Discovery and Characterization of Field Resistance to Bt Maize: Spodoptera frugiperda (Lepidoptera: Noctuidae) in Puerto Rico Author(s) :Nicholas P. Storer, Jonathan M. Babcock, Michele Schlenz, Thomas
Meade, Gary D. Thompson, James W. Bing and Randy M. Huckaba
Source: Journal of Economic Entomology, 103(4):1031-1038. 2010.
Published By: Entomological Society of America
DOI: dx.doi/10.1603/EC10040
URL: www.bioone/doi/full/10.1603/EC10040
BioOne (www.bioone) is a nonprofit, online aggregation of core rearch in the
biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, muums, institutions, and
press.
Your u of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of U, available at www.bioone/page/terms_of_u.
Usage of BioOne content is strictly limited to personal, educational, and non-commercial u. Commercial inquiries or rights and permissions requests should be directed to the
individual publisher as copyright holder.
BioOne es sustainable scholarly publishing as an inherently collaborative enterpri connecting authors, nonprofit publishers, academic institutions, rearch libraries, and rearch funders in the common goal of maximizing access to critical rearch.
F ORUM
Discovery and Characterization of Field Resistance to Bt Maize: Spodoptera frugiperda(Lepidoptera:Noctuidae)in Puerto Rico
NICHOLAS P.STORER,1JONATHAN M.BABCOCK,MICHELE SCHLENZ,THOMAS MEADE, GARY D.THOMPSON,JAMES W.BING,AND RANDY M.HUCKABA
Dow AgroSciences LLC,9330Zionsville Road,Indianapolis,IN46268
J.Econ.Entomol.103(4):1031Ð1038(2010);DOI:10.1603/EC10040
ABSTRACT Transgenic maize,Zea mays L.,event TC1507produces the Cry1F protein to provide protection from feeding by veral important lepidopteran pests,including Spodoptera frugiperda(J.E. Smith)(Lepidoptera:Noctuidae).Reports of reducedÞeld performance against this species in Puerto Rico were investigated,and laboratory bioassays showed that S.frugiperda collected from the affected area exhibited lower nsitivity to the Cry1F protein compared with typical colonies from other regions.The resistance was shown to be autosomally inherited and highly recessive.The Puerto Rico colony was shown to be moderately less nsitive than susceptible laboratory strains to Cry1Ab and Cry1Ac,but the differences in nsitivity were dramatically smaller than for Cry1F.Potential con-tributory factors to the emergence of resistance to Cry1F in Puerto Rico populations of S.frugiperda include the tropical island geography,unusually large population sizes in2006,and drought conditions reducing the availability of alternative hosts.In respon to this resistance incident,the technology providers have stopped commercial sales of TC1507maize in Puerto Rico pending potential reversion to susceptibility.
KEY WORDS Cry1F,transgenic maize,fall armyworm
Transgenic maize,Zea mays L.,and cotton,Gossypium hirsutum L.,crops containing genes from Bacillus thu-ringiensis Berliner(Bt)that express incticidal pro-teins to control speciÞc target pests have been widely deployed in the United States and globally since1996. Since their inception,it has been recognized that target pest populations could evolve resistance to the Bt proteins,thereby reducing the cropsÕlong-term utility(Gould1988,1994,1998;Van Rie1991;Mallet and Porter1992;Roush1994).Resistance manage-ment programs have been implemented wherever the crops have been commercially cultivated in an at-tempt to reduce the rate at which pests may adapt (Sivasupramaniam et al.2007).
TC1507maize expressing the Cry1F protein from B. thuringiensis variety aizawai was registered in the United States in2001for u to control or suppress important primary and condary lepidopteran pests, including fall armyworm,Spodoptera frugiperda(J.E. Smith)(Lepidoptera:Noctuidae)(Siebert et al. 2008a,b).Maize expressing the Cry1F incticidal Bt protein has been grown in Puerto Rico since Novem-ber1996,initially in experimental plots and later also for hybrid development,parent ed production and from2003on commercial silage and dairy farms.S. frugiperda is the most important pest of maize in Puerto Rico,with frequent heavy populations from continuous overlapping generations.In addition,si-lage maize is produced year-round with monthly s
e-quential plantings.Although not well documented, resistance or reduced nsitivity to sprayable products including microbial Bt formulation are known(Euro-pean Plant Protection Organisation;www.eppo. org/QUARANTINE/incts/Spodoptera_frugiperda/ LAPHFR_ds.pdf).Although TC1507maize was not actively marketed on Puerto Rico,silage growers found it provided a safe,economic,and highly effec-tive tool to manage this pest and quickly deployed it on a large percentage of the planted acres in a con-tinuous production scheme.
