsborders. Duringthepresentationareviewoftheachievementswillbemade.However,itisimperativeforthe functioning of ENGL and the whole enforcement process as a whole that the consortium can continuetobeinvolvedinambitiousresearchprogrammesandcanremainabreastofscienceand technologicalcapacitiesinordertocontinuetofulfilitsroleandtodemonstratetotheconsumers and stakeholders that the EC is capable to implement a stringent regulation in a very complex technicalarea. 78 CoExtraInternationalConference 8. SummaryofmainCoExtradeliverables&results,perspectives, informationdissemination&application. YvesBertheau INRA,Versailles,France  Generallyspeaking,GMOproductionanduseisaquitecontroversialsociallydebateditemaround theworld.Thisworldwidecontroversyisalsofacingtheobligationoffreetradeasmademandatory by international treaties like the WTO. It was not the purpose of CoExtra to participate to these debatesbut,asaEUpolicysupportresearchprogram,toprovidealltechnical,economic,scientific andlegalbasisforprovidingtheEuropeanstakeholderswithaccuratedataformakingdecisionand keepingthefreedomofchoicetoEuropeanproducersandconsumers. HerewithyouwillfindthemostimportantresultsandmessagesfromtheseveralWorkpackagesof CoExtra,bothintermsofscientificresultsandpracticalimplementationissuesandsolutions.Some issuesarestillpendingandwillneedfurtherresearch,someofthemalsodependingonthedecisions to be taken by policy makers on, for instance, seeds thresholds and fields’ coexistence to be harmonizedornottotheEuropeanlevel. AsasummaryoftheCoExtraresults,thissummarydoesnotshowallresults,butmoredetailscan be found on the CoExtra website (www.coextra.eu), particularly in the online deliverables and in thepeerreviewedpaperspublishedbyCoExtramembers.  Coexistenceinthefields The first aim of CoExtra on that issue was to test the stability and reliability of biological containment tools like cytoplasmic male sterility (CMS) in maize, cleistogamy in oilseed rape and plastidtransformationintobacco.Therefore,geneflowparametersofCMSmaizeandcleistogamous oilseedrapehavebeenstudiedoverthe4lastyears,underfieldconditionslocatedatdifferentsites inEurope: x x x TheCoExtradatademonstratethatstablecytoplasmicmalesterilityinmaizeisaneffectiveway toreduceoreveneliminateGMpollenmediatedgeneflowtoadjacentfieldsifstableTandC cytoplasms are used. Furthermore, appropriate combinations of CMS hybrids and fertile pollinatorsusedasanagriculturalbiocontainmentsystemcanleadtoasignificantgaininyield. Cleistogamousoilseedrapeasabiologicalmitigationtechniquehasamajorpotentialforlimiting crosspollinationduetothestrongreductionofthepollencloud. Moreover, data mining was performed to gain information about the suitability of chloroplast transformationasacontainmentstrategy.Theoutcomeisthatplastidtransformationprovidesa highlyeffectivetooltodecreasepollenmediatedgeneflowfromtransgenicplants.However,in cases where pollen transmission must be prevented completely, stacking with other containmentmethodsmightbenecessarytoeliminatetheresidualoutcrossingprobability. ThesecondaimofCoExtraworkoncoexistenceinfieldswastogaininformationaboutthemajor driversofmaizepollenflowoverfragmentedlandscapes,throughfieldexperimentsandmodelling. Various factors involved in maize pollen emission and pollen flow were analysed through existing data analysis and field experiments. Tools modelling velocity and pollen concentrations over heterogeneous fields were also developed to assess the crosspollination rates between GM and conventional maize over large distances and in fragmented landscapes. Using new and previously gathereddataastatisticalmodelofpollenemissioninrelationtomicroclimateandaphysicalmodel 79 CoExtraInternationalConference of pollen flow based on fluid mechanics were successfully validated. These results apply on single eventtransformations. x x TheCoExtradatademonstratethatpracticalandtechnicalknowledgeonGMcrosspollination inmaizeishighlyaccumulated.Modelshavebeenvalidatedforlargedistancesandfragmented landscapes. Technical measures could ensure that coexistence at the 0.9% labelling threshold for corn hybrids would be achievable on a longterm basis, as far as seed lots are pure enough. Co existence for maize grain production is feasible and highly dependent on local conditions (e.g. croppingsystems,landscapepatterns)andontheevolutionofpractices(e.g.rateofadoptionof GMvarietiesinaregionandcropmanagement).Furthermore,variouspossibilitiescanbeused in different situations (e.g. timelag of flowering vs. isolation distances) and local operators shouldbeabletochoosethemselvesthebestsolutionsdependingonthelocalconstraints.The issueoffarmersusingfarmssavedseedsandcornpopulationsinsteadofhybridswasaddressed inapartonlegalissues. ThethirdaimofCoExtraworkconcernsseeds.Seedlotsarethestartingpointsinaneverincreasing supply food chain; therefore field experiments of maize seed admixture have been conducted to evaluatetheeffectofseedthresholdsonthefinaloutcrossingrateintheharvestproduct. x x x x The main sources of adventitious presence in nonGM maize are seed impurities, GM cross pollination, and GM kernel transfer via machinery. The average potential rates of adventitious presenceoccurringatvariousstagesduringfarmproductionarerelevanttothe0.9%threshold setbytheEUlabellinglegislation. TheCoExtradatademonstratethatthefinalGMOrateintheharvestproductissimilartothat of the seed admixture for current GM varieties (but will differ with stacked GMOs) and highly dependentonlocalconditions(floweringcoincidence,thesiteandclimaticconditions). CoExtrahasalsoinvestigatedtheimpactofgenestackingonadventitiousGMpresencedueto pollenflowandseedadmixtureaswellasitstranslationintermsofpercentageofGMDNAina nonGMharvest.Weestablished,inthecaseofGMvarietiesbearingonetofourstackedevents, the relationships between the crosspollination rate between GM and conventional fields, the percentageofGMkernelsand the percentageof GMDNAina nonGM harvestaswellasthe relationshipsbetweentherateofseedadmixtureandthepercentagesofGMmaterialinanon GMharvest.Thankstotheserelationships,wesubstantiated,throughseveralexamples,thefact thatthenumberofeventsandthestackingstructureoftheemittingfieldsimpacttheabilityfor anonGMmaizeproducertocomplywithgivenGMkernelorGMDNAunitbasedthresholds. Onalegalandeconomicpointofview,publicresearchpoliciesshouldbedevelopedforinstance on,thebreedingofconventionalvarieties.Moreover,thegeneticresources,asthoseunderthe auspicesofCGIAR,shouldbepreserved.Accordinglyinternationaltechnicalprotectionmeasures shouldbeputinplace,withindemnification,compensationsystemsforhostingcountries.  ConsiderationsderivedfromCoExtrawork: Thepollenflowtobeexpectedtooccurduringthegrowthofcropsisindeedhighlydependenton thecrops’biology: x The seed purity10  is of utmost importance for ensuring coexistence in the fields. Any seed threshold(notyetdeterminedattheEUlevel),shouldbelowerthanthelabellingthresholdbut alsoleaveenoughleewaytomakeitpossiblethecoexistenceatthefieldlevel.Thereisatrade offbetweentheseedpurityandtheadventitiouspresenceintheharvest:thehighertheseed  10 AseedthresholdhasnotyetbeensetupattheEUlevel, 80 CoExtraInternationalConference x x x x x x x x x purity,thelowertheadventitiouspresenceortheeasiertoensurecoexistenceparticularlywhen takingintoconsiderationthethreshold,lowerthanthelabellingthresholdof0.9%,requestedby the companies. This practice of using a practical threshold lower is commonly observed in qualitycontrolofproductioninothersupplychainswhenathresholdisrequiredforqualityor safety purposes. For those supply chains which claim for thresholds lower than the official labellingthreshold,theseedpuritywillbeparticularlycritical. The techniques and procedures for obtaining seeds with low levels of admixture are already available since the GMOs’ seeds sold in numerous countries are also used with high levels of purity.AsobservedinotherresearchprogramssuchastheINRAresearchprogramheldin1999 2000, a low level for seeds threshold admixture might increase the prices of seeds, which is howevernotimpactingthefinalprices,. New sampling plans have to be tested for taking into account the still to be decided seed thresholdandthepracticalthreshold.Sofar,mostofthestudiesofotherresearchnationaland European programs have focused on an expected seed threshold around 0,5% and a kernels labelling threshold of 0.9%. The results of the CoExtra first study for reaching such a 0.1% thresholdareexpectedsoon. Biocontainment measures may facilitate the implementation of individual farm coexistence, provided models taking into account the several environmental conditions and the farmers’ choices (such as individual choices of growing or not GMOs, late arbitrages according to expectablemarketstrends,etc.)factorsinvolvedarecarefullyconsidered. Thepracticalimplementationofbiocontainmentmeasuresdoeshoweverraiseseveralissues: Tillnow,oneoftheratherstableCMSofcorntype(Ttype)isoneofthosealreadyobservedfor hybridproductions,thuswithahighsensitivitytoafungalpathogenfromwhichanepidemicin the70’shadhugeeconomicimpactonseedproduction.Itsusemightbelimitedtothegrowth ofsmallscaledtransgenicfields,e.g.forthesynthesisofpharmaceuticals. Thepracticalimplementationinfarmers’fieldsofsuchmixturesofCMScornandfertilevarieties should be further studied, though higher yields can be expected as observed with the Hybrid PlustechnologystudiedintheCoExtraproject. There is no indication of the rapid commercialization by the seed companies of corn varieties with CMS or oilseed rape with cleistogamy traits. Accordingly, the use of biocontainment methodsisdependingonthefuturereleaseofbiocontainedvarietiesbyseedscompanies.The interest of the seeds companies to release these biocontained varieties is questionable, since morecostly,aswellasitcouldfavourthedevelopmentofhybridsbyfarmersandrevealsome knowhowtotheircompetitors. Duetothe effectofthedefinitionoftheDNAunit asrecommendedby the EC,theincreasing number of stacked genes will rapidly increase the GMO content, measured as HGE11. Accordingly, it may be recommended to use the biocontainment methods to stay on the safe sideoftheGMOcontent. Thefamersusingfarmssavedseedsshouldbenefitfromthesameprotectionmeasures,suchas longdistanceisolation,thatthefarmersprofessionallyproducingforseedscompanies.Farmers whoproducefarmssavedseedsshouldofcoursebenotified,andGMcropsshouldbeproduced withthesameminimumdistancestoavoidanycrosspollinationwithfarmsavedseeds.  Inconclusion,accordingtotheresultsofSIGMEA12modelsandtheresultsofCoExtra,particularly those concerning the practical contractual threshold used by the stakeholders, and the available techniquesandinformationsystems,coexistenceinEuropeanfields,whosesizeisonaveragerather  11 HaploidGenomeEquivalent 12 SustainableIntroductionofGeneticallyModifiedCropsintoEuropeanAgriculture,FP6researchprogram.  81 CoExtraInternationalConference small, would be possible only by using large isolation distances (together with strong information systemoffarmers)orindedicatedproductionareas,beitGMOornonGMO. Thevalidatedbiocontainmenttechniquesmayprovideaneffectivetooltoincreasethebiosafetyof transgenic plants and might be used to reduce dramatically for instance implemented isolation distances. However, in cases where pollen transmission must be prevented altogether (e.g. GMO used for nonalimentary purposes), stacking with other containment methods will be necessary to eliminatetheresidualcrosspollinationrisk.  CoExistenceinthesupplychains Supplychainsmanagement Generallyspeaking,theEuropeancompanieshavenotyetbeenfacingwithcoexistenceastheEuropean GMO production is rather limited and mostly, if not completely, used in feed production. As animals derivedproductsarenotlabelled,coexistenceiscurrentlynotanissue.Thirdcountries,withverylarge fieldshaveimplementedefficienttraceabilityandproductssegregationforexportstowardsthecountries, likeEU,withalabellingthreshold. FrominterviewsconductedwithEuropeanandthirdcountriescompaniesinvolvedincommoditysupply chains,itcanbestatedthatavastmajorityofstakeholders,ifnotall,isusingapracticalthresholdwhich is lower than the labelling threshold (generally from 1/3rd to 1/10th of the labelling threshold, more generally0.1%ofDNAbasedunitGMOcontent).Theseobservationsconfirmthosemadesince2001in other studies on GM and nonGM supply chains (such as third countries IP13 systems). This practice is similartotheonesusedinothersupplychainsmanagement(mycotoxins,allergens,pathogens,etc.).This very common practice of using a practical threshold lower than the official one (for quality or safety purposes) can be explained by the assurance required by stakeholders to protect themselves against samplingandanalyticalmeasurementuncertaintiesinfrontofcontractsorStatecontrols.Inaddition,this practice is rather easy to implement today because the GM pressure is today very weak. It has been difficult to assess what would be the behaviour of stakeholders under different scenarios (combining differenthypothesesonseedthresholds,nonGMdemandorGMpressure). This practical threshold contractually used by the stakeholders conditions the whole supply chain managementandthusthefarms’outcomesandseeds’threshold(s),stilltobedefined.Thisispartlydue totheabsenceofEuropeandefinitionofGMOfree.AllEUmembersStateshavinglegallydefinedGMO freeproductsareusingthe0.1%threshold.  In addition to the analysis of their current strategies, CoExtra has explored how stakeholders could be copingwithcoexistencealongsupplychains,whereGMcropsbedevelopedintheEU. x Inprinciple,stakeholderscanusethreedifferentsegregationstrategiestocopewithcoexistence alongsupplychains: x x If they have dedicated factory plants (strategy 1), they can separate GM and nonGM material but this may lead to increased costs (transportation or underutilisation of some plantsifthemarketdemandchanges). Theycanalsouseseparateproductionlinesinthesamefactoryplant(strategy2),whichis moreflexiblethandedicatedplantsbutnotalwaysfeasible(forexamplestarchfactoriesuse singleproductionlines);  13 IdentityPreservation 82 CoExtraInternationalConference x x x x x x x x Thetemporalspecializationofprocesslines(alternatingbetweenGMandnonGMbatches)is moreflexible,butrequiresregularcleaningofequipmentordowngradingofnonGMbatches (strategy 3). Downgrading involves removing nonGM batches that do not meet a targeted thresholdforGMpresence,andarethereforearedivertedintotheGMsupplychain. In general, segregation of GM and nonGM supply chains is technically feasible, but the organisationofthechain,fromtheupstreamfarmerstothedownstreamstakeholders,playsa criticalroleinmaintaining/improvingtheprobabilityofcompliancewiththeofficialEUlabelling threshold level of 0.9% (with a practical threshold between 1/3 and 1/10 of the labelling threshold).Onthecontrary,upstreamfarmbatchesmaycomplywiththethresholdbut,ifchain managementstrategiesarenotappropriate,thelevelofcomplianceofthefinalproductmaybe verylow. Models have been developed by CoExtra to assess the effect of various variables on the GM adventitiouspresenceinnonGMbatchesandtheprobabilityofcomplianceofnonGMbatches withagiventhreshold,ateachstepofsupplychain(fromthefieldleveltotheenduser).These modelscanbeusedwiththe0.9%labellingthresholdaswellaswithlowerthresholdssuchas theca.0.1%practicalthresholdusedbythestakeholders. Thesupplychainsimulationmodel(basedontheexampleofthestarchsupplymaizechain)can test several management scenarios and compare the various strategies (i.e. automatic downgrading versus each batch processed subsequent to the processing of GM material is automatically put into the GM supply chain if a PCR test indicates the batch does not comply withtherequiredthreshold). Byusinggeneflowmodels,itispossibletoestimatetheadventitiouspresenceofGMmaterial innonGMmaizeatthefarmgate.TheCoExtraresultsshowthatthisinformationhelpsinthe implementation of an automatic downgrading strategy and may therefore save further PCR testing.Thisrequiresstrictverticalorganisationbutcanincreaseoverallprofitability. As the “nonGM” characteristic is not observable by the final consumers, public regulation is necessarytoenforcethecomplianceoffinalproductstothecompulsorylabellingthreshold.This compliancecanbeobtainedthroughpubliccontrolsandpenaltiescostsincaseofnoncompliant nonGMproducts(expostregulation).Itcanalsobeobtainedthroughtestingandsamplingrules imposedtoprivatestakeholders(exanteregulation). WhenGMandnonGMmaterialsareprocessedinthesameproductionline(strategy3),froman economicpointofviewthereisatradeoffbetweenthelevelofcomplianceofthefinalproduct andthenumberofdowngradednonGMbatches.Thistradeoffdependsuponboththerelative valueofthepenaltycostincurredasaconsequenceofnoncompliance(whenanonGMbatch doesnotmeetthethreshold)andthenonGMpricepremiuminthemarketplace. CoexistencebetweenGMandnonGMproductsseemsdifficulttoimplementwithinthesame supplychainswhentheGMpressureishigh.Itisonlyviablefromaneconomicpointofviewif thereisapricedifferentiationbetweenbothproductsinthemarketplace.Thisisnotalwaysthe case, and therefore some stakeholders have stopped segregating GM and nonGM compound animal feedstocks (because products derived from animals fed with GMO’s are currently not labelled).  Documentarytraceability Documentarytraceability(ISO20005:2007)isanimportantpillaroftheEuropeansystemofGMand nonGMcoexistencesystem.Itallowthecosteffectivemanagementofsupplychains,byusingdata fromratherrawmaterials,moreeasilyanalysable,intermsofsamplinganddetectionprocedures, provided critical points are identified along the supply chains and analytical controls are appropriatelymade. 83 CoExtraInternationalConference  Theconceptof“coexistence”isalwaysdirectlyrelatedtotheconceptof“segregation”,whichisthe shapethattheorganizationofthesupplychainsessentiallytakestomakecoexistencepossible.The term “coexistence” is linked with different meanings, which are sometimes confused in several studies. The first one concerns the links between coexistence and segregation and competition strategies. The second one is mostly linked to the problem of coexistence and segregation in relationwithdifferentiationtrendsandGMeventsmultiplication. x Theworkondocumentarytraceabilityshowstheexistenceofthreetypicalformsoforganization systemsforthesupplychainsinthecaseofnonGMOs: x x x x x Thefirstoneisalongand“containerized”supplysystem,whichcanbeobservedinArgentina andBrazil,usingtheoceantransport(generallycalled“hardIP14”). Thesecondsystemisalongbulksupplysystem,alsousingseatransport.Thissystem,usedin Argentina and Brazil to guarantee the European importers with the grains type, is an IP systemofsegregation. ThethirdsystemisanintraEuropeansystem. Since the enforcement of the Regulations 178/2002 and 1830/2003, traceability and labelling arerequiredforGMfoodandfeedproductsinEurope.InArgentinaandBraziltraceabilityofGM foodandfeedisoptionalandnotofficiallyrequired,LabellingisofficiallyrequiredinBrazil.The quality systems and the certification are a voluntary action of a part of the companies or cooperatives, most of whom are attempting to export their products, directly or by the intermediateofgraintraderssuchasADM,Bunge,CargillandDreyfuscompanies. Theexperienceoncoexistenceandtraceability,gatheredintheCoExtraProjectisofparticular relevance to the stakeholders and entrepreneurs, willing to implement new supply chain and quality system. However, these observations have little application for coexistence between farmers,duetothequitelargersizeofnumerousfarmsinthoseexportingcountries.  Economyofsupplychains The interaction of CoExtra partners with the companies has been rather difficult and thus the retrievalofquantitativedatahasbeenalmostimpossible. Generallyspeaking,thecostreductionimpactofgeneralEuropeandirectivesandregulations,such asthe178/02,makingmandatorytheimplementationoftraceabilityinEuropeansupplychains,is not properly estimated by the companies. Moreover, the positive impact of alreadyimplemented traceabilityandcontrols,duetoboththegeneral,orGMOspecificdirectivesandregulations,one.g. companies’image,decreasesofmarketwithdrawalsorrecalls,welfare,ordevelopmentofmarkets niches,impactofGMOandnonGMOsupplychainsorganizationonproductsrelatedtosafetyissues (e.g. management of products for allergens or mycotoxins), is also not properly estimated. On several occasions, the use of analytical controls was overestimated since lowcost documentary traceability is always used. Several third countries have already put in place efficient segregation strategies of GM and nonGM products, in order to gain new markets, which can be used for any valueaddedsupplychains. This situation may be due to either a lack of analytical analyses of the impact of these different legislationsframesortoawillingnessofcompaniestodisclosesuchresults,maybeforconcurrence relatedissues,orboth.  