LlANG Jin-gang ,ZHANG Dan-dan ,Ll Dong-yang, ,ZHAO Sheng-yuan ,WANG Chen-yao ,XlAO Yu-tao,XU Dong,YANG Yi-zhong,Ll Guo-ping,WANG Li-li,GAO Yu,YANG Xue-qing,YUAN Haibin0,LlU Jian,ZHANG Xiu-jie,WU Kong-ming

1 Development Center of Science and Technology,Ministry of Agriculture and Rural Affairs,Beijing 100176,P.R.China

2 State Key Laboratory for Biology of Plant Diseases and Insect Pests,Institute of Plant Protection,Chinese Academy of Agricultural Sciences,Beijing 100193,P.R.China

3 State Key Laboratory of Cotton Biology,Institute of Cotton Research,Chinese Academy of Agricultural Sciences,Anyang 455000,P.R.China

4 Shenzhen Branch,Guangdong Laboratory for Lingnan Modern Agriculture,Genome Analysis Laboratory of the Ministry of Agriculture,Agricultural Genomics Institute at Shenzhen,Chinese Academy of Agricultural Sciences,Shenzhen 518120,P.R.China

5 Institute of Plant Protection and Soil Science,Hubei Academy of Agricultural Sciences,Wuhan 430064,P.R.China

6 College of Horticulture and Plant Protection,Yangzhou University,Yangzhou 225009,P.R.China

7 Institute of Plant Protection,Henan Academy of Agricultural Sciences,Zhengzhou 450002,P.R.China

8 Yantai Academy of Agricultural Sciences,Yantai 265500,P.R.China

9 College of Plant Protection,Shenyang Agricultural University,Shenyang 110866,P.R.China

10 College of Plant Protection,Jilin Agricultural University,Changchun 130118,P.R.China

11 College of Agriculture,Northeast Agricultural University,Harbin 150030,P.R.China

Abstract The fall armyworm (FAW) Spodoptera frugiperda,which originated in the Americas,is advancing across China and threatening the nation’s maize crops.Currently,one widely used tool for its control is genetically modified (GM) Bacillus thuringiensis(Bt) maize.Sufficient content of Bt protein in appropriate plant parts is crucial for enhancing resistance against insect pests.In this study,we conducted a systematic investigation of Cry1Ab levels in Chinese domestic GM maize DBN9936,which has recently obtained a biosafety certificate,and evaluated its efficacy against FAW.Quantification of expression levels of Cry1Ab,via ELISA,indicated a spatio-temporal dynamic,with significant variation of mean Cry1Ab,ranging from 0.76 to 8.48 µg g–1 FW with the Cry1Ab protein level ranked as:V6–V8 leaf＞R1 leaf＞R4 leaf＞R1 silk＞VT tassel＞R4 kernel. Among the nine locations,the Cry1Ab levels in DBN9936 of the Xinxiang,Langfang,and Harbin fields were significantly lower than those from Wuhan and Shenyang,and were slightly,but not significantly lower than those from the other four fields.Furthermore,the artificial diet–Cry1Ab mixture and plant tissue feeding bioassays revealed that DBN9936 has high efficacy against FAW.The insecticidal efficacy of different tissues against FAW larvae reached 34–100% with a descending order of lethality as follows:VT leaf＞R4 leaf＞R1 husk＞R1 silk＞VT tassel＞R4 kernel.Taken together,our results showed that Bt-Cry1Ab maize DBN9936 has potential as a promising strategy to manage FAW.

Keywords:fall armyworm,genetically modified maize,DBN9936,Cry1Ab expression,control efficacy

1.lntroduction

The fall armyworm (FAW),Spodopterafrugiperda(J.E.Smith) (Lepidoptera:Noctuidae),a serious invasive species that is notorious for its polyphagy,has a high capacity for migration and dispersal (Gonçalveset al.2020).In the US,FAW populations from overwintering areas (South Texas and South Florida) migrate annually into regions across the country (Nagoshiet al.2012; Huanget al.2014; Li X Jet al.2019).FAW is the primary pest affecting maize in tropical and subtropical regions of the Americas,inflicting maize yield losses of up to 57.6% in Brazil and 17–72% in Argentina (Eghrariet al.2019; Murúaet al.2019; Gonçalveset al.2020; Souzaet al.2020).More recently,it has become an invasive species threatening maize production in Africa and Asia,resulting in maize yield losses as high as 50%(Bothaet al.2019; Silver 2019; Gonçalveset al.2020; Koffiet al.2020).