As part of the product stewardship of TC1507,the registrants,Dow AgroSciences and Pioneer Hi-Bred International,have tracked product performance in theÞeld by investigating grower reports of less than expected results against the target pests and by di-rectly obrving performance of the trait in hybrid development.In late2006,there were reports from commercial farms and ed companyÞeld stations in Puerto Rico of unusual damage to Cry1F maize hy-brids.Several possible reasons for this damage were considered.Cry1F maize is known to not be immune to S.frugiperda feeding as larvae must consume plant tissue before being affected and large numbers can result in obrvable damage.This is especially true when older instars move to the Bt plants from nearby grassy weeds or maize not transformed to express
1Corresponding author,e-mail:
0022-0493/10/1031Ð1038$04.00/0᭧2010Entomological Society of America
Cry1F protein.Larger larvae consume more plant ma-terial than do smaller larvae before they are affected by the Bt toxins,and they survive longer on the plants. Unusually large populations of S.frugiperda had been recorded that year in Puerto Rico causing extensive damage to conventional maize,and a vere drought from October2006to April2007forced the popula-tions to become concentrated in irrigated crops,of which Cry1F maize was an important component.The high pest pressure probably also rved to further degrade the suitability of alternative host plants and increa the dispersal of adults and larvae to higher quality hosts,including especially inct-protected and irrigated maize.This movement was in addition to the normal larval movement from harvested silage crops.It was therefore considered possible that pest pressure alone could be responsible for the damage obrved in Cry1F maizeÞelds.Furthermore,the maize hybrids involved were temperate varieties not adapted to the tropical environment,leading to some questions as to whether native host plant resistance to S.frugiperda was abnt(normal expression levels of the Cry1F protein were conÞrmed using quantitative (Lepidoptera:Noctuidae)).However,a shift in pest population susceptibility to Cry1F could not be ruled out,so samples of inct populations were collected, and the progeny were tested for Cry1F nsitivity in laboratory bioassays.Here we conÞrm high-levelÞeld resistance to a Bt crop resulting in loss of efÞcacy of the Bt trait underÞeld conditions.
Materials and Methods
Inct Collections.Samples fromÞve S.frugiperda populations were collected from Puerto Rico(PR) (Table1).One population,designated SI,was col-lected from TC1507maize near Santa Isabel in April 2007.This location was also the source of a cond collection(SI2)in February2008.Another popula-tion,designated VEH,was collected from a commer-
cial TC1507maize planting in April2007.A population
designated JD was collected from weed grass in a
non-Bt maize planting in August2007,and another
population,designated SG,was collected from vege-
tables grown in a predominantly vegetable-growing
area near Sabana Grande in February2008.All col-
lections sites from PR were located on the southern
coastal plain parated by a maximum linear distance
of65km.
Two populations from outside Puerto Rico were
ud as control populations(Table1).The population
designated IN was collected from non-BtÞeld maize
at the University of Georgia,Plant Sciences Farm near
Athens,GA,in June of2005and has been cultured at
the Dow AgroSciences inctary in Indianapolis In-
diana since that time.The Benzon population is main-
tained commercially by Benzon Rearch Inc.(Car-
lisle,PA)and was acquired from Ecogen Inc.,
(Langhorne,PA)in2000.Ecogen had previously ob-
tained the population from the United States Depart-
ment of Agriculture Lab in Stoneville Mississippi Ϸ1990.Since1990,this population has had no history of exposure to incticides.Larvae ud for bioassays爱美女人
were hatched from eggs obtained from the IN popu-
lation as needed.
In addition,two populations from the continental
United States that have been expod to Cry1F ex-
presd in Bt cotton and Bt maize underÞeld condi-
tions more typical of U.S.agricultural production sys-
tems were collected for evaluation of nsitivity to
Cry1F(Table1).One population was collected Sep-
tember2007from non-Bt maize at the Coastal Re-
arch Center,Greenville,MS,and one was collected
in September2007from non-Bt maize planting at the
University of Georgia,Plant Sciences Farm.