14 IdentityPreservation(meaningmanagementofnonGMproducts) 84 CoExtraInternationalConference Wecantranslatethislackofaccuratedataasalackofcompanies’willingnesstocarefullycarryout costbenefitanalysesoncoexistenceinordertoincreasecompanies’profits. Coexistence ofGMandnonGMsupplychainsispossibleonlyifallstakeholderscanvalorisetheir production. This is particularly important for animalsderived products which are not labelled, according to whether that animal wasfed with GM or nonGM products. Accordingly coexistence canbeinsuredintheEUonlyifGMOfreelabellingispossible,includinganimalsfedwithnonGM products. According to the results of the analysed food supply chains, only additional costs can thus be expected by organising coexistence between GM and nonGM products in the value chain from productionoffarmcropsuptotheproduction/processinglevelsofthesinglesupplychainsandby maintaining mandatory (or voluntary) thresholds and regulations. Depending on factors like crop requirements, farming, storage and elevating systems, processing strategies, monitoring managements etc, the total additional costs of coexistence and product segregation, for some systems, can increase to 13% of the total product turnover at the gates of rapeseed oil mills or starchindustryprocessingwheatandmaize. However,formostvaluechainsthequestionofcoexistenceisatheoreticaloneatthemoment.The implementation and permanent running of coexistence and segregation systems in the food industry can decrease the additional costs due to savings e.g. in the testing requirements of raw materialsorroutineproceduresduringthedocumentationprocess. Thesegregation,traceabilityandlabellingsystemsformaintainingtheGMOthresholdbelow0.9% hardlyprovidesanysignificantadditionalbenefitsforproducer,retailerorconsumer(asthiswould bethecasee.g.inorganicproduction,fairtradedproductsetc.).Thusitispossiblethatnoactorof thevaluechainmaybe willing topaytheincurredcostsof coexistencemeasuresoccurringalong thelineofthesupplychain. Since European consumers, of the countries studied, rarely accept genetic modifications in food products,theyareunwillingtopayextramoneyforproductdifferentiationinthesenseofalabelled foodproductthatcontainsGMmaterialsbelowthelabellingthresholdof0.9%.Besidesfarmersand seeds companies’ production and croprelated benefits by genetically modified crop varieties like pesticideresistances,anticipatedhigheryieldsorincreasedcontentsofsubstances,thebenefitsfor theconsumerarequitevague,intangibleandhardlyconvincing.Asshownintheconsumersurveys in the countries analysed, the putative health or environmental benefits of GM crops are mainly unknown,uncertainandtheconsumersseesnoreasonforspendingmoremoneyontheseproducts. x x MoreconsumersinDenmark,GermanyandPolandthoughteatingGMfoodsmightharmthem thandidthoseinGBandSpain.Relativelyfewconsumers,ineachstudycountry,agreedstrongly withthestatementthatGMtechnologieswillleadtohealthierfoodandtocheaperfood. ApartfromSpain,consumersinthefourotherstudycountriesrequired'compensation'inorder forthemtochooseGMfoodproducts.Furthermore,thelevelof'compensation'hastobehigher whenGMtechnologyisassociatedwithenvironmentalbenefits,thanwhenitisassociatedwith healthbenefits. The CoExtra results of consumers’ propensity to pay for nonGM products should be usefully compared to those obtained in the consumers’ survey carried out under the coordination of the King’sCollege15.   15 http://www.kcl.ac.uk/schools/biohealth/research/nutritional/consumerchoice. 85 CoExtraInternationalConference Traceabilityandcontrolsinsupplychains By traceability we understand below both the analytical traceability, carried out by analytical methods, and documentary traceability according to its usual standardized meaning (ISO 22005:2007). TheresultsdescribedbelowstronglybenefitedfromtheinvolvementoftheJRC16(IRMMandIHCP institutes)andofnumerousENGL17membersasCoExtrapartners. Efficientandcosteffectivesamplingandtestingapproachesareneededinordertoimplementco existence and traceability, stakeholders need first reliable sampling procedures to obtain representativesamplessecondlyvalidatedmethodsandfinallynovelmethodsduetotheincreaseof thenumberofGMcrops. Samplingrepresentstheinitialstepandinmostcasesthemajorcrucialstepoftheanalyticalchain particularlywhentargetsoranalytesarenothomogeneouslydistributedasforGMOs(seee.g.the Keldaproject18).Theanalysisofsamplesnotrepresentativeofthelotstobeanalyzedforcompliance could get to wrong decision and then to waste of cost and efforts. Development of sampling methodshasbeenanimportantgoalwithintheCoExtraproject. x x x x x Dedicated software tools to support sampling and subsampling plans aimed at GM detection throughthefoodandfeedchainweredeveloped:SISSIanovelapproachtoestimatetheoptimal sample size in experimental data collection and OPACSA (OPtimal ACceptance Sampling by Attributes)anewstatisticaloptimisationsoftwareincludingacostfunctiontofindthecheapest and most reliable mode of analysis by subsampling. It has to be outlined that the EC recommendation for sampling is also based on such subsampling strategy and thus could be adaptedforusingtheOPACSAcostfunctionandoptimisation. Incertaincasesofcoexistenceitisalsoimportanttodetermine,beforeharvestinginthefield, the level of adventitious presence of GMOs in a nonGMO field. Based on the predictions of spatial variability of outcrossing rate, different sampling schemes were developed and validated. After an initial work focusing on the 0.9% labelling threshold, new work has been startedfora0.1%level. Generalcontrolplansshouldbeundertakenwhereseveralanalytescouldbesampled,withlow cost sampling methodologies.  In this regard, the current sampling methodologies for mycotoxins(themoreheterogeneouslydistributedanalyteinalot)couldfulfiltherequisiteofa representativesamplingalsoforGMOsandderivedproducts.Animportantexperimentalwork iscurrentlyunderwaytotestthisassumption. Models have been developed by CoExtra to assess the effect of various variables on the GM adventitiouspresenceinnonGMbatchesandtheprobabilityofcomplianceofnonGMbatches withagiventhreshold,ateachstepofsupplychain(fromthefieldleveltotheenduser). TheexaminationofseveraldatasetsofresultsofthemeasurementofthequantityofGMOsin flourbyPCRbasedmethodscollectedthroughinterlaboratorystudiesshowedthattheuseof the lognormal transformation is necessary to correctly estimate measurement uncertainty of the whole detection process. Uncertainty Profiles built from estimates of measurement uncertainty generallygive arangeof50to200%ofassignedconcentrations formaterialsthat containatleast1%GMO.Thisrangeof50to200%isconsistentwithEuropeanNetworkofGMO LaboratoriesandtheEUCommunityReferenceLaboratory(ENGLandCRL)validationcriteriaand canbeusedasafitnessforpurposecriterionformeasurementmethods.Theeffectofthison  16 JointResearchCenteroftheEuropeanCommission(Geel,BelgiumandIspra,Italy) 17 EuropeanNetworkofGMOLaboratories. 18 http://bgmo.jrc.ec.europa.eu/home/sampling_KeLDA.htm 86 CoExtraInternationalConference x theenforcementofEUlabellingregulationsisthat,ingeneral,analyticalresultsneedtobeless than0.45%todemonstratecomplianceandgreaterthan1.8%todemonstratenoncompliance withalabellingthresholdof0.9%.TheseresultsexplaintheobservationmadeinCoExtrathat companies involved in the food and feed supply chains are using a contractual practical thresholdsaround0.1%forcomplyingwiththeEuropeanlabellingthresholdforGMOs,whichis setat0.9%. Within the project a framework for the analysis of control plans, defined as a test procedure combinedwithasampleacceptancelimit,hasbeendevelopedinordertoenablestakeholders tomakeobjectivechoicesabouttheeffortthatshouldbeputintosamplingandtestinginorder to make objective choices of sampling and testing strategies. The main factors that can affect the reliability are the GMO heterogeneous distribution in the lot and the effect of analytical uncertainty. TheuseofGMOissubjectedtolegalconstraints,eitherwithinbya“deregulation”system(e.g.USA) or an authorizing (e.g. EU) framework. To assess compliance with national and international requirements there is a continuous and increasing need for reliable and cost and timeeffective analyticalmethodsinallareasofanalysis. The reliability of a method is first determined by the validation process, which is the procedure providingevidenceofsuitabilityofananalyticalmethodforitsintendedpurpose.Alllaboratoriesin charge of GMO detection are working under a quality system within an accreditation scheme for which the compliance of the laboratories’ measurement uncertainties (repeatability and reproducibility)withthoseobtainedinvalidatedmethodismandatory.Accordingly,thevalidationof analyticalmethodsandtheimplementationofthevalidationprocess,havebeenkeygoalswithinthe CoExtraproject. x Withinmethodvalidationtwobasicconceptsareprevalent: x x x theglobalapproach,prevalentintheUSAandinotherdetectionareas,inwhichthewhole process from the product to the final measurement outcome is to be validated as a whole and the “modular approach”, in which  the analytical methods are considered as separate “module”([sub]sampling,homogenization,analyteextraction,etc)andeachofthesecanbe validated independently. As such, the “modular approach” provides a good basis for developingacosteffectivevalidation processbythestakeholderandforitsfurtherflexible implementationinroutinelaboratories.Forthispurposeperformancecriteriaandstatistical evaluation tools (such as AMPE: Analytical Method Performance Evaluation software and 'DecisionSupportSystem')havebeenproposed. ThenumberofGMcropsworldwideisincreasingcontinuouslyandacorrespondingincreasingof approved and non approved GMO is an obvious need for screening tools19 for simultaneous detectionofdifferentGMOsinasampleinonestep.TheDualChip®GMOmicroarrayisanovel multiplex screening method for the detection and identification of GMO, based on the use of multiplex PCR followed by hybridization on a microarray. The validation of this novel method wasperformedwithintheframeworkofCoExtra,accordingtoISO5725standard.Furthermore due to the complex nature of a microarray experiment results in many potential sources of variability,afuzzylogicvalidationbasedapproach wassuccessfullyapplied totheanalysisand data interpretation of the chip validation exercise. This microarray can be used not only for GMO screening and identification but also, by using its software using the “matrix approach”, permitssuspicionofthepresenceofunexpectedGMO(generallyunapprovedintheEU).   19 DetectiontargetspresentinseveralGMOs. 87 CoExtraInternationalConference The current legal frame resulted in the establishment of the CRL20 for validating GMOs notifiers’ identificationmethodsofGMOswiththesupportofENGL21.However,thecurrentmandateofthe CRL is restricted to the identification quantitative uniplex PCR22 methods provided by the notifiers while the routine laboratories are also using screening methods and attempting to decreases the analyticalcostsandanalysisdurationbymultiplexingthePCR. Eventthough,mostoftheanalyticalcontrolsaremadeonrawproductsandthatthedocumentary traceability is mostly used for the remaining of the supply chains,  the analytical traceability may impactthecostsandtime(importantforinstancewhendownloadingashipmentbeforeentranceof theproductsintotheEU)ofcontrollingtheGMandnonGMproducts,bytheirdevelopmentcosts, routine use, discrepancies between laboratories or finally by their implementation in accredited laboratories.Thusseveralimprovementstowardsmoreefficientandeffectiveanalyticaltraceability weremadeintheframeofCoExtra: x x x x Developmentofscreeningmethods,asopposedtotheeventspecific23methods,isnotcovered by EU legislation and it represents an additional burden to analytical laboratories. To improve GMOcoveragebythescreeningstep ofanalysis,newscreeningmethodsweredevelopedand arereadyforimplementation,even thoughvalidationhasyet to beputin place.Torecognize false positive results in screening step of detection, methods to detect sequence donor organismsweredeveloped,e.g.methodfordetectionofFigwortmosaicvirus,donorofPFMV sequenceintroducedintoseveralcommercialGMcrops. ToimprovecostandtimeeffectivenessofGMOdetectionseveralmethodsweredevelopedin multiplexformat.MostarequantitativerealtimePCRbased,butsomealternativesystemsfor detection of PCR produced products were also tackled; such as capillary gel electrophoresis (CGE) which separates and identifies PCR products based on length and fluorescent tag. Two realtime PCR multiplex systems are already available as commercial kits andpentaplex24 PCR CGEmethodforidentificationof4mostcommonGMmaizelinesisbeingfullyvalidatedwithin theproject.Also,aduplexsystemforonsitedetectionandquantificationofGT73oilseedrapeis readyforusebythecontrollaboratories. A lot of effort was put into improvements of performance in GMO detection. SIMQUANT, a ‘mostprobablenumber25’statisticscombinedwithrealtimePCRapproach,wasdevelopedand isshowingupto100foldimprovementinlimitofquantification.Togetherwithnewprotocols forDNAextractionfromhighlyprocessedsamples,theefficientcontrolofGMOpresenceisnow possibleinmostoftheprocessedsoybeanlecithinsandoils. NotifiersareprovidingtheCRLwithnumeroustaxon26identificationmethodstobeusedforthe relative GMO content quantification. The reliability of the relative GMO quantification (expressedin%ofGMOcontent)wasaddressedthroughdetailedevaluationofcharacteristics of reference gene methods. Besides strong relationships with EuropaBio27 for harmonizing the taxa reference genes, a guidance document on how to develop and appropriately test new referencegenemethodswasprepared.Similarly,asolutionforindependent,cheapandreliable  20 CommunityReferenceLaboratory,JointResearchCenter,Ispra,Italy 21 EuropeanNetworkofGMOLaboratories,http://engl.jrc.ec.europa.eu/ 22 PolymeraseChainReaction,http://en.wikipedia.org/wiki/PCR 23 GMOidentificationmethod 24 Five(5)PCRinatube 25 Statisticsbasedtechniqueusedinmicrobiology 26 Generallyaspecieslikecorn.Canbealowertaxonomiclevelasforinstancesugarbeet. 27 Europeanassociationofbiotechcompanies. 88 CoExtraInternationalConference x x referencematerialweresearchedforandfoundintheformofplasmidsandgenomicDNA.To improve reliability of GMO quantification a guidance document on different options for systematicDNAqualitycontrolwasprepared.DifferentanalyticalapproachesofrealtimePCR resultsanddifferentquantitativerealtimePCRmachinesandrealtimechemistriesweretested tofindpossiblesourcesofbiasinGMOquantificationandtofacilitateimplementationinroutine laboratories of alternative, more costeffective, detection methods. They were all found to be minor,comparedtothebiasintroducedthroughlessreliablereferencegenemethodsandthe effectoflowDNAquality. Several nonPCR based approaches were also evaluated within CoExtra to check for their performance.Amongseveraltested,loopmediatedisothermalamplification(LAMP)combined with Bioluminescent Assay in RealTime (BART) detection system is promising system for potentialbroaderuseinGMOdetectionitssensitivityandquantificationissimilartoPCR,butis lesssensitivetoinhibitors,cheaper.Themachineforonsitedetectionisavailable.Analternative forimplementingonsitedetection,suchascooperatives,wasalsosuccessfullystudied. During the process of focus groups with stakeholders, the question of how to deal with “botanicalimpurities28”wasraised.Adocumentwasestablishedsummarizingallourknowledge. Thereisunfortunatelynoeasilyapplicabletechnicalalternativetothemicroscopiccountingof representative subsamples. Accordingly, the current practices of adding nonGM products of suchabotanicalimpurity,shouldcontinue,eventhoughratherexpensive. Theincreasingnumber,diversityandcomplexityofGMOsauthorisedwithinand/oroutsidetheEU callsformorerationalstrategiestobeappliedforGMOdetection.Stackingofadded“effect”genes (traits)andpossiblepresenceofunauthorisedGMOsposetwoparticularchallengesinthiscontext. Within CoExtra, several new multiplex methods, detection technologies and strategies have been developedinresponsetothis. x x x Efficient screening based on the “matrix approach”, whose concepts were defined in the previousECFP5GMOchips29researchproject,canbeusedtoidentifythemostlikelysourcesof observedtransgenicmaterialinasample.The“matrixapproach”ishighlyflexible,asindividual screeningmodulesmaybeaddedorsubstituted,dependingonneeds,availabilityandvalidityof modules. Furthermore, both protein and DNA based analytical methods can be exploited with the“matrixapproach”. Multiplex30 detection was mainly achieved through development of well performing oligoplex amplificationreactions(210PCRtargetsamplifiedsimultaneously),wheretheamplifiedtargets were successively pooled and identified simultaneously, e.g. by array hybridisation or colorimetriccapillaryelectrophoresis.Thisstrategyincreasesflexibilitycomparedtomultiplexing at the amplification step, because target interference (DNA sequences) is mainly a problem duringamplification.Expandingthediversityoftargetsthatcanbedetectedinamultiplexassay ismucheasierwhenoptimisationcanbefocusedonoligoplexamplificationmodulesratherthan onamorecomplexmultiplexreaction. UnauthorisedGMOshavebeenobservedseveraltimeswithinandoutsidetheEU.Areviewof the sources and the legal status of various types of unauthorised GMOs and a proposal for a terminologyfortheirclassificationwereproducedinCoExtra.Detectionofsomeunauthorised GMOsmaybeachievedwiththe“matrixapproach”withinthesamescreeningstrategythatmay be applied for routine GMO testing, depending on the specific screening modules applied and the diversity of GMOs in the sample. Another method usable in routine for detecting the unapproved GMOs, the differential quantitative PCR, was also developed. Its main interest is  28 Forinstance:1kgofGMsoybeanina40000metrictonsshipmentofnonGMcorn. 29 http://www.bats.ch/gmochips/contact/index.html 30 SeveralPCRcarriedoutinthesametube,asopposedtouniplexPCR(1PCRpertube). 89 CoExtraInternationalConference x thatthisdetectionusesonlyastatisticaltestonroutinelyused(screeningversusidentification) detectionmethoddata.ThismethodiscurrentlyundervalidationthroughaISO5725organised ringtrial.ThesetwomaindetectionmethodstobeusedfordetectingEUunapprovedGMOs,are accessibletoroutinelaboratories.However,otherunauthorisedorunknownGMOsmayrequire more sophisticated technologies that were also explored within CoExtra, such as analysis of total genomic DNA on high density microarrays without selective amplification or high throughputmRNAsequencing. Stackingof“effect”genes(traits)hasbecomeincreasinglypopularoverthelastfewyears(see its impact on relative, DNA unit based, GMO content above). As a consequence, identification and quantification of GMOs may become less accurate and this in turn may affect the legal compliance of a food or feed product (single GMOs may be EU approved while their stacked counterpart may be not). How gene stacking can be defined and achieved, and its various implications including some legal implications were reviewed in CoExtra. Some proposals for terminology and solutions to cope with the challenges posed by gene stacking were also presented in the review. A statistics based detection method is currently proposed, however withaprobablehigheranalysiscost. Thepolymerasechainreaction(PCR)hasseverallimitations,suchastheneedforspecificprimers, limitedpotentialformultiplexingandneedforthermalcyclers.Alternativetechnologiesthatdonot dependontheuseofPCRwereexploredwithinCoExtra. x x x Multiple displacement (MD) amplification is an isothermal amplification method that may be usedtocreatelargequantitiesofasampleDNA,e.g.forpreparationofreferencematerialfrom limited source material, or to reduce the interference of impurities and DNA damage on microarray hybridisation. CoExtra showed that, because the MD amplification may introduce some bias, i.e. alter the relative copy number ratio of various DNA sequence motifs, it should thusnotbeusedforthepreparationof(reference)materialsforquantitativeanalyses. NASBA31implementedmicroarrayanalysis(NAIMA)combinestheisothermalNASBAtechnique withmultiplexing,potentiallyresultinginsimultaneousamplificationofmultipletargetsthatcan beidentifiedsubsequently,e.g.viamicroarrayhybridisation. TheinterestofdirectanalysisofgenomicDNAviamicroarrayhybridisationwasdemonstrated withoutprioramplificationorwithMDamplificationofthegenomicDNA.Themainadvantage ofthisstrategyisthatthenumberoftargetsthatcanbeanalysedsimultaneouslyisextremely high (> 105), and that very few assumptions need to be made regarding the target sequence priortoanalysis.SuchstrategymightthusbeusedfordetectingEUunapprovedGMOs. TargetspecificbiascouldhavesevereimpactonthereliabilityofGMOanalyses.CoExtratherefore investigated possible pre and postharvest sources of target specific bias. Preharvest sources of bias included the frequency and location ofsubstitutions and insertions/deletions in selected DNA sequence motifs targeted in GMO analyses. Postharvest sources of bias included a number of physicalandchemicalprocessingsuchasheating,lowpHandUVlight.Theresultsindicatethatbias can be a problem for some product types. For these types of products it is proposed that control reactionsareperformedtoassessifbiasislikelyandtodeterminetherangeanddirectionofbias. Notably, bias may be more pronounced with some than with other analytical modules (DNA extraction protocols and specific PCR assays). The modular approach for GMO analysis, which was subject to study in the WP4 part of the CoExtra project, therefore requires that possible bias is covered in validation of the analytical modules. A strategy to implement this was developed in collaborationbetweentwoworkpackagesofCoExtra.   