Recent studies of estimated yield losses in Africa,combined across 12 major maize-producing sub-Saharan countries,indicated that between 4.1–17.7 million tons of maize,with a value of US$1.09–4.66 billion,will be lost annually due to FAW (Li X Jet al.2019).Farmers in Ethiopia estimated maize infestations by the FAW ranging from 24.1–39.4% (with yield reductions of 934 kg ha–1),whereas farmers in Kenya estimated maize infestations ranging from 38–53.9% (with yield reductions of 1 381 kg ha–1) (Kumelaet al.2019; Grooteet al.2020).In Zimbabwe,FAW abundance ranged from 13.7–33.3 larvae per 30 maize plants,with infestation exceeding 94% and leaf,silk and tassel damage levels ranging between 25–50%.The estimated decrease in maize grain yield was 58% (Chimwetaet al.2020).FAW first invaded China in late 2018 and spread rapidly (Sunet al.2021).The actual damage area was about 0.16 million ha,and the production loss of the FAW damaged area was controlled to within 5% in 2019 (Liet al.2020a; Wu 2020; Sunet al.2021).

In the Americas,Bt maize is one of the primary tools used to manage FAW populations (Jegeret al.2017; Souzaet al.2019).Among the Bt maize events,those expressing Cry1Ab,Cry1F,Cry2Ab2,Cry1A.105,and Vip3Aa20 proteins have been effectively used to control FAW (Bothaet al.2019; Xiao and Wu 2019; Moscardiniet al.2020).In the International Service for the Acquisition of Agri-biotech Applications (ISAAA) GM Approval Database (Dec.2019),there are 82,10,48,45,and 46 maize events expressing thecry1Ab,cry1F,cry2Ab2,cry1A.105,andvip3Aa20genes,respectively.However,FAW populations are notorious for rapidly developing resistance to Bt insecticidal proteins produced in genetically modified (GM) maize (Souzaet al.2019; Storeret al.2010).For example,it took only 3–4 years to detect resistance in FAW,since the Bt maize TC1507(producing Cry1F protein) was introduced into Brazil in 2008 (Fariaset al.2014) and Puerto Rico in 2003 (Storeret al.2010).Bt maize TC1507 was first grown commercially in 2005 in Argentina and potential FAW resistance was reported in 2013 (Chandrasenaet al.2018).Although evolution of field resistance in some FAW populations has been reported,Bt maize could reduce the growth and survival of FAW (Silvaet al.2018).Therefore,with limited control options against this pest,the effectiveness of Bt maize against FAW can be important for farmers in places where GM crops have not been widely used (Carzoliet al.2018).

In China,significant economic losses caused by lepidopteran maize pests have been estimated.Collectively,several pest species (Ostriniafurnacalis,Helicoverpa armigera,andMythimnaseparata) cause approximately 10% yield loss in spring maize,20–30% in summer maize,with over 30% of these cases were classified as heavy infestations (Liet al.2020a).Since the implementation of the National GM Variety Development Special Program in 2008,numerous Bt maize events have been developed to effectively control lepidopteran pests,primarily expressingcry1Ab,cry1Ac,cry1Ie,cry1Ah,cry1C,orcry2genes,with some genes modified fromcry1Abandcry1Ia1(cryFLIa)orcry1Ac(mCry1Acandcry1AcM),and some fusion genes,such ascry1Ab/2Aj,cry1Ab/vip3DA,andcry1Ah/cry1Ie(Liuet al.2016; Liet al.2020a).In 2020,Ministry of Agriculture and Rural Affairs of China (MARA) issued biosafety certificates for domestic GM soybean and maize for the first time since 2009.GM insect-resistant (IR)and herbicide-tolerant (HT) maize DBN9936,which was developed by Beijing Dabeinong (DBN) Biotechnology Co.,Ltd.,was one of the newly approved maize hybrids(http://www.moa.gov.cn/ztzl/zjyqwgz/).It was engineered to expressepspsandcry1Abgenes conferring tolerance to the herbicide glyphosate and resistance to lepidopteran pests (Li P Get al.2018).At present,Bt maize shows good application prospects for controlling FAW in China (Zhanget al.2020).In addition,previous research demonstrated that Bt maize C0030.3.5 (i.e.,DBN9936) has a high level of resistance against FAW based on a laboratory bioassay(Zhang and Wu 2019).

