Lirong Zhang,Yunfeng Xu*,Ming-Shun Chen,Zhenqi Sue,Yang Liuf,Yuzhou XuGuixiao Lag,Guihua Bai,*

a Department of Plant Pathology,College of Plant Protection,Hebei Agricultural University,Baoding 071001,Hebei,China

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

c Technological Innovation Center for Biological Control of Crop Diseases and Insect Pests of Hebei Province,Baoding 071001,Hebei,China

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

e State Key Laboratory for Agrobiotechnology and Key Laboratory of Crop Heterosis and Utilization (MOE) and Beijing Key Laboratory of Crop Genetic Improvement,China Agricultural University,Beijing 100193,China

f Institute of South Subtropical Crops,Chinese Academy of Tropical Agricultural Sciences,Zhanjiang 524013,Guangdong,China

g Industrial Crop Research Institute,Henan Academy of Agricultural Sciences,Zhengzhou 450002,Henan,China

Keywords:Wheat Hessian fly resistance Mayetiola destructor KASP GBS

ABSTRACT The Hessian fly (HF,Mayetiola destructor) is one of the destructive pests of wheat (Triticum aestivum L.)worldwide.Resistant cultivars can effectively minimize wheat damage due to this insect pest.To identify new quantitative trait loci (QTL) for HF resistance,a population of recombinant inbred lines (RILs) was developed from a cross between the HF-resistant wheat cultivar ‘Chokwang’ and susceptible wheat‘Ning 7840’,and phenotyped for responses to HF attack.A linkage map was constructed using 1147 single nucleotide polymorphism (SNP) markers generated from genotyping-by-sequencing (GBS).One major QTL,QHf.hwwg-6BS,for HF-resistance was identified on chromosome arm 6BS,which explained up to 84.0% of the phenotypic variation and was contributed by Chokwang.Two RILs showed recombination in the candidate interval of QHf.hwwg-6BS,which delimited QHf.hwwg-6BS to a 4.75 Mb physical interval between 6,028,601 bp and 10,779,424 bp on chromosome arm 6BS of IWGSC Chinese Spring reference genome RefSeq v2.0.Seven GBS-SNPs in the candidate interval were converted into Kompetitive allele specific polymerase chain reaction (KASP) markers.Two of them,KASP-6B112698 and KASP-6B7901215,were validated in a U.S.winter wheat panel.KASP-6B112698 was nearly diagnostic,thus can be used to screen QHf.hwwg-6BS and pyramid it with other resistance genes in breeding programs.

1.Introduction

Hessianfly(HF,Mayetioladestructor)isoneofthemostdestructive pests and can cause significant economic losses in most wheatgrowing regions in the U.S.and worldwide [1–3].Deployment of HF resistance genes have been proven to be the most effective and economical strategy for controlling the pest[4].

To date,37 HF resistance genes(H1 to H36 and Hdic)have been reported from wheat and its relatives[5,6].Among these HF resistance genes,11 genes(H1-H5,H7,H8,H12,H34,H35,H36)are from common wheat[2,6–9];six(H13,H22,H23,H24,H26 and H32)are from Aegilops tauschii;15(H6,H9-H11,H14-H20,H28,H29,H31 and H33) are from durum wheat;and two (H21 and H25) are from Secale cerale L.(RR,2n=2x=14).In addition,H27,H30 and Hdic are from Ae.ventricosa,Ae.triuncialis and cultivated primitive emmer wheat,respectively [5,6,10–13].These genes have been important resistant sources for the improvement of wheat HF resistance in last several decades.Chen et al.[1] found that only five genes,H13,H21,H25,H26,and Hdic,conferred high levels of resistance to six HF populations collected from Texas,Oklahoma,and Kansas;while eight genes,H12,H13,H17,H18,H22,H25,H26,and Hdic,were found highly effective against five HF populations collected from Texas,Louisiana,and Oklahoma [14].Other two surveys revealed that only six genes (H12,H18,H24,H25,H26,and H33) remain to be resistant against the HF biotypes collected from the southeastern U.S.[15,16].Therefore,continuous search for new resistance genes is essential to provide sufficient HF resistant sources for improvement of HF resistance in new wheat cultivars.