31 http://en.wikipedia.org/wiki/NASBA_(molecular_biology) 90 CoExtraInternationalConference Legalandpolicyissues CoExtrawasattemptingtoaddresstheissuesofstakeholdersnotonlyfromatechnicaloreconomic point of view but also from a legal point of view, taking regard not only of generally applicable regulationsgoverningGMOapprovalanduse,butalsoofcontractualmodificationsthereof. Thecoexistenceprojectisanewmodalityofgovernmentoftechniques;itisparticularlyimportant concerningnewtechnologieswhichuntilnowhavebeenmanagedonlyinreferencetopotentialor provenrisks.Thishasmeantthatittendstopreventtheinvoluntaryspreadoftechnologycausing theeliminationofothertechnologies. Thisgovernmentoftechniques’modalitycouldbelinkedtoanobjectiveoftechnologicalpluralism such as the “energy mix”, which could be useful regarding nanotechnologies for example.  The projectitselfisdifficulttocarryout;itisevenhardertofindtheproperrulestomakeitsustainable. CoExtra shows how European authorities have reached this solution aiming at ending the crisis generated by the public’s distrust regarding GMOs food and feed. A CoExtra study analyses the threegovernmentmodalitiesthathavebeentriedouttothisday:the“LawoftheAlliance”which designatesasuppleregulationconceivedbyexperts,industryandadministration;“Lawasseenby theRulers”,representedbythe90/220directive,basedonrisksassessmentwithoutmanagingfarm produced products’ supply chains; the “Law as seen by the ruled”, implemented by the 2003 (1829/03and1830/03)regulatorypackage. Itisfinallyprovedthatcoexistenceisa“moreindepth”formoftraditionalfreedomofcommerce and industry; it lies on aparadox: to insure all a certain freedom, it is necessary to impose strong constraintsandacertainmutualtolerance. Accordingly,CoExtraisconsideringimportant: x x x x x x To officialise the technological pluralism as a global project allowing the reconciliation of knowledgesocietyandrisksocietybythepromotionofamechanisminsuringpublicconfidence. Toconceiverulessothatthispluralismbesustainable. Thecoexistencestrategiesmustfromnowonbethoughtoffromthesupplychainlevelandnot onlyfromfieldcoexistence(presentregulation). Itisessentialtoinsureabetterdistributionofsupplychains’segregationcostsbyestablishinga mainprinciple;thoseintroducinganewtechnologywilltakeinchargethecostsofsegregation fromthefieldtotheconsumer(Neighbourhooddisturbancestheory). Itisimportanttoquicklysolvethequestionofvarioustypesofunknownorunauthorisedevents. Concerning seeds, it is important to quickly solve the matters of 1) the question of fortuitous presentthreshold2)theoneofthefarmer’srighttouse«farmsavedseeds»buttheseseeds risk having an increasing level of unwanted GMOs in some species. 3) the question of the availability of conventional seeds which have been the object of a  traditional technology of plantbreedingtobenefitfromgeneticprogress. Assciencehasbecomeagrowingfoundationofdecisionmaking,disputesmoreandmoreariseon the scientific basis of such decisions, at least when they deal with environmental or health issues. Whatisthequalityofthescientificreportsonwhichthedisputeddecisionrests?Doesthepresent state of scientific knowledge justify this decision? Have all relevant scientific data been taken into account?Wasn’tthepreviousscientificassessmenttooabbreviated? A CoExtra deliverable gives elements in order to better understand and manage these new and decisiveaspectsofriskdecisionmaking.  91 CoExtraInternationalConference Twomainrecommendationsareformulatedwhichcanhavedirectimpactoncoexistencematters. x x Asriskdecisionsaremoreandmoresubmittedtocourts(national,Europeanandinternational), itisofutmostimportancetohaveaclearvisionofwhatisrequiredbythejudgesintermsofrisk assessment. As the judge’s role visàvis science is growing, courts endorse a more disputed role of "arbitrator of good scientific reports", which raises deep stakes that need to be correctly understood.  Aboutliabilityandredressmechanisms: x x x x The legal framework affecting coexistence and traceability was analyzed from various perspectives. European, nonEuropean and international approaches to regulating biotechnology in the food and feed supply chain were compared, including contractual duties and possible liability issues that may arise. Complications arise in particular in international settings with differing national systems, and such problems are aggravated by the fact that market participants may develop overlapping contractual regimes deviating further, even though it may be easier for vertically integrated companies. It shows the unifying effect of EU lawsonasideandofprivatestandardsontheotherside. Whileitisstillunclearhowlossescausedtothirdpartieswillberesolved,particularlyincross bordercases,thesolutionsofferedbyeachcountry’slawsarestronglyinfluencedbyitspolitical attitudetowardsGMfarmingingeneral,andmayamounttoadefactoobstaclethereto. The survey of legal, technical and political issues arising from coexistence and traceability in third countries identified some examples of workable systems and best practices that EU MemberStatesmayusewhenimplementingcoexistenceandtraceabilityrules. Theanalysesclearlyshowedlargediversityintheextenttowhichthirdcountriesareconsidering introducingorinfactimplementingcoexistencemeasures,i.e.tomaintainthreesupplychains. For candidate countries especially, a workable and reliable EU model would be highly appreciated.  CoExtracommunicationwithstakeholders StakeholderopinionsandattitudesoncoexistenceofGMOswithconventionalandorganicsupply chains Mainoutcomesofnationalstakeholderworkshopsandonlinesurveys: Seven stakeholder workshops were organised on the issue of coexistence in seven EU countries, andanonlinequestionnairewaslaunchedtosurveythegeneralattitudesandopinionstowardsco existence. Among a broad spectrum of attitudes and information needs of stakeholders the followingarethemostdominant: x x x ThereisanoverwhelmingwishtohavetheGMlabellingthresholdsforseedsregulated.Thisis overdifferentcountriesanddifferentstakeholders.Withoutthesethresholdsitisdifficulttoset practicalcoexistencemeasures. Thereisageneralconvictionandconcernaboutthecoststhatcoexistenceregimeswillentailin practice.Moststakeholdersareoftheopinionthatcoexistencemeasureswillentailcosts–as anyregulationwillentailcosts–butthereisdifferenceofopiniononhowsignificantthesecosts willbe. Thereisaconcernaboutthepracticalitiesofsamplingandtestingstrategies.Guidancemaybe necessaryhere,andperhapsalsoadiscussiononwhethertestingisnecessaryinallsituations,or thatinmanysituationssamplingwilldo,followedbytestingifaproblemhasarisen. 92 CoExtraInternationalConference x x x x x x A common concern on how to deal with unauthorized events. Nobody would like to be confrontedwithanunauthorizedevent–especiallyonethatisnotauthorizedanywhereinthe world – and there are questions on whether it is possible to prevent contamination with such eventsatalltimes. EspeciallyfromthesideoftheNGOsandorganicfarmers:adiscussiononthelegalmeaningof theconceptsof‘adventitious’and‘technicallyunavoidable’.Thereisgeneralrecognitionofthe fact that the 0.9% is a labelling threshold. But there is difference of opinion on what the consequences of these concepts are for the design of coexistence measures. What should practicalcoexistencemeasuresbeaimingat? Moststakeholdersarenotsupportersofahybridregulatorymodelwithcoexistencerulesboth ontheEuropeanandthecountrylevel,butsomemaystresstheneedforflexibility,especiallyon thepracticallevel. Many stakeholders recommend to monitoring the development of practical coexistence measures and compensation schemes in the different EU member states, with an eye on harmonization and the prevention of competitive advantages and disadvantages for particular farmers. Farmers are inclined to see coexistence regulatory frameworks as yet another set of requirementsthatwillincreasetheamountofpaperworkthattheyhavetodo.Theyarenotin favourofhavingtobecertifiedorlicensedtobeabletogrowGMcrops. Thequestionnairealsoshowsthatalthoughcoexistenceisaneconomicandchoiceissue,some stakeholders perceive, present or use it as an environmental or social issue, especially those stakeholdershavingamorenegativeopinionaboutGMOs.  CoExtradataintegration NumerousdataareissuedfromCoExtraworkandthuscanonlywithdifficultybemadeavailableto thestakeholders,orthecontrolroutinelaboratories.AccordinglyalargepartoftheCoExtrawork wasdedicatedtotheintegrationofdataintoatoolrathermoreeasilyusablebystakeholders.This 32 workwasfocusedontoaquiteuserfriendlyDSS . The outcomes of CoExtra provide a whole range of stakeholders: farmers, EU policy makers, importers,transporters,feed/foodproducers,retailers,consumers,analyticallaboratories,usersof test reports from analytical laboratories, operators and managers of official control with science based,readytouseinformation. The CoExtra Decision Support System integrates some results of the CoExtra project (such as collecteddata,scientificfindings,obtainedknowledgeandexpertise,formulatedrecommendations, developed methods and models, etc.) in a way that is potentially useful for different types of stakeholders. TheDSSprovidesdataandadviceforvariousdecisionquestionsthatoccurinsupplychainsinvolving GMOs,forinstance: x x x Will my (intermediary) product, given a current set of used procedures and materials, containGMOsbelowaspecifiedthresholdlevel? Isthereanypossibilitythatmy(intermediary)productcontainsunapprovedGMOs? Which methods perform best or can be used at all for a given analytical or sampling purpose?  32 DecisionSupportSystem 93 CoExtraInternationalConference x What are the costs associated with maintaining GMO content below some specified threshold? We are using the approach of modelbased DSS. In collaboration between experts and decision analysts,wecreatequalitativemodelsthat: x x x capture and represent expert knowledge in the form of hierarchically structured variables and decisionrules, areabletoassessandevaluatedecisionalternatives,and provide decisionanalytical tools to analyze these alternatives (for instance, finding the advantagesanddisadvantagesofalternatives,andanalyzingtheeffectsofchangesby“whatif” andsensitivityanalysis). Currently,therearesixmodelsimplementedorunderdevelopment: x x x x x AnalyticalModel:aimedattheassessmentofanalyticalmethods,includingDNAextractionand DNAanalysismethods; SamplingModel:assessmentofsamplingplans; Unapproved GM Model: assessing the risk of contamination with unauthorized GMO varieties based on traceability data about the product (for instance, type of product, country of origin, typeandmodeoftransportation); Transportation Model: assessment of potential GM presence due to transportation based on producttraceabilitydata; Dryer and Starch Models: assessing the effect of control parameters (such as using different strategiesforhandlingGMandnonGMbatches)tothecollectionandprocessingofmaize. All together these modules are currently prevalidated by CoExtra partners. A second step of validationshouldbestartedassoonaspossiblewithENGLmembersandsomestakeholdersbefore anyrelease.  Conclusion CoExtraisthelargestECgrantedprojectoncoexistenceandtraceabilityofGMandnonGMsupply chains. CoExtrafocusedonGMOandnonGMOsupplychains.Butthenumberofsupplychainssusceptible ofbeingimplicatedispotentiallyunlimited,evenifverysmalltoday.Theywillallbedifferentfrom oneanother.Itisthereforeimpossibletohaveanexhaustivecount.Asamatterofafact,traceability isthesegregationtool,whichitselfisthetoolforcoexistence.Traceabilityhasbeenstudiedforwhat itis,acomplexregulation,butalsoforitseconomicalandsocialfunction:allowingtrusttoestablish itselfamongactivitiessuspectedforpresentingrisks,rightly orwrongly.Wehereshowthat,atthe intersection of knowledge society and risk society,  juridical systems are trying to establish a confidencesocietytobethelinkbetweentheothertwo. Having as an aim to develop practical implementation of the techniques developed, CoExtra was the first attempt to take into account the several stakeholders’ practices, from seeds to shelves, through consumers’ survey, companies interviews and stakeholders’ focus groups. CoExtra first apprehendedthecurrentpracticesintheEUandthirdcountries,thebottlenecksandthenproposed solutions.CoExtradescribedthusprocesses,developedmodelsandtestedstrategies. Besides experimental work, economic and e.g. pollen flow modelling, whose information can be usedforoptimisingsegregationstrategiesdownstream,CoExtrahasreleasednumeroustechnical andlegalresultsallaimingtofavourcoexistenceandtraceabilityatthelowestcost. 94 CoExtraInternationalConference Suchconsiderationofbothcoexistenceandtraceabilityandtheirrespectiveimpactshasbeentaken intoconsiderationforthefirsttimeinaEuropeanresearchprogramdevotedtothecoexistenceof GMandnonGMproducts. CoExtrahasalsodevelopednewdetectionstrategiessuchasfordetectingstackedorunapproved GMOs. Due to the large number of questions CoExtra embraced, a Decision Support System has beendevelopedtointegratethosedataandfacilitatetheirusebystakeholdersincludinglaboratory analysts.Itsfullvalidationstillremainstobecarriedoutafterthecurrentprevalidation. Someissues,suchashowtodealwith“botanicalimpurities”inroutineanalyses,arehoweverstill pending. Wecanoutlinetheconfirmationthatstakeholdersareusingapracticalthreshold(generallyat0.1%) wellbelowthe0.9%Europeanlabellingthreshold,asusedinotherareashavingasafetyorquality threshold.Thisobservationofthetruelife,ofthereality,ofthedaytodaystakeholderspractices, showsthatthecoexistencebetweenfarmersispossibleonlybyusinglargedistanceofisolationor production(GMornonGMproducts)dedicatedareas,asdeterminedbythemodelsdevelopedin ECfunded SIGMEA project. The technical and legal definitions of such production dedicated areas remain to be done. Biocontainment methods can be helpful but this depends on their rapid commercialavailabilitywhenprovedtobestableandeffective. Generallyspeaking,themethods,strategies,tools,modelsdevelopedinCoExtraforGMandnon GMsupplychainscoexistenceandtraceabilitywillbeusedinthemanagementofnumerousother supplychains,valueaddedornotnichemarkets,harmfulproductssuchasallergensandmycotoxins producingorganismsorpathogens. Thus again a GMO based work provides a good costbenefit ratio, as previously done for instance with PCR applied to the whole supply chains in 1999 or standardization of PCR requirements, for developingsaferandbetterfoodandfeedsupplychains. As for the former FP5 research programs, such as QPCRGMOFOOD33 and GMOchips34, we could expect that CoExtra would have a rather important impact not only on the national and EU legislativeframesbutalsoonsupplychainsmanagement.    33 http://www.vetinst.no/eng/Research/EU-projects/QPCRGMOFOOD  34 http://www.bats.ch/gmochips/  95 CoExtraInternationalConference  PosterAbstracts   P1. AcosteffectiveP35S/Tnosmultiplexscreeningassaywithinternal positivecontrol S.Baeumler,B.Gibfried,M.TranzerandD.Wulff, EurofinsGenescan,[email protected]  AlthoughRealTimePCRmethodsfortheeventspecificidentificationofmostcommerciallyrelevant GMOs are available today, “traditional” screening methods are by far not outdated. In contrast, againstthebackgroundofagrowingnumberofapprovedandcommercializedGMOsthesearestill very valuable tools for a costefficient initial screening step and an indispensible prerequisite for economically bearable analytical strategies. Examples for widely used screening targets are p35S and tnos. As an improvement of existing screening systems a robust and userfriendly multiplex assayhasbeendevelopedfortheseimportantscreeningtargets.Specificallythefeatureofabuiltin “IPC” amplification control (Internal Positive Control) constitutes a significant improvement of qualityasitprovidesacostefficientinhibitioncontrolforhighlyreliableexclusionoffalsenegative results which is specifically important under routine conditions where impurities in sample DNA can’t be eliminated completely. In addition the new format saves consumables and allows higher throughputonlimitedPCRinstrumentcapacity.  A programmed spreadsheet tool facilitates automated data evaluation based on precisely defined acceptance algorithms and eliminates individual and thus subjective evaluation of results largely. This new p35S/tnos triplex assay format however  due to its complexity  requires specific and stringentvalidationinordertoproofitsperformance:Inadditionto“classical”methodparameters ofsimplexmethodslikespecificity,sensitivityorrobustness,furthermultiplexspecificperformance parameters had to be thoroughly addressed. Most importantly noninterference and non competition of the different PCR systems, as well as sensitivity under strongly asymmetric target concentrations had to be validated. Experiments were carried out on ABI 7500 SDS, Stratagene Mx3005P and BioRad iQ5 because multiplex assays call for instrument specific validation. The systemprovedtoperformtospecificationsonallplatformstested.  Acknowledgements: ThisstudywasfinanciallysupportedbytheEuropeanCommissionthroughtheIntegratedProject CoExtra,ContractNo.007158,underthe6thFrameworkProgramme,priority5,foodqualityand safety. 96 CoExtraInternationalConference P2. Theproblemofwhentolabelinpresenceoflowamountsoftransgenic material:thecaseofbotanicalimpurities GilbertBerben1,FrédéricDebode1,EricJanssen1andYvesBertheau2 1 2 CRAW,DépartementQualitédesProductionsagricoles,5030Gembloux,Belgium INRA,Versailles,France  Within the CoExtra project a deliverable was dedicated to the problem of presence of botanical impuritiesinfeedmaterialwithrespecttoenforcementoftheGMlegislationlinkedtolabelling. Afirstsectionisdevotedtoexplainingthesourceoftheproblem,whichfundamentallyislinkedto incompatibilitiesbetweendifferentlegaltexts.Atafirsthandthereareregulatoryprovisionsmade about the requirement of purity, essentially of feed material [1, 2]. As long as the purity level is above95%(mindthattherearesomeexceptionstothisfigurewithsomewhatlowerrates)thefeed material is supposed as pure and there is no need to mention in a label the nature of botanical impuritieswhichbydefinitionareharmlesscomponentsinthefeed.Attheotherhandtherearethe regulatory provisions for GMO labelling if the tolerance level for adventitious and technically unavoidable contamination is exceeded [3]. The problem arises from the fact that for the latter tolerance threshold the unit is generally considered as expressed “per ingredient but grouping all thoseofasameplantspecies”andnotasisthecaseforpurityrequirementsoverthetotalweightof thefeedmaterial. AlthoughthetheoreticalsolutionofexpressingresultsofGMbotanicalimpuritiestowardsthewhole mass of the feed material would solve the problem from a legal viewpoint (such exceptions are foreseen,see[4]),weshowthatbythiswaytechnicalproblemsdoremain. BecausePCRisonlyabletoexpressacontentlevel“peringredient”,thefollowingformulashouldbe usedtotransformthisresultintoacontentexpressedasamassfractionofthefeedingredient: C GMBI 100 u m u c  With: CGMBIgivingin%theGMcontentofthebotanicalpurityintermsofmasstowardstheallmassofthe feedmaterial, m representing the mass ratio of the botanical impurity (i.e. 5,0% becomes 0.050) in the feed material, crepresentingthecontentinGMmaterialofthebotanicalimpurityasmeasuredbyPCR(i.e.1,0% becomes0.010),thusacontentperingredientandnottowardstheallmassofthefeed.  Asecondpartofthedeliverablehandlesthescientificpossibilitieswithrespecttothetermsusedin the above formula. A special focus is given to the techniques that are required for measuring the mass fraction m of the botanical impurity in the product. From the review of techniques made, it clearlyappearsthatwithpresentdaytechnologythereisatechnologicalgapinhowtoquantifywith a sufficient reliability the mass fraction of a botanical impurity in a product. Presently, classical optical microscopy is in fact the main technique used to assess m. The problem of unit type to considerforthedeterminationoftherelativeGMmaterialcontent(c)withinthebotanicalimpurity isalsoassessedwithitsimpactonthefinalresultoftheequationthatisconsideredasthesolution totheproblem.Thebestwayappearstobeworkingintherecommendedunitofcopy%perhaploid genomeequivalents[5]thatshouldthenbeconvertedthroughfixedconversionfactors(definedper speciesforinstance)intorelativemassfractions. 97 CoExtraInternationalConference A discussion also compares in how far these concepts applicable to feed are transposable to food and seeds. The situation in seeds is very comparable to that of feed materials. While in the food sector,theproblemishandledinatotallydifferentwaybecausepuritylevelsofingredientsarenot definedbylawbutthroughcontracts.  References: EuropeanCommission(1996).CouncilDirective96/25/EC. EuropeanCommission(1998).CouncilDirective98/67/EC. EuropeanCommission(2003).RegulationECNr.1829/2003. EuropeanCommission(2006).ReportCOM(2006)626 (http://eurlex.europa.eu/LexUriServ/LexUriServ.do?uri=COM:2006:0626:FIN:EN:PDF). EuropeanCommission(2004).CommissionRecommendation2004/787/EC.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.  98 CoExtraInternationalConference P3. NIRimagingandchemometricsinsupporttothedetectionatthesingle kernellevelofGMO J.A.FernándezPierna,E.Janssen,Ph.Vermeulen,G.Berben,P.DardenneandV.Baeten Walloon Agricultural Research Centre (CRAW), Quality Department of Agricultural products, Chaussée de Namur n°24, 5030Gembloux,Belgium.