Analysis of genetic diversity and genome-wide association study for drought tolerance related traits in Iranian bread wheat

Abstract

Background: Drought is most likely the most significant abiotic stress affecting wheat yield. The discovery of drought-tolerant genotypes is a promising strategy for dealing with the world's rapidly diminishing water resources and growing population. A genome-wide association study (GWAS) was conducted on 298 Iranian bread wheat landraces and cultivars to investigate the genetic basis of yield, yield components, and drought tolerance indices in two cropping seasons (2018-2019 and 2019-2020) under rainfed and well-watered environments.

Results: A heatmap display of hierarchical clustering divided cultivars and landraces into four categories, with high-yielding and drought-tolerant genotypes clustering in the same group. The results of the principal component analysis (PCA) demonstrated that selecting genotypes based on the mean productivity (MP), geometric mean productivity (GMP), harmonic mean (HM), and stress tolerance index (STI) can help achieve high-yield genotypes in the environment. Genome B had the highest number of significant marker pairs in linkage disequilibrium (LD) for both landraces (427,017) and cultivars (370,359). Similar to cultivars, marker pairs on chromosome 4A represented the strongest LD (r² = 0.32). However, the genomes D, A, and B have the highest LD, respectively. The single-locus mixed linear model (MLM) and multi-locus random-SNP-effect mixed linear model (mrMLM) identified 1711 and 1254 significant marker-trait association (MTAs) (-log10 P > 3) for all traits, respectively. A total of 874 common quantitative trait nucleotides (QTNs) were simultaneously discovered by both MLM and mrMLM methods. Gene ontology revealed that 11, 18, 6, and 11 MTAs were found in protein-coding regions (PCRs) for spike weight (SW), thousand kernel weight (TKW), grain number per spike (GN), and grain yield (GY), respectively.

Conclusion: The results identified rich regions of quantitative trait loci (QTL) on Ch. 4A and 5A suggest that these chromosomes are important for drought tolerance and could be used in wheat breeding programs. Furthermore, the findings indicated that landraces studied in Iranian bread wheat germplasm possess valuable alleles, that are responsive to water-limited conditions. This GWAS experiment is one of the few types of research conducted on drought tolerance that can be exploited in the genome-mediated development of novel varieties of wheat.

Keywords: Drought tolerance indices; Genome-wide association study; MTAs; SNP markers; Wheat accessions.

Fig. 1

Box-plot representation of the distribution for grain yield (A), grains number per spike (B), thousand kernel weight (C) and spike weight (D) for Iranian landraces and cultivars in the well-watered and rain-fed environments

Fig. 2

Correlation coefficients between GY_WW, GY_RF and various drought tolerance indices (A), GN_WW, GN_RF and various drought tolerance indices (B), TKW_WW, TKW_RF and various drought tolerance indices (C) and SW_WW, SW_RF and various drought tolerance indices (D) for Iranian landraces and cultivars wheat in the well-watered and rain-fed environments. Abbreviations: GY, grain yield; GN, grains number per spike; TKW, thousand kernel weight; SW, spike weight; WW, well-watered; RF, rain-fed; TOL. tolerance index; MP, mean product; GMP, geometric mean product; STI, stress tolerance index; ATI, abiotic stress tolerance index; SSI, stress susceptibility index; DI, new drought resistance index; HM, harmonic mean

Fig. 3

Hierarchical clustering and heatmap based on GY_WW, GY_RF and various drought tolerance indices (A), GN_WW, GN_RF and various drought tolerance indices (B), TKW_WW, TKW_RF and various drought tolerance indices (C) and SW_WW, SW_RF and various drought tolerance indices (D) for Iranian landraces and cultivars wheat in well-watered and rain-fed environments. Abbreviations: GY, grain yield; GN, grains number per spike; TKW, thousand kernel weight; SW, spike weight; WW, well-watered; RF, rain-fed; TOL. tolerance index; MP, mean product; GMP, geometric mean product; STI, stress tolerance index; ATI, abiotic stress tolerance index; SSI, stress susceptibility index; DI, new drought resistance index; HM, harmonic mean

Fig. 4

Principal component analysis of Iranian wheat germplasm exposed to well-watered irrigation and rain-fed environments using PC1 and 2. Biplot for GY_WW, GY_RF and various drought tolerance indices (A), Biplot for GN_WW, GN_RF, and various drought tolerance indices (B), Biplot for TKW_WW, TKW_RF, and various drought tolerance indices (C), and Biplot for SW_WW, SW_RF and various drought tolerance indices (D). Abbreviations: GY, grain yield; GN, grains number per spike; TKW, thousand kernel weight; SW, spike weight; WW, well-watered; RF, rain-fed; TOL. tolerance index; MP, mean product; GMP, geometric mean product; STI, stress tolerance index; ATI, abiotic stress tolerance index; SSI, stress susceptibility index; DI, new drought resistance index; HM, harmonic mean

Fig. 5

Number of imputed SNPs used in different chromosomes of the wheat genomes (A), number of imputed SNPs used in wheat genomes (B)

Fig. 6

Determination of subpopulations number in wheat genotypes based on ΔK values (A), A structure plot of the 298 wheat genotypes and landraces determined by K = 3 (B). Principle component analysis (PCA) for a total of 298 Iranian bread wheat accessions (C). Cluster analysis using Kinship matrix of imputed data for Iranian wheat accessions (D). The dendrogram of Neighbor-Joining clustering constructed using 43,525 SNPs and 298 Iranian wheat accessions (E)