Most of the previously reported HF resistance genes have been mapped to wheat chromosomes using cytogenetic tools and various DNA markers,including simple sequence repeats (SSR),randomly amplified polymorphic DNA (RAPD),restriction fragment length polymorphism (RFLP),amplified fragment length polymorphism (AFLP),sequence-tagged-site (STS),and single nucleotide polymorphism (SNP) [7–9,17,18].Most of these genes were mapped on chromosome arm 1AS including H3,h4,H5,H6,H9,H10,H11,H12,H14,H15,H16,H17,H19,H28,H29,and Hdic[10,11,13,19,20],and chromosome 2B including H8,H24,H26,H18,H20 and H21 [21–24].Several others were mapped on the other chromosomes including H34 and QHara.icd-6B on chromosome 6B [5,7],H13 and H23 on chromosome 6D [17,23],and H33,H7,H36,H35,H31,H22,H32 and H27 on chromosomes 3A,6A,7A,3B,5B,1D,3D,4D,respectively [5,6,8,24,25].

Several next-generation-sequencing-based genotyping platforms have been used in wheat gene mapping.Genotyping-bysequencing (GBS) that provides genome-wide high-density SNP coverage becomes a more popular genotyping platform for developing DNA markers for fine mapping,gene cloning,and genomic selection[26,27].Using GBS-SNPs,our previous QTL mapping studies located HF resistance genes h4 on 1AS [20],H7 on 6AL,and H8 on 2B[24],H35 on 3BS,and H36 on 7AS[6].Another HF resistance QTL QHf.osu-1Adwas mapped within a 2.7 cM region on chromosome arm 1AS of a hard winter wheat cultivar ‘Duster’ using GBS-SNPs [26].Besides,wheat SNP arrays with different resolutions including 9 K and 90 K SNP arrays have also been developed and used in mapping genes for HF resistance in wheat.Using the 9 K SNP arrays,we identified H34 from a wheat cultivar ‘Clark’and assigned it to chromosome arm 6BS [7].The recent release of the new wheat reference genome sequence[28]provides a powerful tool for identifying genome-wide sequence variations and the construction of high-density physical maps for gene localization.

‘Chokwang’ is a Korean winter wheat cultivar carrying QTL for FHB resistance [29].Previous screening found that Chokwang had HF resistance.The objective of this study was to locate the HF resistance QTL in Chokwang using a high-density GBS-SNP map constructed from a recombinant inbred line (RIL) population of a cross‘Ning 7840′×Chokwang,and to develop user-friendly markers for marker-assisted selection (MAS).

2.Materials and methods

2.1.Plant materials

A population of 179 F8RILs from Ning 7840 × Chokwang was developed by single-seed descent for mapping of HF resistance QTL in Chokwang.Chokwang is a Korean winter wheat cultivar,while Ning 7840(‘Aurora’/‘Anhui 11’//‘Sumai 3’)is a Chinese facultative breeding line that is highly susceptible to HF.Three resistant cultivars,‘Carol’with H3,‘Caldwell’with H6,and‘Molly’with H13,and a susceptible cultivar ‘Danby’,were used as resistant and susceptible controls,respectively,for phenotyping.A U.S.diversity panel including 203 accessions (AM203) [30] was used to validate the genetic effect of the newly developed KASP markers in predicting the HF resistance QTL identified in this study.

