Kai Li,Xinxin Zhang,Jianqiu Guo,Hannah Penn,Tingting Wu,Lin Li,Hong Jiang,Lidan Chang,Cunxiang Wu,Tianfu Han,*

a Soybean Research Institute,Nanjing Agricultural University,Nanjing 210095,Jiangsu,China.

b MOA Key Laboratory of Soybean Biology(Beijing),Institute of Crop Sciences,Chinese Academy of Agricultural Sciences,Beijing 100081,China.

c Luoyang Academy of Agriculture and Forestry Sciences,Luoyang 471022,Henan,China

d Department of Entomology,Louisiana State University,LA 70803,USA.

Keywords:Soybean Staygreen syndrom e Zhengqing DAS-ELISA Viral infection Bean bug(R.pedestris)

A B S T R A C T Staygreen syndrom e or Zhengqing in soybean has recently becom e a m ajor issue for Chinese growers in the Huang-Huai-Hai river basin.Although previous studies revealed that staygreen can be induced w hen pods/seeds are dam aged,it is unknow n w hether virus infection or insect infestation causes staygreen.To determ ine w hether viral infection causes staygreen,a survey of soybean staygreen incidence in the Huang-Huai-Hai river basin was conducted in 2016 and 2017.Diseased samples were collected and analyzed using DAS-ELISA for Soybean mosaic virus,Watermelon mosaic virus,Bean pod mottle virus,Cucumber mosaic virus,and Bean common mosaic virus.The survey show ed that the severity of soybean staygreen syndrom e w as m ost prevalent in Beijing,Henan,Shaanxi,and som e parts of Shandong provinces,w ith yield losses from 0 to nearly 100%,but only a sm all fraction of sam ples were positive for the tested viruses.A field cage experim ent and an insecticide treatm ent field trial w ere conducted to determ ine the contribution of the bean bug,Riptortus pedestris,to staygreen incidence.The field cage experim ent show ed that R.pedestris treatment resulted in shorter plants,m ore empty pods,increased numbers of abnormal seeds,and decreased yields.The field experim ent showed that there w ere few er R.pedestris and less soybean staygreen incidence in fields treated w ith insecticide than in untreated control fields.Together,these results suggest that R.pedestris infestation rather than virus infection induces staygreen syndrom e and that grow ers in this region can m itigate staygreen syndrom e via bean bug control.

1.Introduction

Staygreen syndrom e or Zhengqing(the Chinese com m on nam e),an econom ically dam aging event prevalent across m any crop types,is typically characterized by a lack of tim ely leaf senescence and an increase in seed abortion[1,2].In soybean(Glycine m ax L.[Merr.])specifically,staygreen syndrom e is m anifested as a lack of leaf senescence,pod abortion,and an increase in abnormal seeds[3,4].In recent years,the Huang-Huai-Hai river basin in China has experienced an outbreak of soybean-based staygreen syndrom e,causing great concern am ong soybean grow ers ow ing to high yield losses.

Soybean green stem syndrom e(GSS)in the USA and Japan is sim ilar to soybean staygreen syndrom e in China,and GSS may provide clues to the cause of staygreen syndrome.GSS occurs w hen soybean pods and seeds mature but the stem rem ains green,w hereas staygreen syndrom e in soybean as m anifested in the Huang-Huai-Hai river basin consists of a lack of leaf senescence and an increase in seed abortion.Previous studies have indicated that GSS can be caused by viral infection(by,e.g.,Bean pod mottle virus,BPMV;Soybean mosaic virus,SMV;and Tobacco ringspot virus,TRSV),insect infestation,pod removal,and environmental stress during seed fill[4,5].Based on these GSScauses,a previous report[6]postulated that pod/seed dam age from either disease or an insect w as also to blam e for staygreen syndrom e.Given that SMV is prevalent in the area and has been know n to cause pod/seed abortion,SMV could be one explanation for the ubiquity of staygreen syndrom e.How ever,other investigations[5,6]revealed that pod dam age from a syringe w as a mechanism of soybean staygreen syndrome.This finding[7]suggests that the culprit could be an insect pest,particularly one w ith piercing-sucking m outhparts.

