Qian Zhou, Bingzhi Jiang, Yanbo Chng, Qibin Ma, Qiuju Xia, Z Jiang,Zhandong Cai, Hai Nian,

a The State Key Laboratory for Conservation and Utilization of Subtropical Agro-bioresources, South China Agricultural University, Guangzhou 510642, Guangdong, China

b The Guangdong Provincial Laboratory of Lingnan Modern Agricultural Science and Technology, South China Agricultural University, Guangzhou 510642, Guangdong, China

c The Key Laboratory of Plant Molecular Breeding of Guangdong Province, College of Agriculture, South China Agricultural University, Guangzhou 510642, Guangdong, China

d Guangdong Provincial Key Laboratory of Crops Genetics and Improvement, Crops Research Institute, Guangdong Academy of Agricultural Sciences, Guangzhou 510642,Guangdong, China

e Beijing Genomics Institute (BGI) Education Center, University of Chinese Academy of Sciences, Shenzhen 518083, Guangdong, China

Keywords:Glycine max Powdery mildew Adult-plant resistance Gene Fine mapping

A B S T R A C T Powdery mildew (PM), caused by the fungus Microsphaera diffusa, causes severe yield losses in soybean[Glycine max(L.)Merr.]under suitable environmental conditions.Identifying resistance genes and developing resistant cultivars may prevent soybean PM damage. In this study, analysis of F1, F2, and F8:11 recombinant inbred line (RIL) populations derived from the cross between Zhonghuang 24 (ZH24) and Huaxia 3 (HX3) indicated that adult-plant resistance (APR) to powdery mildew in the soybean cultivar(cv.)ZH24 was controlled by a single dominant locus.A high-density genetic linkage map of the RIL population was used for fine mapping. The APR locus in ZH24 was mapped to a 281-kb genomic region on chromosome 16. Using 283 susceptible plants of another F2 population, the candidate region was finemapped to a 32.8-kb genomic interval flanked by the markers InDel14 and Gm16_428.The interval harbored five genes,including four disease resistance(R)-like genes,according to the Williams 82.a2.v1 reference genome.Quantitative real-time PCR assays of candidate genes revealed that the expression levels of Glyma.16g214300 and Glyma.16g214500 were changed by M.diffusa infection and might be involved in disease defense.Rmd_B13 showed all-stage resistance(ASR)to PM in soybean cv.B13.An allelism test in the F2 segregating population from the cross of ZH24×B13 suggested that the APR locus Rmd_ZH24 and the ASR locus Rmd_B13 may be allelic or tightly linked.These results provide a reference marker-assisted selection in breeding programs.

1. Introduction

Soybean (Glycine maxL.), domesticated in central or northern China 6000-9000 years ago [1], is a source of plant protein and oil for humans and is consumed worldwide[2,3].Powdery mildew(PM),caused by the fungusMicrosphaera diffusa,is a main soybean disease[4-6]in south China,Brazil,and other soybean production regions and causes 30%yield losses in susceptible cultivars in Brazil [7-11].M. diffusaharms mainly leaves and stems by forming white spot-like colonies at the early stage of infection, leading to severe yield losses and low quality [12].

Grau and Laurence[13]found that the pathogen of PM showed differing pathogenicity to different soybean cultivars. Lohnes [14]reported that resistance to PM in soybean was controlled by theRmdlocus, which contains three alleles:Rmd,Rmd-c, andrmd.Rmdprovides adult-plant resistance, whereasRmd-cconfers resistance to PM for the whole life cycle of soybean plants, and plants withrmdare susceptible at all growth stages[7,15-17].According to previous reports[10,17-19],resistance to PM in soybean is controlled by a single dominant gene. Goncalves et al. [7] observed that eight F2populations were all susceptible to PM under natural conditions at the seedling stage, but the inheritance of resistance was in accordance with Mendelian heredity at the adult-plant stage.

