SNP-based identification of QTL for resistance to black point caused by Bipolaris sorokiniana in bread wheat

Qiaoyun Li,Runyu Hu,Zhenfeng Guo,Siyu Wang,Chuang Gao,Yumei Jiang,Jianwei Tang,Guihong Yin*

National Engineering Research Centre for Wheat,College of Agronomy,National Key Laboratory of Wheat and Maize Crop Science,Henan Agricultural University,Zhengzhou 450002,Henan,China

Keywords:Black point Molecular marker Quantitative trait locus Triticum aestivum

ABSTRACT Black point disease caused by Bipolaris sorokiniana is a problem in wheat production worldwide.We aimed to identify major quantitative trait loci (QTL) for resistance to black point and develop molecular markers for marker-assisted selection (MAS).A recombinant inbred line(RIL) population derived from a cross between Wanyuanbai 1 (susceptible) and SN4143 (resistant) was evaluated for black point response at three locations during two years under artificial inoculation with B.sorokiniana,providing data for six environments.Thirty resistant and 30 susceptible RILs were selected to form resistant and susceptible bulks,respectively,that were genotyped by the wheat 660K SNP array;685 singlenucleotide polymorphisms (SNPs) were identified,among which 385 (56.2%) and 115 (16.8%) were located on chromosomes 4A and 2B,respectively.Bulked segregant RNA-Seq analysis identified candidate regions on chromosomes 4A (4.60–40.28 Mb) and 5A (1.22–48.47 Mb).Genetic linkage maps were constructed for chromosomes 2B,4A,and 5A using 59 polymorphic dCAPS and SSR markers.Finally,two QTL,designated QBB.hau-4A and QBB.hau-5A,were detected on chromosomes 4A and 5A,respectively.The resistance allele of QBB.hau-4A was derived from SN4143,and that of QBB.hau-5A came from Wanyuanbai 1.QBB.hau-4A with a large and consistent effect(15.1%)is likely to be a new locus for black point resistance.The markers linked to QBB.hau-4A and QBB.hau-5A have potential application in MASbased breeding.

1.Introduction

Black point is a widespread grain disease in major wheat(Triticum aestivum)-producing countries,including China,the USA,Australia,Argentina,Canada,and Egypt [1–6].Black point infection not only causes substantial economic losses due to discolored grain[7],but also reduces seed germination and inhibits seedling growth[1,4].Some fungal species causing black point may also produce toxic substances [8–10].Consequently,many countries have regulations on levels of black point incidence in wheat seeds.For example,in Australia,infected grains delivered to silos must not exceed 5%[11],whereas in China,where black point grain is considered as one of six sources of defective grain,the proportion of defective grains in first and second grade wheat must be lower than 6% [12].

Currently,most cultivars grown in the Yellow and Huai River wheat area of China(YHWAC)are susceptible to black point under natural field conditions [4],whereas some cultivars show resistance under conditions of artificial inoculation.For example,among the 155 wheat lines exhibiting black point resistance under field conditions during 2010–2012 in the YHWAC,only 6,3,78,and 84 showed resistance when inoculated with Bipolaris sorokiniana,Exserohilum rostratum,Epicoccum sorghinum,and Alternaria alternata,respectively [5].

Apart from differences in resistance among wheat genotypes,fungal species and environmental conditions also influence the level of black point incidence.Multiple pathogens have been identified as the causal agents of black point,among which A.alternata,B.sorokiniana,and Fusarium spp.are most common [1,13–16].In previous studies,we identified eight fungal species,including A.alternata,B.sorokiniana and F.equiseti,as causal agents of black point infection in the YHWAC;and among them,B.sorokiniana and A.alternata were the most prevalent [14].Disease symptoms tend to differ across the causal species [5],and one wheat cultivar may variable responses to different species [7,16].The environmental conditions that influence black point incidence mainly include meteorological factors during the grain-filling stage and agricultural practices [2,17–20].In this regard,significant differences have been found among genotypes,environments,and genotype × environment interactions [3–4,6,20].