Bt proteins are produced in GM crops in a tissue-and time-specific manner (Bakhshet al.2011; Székácset al.2012).Several concerns have been raised regarding the unpredictability or erratic expression of Bt proteins in plants,which is correlated with their insecticidal efficacy against insect pests (Székácset al.2010a; Wang Fet al.2014;Erasmuset al.2019).For Bt maize to be sustainable,it is vital that the Bt protein is expressed in adequate quantities in appropriate plant parts at the requisite time of the growing season.Plant tissue and plant development have been proposed as the main parameters affecting the Cry1Ab contents of GM maize MON810 (Ramirez-Romeroet al.2008).However,it is unclear whether Bt protein levels translate directly to the effectiveness of Bt crops against insect pests (Girón-Calvaet al.2020).Merely estimating Cry protein concentrations gives no indication of the biological activity of the measured protein (Svobodováet al.2017b;Lohnet al.2020).Therefore,in this study,integrated measurements were used to investigate the performance of DBN9936,including evaluating the spatio-temporal expression level of Cry1Ab protein in DBN9936 at nine field locations throughout eastern China,the effect of a simulated artificial diet containing different concentrations of Cry1Ab on FAW,and the effect of feeding DBN9936 tissues to FAW under laboratory conditions.

2.Materials and methods

2.1.Quantitative analysis of Cry1Ab expression

Plant materials Two GM maize varieties,DBN9936 and DBN9858,were used in the study.DBN9858,a non-Bt isoline of DBN9936,served as a non-Bt control.DBN9936 plants express two genes:cry1Abdriven by the constitutive CaMV35s promoter andepspsdriven by the rice OsAct1 promoter.DBN9858 plants also express two genes:epspsdriven by the rice OsAct1 promoter andpatdriven by the constitutive CaMV35s promoter.GM maize DBN9858 passed regulatory approval and received biosafety certificates in July 2020.The seeds of DBN9936 and DBN9858 were both provided by DBN.

Planting designThe two hybrids were grown using best regional practices at nine locations in China in 2019.These nine locations are Pu’er (22°40´N,101°38´E),Yunnan Province; Wuhan (30°44´N,114°46´E),Hubei Province;Yancheng (33°52´N,120°20´E),Jiangsu Province; Xinxiang(35°10´N,113°41´E),Henan Province; Yantai (37°29´N,121°16´E),Shandong Province; Langfang (39°51´N,116°60´E),Hebei Province; Shenyang (41°49´N,123°33´E),Liaoning Province; Changchun (43°48´N,125°24´E),Jilin Province; and Harbin (45°44´N,126°42´E),Heilongjiang Province.Six treatments were planted in a randomized complete block design,and replicated three times at each site.The treatments were a 100% DBN9858 control maize and 100% DBN9936 Bt maize,plus four seed mixtures of non-Bt treatments,mixed with 95,90,85,and 80%DBN9936 and 5,10,15,and 20% DBN9858,respectively.Among the traits in DBN9936,only the lepidopteranactive protein Cry1Ab was measured in this study,while the Roundup Ready herbicide-tolerance trait was not considered relevant.Each plot was 200 m2with a 1–1.5 m plot spacing,and conventional agronomic practices were used at all locations.

Field samplingAccording to the National Information Management and Support System (NIMSS) Project(University of Maryland,Maryland,USA),FAW eats maize leaf,tassel,silk,and kernel.V6–V8 leaf,R1 leaf,R4 leaf,R1 silk,VT tassel,and R4 kernel were collected from each replicated plot at each site (Abendrothet al.2011).When≥50% of the plants in the plot were at a specified growth stage,all samples were collected from randomly selected,healthy,representative plants for each hybrid.

For DBN9936 sampling,the Cry1Ab/Cry1Ac Test Strip(EnviroLogix,Portland,USA) was used to remove the non-Bt plants.At each sampling,V6–V8 leaf,VT tassel,R1 leaf,R1 silk,R4 leaf,and R4 kernel samples were collected from three randomly selected plants for each plot and pooled in a pre-labeled polythene bag,which was kept on dry ice until it was transferred to a freezer (–80°C).For DBN9858 sampling,non-Bt plants were collected from the control plot of each site,and they were used as a negative control.The tissue sampling details are given below:

1) Leaf.The youngest leaf that emerged and was at least 20 cm in length was cut from the whorl.The leaf was cut approximately 20 cm from the leaf tip.Then,the leaf sample (including the midrib) was cut into small sections(approximately 2.5 cm or smaller).

2) Tassel.The tassel was extracted from plants and then each individual tassel sample was cut into small sections(approximately 2.5 cm or smaller).