Bean bug(Fig.1),Riptortus pedestris Fabricius(Hem iptera:Alydidae)[8],has lately received increased interest owing to its potential for soybean dam age and its potential involvem ent w ith staygreen syndrom e[9-11].Bean bug as observed in China is comm on in the summer grow ing season from August to Septem ber and causes greatest dam age betw een the initial seed(R5)and full seed(R6)grow th stages[12,13].Although it is considered a serious pest in soybean production system s in Japan and the Republic of Korea[14],R.pedestris has not been previously reported as contributing to staygreen syndrome[15-17].Nor have sim ilar piercing-sucking insects such as stinkbugs(Hem iptera:Pentatom idae)been reported to contribute to staygreen syndrome or similar phenomena in soybeans[18-20].

To identify the m ain cause of the soybean staygreen phenom enon,w e conducted a tw o-part study of the contribution of both viral infection and R.pedestris infestation to the prevalence of soybean staygreen syndrom e in the basin.Based on those studies,w e postulated that bean bug rather than viral infection w as the prim ary reason for the recent increase in soybean staygreen syndrome in the basin.We conducted a m ulti-province survey of farm er-m anaged fields to estim ate virus presence in conjunction w ith yield losses associated w ith staygreen syndrom e.We serologically analyzed sam ples taken from surveyed fields for soybean virus diseases such as SMV,Watermelon mosaic virus(WMV),BPMV,Cucumber mosaic virus(CMV),and Bean common mosaic virus(BCMV).We tested the contribution of bean bug to the presence of staygreen sym ptom s such as pod abortion and lack of leaf senescence using field cage experim ents and fieldtrials using bean bug-control m ethods.

2.Materials and m ethods

2.1.Staygreen incidence and associated yield losses using surveys of farmer-managed fields

During the soybean seed filling stage(R5 to R6)[21]during the 2016-2017 grow ing seasons,a survey for staygreen incidence was conducted in Beijing,Hebei,Shanxi,Jiangsu,Anhui,Shaanxi,Henan,and Shandong provinces or m unicipalities in the Huang-Huai-Hai river basin(Table 1).Ten random ly selected fields in 4-5 cities or counties(n=40 or 50 fields/province)were investigated in each province,with the exception of Beijing(3 fields),Shanxi(5 fields),Shaanxi(4 fields)and Anhui(20 fields).In each field,30 plants on five diagonal transects(n=150 plants/field)w ere observed for staygreen syndrome occurrence and their yield losses w ere estim ated.Staygreen incidence for each field w as averaged across the 5 transects and the staygreen incidence in each city/county w as represented by the averaged incidence of 10 fields.

2.2.Testing for viruses(SMV,WMV,BPMV,CMV,and BCMV)in collected samples

To determ ine w hether viral infection w as a cause of staygreen syndrome,a subset of typical staygreen soybean plants from the surveyed fields w ith disease sym ptom s w ere sam pled during 2016-2017 for virus detection using serological tests of leaves and pods.Virus detection w as perform ed using double antibody sandw ich enzym e-linked immunosorbent assays(DAS-ELISA)according to the m anufacturer's(ACDiagnostics,Inc.,Elkhart,Indiana,USA)instructions for five viruses:SMV,WMV,BPMV,CMV,and BCMV.The raw DAS-ELISA readings(optical density values)from each sam ple w ere averaged and converted to m ultiples of the negative controls(average reading of the negative sam ples).Sam ples w ith a ratio＞2.0 w ere considered positive for the associated virus.Extracts from non-inoculated,healthy leaves of the SMV-susceptible soybean cv.Nannong 1138-2 w ere used as negative controls and samples of SMV,WMV,BPMV,CMV,and BCMV w ere used as positive controls.No statistical analyses w ere perform ed,as not enough sam ples tested w ere positive.