[email protected]  Since 2000, the Walloon Agricultural Research Centre (CRAW) has acquired expertise in the development of analytical methods based on NIR hyperspectral imaging for particles and single kernelanalysis.TheinstrumentusedisaMatrixNIR®ChemicalImagingSystem(Malverninstruments Ltd)recordingsequentialimageswithanInGaAsarraydetector(240x320pixels)activeinthe900 1700nmrange,thatmeans76800spectraperimage. In the framework of the CoExtra FP6 project (GM and nonGM supply chains: their COEXistence andTRAceability),theCRAWisinchargetoinvestigatethepotentialofNIRhyperspectralimaging together with chemometrics for GMO (Genetically modified organisms) detection. Soybean and barleysamplescomingfromdifferentoriginsandsomebeingtransgenichavebeenanalysedforthis purpose. TheaimistoproduceamethodologyinordertoinvestigatethepotentialofNIRimagingtogether withchemometricsforGMO(Geneticallymodifiedorganisms)detection.Thedatatreatmentofthe spectraldatacollectedcorrespondstounsupervised(PCA)andsupervised(PLSDA)techniques.Inall data sets the results have shown that a good discrimination could be performed according to the varietyandthepresenceofGM.Howeverwithbarleyitisimpossibletodifferentiatethetransgenic lines from the nontransgenic ones when a large diversity of varieties of different origins are considered. From the pattern recognition point of view, more interesting approaches in order to make estimations of the statistical properties based on the images combined with the spectral information has been identified. From the results obtained it appears that next to a merely a qualitative detection, there might be a potential to quantify the GM content in Roundup Ready soybeanatthekernellevel.Thereisatleastacorrelationbutmoreworkisbeingdonetodocument thiscompletelyandtoimprovethecorrelation.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.   99 CoExtraInternationalConference P4. PerformanceofTaqMan®,LNA,CyclingProbeTechnology,Luxand PlexorrealtimePCRchemistriesinquantitativeGMOdetection MetiBuhGašpari1,2§,KatarinaCankar1,3,JanaŽel1,KristinaGruden1 1 Department of Biotechnology and Systems Biology, National Institute of Biology, Vena pot 111, SI1000 Ljubljana, Slovenia 2 PlantSystemsBiology,VIB,TechnologieparkZwijnaarde927,9052Gent,Belgium 3 PlantPhysiology,WageningenUR,Droevendaalsesteeg1,6708PBWageningen,theNetherlands  The realtime polymerase chain reaction is widely used for detecting and quantifying genetically modifiedcomponentsinfoodandfeed.Tocomplywithvariousapplicationsthenumberofdifferent QPCRdetectionchemistriesisincreasing,reachingmorethan20atthemoment.Vastmajorityof the laboratories still uses TaqMan or SYBR Green only and few comparisons were done about the alternative chemistries. In these study Lux, Plexor, Cycling Probe Technology (CPT) and LNA were extensively evaluated and compared using TaqMan chemistry as a reference system. It should howeverbetakenintoaccountthatbothTaqMan®methodswerealreadyproventoberobustand reliablethroughuseinroutineGMOdetection,whilethealternativemethodsweredevelopedand optimizedonlytothedegreedescribedinthisposter. Foreachchemistryampliconsweredesignedonthemaizeinvertaseasareferencegeneandonthe 5´junction in MON810 event. Assays were optimised and compared for their efficiency in PCR amplification, limits of detection and quantification, repeatability, accuracy and specificity. In addition the time investment and costs issues were evaluated. Even though each assay provided satisfactoryperformance,resultssuggestsomearemoresuitableforquantitativeanalysisthanthe others. Of the probe based methods, LNA® chemistry is the most promising, with excellent quantificationlimitsandefficiency.Verygoodrepeatability,evenforlowcopynumbers,isreflected inhighprecisionandaccuracyofmeasurements.LNA®methodscanbeeasilytransferredfromthe widely used and certified TaqMan® methods should this prove beneficial for some applications. BecauseLNA®probesaremuchshortertheycouldbeespeciallyappropriatewherehighspecificityis needed(e.g.onlyonenucleotidedifferenceinthesequence).Theyarealsolikelytobeusedwhere thesequencesaresuchthatthedesignofacommonTaqMan®probeisdifficultorevenimpossible, forexampleindetectingjunctionsbetweenGMinsertandplantDNA. Due to some performance characteristics it is not likely that Lux™ or Plexor™ chemistries would replace the probe based chemistries in the quantification of GMO content, especially for samples with multiple ingredients. With the probe absent, a perfect specificity is even harder to achieve, whichalsoshowedasslightcrossreactivenessinoneofPlexor™designs.Plexor™chemistryhowever performed well when considering LOD. In addition it was the most robust against inhibitory substancesofallthechemistriestestedandprovedpracticalforroutineuse.Webelievethatwith additional effort put in design of specific primers Plexor™ technology provides an appropriate and affordableapproachforqualitativeanalysis. Our results suggest that probe based TaqMan® and LNA® technologies are best for quantitative analysis. Primer based Plexor™ on the other hand could be the method of choice for qualitative analysisifappropriatelydesignedtoassurespecificityofthemethod.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety. 100 CoExtraInternationalConference P5. GMOanalysis:towardsassuringconfidenceinaresult MalcolmBurns LGCLaboratoryoftheGovernmentChemist,UK,[email protected]  TheenforcementofcurrentEUlegislationforthecorrectlabellingoffoodmaterialscontainingGM derivedingredients,dictatesthatallstakeholdersneedaccesstoaccuratetechnologiesandmethods inordertodeterminethecorrectlevelofGMingredientspresentinfoodsamples. The advent of modern molecular techniques, including high throughput realtime PCR and array based methods, have meant that a number of different and highly sensitive techniques are becoming increasingly available to conduct such GM analysis. However, whilst the technology behindGMdetectionhasbecomeincreasinglyadvanced,thereisaneedtostandardisemetrological aspectsassociatedwithmethodvalidationandexperimentaldesign,ifconfidenceistobeattributed toresultsarisingfromthesenewtechnologies. Theareaofstandardisationofdataanalysisandinterpretation,isoftenanoverlookedareaofthe analyticalapproach,andcancausesignificantmeasurementuncertaintyassociatedwithanalysisof GM. If results are to be interpreted correctly at both national and international level, confidence mustbeaffordedtothevaluesgiven,andastandardisedwaytoevaluatingdatafromGManalysis can be regarded as a critical step in helping facilitate this. The work presented here outlines two suchpublishedapproaches,designedtohelpaffordbetterconfidenceinresultsassociatedwithGM analyses.  LimitofDetection TheLimitofDetection(LOD)isacriticalperformancecharacteristicofanassaythatrequirescareful evaluationduringmethodvalidation.OneaccepteddefinitionfortheLODofanassayisbasedonthe mean value of blank determinations, plus a derivation of their standard deviation. However, this formalcalculationfortheLODdoesnottakeintoaccountatypicaldatasetsthataregeneratedfrom realtimePCRtechniques,whichcanbenonnormallydistributed,truncated,andheteroscedastic. The LOD can also be defined as the lowest amount of analyte that can be distinguished from a background response, on 95% of occasions. Experimental data for the quantitation of Genetically Modified (GM) material were produced using realtime PCR, in order to model the LOD. A bootstrappingcomputersimulationcalculatedtheprobabilitiesofdetectingPCRpositivetestresults from these data sets, and computer modelling defined a function from the resulting probability plots.TheLODwasmodelledinresponsetochangesinsamplereplicationlevelandcyclethreshold values. ThebootstrappinganddatamodellingapproachwasshowntoaccuratelypredicttheLODassociated withrealtimePCRanalyses,andtheapproach’sbroadapplicabilityshouldbeofgeneralinterestto laboratoriesconductingtraceleveldetection.  Samplereplication Thelevelofsamplereplicationwithinanyassayisafundamentalaspectthatneedstobeconsidered whenproducingresultswithhighconfidence.Anovelapproachwasusedtoevaluatetheoptimum numberofsamplereplicatestouseinGManalysis,usingrealtimePCRasamodelsystem.Thework modelled the change in precision associated with the estimation of GM content of sample unknowns, in response to changes in the level of replication associated with both calibrants and sample unknowns. Using an experimentally derived data set, it was shown that it was possible to reducethesamplelevelofreplicationfromsixtothreePCRreplicates,withoutasignificantchange 101 CoExtraInternationalConference inthemeanvalueorvariabilityoftheexpressedresult.Theuseofsuchanapproachcanfacilitate theuseoftheminimumnumberofreplicatesinordertoproduceanaccurateresult,thussavingon importantresourcesinvolvedinquantitationassays.  Conclusion Method validation, and the implementation of appropriate experimental designs to support such validation, are two fundamental principles used to provide objective evidence for the “fitness for purpose“ofamethodorresult.TheLODisacriticalperformancecharacteristictoevaluateduring method validation. A novel approach to evaluating the LOD has been published, that overcomes someofthelimitationsoftraditionaldefinitionsforLODwhichPCRdoesnotconformto. Inexperimentaldesign,aperpetualquestioniswhatlevelofreplicationshouldbeimplemented.The answerisoftenbasedonabalancebetweenaffordingconfidenceinaresult,andbeingcostefficient with sample throughput. A modelling approach has been published that shows a reduction in the levelofsamplereplicationdoesnotnecessarilycauseasignificantreductioninaccuracyassociated witharesult.  References: Modelling the Limit of Detection in realtime quantitative PCR. M. Burns and H. Valdivia. European Food Research and Technology(2008)226(6):15131524.DOI:10.1007/s002170070683z A simulation approach to assess the minimal number of realtime PCR replicates for GM quantification. M. Burns, H. Valdivia.EuropeanFoodResearchandTechnology(2008)Volume227(6):17211727.DOI:10.1007/s0021700808996  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.          102 CoExtraInternationalConference P6. DetectionofBacillusthuringiensisbyrealtimePCR DebodeFrédéric,JanssenEric,ArranzEsther,RoulezDenis,AncionCécile,AntoineGaëlle,HulinJulie andBerbenGilbert CRA–WDépartementQualitédesproductionsagricoles.24,chausséedeNamur5030Gembloux.Belgium  CoExtraprojectWP5workpackage Amongthegeneticmodificationsintroducedinplants,insecticidalprotectionisacommontrait.This ismadepossiblethankstotheintegrationofacrygene(isolatedfromBacillusthuringiensisstrains). Following the origin of the cry gene, this protection can specifically be oriented against lepidopterans,coleopteransormosquitosandsimulies. Bacillus strains and their spores are largely widespread in the environment. They can be found in soil, plants, food and animals. Some species can be present in all kinds of food and cause food poisoningandtoxiinfections(somespeciesareextremelypathogens). TheaimofthisstudyistodevelopaRealTimePCRtestforBacillusthuringiensisinordertodetecta falsepositivescreeningresultduetothepossiblepresenceofthebacterium. Beforethesettingupofthetests,itwasimportanttotakeintoaccountthattheuniquecriteriumto distinguish the Bacillus thuringiensis strains from other Bacillus strains is the presence of a parasporal crystal inclusion (protein) that appears during sporulation . Some sequences ofBacillus thuringiensiscanalsobeclosertootherBacillusstrainsthaninsidetheBacillusthuringiensiscluster. Moreover,theconjugationbetweenBacillusstrainsisnotlimitedtothesubspeciesbutispossible betweenspecies. EightcouplesofprimersandprobesweredesignedfordetectionofBacillusthuringiensis: x x x OnesystemtargetingtherpoCgene.ThiswasbasedonalignementofsequencesfromBacillus thuringiensis sp. kurstaki, sp. tolworthi, sp. thuringiensis, sv israelensis, alvei and mycoïdes (sequencesofrpoCgenesprovidedbyINRA,France); TwosystemsbasedongyrBgeneandestablishedonsequencesofYamadaetal(1999); Fivesystemsbasedoncrygenes.TheinformationavailableoncryIA(a),cryIA(b)andcryIA(c)was collectedanddifferentprimersandprobeswereselectedinregionssupposedasnotintroduced intransgenicplants Tests were done on different strains of Bacillus (Bacillus cereus, Bacillus mycoides, Bacillus weihenstephaniensis, Bacillus thuringiensis aizawai, Bacillus thuringiensis israelensis, Bacillus thuringiensis kurstaki and Bacillus subtilis), on commercial insecticides containing Bacillus thuringiensis aizawai, Bacillus thuringiensis israelensis or Bacillus thuringiensis kurstaki and on differentfoodandfeedproductsofthemarket. ResultsshowthatrpoCisnotthebesttarget,gyrBtargetcouldbeusefultodistinguishtheBacillus thuringiensis israelensis strains and that the best choice for detection of the Bacillus thuringiensis kurstaki strains is among the cry genes. The selection among the different targets will be possible oncetestswillbedoneonadditionalstrainsandcontaminatedsamples(laboratorymadesamples). Concerning the tests done on food and feed samples to determine if there was a constant signal observed due to the possible contamination of food and feed products by Bacillus strains, we observedthatmostfoodandfeedsamplesarenegative.Signalsarelateforambiguousorpositive samples.Themostpositivesample(hamburger)wasabadconservedsample.Thisindicatesthatthe riskofpositivesampleismorelocatedinbadconservedproducts(bacterialproliferation).  103 CoExtraInternationalConference Testswerealsodonetoknowifsomeexistingcrytargets(fordetectionofGMplants)giveapositive signalwiththeBacillusstrains.ThepatentedprimersofEppendorfArrayTechnology(EAT,Namur, Belgium)usedforcrydetectionbymicroarraysandofISP(Brussels,Belgium)usedforcrydetection byRealTimePCRwithSYBRGreen,theprimersdevelopedbyMatsuokaetal.forclassicalPCRand theprimersofCRAWfordetectioninRealTimePCRwithTaqMan®probes(developedwithinthe BelgianGMOdetecproject)weretestedonBacillusstrainsandcommercialBtinsecticides.Nosignal wasobserved.Thisindicatesthatexisting“transgenic”primersaredoingagooddistinctionbetween naturalandintegratedsequences.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.                           104 CoExtraInternationalConference P7. Developmentofanewprobeforqualitativeidentificationand quantificationofBt11maize G.CandanGurakan,HoumanJabbarifarhoudandRemziyeYlmaz 1 MiddleEastTechnicalUniversity,FoodEngineeringDepartment,06531Ankara,Turkey. MiddleEastTechnicalUniversity,BiotechnologyDepartment,06531Ankara,Turkey. 3 MiddleEastTechnicalUniversity,CentralLaboratory,MolecularBiologyandBiotechnologyR&DCenter,06531Ankara, Turkey.[email protected] 2 Inthisstudy,anewsystemconsistingofasetofnewprimers,aTaqManprobehasbeendeveloped for identification and quantification of Bt 11 maize. Additionally, this system including TaqMan® probewascomparedwiththatofincludingSYBRGreen™I.                      105 CoExtraInternationalConference P8. DevelopmentofconstructspecificTaqManrealtimePCRfordetection andquantificationoftransgenicBt11maize(Zeamays) G.CandanGurakan1,HoumanJabbarifaroud2andRemziyeYlmaz3 1 MiddleEastTechnicalUniversity,FoodEngineeringDepartment,06531Ankara,Turkey. MiddleEastTechnicalUniversity,BiotechnologyDepartment,06531Ankara,Turkey. 3 MiddleEastTechnicalUniversity,CentralLaboratory,MolecularBiologyandBiotechnologyR&DCenter,06531Ankara, Turkey. 2  QualitativeandquantitativePolymeraseChainReaction(PCR)techniquesareutilizedfordetection andquantificationofgeneticallymodifiedcropsinfoods.Manycountrieshaveissuedregulationsto labelfoodsthatincludegeneticallymodifiedcrops.PCRMethodsusedindetectionofGMproducts are required to amplify a specific target gene. Bt11 maize is a genetically modified corn and is resistant against lepidopteran insects. The Bt11 transformation event has been produced by using plasmid pZO1502 containing a truncated synthetic cry 1A(b) gene encoding Btk endotoxin. It also containsasyntheticpatgenetoallowtransformantselectiononglufosinateammonium.35SCaMV isthepromoter,nosterminationsequencesareincludedandintronsIVS2orIVS6areincorporated toenhanceexpression. TheobjectivesofthisstudyistoscreenGMmaizeinsamplesinTurkey,comparemethodsusedfor the quantification of Bt11 maize from few GM positive samples from the markets and to initialize quantification studies in Turkey by developing TaqMan probes, to decide cost effective, reliable method(analyzinghighernumbersofsamples)toquantifyGMmaizestartingfromBt11maize. Inthisstudy100cornsampleswerecollectedfrommarketsthoughTurkeyfrom2006to2008.The samples were screened for genetic modifications using 35S, nos, PAT, Bar. Then, end point PCR identificationsforBt11werecarriedoutbyvalidatedprimersandBt11CertifiedReferenceMaterials (CRMs)withGMOcontentsof0%,0.1%,0.5%,1%,2%and5%wereobtainedfromJRCIRMM.Bt11 maizewasdetectedin4samples. For quantification studies, corn kernels were grinded by electric blender and according to CRL validationreportCTABmethodwasappliedforDNAextractionfromtestsamples(100mg).DNAwas extracted by CTAB solution and after purification with chloroform it was precipitated out with isopropanol.DNAconcentrationsweresetto40ng/ulafterUVspectrophotometry.PCRtestswere conducted by primers specific for the maize zein gene to verify specificity and quality of DNA. Specific primers and corresponding probe labeled with 5’FAM and 3’TAMRA were designed for amplificationofa93bpfragmentoftheIVS6/Cry1A(b)junctionregionintheBt11geneconstruct. Considering the required conditions for designing of specific primers and probe for TaqMan real time PCR assay, online Primer3 software was used for designing primers and probe. The DNA and reaction reagent mixture with a total volume of 25 μl were incubated in PCR thermocycler (ABI 7500)underthefollowingprogram:initialdenaturationin95oCfor10minfollowedby40cyclesof amplification, each consisting of 95oC for 15 sec, 60oC for 30 sec and 72oC for 32 sec. Analyses of amplification curves were used for calculating of. Standard curves were set up on the bases of ThresholdcyclevaluesofBt11CertifiedReference Materials. GMOquantificationofBt11positive sampleswascalculatedbyalimitofquantificationoflessthan0.1%(w/w).Toconfirmtheresults, one more DNA extraction was performed for each Bt11 positive samples and three parallel qPCR applications were carried out for each extract. The results of parallel samples showed correlation witheachother. Furthermore,SYBRGreen™Irealtimeassayswereusedtoverifytheresultobtainedinthisstudy. SYBRGreenIrealtimePCRtechniqueswerealsoutilizedtoconfirmthequantificationresults.Inthis study,ItwasalsoconfirmedthattheBt11sampleswerenotBt10.Comparingtheoverallresults,it 106 CoExtraInternationalConference was concluded that primer and TaqMan® probe set developed in this study can be used as a functionalmethodfordetectionandquantificationofBt11maizeline.               107 CoExtraInternationalConference P9. Stateoftheartonsamplepreparationandassessingthevalidityof proceduresderivingtestportionfromlaboratorysamples Janssen1Eric,Debode1Frédéric,Onori2Roberta,Ancion1Cécile,Antoine1Gaëlle,ArranzRivera1 Esther,KayokaMukendi1Nicaise,Roulez1DenisandBerben1Gilbert 1 2 WalloonAgriculturalResearchCentre,Gembloux,Belgium. ItalianNationalInstituteforHealth,Roma,Italy.  One of the main objectives of this work is to provide “real world” and empirical data concerning sample preparation and procedures deriving test portions from laboratory samples on one hand, and confronting these data to standards and guidelines generally followed in the GMO detection fieldontheotherhand.Thisworkcouldhelptoimplementthesestandardsorguidelinesbutshould alsoprovideinformationtobetakenintoconsiderationinanydecisionsupportsystemusedinthe contextofGMOdetection,oranymodularanalyticalschemeusingDNAtargetsasanalytes. Thefirstpartofthisabstractpresentsabriefoverviewofthestateoftheartconcerningthesample preparation(especiallytestportion).Thestateoftheartonsamplepreparationhasbeencarriedout using a questionnaire distributed to CoExtra partners and ENGL members. The questionnaire was also extended to the Italian network of GMO laboratories and one Turkish laboratory (neither Co ExtrapartnersnorENGLmembers).Thirtysixlaboratoriesansweredthequestionnaire.Fourteenare CoExtrapartners,twentyfourareENGLmembers.TwelveofthemareCoExtrapartnersandENGL members.TenlaboratoriesareneitherCoExtrapartnersnorENGLmembers.Thepresentstateof the art focus on the test portions properties : size, number of replicates, final particles size. This abstractconcernsonlygrains(seedsandkernels). Concerningthetestportionsnumber,ISOStandardsonGMOdetectionrequire2testsportionsto perform two independent DNA extracts per sample. The study shows that this requirement is generally followed by laboratories (one laboratory performs three tests portions). For the test portion size, DNA extraction protocols described in ISO standards annexes are usually written for 200300mgtestsportionssizes.ISOstandardsrecommendnottoexceed2gtestportion.Mostof laboratoriesperform200mgto2gtestsportionsEightlaboratoriesperform100mgtestsportions. Onlyonelaboratoryperforms10gtestportions.Concerningfinalparticlessizes,ISOstandardsdon’t haveanypreciserequirementbutAFNORXPV03recommendstousea<0.5mmfinalparticlesize (as much as possible). This study shows that the final particle size is not always known. Thirteen laboratories reported an unknown final particles size. Nineteen laboratories reported a value (between 0.08 and 4.0 mm). Thirteen laboratories reported particles size d 0.5 mm. One lab reported 0.75 mm, one lab < 1.0 mm and one lab 4.0 mm. Other labs reported variable values (between0.3and1.0mm).Threelabsdidn’treportanyvalue.Onelabdoesn’tperformanalysison grains. Theseresultsshowsthat,whendataareavailable,ISOand/orAFNORstandardsonGMOdetection are quite well followed. However, the differences observed between some modalities of sample preparation modules (or submodules) could lead to different analytical performances (if it’s not demonstratedthatthecombinationofdifferentmodalitiesoftwoormoremodulesareequivalent intermsofperformancecriteria). The second part of this work consists in assessing the validity of procedures deriving test portions from laboratory samples, taking into consideration the main conclusions coming from the above stateoftheart,especiallyconcerningthetestportionsize.Theaimofthispartisfromamodular pointofview,toanswertothequestion:“isatestportionrepresentative–fromaquantitativepoint ofviewofalaboratorysample?”