Response to black point in wheat is a quantitative trait,with intermediate incidence in F1plants between resistant and susceptible parents[6].The incidence in F2,recombinant inbred line(RIL)and doubled haploid(DH)populations shows continuous distributions and transgressive segregation [3,6,21–23],indicating that resistance is conferred by polygenes.

Genetic studies on black point resistance in wheat began early in this century [22] and increased rapidly in the past five years,concomitant with advances in sequencing and chip technologies[3,6,22,24,25].Using a RIL population derived from cross Linmai 2×Zhong 892 and the wheat 90K SNP array,Liu et al.[3]identified nine quantitative trait loci(QTL)for black point resistance on chromosomes 2AL,2BL,3AL,3BL,5AS,6A,7AL (2),and 7BS,each explaining 3.7%–15.6% of the phenotypic variances.They also detected 25 resistance loci for black point based on a genomewide association study(GWAS)of 166 elite wheat cultivars grown in ten environments using the wheat 90K and 660K SNP arrays[24].Although more than 200 loci for black point resistance have been reported[3,6,22,24,25],major QTL associated with resistance to this disease remain to be ascertained.Relatively few studies have examined the use of molecular markers in selection of black point resistance in wheat breeding.

As black point develops during grain filling the response can be evaluated only after flowering,and the symptoms are evident only after harvest and threshing [5,6,20,26].Assessment is timeconsuming and the results can only serve as a reference for planning crosses in the following year in the YHWAC.Moreover,under natural field conditions,results vary among locations and years[4–6,16,24].Accordingly,in the absence of reliable large-scale,efficient tests of black point levels,markers linked to resistance genes would be a valuable tool for plant breeders.However,few molecular markers are currently available for use in wheat breeding thus restricting the rate of progress in achieving resistance.

We previously identified 36 highly resistant and 7 highly susceptible lines among 403 wheat genotypes based on their black point incidence under natural field conditions from 2010 to 2012[4],and subsequently developed eight RIL populations from crosses between resistant and susceptible lines.In the present study,we used a RIL population derived from a cross between Wanyuanbai 1 (susceptible) and SN4143 (resistant) to map QTL for resistance to black point caused by B.sorokiniana.Our objectives were to:(1) identify major QTL for resistance to black point,and (2) develop molecular markers for reducing the level of black point in wheat breeding.

2.Materials and methods

2.1.Mapping population and field trials

The 327 F2:7RILs used for QTL mapping were derived from the cross Wanyuanbai 1 × SN4143.The average incidences of black point in Wanyuanbai1 and SN4143 were 53.7% and 5.8%,respectively,following inoculation with B.sorokiniana from 2016 to 2018 [16].

The RILs and parents were planted at three locations,Neihuang(35°94′N,114°88′E),Xingyang (34°79′N,113°35′E),and Xiping(33°63′N,115°17′E) in Henan province in randomized complete blocks with three replications during the growing seasons 2018–2019 and 2019–2020 (hereafter referred to as 2019 and 2020,respectively).Seeds were planted between the mid-October and early November,and grains were harvested by the end of May or at the early June.Plots consisted of single 1 m rows with approximately 30 plants in each row spaced 20 cm apart.Field trials were managed according to the local practices.

2.2.Evaluation of black point incidence

The pathogen,Ta-BP33 isolate of B.sorokiniana (Fig.1a,b),was cultured as reported previously[6,14].Conidial suspensions of Ta-BP33(1×105conidia mL-1)were prepared following Mahto et al.[27].Five spikes at mid-anthesis in each plot (Zadoks GS 65 [29])were treated with the conidial suspensions using a hand sprayer.The spikes were covered with transparent plastic bags for 5 days to maintain humidity after inoculation [5,16].