The first reported PM resistance gene in soybean was identified by Lohnes [15]. To date, six powdery mildew resistance loci identified in six soybean cultivars have been located on chromosome 16, among which the PM resistance gene in Williams was closely linked with theRps2gene controlling resistance to phytophthora root and stem rot.Rmd_PI243540in soybean cv.PI243540 was mapped to a 10.9-cM interval delimited by the simple sequence repeat (SSR) marker Sat_224 and the singlenucleotide polymorphism (SNP) marker BARC-02187504228 [17].Rmd_PI567301Bin soybean cv. PI567301B was mapped between the SSR markers BARCSOYSSR_16_1272 and BARCSOYSSR_16_1298 within 1.4 cM [18].Rmd_V97-3000in soybean cv.V97-3000 was mapped to a 7.7-cM region flanked by SSR markers Satt547 and Sat_396[19].Rmd_B3 was located within 11.7 cM between GMES6959 and Satt_393 [10]. Using a high-density genetic linkage map,Rmd_B13in soybean cv. B13 was mapped to a 188.06-kb region that contained 17 disease resistance genes[10].Although all the reported resistance loci to PM were mapped at the end of chromosome 16,it was unclear whether the reported resistance to PM acted at all developmental stages or only at the adult-plant stage. The relationships between the reported genes remain to be determined.

Plant resistance may be divided into all-stage resistance (ASR)and adult-plant resistance (APR) according to the growth stage of resistance expression[20].APR refers to the phenomenon in which plants show resistance to infection, growth, and reproduction of pathogens at the adult stage but are susceptible at the seedling stage [21]. Examples are wheat APR to stripe rust (caused byPuccinia striiformis), rice APR to white leaf blight (caused byXanthomonas oryzae), soybean APR to soybean blight (caused byPhytophthora capsici), andArabidopsisAPR to Pseudomonas syringae [22-25]. Durable disease resistance is indispensable for cultivars grown in the same area over time, especially in suitable environments for disease occurrence. Many wheat cultivars with durable resistance show APR [26]. Scientists and breeding institutions pay attention to APR and try to use it in breeding [27,28].

The cv. ZH24 is susceptible to PM at the seedling stage and resistant at the adult-plant stage under suitable conditions. The objectives of the present study were to characterize the inheritance of the PM resistance gene, to identify candidate genes in ZH24 by fine mapping, and to test the allelism of PM resistance genes in ZH24 and a resistant cv. B13.

2. Materials and methods

2.1. Plant materials

ZH24 shows APR to powdery mildew.HX3 is highly susceptible to PM.B13 shows resistance at all plant growth stages[10].All soybean cultivars were provided by the Guangdong Subcentre of the National Center for Soybean Improvement, South China Agricultural University.

Fifteen F1plants, 1170 F2plants, and 166 F8:11recombinant inbred lines (RILs) were developed from the cross of ZH24 and HX3.Another segregating F2population of 204 plants derived from the cross of ZH24 and B13 was used to test the allelism of PM resistance genes in ZH24 and B13 following Lu et al. [29].The RIL population was developed by a modified single-seed method [30,31].These populations were used for disease evaluation and genetic analysis.

2.2. Evaluation of genetic materials for PM resistance

Fifteen F1plants from the cross of ZH24 and HX3 as well as the parent plants were planted on a farm in Guangzhou, Guangdong province, China on December 10, 2019. A total of 1170 F2plants from the cross of ZH24 and HX3 and the parents were planted in plastic pots 30 cm deep and 20 cm in diameter and placed in the greenhouse in Guangzhou on December 20, 2020.Each of the parents was planted among the F2plants as resistant and susceptible controls.

Field tests of artificial PM infection were performed using the 166 F8:11RILs of the ZH24×HX3 population at the farm in Guangzhou. The RILs were planted on January 7 and February 4 of 2016 and February 4 and March 13 of 2017,together with both parents,following a randomized complete block planting scheme with three replications. For each RIL and the parents, 25 plants per row, with 50 cm between rows and 10 cm between plants, were used. The 204 F2plants from the cross of ZH24 and B13 and their parents were planted in pots 30 cm deep and 20 cm in diameter in the greenhouse in Guangzhou in January 2020. The design was a randomized complete block planting scheme with three replications.