2.2.HF resistance evaluation

The RIL population,their parents,and the four controls were evaluated for reactions to HF biotype‘Great Plains’(GP),a predominant biotype in the U.S.Great Plains [1,31],in two greenhouse experiments,designated as Exp 1 and Exp 2 hereafter,at Kansas State University,Manhattan,KS,in fall 2018.The AM203 diversity panel was also phenotyped in three greenhouse experiments using the same method at the same location in fall 2013,spring 2019,and fall 2019,respectively.Each planting tray contained a mixture(2:1:1)of soil,sand,and vermiculite as the growth medium and 12 rows were uniformly made on the medium for planting.In each experiment,approximately 20 seeds per experimental line were planted in a half row together with the four controls in the two middle rows on each tray.The greenhouse temperature was set at 18 °C with a 14 h/10 h (light/dark) photoperiod.At the 1.5-leaf stage,HF biotype GP adults were released to seedlings by putting HF stocks under a cheesecloth tent along with trays containing seedlings for infestation.The tent allowed the flies to move only within the tent and maintained appropriate humidity during HF infestation,and then was removed after about one week when larvae hatched from eggs.HF damage on infested seedlings was scored two weeks after the tent was removed.Resistant plants usually grew with light green leaves and dead larvae at the bottom of the leaf sheath although some tiny larvae could survive on some resistant plants.Susceptible plants were stunted with dark green leaves and with larvae growing between leaf sheaths at the base.The percentage of resistant plants per line was calculated as the resistance scores for QTL analysis.Phenotypic ratings were rechecked one week after first scoring to ensure data accuracy.

2.3.DNA extraction and GBS analysis

Leaf tissues at the two-leaf stage were collected from the RILs and their parents and put into 1.1-mL 96-deep-well plates with a 3-mm stainless bead in each well.Leaf tissues were freeze-dried for 48 h and then ground for 3 min at 30 cycles per sec to fine powder in a Mixer Mill (Retsch GmbH,Germany).DNA was isolated using the cetyl trimethyl ammonium bromide (CTAB) method[32].DNA quality was checked by electrophoresis on a 1% agarose gel and quantified using a Quant-iT PicoGreen dsDNA assay kit following the manufacturer’s protocol (Life Technologies,Grand Island,NY,USA).All DNA samples were normalized to 20 ng μL-1for GBS library construction.

The GBS libraries were constructed from the 179 RILs and parents using HF-PstI and MspI restriction enzymes following a previous reported protocol [33].GBS-SNPs were called using a reference-free Universal Network-Enabled Analysis Kit (UNEAK)pipeline implemented in the TASSEL [34–36].The GBS-SNPs with ＜20% missing data and ＜20% heterozygotes were used for further linkage map construction.

2.4.Linkage map construction and QTL analysis

The BIN function in QTL IciMapping 4.1[37]was used to remove redundant markers.Markers with the least missing data points were retained in each bin.The genetic linkage map was constructed using the regression mapping algorithm and Kosambi mapping function in JoinMap 4.0[38].Linkage groups were identified at a minimum LOD score of 3.0 [39].Linkage groups were assigned to corresponding chromosomes by blasting GBS-SNP tag sequences in the IWGSC reference genome RefSeq v1.0 [28].Composite interval mapping (CIM) in WinQTLCart software v2.5 was used to scan QTL at a 1-cM window and claimed to be significant at a LOD score of 3 [39,40].

2.5.Conversion of GBS-SNPs into KASP makers

To fill up possible missing data and more accurately determine QTL location,the GBS-SNPs within a major QTL interval were converted into KASP assays to re-genotype the mapping population.Allele-specific primers for each KASP marker were designed based on the GBS sequences carrying the SNPs using Primer3 (https://bioinfo.ut.ee/primer3-0.4.0/).Two different tail sequences,GAAGGTGACCAAGTTCATGCT and GAAGGTCGGAGTCAACGGATT,were added to the 5′-end of the two forward primers,respectively,to match with the FAM-and HEX-fluorescence-dye-labeled sequences in the KASP reaction mix.

KASP assays were performed in a Veriti Thermal Cycler(Applied Biosystems,Foster City,CA,USA)using a 4-μL reaction mix consisting of 1.94 μL of 2× PACE Genotyping Master Mix (www.3crbio.-com),0.06 μL primer mix,and 2 μL DNA.The PCR profile started with 15 min at 94 °C,followed by 32 cycles of 94 °C for 20 s and 55–60 °C for 1 min.PCR products were scanned in a FLUOstar Omega microplate reader (BMG Labtech Inc.,Cary,NC,USA) and data were analyzed using Klustercaller software v3.4.1.39 (LGC group,Teddington,UK).

