Genome-Wide Association Analysis and Molecular Marker Development for Resistance to Fusarium equiseti in Soybean

Abstract

Fusarium root rot, caused by Fusarium equiseti, poses a significant threat to soybean production. This study aimed to explore the genetic basis of resistance to Fusarium equiseti root rot (FERR) by evaluating the resistance phenotype of 346 soybean germplasms and conducting a genome-wide association study (GWAS) using 698,949 SNP markers obtained from soybean germplasm resequencing data. GWAS analysis identified 101 SNPs significantly associated with FERR resistance, distributed across nine chromosomes, with the highest number of SNPs on chromosomes 13 and 20. Further gene-based association and allele variation analyses identified candidate genes whose mutations are closely related to FERR resistance. To accelerate soybean FERR resistance breeding screening, we developed CAPS markers S13_14464319-CAPS1 and S15_9215524-CAPS2, targeting these SNP sites, and KASP markers based on the S15_9205620-G/A, providing an effective tool for marker-assisted selection (MAS). This study offers a valuable theoretical foundation and molecular marker resources for the functional validation of FERR resistance genes and soybean disease resistance breeding.

1. Introduction

Soybean root rot is a globally distributed and highly destructive disease that poses significant economic losses to soybean production worldwide [1]. The pathogen exhibits strong persistence and latency in the field, capable of surviving for extended periods in the rhizosphere soil and reactivating under favorable environmental conditions to infect plants. This latent nature makes the disease particularly difficult to manage. Once an outbreak occurs, it not only compromises the yield of the current growing season but also deteriorates soil health, potentially causing long-term negative effects on subsequent crop cultivation [2].

Fusarium species are important pathogens responsible for root rot in various crops worldwide, including F. equiseti, F. oxysporum, F. graminearum, F. solani, and F. asiaticum, which exhibit regional variation in both prevalence and pathogenicity [1,3,4]. F. equiseti has recently gained attention as an emerging and important pathogen causing root rot in soybean production. It is widely distributed in soil and demonstrates strong environmental adaptability, enabling it to survive and infect under diverse ecological conditions. Infection by F. equiseti leads to discoloration and necrosis of soybean roots, ultimately resulting in stunted plant growth and significant yield losses [5]. According to previous reports, F. equiseti has been identified as a causal agent of root rot in multiple host plants, including soybean [6], goji berry (Lycium barbarum) [7], cowpea [8], and chickpea (Cicer arietinum) [8], indicating its broad host range and pathogenic potential across diverse crop species. Apart from root rot, F. equiseti is also capable of causing various plant diseases, including stem rot, fruit rot, wilt, and leaf spot [9,10,11]. Its broad pathogenic spectrum poses a potential threat to multiple crop species, underscoring the need for comprehensive disease management strategies. In soybean production, fungicides are commonly used to prevent and control Fusarium infections. However, this practice can lead to a reduction in soil microbial diversity, which is detrimental to sustainable agricultural development [12]. Breeding Fusarium-resistant soybean cultivars offers a sustainable long-term solution, yet its effectiveness hinges on detailed knowledge of pathogen distribution and virulence diversity.

Genome-wide association study (GWAS) has become a powerful approach for dissecting the genetic architecture of complex quantitative traits and identifying key functional genes in crops [13]. Moreover, the development of reliable molecular markers based on GWAS-identified loci is essential for enabling marker-assisted selection and precision breeding [14]. In soybeans, numerous studies have successfully employed GWAS to identify SNPs associated with stress tolerance and disease resistance, demonstrating its effectiveness in genetic improvement efforts. A GWAS conducted on 460 soybean accessions identified 99 SNPs associated with Cercospora leaf blight (CLB) severity and 85 SNPs with disease incidence [15]. Similarly, salt tolerance was evaluated in 563 soybean germplasm accessions from over 20 countries, and genome-wide association analysis identified 10 SNPs and 11 candidate genes significantly associated with salt tolerance [16]. In soybean root rot research, a genome-wide association study of 350 accessions identified eight SNPs significantly associated with Fusarium oxysporum root rot (FORR) resistance, primarily on chromosome 6, and highlighted Glyma.06G088400 as a candidate gene [17].