All plants were challenged withM. diffusaas described previously [10]. AM. diffusaspore suspension containing 1 × 105cfu mL-1was prepared by washing spores from the leaves of susceptible plants planted in the greenhouse with pure water and spraying them onto each plant of F1, F2, RILs and the parent plants at soybean stage V1(first trifoliolate).The disease reactions of each plant were recorded at 15 and 30 days after inoculation and postflowering[17,20,32].Plants were scored as susceptible(S)if white PM colonies were present on the leaves,similar to those of the susceptible parent, and resistant (R or APR) if no white PM colonies were visible on any leaf, similar to the resistant parent. Each RIL line with all resistant plants was designated homozygous resistant(R), and a family with all susceptible plants was considered homozygous susceptible (S).

The chi-square test was used for genetic analysis of soybean powdery mildew resistance. The segregation patterns of phenotypes in the mapping population were tested for goodness of fit to Mendelian segregation [33].

2.3. Preliminary mapping of PM resistance gene(s) in ZH24

A linkage map was constructed by the Guangdong Subcentre of the National Center for Soybean Improvement, South China Agricultural University using the F8RILs of the cross between ZH24 and HX3 [31]. Using sequencing data of RILs and their parents,all polymorphic SNP sites in the RILs were integrated into recombination bin units. The map contained 2886 bin markers and was 3174 cM in length, and the mean distance between adjacent bin markers ranged from 0.86 to 1.43 cM among chromosomes.WinQTLCart 2.5 (North Carolina State University, Raleigh, NC,USA)was used to map QTL for APR to PM.Composite interval mapping (CIM) was performed to identify resistance QTL in the whole genome and estimate the additive effect and phenotypic variation explained by each QTL. The LOD thresholds for QTL significance were determined by permutation testing (1000 cycles) with a genome-wide 5% level of significance, and the location of a QTL was described according to its LOD peak location and the surrounding region within a 95%confidence interval[10].The position on the genetic map of the resistance locus was determined from the phenotypes and genotypes of recombinants.

2.4. Genomic DNA extraction and DNA marker development and selection

The cetyltrimethyl ammonium bromide (CTAB) method was used to extract genomic DNA from young unfolded trifoliolate leaves of F2plants at stage V2 (second trifoliolate), and DNA was stored at -20 °C. After potential gene locations were identified,resistance gene(s) were located by partial genome mapping with recessive-class analysis (RCA) [34].

First,18 SSR markers in the potential gene region in the soybean reference genome Williams 82.a2.v1 from the SoyBase (https://www.soybase.org/) soybean SSR database and 12 SSR markers linked to the PM resistance locus in previous studies [10,17-19]were selected. Four polymorphic SSR markers (Satt547, Gm502,BARCSOYSSR_16_1288, and BARCSOYSSR_16_1295) were further genotyped in the 283 susceptible F2plants derived from ZH24 and HX3.

Second, the sequence information of the reference genome in the same genomic region was used to develop 20 new SSR markers.Using resequencing information in the mapping region of ZH24 and HX3, 10 new InDel molecular markers were developed and their polymorphisms were identified.By bulk-segregation analysis,seven polymorphic markers (BARCSSR-02, InDel68, InDel14,Gm16_428, BARCSSR-07, BARCSOYSSR_16_1291, and Gm683)were selected from the 30 newly developed markers to identify PM resistance genes.DNA samples of 10 resistant and 10 susceptible RILs were mixed in equal concentrations to construct a resistant bulk (R bulk) and susceptible bulk (S bulk) using the phenotypic data of 166 RILs (ZH24 × HX3). Primers for fine mapping were designed using Primer-BLAST at NCBI (https://www.ncbi.nlm.nih.gov/tools/primer-blast/) (Table S1).

2.5.Fine mapping of the PM resistance gene(s)in ZH24 based on DNA markers

Further polymorphic markers were identified by RCA in the 283-plant F2population. The flanking markers InDel14 and Gm16_428 linked to the APR locusRmd_ZH24were identified in the 283 susceptible F2plants and the 166 RILs to confirm the association between the markers and the resistance gene, using PCR and genotyping by the SSR marker method of Ou et al. [35]. The predicted genes in the target regions were matched with their annotations in the reference genome in SoyBase and the putative functions of candidate genes were identified by BLAST search against the TAIR protein database (http://www.arabidopsis.org/)[36].