KASP primers were first screened for polymorphisms between parents,and polymorphic markers were then used for genotyping the RIL population.The GBS-SNP data were replaced by the corresponding KASP marker data to reconstruct the linkage group and remap the QTL.The sequences harboring GBS-SNPs within the candidate QTL region were used to construct two physical maps by aligning their sequences to the IWGSC RefSeq v1.0 [28] and v2.0(https://www.wheatgenome.org/) wheat assembly.Based on the GBS-SNP data of the RILs,the lines with a recombination within the QTL region were identified and used to redefine the candidate QTL region according to the physical maps.The developed KASP markers were also used to screen the AM203 panel to validate their prediction accuracy of the HF resistance QTL in the panel.

3.Results

3.1.Reactions of the RIL population and their parents to HF infestation

The resistant parent Chokwang and the HF resistant controls Carol(H3),Caldwell(H6),Molly(H13)showed resistance to HF biotype GP,whereas the susceptible parent Ning 7840 and the susceptible control Danby were completely susceptible to HF biotype GP.Chokwang had 80% and 75% resistant plants in Exp 1 and Exp 2,respectively.The HF resistance scores of the RILs showed a bimodal distribution with two peaks at 0 and ＞90% of resistant plants per RIL,respectively (Fig.1).Comparing the resistance scores of RILs with their parents,the RILs were arbitrarily classified into four classes:highly resistant as for Chokwang with HF resistance scores ＞ 80%;moderately resistant with HF resistance scores between 40% and 80%;moderately susceptible with HF resistance scores between 1% and 40%;and completely susceptible with 100% susceptible plants.Based on this classification,the numbers of the RILs in the four classes are 59,29,25,and 66,respectively,in Exp 1;and 66,23,21,and 69,respectively,in the Exp 2.When the resistant and moderately resistant groups were combined as the resistant group,and moderately susceptible and highly susceptible groups were combined as the susceptible group,the segregation ratio between the resistant and susceptible groups fit 1:1 for single gene segregation (=0.05 ＜6.63,P ＞0.01)(Table 1),suggesting one major locus conditioning the HF resistance in Chokwang.

3.2.Linkage map construction

GBS analysis on the 179 RILs and their parents generated 14,028 SNPs with 2748 SNPs at ＜20% missing datapoint rate.The 2748 GBS-SNPs were binned to 1158 bins for map construction,and 1147 of them were finally mapped to 34 linkage groups that covered a total genetic distance of 2088.8 cM at an average marker density of 1.8 cM per marker (Table S1).The 34 linkage groups were anchored to 21 wheat chromosomes with the most SNPs(1) on 3B spanning 127.8 cM and the least SNPs (7) on chromosome 4D spanning 17.4 cM.

3.3.HF resistance QTL mapping and KASP marker development

The phenotypic data for HF resistance from the two experiments and their mean value were used for QTL analysis.Composite interval mapping (CIM) identified one significant QTL,designated as QHf.hwwg-6BS,on chromosome 6B in both experiments and the mean value,which agreed with the frequency distribution of resistance scores and the χ2-test result from the RIL population.QHf.hwwg-6BS was flanked by GBS15120 and GBS21561 on the distal region of 6BS.

The sequences of 10 GBS-SNP markers in the QHf.hwwg-6BS region were selected to design KASP primers and seven GBS-SNP markers (GBS25650,GBS10252,GBS26836,GBS28451,GBS9198,GBS7626 and GBS29023) were successfully converted to KASP markers and renamed as KASP-6B112698,KASP-6B2147503,KASP-6B3625151,KASP-6B7901215,KASP-6B8870440,KASP-6B9195610,and KASP-6B10680899,respectively,based on their chromosome names(6B)and physical positions on Chinese Spring reference genome RefSeq v1.0 (Table 2).These KASP markers showed the same segregation pattern as their corresponding GBS-SNPs,but filled up missing datapoints of GBS-SNPs in the RIL population.

Fig.1.Frequency distribution of HF resistance scores of 179 Ning 7840×Chokwang recombinant inbred lines evaluated in the two greenhouse experiments of fall 2018,and the mean values of the two experiments.

Table 1 Chi-square (χ2) test of one gene or two genes segregation ratios of HF resistance in Ning 7840 × Chokwang recombinant inbred line population in the two greenhouse experiments of fall 2018,and the mean values of the two experiments.