Cleaved Amplified Polymorphic Sequence (CAPS) and Kompetitive Allele-Specific PCR (KASP) are two efficient types of molecular markers, widely used in gene mapping, germplasm evaluation, and molecular breeding due to their high specificity and ease of operation. Developing CAPS or KASP markers based on GWAS-identified loci enables rapid and high-throughput detection of target genes, thereby improving the efficiency of genetic improvement and offering significant practical value in crop breeding programs. Recently, researchers have reported molecular markers related to resistance to soybean powdery mildew, soybean root rot, and soybean mosaic virus [18,19,20]. A GWAS of 264 Chinese accessions under PEG-induced drought stress identified ninety-two SNPs and nine candidate genes, and developed two linked KASP markers [21]. Similarly, GWAS analysis of 356 soybean germplasm materials identified 150 SNPs associated with FORR resistance in 41 QTLs, and three functional molecular markers were developed based on candidate genes and favorable haplotypes, including two CAPS molecular markers and one KASP molecular marker [19]. Li et al. [20] used NGS-based BSA technology to locate the dominant resistance gene of Heinong 84 to soybean mosaic virus N3 type (RSMV-N3) on chromosome 13 and developed two pairs of CAPS markers related to RSMV-N3. Although researchers have made considerable progress in developing molecular markers for resistance to powdery mildew, root rot, and mosaic virus in soybean research, research specifically targeting soybean FERR resistance remains limited.

Given the serious threat of soybean FERR caused by F. equiseti to soybean production, developing disease-resistant varieties has become an effective way to address this problem. Although several SNP loci associated with soybean disease resistance have been identified through GWAS, genetic research on soybean FERR resistance is still limited. Therefore, we evaluated FERR resistance in 346 soybean germplasms and used GWAS to analyze SNP loci associated with FERR resistance and then developed CAPS and KASP markers. The results provide new perspectives and tools for soybean disease resistance breeding.

2. Materials and Methods

2.1. Plant Materials

A total of 346 soybean (Glycine max) germplasm accessions, obtained from the Soybean Research Institute of Northeast Agricultural University (NEAU), Harbin, China, were used in this study. The majority of these accessions originated from China, while others were introduced from the United States, Italy, Ukraine, and Japan. All experiments in this study were conducted in greenhouse facilities in November 2023 at the Soybean Research Institute of NEAU, Harbin, China.

2.2. Culture and Inoculation of Fusarium equiseti

The F. equiseti isolate used in this study, designated as strain FE17, was also provided by the Soybean Research Institute of NEAU. This isolate was originally recovered from the root tissue of a diseased soybean plant collected from a field in Harbin, Heilongjiang Province, China, and was subsequently identified based on morphological and molecular characteristics. For long-term storage, the fungal isolate was maintained in the dark at 4 °C on potato dextrose agar (PDA) medium.

F. equiseti was propagated using PDA medium. The fungus block was inoculated in the center of the PDA and incubated at 25 °C in the dark for 7 days. Sorghum seeds (Sorghum bicolor (L.) Moench) were washed and soaked with distilled water and then placed in a 150 mL Erlenmeyer flask and autoclaved. The mycelium blocks after 7-day PDA culture were inoculated into sterilized sorghum seeds and cultured in the dark for 2 weeks. The flask was shaken every 24 h to promote uniform growth of F. equiseti.

Sorghum grains infected with F. equiseti were crushed with a grinder and thoroughly mixed with sterilized vermiculite at a ratio of 1:50 (volume ratio) and placed in 250 mL seedling boxes for use as inoculation substrate. Four soybean seeds were sown in each box. Five replicates were set for each variety. When the soybean seedlings emerged from the soil, the seedlings were thinned to 3 plants per box. The experiment was conducted using a randomized block design. The experiment was conducted under greenhouse conditions with a 12 h light period at 25 °C and a 12 h dark period at 18 °C.

