Xingyng Xu,Genqio Li ,Guihu Bi ,Brett F.Crver ,Ruolin Bin ,Amy Bernro ,J.Sott Armstrong

a USDA-ARS Wheat,Peanut,and Other Field Crop Research Unit,Stillwater,OK 74075,USA

b USDA-ARS Hard Winter Wheat Genetics Research Unit,Manhattan,KS 66506,USA

c Department of Agronomy,Kansas State University,Manhattan,KS 66506,USA

d Department of Plant and Soil Sciences,Oklahoma State University,Stillwater,OK 74078,USA

Keywords: Wheat Greenbug resistance gene Gb1 KASP markers Linkage analysis Genotyping-by-sequencing

ABSTRACT Greenbug(Schizaphis graminum,Rondani)is a serious insect pest in many wheat growing regions and has been infesting cereal crops in the USA for over a century.Continuous occurrence of new greenbug biotypes makes it essential to explore all greenbug resistant sources available to manage this pest. Gb1, a recessive greenbug resistance gene in DS28A,confers resistance to several economically important greenbug biotypes and is the only gene found to be resistant to greenbug biotype F.A set of 174 F2:3 lines from the cross DS28A×Custer was evaluated for resistance to greenbug biotype F in 2020 and 2022.Selective genotyping of the corresponding F2 population using single nucleotide polymorphism(SNP)markers generated by genotyping-by-sequencing (GBS) led to the identification of a candidate genomic region for Gb1.Thus,SSR markers previously mapped in this region were used to genotype the entire F2 population,and kompetitive allele specific PCR (KASP) markers were also developed from SNPs in the target region.Gb1 was placed in the terminal region of the short arm of chromosome 1A,and its location was confirmed in a second population derived from the cross DS28A×PI 697274.The combined data analysis from the two mapping populations delimited Gb1 to a ＜1 Mb interval between 13,328,200 and 14,241,426 bp on 1AS.

1.Introduction

Greenbug [Schizaphis graminum(Rondani)] has been reported worldwide with wide distribution in southern Europe,the Middle East,central western and central Asia,Africa,and North and South America [1,2].Wheat infestations have occurred for more than a century in the USA [3].When heavy infestations occur in fall to early spring,the growth of wheat plants can be seriously inhibited or even killed,which can lead to significant yield losses [4].The yield losses,which are largely dependent on infested aphid number and infestation period,mainly result from reduced spike number m-2and seed number per spike with a mean of 0.51 kg ha-1per greenbug-day in years having normal precipitation and 1.17 kg ha-1per greenbug-day under severe drought condition[5].Moreover,greenbug is also a common vector of barley yellow dwarf virus [6],a serious viral disease to wheat production.

Enormous efforts directed to searching for greenbug resistance led to the discovery of several greenbug resistance genes in wheat.Among them,Gb1is the first wheat greenbug resistance gene initially identified from the hexaploid wheat accession Dickinson Selection 28A (DS28A) and the landrace PI 70715 (formerly CI 9058) in the 1950s [7].Other wheat greenbug resistance genes includeGb2andGb6from rye [8,9],Gb5fromT.speltoides[10],andGb3,Gb4,Gb7,andGb8fromAe.tauschii[11-14].In addition,a few temporarily named genes such asGba,Gbb,Gbc,Gbd,Gbx1,andGbzwere identified fromAe.tauschii[15,16],and another one,Gby,was discovered from Sando’s selection 4040,a line derived from multiple crosses including one between Chinese Spring(T.aestivum)andLophopyrum ponticum(Podp.)A.Love[17].

Many greenbug biotypes have been identified,and each of them shows avirulence to only a few resistance genes.For example,greenbug biotype B is avirulent toGb2andGb6,whereas biotypeGis avirulent to onlyGb6[18].Given that greenbug biotypes often vary across field environments,pyramiding multiple greenbug resistance genes is imperative for breeding durable greenbug resistant cultivars.Advances in wheat genomics,especially release of the Chinese Spring reference genome sequence[19],make it feasible to develop molecular markers closely linked to these genes and tag them in wheat breeding.Currently,all greenbug resistance genes exceptGb1have been mapped to specific genomic regions[10,11,12,14,20,21],and high-throughput molecular markers are available for some of these genes [20,22].