The linkage map of chromosome 6B was updated by replacing the seven GBS-SNPs in the QTL region with the corresponding KASP markers and the new 6B linkage map contained 78 SNP markers(Fig.2a).Using the new map,QHf.hwwg-6BS was remapped to a 0–8.6 cM interval between KASP-6B112698 and GBS21561,with the QTL peak at KASP-6B7901215 (6B:7,901,215 bp) (Fig.2a;Table 3).The flanking markers KASP-6B112698 and GBS21561 were aligned to 112,698 bp and 19,318,874 bp,respectively,on chromosome 6B based on IWGSC reference genome RefSeq v1.0(Table 3).This QTL explained 80.2%to 84.0%of the phenotypic variation with high LOD scores from 71.4 to 80.9 in the two experiments and the mean value from the two experiments (Table 3).Chokwang contributes the allele for increased HF resistance.

3.4.Delimiting candidate interval of QHf.hwwg-6BS

By blasting the sequences of the three GBS-SNP tags and seven KASP markers in the QHf.hwwg-6BS region against the IWGSC reference genome RefSeq v1.0,nine of them were aligned to the short arm of 6B and used to construct a physical map(Fig.2b).Genotypic analysis of the RIL population using the nine markers identified two RILs (NCk104 and NCk068) with homozygous recombination within the QHf.hwwg-6BS region.The RIL NCk104 has the recombination between GBS23947 (3,824,582 bp on 6BS) and GBS23103(7,782,286 bp on 6BS),and carries the Ning 7840’s allele at GBS23947 and the Chokwang’s allele at GBS23103 (Fig.2b).This RIL showed HF resistance with 86.1%HF resistant plants to the biotype GP.Combined analyses of the genotypic and phenotypic data of NCk104 moved the left flanking marker for QHf.hwwg-6BS to GBS23947 and suggested that the left border of the candidate QTL interval is between 3,824,582 bp and 7,782,286 bp.

The RIL NCk068 has another recombination between KASP-6B7901215 or GBS28451 (7,901,215 bp on 6B) and KASP-6B8870440 or GBS9199 (8,870,440 bp on 6B),and carried the Chokwang’s allele at KASP-6B7901215 and Ning 7840’s allele at KASP-6B8870440 (Fig.2b).NCk068 showed complete resistance to HF biotype GP,suggesting that KASP-6B8870440 is the right flanking marker for QHf.hwwg-6BS and the right border of the QTL candidate interval is between 7,901,215 bp and 8,870,440 bp on 6BS.Based on these data,QHf.hwwg-6BS can be delimited to a 5,045,858 bp physical interval between 3,824,582 bp and 8,870,440 bp on the terminal end of chromosome arm 6BS.Searching for annotated candidate genes in this region using the IWGSC RefSeq annotation v1.1 (IWGSC 2018) identified 96 highconfidence candidate genes (Table S2).Among them,34 genes are annotated as plant disease resistance genes.

The recent version of IWGSC RefSeq v2.0 was also used to estimate the possible physical interval for QHf.hwwg-6BS (Fig.2b).In the new map,KASP-6B112698 was mapped to 6,028,601 bp on 6B as the left flanking marker for QHf.hwwg-6BS based on the genotypic and phenotypic data of NCk104.KASP-6B8870440 was moved to 10,779,424 bp as the right flanking marker based on the information of NCk068.Therefore,QHf.hwwg-6BS is delimited to a 4,750,823 bp physical interval between 6,028,601 bp and 10,779,424 bp based on IWGSC RefSeq v2.0.There are 69 highconfidence candidate genes in the region with 29 plant disease resistance-related genes according to the IWGSC RefSeq annotation v1.1 (Table S2).

3.5.Validation of KASP markers

To evaluate the effectiveness of the markers in the QHf.hwwg-6BS region on marker-assisted selection,all the seven KASP markers were used to screen the AM203 diversity panel.Significant differences in HF resistance were found between the two groups of accessions carrying the contrast alleles at markers KASP-6B112698 (P=0.00002) and KASP-6B7901215 (P=0.010),respectively (Fig.3).The differences for the other five markers were not significant.Among the two significant markers,KASP-6B112698 detected 11 of the 12 (91.7%) accessions that carry the resistance marker allele and showed HF resistance (Table S3),therefore,this marker is near diagnostic for the QTL.For KASP-6B7901215,51 out of 123 accessions carrying the resistance alleles had HF resistance (Table S3).When both markers were combined to analyze the panel,the detection accuracy was not improved compared to the result from the single marker KASP-6B112698.