2.6. Quantitative real-time polymerase chain reaction (PCR) analysis of candidate genes

To characterize the expression of candidate genes,the ZH24 and HX3 plants at stages V1 and R1 (beginning flowering)were inoculated at the same time with anM.diffusaspore suspension containing 1 × 105cfu mL-1and kept in a growth chamber at 23 °C and 75% relative humidity with a 16 h light/8 h dark photoperiod.The leaves from three replications at 0, 6, 12, 24, 48, and 72 h post-inoculation were sampled and stored at -80 °C. Total RNA was extracted using TransZol (TRANZ, Beijing, China), and 1 mg of total RNA was reverse-transcribed to first-strand cDNA using HiScript II Q select RT SuperMix with gDNA wiper (Vazyme, Nanjing, Jiangsu, China).

Candidate genes in the target region were predicted in Soybase using the Williams 82 reference genome. Quantitative real-time PCR(qPCR)was performed to obtain the expression profiles of candidate genes using primers designed with Primer Premier 5.0(Premier, Vancouver, Canada). The housekeeping geneActinwas used as a control. The specific primers for each gene are listed in Table S2.

qPCR was performed with a CFX96 Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA) using ChamQ Universal SYBR Master Mix (Vazyme, Nanjing, Jiangsu, China). All reactions were performed in 20-μL volumes containing 1 μL cDNA as a template.The thermal cycling conditions consisted of 95 °C for 3 min, followed by 40 cycles of 95 °C for 10 s, 55.0-63.3 °C (depending on the primers)for 10 s and 72°C for 30 s.Three independent biological repeats were used. The qPCR data were evaluated using the 2-ΔΔCTmethod [37].

3. Results

3.1. Genetic analysis of resistance to PM

ZH24 was susceptible to PM at the seedling stage but shows resistance at the adult-plant stage(R1),and HX3 showed a susceptible reaction to PM at all stages (Fig. 1). The reaction to PM of ZH24 changed from susceptible to resistant at growth stage V6(sixth trifoliolate), and HX3 was susceptible at all stages (Fig. S1).The disease evaluations of the RIL population from the cross of ZH24 and HX3 in multiple environments were consistent over multiple years. The 15 F1plants derived from the cross of ZH24 and HX3 were susceptible to PM at the seedling stage and resistant at the adult-plant stage (Table 1). Of the 1170 F2plants derived from the cross of ZH24 and HX3, 887 showed adult-plant resistance to PM, and 283 showed susceptible reactions, fitting a 3:1 resistance: susceptibility segregation ratio (χ2= 0.41,P= 0.55) of a single-dominant-gene trait (Table 1). Among the 166 RILs screened in the field, the phenotypic reaction was the same between years and sites:93 RILs were resistant and 73 susceptible,fitting the ratio of 1:1 (χ2= 2.41,P= 0.12) expected for the segregation of a single gene (Table 1). These results suggested that the APR of ZH24 was controlled by a single dominant gene,temporarily designatedRmd_ZH24.

3.2. Preliminary mapping of Rmd_ZH24

Based on the high-density map constructed with bins as markers and using CIM for PM resistance gene localization, a single PM resistance locus was detected on chromosome 16 (Fig. 2), with an LOD value of 25.5, explaining 32.1% of the phenotypic variance. In the high-density linkage map,Rmd_ZH24was placed in bin133-bin135 on chromosome 16 in two recombinants (Figs. 2, 3).Rmd_ZH24was located in a 281-kb region between 37,127,628 and 37,408,683 bp in the reference genome Williams 82.a2.v1(Fig. 3).

3.3. Fine mapping for delimitation of the candidate genomic region

To narrow the region coveringRmd_ZH24, 283 susceptible plants of the F2population were used to map the targeted locus further. Four of the 30 SSR markers in the delimited target region,BARCSOYSSR_16_1288, BARCSOYSSR_16_1295, Gm502, and Satt547,were polymorphic between the parents ZH24 and HX3.These markers were amplified in plants of the F2population of ZH24 and HX3 to identify linkage between the phenotype and genotype.Rmd_ZH24was linked to BARCSOYSSR_16_1288, BARCSOYSSR_16_1295, and Gm502 and located between the first two markers (Fig. 4).