Inoculated spikes were harvested at GS 92 [28],sun-dried and hand threshed.The number of total and diseased kernels (black or brown discoloration area ＞1 mm2)was recorded,and the black point incidence was determined using the following equation:

Incidence (%)=number of diseased kernels/number of total kernels × 100

Thirty lines with the lowest and 30 with the highest average incidence across six environments were selected for BSA and BSR-Seq analysis in 2020.

Analysis of variance(ANOVA)was performed using SPSS statistical software (Version 20.0;IBM Corporation,Armonk,NY,USA).The black point incidence was arcsine square root-transformed to equalize variance in the average incidence among lines.Correlation analysis for black point incidence among different years and locations was based on correlate-bivariate correlation [6].

2.3.DNA extraction and BSA

Wheat leaves were harvested from the RILs and the parents to extract genomic DNA using the cetyltrimethylammonium bromide method [29].Equivalent DNA mixtures from 30 resistant and 30 susceptible RILs were used to form resistant and susceptible pools,respectively.The DNA pools,along with the parental DNA,were genotyped using the 660K SNP array at the China Golden Marker Biotech Co.,Ltd.(Beijing,China,http://www.cgmb.com.cn).

SNP genotyping and clustering were processed using Affymetrix Axiom Analysis Suite software(Thermo Fisher Scientific,Waltham,MA,USA).A dish QC (DQC) value ≥0.82 and call rate (CR) ≥90 were used as the criteria for SNP filtering.The high-resolution SNPs between contrasting pools and between parents were assumed to be associated with the black point response.Using the wheat 660K SNP array,we identified the physical positions of SNP markers based on the Chinese Spring reference genome sequences v1.0(IWGSC,http://www.wheatgenome.org/).

2.4.BSR-Seq analysis

Kernels from 30 resistant and 30 susceptible lines inoculated with B.sorokiniana were harvested at 14 h and 12 d post inoculation,chilled in liquid nitrogen,and stored at -80 °C for total RNA extraction using the TRIzol method [30].The concentration and integrity of RNA were assessed using a ND-1000 spectrophotometer (NanoDrop Technologies,Wilmington,DE,USA) and a Gel Doc XR+imaging system (Bio-Rad Laboratories,Inc.,Hercules,CA,USA).The cDNA libraries of the two parents and the two bulks sampled at 14 h and 12 d were constructed using an NEBNext Ultra RNA Library Prep Kit for Illumina (New England Biolabs,Ipswich,MA,USA).RNA sequencing was performed using the Illumina Novaseq 6000 platform according to the manufacturer’s protocol(Biomarker Technologies Corporation,Beijing,China).The raw RNA-Seq reads obtained were filtered to remove low-quality nucleotides,reads with adapters,or those with more than 10%unidentified nucleotides,and the resultant reads were aligned to the Chinese Spring reference genome published by the IWGSC using STARv2.4.0j software [31] with default parameters.SNP and InDel were called using GATK software [32].

Fig.1.Fungal and conidial morphologies of Bipolaris sorokiniana isolate Ta-BP33,phenotypic reactions of the resistant and susceptible parents and lines,and distribution of SNPs associated with black point resistance on chromosomes based on bulked segregant analysis and the 660K SNP chip.(a,b)Colony and spore morphologies of isolate Ta-Bp-33 grown on PDA medium at 25°C for 7 days in the dark.(c),(d),(e)and(f)are the resistant parent,susceptible parent,resistant line(WS114)and susceptible(WS57)line,respectively,following inoculation with B.sorokiniana at Xingyang in 2020.Photographs were taken of seed harvested at 2 months post inoculation.(g)Distribution of SNPs associated with black point resistance on chromosomes.(h) Distribution of associated SNPs on chromosome 4A.