.Inotherwords“doesthetestportionsizeinfluencequantitative PCRresponseandthen,themeasurementuncertainty?”. 108 CoExtraInternationalConference  Tocarryoutthistask,soybeans(grains)havebeenchosenasamodelmatrix.Samples(containing ~1000 grains each) have been built by spiking non GM soybeans with RR soybeans. Seven levels (from0.1to1.8%inweight)weresetup.Foreachlevel,fivetestportionsizes(from50to800mg, taking into consideration ISO standards guidelines and the above state of the art) have been analysed. Firstanalysisofresultsshowsthat,fortherespectiverangesofbothGMcontentsandtestportion sizes,theinfluenceoftheGMcontentseemsmoreimportantthanthetestportionsize,especiallyin termsofvariabilityontheGMcontentmeasurement.Themajorpreliminaryconclusionofthiswork is that the different combinations “GM %test portion size” are fit for purpose (together with the severalanalyticalmodulesused)inthecontextofthepresentGMOregulationframework.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety. 109 CoExtraInternationalConference P10. DesigningthePCRmarkersAgrobacteriumtumefaciensgallforming strains. Spiridonova1Elizaveta,Bertheau2YvesandKuznetsov1*Boris. 1 CenterBioengineeringRAS,Pr.60letyaOktyabrya,71,Moscow,117312,Russia, 2 InstitutNationaldelaRechercheAgronomique(INRA),RD10RoutedeSaintCyrF78026,Versaillescedex,France *[email protected]  Abbreviations. DIRPCR – Diverged Interspersed Repeats PCR, SCARPCR  Sequence Characterized AmplifiedRegionPCR Agrobacterialplanttransformationisoneofthecommonmethodsinmodernplantbiotechnology. Thereareseveralproblemsassociatedwithbacterialstrainsinvolvedintransformationprocess.First ofall,theefficacyofanynewplantspeciestransformationstronglydependsupontheagrobacterial strainusedandthesearchofnewhighlyeffectivebacterialstrainfromnaturalhabitatsisaninstant process.Fastanddirectdetectionofcrowngallformingagrobacteriuminnaturalspecimencouldbe substantially facilitated with simple and reliable PCR assay. Second, there are special biosafety requirementsduringfieldtrialsofbiotechplantsaimedtopreventintrogressionofbacterialstrains usedfortransformationintonaturalhabitats.ThesensitiveandreliablePCRassayonthepresence oftransformationallyactiveAgrobacteriumtumefaciens(At)intheplantmaterialisquitenecessary forsuchtesting.TheaimofthisstudywasdesigningthePCRmarkerspecificforthetumefaciens(At) strains, testing its specificity within Rhizobium/Agrobacterium group and designing the combined duplex PCR assays for detection of crown gall forming A.tumefaciens strains in bacterial and plant samples. TheaimofthisstudywasdesigningthePCRmarkerspecificfortheA.tumefaciensstrainsbelonging to the biovar 1 according to classification of Keaneetal [1] and testing its specificity within Rhizobium/Agrobacterium group. All laboratory strains commonly used for bacterial plant transformationbelongtothisbiovar.Fourlaboratorystrains(EHA105,AGL0,LBA4044,GV3101)and 16fieldisolatesfromRhizobium/Agrobacteriumgroupwerestudied. The fingerprints of DNAs of studied bacterial strains were obtained using DIRPCR method [2]. SeveralprimersweretestedontheirabilitytoproducethevaluablespectraofPCRfragments. The DNA band from DIRPCR profiling obtained with KRPN 2 primer which was common for all laboratorystrainswaschosenforfurtherpurification,cloningandsequencing.Theconsensusclone sequence corresponded to the region of circular chromosome of A.tumefaciens strain C58. This region includes two oppositely oriented ORFs with undetermined functions and intergenic spacer. Designing the SCAR primer system, we placed forward primers within the first ORF and reverse primersinsidethesecondORF.Thedesignedprimers(6primercombinations)weretestedonDNAs ofallstudiedstrains.IntheresultofPCRtheexpectedfragmentsforeachprimerpairwereobtained only for agrobacterial strains belonging to biovar1. Primer pair N2f2r2 was chosen for the futher workduetoitshighestspecificity. After optimization of PCR reaction conditions, primers N2f2r2 were shown to allow the specific detection of A.tumefaciens strains belonging to biovar1 and thus are proposed to be the SCAR primers. The specificity of SCAR primers N2f2r2was tested in PCR on 10 variuos plant DNA matrices (soya, maize,sugarbeet,potato,tomato,cabbage,pea,durumwheat,barley,cucumber).Theconvenience oftheplantDNAsforPCRwastestedinthereactionwith18SrDNAspecificprimers.Inallcasesthe correspondingPCRproductswereobtained.NoamplificationwasobservedwithN2f2r2primers.  110 CoExtraInternationalConference Then PCR with N2f2r2 primers were performed on mixed DNA matrices, when every between 10 tested plant DNAs was mixed with A.tumefaciens DNA in ratio 95ng of plant DNA to 5 ng of bacterial one per reaction. The presence of the target PCR bands and the absence of nonspecific PCR products in these reactions proved the specificity of N2f2r2 primers in reactions on complex mixedmatrices. TheduplexreactionswithN2f2r2–18Sfr2primerswereperformedonmixedplant/bacterialDNAs andcorrespondingpureplantandbacterialDNAsasapositivecontrols.Asinglebandsofthecorrect sizes were detected in reactions with pure DNAs. Two bands of correct sizes were detected for reactionswithmixedplantAgrobacteriumDNAs. Proposed duplex PCR approach with newly developed primer system allows to substantially accelerateandincreasetheaccuracyofprimaryplanttransformantsscreening.  References: Keane,P.J.,Kerr,A.andNew,P.B.(1970)Crowngallofstonefruit.II.IdentificationandnomenclatureofAgrobacterium isolates.Aust.J.Biol.Sci.,23,585–595 TsygankovaS.V.,IgnatovA.N.,BoulyginaE.S.,KuznetsovB.B.,KorotkovE.V.(2004)Geneticrelationshipsamongstrainsof Xanthomonascampestrispv.campestrisrevealedbynovelrepPCRprimers.EuropeanJournalofPlantPathology,110,8, 845853.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety. 111 CoExtraInternationalConference P11. Arapid&simplepointofusediagnosticforGMOdetectioninplants GuyKiddle,ManuelaRizzoli,OlgaGandelman,CathalJ.McElgunn, CathyMoore,LaurenceC.Tisi& James,A.H.Murray LumoraLtd,CambridgeshireBusinessPark,DenmarkHouse,AngelDrove,Ely,CambridgeshireCB74ET,UK  In this investigation, the utility of two commonly used nucleic acid amplification technologies (iNAATs)werejudgedfortheirabilitytodetectcommongeneticallymodifiedelementsinplanta.An isothermal amplification technology (Loopmediated amplification; LAMP) coupled to a realtime bioluminescent reporter (BART) was compared to realtime PCR (qPCR). The rate of amplification, detection limit, and robustness of each technique were assessed against a variety of samples obtained using conventional or novel DNA extraction procedures. Both amplification techniques weresensitiveenoughtoroutinelydetect0.1%GMwithinablendedmaizeseedreferencesamples. UnlikeqPCR,LAMPBARTwasalsorobust,toleratingtheusualPCRinhibitorsfoundinunprocessed extracts, making this technology a more suitable high throughput technique. Moreover, the isothermal nature of LAMPBART, together with the simplistic instrumentation and extraction procedure,allowedtheanalysistobeperformedwithinafieldenvironment.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.                 112 CoExtraInternationalConference P12. Developmentofanintegratedplatformforthedetectionofmaterials derivedfromgeneticallymodifiedcropsinfoodandfeedproducts LievensA.,A.LeundaCasi,E.Piednoir,G.MbongoloMbella,M.SneyersandM.VandenBulcke ScientificInstituteofPublicHealth,J.Wytsmanstraat14,B1050Brussels([email protected])  The production of Biotech Crop Products (BCP) has become an important global business for food and feed products. A commonground of understanding of the GMnature of the sold/bought productsisconsideredessential.Indeed,asufficientlevelofinformationontheGMcontentpresent intheproductmustbeavailabletocomplynotonlywithlegalandsafetyobligationsbutmustalso beinlinewithbothseller's/buyer'srisksandbenefits. To date, the key technology applied in GMO detection, identification and quantification in the European Community is the “Polymerase Chain reaction” (PCR). In view of the recent recommendationbytheECtoapplythe"HaploidGenomeCopynumber"(HGC)asthestandardDNA detection unit, all routine screening and identification methods are focused on DNA technologies (e.g.RTPCR,microarrays,(bio)chips,etc). The marginal cost of GMO analysis could be significantly reduced if 1°) cheaper denominator reference materials to be used in a qualitative/quantitative GMO detection (e.g. monospecific targetplasmids(MTPs))wouldbeavailable(M.VandenBulckeetal,2006),and2°)acosteffective, technicallysimplebutrobustGMeventscreeningapproach,allowingtooptimizetheidentification ofallEUauthorizedGMOeventspresentinthesample,couldbedeveloped. The Scientific Institute of Public Health (further abbreviated as ISP) has developed a 96well plate matrixbased decision model applying a standard SYBR®GREEN QPCR amplification strategy, using MTPsastargetdenominators.Aminimalsetofscreeningelementshasbeenidentified,takingthe ECauthorizedGMOUniverseasstartingsource(date:June2005).MTPsforthedifferentelements havebeendevelopedandhavebeensubmittedundersafedepositattheBCCM(Ghent,Belgium). The potential application of the combination of subsets of these MTPs as a reference material for GMO screening in maize and soybean, has recently been demonstrated in a 96well plate liquid phaseformat(A.Lievensetal.,2007). Here,wereportthedescriptionofthebasicelementsofthedevelopedGMOdetectionplatformand present some data on the results obtained with this new GMO screening system; designated as "CoSYPS"(for"CombinatorySybrgreenqPCRScreening").  Acknowledgments: This work was partially performed within the Work Package 5 (Coordination K. Gruden, NIB, Slovenia) of the EC granted project CoExtra (contract 007158). Cosponsors of this work are the BelgianFederalMinistryofPublicHealthandtheEnvironment(GMOdetecproject)andtheECJRC InstituteofHealthandConsumerProtection(Ispra,Italy).      113 CoExtraInternationalConference P13. UseofpJANUS¥02001asCalibratorPlasmidforGTS4032(Roundup ReadySoybean)Detection:AnInterLaboratoryTrialAssessment Lievens1A.,DeBernardi1D.,Moens1,W.,VandenBulcke1,M.Bellocchi2G.,Savini2C.,Mazzara2M., VandenEede2G. 1 2 ScientificInstituteofPublicHealth,J.Wytsmanstreet14,1050Brussels,Belgium ECJointResearchCentre,InstituteforHealthandConsumerProtection,viaE.Fermi2749,21027,Ispra(VA),Italy  This study aims to evaluate the use of dualtarget plasmids as calibrators in the quantification of GMmaterialsinfoodandfeedproducts.Asamodelsystem,RoundupreadysoyGTS4032(RRS), themajorGMmaterialonthemarket,waschosen. Two dual target plasmids, designated as pJanus¥02001 and pJanus¥Monsanto, comprising part ofajunctionregionoftheGMsoyEventGTS4032andtheendogenoustaxonspecificlectingene wereevaluatedforuseasreferencecalibrator.TheefficiencyofbothplasmidsasDNAtemplatewas tested for using various RTPCR methods applying both SYBR®GREEN PCRmethods and Taqman chemistries.Basedonthislimitedevaluation,bothplasmidsareshowntoperformequallywell.The pJanus¥02001plasmidwasthepreferredchoiceforfurtherstudyasthisplasmidoffersabroader versatilityinchoiceofPCRmethodstobeusedinthequantification. ThepJanus¥02001plasmidwaspasseddownaninterlaboratorytrialforperformancecomparison togenomicDNAofleaftissuefromRRS.Intotal11laboratoriesparticipatedinthestudy.Thedata generated in the interlaboratory trial have been analyzed both by conventional statistic & fuzzy logicapproaches. Data analysis confirmed that both the plasmid and the genomic DNA perform equally well as calibratorinastableandhomogenouswaythroughoutthedifferentlaboratoriesinvolvedinthetrail andthroughoutthedifferentquantitiesoftestedGM%(0,2%,1,8%and4,4%). Theseresultsindicatethatplasmidcalibratorsmayrepresentacostefficientvaluablesubstitutefor genomic DNA or reference powders as calibrators in the detection and quantification of RRS in products.  Acknowledgments: ThisstudywasperformedwithintheWorkPackage4(Coordinator:R.Onori(ISS,Rome,Italy)ofthe ECgrantedprojectCoExtra(contract007158).           114 CoExtraInternationalConference P14. Testingthe“ModularApproach”:anexamplewithRoundUpReady Soybean Bellocchi1 G., Foti1 N., Mazzara1 M., Savini1 C., Van den Eede1 G., De Giacomo2 M., Palmaccio2 E., Onori2R.,Miraglia2M. 1 European Commission Joint Research Centre, Institute for Health and Consumer Protection, Molecular biology and genomicsUnit,viaE.Fermi2749,21027Ispra(VA),Italy 2 ItalianNationalInstituteofHealth,DepartmentofVeterinaryPublicHealthandFoodSafety,GMOandMycotoxinsUnit, vialeReginaElena299,00161Rome,Italy  The concept of “modularity” for the analytical procedures and validation of methods in GMO analysis lies on the basic idea that GMO analysis consists of a limited set of distinct steps that represent a certain elementary unit with the process, called “module”. Modularity implies independency and flexibility of combining modules on the one hand, and uniformity and harmonisationontheotherhand.Ifmodularvalidationistobeapplied,fitforpurposeprocedures andgeneralacceptanceofminimumrequirementsforeachmoduleareneededinordertoevaluate theuncertaintiesassociatedwitheachmodule. If the realtime PCR measurement isnot influenced by the type of extraction method applied, the validationoftherealtimePCRmeasurementcanbeperformedonDNAextractedwithanymethod and from any type of matrix, but scientific evidence for this approach has still to be provided. Indeed,differentextractionmethodscaninfluencetheDNAquantificationinfoodproductsthrough realtimePCR.Aswell,thequantificationofGMOcanalsobeaffectedbythedegreeofprocessingof thematrixfromwhichgenomicDNAisextracted.Inthisstudy,anexperimentwasperformedtotest the independence between modules. A balanced experimental plan was designed where three methodsbasedondifferentmechanismsofDNAextractionandpurification(CTAB,Dellaporta,and Wizard)andthreematricesofdifferentnature(feed,biscuits,CRM1%)werefullycombined.DNA yieldswereestimatedbyspectrophotometricandfluorometricdeterminations.Criteriawereapplied toassessDNAqualityforthepossiblepresenceofinhibitorycompounds.The%valuesofGMDNA were processed via statistical and fuzzybased approaches. The findings of this study generally support the independence between modules when appropriate performance criteria are met by DNAextractssothattheycanbefitforthepurposeoftheanalyticalexercise,independentlyfrom the previous matrixDNA extraction combination. However, independence cannot be confirmed whenhighlyprocessedmaterial(biscuits,inthisstudy)iscoupledwithamagneticbeadssystemfor DNAextraction.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.       115 CoExtraInternationalConference P15. NAIMA:afastquantitativemethodforhighthroughputGMO diagnosticsinfoodandfeedstuffs DanyMorisset,TinaLikar,DavidDobnikandKristinaGruden Dept.ofBiotechnologyandSystemsBiology,NationalInstituteofBiology,Vecnapot111,Ljubljana1000,Slovenia  In the future, GMO testing will become economically difficult to cope with increasing numbers of approvedGMlines.Severalmultiplexingapproachesarethereforeindevelopmenttoprovideacost efficient solution. Despite its high sensitivity and specificity, PCR technology has some limitations, includingthelackof true multiplexing solutions.Furthermore,theaccuracy ofquantificationusing qPCRhasitslimitsduetoitsexponentialamplificationnature. In order to alleviate some of these inconveniences associated with PCR technology, we have investigatedtheabilityofNASBAtechnologytobeusedforGMOdiagnostics. Therefore,wehavedevelopedanovelfastmultiplexquantitativeDNAbasedamplificationmethod suitable with detection on microarray. This new method named NASBA Integrated Multiplex Amplification(NAIMA)allowsfastamplificationoftargetDNA(15to60min)inamultiplexfashion. Sensitivity(AbsLODis220copies,allGMtargetsdetectedbetween0.1%and100%)andspecificity are suitable with requirements for GMO detection. The method allows quantification with same precisionasforsingleplexqPCRinthetestedGMrange(0.1%20%)onfoodandfeedsamples.In combinationwithmicroarraydetection,NAIMAallowsfastidentificationandquantificationofGMOs infoodandfeedsamples.Theconceptcouldeasilybeextendedtodomainswherediagnosticsrely onDNAbasedsequencedetection.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety. 116 CoExtraInternationalConference P16. GMOversusmycotoxinssamplingplan:apragmaticapproach M. De Giacomo1; R. Onori1; E. Palmaccio1; M. De Vivo1; C. Di Domenicantonio1; C. Brera1; M. Miraglia1;A.Malcevsky2;N.Marmiroli2,E.Prentera1. 1 ItalianNationalInstituteofHealth,DepartmentofVeterinaryPublicHealthandFoodSafetyGMOsandMycotoxinsUnit, Rome,Italy, 2 UniversityofParma,Parma,Italy  The application of Recommendation 787/2004 for the GMOs sampling in the food and feed chain raised some difficulty and sometime the worthlessness / unfeasibility of its implementation. It shouldbeconsideredthatthecontrolplanisanactiontobeundertakenwithaglobalperspective implying the implementation of control procedures suitable for several analytes. In addition, it wouldbedesirabletomakeavailableareliable,easy,andlowcostsamplingmethodologies. This study aims at verifying if the current sampling methodologies for mycotoxins (the most heterogeneouslydistributedanalytesinalot)couldfulfiltherequisiteofarepresentativesampling also for GMOs and derived products. This should minimize the costs for the control plans for mycotoxins/GMOsandcouldprovideatemplatefortheharmonizationofsamplingproceduresalso in accordance with trends raised during recent “ISO/IWA on Bulk Commodity Grain Sampling” meetingheldinSeattleMay2008. Theworkimpliesthepragmaticsimultaneousimplementationofthreesamplingmethodologiesfor soybeansandflourconsignments:mycotoxinssamplingplanaccordingtoRegulation401/2006,the GMOsamplingaccordingtoRecommendation787/2004andamoreintensivesystematicsampling that can evaluate as closest as possible the “true” GMO content of the lot (mean value) and can allowtheevaluationoftheGMOdistributioninthelot. TheresultsofmycotoxinsandGMOssamplingwillbecomparedwiththemeanconcentrationofthe intensivesampling,throughappropriatestatisticaltest(ttest),andthroughresamplingtechniques. ThesamplingprocedureswerecarriedoutwithmanualdynamicsamplingonshipmentsatlowGMO level: 4 lots (5000 tons each) of soybean grains and 2 lots (5000 tons each) of soybean flour. For mycotoxins and GMOs sampling plans, duplicate sampling were also performed for evaluating samplingprecision. Theincrementalandthebulksampleswerehomogenizedbydrymilling(Romerlabs.Inc.Rasmill) and a representative portion was further homogenised by dry milling (Retsch ZM 200 centrifugal mill, sieve 0.5 mm) to obtain the analytical sample. Genomic DNA was extracted from 200 mg of sample by Qiagen plant DNA extraction minikit, according to producer instructions. The concentration of the extracted DNA was determined by NanoDropTM ND1100 (Euroclone s.p.a., Milan, Italy). The extracted material was then diluted to a final concentration of 80 ng/ul for the followingRealTimePCRanalysis. Quantification of Roundup Ready Soybean (RRS) was performed by realtime PCR with TaqMan chemistry ( ABI PRISM 7700 Sequence Detection System 1.9.1) by means of a constructspecific method (ISO 21570: 2005 Annex C.4). According to the limit of quantification of the method ( 20 genomiccopiesforRRS)thepracticalLOQwascloseto0.008%RRS. Furthermore, samples were analyzed by qualitative method (lateral flow strip, LOD 0.1% RRS) according to sub sampling strategies deriving from cost based statistical tool OPACSA (OPtimal ACceptanceSamplingbyAttributeofgrains)basedonOCC(OperationCharacteristicsCurves). The poster illustrates the methodologies applied and summarizes the results of the first sampling performedatRavennaharbouronaGMOfreesoybeanlotof5,000tons.  117 CoExtraInternationalConference  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.         