Genetic diversity is essential to sustaining human life,on which global food security largely depends.Currently,Gb3,which confers resistance to greenbug biotypes C,E,H,I,and K,but susceptibility to biotypes B,F,G,NY,FL1,SC and KS1 [18],has been widely deployed in experimental and commercial germplasm in the USA.More greenbug resistance genes,especially those conferring resistance to greenbug biotypes virulent toGb3,are urgently needed to broaden genetic diversity and provide durable greenbug resistance.

Gb1confers resistance to several critical biotypes and is the only gene providing resistance to greenbug biotype F.However,Gb1is a recessive gene and thus not amenable to traditional phenotypic selection in early generations.Therefore,marker-assisted selection is needed to improve selection efficiency withGb1in wheat breeding populations.The objectives of this study were to determine the chromosome location ofGb1and develop genomic tools for selection of the gene in cultivar development.

2.Materials and methods

2.1.Plant materials

An F2population and 174 F2:3lines derived from DS28A × Custer,and an F2population and 219 F2:3lines from DS28A × PI 697274 (formerly PI 595379-1),were used to mapGb1.DS28A and PI 697274 carryGb1andGb8,respectively,while Custer,released by Oklahoma State University in 1994,is highly susceptible to greenbug.DS28A is a spring wheat with some degree of cold hardiness [7].PI 697274 is susceptible to greenbug biotype F.In addition,86 experimental wheat lines from the 2019-2020 USDA-ARS Hard Winter Wheat Regional Performance Nursery (https://www.ars.usda.gov/plains-area/lincoln-ne/wheatsorghum-and-forage-research/docs/hard-winter-wheat-regionalnursery-program/research/)were used to evaluate the usefulness of KASP markers developed in this study.

2.2.Evaluation of F2:3 lines to greenbug infestation

The DS28A×Custer F2:3population was evaluated for response to greenbug biotype F in 2020 and 2022 at the USDA-ARS Wheat,Peanut,and Other Field Crop Research Unit.The DS28A × PI 697274 F2:3population was evaluated for response to greenbug biotype F in 2022 using the same protocol to confirm mapping results from the DS28A × Custer cross.

Greenbug biotype F,which is virulent to all known greenbug resistance genes exceptGb1,was cultured on barley cultivar Eight-Twelve (PI 537437).About 60 seeds were planted into 15-cm pots filled with sand.The culture plants,which were enclosed in a cylindrical plastic cage fitted with muslin cloth ventilation on the top and sides,were infested with about 50 greenbug aphids at the two-leaf stage.These plants were used to infest the test plants following four weeks of aphid development.

The greenbug phenotyping assay was conducted in a greenhouse supplied with daylight of 16 h at 22 ± 2 °C using a randomized complete block design with two replicates.In each replicate,15 seeds per line were planted in three cells of a 73-cell growing tray filled with Sunshine Red-earth growing mix (Growing Systems,Inc.).The resistant(DS28A)and susceptible(Custer)controls were planted in one and two cells,respectively,in each tray.The culture plants with abundant greenbugs were placed close to each row of the test plantlets one day after their emergence.Each plant was initially infested with about 10 to 15 aphids and scored using a binary scale after two weeks infestation when the susceptible control died,in which asymptomatic healthy plants were rated as resistant and plants that turned yellow or completely died were rated as susceptible.Genotypes of F2plants were inferred from phenotypes of their corresponding F2:3progenies.

2.3.Selective genotyping of the F2 population derived from DS28A × Custer

Plant leaves were taken from each F2plant at the two-leaf stage and freeze-dried in a lyophilizer (Sp Scientific) for two days.The dried leaves were ground at 1500 r min-1for 1 min using a MiniG Automated Tissue Homogenizer (SPEXSamplePrep).Genomic DNA was extracted using a method described by Dubcovsky et al.[23]and normalized to 10 ng μL-1.Together with two parents,a subset of 36 F2DNA samples were randomly chosen to develop GBSderived SNP markers following Xu et al.[20],leading to the identification of a large set of genome-wide SNPs.The Excel ‘‘count”function was used to reveal the distribution of DS28A and Custer alleles at each SNP locus in the F2sub-population.A candidate genomic region that may harborGb1was then identified.