SNPs with a P-value ≤0.01,exhibiting different major alleles in both pools and a product of major allele coverage ratio in both pools ≥90%,as well as a minor allele in each pool ≤4 reads,were defined as linked SNPs and selected for further analysis of candidate regions.The regions associated with the black point resistance were analyzed using Euclidean distance(ED)and SNP-index methods[33,34].The regions determined using both methods were considered to be candidate regions identified by RNA-Seq.

2.5.Development of dCAPS markers

SNPs associated with black point resistance on chromosomes 2B,4A,and 5A identified by BSA and BSR-Seq analyses were selected to develop derived cleaved amplified polymorphic sequence (dCAPS) markers.DNA sequences containing SNPs associated with the black point resistance were searched using URGI BLAST (https://urgi.versailles.inra.fr/blast/blast.php),along with sequences flanking the SNPs (71 bp).Forward primers were designed using dCAPS Finder 2.0 software (http://helix.wustl.edu/dcaps/dcaps.html) based on the mutant alleles and sequences flanking SNPs.Reverse primers were designed based on sequences containing SNPs using Primer Premier software(V5.0;http://www.premierbiosoft.com/primerdesign/index.html).The primers were validated for polymorphisms in the resistant and susceptible parents.One hundred and sixty-three wheat microsatellite markers on chromosomes 2B,4A,and 5A were selected for testing polymorphisms between the parents and between the bulks.Information on these markers was obtained from the Grain Genes database(http://wheat.pw.usda.gov/).

Amplifications were performed using a 96-well thermocycler(TCA0096;Thermo Fisher Scientific,Vantaa,Finland).The PCR mixture(10 μL)contained 50 ng of genomic DNA,200 μmol L-1of each dNTP,1.5 mmol L-1of MgCl2,0.25 μmol L-1of each primer,and 1.25 U Taq DNA polymerase (TaKaRa Bio,Beijing,China).Samples were amplified using the following PCR profile:a single cycle denaturation at 94 °C for 3 min;followed by 35 cycles at 94 °C for 30 s,55–65 °C (depending on primers) for 30 s,and 72 °C for 30 s;and a single cycle extension at 72°C for 5 min.The amplified products were digested according to instructions provided for each restriction endonuclease(TaKaRa Bio).The enzyme-digested products were separated using 8% polyacrylamide gels and visualized by silver staining.

2.6.Linkage map construction

Polymorphic dCAPS and SSR markers were used for genotyping the mapping population.Linkage groups were generated using Joinmap V4.0 software [35] (http://www.kyazma.com),and genetic distances (in centimorgans,cM) between markers were estimated based on the Kosambi mapping function [36].Genetic maps were constructed using MapChart V2.2 software [37](http://www.earthatlas.mapchart.com).The QTL analysis was conducted by the composite interval mapping method using QTL Cartographer V2.5 software,with an LOD threshold of 2.5 based on 2000 permutations at P ＜0.01 [38] (http://statgen.ncsu.edu/qtlcart/WQTLCart.htm).Effects of QTL were estimated as the proportion of phenotypic variance (R2) explained by a single QTL.QTL identified in individual environments with overlapping 10 cM intervals were considered common.

2.7.Validation of QTL

The potential application of two QTL for black point resistance in wheat breeding was examined in the RIL population,and 122 wheat lines that were evaluated for black point response in 2016 and 2017[5].QBB.hau-4A and QBB.hau-5A were tested in both populations using the closest linked markers BP-4A-d20 and Xbarc10,respectively.The significance of difference between lines with different combinations of alleles at the QBB.hau-4A and QBB.hau-5A loci were assessed using Duncan’s multiple range t-test at P ＜0.01 (SPSS Statistics 20.0,IBM Corporation,Armonk,NY,USA).