2.3. Evaluation of Soybean Resistance to Fusarium equiseti

The criteria for soybean identification and evaluation used in this study were adapted and modified from the method described by Chang et al. [3]. The soybean root system 15 days after inoculation was used as the evaluation object to assess the resistance of soybean varieties to FERR. Based on the symptoms and lesion development of F. equiseti infection, a disease severity scale of 0 to 7 was adopted, where 0 = no visible symptoms; 1 = slightly browned taproot, healthy lateral roots, normal plant growth; 3 = taproot mostly browned, brown spots on lateral roots, slightly inhibited aboveground; 5 = completely browned taproot, obvious brown spots on lateral roots, restricted growth of aboveground; 7 = taproot dark brown and fractured, blackened lateral roots, plant death or failure to form seedlings. Fifteen plants of each variety were randomly selected for a disease grade survey, and their disease severity index (DSI) was calculated accordingly. Three independent replicates of the experiment were conducted to ensure the reliability of the results. The DSI of each soybean accession was calculated using formula (1). Based on the DSI value of each variety’s resistance to F. equiseti, the resistance of soybean germplasm resources to F. equiseti was divided into five levels: high resistance (HR, 0 < DSI ≤ 10), moderate resistance (MR, 10 < DSI ≤ 20), moderately susceptible (MS, 20 < DSI ≤ 30), susceptible (S, 30 < DSI ≤ 60), and highly susceptible (HS, 60 < DSI < 100). The phenotypic data were analyzed using IBM SPSS Statistics 26.0 and Microsoft Excel 2021. To assess the heritable portion of the trait, the broad-sense heritability (H²) was calculated using formula (2) [22]. In the formulas (2), ${\mathsf{\sigma }}_{\mathrm{G}}^2$ and ${\mathsf{\sigma }}_{\mathrm{P}}^2$ represent the genetic variance and phenotypic variance, respectively. Additionally, MSG, MSE, and rep correspond to the mean square of genotype, mean square error, and the number of replications, respectively.

$$\mathrm{D}\mathrm{S}\mathrm{I}={\displaystyle \frac{\sum (\mathrm{R}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{v}\mathrm{a}\mathrm{l}\mathrm{u}\mathrm{e}\times \mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r} \mathrm{o}\mathrm{f} \mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{t}\mathrm{s} \mathrm{w}\mathrm{i}\mathrm{t}\mathrm{h} \mathrm{t}\mathrm{h}\mathrm{i}\mathrm{s} \mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{g})}{\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{n}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r} \mathrm{o}\mathrm{f} \mathrm{p}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{t}\mathrm{s} \mathrm{s}\mathrm{u}\mathrm{r}\mathrm{v}\mathrm{e}\mathrm{y}\mathrm{e}\mathrm{d}\times 7}}\times 100\mathrm{\%}$$

(1)

$${\mathrm{H}}^2={\displaystyle \frac{{\mathsf{\sigma }}_{\mathrm{G}}^2}{{\mathsf{\sigma }}_{\mathrm{P}}^2}},{\mathsf{\sigma }}_{\mathrm{G}}^2=\left({\displaystyle \frac{\mathrm{M}\mathrm{S}\mathrm{G}-\mathrm{M}\mathrm{S}\mathrm{E}}{\mathrm{r}\mathrm{e}\mathrm{p}}}\right),{\mathsf{\sigma }}_{\mathrm{P}}^2=\left({\displaystyle \frac{\mathrm{M}\mathrm{S}\mathrm{G}-\mathrm{M}\mathrm{S}\mathrm{E}}{\mathrm{r}\mathrm{e}\mathrm{p}}}\right)+\mathrm{M}\mathrm{S}\mathrm{E},{\mathrm{H}}^2={\displaystyle \frac{\left(\frac{\mathrm{M}\mathrm{S}\mathrm{G}-\mathrm{M}\mathrm{S}\mathrm{E}}{\mathrm{r}\mathrm{e}\mathrm{p}}\right)}{\left(\frac{\mathrm{M}\mathrm{S}\mathrm{G}-\mathrm{M}\mathrm{S}\mathrm{E}}{\mathrm{r}\mathrm{e}\mathrm{p}}\right)+\mathrm{M}\mathrm{S}\mathrm{E}}}$$

(2)

2.4. Genotyping

Genomic DNA was isolated from the young leaves of 346 soybean germplasms using the CTAB (cetyltrimethylammonium bromide) extraction method [23]. DNA quality and concentration were evaluated using a NanoDrop™ One spectrophotometer (Thermo Fisher Scientific, Inc., Waltham, MA, USA). Samples exhibiting an OD260/OD280 ratio between 1.8 and 2.0 were deemed suitable for genotyping. After SNPs with MAF < 5% and missing rate > 10% were removed, 698,949 high-quality loci were retained for the GWAS. SNPs were scattered across all 20 chromosomes of soybean (Supplementary Table S1).