2.4.Genotyping the DS28A × Custer F2 population using simple sequence repeat (SSR) markers

Selective genotyping of F2plants revealed a genomic region thatGb1may reside.Therefore,SSR markers previously mapped in this region were selected to genotype the entire F2population using a previously described protocol based on a LI-COR 4300 DNA analyzer[14],and theGb1gene was subsequently mapped to a specific interval.

2.5.Development of kompetitive allele specific PCR (KASP) markers

GBS-SNPs in the target region were selected to develop KASP markers in the DS28A × Custer F2population.Primers were designed using the PolyMarker program(http://www.polymarker.info),which blasted the target sequences against the wheat reference sequence IWGSC RefSeq v2.1 [19] to identify unique GBSSNPs in the target region for marker development.The newly developed KASP markers,designated with prefix ‘‘stars-KASP”(Stillwater,ARS) and numbered consecutively,were tested using two parents and a subset of 36 F2DNA samples.KASP markers that distinguished the two parental alleles and heterozygous genotypes were selected to genotype the entire F2population using a protocol based on the ABI ViiA7 Real-time PCR system (Thermo Fisher Scientific,Waltham,MA,USA).In brief,a 5 μL PCR mixture was prepared using the OT2-Pipitting Robot (Opentrons,New York,NY,USA) with each reaction containing 10 ng genomic DNA,2.5 μL KASP-TF V4.0 2X Master Mix with Low Rox (Biosearch Technologies),6 mmol L-1of each allele-specific primer,and 15 mmol L-1of the common primer.The PCR started at 30 °C for 1 min and 94 °C for 15 min,followed by 40 cycles of 94 °C for 20 s and 60°C for 1 min,with a final step of 30°C for 1 min.KASP data were scored using the ABI ViiA 7 software.

2.6.Linkage analysis

MAPMAKER 3.0 software[24]was employed to conduct linkage analysis using the Kosambi function[25]to convert recombination frequency to genetic distance in centimorgan (cM).A logarithm of the odds (LOD) score of 3.0 was used as the threshold for linkage detection.MapChart 2.2 software [26] was used to draw the linkage map.In addition,the goodness-of-fit test was performed to determine whether the observed phenotypic data fit the expected segregation ratio in F2.

3.Results

3.1.Responses of DS28A × Custer F2:3 lines to greenbug biotype F

The two parents and their 174 F2:3progenies derived from DS28A×Custer were evaluated for responses to greenbug biotype F.DS28A showed high resistance without obvious plant-tissue damage,while Custer died after two weeks’ infestation.Thirty and 51 F2:3lines exhibited homogeneous resistance and homogeneous susceptibility,respectively,and the remaining 93 lines segregated for resistance and susceptibility.The χ2test indicated single-gene segregation in the population with the expected 1:2:1 segregation ratio atP＞0.05 (χ2=5.64;df=2).

3.2.Selective genotyping of the DS28A × Custer F2 population

Thirty-six F2plants were randomly selected from the DS28A × Custer population and genotyped using the GBS approach.The F2:3phenotypic data implied 7 homozygous resistant,5 homozygous susceptible,and 24 heterozygousGb1genotypes.A total of 116,176 SNPs were identified,with 35,042-75,160 SNPs per plant.A subset of 58,088 SNPs with minor allele frequency greater than 0.2 was analyzed.Of these,one SNP,designatedS1A_2619907219,was likely associated withGb1,as evidenced by the fact thatS1A_2619907219co-segregated withGb1in 34 of the 36 F2plants.All resistant and susceptible F2plants carried DS28A and Custer alleles,respectively,and all but two heterozygous F2plants carried both alleles at this SNP locus.This indicated thatGb1should reside close toS1A_26199072,located at 26,199,072 bp in the short arm of chromosome 1A.

3.3.Mapping of the greenbug resistance gene Gb1

A set of SSR markers previously mapped on chromosome 1A were evaluated for polymorphism between the two parents,and 13 polymorphic SSRs,of which nine were recently developed in our lab(Table S1),were subsequently used to genotype the F2population.Linkage analysis indicated thatGb1was 4.9 cM proximal toXwmc818and 24.2 cM distal toXstars948(Fig.1).Further searching the primer sequences ofXwmc818against the IWGSC RefSeq v2.1 revealed its reverse primer sequence at 13,328,177-13,328,200 b p.Therefore,Gb1resides in a 21-Mb interval between 13,328,200(Xwmc818) and 34,345,763 bp (Xstars948) in the IWGSC RefSeq v2.1.