3.Results

3.1.Phenotypic evaluation

In 2019 and 2020,the mean black point incidence was 5.7% for the resistant parent SN4143,and 42.4% for the susceptible parent Wanyuanbai 1 across six environments (Table S1;Fig.1c,d).Continuous distribution of black point incidence among the 327 RILs ranged from 0.2% to 70.6% across the six environments,with a mean of 17.7% (Fig.S1).The black point incidence in individual lines was significantly correlated (r=0.32–0.73,P ＜0.01) among the three locations (Table S2),and the analysis of variance of disease incidence revealed significant differences (P ＜0.01) among lines,years,locations,and lines × years,lines × locations,years × locations,and lines × locations × years interactions(Table S3).

3.2.Resistance loci detected by BSA

The two parental genotypes and the resistant and susceptible bulks were screened using the wheat 660K SNP array.After filtering out SNPs with the criteria of DQC ＜0.82,and CR ＜90,we selected 344,168 SNPs as the high-resolution SNPs from the 660,009 SNPs included in the 660K SNP array.Six hundred and eighty-five SNPs showing polymorphisms between the two parents and between the two bulks were considered to be potentially associated with the resistance to black point(Table S4).These SNPs were located on chromosomes 1A(10),1B(15),2B(115),2D(2),3A(1),3B (4),3D (2),4A (385),4D (1),5A (22),5B (50),5D (23),7A(16),7B (38),and 7D (1),among which chromosomes 4A and 2B contained 56.2% and 16.8% of the associated SNPs,respectively,indicating that major QTL for black point resistance may exist on these chromosomes (Fig.1g,h).

3.3.BSR-Seq analysis

BSR-Seq analysis of the two parents and two bulks generated 311,088,144 clean reads,with an average percentage of Q30 value exceeding 92.4%(Table S5).After quality control,98.6%of the reads were aligned to the Chinese Spring reference sequences,among which 64.8% were uniquely mapped (Table S6).Subsequent SNP calling identified 8745 and 3065 genome-wide high-quality SNPs between the resistant and the susceptible bulks at 14 h and 12 d,respectively (Tables S7,S8).Using the Euclidean distance method,regions associated with black point response were located on chromosomes 4A (4.60–624.11 Mb) and 6B (290.72–295.77 Mb and 349.46–391.68 Mb) at 14 h,and 5A (1.22–66.09 Mb) at 12 d,respectively.Using the SNP-index method,associated regions were identified on chromosomes 4A (4.60–40.28 Mb) at 14 h,and 5A(1.22–48.47 Mb) at 12 d.Based on the results from the two methods,candidate regions were mapped on chromosomes 4A (4.60–40.28 Mb) and 5A (1.22–48.47 Mb),which represented corresponding regions identified by both methods (Table 1).

3.4.Black point resistance-associated QTL on chromosomes 4A and 5A

The BSA and BSR-Seq analyses of the RIL population showed that QTL on chromosomes 2B,4A,and 5A were important for resistance to black point caused by B.sorokiniana.One hundred and sixty-three SSR and 185 dCAPS markers located on chromosomes 2B,4A,and 5A were used to test the resistant and the susceptible parents,and 35 dCAPS and 24 SSR markers showing polymorphisms between the parents were identified (Table S9;Fig.S2)and then used for genotyping the RILs.

Combining the black point incidences in the six environments(Table S1) and genotyping data for the entire RIL population(Table S10),two QTL for black point resistance,designated QBB.hau-4A and QBB.hau-5A,respectively,were mapped on chromosomes 4A and 5A,explaining 3.3%–15.1% of the phenotypic variances.The resistance allele of QBB.hau-4A was derived from the resistant parent SN4143,whereas that of QBB.hau-5A was from the susceptible parent Wanyuanbai 1 (Table 2;Fig.2).

QBB.hau-4A with a large resistance effect was identified across the three environments,accounting for 4.4%–15.1% of the phenotypic variances;it was flanked by BP-4A-d11 and BP-4A-d20 at a genetic distance of 9.6 cM (Table 2;Fig.2a).The QTL on chromosome 5A,detected at all three environments,was flanked by markers Xbarc232 and Xbarc10 at a genetic distance of 19.1 cM(Table 2;Fig.2b).