2.3.Riptortus pedestris infestation field cage experiments

The effects of R.pedestris on staygreen incidence w ere m easured at Beijing(GPS coordinates 39°57′N,116°19′E),and Luoyang(34°38′N,112°29′E)experimental field sites.The soybean cultivars Zhonghuang 30(Beijing,China)and Yudou 22(Luoyang,Henan,China)w ere sow n at a density of 60 plants per field cage w ith a row w idth of 20 cm on June 26,2016 in Beijing and June 22,2016 in Luoyang.At the tim e of pod initiation(R3)[21],field cages m ade of steel pipes and covered w ith nylon m esh(w ith pore size 0.85 m m)fine enough to exclude most arthropods,but permitting light and rainfall penetration,w ere erected around the plants.The cages(1.0 m×3.0 m×1.5 m)w ere buried to 0.5 m for stability.In the control treatm ent,plants w ere treated tw ice per w eek w ith 2.5 m L L-1of acetam iprid(an alpha-chloro-Nheteroarom atic com pound)(Lyuyiyuan Technology Developm ent Co.Ltd.,Beijing,China)in w ater.In the R.pedestris inoculation treatm ent,plants w ere sprayed w ith an equal am ount of w ater follow ed by the addition of 150 w ild-caught adult bean bugs over 3 days(50 per day).The plants w ere allow ed to interact with the respective treatment for 87 days(119 days after sow ing),until harvest on October 15,2016(w hen control plants had fully m atured,R8)[21].The status of the plants w as evaluated by observing the leaf color,counting all pods,em pty pods,all seeds,intact seeds,and m alform ed seeds,and m easuring plant height.Control and inoculation treatm ents w ere com pared using Duncan's new m ultiple range test in SPSS 17.0(SPSS,Chicago,IL,USA).Differences w ere considered significant at P＜0.05.

2.4.Insecticide field trials of R.pedestris in soybean fields

The field experim ent w as com pleted in 2017 on the Chunguang fam ily farm in the tow n of Xinqiao in Shangqiu city,Henan province.Three replications of the soybean cultivar Hedou 19 w ere established for both the R.pedestris prevention and control treatm ents in a com pletelyrandomized design.The total area of the soybean field w as approximately 1.2 ha and that of each plot 1500 m2.The prevention treatm ent consisted of an insecticidal spray(fipronil and deltam ethrin)applied per label recom mendations once a w eek for three w eeks at the start of flow ering.Staygreen incidence w as recorded before soybean harvest at each future insect collection location.Three days after each insecticide application,the total num ber of R.pedestris w as recorded for all treatments.Insects were collected w ith insect sweep nets at five points(in an“X”configuration);at each point,and 20 plants w ere sampled for a total of 100 plants per treatment replicate.Insecticide and untreated control samples w ere com pared by ttest w ith SAS9.4(SASInstitute Inc.,Carry,North Carolina,USA).Differences w ere considered significant at P＜0.05.

Table 1-Incidence of staygreen syndrom e and estim ates of dam age(average yield loss)in soybeans in the Huang-Huai-Hai river basin during 2016 and 2017.

3.Results

3.1.Staygreen incidence and estimated yield loss in farmermanaged fields of the Huang-Huai-Hai river basin

Staygreen incidence and estimated yield loss in the Huang-Huai-Hai river basin varied by province(Table 1).Staygreen syndrome w as not found in the sam pled sites of Hebei province in either 2016 or 2017.How ever,staygreen w as present to some degree in all other sampled provinces,w ith the m ost severely affected areas in Henan and Shandong provinces,w here som e fields experienced up to 90%incidence and 80%yield loss in 2016(Fig.2).Within the m ost highly affected provinces,staygreen incidence w as not uniform across the entire area.For instance,in Shandong,staygreen syndrome w as observed m ainly in the northw est.Riptortus pedestris w as abundant in fields w here soybean staygreen syndrome occurred.On the experimental farm of Luoyang Academ y of Agricultural Sciences,the R.pedestris population w as as high as 4-5 individuals/m2in m id-to late August w hen soybean plants were at podding stage(R4).Thereafter,R.pedestris abundance continued to increase and w as still high in Beijing in early October.Riptortus pedestris was observed feeding on pods of a July-planted soybean plant on October 3,2016(Supplem entary file S1).