3) Silk.Previously bagged ear shoots were selected.The selected ear with the shoot bag was removed from the plant and moved to a pollen free environment to obtain the silk.The silk was cut from the ear into small sections(approximately 2.5 cm or smaller).

4) Kernel.One expression kernel sample was collected from one ear and the ear was shelled.Then,a representative sub-sample was collected from each group of 15 kernels.

Cry1Ab ELlSAThe contents of Cry1Ab protein of GM maize tissues were quantified using a sandwich ELISA with the Cry1Ab/Cry1Ac Quantiplate Kit (Envirologix,Portland,USA),performed according to the manufacturer’s instructions.All tissue samples were ground into lyophilized powder using a Retsch MM400 tissue lyser (Retsch GmbH,Haan,Germany),then stored in a freezer (–80°C) until needed.The extraction buffer phosphate-buffered saline Tween-20 (provided with the kit) was added to lyophilize the tissue powder at a ratio of 1:10 (mg sample:µL buffer).The extracted samples were centrifuged at 4°C,and the supernatants were collected and mixed for five treatments of DBN9936,which were further diluted into proper concentrations for ELISA assays.The optical density (OD) values were read at 450 nm with an Infinite M200 Pro Microplate Reader (Tecan,Männedorf,Switzerland).For statistical modeling,simple regression analysis was carried out in Microsoft Excel 2016 (Microsoft Corp.,Redmond,USA).Data were presented as the amount of protein in fresh weight (FW) of V6–V8 leaf,VT tassel,R1 leaf,R1 silk,R4 leaf,and R4 kernel.Standard curves were obtained at concentrations of 1,0.82,0.64,0.46,0.28,and 0.1 µg L–1of Cry1Ab standards (linear regression).

2.2.Uptake of Cry1Ab by FAW during artificial dietincorporation bioassay

Specimens of FAW were sampled from Dehong Autonomous Prefecture,Yunnan Province of China in January 2019.The population was reared on the diet described below without any Bt proteins or insecticides in the laboratory.Larvae were reared to the pupal stage on an artificial diet of soybean and wheat,and the pupae were kept separately in a glass tube for adult emergence.The adults were placed in cages(40 cm×25 cm×25 cm) and fed 5% sugar solution.The top of each cage was covered with white medical gauze for oviposition,and the gauze was changed every 24 h.The gauze containing eggs was placed into a zip lock bag.All pest colonies were maintained in an insect chamber with a controlled environment of (26±1)°C,(60±10)% RH,and a photoperiod of 16 h:8 h (L:D).

The Cry1Ab protein for bioassay was purchased from Case Western Reserve University (Cleveland,USA) and stored at–80°C.Susceptibility of FAW to Cry1Ab was determined using a diet incorporation bioassay in a 24-well plate.Treated diets contained a final concentration of 0,1,2,4,or 8 µg mL–1of Cry1Ab protein.Diets were then cut into small pieces (0.8 g/piece).Three replicates were used for each treatment and the control.Each well in a 24-well plate contained one piece of the corresponding diet and one neonate FAW.Based on head capsule and body size,the number of dead larvae in specific instar stages was counted after 6 days.A larva was considered dead if it was incapable of movement after being placed on its dorsal surface and prodded with a camel hair bush (Hardkeet al.2011).

2.3.Uptake of Cry1Ab by FAW during the plant tissue test

The FAW larval population used for the tissue-based bioassay was collected from maize fields in Jiangcheng,Yunnan (22°40´N,101°38´E) in January 2019.The population was reared constantly for 16 generations until April 2020.The two lines of GM maize (DBN9936 and DBN9858) used for the bioassay were planted previously in the experimental field of Jiangcheng Station,January 2020.Three replicate samples of different maize tissues,including VT leaf,VT tassel,R1 silk,R1 husk,R4 leaf,and R4 kernel,were sampled randomly from different growth stages of the two maize lines (DBN9936 and DBN9858),and the Cry1Ab/Cry1Ac Test Strip (EnviroLogix,Portland,USA) was used to identify the Bt maize.Different tissues of the same growth stage were sampled from the same plant of each maize line.

Newly hatched larvae of the F16generation were collected and divided randomly into two groups.They were placed in round transparent plastic boxes (10 cm in diameter and 8 cm in height) to infest different maize tissues of two maize lines.The tests were conducted in an environmental chamber maintained at (25±1)°C,(70±5)% RH,with a photoperiod of 16 h:8 h (L:D).Three repeats were set for each group,and each repeat contained 30 larvae for one specific tissue.The numbers of deaths and survivors of FAW larvae in different groups and tissue treatments were calculated on the 3rd and 5th days after infestation.