Populations cultured at Dow AgroSciences were
initiated from collections of at least100individuals
each with the exception of the SI2collection,which
was initiated from75pupae.SI and VEH populations
were each initiated by rearingÞeld collected larvae to
adulthood from which progeny were hatched en
mas onto inct diet treated with10,000ng/cm2of
truncated Cry1F protein.Little mortality or stunting
was obrved and larvae were reared until they
reached third stadium at which time they were trans-
ferred to untreated diet to complete development.All
other populations were colonized directly fromÞeld-
collected larvae or eggs and maintained on diet with
no exposure to Cry1F.
Test Materials.Cry1F protein(sample TSN104550,
35%active ingredient)and Cry1Ac protein(sample
TSN10259114%active ingredient)proteins were het-
erologously produced by Dow AgroSciences in Pudomonasfluorescens and concentrated as lyoph
i-lized powders.The Cry1Ab protein sample was de-livered as a puriÞed protein solution of375g/ml concentration.Bioassay solutions were prepared by adding protein to10mM3-(cyclohexylamino)pro-panesulfonic acid(CAPS),pH10.5,buffer to gen-erate the highest tested diet overlay concentrations of either10,000or3,333ng/cm2.Lower concentra-
Table1.Source,collection date,and founding pop size of collections of(S.frugiperda)populations evaluated for nsitivity to Cry1F protein overlaid on artificial diet
Pop code Collection location
Date of
collection
N
(initiation)
SI and SI2Santa Isabel PR April2007and
July2007
163and75
VEH VEH Farms PR April2007113
JD ICIA farm Juana Diaz PR Aug.2007125
SG Sabana Grande,PR Feb.2008125
Benzon USDA,Stonesville
Mississippi via Ecogen,
Inc.,Langhorne,PA
1990n/a a
IN University of Georgia,
Plant Sciences Farm
Athens,GA
June2005100
MS Dow AgroSciences
Costal Rearch
Center,Greenville,MS
Sept.2007100
GA University of Georgia,
Plant Sciences Farm
Sept.2007100
a Information on the collection size of the Ecogen colony is not
available.
插入1032J OURNAL OF E CONOMIC E NTOMOLOGY Vol.103,no.4
tions of each protein were generated by three-fold rial dilutions with additional10mM CAPS,pH 10.5,buffer.
Diet Overlay Bioassays.Warm multispecies inct diet(Southland Products,Lake Village,AR)was dis-pend into standard128-well bioassay trays(Bio-Serv,Frenchtown,NJ)at a rate of1.0ml/well.The surface area of diet in each well was1.5cm2.Each test well was treated by applying40l of liquid sample onto the diet surface.When all of the wells within a tray were treated,each tray was tilted from left to right and front to back to ensure that the liquid sample completely coated the surface of the diet.Both un-treated(no buffer)and untreated buffer checks were included in the assay.After application,the trays were placed into a biosafety chamber and the samples were allowed to dry.Careful oversight of the sample drying process ensured that the diet did not desiccate and crack or pull away from the edges of the well thereby exposing untreated diet.
A minimum of16,and up to32replicate wells were treated for each protein concentration by population combination per test date.Tests were further repli-cated over a minimum of three parate test dates.At leastÞve concentrations were ud for each popula-tion bioassay,with the highest concentration being 3,333ng/cm2in most bioassays(although one t was run at the higher concentration of10,000ng/cm2). Neonate larvae that had been allowed to feed on untreated diet for2
Ð4h were individually transferred by hand to the treated diet wells at a density of one per well using aÞne artistÕs paintbrush.Air-permeable lids were placed over the wells after infestation.The bio-assay trays were incubated at28ЊC(photoperiod of 16:8[L:D]h)for7d after which the incts were graded for mortality(larvae showing no movement of any body part when prodded).For each replicate,live larvae from each inct colony on each concentration of each protein were pooled and weighed after7d. Average larval weight was calculated bad on the number of tested larvae.Growth inhibition was cal-culated by dividing the treatment average surviving larval weight per inct bioassayed by the average surviving larval weight per inct bioassayed in the
buffer check and subtracting from1.The values were then expresd as a percentage.
Data Analysis.Mortality data for each inct strain on each protein were analyzed by log-probit analysis to estimate the LC
50
(concentration required to kill 50%of the incts)for each population.Linear re-gression was ud toÞt probit-transformed growth inhibition data to log-transformed protein concentra-tion;the concentration required to cau50%growth
inhibition(GIC
50
)was estimated for each inct strain.All analys were conducted in MINITAB ver-sion12.21(Minitab Inc.,State College,PA). Inheritance.Cry1F nsitivity in the SI colonies indicated high survival of neonate larvae at a10,000 ng/cm2challenge do ud at the time of laboratory colonization.(This conclusion was subquently con-Þrmed using do respon[Table2].)To evaluate inheritance of this respon to Cry1F,the F1progeny from reciprocal cross-mass matings of the SI popula-tion with the IN susceptible population(SI(mated with IN&and IN(mated with SI&)were bioassayed, and the resulting doÐrespon statistics were com-pared.