2.5. Population Structure Evaluation and Linkage Disequilibrium Analysis

The population structure of the germplasm accessions was assessed through principal component analysis (PCA), performed using the GAPIT version 3.0 software package [24]. Linkage disequilibrium (LD) between SNP pairs was calculated as squared allele frequency correlation (r²) based on SNPs with MAF ≥ 5% and missingness ≤ 10% used by TASSEL version 5.0 [25]. Unlike GWAS analyses, missing genotypes were not imputed before LD analysis. SNPs with MAF ≥ 0.05 and SNP completeness ≥ 80% were used in the analysis.

2.6. Genome-Wide Association Study

The GWAS analysis was conducted using the GAPIT 3.0 package in conjunction with R 4.4.0 and RStudio (version 2024.04.0+735) software [24]. To reduce false positives in GWAS, Mixed Linear Model (MLM) was employed [26]. The model accounts for both fixed and random effects, enhancing the accuracy and reliability of association analysis. A significance threshold of −log₁₀(p) ≥ 4.0 was used to identify SNP loci significantly associated with FERR [27]. SNPs exceeding this threshold were considered significant sites associated with FERR for subsequent candidate gene mining and functional annotation analysis.

2.7. Identification of Candidate Genes and RT-qPCR Assay

Candidate genes within a 200 kb genomic interval (100 kb upstream and downstream) of each significant QTN were identified and annotated based on the soybean reference genome Wm82.a2.v1 [24]. SNP variation analysis was conducted on candidate genes from 30 FERR-resistant varieties (DSI 4.40–14.29) and 30 FERR-sensitive varieties (DSI 36.51–73.33) using genome resequencing data (Supplementary Table S3). The analysis covered exon regions, 5′ UTR, 3′ UTR, introns, and upstream and downstream regulatory sequences. The general linear model (GLM) method in TASSEL version 5.0 [28] software was used to carry out gene-level association analysis to identify significant SNPs or haplotypes associated with FERR resistance. The significant difference between the allelic variation SNP sites and FERR resistance was evaluated by t-test.

Soybean seeds were inoculated with F. equiseti after 4 days of culture in vermiculite, and samples were collected from the taproot 72 h after inoculation. As a control treatment, sterilized sorghum grains were crushed and combined with sterilized vermiculite. Total RNA was extracted with the Easy Pure Plant RNA Kit (Quanshijin, Beijing, China) according to the manufacturer’s instructions. Subsequently, the RNA was reverse transcribed into cDNA using the PrimeScript™ RT Kit (with gDNA Eraser, Takara Bio Inc., Kusatsu, Shiga, Japan). RT-qPCR primers were designed using the Primer-BLAST tool available on the NCBI website (Supplementary Table S2). Transcript levels were normalized using the 2^−ΔΔCt method, with actin serving as the housekeeping gene for comparison. RT-qPCR was conducted with the One-Step Real-Time RT-PCR Kit (Toyobo, Osaka, Japan). The reaction system was set up as follows: 94 °C pre-denaturation for 5 min, followed by 45 cycles (94 °C denaturation for 30 s, 60 °C annealing for 30 s, and 72 °C extension for 40 s). Melting curve analysis was performed from 55 °C to 100 °C, followed by cooling at 72 °C for 10 min. Three biological replicates and three technical replicates were set for each sample.

2.8. Development of CAPS and KASP Molecular Markers

The genome sequences of Glyma.13G048800, Glyma.15G117200, and Glyma.15G117100 were downloaded from the Phytozome (https://phytozome-next.jgi.doe.gov, accessed on 10 July 2024.) database, and the significant SNP sites S13_14464319 and S15_9215524 were used as target sites. DNA from 10 FERR-resistant and 10 FERR-sensitive varieties were extracted using the 2 × CTAB method [23]. CAPS primers were designed on both sides of the single nucleotide polymorphism site using Primer Premier v5.0 with default settings (

Table 1. Primers used in the present study.

Primer Names	Sequences(5′-3′)	Usage
S13_14464319-CAPS1-F	TTGATTGTAAGCAATTCAGGTCT	CAPS assay
S13_14464319-CAPS1-R	CTCATTTGTCTTTCAGTTGTTGG	CAPS assay
S15_9215524-CAPS2-F	CTAGTTGTTGCAAGTGGTGTGG	CAPS assay
S15_9215524-CAPS2-R	CGGCCCTCTGAAATCAAGATC	CAPS assay
S15_9205620-F1	GAAGGTGACCAAGTTCATGCTCTGCAAATACTTGCCGGCACTGTCG	KASP assay
S15_9205620-F2	GAAGGTGACCAAGTTCATGCTCTGCAAATACTTGCCGGCACTGTCA	KASP assay
S15_9205620-R	CATCTTGAACAATTGATGAGCATCAGATT	KASP assay

). The purified DNA fragments of each variety were digested with restriction endonucleases PciI and AflII. Polymorphism was identified by resolving the products on a 2% agarose gel.