Table 2 Primer sequences of KASP markers used in this study and the physical locations of the SNPs in IWGSC RefSeq v1.0.

Table 3 Chromosome location and effect of the HF resistance quantitative trait locus (QTL) QHf.hwwg-6BS detected in the Ning 7840 × Chokwang recombinant inbred line population evaluated in fall 2018.

Fig.2.Linkage and physical mapping of the HF resistance QTL QHf.hwwg-6BS.a) Linkage group of the chromosome 6B and the composite interval mapping of the HF resistance QTL QHf.hwwg-6BS;b) Delimit QHf.hwwg-6BS to a 5.05 Mb physical interval on chromosome arm 6BS of IWGSC RefSeq v1.0 and a 4.75 Mb physical interval on chromosome arm 6BS of IWGSC RefSeq v2.0 based on the genotype and phenotype of two RILs that have recombination in the candidate interval of QHf.hwwg-6BS.The bars with black and gray color correspond to HF resistant Chokwang genotype and susceptible Ning 7840 genotype,respectively.The number of seedlings representing the number of resistant seedlings and the total number of screened seedlings.The two markers with italic names are GBS markers,while the other seven are KASP markers developed in the present study.

Fig.3.Genotyping result of KASP markers in diversity panel.(a)Allele distribution of KASP marker KASP-6B 112698;(b)Hessian fly resistance score(%)of accessions from the 203 U.S.diversity panel carrying resistant(R)and susceptible(S)alleles at the KASP markers KASP-6B112698 and KASP-6B7901215.In(a),the green dots are samples from resistant genotypes as Chokwang;the blue dots are susceptible genotypes as Ning 7840;and the black dots are H2O controls.

4.Discussion

4.1.Genetic relationships among three QTL mapped on 6BS

To date,37 HF-resistance genes have been formally named[5,6],and several other QTL have been reported [7,26,41–43].Among them,only four HF resistance genes and one QTL have been mapped to B genome,including H20 on 2B [21],H35 on 3BS [6],H31 on 5BS [9],H34 [7] and QHara.icd-6B on 6BS [5].Based on the physical locations of the flanking markers on 6BS in IWGSC reference genome RefSeq v1.0,H34 was located between 8.409 and 15.134 Mb in ‘Clark’ [7],and QHara.icd-6B was delimited to the interval of 0.167–11.299 Mb in a durum wheat [5];therefore,the two QTL regions overlapped each other.In this study,QHf.hwwg-6BS was identified in a similar region on 6BS in Chokwang(Table 3;Fig.2a).Genotypic and phenotypic analyses of the RIL population of Ning 7840 × Chokwang identified two critical recombinations in the candidate interval of QHf.hwwg-6BS from the two RILs,which delimited QHf.hwwg-6BS to a 5.05 Mb physical interval between 3,824,582 bp and 8,870,440 bp of chromosome 6B based on IWGSC RefSeq v1.0 (Fig.2b;Table 3).The QHf.hwwg-6BS flanking region is within the QHara.icd-6B candidate region,but only partially overlaps with the H34 flanking region.More recently,a new version of Chinese Spring wheat genome reference,IWGSC RefSeq v2.0,has been released,and the physical candidate region of QHf.hwwg-6BS in the new version of reference is between 6.029 and 10.779 Mb that is within the QHara.icd-6B candidate region (5.276–13.789 Mb),and partially overlaps with the H34 flanking region(10.251–18.148 Mb).The difference in the physical intervals of QHf.hwwg-6BS between the two versions of the Chinese Spring reference genome is most likely due to improved reference sequence assembly accuracy in RefSeq v2.0.Therefore,QHf.hwwg-6BS is most likely located in the 4.75 Mb(6.029–10.779 Mb)interval (Fig.2b;Table S2).These data suggested that the three QTL identified on chromosome arm 6BS most likely are in the same physical region.However,they were from different origins:H34 from a U.S.bread wheat cultivar,QHf.hwwg-6BS from a Korean bread wheat cultivar,and QHara.icd-6B from durum wheat.Further allelic tests and cloning of these genes will provide conclusive evidence to determine their relationship.