2.4.Statistical analysis

The corrected mortalities of FAW larvae were calculated using the following formula:Corrected mortality (%)=(Treatment group mortality–Control mortality)/(1–Control mortality)×100.Unpaired Student’st-tests were used to compare the means of two groups.One-way ANOVA was used to assess the significant differences between the groups.If the result from the ANOVA was significant,post hoc testing was performed using the LSD test (Denget al.2018).The data were generated as mean±standard error (SE),withP≤0.05 considered statistically significant.All statistical analyses were performed using SPSS 19.0 (SPSS Inc.,Chicago,USA).

3.Results

In total,the Cry1Ab protein levels of 156 samples of DBN9936 were determined for nine locations.The content of Cry1Ab in different tissues varied substantially among locations.Also,as expected,the content of Cry1Ab was below the limit of quantification for all tissues in the samples of non-Bt maize DBN9858.

3.1.Cry1Ab protein levels of DBN9936 in different tissues

At Pu’er,Wuhan and Yancheng,the average contents of Cry1Ab protein followed the pattern:silk＞leaf＞tassel＞kernel; there was no R4 samples in Pu’er.At Xinxiang,Langfang and Shenyang,the average contents of Cry1Ab protein in different tissues were ranked as leaf＞silk＞tassel＞kernel.At Yantai,Changchun and Harbin,the average contents of Cry1Ab protein were the highest in the leaf,followed by tassel,silk,and kernel (Table 1).

To obtain an overall seasonal expression pattern of Cry1Ab in DBN9936,statistical analysis was performed on data pooled across nine locations.Overall,as maize plants matured,the average content of Cry1Ab gradually decreased (P≤0.05).The highest Cry1Ab level was found in V6–V8 leaf,followed by R1 leaf,R4 leaf,R1 silk,VT tassel,and R4 kernel.The Cry1Ab level in the R4 kernel was significantly lower than that in all other tissues (Fig.1).For the temporal characteristics,the content of Cry1Ab protein was the highest at the V6–V8 stage (8.48 µg g–1FW) and the lowest at the R4 stage (3.42 µg g–1FW).For the spatial characteristics,the results indicated variable Cry1Ab content among different tissues,with the maximum content in the V6–V8 leaf (8.48 µg g–1FW),and the lowest content in R4 kernel (0.76 µg g–1FW) (Fig.1).

3.2.Cry1Ab protein levels of DBN9936 in different locations

The maximum/minimum fold changes of the Cry1Ab content from different locations in each tissue (V6–V8 leaf,VT tassel,R1 leaf,R1 silk,R4 leaf,and R4 kernel)were approximately 3,3,5,7,2,and 5,respectively(Table 1).The analysis of the mean content of Cry1Ab from the different tissues in different locations revealed no significant differences among Pu’er,Wuhan,Yancheng,Yantai,Shenyang,and Changchun,or among Pu’er,Yancheng,Xinxiang,Yantai,Langfang,Changchun,and Harbin.Among the nine locations,the Cry1Ab levels in DBN9936 of the Xinxiang,Langfang,and Harbin fields were significantly lower than those from Wuhan and Shenyang,and were slightly,but not significantly lower than those from the other four fields (Fig.2).

3.3.Effect of simulated artificial diet containing Cry1Ab on FAW

To simulate Cry1Ab expression in DBN9936,the effect of Cry1Ab on FAW was assessed in the artificial diet containing Cry1Ab protein.All four concentrations of Cry1Ab protein had strong toxicity to FAW (Fig.3).Percent corrected mortality of FAW (60.32–83.33%) increased with increasing concentrations of Cry1Ab protein (1–8 µg mL–1).The results also showed that the development of FAW neonates was clearly inhibited with increasing concentrations of Cry1Ab protein (Fig.3).In total,13.89,9.03,4.17,and 1.39% of FAW larvae reached the 3rd instar larval stage in the presence of 1,2,4,and 8 µg mL–1Cry1Ab protein,respectively.In contrast,83.33% of FAW larvae reached the 3rd instar larval stage in the negative control (0 µg mL–1Cry1Ab).

3.4.Laboratory bioassay of DBN9936 tissues fed to FAW

Fig.1 Mean expression levels of Cry1Ab in various tissues with a moving average trendline.The different lowercase letters on the SE bars indicate significantly different means (P≤0.05) (n=9 except for the tissues at the R4 stage,for which n=8).