For KASP molecular markers, primers targeting the SNP site S15_9205620-G/A significantly associated with FERR were designed using the Primer-BLAST tool on the NCBI website (https://www.ncbi.nlm.nih.gov/, accessed on 23 July 2024.). The primers were designed as two specific forward primers (F₁ and F₂) and a universal reverse primer (R). The F₁ and F₂ primers contained 6-carboxyfluorescein (FAM) and hexachloro-6-methylfluorescein (HEX) fluorescent linker sequences, respectively. Genomic DNA was extracted using the 2 × CTAB method and PCR amplification was performed using KASP V4.0 2 × Mastermix (LGC, Teddington, UK) on a quantitative real-time PCR system (ABI7500) according to the manufacturer’s instructions. Primers F₁, F₂, and R were used to specifically identify these KASP markers for genotyping, as described in Table 1.

3. Results

3.1. Germplasm Evaluation of Fusarium equiseti Root Rot

We inoculated soybean germplasms with Fusarium equiseti root rot (FERR) and observed a broad range of phenotypic variations in disease response, quantified using the disease severity index (DSI). The DSIs exhibit considerable variation, ranging from 4.40 to 73.33, with a mean of 25.06. The broad-sense heritability (H²) of DSI was calculated as 0.67, suggesting that genetic effects played a major role in phenotypic variation. Its continuous, unimodal distribution aligns with the anticipated pattern of a quantitative trait, indicative of polygenic regulation (Figure 1, DSI refer to Supplementary Table S3).

Natural populations of soybean exhibit substantial genetic diversity, which is crucial for the success of GWAS in uncovering the genetic architecture underlying complex traits. No completely immune accessions were found in the FERR resistance screening. Of the soybean germplasms, 176 showed a moderately susceptible phenotype, accounting for 50.9% of the total soybean germplasms, while 17 soybean germplasms showed high resistance, accounting for 4.9% of the total soybean accessions. The germplasm resources for FERR resistance in soybean are highly limited (

Table 2. Evaluation of resistance to FERR in soybean germplasms.

DSI	Type of Reaction	Soybean Germplasms	Percentage (%)
DI = 0	Immune, I	0	0.0
0 < DI ≤ 10	High Resistant, HR	17	4.9
10 < DI ≤ 20	Medium Resistant, MR	73	21.1
20 < DI ≤ 30	Medium Susceptible, MS	176	50.9
30 < DI ≤ 60	Susceptible, S	78	22.5
DI ≥ 60	High Susceptible, HS	2	0.6

).

3.2. Population Structure and LD Analysis

We obtained a total of 698,949 SNPs by genotyping 346 soybean germplasms based on resequencing data. These SNPs are located across 20 chromosomes in soybean. To ensure the accuracy of the GWAS analysis, this study used GAPIT 3 for association analysis and calculated LD decay to evaluate linkage disequilibrium of genetic variation in soybean populations. The average LD decay distance was 221 kbp, indicating rich genetic diversity among different soybean varieties (Figure 2A). Principal component analysis (PCA) of genome-wide SNP data revealed significant population stratification, with the first three principal components collectively contributing to 17.67% of the variation in the mapping population (Figure 2B,C). The soybean genetic relationship matrix indicated a low degree of relatedness among the 346 soybean germplasms evaluated (Figure 2D).