There were 156 GBS-SNPs in the target genomic region(3,328,200-34,345,763 bp).These SNPs were examined,and primers were designed to convert a subset of them to KASP markers.However,only five of them,namelyXstars-KASP184,Xstars-KASP190,Xstars-KASP194,Xstars-KASP199,andXstars-KASP205(Table S2) generated robust data,with the other primers failing to amplify PCR products from DS28A.This indicated significant sequence variability between DS28A and Chinese Spring in this region.These five markers were used to genotype the F2population.Linkage mapping based on KASP and SSR markers placedGb1in a 4 Mb-interval between 13,328,177 bp (Xwmc818) and 17,413,030 bp (Xstars-KASP184) (Fig.1).

3.4.Confirming the chromosome location of Gb1 in the DS28A × PI 697274 population

Of the markers flankingGb1,Xwmc818is a dominant marker that cannot amplify PCR products from DS28A,and the genetic distance betweenGb1andXstars-KASP184is over 10 cM.Therefore,we further phenotyped 219 F2:3lines from DS28A × PI 697274,with the aim to confirm the genomic location ofGb1and develop additional markers worthy of selection.Genotyping the two parents withGb1-flanking markers found that onlyXstars-KASP190was polymorphic between DS28A and PI 697274.Deep sequencing of the two parents identified 43,233 GBS-SNPs,with 14 in the interval harboringGb1(between 13,328,177 bp and 17,413,030 bp).One SNP at 14,241,426 bp was successfully converted to a KASP marker (Xstars-KASP170),while the others were not converted due to either sequence variation among genotypes or lack of fitness of sequences flanking the SNPs to design primers.Linkage analysis indicated thatGb1was 5.2 and 6.7 cM distal toXstars-KASP170andXstars-KASP190,respectively (Fig.1).Combining data from the two mapping populations,we conclude thatGb1resides in an interval of 913,326 bp on 1AS between 13,328,200(Xwmc818)and 14,241,426 bp(Xstars-KASP170)(Fig.1).

3.5.Distribution of DS28A alleles at four KASP marker loci in U.S.Wheat breeding lines

To determine the usefulness of SNP markers flankingGb1for breeding applications,86 experimental wheat lines from the 2019-2020 USDA-ARS Hard Winter Wheat Regional Performance Nursery were genotyped usingXstars-KASP170,Xstars-KASP184,Xstars-KASP190,andXstars-KASP194.Our results indicated that 20.0%,14.1%,8.5%,and 27.2%of these lines carried the DS28A alleles at theXstars-KASP170,Xstars-KASP184,Xstars-KASP190,andXstars-KASP194loci,respectively.Given that most contemporary breeding lines carry the alternative alleles at these loci,polymorphism is expected at one or more KASP loci in most breeding populations derived from DS28A and U.S.wheat accessions.Therefore,these markers should be useful for trackingGb1in marker-assisted breeding.

4.Discussion

Gb1is the first greenbug resistance gene identified in wheat[7].Unfortunately,Gb1has never been used in cultivar development because a new greenbug strain overcame the gene soon after its discovery[27].The new greenbug strain virulent toGb1was found in a greenhouse in Stillwater,OK and designated as greenbug biotype B,presuming that naturally occurring greenbugs in the field were avirulent toGb1,thus constituting biotype A.Likely,the loss of effectiveness to greenbug biotype B and other biotypes identified in earlier years led to the perception thatGb1was overcome by new greenbug biotypes and therefore unworthy for wheat breeding.However,further study found that the occurrence of new greenbug biotypes is a result of natural genetic variability maintained on non-cultivated grass hosts rather than a result of selective pressure from exposure to deployed resistant cultivars or hybrids [28].Recent studies indicated thatGb1confers resistance to many economically important greenbug biotypes such as A,F,J,NY,KS1,KS2,TX1,TX2,TX3,TX4,TX5,TX6,TX7,TX10,WY81 and WY12 MC[18,29],and is the only gene conferring resistance to biotype F.Therefore,Gb1remains a valuable greenbug resistance gene for breeding wheat cultivars with a wide spectrum of resistance to greenbug.More importantly,Gb1is the only greenbug resistance gene that originated from bread wheat.Compared with those from wheat relatives,Gb1can be more easily used incultivar development,although pre-breeding is still needed to transferGb1to locally adapted breeding lines or cultivars because the yield and quality traits of DS28A and PI 70715 may not meet the requirements of modern wheat cultivars.