118 CoExtraInternationalConference P17. Approachestomonitortheadventitiouspresenceoftransgenesinex situcollectionsofnationalgenebanks GurinderJitRandhawa,MonikaSingh ReferralCentreforMolecularDiagnosisofTransgenicPlantingMaterials;NationalResearchCentreonDNAFingerprinting, NationalBureauofPlantGeneticResources,PusaCampus,NewDelhi110012,India [email protected];[email protected]  Withthedramaticincreaseinthecommercialcultivationoftransgeniccrops,theconcernsregarding theirpotentialimpactsonenvironmentandhumanhealthare requiredtobeaddressedinproper perspective.Theseissueshavemorerelevancetodevelopingcountries,particularlyIndiabeingrich in biodiversity and centre of origin of many crops. National Bureau of Plant Genetic Resources (NBPGR)isthenodalinstituteatthenationallevelformanagementofindigenousandexoticplant geneticresourcesforfoodandagricultureandtocarryoutrelatedresearch.NationalGeneBankat NBPGR conserves more than 3,68,584 germplasm accessions of field and horticultural crops and their wild relatives. With the commercial release of five events of Bt cotton in India and their cultivation on an area of 7.6 million hectares, an appropriate approach to study the probability of adventitious presence of transgenes into ex situ collections of the Gene Banks needs to be established. The main aim of Gene Banks is to collect, conserve and make genetic resources availabletothebreeders,sothemaintenanceofthepurityofgeneticidentityoftheaccessionsisof critical importance. Therefore, all possible efforts need to be made to prevent the adventitious presence of transgenes in the accessions conserved in the Gene Banks for the posterity. The differentapproachestomonitortheadventitiouspresenceoftransgenesinexsitucollectionsneed tobedevelopedaimingtominimizethegeneflowofthetransgenes.Themajorareaofadventitious presenceoftransgenesisthecollectionandacquisitionstageasgeneticresourcesmayhavebeen exposedtogeneflowoutsidethecontroloftheGeneBank.Sothestrategiesshouldaimtominimize the gene flow of the transgenes at these stages. As a part of risk analysis, while collecting or acquiringnewaccessionsbyothermeans,theGeneBanksshouldconsiderbeforetesting,whether transgenicevents(boththecommercialaswellasunderresearch)intherelevanttaxaarelikelyto bepresentintheareaofexploration/collection.TorandomlytestedtheaccessionsfromtheGene Bank, the reliable, sensitive and costeffective qualitative as well as quantitative methods for detecting the transgenes are needed to ensure seed quality and to monitor for its adventitious presence in nonGM seed lots. Conservation of ex situ collections with minimal adventitious presence of transgenes in the National Gene Banks would be urgently required to sustain the biodiversitywhilefullyharnessingthebenefitsoftransgeniccrops.     119 CoExtraInternationalConference P18. Monitoring the adventitious presence of transgenes in ex situ cotton collectionsoftheNationalGeneBank GurinderJitRANDHAWA1,,RashmiCHHABRA1,MonikaSINGH1,RuchiSHARMA1,AnjaliKAK2 1 ReferralCentreforMolecularDiagnosisofTransgenicPlantingMaterials,NationalResearchCentreonDNAFingerprinting 2 GermplasmconservationDivison NationalBureauofPlantGeneticResources,PusaCampus,NewDelhi110012,India [email protected];[email protected]  Geneflowthroughseed/pollendispersalfromgeneticallymodified(GM)cropstononGMcropsor totheirwildandweedyrelativesisoneofthemajorconcernsrelatedtoecologicalriskspertaining to GM crops. The introductions of transgenic DNA constructs into the agricultural field provide unique markers to measure the introgression of transgene in ex situ collections of Gene Banks. Bt cottonexpressinginsectresistancecrygeneisthefirstbiotechcropthathasbeencommercializedin Indiain2002andfiveeventsofBtcotton,viz.,MON531,MON15985,Event1,GFMcry1AandBNBt arecurrentlyplantedonanareaof7.6millionhectares.NationalGeneBankatNationalBureauof Plant Genetic Resources, New Delhi conserves more than 3,68,584 germplasm accessions of field andhorticulturalcropsandtheirwildrelativesincluding5,443ofcottongermplasm.Thequalitative methods for detecting GM cotton are needed to ensure seed quality and to monitor for its adventitious presence in nonGM seed lots. The present study reports on the monitoring of adventitious presence of Bt cotton in in the ex situ collection comprising 50 accessions of cotton collected from the different cotton growing regions of India that have been conserved in National GeneBank.Hundredseedsofeachaccessionweretakenandhomogenizedthoroughlytoformthe seed powder. The efficacy of all the homogenized samples for amplification by polymerase chain reaction(PCR)wasdeterminedusingtheprimerpairspecificforcottonspecifcendogenousfsACP geneencodingfiberspecificacylcarrierproteinandallthesamplesamplifiedtheproductof112bp for fsACP gene. The polymerase chain reaction (PCR) based assays for detection of Cauliflower Mosaic Virus 35S promoter, nptII marker gene and cry1Ac gene were performed and the desired ampliconsof112bpfor35Spromotersequence,215bpfornptIIgeneand203bpforcry1Acgene were detected in the Bt cotton events used as positive amplification control, whereas no amplificationwasobservedinnonGMcottonandwatercontrolusedasnegativecontrols.Allthe50 testsampleswerescreenedandnoampliconsfornosterminator,nptIImarkergeneandcry1Acgene were detected. Furthermore, detection of different Bt cotton events was undertaken to check the transgenicityofexsitucottonaccessions,whichconfirmstheabsenceoftransgeneconstructsinex situ seed lots of all the 50 accessions tested for adventitious presence of transgene, indicating transgenic DNA has not been introgressed in any of cotton accession collected from the regions whereGMcottonhasbeencommercialized. 120 CoExtraInternationalConference P19. MoleculardiagnosisofcommercializedorunapprovedBtcropsofIndia usingqualitativeandquantitativePCRassays GurinderJitRANDHAWA,MonikaSINGH,RashmiCHHABRA,RuchiSHARMA ReferralCentreforMolecularDiagnosisofTransgenicPlantingMaterials National Research Centre on DNA Fingerprinting, National Bureau of Plant Genetic Resources, Pusa Campus, New Delhi 110012,India [email protected];[email protected]  To date, five events of Bt cotton, viz., MON531 with cry1Ac gene, MON15985 with cry1Ac and cry2Abgenes,Event1withcry1Acgene,GFMcry1Awithfusedcry1AgeneandBNBtwithtruncated cry1AcgenehavebeencommercializedinIndiaandotherBtcropssuchasBtbrinjal,Btcauliflower, BtokraandBtricearecurrentlyunderdifferentstagesoftestingunderlimitedandlargescalefield trials.Globally,areaundercultivationofgeneticallymodified(GM)cropshasdramaticallyincreased from1.7millionhectaresin1996to125millionhectaresin2008.Despitethepotentialbenefitsof GMcrops,perceivedenvironmentalandhealthrelatedissuesassociatedwithGMcropshavetobe addressedinproperperspective.ToenforceaneffectivemonitoringandtraceabilitysystemforGM crops, it is necessary to develop sensitive and reliable GM detection methods. Polymerase chain reaction (PCR) methods are being routinely and widely used for GM detection. The present study aimedatthedevelopmentofPCRbasedqualitativeandquantitativediagnosticassaysfordetection ofBtcottonevents,Btbrinjalwithcry1Acormodifiedcry1Abgene,BtcauliflowerandBtriceandBt okra with cry1Ac gene. The multiplex PCR assays for simultaneous detection of specific cry gene, Cauliflower Mosaic Virus 35S promoter, nos terminator and selectable marker gene, viz., nptII (neomycin phophotransferase) or aadA (aminoglycoside3’adenyltransferase) along with the respective endogenous reference genes in different Bt crops such as fsACP gene encoding fiber specific acyl carrier protein in cotton, SRK (Sreceptor kinase) gene in cauliflower, fructosidase gene in brinjal and TubA (tubulin) gene in rice were developed.The quantitative real time PCR assays have also been reported for estimation of copies of specific transgene integrated in the differentBtcrops.TheassaysfordetectionofspecificGMtrait/transgeneupto0.01%havealsobeen developed and validated that meet the Supreme Court of India’s stipulations for developing a protocol for testing contamination to a detection limit of 0.01%. The developed qualitative and quantitative PCR assays provide a robust, costeffective and sensitive method for diagnosis of differentBtcrops.  121 CoExtraInternationalConference P20. MultiplexingofSIMQUANT CamillaSkjæret,ArneHolstJensen,TorsteinTengsandKnutG.Berdal NationalVeterinaryInstitute,SectionforFoodBacteriologyandGMO,POBox750Sentrum,0106Oslo,Norway  Aim TheideaistodeveloptheSIMQUANTapproach(Berdaletal.,2008)forGMOquantificationintoa multiplexquantificationormultiplexthresholdquantificationscheme.Thisapproachmaybeusedas a screening strategy to identify whether the total percentage of several GMOs of one species is above a threshold. The strategy is similar to SIMQUANT, but rather than scoring the absence or presence of one GMO analyte in a PCR, the score is made using a mixture able to detect several GMOsatasinglemoleculelevel.Thisisachievedbymixingseveralprimersandprobestogether.The resultwillnotbeabletodecidewhichGMOsthatispresent,butratherthetotalconcentrationof themultiplexedGMOs.  MaterialsandMethodology The multiplexed methods are all singleplex validated as part of the European procedure for GMO authorization,orvalidatedaccordingtoothercollaborativeprojectsandusedinourlaboratoryfor accredited GMO detection. In this study we have mixed 8 different GMmaize methods: Bt176, Mon810,NK603,Mon863,TC1507,DAS591227,T25andGA21. Firstwewantedtocomparethesensitivityofthemultiplexmethodwiththesensitivityofvalidated singleplex PCRs for all the GMOs at a concentration of one target DNA molecule using certified referencematerials(IRMM,Belgium)materialforeachmaizeevent.ForthesingleplexrealtimePCR we used TaqMan Universal PCR MasterMix (Applied Biosystems), with primer and probes concentrationsandPCRcyclingconditionsaccordingtovalidatedsingleplexmethods.Formultiplex realtime PCR we used QuantiTect Multiplex PCR MasterMix (Qiagen). We tested both the same concentrations of primer and probes as singleplex (A), or made a simple modification of the concentrationsoftheprimersandprobestoaccommodateamultiplexsituation(B).Therealtime PCRcyclingconditionswasrunaccordingtoQuantiTectMultiplexPCRMasterMix’sprotocol. FinallywequantifiedseveralGeMMaproficiencytestsandseveralCRMswiththemodifiedmultiplex approach (B), and compared these results with the results from traditional realtime PCR quantificationsandGeMMa’sassignedvalues.ThePCRformingunit(PFU)numberforthereference geneAdh1wasdetermineda priori,andthesamplesweresuccessivelydilutedtoa concentration wherethepresenceof(onaverage)oneGMtargetPFUperPCRcorrespondedto0.9%GMO.  Resultsanddiscussion To compare the sensitivity of Multiplex SIMQUANT with Singleplex SIMQUANT we used the same batchofonePFUDNApertestportionforeachGMmaize.ThenumberofpositivePCRsoutof10 parallelswascompared.MultiplexA(noprimerandprobemodifications)showedsimilarsensitivity as the singleplex for all maize GMOs, except GA21, while multiplex B had similar sensitivity as the singleplexforallmaizeGMOs. The accuracy of multiplex SIMQUANT method was acceptable when tested on several samples of knownGMOconcentrations(GeMMaandCRMs).Thiswork,intendedasaproofofprinciple,shows thatmultiplexingofSIMQUANTispossiblewithnoorminoroptimization.  122 CoExtraInternationalConference  References: Berdal,K.G.,C.B.Andersen,T.TengsandA.HolstJensen.(2008)Improvinghundredfoldthelimitofquantification(LOQ) ofGMOanalysesbyqualitativesinglemoleculequantification(SIMQUANTPCR).EuropeanFoodResearchandTechnology, 227:11491157.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.             123 CoExtraInternationalConference P21. Useofcomputationalsubtractiontosearchforunknowngenetic modifications Torstein Tengs 1, Haibo Zhang 1,2, Arne HolstJensen 1, Jon Bohlin 3, Melinka A. Butenko 4, Anja BråthenKristoffersen3,HildeGunnOpsahlSorteberg5andKnutG.Berdal1 1 NationalVeterinaryInstitute,SectionforFoodBacteriologyandGMO.POBox750Sentrum,0106Oslo,Norway 2  School of Life Science and Biotechnology, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, P.R. China. 3 NationalVeterinaryInstitute,SectionforEpidemiology,POBox750Sentrum,0106Oslo,Norway 4 UniversityofOslo,DepartmentofMolecularBiosciences,POBox1041,Blindern,0316Oslo,Norway 5 AgriculturalUniversityofNorway,DepartmentofPlantandEnvironmentalSciences,POBox5003,1432Ås,Norway  Whengeneratingageneticallymodifiedorganism(GMO),theprimarygoalistointroducewithgene technologyoneorseveralnoveltraitstothetargetorganism.AGMOwilldifferfromitsnonGMO parent in that its pool of transcripts is altered. Currently, there are no methods that reliably can determineifanorganismhasbeengeneticallymodifiedifthenatureofthemodificationisunknown. WehaveusedcomputationalsubtractionandhighthroughputcDNAsequencingtodetermineifan organismisgeneticallymodifiedaswellastodefinethenatureofthemodification.Webelievethat thisapproachwillrepresentapowerfulnewstrategywherefewerassumptionswillhavetobemade comparedtomethodscurrentlyinuse.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.        124 CoExtraInternationalConference P22. EffectofdifferentstorageconditionsonPCRamplificabilityofgenomic DNAextractedfrompelletscontainingmaizeMON810maize BarbauPiednoir1 E., E. Vandermassen1, D. Van Geel1, M. Van den Bulcke1 and Nancy Roosens1, A. Malcewsky.2andN.Marmiroli2 1 ScientificInstituteofPublichealth,J.Wytsmanstraat14,B1050Brussels,Belgium 2 DepartmentofEnvironmentalSciences,UniversityofParma,Parma,Italy  In2009,theEuropeanCommissionhaspublishedthesecondreportonthecoexistenceofGMcrops with conventional and organic farming, outlining the activities undertaken in response to the Council's request and providing an update of the state of implementation of national coexistence measures. In view of further enhancing the efficiency of national coexistence measures, the EuropeanCoexistenceBureau(ECoB)isdevelopingsuitablecodesofgoodpracticestobeadopted by the stakeholders involved in the production, harvesting, storage, processing and marketing of GM. Within such a framework of enforcement purposes and liability issues, GM traceability along the supply food and feed chains is an essential prerequisite. GMO detection, based on DNA technologies, plays a key role in this process often implying the storage of samples for up to 6 months(RecommendationEC/787/2004,RegulationEC/172/2002). Within the CoExtra Project, the University of Parma and the Scientific Institute of Public Health, have explored the effect of time and different storage conditions on DNA extraction and QPCR amplification efficiency in feed samples. Fodder pellets were prepared from a mixed flour (composed of 48% maize, 25% bran, 10% barley, 9% soybean and 8% sunflower flours in weight)) containingthree(0.1%,1%and5%)differentpercentagesofmaizeflourMON810. IntheUPARexperiments,thepelletshavebeenstoredat20C°,4C°andRTandextractionshave been performed at T0, T1 (3 weeks) and T2 (6 months). using a CTAB protocol (ISO 21571:2005) previously validated by JRCIHCP. The quality of DNA extraction efficiency has been monitored in terms of yield (ng/micro litre), 260/230 and 260/280 ratios. The evaluation of PCR amplification efficiency has been carried out through both end point PCR and RTPCR analysis. The results from theseanalysesdemonstratedthatgDNArecoveryandamplificabilityremainsconstantovertimein allexperimentalstorageconditions.AslightdecreaseinthequalityofextractedDNAwashowever observed. In the IPH experiments, the mixed flour (as dry powder and admixed with liquid soup) and the fodderswerestoredforupto6monthsat+4°C(withandwithoutsilicagel),20°Cand80°C.The assessment of the conservation of the different matrices was carried out at different time points: day0,week3,month2,month4andmonth6.gDNAwasextractedbyavalidatedCTABprotocol. The SYBR®Green QPCR methods applied in the Cosyps screening system involving 11 markers (kingdom,species,genericrecombinant,traittargets)andappropriateTaqMan®identificationqPCR analysis for the present GMevents were used for determining the influence of the respective storageconditionsonGMOdetection.Theresultsofthisstudyshowthatflourandfodderarewell conservedforupto6monthinallconditionsbutthatinliquidsamplesfreezingat20°Cand80°Cis recommended.  References: ECreportof2April2009onnationalstrategiestoensurecoexistenceofgeneticallymodifiedcropswithconventionaland organicfarming..Pleasesee:http://ec.europa.eu/agriculture/coexistence/index_en.htm Commission Recommendation EC/787/2004 of 4 October 2004 on "technical guidance for sampling and detection of geneticallymodifiedorganismsandmaterialproducedfromgeneticallymodifiedorganismsasorinproductsinthecontext ofRegulation(EC)No1830/2003" 125 CoExtraInternationalConference Regulation EC/178/2002 of the European Parliament and of the Council of 28 January2002 laying down" the general principlesandrequirementsoffoodlaw,establishingtheEuropeanFood"  Acknowledgements: This work was performed within the framework of Work Package 4 of the CoExtra Project (Co coordinator: R. Onori (ISS, Rome, Italy), contract 007158, under the 6th Framework Programme, priority5,foodqualityandsafety.                     126 CoExtraInternationalConference P23. MultiplexDNADetectionSystemForIdentificationOfGenetically ModifiedOrganisms(GMOs)InFoodAndFeedChains;CoExtraWP6 results Jeroen VAN DIJK, Gabriella UJHELYI, Theo PRINS, Marleen VOORHUIJZEN, Angeline VAN HOEF, HenriekBEENEN,HenkAARTS,EstherKOK. RIKILTInstituteofFoodSafety,Wageningen,Netherland  TheadoptionofGMcropshascontinuouslyincreasedoverthelastdecadewith125millionhectares ofthesecropsgrownin2008worldwide.Thereisagreatdemandfromboththegeneralpublicand theorganicfarmingcommunityforthepossibilitytochoosefromgeneticmodification(GM)freeor GMcontainingfoodstuffsandfortheabilitytogrowGMfreecrops.Coexistenceisawayofallowing farmers to choose between conventional, organic and GM crop production and demands a traceability system. In Europe, such traceability is legally mandated for food and feed originating from or containing GMOs (EU directive 1829/2003). In order to monitor and enforce compliance withcoexistenceregulations,authoritiesrequiretheabilitytotrace,detectandidentifyGMOs. ConventionalrealtimePCRreachesa0.1%detectionlevelformosttargets.However,thenumbers ofdifferentapprovedandunapprovedGMplantsmakedetectionandidentificationofGMmaterial in food a timeconsuming and expensive puzzle in cases when many subsequent real time PCR reactionshavetobeperformedforafinalidentification.Thereisaclearneedforamethodthatcan identifymanyDNAtargetswithinalimitedsetofexperimentsandatasensitivelevel. Padlockprobe(PLP),ligationbasedmultiplexdetectionprovidesapromisingmethodtomeetallthe demandsofGMdetection.ThismethodisbasedonthedetectionofauniqueDNAsequencebya PLP in isolated plant DNA. Only when both ends of the PLP hybridize juxtaposed to their specific complementarytargetsequence,ligationcanoccurandwillresultinacircularmolecule.Universal primer sites in the PLP then enable amplification of only the circularized probes. Only amplified probeswillyieldasignalwhenthepoolofPCRproductsishybridizedtoamicroarray. PLPshavebeendevelopedfor29targetsalready,includingGMplantspeciesandseveralGMevents, elements and constructs. We have detected positive targets in mixtures up to 13 DNA targets. DuringthePLPexperiments0.1%detectionlevelhasbeenreachedsofarincaseofelements,and 1%incaseofevents.Similarresultswerereachedduringtransferofthemethodtoanotherinstitute. Further aims are to design padlock probes for more targets and to lower the detection level. An alternative,socalledrollingcircleamplificationmethodisalsobeingexplored.Thelatestresultswill bepresented.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.  