3.3. Genome-Wide Association Analysis to Identify Loci Associated with Resistance to Fusarium equiseti in Soybean

A total of 346 soybean germplasms were inoculated with F. equiseti to assess the DSI of root rot. Using 698,949 SNP markers from these accessions, a GWAS was performed employing a mixed linear model (MLM). The Manhattan plot illustrates the distribution of SNPs across various chromosomes. Genome-wide analysis identified 101 SNPs significantly associated with FERR resistance. These significant loci were randomly distributed across nine chromosomes in soybean. Among these SNP loci significantly associated with FERR, the largest numbers were located on chromosomes 13 and 20, with 23 and 50 SNPs, respectively. In contrast, no SNPs significantly associated with FERR were identified on chromosomes 1, 2, 3, 4, 8, 10, 11, 12, 14, 16, and 17 (Figure 3). These SNPs significantly associated with FERR were located on 28 QTLs (Supplementary Table S4). After excluding genes without annotation information within 200 kb upstream and downstream of these SNPs, 176 candidate genes related to FERR resistance were identified (Supplementary Table S5). GO enrichment analysis showed that these genes related to soybean FERR resistance were mainly involved in cell wall organization, ion transport, and transcriptional regulation (Supplementary Figure S1).

3.4. Gene-Based Association Analysis Reveals SNPs Associated with FERR Resistance

To further evaluate SNPs associated with resistance to FERR in candidate genes, we conducted 60 soybean germplasm resequencing data (30 lines with higher/lower FERR resistance levels) and their DSI association analysis using the GLM method. The analysis identified a total of six candidate genes, including Glyma.13G048800, Glyma.13G047600, Glyma.13G047700, Glyma.15G117200, Glyma.15G117100, and Glyma.18G076200, which contained 55 SNPs significantly associated with FERR resistance. These SNPs were found in several regions of the genes, including the promoter, exons, introns, and downstream regions (Supplementary Table S6). To investigate the potential functional effects of these candidate genes, we further analyzed the allelic variations within the exonic and upstream regions of the identified genes and their correlation with the DSI.

In the exon region of the candidate gene Glyma.13G047600, soybean germplasms carrying the S13_14299769-G allele exhibited extremely significantly higher mean DSI compared to those carrying the S13_14299769-A allele (Figure 4A). Similarly, in the exon region of Glyma.15G117100, soybean germplasms carrying the S15_9205620-G allele exhibited extremely significantly lower mean DSI compared to those carrying the S15_9205620-A allele (Figure 4D). Likewise, in the exon region of Glyma.18G076200, soybean germplasms with the S18_7214157-T allele had extremely significantly lower mean DSI compared to those carrying the S18_7214157-C allele (Figure 4F). Conversely, in the intron region of Glyma.15G117200, soybean germplasms carrying the S15_9215524-A allele had an extremely significantly higher mean DSI than those carrying the S15_9215524-G allele (Figure 4E). In the upstream region of the candidate gene Glyma.13G047700, soybean germplasms carrying the S13_14302394-C allele showed extremely significantly higher mean DSI than those carrying the S13_14302394-T allele (Figure 4B). Similarly, in the upstream region of Glyma.13G048800, soybean germplasms with the S13_14464319-A allele showed extremely significantly higher mean DSI compared to those carrying the S13_14464319-C allele (Figure 4C).

3.5. Candidate Gene Expression Analysis Under FERR Inoculation

To further investigate the potential roles of candidate genes in FERR resistance, we examined the expression levels of these genes in soybean materials with varying resistance to FERR. Specifically, the resistant material DN8004 and the sensitive material JF3 were subjected to F. equiseti inoculation. Analysis of the relative expression levels of these genes provided insight into the genetic factors underlying FERR resistance. Following inoculation with F. equiseti, samples were collected 72 h post-inoculation and subjected to RT-qPCR analysis. Significantly increased expression of Glyma.15G117100 was detected in DN8004 compared to JF3 following inoculation with F. equiseti. However, after F. equiseti inoculation, the expression levels of Glyma.13G048800 and Glyma.15G117200 in DN8004 were lower than those in JF3. In addition, there was no significant difference in the expression of Glyma.13G047600 and Glyma.18G076200 between the two materials (Figure 5).