Mortality of FAW larvae feeding on VT leaf,VT tassel,R1 silk,R1 husk,R4 leaf,and R4 kernel of non-Bt maize DBN9858 were respectively 11.11,15.56,18.89,12.22,11.11,and 10.00% after 3 days,and 22.22,32.22,27.78,31.11,23.33,and 22.22% after 5 days.Five days after infestation,the larval mortality of FAW fed on tissues of the control non-Bt maize DBN9858 did not reach 33%.During the entire bioassay period,mortality was higher for FAW reared on Bt maize DBN9936.On the 3rd day after infestation,the corrected mortality of FAW fed on VT leaf was significantly higher than FAW fed on the other tissues.On the 5th day after infestation,a higher corrected mortality of FAW larvae was observed on VT leaf and R4 leaf.A few larvae (1.45%) survived on the R4 leaf,while no survivors of FAW were observed on the VT leaf after 5 days (Fig.4).The descending order of lethality was ranked as follows:VT leaf＞R4 leaf＞R1 husk＞R1 silk＞VT tassel＞R4 kernel.

Fig.2 Mean expression levels of Cry1Ab in different locations in China with a moving average trendline.The different lowercase letters on the SE bars indicate significantly different means(P≤0.05) (n=6 except for Pu’er,for which n=4).

Fig.3 Stacked bar charts of the effect of a simulated artificial diet containing Cry1Ab protein on fall armyworm after 6 days.Cry1Ab concentrations in the artificial diet were 8,4,2,and 1 µg mL–1,which correspond to the mean Cry1Ab levels in the leaf,silk,tassel,and kernel,respectively.

Fig.4 Corrected mortality of larvae of fall armyworm fed on different maize tissues of DBN9936 on the 3rd and 5th day.The different lowercase letters on the SE bars indicate significantly different means (P≤0.05) (n=3).

4.Discussion

The variation of Bt protein level in different regions (spatial)and plant developmental stages (temporal) might enhance the probability of survival of several pest species (Nguyen and Jehle 2007,2009; Székácset al.2010a,b; Trtikovaet al.2015; Romeiset al.2019).To evaluate the spatial and temporal expression of Cry1Ab protein in GM maize DBN9936,V6–V8 leaf,VT tassel,R1 leaf,R1 silk,R4 leaf,and R4 kernel were used in ELISA assays.The contents of Cry1Ab in each tissue from different locations were clearly different (P≤0.05).Similarly,on average 20-fold differences were observed in Cry1Ab levels of MON810 maize,depending on cultivation location (Székácset al.2010b).Among the nine locations,the Cry1Ab levels in DBN9936 of the Xinxiang,Langfang,and Harbin fields were significantly lower than those from Wuhan and Shenyang,and were slightly,but not significantly lower than those from the other four fields.Overall,with the growth of maize,the mean Cry1Ab level has a gradually decreasing trend(P≤0.05).This is consistent with a previous report where the highest Bt protein level in maize was detected in leaf,followed by tassel,silk,and kernel (Sunet al.2018).The kernel of Bt maize seems to produce a low concentration of Cry toxin and the levels of Cry1Ab protein in grain were approximately 0.3–0.7 mg kg–1(Kochet al.2015; Silvaet al.2018).

A similar downward trend of Cry1Ab levels from V6–V8 to R4 phenological stages was also seen in Bt maize MON810 (Székácset al.2010a).Furthermore,a previous observation in GM maize MON88017 showed that the highest Cry3Bb1 protein contents were in young tissues,and the lowest Cry3Bb1 contents were in kernel (Nguyen and Jehle 2009).Similarly,for moCry1F maize line 6275,leaf samples had a higher level of Cry1F expression compared to grain samples (US-EPA 2010).Despite the variation of Bt protein expression found in different tissues,the pest control properties are unlikely to be affected significantly,at least until the first 100–115 days after sowing (Dong and Li 2007).However,these results also suggest that long-term feeding by FAW on GM maize with variable Bt expression will increase the risk of resistance (He and Wang 2020; Liet al.2020b).