3.5.Application of QTL for MAS

QBB.hau-4A and QBB.hau-5A were evaluated using the closest linked markers BP-4A-d20 and Xbarc10,respectively.Among the RIL population,markers BP-4A-d20 and Xbarc10 identified 173 and 157 lines with resistance alleles at the loci QBB.hau-4A and QBB.hau-5A,respectively.The black point incidence of lines with resistance alleles was significantly lower than those with susceptible alleles(P ＜0.01).The incidence of 73 lines with resistance alleles at both loci was the lowest (11.5%),and that of the lines with resistance alleles at either QBB.hau-4A or QBB.hau-5A was 15.7%and 16.4%,respectively,which was significantly lower than the lines without resistance alleles at both loci (P ＜0.01,Table 3).

The two markers were also validated on 122 wheat lines inoculated with B.sorokiniana in 2016 and 2017[5].Sixty lines with the resistance allele at QBB.hau-4A had a mean black point incidence of 17.1%,lower than that of nine lines with the resistance allele at QBB.hau-5A only(21.4%),and 33 lines with neither resistance allele(22.9%).The black point incidence of 20 lines with the resistance alleles at both loci (11.3%) was significantly lower than that of the lines with the resistance allele at QBB.hau-5A and lines with neither resistance allele (P ＜0.01,Fig.3;Table S11).

Table 1 SNPs and regions associated with resistance to black point caused by Bipolaris sorokiniana detected via RNA-Seq and bulked segregant analyses.

Table 2 QTL associated with resistance to black point in the Wanyuanbai1/SN4143 RIL population.

Table 3 Effects of QBB.hau-4A and QBB.hau-5A on black point incidence caused by Bipolaris sorokiniana.

Fig.2.Genetic linkage map showing QTL on chromosomes 4A(a)and 5A(b)for resistance to black point in the Wanyuanbai 1×SN4143 RIL population.LOD contours were obtained by composite interval mapping.The LOD scores are indicated on the×axis,and a scale of genetic distance(in centimorgans)is provided.A LOD threshold of 2.5 is indicated by solid vertical lines.The best linear unbiased estimate(BLUE)values for black point scores were calculated for the six environments.XY,XP,and NH denote the experimental sites,Xingyang,Xiping,and Neihuang,respectively.

4.Discussion

Although black point resistance could be improved by transferring resistance QTL into high yielding wheat cultivars using MAS technology [22,23] the current paucity of both relevant molecular markers and resistant germplasms restricts the development of resistant wheat cultivars.The severity of black point is influenced to a large extent by environmental conditions[2,17–20],and multiple pathogens have been implicated as causal agents in wheat[1,13–16].Differences in the reported host-pathogen interactions suggest that the underlying mechanisms of resistance might not be identical [5,7,16].Other researchers believe that black point is caused by stress and that the pathogens that are isolated are present as saprophytes.Black point incidence under natural field conditions tends to differ among locations and years probably as a consequence of differences in the causal pathogens and the meteorological factors during grain filling [3,16].In this regard,the mapping results based only on disease incidence under natural field conditions can be unreliable due to inconsistent phenotypic data.In most previous studies black point response was largely based on such data [3,22,24,25].

Fig.3.Tests of 122 wheat lines with markers linked to the two QTL associated with resistance to black point.Bars denoted by different letters are significantly different at P ＜0.01.4A and 5A refer to QBB.hau-4A and QBB.hau-5A,respectively.

In order to pyramid QTL in high-yielding wheat cultivars by MAS it might be necessary to identify multiple QTL and develop molecular markers for the different causative pathogens,particularly A.alternata and B.sorokiniana than are more commonly isolated.We used a RIL population derived from the resistant parent SN4143 and the susceptible parent Wanyuanbai 1,which were screened for several years after inoculation with B.sorokiniana[16] to identify loci associated with black point resistance under inoculated conditions.Twenty seven of the 327 RILs showed resistance in all six environments,providing resistant germplasm for use in breeding.