3.6. Development of CAPS Markers Associated with FERR Resistance in Soybean

CAPS marker detection is a reliable method for identifying SNP variations and has significant applications in crop breeding. Based on the SNP variations at S13_14464319 and S15_9215524, we designed two pairs of primers to amplify DNA fragments via PCR, followed by restriction enzyme digestion, for the rapid identification of FERR resistance. To validate the SNP-based CAPS markers, 20 extreme accessions were selected, including 10 resistant and 10 susceptible germplasm resources, for the specific amplification of DNA fragments carrying haplotypes (Supplementary Figure S2). S13_14464319-CAPS1 distinguishes the A and C alleles based on PciI restriction enzyme digestion. The PciI enzyme recognizes the (A)CATGT sequence and cleaves it, while the (C)CATGT sequence is not cleaved by PciI. Digestion with PciI produced two fragments of 325 bp and 449 bp, corresponding to the homozygous AA genotype found in the ten susceptible germplasm resources. In contrast, the undigested 774 bp fragment indicated the homozygous CC genotype present in the ten resistant germplasm resources (Figure 6A). S15_9215524-CAPS2 A and G alleles were distinguished by AflII restriction endonuclease digestion. AflII recognizes and cuts the CTTA(A)G sequence, but not the CTTA(G)G sequence. The 450 bp and 203 bp digestion products of the DNA fragment after AflII restriction endonuclease digestion corresponded to the homozygous AA genotype found in susceptible germplasm resources, while the undigested 653 bp fragment represented the homozygous GG genotype found in resistant germplasm resources (Figure 6B). The susceptible germplasm was successfully digested by PciI and AflII and produced the expected banding pattern. In contrast, no digestion was observed in the resistant germplasm. These results indicate that the identified SNP loci are closely associated with soybean resistance to FERR and can serve as potential markers for MAS.

3.7. KASP Marker Development for FERR Resistance in Soybean

Based on the GWAS results of 346 soybean germplasms, we developed a KASP marker targetingS15_9205620-G/A site located in the exonic region of the Glyma.15G117100. This SNP was found to be significantly associated with DSI for FERR. Notably, soybean germplasms with different genotypes exhibited significant differences in DSI. Specifically, soybean germplasm carrying the GG allele (represented by blue dots) showed significantly lower DSI, while soybean germplasm carrying the AA allele (represented by red dots) showed the opposite trend. Further statistical analysis confirmed that soybean lines carrying the AA genotype had significantly higher DSI than lines carrying the GG genotype. The clustering of different genotypes highlights the effectiveness of the KASP marker in achieving precise SNP-based genotyping at this locus. The development of this molecular marker provides a valuable scientific basis for further exploring FERR resistance in soybean (Figure 7A–D).

4. Discussion

Fusarium species are well-recognized phytopathogens implicated in a broad spectrum of destructive plant diseases worldwide, notably including soybean root rot, sudden death syndrome, wilt, and damping-off [29,30,31]. The incidence and intensity of Fusarium root rot are greatest in areas with high temperatures and moisture, and are further aggravated under conditions of increased rainfall and relative humidity [32]. Environmental factors are one of the key reasons for the differences in the main pathogens of root rot in different regions, which makes it difficult to breed soybean varieties with broad-spectrum resistance to root rot. In agricultural production, the application of quicklime and chemical control are the main means of preventing and controlling root rot, but these methods may cause soil compaction and acidification, thereby triggering soybean stress and resulting in reduced production. The pathogen of soybean root rot has strong environmental adaptability and latency. This pathogen persists in soil for extended periods and readily re-emerges under conducive conditions, thereby sustaining continuous infection cycles and posing significant challenges to disease control. Recent studies have shown that Pseudomonas chlororaphis IRHB3, Bacillus subtilis HSY21, and Bacillus velezensis BVE7 are effective in controlling soybean FORR [33,34,35]. At present, no effective control measures have been found for soybean root rot caused by F. equiseti. Therefore, this study used the GWAS method to explore soybean FERR-related candidate genes. By developing CAPS and KASP molecular markers, resistance genes and varieties can be efficiently screened to promote precision breeding.