127 CoExtraInternationalConference P24. TheCoExtrawebsite,akeytoolintheCoExtraexternal communicationstrategy P.Daubresse1,K.Minol2andD.Breyer1 1 ScientificInstituteofPublicHealth,DivisionofBiosafetyandBiotechnology,RueJ.Wytsmanstraat14,B1050Brussels Belgium,Phone:+3226425293,Fax:+3226425292,Email:[email protected] 2 GeniusGmbH,Darmstadt,Germany Communicatingandinteractingwiththepublicaboutresearchisofvitalimportance.Sciencethatis communicatedpoorlyremainsunrecognized.ParticipantsinEUfundedprojectsareencouragedby theEuropeanCommissiontopromotemoreandbettercommunicationonscienceandresearch,by payingparticularattentiontothe"publiccommunication"dimensionoftheirwork. Disseminating and facilitating access to sciencebased information has been therefore one of the major objectives of CoExtra, a Europeanfunded project addressing coexistence of genetically modifiedorganismsandnongeneticallymodifiedorganismsinEuropeandtheirtraceability.Tothis end, a dynamic and interactive website (http://www.coextra.eu) has been developed as the core elementoftheCoExtraexternalcommunicationstrategy.Thiswebsitehasbeendesignedtomakeit attractive and accessible to a large audience in a very simple and practical manner, building on practical experiences gained in the development of other websites related to biotechnology and geneticallymodifiedorganisms. Thewebsitewasdevelopedtakingintoaccountthat"thepublic"isnotahomogeneouspopulation; ratheritencompassesnumeroussubgroups,eachofthemconstitutingadistinctaudienceseeking information that answers their questions and concerns with an appropriate level of detail. Accordingly, The CoExtra website has been structured to allow 3 main readership levels: Level 1 corresponding to the most accessible pages and providing general and popularized information (suchasnewsandreportages);Level2offeringinformationfornonspecialistsaboutthedozensof research projects within CoExtra; Level 3 providing for the more expert readers the detailed scientificdatafromtherunningprojectsincludingthemostrecentresults,reportsandpublications, andthelistofpartners/institutionsinvolved. Anotherimportantaspectofthewebsiteisthatitsuppliesbackgroundinformationonprogressin the implementation of coexistence and traceability measures in various European countries ("countrysections").ThispartofthewebsiteisavailableinseveralEuropeanlanguagestoovercome potentialbarriersoftheusersbyallowingaccesstolocalinformationintheirnativelanguage. Last but not least, the website also provides forvarious permanent tools allowing multidirectional interactionwithitsvisitors(electronicnewsletter,onlinediscussionforum...). Content is displayed using a webbased platform, based on a sophisticated Content Management System. In order to maintain consistent management policy in content edition, an Editorial Office (responsibleforthepublicinformationlayer)andanEditorialBoard(responsibleforthereviewand endorsementofcertaintypesofdocumentsbeforetheyarepublished)havebeenestablished. ThefrequencyandprofileoftheuseoftheCoExtrawebsitehavebeenmonitoredonaregularbasis all along the duration of the project through the use of various indicators. Data indicate that the "popularity" of the CoExtra website as well as its ranking in major search engines for relevant keywords have gradually increased over time. The interest of the public goes mainly on local information ("country sections") and on content written in journalistic style (news, reportages), while scientific results attract fewer visitors, most probably due to the low amount of detailed scientificdatacurrentlyavailableonthewebsite.  128 CoExtraInternationalConference Broadlyspeaking,onecanconsiderthatthewebsitemetitsobjectives:providingbalancedandfact based communication and contributing to improve awareness and understanding about co existenceaspectsofGMO.Technicallyspeaking,thewebsiterepresentsalsoapowerfulwebbased communicationplatformthatwillremainactiveforthenext5years.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.             129 CoExtraInternationalConference P25. Influenceofthe(nonGM)soybeanpriceoncompoundfeedprice NicolasGryson1,2andMiaEeckhout2 1 Laboratory AgriFing, Faculty of Biosciences and Landscape Architecture, University College Ghent, Ghent University Association,Voskenslaan270,9000Ghent,Belgium,[email protected] 2 LaboratoryofCerealandFeedTechnology,FacultyofBiosciencesandLandscapeArchitecture,UniversityCollegeGhent, GhentUniversityAssociation,Voskenslaan270,9000Ghent,Belgium.  The added cost for the production of nonGM feed compared to GM feed is mainly related to its composition. The amount of (nonGM) soy used, which depends on the target animal, will highly influencethefinalpriceofthecompoundfeed.DatashowthatduringtheperiodJanuary2006till December2008theevolutionofthesoybeanmarketpricehasasimilarpatternasthe compound feedprice.However,bytheendof2007,theGMandnonGMsoybeanpricesincreasedsignificantly. Asaresult,benefitsforagriculturalcompanieshaveincreasedandgainmarginsofcompoundfeed manufacturerswerereduced.Manybelievethatthisincrease,whichalsoaffectedtheGMandnon GM soybean prices, was due to the size of the South American acreage, the continued strong Chinese demand for soybeans and the decreased US supply. However, the US acreage used for biodieselproductionisstillrelativelysmall.Moreover,allcommoditypricesincreasedinthatperiod andcouldthereforebeexplainedbyspeculationsontheglobalmarket.Anno2009,thecommodity priceswentmoreorlessbacktotheirstatusof2006. Atthesametime,thepricepremiumfor1tonofnonGMsoyincreasedfrom12€inJanuary2006to 77€inJanuary2008duetothedecreasedavailabilityanduncertaintyregardingBrazil’spotentialto produce nonGM soy. This is the major factor which sets the price for nonGM soy. Again, this increased premium further decreased the margin gains of compound feed producers using conventionalnonGMsoy. Twomajorfactorsshouldthereforebetakenintoaccountinordertomaintainthecoexistenceof GM and non GM soy for its use in compound feed products in the future: (1) a sufficiently high premiumorincentiveforfarmersproducingnonGMand(2)foodproductsderivedfromanimalsfed withGMshouldbelabelled.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.  130 CoExtraInternationalConference P26. The costeffectiveness of the coexistence of GMHT35 oilseed rape in Ireland:ananalysisofcropmanagementstrategies. ConorKeelan1,FionaS.Thorne1,andEwenMullins2 1 RuralEconomyResearchCentre,Teagasc,MalahideRoad,Kinsealy,Dublin17. 2 OakParkResearchCentre,Teagasc,Carlow.   ThecoexistenceofGeneticallyModified(GM)andnonGMcropsspecificallyreferstotheabilityof farmers to make a practical choice between conventional, organic and GM crop production, in compliance with the legal obligations for labelling and/or purity standards. In 2005, the Irish Governmentpublishedasetofcropspecificrecommendationstofacilitatetheestablishmentofan Irishcoexistenceframework(McGilletal.,2005).However,thisreportomittedspecificguidelinesfor thecoexistenceofGMandnonGMoilseedrapeduetothescarcityofIrishspecificresearchandthe natureofthecropwithregardtogeneflow. Followingfromthisknowledgegapresearchwasinitiated todevelopproductionmeasuresforthe cultivation of GMHT oilseed rape in Ireland that were both agronomically sustainable and economically viable. The objective of this paper is to outline the costeffectiveness of different methods by which coexistence between GM and nonGM oilseed rape can be achieved. Typical methodsincludeisolationdistances,pollenbarriersandrotationintervalstogetherwithspecificcrop management techniques. The GeneSys spatial model was used to generate computer simulations that were agronomically sustainable in an Irish context whereby gene flow was minimised and coexistencethresholdswerenotexceeded.Eachofthesesimulationswerethenanalysedusingthe AnnualizedNetPresentValue(ANPV)method.Thereforeeachsimulationwasassessedfrombothan agronomicandeconomicviewpoint. ThetraditionalcroprotationforcultivatingoilseedrapeinIrelandistogrowoilseedrapefollowed bythreerotationsofwinterwheat.Thiswasregardedasthebaselinemodelforthepurposeofthe analysis.Specificallyfourdifferentcropmanagementregimeswereanalysed: x x x x Impactofintroducingaspringcropintherotation,specificallyspringbarley.Thebaselinemodel is compared with a rotation of oilseed rape, winter wheat and two rotations of spring barley. Thisscenarioalsoexaminestheimpactoflengtheningtherotationcyclefromfourtosixyears. Increasing weed control with management use by increasing the mortality rate to 99% for volunteercontrolincerealcrops. Introducingfieldbordermanagementregimesandchangingmanagementtimesbyconducting hedgecuttinginNovemberratherthaninMay. Impact of reduced seed loss at harvest through use of better machinery and lower combine settingsatharvesttime. The Net Present Value (NPV) method is a standard approach for evaluating investment opportunities.Itisbuiltontheprincipleofthetimevalueofmoneywhichstatesthatapound/euro todayisworthmorethanapound/eurotomorrow.Projectedfuturecashflowsarediscounted,using anappropriatediscountrate,soastoestimatetheirpresentvalue.Eachofthepresentvaluesare then summed together and should the NPV be positive then the investment is deemed to be worthwhile.However,useofthisapproachbecomesproblematicinevaluatingprojectswhichhave differentlengthsofyears.Forexample,a6yearinvestmentinaprojectgeneratingapositiveNPV canbeassumedtobemorefavourableovera4yearinvestmentinthesameproject. Toproperly differentiateanddistinguishbetweeninvestinginprojectswithdifferentlengthsanAnnualizedNet   131 CoExtraInternationalConference Present Value (ANPV) is computed. The ANPV is computed using the same formula as the NPV. However, once the NPV is computed this figure is discounted by using the corresponding annuity factorsoastoarriveattheANPVoftheinvestment.TheinvestmentwhichhasthehighestANPVis consideredthemostrationaltochoosebyaprofitmaximisinginvestor.ToivonenandTahvanainen (1998)statedthattherequiredinterestfrominvestmentinagricultureisusuallyinthe35%range andassumeda5%discountrateintheiranalysis.AnumberofrecentstudiesusingtheNPVmethod toevaluateinvestmentinvariouscropshavealsouseda5%discountrate(Gooretal.,2000;Styles etal.,2008).Thereforea5%discountratewasalsoadoptedinthisanalysissimilartothatusedin otherrecentliterature. While the results presented above show that the ANPV of the baseline model is larger than the rotationwhichincludesspringbarleybothANPVsaregreaterthanzero. In each of Scenario 2 and 3 the ANPV for the baseline model was significantly higher than that associatedwithamodelwithgreaterweedcontrol(Scenario2)andamodelwithadditionalhedge cutting expenditure (Scenario 3). Scenario 3 is likely to be prohibitive with regard to Ireland as changingcuttingpatternstomonthssuchasNovemberiscontrarytoenvironmentallegislation.By contrastamortalityrateof99%forvolunteercontrolcanbeexpectedtohavesignificantagronomic benefits in reducing the incidence of volunteers in subsequent years of the rotation interval. However, in the absence of additional Irish agronomic studies, or field trials, on this subject it is difficultatpresenttorecommendScenario2aheadofanyoftheothersduetocostissues. InrelationtoScenario4theresultsindicatethataminimumyieldgaingreaterthanorequalto3% would be required for the ANPV for scenario 4 to be greater than the baseline model. Additional agronomicanalysissuchasanextensionofthisstudytofieldtrialsmayhelpdeterminewhethera yieldgainofthelevelsdiscussedinthispaperwillarisefromadoptingnewertechnologytoharvest GMHT oilseed rape. Regardless any reduction in the level of seed loss at harvest time can be assumedtolowerthelikelyincidenceofvolunteersinsubsequentyearsoftherotationinterval. Astheyieldandpriceassociatedwithspringbarleyarelowerthanthatofwinterwheat,rotations usingwinterwheatasthesolebreakcropwillalwaysreportahigherANPVthanthosewhichalso include spring barley. This is demonstrated in Scenario 1 as the baseline model appears to be the rationalinvestmentfortheprofitmaximisingproducer. Theseresultscanberegardedasaconsideredfirststepinassessingtheeffectivenessofalternative coexistencestrategiesinhelpingtoensurethatcoexistencetolerancethresholdsarenotexceeded. Theanalysisalsohashighlightedhoweconomicandagronomiccostsandbenefitsmustbeexamined togethersoastoobtainamorecompletepicturefromadoptionofanewtechnologysuchasGM oilseedrape.     132 CoExtraInternationalConference P27. ModellingcoexistencebetweenGMandnonGMsupplychains BaptisteLecroart1,AntoineMesséan1andLouisGeorgesSoler2 1 2 INRAEcoInnov,ThivervalGrignon,France, INRAALISS,Ivry,France  Introduction Coexistence is an approach allowing farmers to choose between conventional, organic and geneticallymodified(GM)cropsandallowingconsumerstochoosebetweendifferentfoodproducts subjecttoobligationsregardinglabellingandpurity.CoexistencebetweenGMandnonGMsupply chains is a complex issue, because adventitious mixing of GM material with nonGM product can occuratany oneofthestagesofproductionandanywherealong thesupplychain,fromthefield where the crop is grown to its handling and processing. Another major facet of GM and noGM coexistenceisthefactthattheGMcontentofaproductisnotavisibleattribute.Meantobridgethe gap in information do exist (product testing, using model), but they are subject to error. In this paper,wepresentasimulationmodelofthecoexistencebetweenGMandnonGMproductsalong supplychains.Morespecifically,theframeworkofthemodelisinspiredbythestarchmaizesupply chain. The aim of this model is to assess the ability of the supply chain to provide final nonGM product compliant with a required threshold (0.9% labelling threshold for example) and to discuss theimpactofthemeanstobridgetheinformationgaponthisprobabilityofcompliance.  MaterialsandMethods The model simulates GM and nonGM flows, and takes into account admixture and dilution functionsbetweenGMandnonGMbatchesalongthesupplychain.Inspiredontheexampleofthe starchmaizesupplychain,threekeystagesofthesupplychainareconsidered:grainproductionat field level, grain collection (including drying), and processing. Firstly, the MAPOD geneflow model (Angevin et al., 2008) is used to simulate GM adventitious presence in nonGM harvests due to crosspollination between GM and nonGM maize. Within the downstream supply chain, there is onlyonedryerandoneprocessingplant.Hence,GMandnonGMmaterialaresuccessivelyhandled in the same equipments. On the contrary, storage capacities are considered nonlimiting in the model and admixture due to storage equipments is considered negligible. At the maize collection level,themodelsimulatesontheonehandadmixturebetweenseveralbatchesblendedinasame bin, and on the other hand admixture between succeeding batches during drying process. Finally, the model simulates admixture between succeeding batches at processing. We have adopted a compartmentalmodellingapproachoftheprocesstoquantifyrisksofadmixture. StakeholdersdefinethefrequencyatwhichGMandnonGMflowalternateatdryingandprocessing levels (scheduling parameters). GM and nonGM batches are then randomly ordered according to thesevariables. Oncesequencesofbatcheshavebeenscheduled,uncertaintyremainsabouttheGMcontentofthe batches,allthemorethatitisnotavisibleattribute.Threekindsofcontrolsystemmightbesetup inthemodel: 1. Simpletraceability:thissystemallowsstakeholderstoidentifywhetherthebatchescomesfrom eitherGMornonGMvarieties. 2. Automaticdowngrading:thesimpletraceabilitysystemissupplementedbyrulesonautomatic downgradingofnonGMbatchesdriedand/orprocessedafterGMbatches. 3. PCRTesting:inadditiontothesimpletraceabilitysystem,testingisusedtogatherinformation on the nonGM batches. The model takes into account the fact that testing can be inaccurate (Starbird,2007).Weassumedaproportionalerrorbysimulatingmeasurementuncertaintywith alognormaldistribution.Testingcanbecarriedoutbeforeand/orafterprocessing. 133 CoExtraInternationalConference Two contrasted sets of admixture parameters (at drying and processing levels) were taken into accountforthesimulations,correspondingtolowandhighlevelofadmixturebetweensucceeding lots.Inaddition,previousstudieshavehighlightedthatthedistributionofGMadventitiouspresence innonGMharvestsisquitevariableamongregions(LeBailetal.,submitted).Thus,threecontrasted distributionsoftheGMadventitiouspresenceinnonGMharvestweretakenintoaccount,inorder to assess the effect of the input purity rate on the output purity rate. As far as the scheduling scenarioswereconcerned,twovaluesoftheschedulingparametersweretakenintoaccount:10and 100. Forscenarios2(automaticdowngrading)and3(PCRtesting),themodelidentifiesthestrategythat maximisestheprofit.Profitdependsonthenumberofbatchesofeachtype(GMandnonGM),on the testing cost and, on the probability that nonGM batches are compliant with the required threshold,accordingtoclientstesting.Clienttestingisperformedseveraltimesandthemeanvalue isconsideredfortheprofitcalculation.  Resultsanddiscussion Workonthesimulationmodelisstillongoing.Nonetheless,firstsimulationresultsshowthatchain organization,fromtheupstreamproducerstothedownstreamstakeholders,playsacrucialrolein maintainingorimprovingthenonGMproductcompliancewiththelabellingthreshold.Inaddition, modelshouldallowcomparingvariousstrategies.  References: Angevin F., Klein E.K., Choimet C., Gauffreteau A., Lavigne C., Messéan A., Meynard J.M., 2008. Modelling impacts of croppingsystemsandclimateonmaizecrosspollinationinagriculturallandscapes:TheMAPODModel.EuropeanJournal ofAgronomy,28(3):471484. LeBailM.,LécroartB.,GauffreteauG.,AngevinF.,MesséanA.,2009.Effectofthestructuralvariablesoflandscapesonthe risksofspatialdisseminationbetweenGMandnonGMmaize.submittedtoEuropeanJournalofAgronomy. StarbirdS.A.,2007.Testingerrors,suppliersegregation,andfoodsafety.Agriculturaleconomics,36,325334.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.          134 CoExtraInternationalConference P28. Supply chain description and analysis for maize, potatoes and fresh tomatoesinSlovenia Vladimir Megli, Peter Dolniar, Zoran ergan, Kristina Ugrinovi , Jelka Šuštar Vozli and Marjeta Pintar AgriculturalInstituteofSlovenia,Hacquetova17,1000Ljubljana,Slovenia  Sloveniaischaracterisedbylargegeographicaldiversitywhichresultsinthedistributionandsizeof farmland, which is in most parts characterized by a very small size of less than two ha. Due to expectedmoveofGMplantproductionintotheSlovenianfarming,thereisaneedtoconstructthe scientificandtechnologicalbasisforGMproductionandtoorganizetheagriculturalproductionina way, that will enable the coexistence between conventional, organic and GM production. The overall objective of our work is to facilitate coexistence along the feed and food chains by characterising the organisational schemes of supply chain product management from the farm to the shelf. Consumers need to have information whether they are going to choose GM or non GM foodorfeed.That’swhycoexistencecannotberestrictedonlyonproductionfieldbutaswelltothe wholesupplychainofthespecificcroporproduct.Thereisaneedtoestablishasystemthatallows separationofdifferentproductionsystems,thatfeedafterwardstothesupplychainsandwiththat minimizestheriskofmixing.Threecasestudiesaretryingtodeterminegenericresults. Maize is the most widespread field crop in Slovenia. Present range of maize production is a reflection of specialization and concentration in agriculture. In the case of Slovenia, field crop production supports wellformed animal husbandry with a comprehensive part of voluminous and concentrated feed. Since 1985 its share in crop rotation has been over 40%. Majority of approx. 73.000 ha of maize is grown for grain (58%), the rest is grown for silage (42%). Almost all the productionisintendedforanimalfeedingandlessthan3%isusedforhumanconsumption.Maize doesappearasmarketgoodinasmallshare(upto20%oftotalgrainproduction).Thereisacritical pointrelatedtoseedsinthecasethereisadmixtureofGmintheseedpurchased.Thiswouldalso have an impact on sowing and harvesting machinery as well as transport facilities particularly if machinery is used jointly with other farmers. From the two supply chains described we can conclude,thatformaizesilagesupplychaininSloveniathereisaverylowriskofcontaminationand admixture due to the fact that no silage is being imported and thus there is no risk of mixing the silagewhenthecommodityarrivestoSlovenia.However,somefarmershaveonfarmstoragefrom wheretheymaysellsilagetootherfarmers.Thisconstitutesacriticalpointintermsofadmixture. Potato is as well an important field crop, which has a long tradition in Slovenia. Due to problems with diseases and pests we can expect interest of producers to use resistant genetically modified varieties.Theharvestedpotatoesaresolddirectlytotheconsumerortotheretailersorwholesalers andfurtherdistributedtowardsthefinalconsumer.Thesaletotheindustryandtheprocessingof potatoes is of minor importance in Slovenia. Only small proportion of seed potato is produced in Slovenia.MostofitisimportedfromnorthernandwesternEuropeancountries.Potatoproduction inthelast15yearsdroppedfrom13.000hain1992to5.400hain2007.Webelievethatwehave reached the dewpoint in the Slovenian potato production. The acreage will probably stay at the present level, but the total number of potato producing farms will still decrease. The quantity of importedwareandseedpotatoisdirectlyconnectedtothequantityofwarepotatoproducedina previousyear.Thelargestimport(194000tons)wasnoticedinayear2004.Thepotatohorizontal supplychainisnotwelldefined.Themarketisnotwellorganised,thereforetherearealotofsupply chainswhichcan'tbetraced.Practicallyitisimpossibletoestablishtheimportanceofeachsupply chain.Sufficienttraceabilityfromtheproducertotheconsumerisprescribedbythelegislationand isobligatory.Theproblemsmightoccurbecauseoflargenumberofsmallfarmersandotherchains 135 CoExtraInternationalConference whichcan'tbeadequatelycontrolled.Theeasiestwaytoapplytraceabilitysystemwouldbeforthe integratedmanagementsystem,whichisontheotherhandnotobligatoryforallfarmers. InSloveniaanaverageproductionoftomatoesinthelastfiveyearperiodfromwasaround4000t. Generally, seeds (category: standard) are imported by the representatives of different seed companies and further sold to producers of transplants or directly to farmers and gardeners who producethetransplantsthemselves.Thetransplantsproducedbyspecializedplantsaresoldtothe retailers of transplants or directly to the growers and gardeners. Harvested tomatoes are sold directly to the consumer or to the groceries and canteens or to the wholesalers and farmers cooperatives who afterwards sell to the groceries and canteens. The sale of industry and the processing tomato is of minor importance in Slovenia. The import of fresh tomatoes in Slovenia is between10and13thousandtonsperyear,whichismorethantwotimeshigherthanthevolumeof the domestic production. The production of tomatoes in Slovenia appears to be segmented and through that without overall control. In Slovenia there is one existing certified quality system that addressestheIntegratedProductionofVegetablesonly.Thetraceabilityoftheproductionprocess, fromseedtothesaleoftomatoesisinSlovenianotwelldefined.