Then, 20 new markers based on the sequence data of Williams 82 in Soybase were developed and tested. The markers BARCSSR-02, Gm428, BARCSSR-07 and Gm683 showed polymorphisms between ZH24 and HX3 and between two DNA bulk pools. Ten other InDel markers were developed from the resequencing data in the target interval of ZH24 and HX3. Two InDel markers(InDel68 and InDel14) in the interval between BARCSOYSSR_16_1288 and Gm428 were identified as polymorphic in this screening. Consequently, seven polymorphic markers (five SSR and two InDel markers)were further verified in an F2mapping population from ZH24 and HX3 containing 283 susceptible plants for linkage analysis by RCA (Table S1). The genotypes and phenotypes of susceptible recombinants showed that the APR locusRmd_ZH24was located between InDel14 and Gm16_428 on chromosome 16 at genetic distances of 0.29 and 0.15 cM, respectively(Fig. 4).

Fig.1. The phenotypic reaction of the parents Zhonghuang 24(ZH24)and Huaxia 3(HX3) to M. diffusa. ZH24 was susceptible at the seedling stage but showed resistance at the adult-plant stage (R1). HX3 showed a susceptible reaction at all stages. S, seedling stage; A, adult-plant stage.

Fig.3. Genomic location of the Rmd_ZH24 locus.Rmd_ZH24 was mapped to a 281-kb genomic region between 37,127,628 and 37,408,683 bp on chromosome 16 in the reference genome Williams 82.a2.v1. Recombinant inbred lines with recombination near the Rmd_ZH24 locus are shown with blue and red bars representing homozygous genotypes from ZH24 and HX3, respectively. Lines 3 and 177 were PM-susceptible plants (S). R, adult-plant resistance; S, susceptible.

3.4. Identification of candidate genes in the delimited region

The sequences of the two markers InDel14 and Gm16_428 tightly linked to the APR locusRmd_ZH24were searched by BLASTN against the Williams 82.a2.v1 reference genome in Soybase. The physical size of the region between markers at physical positions 37,202,495 bp and 37,235,283 bp was 32.8 kb (Fig. 4).A BLAST search showed five genes (Glyma.16g214300,Glyma.16g214400,Glyma.16g214500,Glyma.16g214600, andGlyma.16g214700) annotated in the Williams 82.a2.v1 reference genome (Table 2). The putative functions of these predicted genes were annotated by BLAST searches against the TAIR protein datasets.Glyma.16g214400belongs to the Exo70 gene family and the other four genes belong to the Toll-interleukin receptor (TIR)-nucleotide-binding site (NBS)-leucine-rich repeat (LRR) resistance gene family, which contains the most common disease-resistance(R) genes in plants. These five candidate genes were suggested to be candidate genes forRmd_ZH24.

Table 1Genetic analysis of resistance gene against PM in ZH24 using crosses of ZH24 × HX3 and ZH24 × B13.

Table 2 Annotations of candidate genes in the region of Rmd-ZH24 on soybean chromosome 16.

Fig.2. Results of Rmd_ZH24 locus analysis by composite interval mapping in the F8-derived RILs from a cross of ZH24×HX3.(A)The LOD value distribution over the genome in the RIL population.Add,additive effect;Chr.,chromosome;cM,centimorgan.(B)Rmd_ZH24 was mapped at the site between bin133 and bin135 of chromosome 16 on the high-density genetic linkage map, explaining 32.1% of phenotypic variance.

Fig.4. Fine mapping of the Rmd_ZH24 locus.(a)Preliminary mapping of Rmd_ZH24.(b)Fine mapping based on nine markers and recombinants in a segregating population of 283 susceptible F2 plants from the cross of ZH24 and HX3.(c)Genetic distances between the APR gene and flanking markers and five candidate genes on Williams 82.a2.v1.

3.5. Expression profiles of candidate genes

To identify genes induced after infection withM. diffusa, the expression patterns of the five predicted genes were characterized by qPCR. Compared with the control (0 h), all the candidate genes were differentially expressed between the adult plant-resistant cv.ZH24 and the susceptible cv. HX3 at the seedling and adult-plant stages (Fig. 5).