4.1.Use of BSA and BSR-Seq to map QTL for black point resistance

The BSA and BSR-Seq technologies are effective for identifying resistance genes in wheat.However,they were mostly used for qualitative traits [39].Here,we detected loci for black point resistance,a typical quantitative trait using both methods.In our previous study,23 major marker-trait associations(MTAs)for resistance to black point following inoculation with B.sorokiniana were detected by GWAS in an incomplete diallel cross population;each explaining more than 11% of the phenotypic variance [6].The 21 major MTAs in the resistant parent SN4143 were located on 13 chromosomes.Based on the result of BSA,BSR-Seq,genetic linkage map construction and QTL analysis,two QTL for black point resistance were mapped on chromosomes 4A and 5A,and the resistance alleles of QBB.hau-4A and QBB.hau-5A were contributed by the parents SN4143 and Wanyuanbai 1,respectively.This is consistent with the observation that some of the RILs had higher resistance than the resistant parent SN4143(Figs.1c–f,S1).The physical positions of markers linked to QBB.hau-5A were close to those of MTAs previously identified on chromosome 5A[6].Multiple studies have mapped QTL for resistance to black point on chromosomes 4A and 5A [3,22,25].QBB.hau-4A and QBB.hau-5A explained 3.3%–15.1% of the phenotypic variance,levels of variation similar to those reported previously [3,22].These results suggest that BSA and BSR-Seq technologies combined with a genetic linkage map to identify QTL for quantitative traits are feasible,and help to reduce they workload of QTL mapping and ensure reliability of results.

The locus on chromosome 4A was identified by BSR-Seq only at 14 h post inoculation and that on chromosome 5A only at 12 d post inoculation.B.sorokiniana is an important pathogen because it can cause multiple diseases in wheat,such as black point,root rot,and leaf spot [14,40].Plants have evolved complex mechanisms in response to fungal infection to allow rapid modulation of biological processes and cellular functions,such as increased antioxidase activity and induced expression of various genes involved in oxidative stress,such as genes associated with pathogenesis-related proteins,reactive oxygen species (ROS)-scavenging enzyme,the ascorbate–glutathione cycle,and glutathione peroxidase cycle[40–42].Besides the ability to counteract oxidative stress during the development of black point,it was reported that enzymatic browning in wheat kernels brings about the production of blackening compounds causing black point symptoms[43].Phenolic acids are natural compounds in wheat kernels.In addition to being substrates for peroxidase and polyphenol oxidase in enzymatic browning,phenolic acids are strong antioxidants[44].The activity of antioxidant enzymes in the resistant lines was higher than that in susceptible lines [43];and excessive ROS caused by B.sorokiniana was eliminated more efficiently in resistant lines than in susceptible lines,indicating that more phenolic acids are important to reduce the excessive ROS in susceptible kernels.Some candidate genes identified for black point resistance are involved in physiological processes,including genes related to enzymes for enzymatic browning,signal transduction pathways of plant hormones and defense mechanisms during stress response [3,6,24].These results indicate that the biochemical pathways involved in black point resistance are complex and may change change at different times after inoculation with B.sorokiniana.Thus,the candidate regions associated with black point resistance detected by BSRSeq at 14 h and 12 d were inconsistent.Black point incidence is associated with both the causal pathogen and environment[14,19,20].At Xingyang in 2020,the date for BSR-Seq sampling at 14 h and 12 d after inoculation ranged from April 30 to May 2 and from May 12 to May 14,respectively,due to different flowering times of the RILs.Compared to the sampling time at 14 h,the daily average temperature and daily sunshine duration increased by 7.6 °C and 2.1 h,respectively,while the precipitation and daily relative humidity decreased by 1.1 mm and 27.0%,respectively.Such variation in meteorological factors could result in different gene expression between the resistant and the susceptible lines.Therefore,the candidate regions detected by BSR-Seq were inconsistent between 14 h and 12 d after inoculation and need to be further studied.