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.                   136 CoExtraInternationalConference P29. PreferenceheterogeneityamongGermanconsumersregardingGM rapeseedoil MarinaZapilko,AgnesKlein,AndreasGabrielandKlausMenrad ScienceCentreStraubing,UniversityofAppliedSciencesWeihenstephan,Straubing,Germany  Consumeracceptanceisadeterminingfactorfortheprofitabilityandthedevelopmentpotentialof GMfood.AlthoughtestsintheUSAhavenotfoundsafetyproblemsderivingfromGMOsandGM ingredients in food products the acceptance towards the application of genetic engineering in the agrofoodsectorisstilllowintheEUespeciallyamongGermanconsumers.Intheopinionofmost European consumers there is nothing to gain by GMOs, but instead serious disadvantages could occur(EvensonandSantaniello,2004;Gaskelletal.,2006). Using the example of rapeseedoil this poster investigates the acceptance of German consumers towards GM food. In order to analyze this question 319 German consumers were interviewed in spring2007regardingtheirknowledge/trustinGMtechnology,theirattitudestowardsorganicand GM products as well as regarding their sociodemographic characteristics. In addition respondents had to make choices between several alternative rapeseedoil products within a DiscreteChoice Experiment.Theproductswerecomposedofdifferentlevelsoftheattributespriceandorigin.The third attribute taken into account was production technology comprising the levels conventional, organic,GMwithassociatedhealthbenefitsandGMwithassociatedenvironmentalbenefits. By means of latent class analysis three different consumer segments could be identified. The first clusterconsistsofconsumerswhosetspecialvalueonorganicallyproducedfoodproductswhilefor thesecondgroupofrespondentscheappricesarethemostdecisivepurchasecriterion.Forthethird segmentofconsumersnoneoftheattributesisofoutstandinginterest,buttheypreferindeedGM rapeseedoilwithassociatedhealth benefits.Allrespondentsfavourlocallyproducedrapeseedoil, butconsumersofthedifferentclusterssignificantlydifferintheirriskperceptionofGMtechnology aswellasintheirattitudestowardsthefeasibilityoflawsandregulationstoprotectconsumersfrom risks of GM food, the negative impacts of agriculture for the environment and the assessment of pricesoforganicallyproducedfood.Onthebasisoftheobtainedresultsthisposterwilladditionally give recommendations regarding special marketing activities for the different achieved consumer segments.  References: Evenson,R.E.andV.Santaniello(2004):Consumeracceptanceofgeneticallymodifiedfood.CABInternationalPublishing, Cambrige,Oxon. Gaskell,G.et.al.(2006):Eurobarometer64.3EuropeansandBiotechnologyin2005:PatternsandTrends.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.     137 CoExtraInternationalConference P30. Costs of coexistence and traceability systems in the food industry in GermanyandDenmark KlausMenrad1,AndreasGabriel1,MortenGylling2, 1 2 ScienceCentreStraubing,UniversityofAppliedSciencesWeihenstephan,Straubing,Germany DanishInstituteofFoodandResourceEconomics,Copenhagen,Denmark  IncontrasttotheincreasinguseofGMplantsinworldwideagriculture,theacceptanceofGMfood isstilllowintheEuropeanUnion(EU).Inordertoensurefreedomofchoiceforconsumersandusers ofGMandnonGMproducts, GMfoodandfeedproductshaveto belabeledtocontain GMOsor GMmaterialincaseatolerancethresholdof0.9%isexceededforEUauthorizedGMOs. ThisposteraimstoquantifythecostsoftraceabilityandcoexistencesystemsforGMfoodfromthe seedtotheretaillevelforsugar,wheatstarchandrapeseedoilforhumanconsumptioninGermany and Denmark respecting the 0.9 % threshold for labelling of GM food. The cost calculation for traceabilityandcoexistencemeasuresaredonewithaspecificmodelwhichfollowstheprincipleof aggregatingallincurredcostforcultivatingandtransportingcropsorprocessingoftherawmaterial croponthedifferentlevelsofthevaluechainandofincreasingthepriceofthefinalproductateach level.Thusallcoexistenceandtraceabilitycostsemerginginthevaluechainaretransferredtothe finalendproducts. Altogetherthemeasurestoensurecoexistenceandtraceabilityleadto5%to8%higherpricefor GMOfreerapeseedoilinGermanyand8%to10%higherpricesinDenmark.IncaseofGMOfree sugarthepriceloadingamountsto2%to5%andincaseofGMOfreewheatstarchto8%to11% eachrelatedtothecurrentpriceoftherespectiveproductinGermany.InDenmarkthepriceloading fornonGMwheatflourforbakingwillbeatthelevelof7%to8%.ForGMOfreesugartheanalog figure lies between 0.3 % and 2 % in Denmark. Finally recommendations for practical implementationandhandlingofcoexistencesystemswillbegiven.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.           138 CoExtraInternationalConference P31. AnalysisoftheextracostsgeneratedonFrench“LabelRouge”chicken supplychainbynonGMfeedpolicy. JulienMilanesi UniversitédePauetdesPaysdel'Adour,France.[email protected]  Theposterpresentstheresultsofastudyconductedabouttheextracostsborneby“LabelRouge” French chicken producers who chose GMfree feed policies. This research estimated these extra costs and studied their distribution along the supply chain through different scenarios. The poster showsanintermediatescenario,whereextracostsonGMfreesoyareconstants(atthe2008level) andGMcornisauthorizedinFrance.Thepostershows,atthemiddle,adiagramofthesupplychain, fromthecropproductionofcornandsoy,tochickensalesinsupermarkets.Thescenariopresented inthediagramisdetailedthroughboxeslocatedatitsside. This research has been accomplished between January and July 2008 by the Research Centre on ManagementoftheUniversitédePauetdesPaysdel’AdourandfinancedbytheNGOGreenpeace France. In addition to secondary data and literature review, the investigations included several surveys and meetings carried out in the sectors of chicken production and crops production and markets. The data concerning the chicken industry are based on information collected from three chickenproducersorganizationsrepresenting40%oftheFrenchchicken“LabelRouge”production. Theproducersofchicken“LabelRouge”whohaveadoptedGMfreepoliciesarecurrentlypayinga premiumtopurchasenonGMsoy.Thepostershowstheevolutionofthispremiumduringthepast years.Sincethemiddleof2008,thevalueofthepremiumisstabilisedaround2530€perton.The amountofpremiumchoseninthescenariois30€perton. As shown by several researches and as it has been experienced on other crops and countries, an authorization and generalization of GM corn in France would also translate intoextra costs to purchase GM free corn for the poultry producers. Theefforts made by nonGM corn producers to protectandcontroltheirproductionwouldindeedbetranslatedinmarketsegmentationbetween GM and nonGM corn. The poster shows the different measures necessary to comply with the thresholdof0,9%.TheircosthighlydependonthecharacteristicsofthefarmandontheGMOcrops shareontheregion,itcanvaryfrom5to37€perton.Theamountchoseninthescenariois10€per ton. Theposterthenshowsthedistributionoftheseextracostsonthedifferentlevelofthesupplychain, fromthefeedindustriestotheslaughterhouses,throughthefarmers.Itshowsthatwithextracosts of30€pertonforsoyand10€pertonforcorn,theextracostsattheendofthesupplychainwould be5,7centsperkiloofchickenmeat. At the bottom of the diagram, the poster shows what is at stake concerning distribution and labelling.In2008,retailersdidn’taccepttopaytoproducersincreasesofmorethan2or3centsper kiloforthiskindofchicken.Theyarguedthattheycouldn’tincreasetheirbuyingpricebecause,asit is impossible in the current French framework to show through labelling GMfree characteristic to consumers,theycouldn’tincreasetheirsellingprice. Without this possibility to transfer the extracosts to the consumers, the chicken producers would thereforehavetobearthem,butitwouldbeeconomicallyunsustainableifthisextracostsreach5 or6centsperkiloasshowninthescenariopresentedintheposter.Withoutpossibilityoflabelling, thechickenLabelRougeFrenchindustrywithGMfreefeedpolicywouldthereforenotsustainthe authorizationofGMcorninFrance. 139 CoExtraInternationalConference As a conclusion, the poster draws new perspectives of research on this issue, focusing on the concept of externalities. Considering extracosts as production externalities could indeed lead to implementotherpublicpoliciesthanlabelling.  140 CoExtraInternationalConference P32. Towardsanoptimalmanagementregimetofacilitatethecoexistence ofGMandnonGMoilseedrapeinIreland EwenMullins1,,YannTricault2,PaulFlanagan1,NathalieColbach2,ReamonnFealy3 1 PlantBiotechnologyUnit,TeagascCropsResearchCentre,OakPark,Carlow,Ireland 2 UMRBiologieetGestiondesAprentices,INRA,17rueSully,BP86150,21065DijonCedex,France 3 SpatialAnalysisUnit,TeagascRuralEconomyResearchCentre,Kinsealy,Dublin,Ireland  Capitalising on the existing knowledge base generated through the development of GENESYS, the objective of this research was to utilise the most recent model version to develop an optimal managementstrategyforpotentialearlyadoptersofGMHToilseedrapeinIreland.Theconsidered landscape was Bridgetown, Co. Wexford, which is situated in the south east of Ireland. A digitised map was created from a suite of aerial photographs (provided by Ordinance Survey Ireland, www.osi.ie) using ArcView v3.2. The map consisted of 1704 plots identified as cropped fields (607 plots),hedgerows(972plots)ornovegetationareas(roads,farms...58plots).Thenumberoffallow (uncropped)fieldsacrossthelandscapewassetat10%(n=67).Owingtothedistinctfieldshapein thelandscape,thehedgerowsofeachfieldwerecreatedusingthemanualtracingtoolinArcView v3.2. Foreachsimulatedcroppingsystem,threeregionalproportions(5%,15%and30%)ofGMOSRsown inOSRfieldsweresimulatedfor.OSRfieldswerecultivatedwitheitherawinterconventionalvariety or a winter GM herbicide tolerant variety homozygous for a dominant transgene conferring resistance to glyphosate. Crops were randomly allocated on each cropped field based on (i) the regional proportion of GM OSR and (ii) the number of fields with the simulated rotation. For two coexistingcroppingsystems(GMandnonGMOSR),theallocationofGMtoconventionalfieldsfora 5% regional proportion was 577 and 30 respectively, 516 (GM) to 91 (nonGM) for 15% and 425 (GM)to182(nonGM)for30%.Foreachsimulation,thestartingcropwasrandomlyderivedfrom theavailablecropswithinthemanagedrotation.Simulationsexaminedtheimpactontemporaland spatialgeneflowacrossthelandscapeof: x x x Alternativecroppingrotations; Hedgerowmanagement; Improvedherbicideefficiencyduringcerealcrops Inbrief,GENESYSconfirmedthattheinclusionofsuccessivewinterwheatcropsfollowingaGMHT oilseedrapecrop(Scenario1OSR/WW/WW/WW)presentsahighriskscenariothatwillnegatethe potential for efficient coexistence at a regional level. Extending the duration of the rotation to 6 years(Scenario3–OSR/WW/WW/WW/WW/WW)providednosignificantdifference(P>0.05),with 16.46%, 47.82% and 85.48% of fields possessing GM admixture >0.9% for a 5%, 15% and 30% regional adoption respectively. In contrast, the substitution of winter wheat with a spring barley crop in year 3 and 4 of both Scenario 1 and 3 reduced the harvest impurity 4.1fold for a 4 year rotation(Scenario2–OSR/WW/SB/SB)and1.5foldfortheextended6yearrotation(Scenario4– OSR/WW/SB/SB/WW/WW). The influence of an alternative spring crop to barley was assessed by simulating the impact of potato (Scenario 5 – OSR/WW/POT/SB) or maize (Scenario 6 – OSR/WW/MAIZE/SB) management in the third year of the four year rotation. In both cases, the levels of gene flow across the landscape were comparable to Scenario 2, with neither maize nor potatocultivationdecreasingthedegreeofharvestimpuritiesorthe%offields>0.9%foranyofthe threeadoptionlevels.  141 CoExtraInternationalConference Theintroductionoffieldbordermanagement(cuttingand/orherbicidetreatmentinMay)reduced thelevelofharvestimpurityinneighbouringnonGMoilseedrapecropsfrom3.5%toapproximately 2% (for OSR/WW/WW/WW rotation – 30% GM adoption) but this approach may conflict with existing EU environmental directives via REPS. By increasing the herbicide efficiency in a single applicationthroughwinterwheatcropsatseedlingstage,harvestcontamination(HC)didnotexceed the 0.9% threshold at either the 15% (0.39% HC) or the 30% (0.76% HC) regional adoption level. Includingaherbicidetreatment(95%efficiency)atadultvolunteerstageprovidedforacomparable decreaseintransgeneflowrelativetothesingleapplicationwith95%herbicideefficiency.Critically, for % HC (5% regional uptake) there was no difference between adopting a spring crop rotation (Scenario 2, 5 or 6) and improving the efficacy of volunteer control in the standard winter wheat rotation(Scenario1)viaenhancedherbicideefficiencyorincreasingthenumberofapplications. Thesignificanceoftheseresults,alongwithdatasetsfromadditionalsimulationswillbediscussedin light of developing an optimal coexistencebased management strategy for the potential early adoptersofGMHToilseedrapeinIreland.                        142 CoExtraInternationalConference P33. BrazilianGMOFreeAreasExperimentandtheReleaseofRRSoybeans RoseliRochadosSantos,AnaPaulaMyszczuk,FredericoGlitz UniversidadeEstadualdoParana,Brasil.   This presentation aims to examine how the release of the Roundup Ready soybeans commercialization has been done in Brazil and the resistance of the organized society to this fact. Forthat,theBrazilianlegislationonBiosafetyisconsidered,astheapplicationsprocessforrelease and commercialization of GM soy by Monsanto to the National Biosafety Technical Commission (CTNBio).Thelawsuitsmadebytheconsumers’organizationstopreventsuchreleaseandthemain Courtdecisionsonthiscasearealsoexamined. Inconclusionitwillbeverifiedthat,althoughorganizedsocietyandvariousstatesoftheFederation tried to establish a prohibition on the use of GMO or create some GMOs free area, the pressure from farmers and industry and the lack of effective supervision of the Federal Government, has allowedtheRRsoybeantobereleasedandspreadinallcountry.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.                     143 CoExtraInternationalConference P34. AbibliometricsapproachonSoybeanResearchinBrazil RoseliRochadosSantos,EvelinLucht UniversidadeEstadualdoParana,Brazil.   Brazilian nonGMO soybean production faces challenges, especially with seed. With the big emphasis on GM soybeans in recent years, private and public seed research have focused their breedingeffortsonGMvarieties.SeveralactorsarecomplainingaboutthenonGMimprovements scarcityinBrazil.Isittrueornot? The aim of this research is to present the findings reached by the research into the scientific and academic Brazilian production about research on soybean genetic plant breeding in the period of 2000 to 2009, and to classify them according to quantity, chronology of publication, author’s function, source, kind of study, topic, and key words. The research has been carried out taking a bibliometricsmethodology.Theworksanalyzedconsistedofallpapersabouttheissuethatreached thehighestgradeintheCapes,orA,andalsofromanetworkofinstitutionsthatstudytheissue,like Embrapa, Ocepar/Coodetec, FT, Indusem, Cotia, FECOTRIGO, IPAGRO, EMGOPA, EPAMIG, EMPA, EPABA,EMAPA,EMPASC,EMPAER,IPAGRO.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.                 144 CoExtraInternationalConference P35. TheAgroindustrialChainofSoybeaninBrazil:BriefNotesonthe ContractofSale RoseliRochadosSantos,AnaPaulaMyszczuk,FredericoGlitz UniversidadeEstadualdoParana,Brazil.  This presentation aims to do a juridical analysis on the commercialization of soybeans that is performedinBrazil,throughnetworkscontracts.Suchoperationsareconductedinawellstructured andcomplexagroindustrialjuridicalsystem,withtheestablishmentofcontractualrelationsamong severalagentsofthesoybeansupplychain.Oneofthemostusualcontractsusedinthisnetworkis the contract of soybeans sale. The form and objectives of these agreements will depend on the economicneedsinvolved.However,itiscertainthatBraziliancourtshaverecognizedthecontractual practicesthattookplaceonthesetopics.  Acknowledgements: This study was financially supported by the European Commission through the Integrated Project CoExtra, Contract No. 007158, under the 6th Framework Programme, priority 5, food quality and safety.                      145 CoExtraInternationalConference P36. Time Requirements and Financial Expenditures for Coexistence Measures and Their Impact to Profitability of Genetically Modified PlantsinSwitzerland DanielWolfandGregorAlbisserVögeli AgroscopeReckenholzTänikonResearchStationART,Tänikon,CH8356Ettenhausen. Email:[email protected].  The ban on the commercial cultivation of genetically modified plants (GMPs) in Switzerland is expected to stay in force until 2013. Should cultivation be authorised in the medium term, the decision whether or not to grow GMPs will be up to the farmers. As in the EU, coexistence regulations for protecting conventional and organic crop farming would then also have to be observedinSwitzerland,witharesultantimpactonprofitability.Thispaperthereforepresentsinitial time, effort and cost calculations for Switzerland for the most import measures in this regard. In order to cover the potential ranges, they are specifically calculated – to varying degrees (mildto severerestrictions)ortechnicalfeasibility–forthestructuralconditionsofSwissagriculture.Table1 givestheinitialresultsfortimerequirementsandtheestimatedcosts. In order to clarify the extent to which GMP crops are profitable under Swiss conditions, gross marginsforcornborerresistantmaize,herbicidetolerantmaizeandoilseedrapearecalculatedfor different model farms, with special consideration being given to agricultural structures and farm conditionsaswellas the price levelinSwitzerland.Theseprofitability calculationsalsoencompass the temporal and financial effects of the coexistence measures. The profitability analyses are specifically calculated to varying degrees (farm size, agricultural structure) for the structural conditionsofSwissagriculture.Inaddition,otherpotentialinfluencingfactors(seedprice,changes in yield, etc.) are varied in order to illustrate the range of potential profitability for Switzerland, whilstkeepingthebasicpolicyconditions(i.e.thecoexistenceregulations)constantforthesakeof better comparability. In order also to highlight the sensitivity of the influencing factors to the profitability of GM plants, the calculation is also supplemented with a Monte Carlo simulation, allowingforidentificationofthecriticalinfluencingvariables. Initial results show that the cultivation of Bt corn becomes economically worthwhile compared to the cultivation of nontreated maize once there is a lighttomoderate (1025%) cornborer infestation,providedthattheseedpricepremiumdoesnotexceed25%.ForsmallSwissfarms(<15 haarableland)thecultivationofBtcornisonlyprofitable,ifthecornborerinfestationisstrong. 146 CoExtraInternationalConference Table1: Estimateofpossibletimerequirementsandfinancialexpenditureforpotentialcoexistence measuresinSwitzerland Timerequirements Financialexpenditure Comments Measure unfavourable 15min/field 80min/field 41min/field 295min/field 9min/year 84min/year 7CHF/year 294CHF/year sower 35min/year 210min/year 23CHF/year 1790CHF/year combineharvester 13min/field 36min/field 6min/year 9min/year * Monitoringof volunteers* unfavourable 50m 5CHF/field Isolationdistance Specialmachine cleaning** favourable  favourable 20CHF/field 300m none Percheck Onlineregistration 0.50CHF/year 2.00CHF/year Permission 34min/year 46min/year 50min/year 83min/year 5CHF/year 10CHF/year Written notificationof5 neighbours Informationand training ½day 2days 8CHF/year 28CHF/year Trainingonceevery 10yearsonly Documentation 25min/year 38min/year 3CHF/year 15CHF/year  Notificationof neighbours WrittenPermission *Dependingonstatutoryregulation,cropandagriculturalstructure *Dependingontechnicalcircumstances(cultivationtechnique,cropandmechanisation(ownmachinery,hiredmachinery, agriculturalcontractor))andthecoexistenceregulationsinforce  The cultivation of herbicidetolerant maize and oilseed rape has an economic advantage in labour productivity for medium and large Swiss farms. Restrictions are that the weed pressure is low or moderate(noadditionalspecialherbicidesexceptfortotalherbicidesareused)andifthecultivation systemischangedfromploughingtononploughtillage.Otherwisethelabourandfinancialsavings in herbicide use do not compensate the expenditures for coexistence. At all events, large Swiss farms(<35haarableland)withroundedofffieldsandlargescaledagriculturalstructurecouldhave the same gross margin and better labour returns in ploughing systems both for conventional and herbicidetolerantculturesandiftheseedpricepremiumdoesnotexceed30%. 147 CoExtraInternationalConference 148 CoExtraInternationalConference    149 Thankyoutothefollowingsponsorsfortheirgeneroussupportofthe CoExtraFinalConference:  Biochip Systems