At the seedling stage, the expression ofGlyma.16g214300in ZH24 was sharply upregulated by the pathogen and reached its highest level at 24 hpi (hours post-inoculation), and then the expression levels in parents ZH24 and HX3 were all downregulated at 48 and 72 hpi. At the adult-plant stage, the transcript levels ofGlyma.16g214300were higher in ZH24 than in HX3 except 72 h after treatment with the pathogen. Following inoculation, the expression pattern ofGlyma.16g214400at the seedling stage was similar to that ofGlyma.16g214400at the adult plant stage, and the transcript level of this gene was upregulated after inoculation.

The expression level ofGlyma.16g214500was relatively low in HX3 and did not change significantly at 6, 24, and 48 hpi in ZH24 at the seedling stage. In contrast, the expression ofGlyma.16g214500was upregulated rapidly in ZH24 at 6 and 12 hpi at the adult-plant stage and its transcript level was nearly 17-fold higher than that in HX3 at 6 hpi.

However, the transcript level ofGlyma.16g214600was significantly higher in HX3 than in ZH24 at both the seedling and adult-plant stages after treatment. A similar expression pattern was observed for the geneGlyma.16g214700, which was higher in HX3 except at 72 hpi at the seedling stage. There were no significant differences between the expression levels ofGlyma.16g214700in ZH24 and HX3 at 6, 12, 24, and 72 hpi at the adult plant stage.

Thus,Glyma.16g214300andGlyma.16g214500were upregulated byM.diffusain the adult plant-resistant cv.ZH24 and may be associated with the APR of ZH24 during the interaction between the PM pathogen and soybean.

3.6. The relationship between Rmd_ZH24 and Rmd_B13

To clarify the relationship betweenRmd_ZH24andRmd_B13,an F2population consisting of 204 plants was developed from a cross between ZH24 and B13.Among these plants,151 plants were resistant and 53 susceptible to PM at the seedling stage,showing a good fit(χ2=0.10,P=0.8)to the 3:1 ratio expected for single-dominantgene inheritance (Table 1). Resistance reactions of all F2plants were also observed at the adult plant stage,and there were no susceptible plants. These results indicate that the resistance of ZH24 to PM differed from that of B13 and thatRmd_ZH24andRmd_B13may be allelic or tightly linked genes.

4. Discussion

4.1. Comparison of adult plant resistance genes with other PM resistance genes on chromosome 16

In previous studies, resistance genes providing all-stage resistance to PM in cv. V97-3000, PI567301B, PI243540, and B13 were all mapped to the end of chromosome 16.According to the physical map Glycine_max_v2.0,Rmd_V97-3000in cv. V97-3000 was located at 34,035,391-37,631,694 bp,Rmd_PI24540in cv.PI24540 at 34,258,523-36,750,257 bp,PMD_PI567301Bin cv.PI567301B at 37,249,583-37,370,175 bp andRmd_B13in cv. B13 at 37,102,014-37,290,074 bp [10,17-19].Rmd_B3in cv. B3 was mapped between 36,221,397 and 37,631,694 bp in Williams 82.a2.v1 reference genome[10].ZH24 was determined to be an adult plant-resistant cultivar and was distinct from other reported PMresistant cultivars. In present study,Rmd_ZH24was positioned between 37,127,628 and 37,408,683 bp by the high-density genetic map, andRmd_ZH24covered the same regions asRmd_B3,Rmd_B13,Rmd_V97-3000, andPMD_PI567301Bbut not asRmd_PI24540(Fig. S2).The researchers considered that PI24540 may carry a different PM resistance gene, although it shares the same phenotype as the other four cultivars.The locations and distances between markers and loci on genetic maps constructed by different mapping populations may change for various reasons,such as insertions,deletions,translocations,or chromosomal modifications in the parents[38].The results of our study and previous studies suggest thatRmd_ZH24,Rmd_B13,Rmd_B3,Rmd_V97-3000,PMD_PI567301B,andRmd_PI24540on chromosome 16 may be allelic or tightly linked genes [10,17-19]. However, further confirmation of the relationship of those resistance genes is needed. The mapping region ofRmd_ZH24was the narrowest,and the physical distance was 32.8 kb according to the reference genome. These results will help us further understand the characteristics of APR genes.