3.2.Serological assay of diseased samples for SMV,WMV,BPMV,CMV,and BCMV

The serological results show ed that no sam ple w as positive(0/114)for WMV,suggesting that soybean staygreen syndrom e was not associated with WMV infection.How ever,a few(15/1062)w eak positives(1.4%overall)w ere obtained for SMV including sam ples from seven leaf and eight pod sam ples,from Henan,Shanxi,and Shandong provinces.The serological test for BPMV resulted in four positive samples(of 114)from one leaf and three pod sam ples,all from Shandong province,bringing the positive BPMV rate to 3.5%.The serological test for CMV resulted in 10 positive sam ples(of 798)from four leaf and six pod samples,resulting in an overall positive rate of 1.25%for CMV.Tests for BCMV resulted in six positive sam ples(of 798)from tw o leaf and four pod sam ples(0.8%)(Table 2).No direct corelation w as found betw een soybean staygreen syndrom e and virus infection w hen sam ples w ere pooled for all viruses,as a m ajority(＞98%,2717/2752)of staygreen-infected plant sam ples did not contain any virus.

3.3.Response of soybean plants to R.pedestris in field cages

Plants grew w ell and flow ered norm ally in all treatm ent groups.The control treatment produced taller plants than the R.pedestris-inoculated treatm ent in both experim ental locations(Table 3).How ever,there w as a location difference in the relationship betw een treatm ent and soybean node num ber.Whereas the control group produced normal pods(in both num ber and quality),the inoculation treatm ent increased the num ber of aborted pods and abnorm al seeds and reduced the num ber of intact seeds at both locations.Control plants reached full maturity normally(becam e senescent and dried enough for harvest)by October 15,2016.In contrast,R.pedestrisinoculated plants carried green pods and leaves until the end of the experim ent(Fig.3a,b,c),seed abortion w ith empty cavities in the pods(Fig.3d,e,f),and non-aborted seeds that were of low com m ercial value ow ing to poor quality.Overall,R.pedestris infestation dam aged both soybean pods and seeds and resulted in staygreen syndrom e in the plants.

Fig.2-Severe staygreen-affected soybean p opulation in Jining city,Shand ong province.a)View of the field.b)A typical soybean staygreen p lant.

Table 2-Virus detection for SMV,WMV,BPMV,CMV,and BCMV using DAS-ELISA from farm er-m anaged field sam p les.

3.4.Soybean staygreen incidence and prevention of R.pedestris infestation in field plots

Insecticide-treated(for R.pedestris infestation prevention)fields contained significantly few er R.pedestris than untreated fields(Table 4).The corresponding soybean staygreen incidences w ere also low er in the treated fields than in the untreated control fields.

4.Discussion

4.1.Soybean staygreen syndrome in Huang-Huai-Hai river basin

Soybean staygreen syndrom e has recently becom e extrem ely prevalent in the Huang-Huai-Hai river basin,China,resulting in high losses to farm ers due to this biotic stress.In the present study,w e conducted a survey of the incidence of this phenom enon in farm er-managed fields throughout the Huang-Huai-Hai river basin,perform ed virus assays using sam ples from those fields,and conducted an insect inoculation field cage experim ent and an insect prevention field trial to investigate the factors causing staygreen syndrom e in soybeans.The survey in 2016 and 2017 show ed that there w ere greater levels of staygreen syndrome in Beiing,Henan,Shaanxi,and som e parts of Shandong province.Also,R.pedestris was abundant in surveyed fields affected by severe staygreen syndrom e.Interestingly,staygreen incidence in the Hebei province w as high in 2015(personal com m unication)but low in 2016 and 2017,corresponding to farm ers'use of chem ical controls for R.pedestris.These results obtained from field observations suggest that the occurrence of soybean staygreen syndrom e is associated w ith infestation by R.pedestris.

