FHB resistance distribution in Chinese wheat cultivars
Investigation of FHB in 406 wheat cultivars indicated that the mean disease index (DI) of and II was 2.47 (±0.52) and 2.33 (±0.59) (Fig. 1a), and more than 95% wheat cultivars showed FHB DI of more than 2 (Fig. 1b). In Panel I, 0.5%, 37%, 57% and 5.5% of the accessions showed FHB DI of 0 to 1, 1 to 2, 2 to 3, and 3 to 4, respectively. In Panel II, 4%, 45%, 45%, and 6% of them showed FHB DI of 0 to 1, 1 to 2, 2 to 3, and 3 to 4, respectively. The correlation coefficients between the two replicates in Panel I and Panel II are 0.73 and 0.68, respectively. Results showed that cultivars with FHB DI of 1 to 3 were predominant, and there were small number of accessions with good FHB resistance. For example, 8 cultivars (Shengyuan 928, Jimai 36, Jimai 32, Youmai 2, Yunong 201, Changnong 339-5-1, Dan 98-4025, and Beijing 841) showed FHB DI of 1 in multiple replications, and they could be used as FHB-resistant germplasms in wheat breeding.
Genome-wide association study of wheat FHB resistance
After filtering, a total of 395,782 SNPs in Panel I and 41,561 SNPs in Panel II were used for GWAS analysis (Fig. 2). A total of 321 significant SNPs with PVE of 2.71% to 13.32% were detected and they mainly distributed on chromosomes 2B, 5A, and 7A (Table S2).
In Panel I, 204 SNPs with PVE of 2.71% to 9.76% were significant in two replications. Among them, 55, 24, and 49 significant SNPs mainly distributed on 2B (9,740,162 to 68,200,954 bp), 5A (329,828,930 to 595,372,995 bp), and 7A (515,126,041 to 537,557,064 bp) (Fig. 2a; Fig. 3a; Table S1). In Panel II, 117 SNPs were significant in two replications with the PVE of 5.74% to 13.32%, and 36 of them mainly distributed on 5A (329,828,930 to 595,372,995 bp) (Fig. 2; Fig. 3b; Table S1). Combination of Panels I and II showed that there was an important genetic locus on 5A named as gFHB-5A (329,828,930 to 595,372,995 bp) for FHB resistance in two panels, and 60 significant SNPs located in this interval (Table 1; Table S1). Furthermore, gFHB-2B (9,740,162 to 68,200,954 bp) and gFHB-7A (515,126,041 to 537,557,064 bp) were also the important FHB resistance loci in Panel I since there were 55 and 49 significant SNPs in intervals of gFHB-2B and gFHB-7A, respectively (Fig. 2a; Table S1). In addition, some other FHB resistance loci on 3A (1,326,004 to 12,813,934 bp), 3D (2,597,023 to 92,375,516), 5D (426,001,273 to 544,697,840 bp), and 7B (675,149,692 to 714,012,248) were also detected in two panels.
Haplotype analysis of gFHB-5A in Chinese cultivars
Haplotype analysis (Fig. 4a, Table 2) in Panel I showed that 21 significant SNPs on 5A (577,926,236 to 588,752,395 bp) developed a 10.83-Mb block (PI_Hap_5A1) with two haplotypes (Table 2). Phenotypic comparison indicated that cultivars with PI_Hap_5A1a (2.37 to 2.4) showed a stronger FHB resistance than those with PI_Hap_5A1b (2.75 to 3.45) (Fig. 4c). In Panel II, 17 significant SNPs on 5A (331,401,559 to 422,077,781 bp) formed a 90.68-Mb block (PII_Hap_5A1) and 7 significant SNPs (591,155,558 to 595,372,995 bp) on 5A formed a 4.22-Mb block (PII_Hap_5A2) (Fig. 4b). PII_Hap_5A1 contained two haplotypes (Table 2). Phenotypic comparison indicated that cultivars with PII_Hap_5A1a (2 to 2.24) showed a stronger FHB resistance than those with PII_Hap_5A1b (2.55 to 2.64) (Fig. 4d). PII_Hap_5A2 contained two haplotypes, and phenotypic comparison indicated that cultivars with PII_Hap_5A2a (2.06 to 2.1) showed a stronger FHB resistance than those with PII_Hap_5A2b (2.5 to 2.52) (Fig. 4e). Therefore, cultivars with PI_Hap_5Aa, PII_Hap_5A1a, and PII_Hap_5A2a showed relatively stronger resistance to FHB. Pyramiding analysis indicated that cultivars with PII_Hap_5A1a/PII_Hap_5A2a (1.7 to 2) showed significantly stronger FHB disease resistance than those with PII_Hap_5A1b/PII_Hap_5A2b (2.8 to 2.9) in Panel II (Fig. 4f, Table 2). PII_Hap_5A1a, PII_Hap_5A1a, and PII_Hap_5A2a could be used for marker-assisted selection in FHB resistance breeding.
Resistance alleles at the gFHB-5A locus were becoming popular in China
All cultivars from Panel I were developed after 2014, and 67.5% of them possessed FHB-resistant allele PI_Hap_5A1a, indicating that it has been widely used in the Huanghuai valley (Fig. 5a). Based on released years of cultivars, Panel II could be divided into three periods, i.e. before 1980s, 1980s to 2000s, and after 2000s. Percentages of FHB-resistant allele PII_Hap_5A1a are 11.1%, 11.4%, and 14.8% in the cultivars of before 1980s, 1980s to 2000s, and after 2000s, respectively (Fig. 5b). Moreover, percentages of FHB-resistant PII_Hap_5A2a are 11.1%, 16.5%, and 22.2% in the cultivars of before 1980s, 1980s to 2000s, and after 2000s, respectively (Fig. 5b). It suggested that FHB-resistant alleles at the gFHB-5A locus were becoming prevalent with time in Chinese cultivars.
Linkage mapping for FHB resistance in UP population
Linkage analysis in the RIL population UP showed that a QTL (FHB.hau-4D) on chromosome 4D was detected for FHB resistance and it was flanked by markers wPt-8836 and wPt-4572 with PVE of 8.89%. Wheat lines with FHB.hau-4D_AA (2.37) showed a stronger FHB resistance than those with FHB.hau-4D_BB (2.73) (Fig. 6). Analysis of physical position in the Chinese Spring genome database indicated that FHB.hau-4D was mapped to be around 9431967-bp position on 4D.