Differentially evolved drought stress indices determine the genetic variation of Brassica napus at seedling traits by genome-wide association mapping

Abstract

Drought seriously curtails growth, physiology and productivity in rapeseed (Brassica napus). Although drought tolerance is a complex trait, efficient phenotyping and genotyping has led to the identification of novel marker-trait associations underlying drought tolerance. A diverse panel of 228 Brassica accessions was phenotyped under normal (without stress) and water-stress conditions, simulated by polyethylene glycol (PEG-6000) (15% PEG stress) at the seedling stage; stress tolerance index (STI) and stress susceptibility index (SSI) values were acquired. Genome-wide association studies (GWAS) using 201 817 high quality SNPs identified 314 marker-trait associations strongly linked with drought indices and distributed across all nineteen chromosomes in both the A and C genomes. None of these quantitative trait loci (QTL) had been previously identified by other studies. We identified 85 genes underlying these QTL (most within 100 kb of associated SNPs) which were orthologous to Arabidopsis genes known to be associated with drought tolerance. Our study provides a novel resource for breeding drought-tolerant Brassica crops.

Keywords: Brassica napus; Candidate genes; Drought tolerance; Genome-wide association studies (GWAS); Specific length amplified fragments (SLAFs).

Graphical abstract

Fig. 1

Distribution of seedling growth-related phenotypes in 228B. napus accessions. Histograms of the distribution of the different phenotypes measured in control (CK) and drought-stressed plants (PEG 15%) in 228 accessions with 3 replicates.

Fig. 2

Quantile-quantile plots from GWAS of stress susceptible index (SSI) and stress tolerance index (STI) under a general linear model (GLM) and mixed linear model (MLM). The green curve indicates the observed negative log p-values (Y-axes) of marker-trait association and the red line represents expected p-values (X-axes). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

Manhattan plots representing the SNP markers-traits associations (MLM) under stress susceptible index (SSI) and stress tolerance index (STI) indices. The X-axes indicates the nineteen chromosomes from (left to right) and the Y-axes represents -log₁₀ (p) values of the SNP marker. The blue dashed horizontal line depicts the genome-wide significance threshold. Different colors displayed each chromosome. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 4

Comparison of the proportion of SNPs among SSI and STI based on GLM and MLM. (A) Association of homologous loci on drought-related traits based on SSI index among GLM and MLM. (B) Association of homologous loci based on the STI index among GLM and MLM. Salmon color bars represent SSI index while saddle brown color bars represent STI index among GLM and MLM models. SNP markers distributed in the genome are depicted in different colors bars. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 5

Percentage of superior and inferior alleles for four seedling traits related to drought stress under SSI and STI indices. The green bars represent the superior alleles, while red bars represent inferior alleles in each aspect of GWAS analysis; (A) SSI_GLM, (B), SSI_MLM, (C) STI_GLM, (D) STI_MLM. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 6

Distribution pattern of candidate genes and their corresponding SNPs on the chromosomes associated with drought tolerance. The abbreviations of orthologous genes in Arabidopsis thaliana are shown in brackets after the candidate genes. SNPs are marked in red. Numbers represent the relative distances in the genome, 1 = 1 kb.