4.2.Relationship between virus infection and soybean staygreen syndrome in the Huang-Huai-Hai river basin

Serological assays of sam ples w ith staygreen for virus incidence(SMV,WMV,BPMV,CMV,and BCMV)show ed thatno sam ple w as positive for WMV and that only a sm all fraction(＜2%)of the sam ples w ere positive for SMV,BPMV,CMV,and BCMV.Nor did the few virus-positive sam ples show geographical overlap w ith staygreen incidence.In our previous studies[22-24],thousands of soybean plants w ere inoculated w ith m ultiple SMV strains but no staygreen syndrome w as observed post-inoculation.We accordingly conclude that staygreen syndrom e in soybean is not associated w ith SMV,WMV,BPMV,CMV,or BCMV.

Table 3-Soybean developm ent com pared betw een control and Riptortus pedestris-inoculated treatm ents in 2016.Letters d enote significant(P＜0.05)differences by Duncan's test betw een treatm ents in a location.

Fig.3-Ap pearance of soybean plants in a field cage ex perim ent that included Riptortus pedestris exp osure.a)Pre-harvest view of the cage w ithout R.pedestris inoculation(control).b)Pre-harvest view of the cage w ith R.pedestris inoculation.c)Whole-p lant app earances of control(left)and inoculated(right)treatm ents at harvest.d-f)Com parison of intact(d)and shelled pod s(e)as w ell as seed d evelopment(f)under control(top/left)and inoculation(bottom/right)treatments.

4.3.Relationship between insect infestation and soybean staygreen syndrome in the Huang-Huai-Hai river basin

In the R.pedestris field cage experim ents,insect dam age consistently induced staygreen syndrome in soybean plants regardless of variety or location.The resulting staygreen dam age reduced plant height,induced abnorm al seed production,and increased the num ber of empty pods.Given the results of previous studies[6,7],the m echanism could be the puncturing of the pod by the bug's m outhparts,resulting in seed death and preventing horm onal signals from the seed to the leaf[25],ultim ately delaying soybean m aturation and harvest[6].How ever,this insect is also know n for its bacterial endosymbiont[26,27],w hich could potentially exacerbate the syndrome upon m outhpart dam age and introduction of the insect's saliva[28].Alternatively,the insect could transm ite a plant disease w hich w as not tested in the present study;in that case,staygreen occurrence in soybeans could further increase.

4.4.Future R.pedestris management for soybean staygreen syndrome prevention in the Huang-Huai-Hai river basin

Given that R.pedestris appears to be the prim ary reason for the prevalence of staygreen syndrom e in China,control of this insect pest in future will be necessary.Pest management tactics for control of R.pedestris are already w ell studied and im plem ented in agricultural system s in Japan and the Republic of Korea[14,15,29,30]but would need to be adapted to the Chinese system.For instance,know n chem ical control and parasitoids that provide biological control w ould need to be studied for safe and effective introduction to the targetregion[31-35].In addition to R.pedestris management strategies,soybean-breeding program s have been show n to develop plant resistance to staygreen syndrom e[36,37]and could be studied in conjunction w ith other insect m anagem ent techniques for effective control of staygreen syndrom e.

Table 4-Abundance of Riptortus pedestris and staygreen incidence in untreated control fields and insecticide-treated(for p revention of Riptortus pedestris infestation)fields in 2017.Letter differences denote significant(P＜0.05)differences by ttest betw een treatm ents in a location.

5.Conclusions

Using serological assay of staygreen soybean plant sam ples from farmer-managed fields using the DAS-ELISA method,w e found no direct association betw een soybean staygreen syndrom e in the Huang-Huai-Hai river basin and virus infection by any of five tested viruses.From field cage infestation and in-field prevention experiments,w e conclude that the piercing-sucking insect R.pedestris is the prim ary cause of staygreen syndrom e in the Huang-Huai-Hai river basin.We recommend that grow ers in this region mitigate staygreen syndrom e via insect-control m ethods.

Supplem entary data to this article can be found online at https://doi.org/10.1016/j.cj.2018.07.008.

Acknow ledgm ents

This project w as supported by the National Key Research and Developm ent Program of China(2017YFD0101400)to T.Han and China Agriculture Research System(CARS-04)to T.Han and K.Li.We thank colleagues of the CARS soybean experim ental stations in the Huang-Huai-Hai river basin for their help in the field surveys.