4.2.Annotated genes in the QHf.hwwg-6BS region

Several previous studies attempted to predict putative candidate genes for QTL based on the physical locations of their flanking markers.This could facilitate a better understanding of the possible genes underlining target QTL.However,the QTL mapping data need to meet the following assumptions:1) using a high-density genetic map and accurate phenotypic data for QTL mapping;2)QTL with a major effect and contrast alleles that can be phenotypically separated easily;3)high collinearity in the physical maps of the QTL regions between the reference genome and QTL donor’s genome;and 4) the candidate interval is reasonably small.

In the present study,QHf.hwwg-6BS interval was delimited to a 4.75 Mb physical interval of IWGSC reference genome RefSeq v2.0.Searching the annotated candidate genes in the interval found 69 high-confidence genes (Table S2).Among the 69 annotated genes,29 are plant disease resistance-related genes that encode NBS(nucleotide-binding site)-LRR (leucine-rich repeat)-like resistance proteins,protein kinases or receptor kinases,F-box proteins,transmembrane proteins,a senescence-associated protein DIN1,and a Hfr-2-like protein.Since the current candidate resistance gene list is still too long to determine the final candidate gene underlining QHf.hwwg-6BS,further work to finely map and clone QHf.hwwg-6BS will pinpoint the causal candidate gene for the QTL.

4.3.Molecular markers for HF resistance improvement

KASP markers are more suitable for marker-assisted selection in many breeding programs due to its low cost per datapoint and suitable for medium to high throughput screening.More and more KASP markers have been developed for marker-assisted selection in wheat breeding.For HF resistance,KASP markers are available only for several HF resistance genes or QTL,including H32 [31],QHara.icd-6B [5],h4 [20],H7 [24],and H35 and H36 [6].However,most of the markers for these genes or QTL are populationspecific and have not been validated in a diversity panel.

In this study,seven GBS-SNPs close to QHf.hwwg-6BS were successfully converted to KASP markers(Table 2)and used to screen a diversity panel including 203 U.S.winter wheat elite breeding lines.Two of them (KASP-6B112698 and KASP-6B7901215)showed significant difference in HF resistance between accessions carrying the contrast alleles (Fig.3b).Among them,KASP-6B112698 showed better prediction accuracy and 11 of 12 accessions with the positive allele showed HF resistance,which supports that the reference RefSeq v2.0 is more accurate than RefSeq v1.0 because KASP-6B112698 is 0.5 Mb closer to the QTL in RefSeq v2.0,whereas 3.5 Mb farther to the QTL in RefSeq v1.0 than KASP-6B3625151 (Fig.2b).The higher prediction accuracy of KASP-6B112698 also supports that the causal gene of QHf.hwwg-6BS is closer to KASP-6B112698 than the right flanking marker KASP-6B7901215.Therefore,KASP-6B112698 can be used for screening of QHf.hwwg-6BS and pyramiding it with other resistance genes.

CRediT authorship contribution statement

Lirong Zhang:Writing–original draft,Investigation,Data curation,Formal analysis.Yunfeng Xu:Writing– original draft,Writing -review &editing,Investigation,Data curation,Formal analysis.Ming-Shun Chen:Investigation,Data curation,Formal analysis.Zhenqi Su:Investigation,Data curation,Formal analysis.Yang Liu:Investigation,Data curation,Formal analysis.Yuzhou Xu:Investigation,Data curation,Formal analysis.Guixiao La:Investigation,Data curation,Formal analysis.Guihua Bai:Conceptualization,Resources,Supervision,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 is contribution number 21-281-J from the Kansas Agricultural Experiment Station.This project is partly funded by the National Research Initiative Competitive Grant (2017-67007-25939)from the U.S.Department of Agriculture,National Institute of Food and Agriculture.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.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.2021.08.004.