4.2.Comparison of the identified QTL with those reported previously

Lehmensiek et al.[22]identified a QTL for black point resistance on chromosome 4A closely linked to SSR marker Xwmc48(98.7 Mb).Fifty-five SNP markers were associated with black point resistance [6,25] at five loci (11.8–21.9 Mb,146.6 Mb,634.8–661.3 Mb,693.3 Mb,and 712.8 Mb) based on linkage disequilibrium (LD) decay distance [6,24,45].The physical positions of markers BP-4A-d11 and BP-4A-d20 linked to QBB.hau-4A are at 47.2 and 40.4 Mb,respectively,18 Mb from loci reported in previous studies [6,22,24].Based on the LD decay distances the loci at positions greater than 11 Mb should be considered different[24].Therefore,QBB.hau-4A is likely a new QTL for resistance to black point caused by B.sorokiniana.

QBB.hau-5A was closely linked to the SSR marker Xbarc10(Table 2;Fig.2b).A QTL for black point resistance closely linked to the SSR marker Xgwm443 was previously located on chromosome 5A in the DH populations [22].QBp.caas-5AS for black point resistance in four environments was tightly linked with the SNP marker IWA3349 (110.7 Mb) [3].Another six loci for black point resistance were detected on chromosome 5A (8.2–32.9 Mb,93.9 Mb,453.4–461.6 Mb,525.0–535.8 Mb,567.9–605.7 Mb,and 663.3–698.6 Mb) [6,24,25].The physical positions of SSR marker Xgwm443 (11.31 Mb) and SNP marker Ra_c10762_1137 (8.2 Mb)were close to Xbarc10 (2.1 Mb) linked to QBB.hau-5A,indicating that QBB.hau-5A could be identical to the previously identified locus [22,25].

4.3.Potential application of molecular markers for black point resistance in wheat breeding

We analyzed the effects of QBB.hau-4A and QBB.hau-5A on black point response using the flanking markers.Among four groups with different combinations of alleles at loci QBB.hau-4A and QBB.hau-5A in the RIL population,the black point incidence of lines with the resistance alleles at both QBB.hau-4A and QBB.hau-5A was lower than any of the other lines.

The 122 lines evaluated by Li et al.[5]for black point resistance were also assayed for the presence of QTL QBB.hau-4A and QBB.hau-5A.The black point incidence of the lines with a resistance allele at locus QBB.hau-4A or QBB.hau-5A was lower than that of the lines with neither resistance allele.The black point incidence of lines with both resistance alleles was significantly lower than that of the lines with QBB.hau-5A only or lines without resistance alleles at both loci,implying that black point incidence could be effectively reduced by transferring these QTL to high-yielding wheat cultivars.RILs with both resistance alleles represent resistant germplasms and the markers linked to QBB.hau-4A and QBB.hau-5A can be used in MAS to reduce the incidence of black point in breeding programs.

CRediT authorship contribution statement

Qiaoyun Li:Conceptualization,Formal analysis,Writing–original draft.Runyu Hu:Data curation,Methodology,Software.Zhenfeng Guo:Data curation,Methodology,Software.Siyu Wang:Data curation,Methodology,Software.Chuang Gao:Data curation,Methodology,Software.Yumei Jiang:Data curation,Methodology,Software.Jianwei Tang:Investigation,Validation,Visualization.Guihong Yin:Project administration,Writing– review &editing,Funding acquisition.

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 work was supported by the National Natural Science Foundation of China (32171983) and the Special Fund for Key Agricultural Projects in Henan Province (201300110800).We thank Prof.Xianchun Xia for critical review of the manuscript.

Availability of data and material

The datasets generated and/or analyzed during the current study are available on request.

Appendix A.Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2021.09.007.