Fig. 7

Overview of the linkage disequilibrium (LD) within the whole association panel per genome using imputed SNPs

Fig. 8

GWAS results (MLM method) for GY_WW, GY_RF, various drought tolerance indices, PCA1 and PCA2 (A), GN_WW, GN_RF, various drought tolerance indices, PCA1 and PCA2 (B), TKW_WW, TKW_RF, various drought tolerance indices, PCA1 and PCA2 (C) and SW_WW, SW_RF, various drought tolerance indices, PCA1 and PCA2 (D) of Iranian landraces and cultivars wheat in well-watered and rain-fed environments. Abbreviations: GY, grain yield; GN, grains number per spike; TKW, thousand kernel weight; SW, spike weight; WW, well-watered; RF, rain-fed; TOL, tolerance index; MP, mean product; GMP, geometric mean product; STI, stress tolerance index; ATI, abiotic stress tolerance index; SSI, stress susceptibility index; DI, new drought resistance index; HM, harmonic mean; PCA, principal component analysis

Fig. 9

Circular Manhattan plots (MLM method) to draw common regions associated with A = GY_WW, GY_RF, various drought tolerance indices, PCA1 and PCA2, B = GN_WW, GN_RF, various drought tolerance indices, PCA1 and PCA2, C = TKW_WW, TKW_RF, various drought tolerance indices, PCA1 and PCA2 and D = SW_WW, SW_RF, various drought tolerance indices, PCA1 and PCA2. Inner to outer circles represent average trait and breeding Values including Y_WW, Y_RF, TOL, MP, GMP, STI, ATI, SSI, DI, HM, PCA1 and PCA2, respectively. The chromosomes are plotted at the outmost circle where thin dotted blue and red lines indicate significant level at P value < 0.001 (− log10 (p) > 3) and < 0.00001 (− log10 (p) > 5), respectively. Green and red dots indicate genome-wide significantly associated SNPs at P value < 0.001 and < 0.00001 probability level, respectively. Scale between ChrUn and Chr1A indicates − log10 (p) values. Colored boxes outside on the top right side indicate SNP density across the genome where green to red indicates less dense to dense. Abbreviations: GY, grain yield; GN, grains number per spike; TKW, thousand kernel weight; SW, spike weight; WW, well-watered; RF, rain-fed; TOL, tolerance index; MP, mean product; GMP, geometric mean product; STI, stress tolerance index; ATI, abiotic stress tolerance index; SSI, stress susceptibility index; DI, new drought resistance index; HM, harmonic mean; PCA, principal component analysis

Fig. 10

GWAS results (mrMLM method) for GY_WW, GY_RF, various drought tolerance indices, PCA1 and PCA2 (A), GN_WW, GN_RF, various drought tolerance indices, PCA1 and PCA2 (B), TKW_WW, TKW_RF, various drought tolerance indices, PCA1 and PCA2 (C) and SW_WW, SW_RF, various drought tolerance indices, PCA1 and PCA2 (D) of Iranian landraces and cultivars wheat in well-watered and rain-fed environments. Abbreviations: GY, grain yield; GN, grains number per spike; TKW, thousand kernel weight; SW, spike weight; WW, well-watered; RF, rain-fed; TOL, tolerance index; MP, mean product; GMP, geometric mean product; STI, stress tolerance index; ATI, abiotic stress tolerance index; SSI, stress susceptibility index; DI, new drought resistance index; HM, harmonic mean; PCA, principal component analysis

Fig. 11

Circular Manhattan plots (mrMLM method) to draw common regions associated with A = GY_WW, GY_RF, various drought tolerance indices, PCA1 and PCA2, B = GN_WW, GN_RF, various drought tolerance indices, PCA1 and PCA2, C = TKW_WW, TKW_RF, various drought tolerance indices, PCA1 and PCA2 and D = SW_WW, SW_RF, various drought tolerance indices, PCA1 and PCA2. Inner to outer circles represent average trait and breeding Values including Y_WW, Y_RF, TOL, MP, GMP, STI, ATI, SSI, DI, HM, PCA1 and PCA2, respectively. The chromosomes are plotted at the outmost circle where thin dotted blue and red lines indicate significant level at P value < 0.001 (− log10 (p) > 3) and < 0.00001 (− log10 (p) > 5), respectively. Green and red dots indicate genome-wide significantly associated SNPs at P value < 0.001 and < 0.00001 probability level, respectively. Scale between ChrUn and Chr1A indicates − log10 (p) values. Colored boxes outside on the top right side indicate SNP density across the genome where green to red indicates less dense to dense. Abbreviations: GY, grain yield; GN, grains number per spike; TKW, thousand kernel weight; SW, spike weight; WW, well-watered; RF, rain-fed; TOL, tolerance index; MP, mean product; GMP, geometric mean product; STI, stress tolerance index; ATI, abiotic stress tolerance index; SSI, stress susceptibility index; DI, new drought resistance index; HM, harmonic mean; PCA, principal component analysis

Fig. 12

The KEGG pathway of ascorbate and aldarate metabolism (A) and pathway of biosynthesis of amino acids (B). The authors declare all that permissions were obtained for the appropriate copyright KEGG image depicted

Fig. 13

The geographical location of the study area (a) and the geographic distribution of Iranian wheat landraces collected between 1931 and 1968 years (b)

Fig. 14

Climatic data in the studied environments