4.2. Candidate gene analysis

Of the five annotated genes in the mapping region, four belonged to the TIR-NBS-LRR (Toll-interleukin receptornucleotide-binding site-leucine-rich repeat)family,which function in plant disease resistance (Table 2). The NBS-LRR gene family is a large family of plant disease resistance genes [39]. These genes have the same functional descriptions and are involved in the same biological and metabolic processes(Fig.S3).Adult-plant resistance to PM is probably provided by one gene or several tightly linked genes. These genes are functional candidates forRmd_ZH24. The quantitative real-time PCR results showed that the candidate geneGlyma.16g214300was upregulated in ZH24 compared with HX3.The expression level ofGlyma.16g214500increased rapidly at 6 and 12 hpi at the adult plant stage, but its expression was not induced quickly at the seedling stage. Nonsynonymous mutations and frameshift mutations may be used to assess candidate resistance genes. There were many sequence variations between parents ZH24 and HX3 on chromosome 16. The four candidate genes had 40-156 SNPs and InDels between parents ZH24 and HX3(Fig. S4). There were several nonsynonymous variants in the three exons ofGlyma.16g214300.Glyma.16g214500also had nonsynonymous variants and splice region variants in its coding sequence.These candidate genes may participate in adult plant resistance of soybean cv. ZH24 to PM.

4.3. The possible mechanism of adult plant resistance

Some plants are susceptible to disease at the seedling stage but resistant at the adult-plant stage, and the cause of APR resistance has been reported: the plant resistance gene is expressed at the adult-plant stage, and the physiological and biochemical changes of plants over the growth period may lead to changes in disease resistance [40].

According to the distinguishing mechanism of extracellular and intracellular pathogens, researchers have characterized nine resistance mechanisms of plants [41]. NLRs recognize effectors of pathogens directly or indirectly and trigger downstream resistance infection [42,43]. Mutation of genetic elements occurring in the cellular pathways of host plants results in durable resistance against a broad range of pathogens, and this resistance usually occurs at the later growing stage of plants. The barley powdery mildew resistance geneMLOand rice blast resistance genePi21suppress the plant defense response, but the recessive loss-offunctionmloandpi21confer resistance to plants [44,45]. These APR genes often confer partial resistance against a broad range of pathogens. The candidate genesGlyma.16g214300andGlyma.16g214500belong to the TIR-NBS-LRR resistance gene family,but how they are involved in the APR still needs further study.Some studies [39,46-50] have also suggested that APR is associated with the accumulation of secondary metabolites in plants induced by pathogens. Thus, the expression stage of R gene and the accumulation of secondary metabolites are probably linked to APR in plants.

5. Conclusions

An adult-plant PM resistance locus,Rmd_ZH24, was finemapped to a 32.8 kb interval in chromosome 16, containing five genes in the reference genome sequence of Williams 82.a2.v1.Glyma.16g214300andGlyma.16g214500predicted in ZH24 were induced byM. diffusaand may be involved in disease defense. An allelism test suggested thatRmd_ZH24was different from the allstage resistance geneRmd_B13,and these two genes may be alleles or tightly linked to each other. The mapping information for the PM resistance locus may be useful for marker-assisted selection in soybean breeding.

CRediT authorship contribution statement

Qian Zhou:Writing - original draft, Methodology, Investigation, Data curation.Bingzhi Jiang:Writing - review & editing,Methodology, Resources, Conceptualization.Yanbo Cheng:Resources, Supervision.Qibin Ma:Writing - review & editing,Supervision.Qiuju Xia:Software.Ze Jiang:Investigation, Data curation.Zhandong Cai:Project administration.Hai Nian:Conceptualization, Writing - review & editing, Supervision, Resources,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 (31971966), the Key-Areas Research and Development Program of Guangdong Province(2020B020220008),and the China Agriculture Research System (CARS-04-PS09).

Appendix A. Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2021.12.003.