According to ISAAA GM Approval Database (Dec.2019),Bt11 and MON810 were the two most commonly used IR maize events,and they both express the insecticidal protein Cry1Ab.They have 50 and 29 maize hybrids,respectively,in total accounting for 38% of commercial GM IR maize.Sunet al.(2021) found that suitable FAW caterpillar development occurs when maize is at the 7th to 12th leaf stages.The mean Cry1Ab level of MON810 in leaf was 9.76 µg g–1FW in the Monsanto document (Monsanto 2002) registered in AGBIOS (Merrickville,ON,Canada),while for Bt11 it was 4.97 µg g–1FW in the Australia New Zealand Food Authority(ANZFA) draft risk analysis report (ANZFA 2000).We found that the mean Cry1Ab level of DBN9936 in leaf among nine locations was 7.45 µg g–1FW,which is between 9.76 µg g–1FW (MON810) and 4.97 µg g–1FW (Bt11).Although MON810 has contributed to FAW management in Brazil,it was a moderate-dose Bt event for this insect (Omotoet al.2016).

To assess the efficacy of Bt maize DBN9936 against FAW,larval mortality was measured using two assay methods in our study:diet-incorporated and maize tissue bioassays.The corrected mortality of FAW on leaf was 83.33% in diet incorporation assays (8 µg mL–1concentration of Cry1Ab) and 100% in the tissue assays (Figs.3 and 4).Similarly,Zhang and Wu (2019) found that the corrected mortality of C0030.3.5 (i.e.,DBN9936) leaf against 1st–4th instar larvae of FAW (sampled from Dehong,Yunnan,China)ranged from 13.33 to 65.41%.Together with bioassay data,the tolerance of Bt maize DBN9936 to FAW was ranked as:VT leaf＞R4 leaf＞R1 husk＞R1 silk＞VT tassel＞R4 kernel,consistent with the results of Cry1Ab protein expression analysis.Studies onSpodopteralittoralisalso showed that the biological activity of Cry1Ab was consistent with ELISA measurements (Obristet al.2006; Svobodováet al.2017b).In addition,it was reported that silk and kernel tissues had a positive effect on survival and development of FAW larvae (Brazil populations) in the reproductive stage of maize (Pannutiet al.2016).Our study showed that R1 silk and R4 kernel of DBN9936 can produce a high enough dose of Cry1Ab protein to kill the FAW (Fig.4).Overall,our bioassay results indicated that the expression level of the Cry1Ab protein in DBN9936 tissues was sufficient to control the FAW invading China.

In an attempt to correlate the Cry1Ab protein concentration with the insecticidal efficacy of DBN9936,we used both ELISA and bioassay methods to associate Bt protein levels with toxicity to FAW.We found that the insecticidal efficacies to FAW larvae were ＞98% on VT leaf and R4 leaf,＞61%on R1 silk,＞37% on VT tassel,and ＞34% on R4 kernel,which is largely consistent with the differences of the mean Cry1Ab protein content in these plant parts (Figs.1 and 4).Our study clearly demonstrated that either of these two methods can be used to assess the insecticidal efficacies of Bt maize to FAW larvae.However,previous studies also suggested that the toxicity of Bt cotton may not be simply related to the amount of Bt protein in the plants,which may also apply to Bt maize (Dong and Li 2007; Girón-Calvaet al.2020).Therefore,further multidisciplinary research is required to generate reliable and robust data to understand the relationship between Bt insecticidal protein levels and effectiveness of Bt crops against insect pests.

FAW has developed resistance to multiple pesticides,from 2008 to 2015,and there are five different Bt insecticidal proteins (Cry1Ab,Cry1A.105,Cry2Ab2,Cry1F and Vip3Aa20) present in GM maize,so both single and pyramided combinations have been launched for FAW control.The Food and Agriculture Organization of the United Nations (FAO) is aware that Bt maize can decrease damage from FAW,but,unfortunately,this pest may rapidly evolve resistance to some Bt insecticidal proteins within 2–3 years,except to Vip3Aa20.Instances of field evolved resistance suggest that certain assumptions of the high-dose/refuge strategy have not been met (Vélezet al.2016).Indeed,after Bt maize was first introduced to control FAW,larvae practically disappeared from maize fields (Fatorettoet al.2017).Now that FAW have arrived in southern China,there is a very high possibility that they will invade eastern China on an annual basis (Li X Jet al.2019).More than 30 provincial areas in China were predicted to be invaded by FAW larvae from the end of August to the beginning of September 2020 (Wang and Lu 2020).Li G Pet al.(2019)found that the FAW strain that invaded China in 2019 has not developed significant resistance to Cry1Ab.Zhanget al.(2020) also showed that although the invading FAWs carry resistance to both organophosphate and pyrethroid pesticides,they continue to be sensitive to Bt-Cry1Ab maize in China.A recent report indicated that the planting of Bt maize across a large area of China would definitely help to control the FAW (Silver 2019).Consistent with the above findings,our study confirmed that Bt maize (e.g.,DBN9936)could have slowed the advance of FAW across China.