Genome-wide association studies (GWAS) have become an effective tool for analyzing the genetic basis of complex quantitative traits [36]. The rapid development of modern molecular biology techniques, especially the widespread application of GWAS, has promoted the application of molecular marker technology in crop breeding, covering the molecular breeding research fields of wheat, rice, corn, sorghum, rapeseed, and other crops [14]. Soybean disease resistance is manifested as a variety of complex quantitative traits, which are usually regulated by multiple genes and affected by the interaction between genotype and environmental factors. In the soybean root rot GWAS-related study, Sang et al. [17] used 52,041 SNP markers to genotype 350 soybean varieties, and through genome-wide association analysis, identified eight loci significantly associated with resistance to FORR, which were mainly located on chromosome 6. Lin et al. [37] inoculated two recombinant inbred lines with Pythium sylvaticum and identified a total of 7 SNP markers significantly associated with this pathogen through GWAS analysis. These markers were mainly located on chromosomes 10, 18, and 20. Zhao et al. [38] evaluated the resistance of 225 soybean varieties and 109 recombinant inbred lines (RILs) to Phytophthora sojae race 1. Through genome-wide association analysis, they detected eight quantitative trait nucleotides significantly associated with soybean resistance to Phytophthora sojae race 1 and found co-localized resistance genes on chromosome 3. Although previous studies have explored different pathogens of soybean root rot, GWAS studies on soybean root rot caused by F. equiseti are still relatively lacking. In this study, 346 soybean resources were inoculated with F. equiseti, and the disease index was evaluated. GWAS analysis was performed based on the resequencing data of these soybean germplasms. It was found that 101 SNP loci significantly associated with FERR resistance were located in 28 QTL regions. Functional annotation and gene-based association analysis of the genes within the QTL identified six candidate genes, including Glyma.13G048800, Glyma.13G047600, Glyma.13G047700, Glyma.15G117200, Glyma.15G117100, and Glyma.18G076200. Glyma.13G04880 contains five allelic variants that are annotated as glycosyl hydrolase proteins, a class of genes found to be associated with the recognition of receptor toxins [39]. Glyma.15G117200 contains 36 allelic variants and is annotated as a protein containing an agenet domain, which is related to pathogen recognition and defense mechanisms [40]. Glyma.15G117100 contains five allelic variants and is annotated as a homolog of the yeast transcriptional activator ADA2. Studies have shown that this type of gene is highly expressed in root meristems and is essential for cell division and growth [41]. Glyma.18G076200 has three allelic variants and is annotated as a lectin receptor-like kinase. Studies in the field of disease resistance have shown that its homologs are involved in the response of rice to rice blast fungus [42]. Glyma.13G047600 contains five alleles and is annotated as a proline-rich family protein. Related studies have shown that this type of gene may be involved in the defense against Fusarium wilt [42]. Our study identified five candidate genes associated with soybean FERR resistance, providing a basis for the development of CAPS and KASP molecular markers.

Developing molecular markers using the polymorphism of SNP sites can effectively shorten the breeding cycle. CAPS molecular markers are a simple and reliable method for detecting SNP variations and are frequently used in crop breeding. CAPS molecular markers have a wide range of applications in plants, such as in the study of tomato leaf curl disease [43], cucumber powdery mildew [44], and pepper resistance to Phytophthora [45]. Similarly, related effects have been reported in soybean, including studies on soybean mosaic virus [20] and powdery mildew (PMD) [18]. The two CAPS markers, S13_14464319-CAPS1 and S15_9215524-CAPS2, developed in this study were significantly correlated with FERR resistance and could effectively distinguish between resistant and susceptible genotypes, thereby accelerating the process of disease-resistant breeding. KASP molecular markers are a method for SNP typing through highly sensitive fluorescence detection. Recently, research on accelerating breeding progress by developing KASP markers in soybean has gradually emerged. In the research of soybean pod dehiscence tolerance [46], main stem node number [47], salt stress [48] and soybean cyst fungus resistance [49], KASP molecular markers are gradually being developed and applied. In our study, we developed an FERR-related KASP molecular marker based on S15_9205620-G/A, which can effectively achieve accurate SNP genotyping and is of great value for the discovery of soybean FERR-resistant germplasm. In the future, with the continuous development of molecular marker technology, more efficient and accurate tools will be provided for soybean disease-resistant breeding, thereby accelerating the selection and promotion of disease-resistant varieties.

5. Conclusions

In this study, we evaluated the resistance of 346 soybean accessions to FERR and combined the resequencing data of these accessions to conduct GWAS, identifying a total of 101 SNPs significantly associated with FERR resistance. These SNPs are distributed on nine chromosomes, with the largest number of SNPs associated with FERR on chromosomes 13 and 20. Gene-based association analysis identified SNP variations in six candidate genes, and the allelic variations in these genes are closely related to FERR resistance. In addition, we successfully developed two CAPS markers based on these SNPs, S13_14464319-CAPS1, and S15_9215524-CAPS2. We also developed a KASP marker based on S15_9205620-G/A. These molecular markers can be used as MAS in future breeding programs to improve soybean FERR resistance. These findings have enhanced our theory of soybean FERR resistance and provided valuable molecular tools for breeding resistant varieties.