Identification of Gene-Allele System Conferring Alkali-Tolerance at Seedling Stage in Northeast China Soybean Germplasm

Abstract

Salinization of cultivated soils may result in either high salt levels or alkaline conditions, both of which stress crops and reduce performance. We sampled genotypes included in the Northeast China soybean germplasm population (NECSGP) to identify possible genes that affect tolerance to alkaline soil conditions. In this study, 361 soybean accessions collected in Northeast China were tested under 220 mM NaHCO₃:Na₂CO₃ = 9:1 (pH = 9.8) to evaluate the alkali-tolerance (ATI) at the seedling stage in Mudanjiang, Heilongjiang, China. The restricted two-stage multi-locus model genome-wide association study (RTM-GWAS) with gene-allele sequences as markers (6503 GASMs) based on simplified genome resequencing (RAD-sequencing) was accomplished. From this analysis, 132 main effect candidate genes with 359 alleles and 35 Gene × Environment genes with 103 alleles were identified, explaining 90.93% and 2.80% of the seedling alkali-tolerance phenotypic variation, respectively. Genetic variability of ATI in NECSGP was observed primarily within subpopulations, especially in ecoregion B, from which 80% of ATI-tolerant accessions were screened out. The biological functions of 132 candidate genes were classified into eight functional categories (defense response, substance transport, regulation, metabolism-related, substance synthesis, biological process, plant development, and unknown function). From the ATI gene-allele system, six key genes-alleles were identified as starting points for further study on understanding the ATI gene network.

Keywords: Northeast China soybean germplasm population (NECSGP); alkali tolerance; gene function; gene–allele matrix; gene–allele sequence marker (GASM); optimal cross design; restricted two-stage multilocus genome-wide association study (RTM-GWAS); soybean (Glycine max (L.) Merr.).

Figure 1

ATI gene–allele system identified in the NECSGP using GASM-RTM-GWAS. (a): Frequency distribution of the ATI at the seedling stage averaged over three environments in the NECSGP. (b): Manhattan (left) and Q-Q plot (right) of ATI candidate genes identified in the NECSGP. The horizontal solid red line indicates the threshold of p < 0.05, the upper colored dots are the significant candidate genes mapped in different chromosomes. (c) Phenotypic contribution of genes detected for ATI. The vertical axis indicates genetic contribution of a gene, while for the horizontal axis, the genes with a green color have an R² value ≥ 1.0% and for those with a red color R² < 1.0%. (d) Distribution of allele effect values of ATL alleles of each gene. (e) Frequency distribution of allele number per ATI gene. (f) Recombination potential prediction of ATI in the NECSGP based on the linkage model. (g) Gene ontology classifications of ATI in the NECSGP; I: Defense response; II: Material transport; III: Regulation related; IV: Metabolic process; V: Biosynthetic process; VI: Biological process; VII: Development; and VIII: unknown. (h) ATI gene–allele matrix for the NECSGP. The horizontal axis represents the varieties of each subregion which are arranged in ascending order based on their phenotypic value, while the vertical axis represents the alleles on the chromosomes. The allelic effects are represented by different colors; warm colors represent positive values and cool colors represent negative values. The darker the color, the greater the absolute value of the ATI value. (i) ATI gene–allele matrix of the top 10 tolerant and top ten sensitive varieties at the seedling stage.

Figure 2

Venn diagram of allele number shared by subregions and protein–protein interaction network of the ATI gene system in the NECSGP. Note: (a) Venn diagram of allele number shared by subregions. (b) Protein–protein interaction (PPI) networks. (i) The largest PPI network, containing 41 genes/proteins; the red and yellow nodes in the middle part with more links are hub genes. (ii) the blue nodes comprise the second largest network with 7 genes/proteins. (iii,iv) the deep pink nodes and bright green nodes, represent two small networks, each with only two genes/proteins.