A slightly low content of Cry1Ab protein in DBN9936 tissues was found in the late growing season,which may affect the control efficacy of Bt maize for FAW.However,because DBN9936 is highly effective at controlling FAW in the early and middle growing seasons,taking into account that the egg masses of FAW are mainly distributed at the early whorl and silking stages (Huanget al.2020),the effectiveness of Bt maize against FAW may not be an issue as maize matures.FAW has the potential to develop resistance in the late growing season,and as many GM Bt crops providing poor control of target pests late in the growing season,care must be taken when introducing Bt maize hybrids in China (Wang Y Net al.2014; Abrahamset al.2017; Liet al.2020b).Field-evolved pest resistance to Bt maize by FAW has been reported in Puerto Rico,Brazil,Argentina and the USA (Storeret al.2010,2012; Sisayet al.2019; Gonçalveset al.2020).For sustainable use of Bt maize,appropriate strategies have to be developed to delay the evolution of FAW resistance including the highdose/refuge and gene pyramiding strategies.It is feasible to increase the expression level of thecry1Abgene or pyramid insecticidal genes such ascry1Aborcry1Ftogether withvip3A(Wang Y Net al.2014b; Wanget al.2016; Liet al.2020b).Careful consideration should be taken in deploying Bt maize in the tropical and subtropical regions of China where FAW reproduces year-round (He and Wang 2020).

To achieve the goal of low-cost,green,and sustainable management of FAW in China,Wu (2020) has proposed developing a “two-step” strategy including Bt maize in the second step,as well as an integrated pest management(IPM) strategy.Previous reports also suggested that when using Cry1Ab maize,other IPM practices should be adopted to minimize the yield losses caused by FAW (Silvaet al.2018; Liet al.2020b).Given the current distribution of FAW in China and the documented speed of insects evolving resistance,it is important to adopt a proactive insect resistance management (IRM) plan,following commercial planting of Bt maize DBN9936,especially in refuge implementation (Li Y Het al.2018; He and Wang 2020).There are no significant alternative wild or cultivated host plants that serve as natural refuges for FAW,and it is difficult to implement the structural refuge strategy in China with small-scale farms and millions of individual growers.Taking all these factors into consideration,a seed mix refuge(mixing 5–10% non-Bt seed into a bag of Bt maize) may be a more suitable option (Wuet al.2019).A seed mix refuge combined with biological control (e.g.,Trichogramma pretiosumRiley) is another promotion strategy to increase the control efficiency of FAW (Silvaet al.2020).Additional evidence with generalist predators and FAW can also be compatible with Bt maize (Dutraet al.2012; Svobodováet al.2017a),and support an IPM approach (Romeiset al.2019).Moreover,the combination of entomopathogenic nematodes and Bt maize silks could enhance the mortality of FAW prepupae,as proposed by Prasannaet al.(2018).

Because of the commercial-scale planting of GM crops,it is essential to continuously monitor their performance in the ecosystem,particularly in its relationship to target gene expression (Cheemaet al.2016; He and Wang 2020).It is well known that subsequent traditional breeding with the approved GM event may lead to expression levels of inserted genes considerably different from the expression levels of the initially authorized GM event (Székácset al.2010a; Koket al.2014).In this case,a long-term monitoring system should be designed to detect the efficiency of DBN9936 to control the target pests,and to assess the insect resistance efficiency of the breeding lines derived from DBN9936.

5.Conclusion

Bt maize has been widely used for control of insect pests in many countries,but not yet in China.This paper provides the first large-scale expression profile of Cry1Ab protein in the newly approved GM maize DBN9936,at nine locations in China.The content of Cry1Ab in different maize tissues was ranked as V6–V8 leaf＞R1 leaf＞R4 leaf＞R1 silk＞VT tassel＞R4 kernel.Bioassay results indicated that the GM maize DBN9936 has high efficacy against the invading population of FAW in China,and would therefore provide important baseline data as a reference for commercial plantings of Bt maize,as well as future pest management in China.

Acknowledgements

This work was funded by the National Genetically Modified Organism New Variety Breeding Program of China(2019ZX08012-004).We are grateful to Prof.Li Yunhe from Institute of Plant Protection,Chinese Academy of Agricultural Sciences for his helpful comments and suggestions.

Declaration of competing interest

The authors declare that they have no conflict of interest.