SinceGb1is recessive,a marker-based scheme is more efficient to selectGb1than classical phenotypic selection in cultivar development.In this study,we mappedGb1to an interval of less than 1 Mb in the short arm of chromosome 1A and developed a few KASP markers linked toGb1,includingXstars-KASP170,Xstars-KASP184,Xstars-KASP190,andXstars-KASP194.The genetic distances betweenGb1and these markers ranged from 5.2 to 14.1 cM (Fig.1).Thus,reasonable selection accuracy should be achieved when these markers,especiallyXstars-KASP170,are used to tagGb1.However,our efforts to develop KASP markers co-segregating withGb1failed because of significant sequence variation between DS28A and Chinese Spring in the target region,evidenced by the fact that many primers based on Chinese Spring could not amplify a PCR product from DS28A.Further sequencing the DS28A genome is ideal for developing markers more closely linked toGb1.

Given that each greenbug resistance gene confers resistance to only a few greenbug biotypes and new biotypes continue to emerge,marker-assisted gene pyramiding is preferred to breed durable greenbug resistant cultivars.Of the known greenbug resistance genes,Gb7was identified from a synthetic hexaploid wheat line W 7984 [12],and KASP markers closely linked toGb7,synopGBS773-KASP and synopGBS1141-KASP,were developed for breeding applications [22].Gb5was originally identified in the wheat-T.speltoidestranslocation segment 7S#1 that is about 702.28-717.54 Mb in length and accounts for 96.4% of chromosome 7A[10,20].Deleterious linkage drag associated with this long alien chromosome segment madeGb5unworthy in wheat breeding.However,the development of 7S#1L,a shortened wheat-T.speltoidestranslocation that reduced theT.speltoidessegment to 79.5-87.8 Mb,makesGb5a usable source of resistance in cultivar development [20,30].Three KASP markers specific to the wheat-T.speltoides7A-7S translocation segment 7S#1L (KASP-Gb5-1,KASPGb5-2,andKASP-Gb5-3) have been developed and can be used to selectGb5[20].

Gb2andGb6were transferred from rye cv ‘‘Insave” to wheat lines‘‘Amigo”and‘‘CRS1201”,respectively,by the 1AL.1RS translocation,and both genes reside in the satellite region of the 1RS chromosome arm,withGb6distal toGb2[21].More recently,a KASP marker,KASP-1RS-1,was developed[31]to target the 1RS chromosome arm whereGb2andGb6reside.This marker has the potential to effectively selectGb2andGb6.In addition,Gb3is allelic toGb4and they were mapped to the long arm of chromosome 7D.AFLP,STS,and CAPS markers closely linked toGb3were developed[11,32].Similarly,SSR markers associated withGb8are available[14].Currently,a project aimed at developing durable greenbug resistant cultivars by pyramiding these greenbug resistance genes has been initiated.The KASP markers developed in this and previous studies lay a solid foundation for this project,and further development of KASP markers forGb3andGb8is essential for the success of this project.

CRediT authorship contribution statement

Xiangyang Xu:Conceptualization,Supervision,Formal analysis,Writing-original draft.Genqiao Li:Investigation,Formal analysis,Writing-review&editing.Guihua Bai:Conceptualization,Supervision,Writing -review &editing.Brett F.Carver:Resources,Writing -review &editing.Ruolin Bian:Data curation,Software,Writing-review &editing.Amy Bernardo:Investigation,Writing-review &editing.J.Scott Armstrong:Resources,Writing -review &editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This study was supported by the USDA ARS CRIS Project(3072-21000-009-00D).Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the USDA.The USDA is an equal opportunity provider and employer.

Appendix A.Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2023.02.002.