Dominance Asssment.Mortality and growth in-hibition data for IN S and SI R populations were ud to calculate dominance of resistance using the meth-ods for effective dominance at aÞxed concentration
(D
ML
)
in Bourguet et al.(2000):D
ML
ϭ(ML
F1
ÐML
IN
)/(ML
SI
ÐM L
IN
),where ML
IN
,ML
SI
,and ML
素丸子F1 are the mortality levels at3,330ng Cry1F/cm2for the IN population,SI population,and F1progeny of the populations,respectively.Estimates of D
LC
(e.g.,
bad on LC
50
values for the three genotypes)could not reliably be assd becau signiÞcant mortality or growth inhibition of the SI population did not occur at the top concentration tested in the replicated bi
o-
assays and LC
礼仪之争50
values could not be calculated.D
LC values are in any ca of little value for resistance management(Bourguet et al.2000).A cond mea-sure of effective dominance,here termed D
GIL
,was
calculated using the methods for D
ML
but applying the growth inhibition endpoint rather than the mortality
Table2.Concentration-respon offield and laboratory populations of S.frugiperda to Cry1F protein overlaid on artificial diet,as measured by both mortality(LC50)and growth inhibition(GIC50)
Pop
No
replicates a
Mortality Growth inhibition
SlopeϮ
SE
LC50b
(95%CL)
Sensitivity
ratio c
2SlopeϮ
SD
GIC50b
(95%CL)
卷上珠帘总不如Sensitivity
ratio c
R2
(%)
SI6nc dϾ10,000Ͼ34.6nc ncϾ10,000Ͼ356nc VEH2Ͼ10,000Ͼ34.6nc ncϾ10,000Ͼ356nc JD1ncϾ3,333Ͼ11.5nc ncϾ3,333Ͼ119nc SG1ncϾ3,333Ͼ11.5nc ncϾ3,333Ͼ119nc MS1 1.26Ϯ0.14420(286Ð648) 1.49.7 1.44Ϯ0.1121.8(14.1Ð33.6)0.7897.1 GA1 1.68Ϯ0.19267(193Ð373)0.92 4.8 1.24Ϯ0.2814.0(2.74Ð71.1)0.5077.2 Benzon5 2.47Ϯ0.20428(367Ð
495) 1.520.8* 1.29Ϯ0.1219.7(11.4Ð30.4)0.7083.2 IN5 1.06Ϯ0.14289(223Ð355)26.3* 1.69Ϯ0.1828.1.(14.7Ð46.0)77.4
*SigniÞcant at PϽ0.01.
a Each replicate consisted of16Ð32incts at each ofÞve to ven concentrations of Cry1F protein.
b Nanograms of Cry1F/cm2diet.
c LC
50or GIC50relative to IN population.
d nc,not calculated du
e to insufÞcient do respon.
August2010S TORER ET AL.:R ESISTANCE TO B T M AIZE IN S.frugiperda IN P UERTO R ICO1033
endpoint.D
卡尔森酒店ML and D
GIL
were calculated at3,333.3
ng/cm2becau at this concentration mortality and growth inhibition of the susceptible strains were Ͼ99%(butϽ100%)for the IN population but very low
for the SI population,similar to the effects of TC1507 maize en on susceptible and resistant S.frugiperda populations in theÞeld.
Population Sensitivity to Cry1Ab and Cry1Ac Bt Proteins.The IN and SI populations were evaluated for mortality and growth inhibition resulting from exposure to Cry1Ab and Cry1Ac protein using the bioassay methods described above.The proteins were evaluated to understand nsitivity in the SI population to other Bt proteins that have been ud in Puerto Rico.Sensitivity ratios were calculated using concentrationÐrespon statistics bad on growth in-hibition(estimated as above)with the resistant pop-
ulation GIC
50as the numerator and susceptible pop-
ulation GIC
50as the denominator.When growth
inhibition was notϾ50%at the highest protein con-centration tested,the concentration respon statistic ud for generating nsitivity ratios was this highest rate.A nsitivity ratio was regarded as signiÞcantly different from1.0(equal susceptibility)if there was no overlap in the95%conÞdence intervals(CI)of the estimate of the values.However,becau the strains are from different collections with different histories and different background genetics,the numbers should not be regarded as true resistance ratios.
TC1507Efficacy Trials.Inct efÞcacy of TC1507 maize was evaluated in1999before commercialization and in2009after unexpected damage was reported.In 1999,a line of TC1507was planted at the Mycogen SeedsÞeld station in Santa Isabel,PR,and subject to natural infestation.In total,1,233TC1507plants were rated for S.frugiperda damage to the whorl.Whorl damage was scored using a1Ð9rating scale,whereby a score of1indicates a whorl totally destroyed by inct feeding,and a score of9indicates no damage. S.frugiperda infestation was conÞrmed through ob-rvations to da
mage to non-Bt plants in the sameÞeld.
A replicatedÞeld trial(four replication in a random-ized complete block design,plot size of3by6m)was planted on18December2009and included a TC1507 line(stacked with Cry34/35Ab1event DAS-59122-7 conferring corn rootworm(Diabrotica spp.)protec-tion and EPSPS event NK603conferring glyphosate tolerance)and a near non-Bt isoline(event NK603 alone).The lines were included as part of a larger Þeld trial which included other maize lines.On2 January2010,S.frugiperda damage to10plants per plot was rated using the0Ð9Davis scale(Davis et al. 1992),whereby a score of0indicates no damage and a score of9indicates a whorl totally destroyed(note this is the rever of the scale ud in1999).The data from this trial were analyzed by analysis of variance (ANOVA),with means paration using the StudentÐNewmanÐKeuls test in ARM7(Gylling Data Manage-ment,Inc.,Brookings,SD).
Results
Survey of S.frugiperda Populations.All populations春笋炒肉片
sampled from Puerto Rico for this study were inn-sitive to Cry1F bad on bioassay results(Table2). The IN and Benzon populations were not signiÞcantly different in mortality or growth
什么肩接踵
respons to Cry1F bad on overlapping conÞdence limits for calculated mortality and growth inhibition statistics.Sensitivity ratios for Puerto Rico populations wereϾ34bad on mortality andϾ350bad on growth inhibition.Mor-tality and growth inhibition at the top concentration tested were both muchϽ50%for the SI population, compared with LC
50
and GIC
50
estimates of289and 28.1ng/cm2,respectively,for the IN population.The SI population produced a respon typical of the other Puerto Rico populations and therefore was an appro-priate choice for inheritance and dominance evalua-tions.The IN population was chon as a control for the remaining experiments as it is reprentative of the other continental populations and was readily avail-able from the Dow AgroSciences inctary. Populations collected in2008from the southern continental United States showed relatively uniform nsitivity to Cry1F in the bioassays.Their nsitiv-ity to Cry1F was not signiÞcantly different from the susceptible Benzon and IN populations bad on over-
lapping95%conÞdence intervals for LC
50
and GIC
50 estimates(Table2).
Inheritance of Resistance.There was no difference between LC or GIC statistics for reciprocal F1cross of IN S and SI R populations bad on overlapping conÞdence limits.Additionally a comparison of re-gression lines from reciprocal cross indicated that both slope and intercept were not different.Thus, the data and subquent data generated from the F1cohorts were pooled for analysis.Collectively, the results indicate autosomal inheritance of Cry1F in the SI R population(Table3).
Dominance.As a result ofÞnding no x linkage associated with the inheritance of Cry1F,F1data were pooled.Figure1shows the Cry1F concentrationÐre-spon curves for the three genotypes.There were differences in mortality and the F1population was signiÞcantly less nsitive than the IN population to
Cry1F at the LC
50
bad on a comparison of conÞ-dence intervals(Table3;Fig.1A).However,the growth inhibition respons of the pooled F1and IN populations were not signiÞcantly different bad on overlapping95%conÞdence intervals(Table3;Fig. 1B).The SI population was uniformly innsitive to Cry1F concentrations of up to3333ng/cm2(in one the one bioassay run at10,000ng/cm2,this population showed no signiÞcant mortality but inconsistent growth inhibition).No concentrationÐrespon sta-tistics were generated for this population.The calcu-
lation of D
ML
(and here also D
GIL
)summarized by Bourguet et al.(2000)generates values between0and 1,which indicate a range of res
pons from fully re-cessive to fully dominant,respectively.The evaluation of Cry1F-resistance bad on growth inhibition at 3,333ng/cm2indicated highly recessive resistance,
1034J OURNAL OF E CONOMIC E NTOMOLOGY Vol.103,no.4
