Genome-Wide Association Study (GWAS) for Identifying SNPs and Genes Related to Phosphate-Induced Phenotypic Traits in Tomato (Solanum lycopersicum L.)

Abstract

Phosphate (P) is a crucial macronutrient for normal plant growth and development. The P availability in soils is a limitation factor, and understanding genetic factors playing roles in plant adaptation for improving P uptake is of great biological importance. Genome-wide association studies (GWAS) have become indispensable tools in unraveling the genetic basis of complex traits in various plant species. In this study, a comprehensive GWAS was conducted on diverse tomato (Solanum lycopersicum L.) accessions grown under normal and low P conditions for two weeks. Plant traits such as shoot height, primary root length, plant biomass, shoot inorganic content (SiP), and root inorganic content (RiP) were measured. Among several models of GWAS tested, the Bayesian-information and linkage disequilibrium iteratively nested keyway (BLINK) models were used for the identification of single nucleotide polymorphisms (SNPs). Among all the traits analyzed, significantly associated SNPs were recorded for PB, i.e., 1 SNP (SSL4.0CH10_49261145) under control P, SiP, i.e., 1 SNP (SSL4.0CH08_58433186) under control P and 1 SNP (SSL4.0CH08_51271168) under low P and RiP i.e., 2 SNPs (SSL4.0CH04_37267952 and SSL4.0CH09_4609062) under control P and 1 SNP (SSL4.0CH09_3930922) under low P condition. The identified SNPs served as genetic markers pinpointing regions of the tomato genome linked to P-responsive traits. The novel candidate genes associated with the identified SNPs were further analyzed for their protein-protein interactions using STRING. The study provided novel candidate genes, viz. Solyc10g050370 for PB under control, Solyc08g062490, and Solyc08g062500 for SiP and Solyc09g010450, Solyc09g010460, Solyc09g010690, and Solyc09g010710 for RiP under low P condition. These findings offer a glimpse into the genetic diversity of tomato accessions' responses to P uptake, highlighting the potential for tailored breeding programs to develop P-efficient tomato varieties that could adapt to varying soil conditions, making them crucial for sustainable agriculture and addressing global challenges, such as soil depletion and food security.

Keywords: GWAS; SNP; gene; genotype-by-sequencing; phosphate uptake; tomato.

Figure 1

(a) Tomato accessions in germination sheet roll under the control (40 ppm) and low phosphate (LP, 4 ppm) conditions, (b) tomato accession grown in germination sheet roll at 14 days after sowing (14 DAS), (c) tomato young plant shown at 14 DAS, (d) kinship matrix of tomato accessions showing closeness among themselves, (e) three-dimensional principal component analysis (PCA) illustrated the population structure based on first three PCA components, (f) a linkage disequilibrium (LD) plot representing the average genome-wide LD decay in the panel with genome-wide markers in tomato. The Y-axis values represent the squared correlation coefficient r², while the X-axis represents the physical distance in kilobase (kb), (g) heat map representing the marker density in 12 chromosomes.

Figure 2

Correlation between traits analyzed. A linear correlation between shoot height and shoot inorganic phosphate content under the control (40 ppm) and low phosphate conditions (LP, 4 ppm) (a), primary root length (PRL) and root inorganic phosphate content in high (control) and low phosphate conditions (b), root iP content uptake efficiency and shoot height (c), and root iP content uptake efficiency and primary root length (PRL) (d).

Figure 3

Plant Biomass and GWAS. (a) Plant biomass (in g) of tomato accessions under the control, (b) Manhattan plot showing significant SNPs, viz. SSL4.0CH10_49261145 and SSL4.0CH06_8670859, is located on chromosome numbers 10 and 6, while the green horizontal line indicates the significance threshold, (c) quantile–quantile (Q-Q) plots based on observed versus expected −log₁₀ (p-values) of plant biomass, (d) whisker plot showing significant SNPs pattern (e) showing significant SNPs distribution pattern on chromosome number 6 and 10. Different colors in (e) indicate different chromosomes.

Figure 4

Shoot iP and GWAS. Shoot inorganic phosphate content (SiP, ppm/mg fresh weight) of tomato accessions under control (a) and low phosphate treatment (b), Manhattan plots showing significant SNPs, viz. SSL4.0CH08_58433186 located on chromosome number 8 in control (c), and significant SNPs distribution pattern on chromosome number 8 (d), Manhattan plot showing SNP SSL4.0CH08_51271168 on chromosome 8 under low phosphate condition (e), and significant SNPs distribution pattern on chromosome number 8 (f), the green horizontal line indicates the significance threshold. The quantile–quantile (Q-Q) plots based on observed versus expected −log₁₀ (p-values) and whisker plots show the behavior of the SiP in control (g,h) and low phosphate condition (i,j).

Figure 5

Root iP and GWAS. Root inorganic phosphate content (RiP, ppm/mg fresh weight) of tomato accessions under control (a) and low phosphate treatment (b), Manhattan plots showing significant SNPs, viz. SSL4.0CH04_37267952 and SSL4.0CH09_4609062, located on chromosome number 4 and 9 in control (c), and significant SNPs distribution pattern on chromosome number 9 (d), Manhattan plot showing SNP SSL4.0CH09_3930922 on chromosome 9 under low phosphate condition (e), and significant SNPs distribution pattern on chromosome number 9 (f), the green horizontal line indicates the significance threshold. The quantile–quantile (Q-Q) plots based on observed versus expected −log₁₀ (p-values) and whisker plots show the behavior of the RiP in control (g,h) and low phosphate condition (i,j).

Figure 6

STRING predicted the interaction of proteins. The genes and their protein IDs (shown in bold) show the interaction of proteins of identified candidate genes using STRING database version 12.0 for plant biomass (a), shoot inorganic phosphate (SiP, (b)) under control, and SiP in low phosphate (c,d), root inorganic phosphate (RiP) in control (e–l). The red circle represents the identified candidate gene. The different selection parameters used are shown in colored lines.

Figure 7

STRING analysis of proteins of candidate genes. The genes and their protein IDs (bold) show the interaction of proteins of the identified candidate gene using STRING database version 12.0 for root inorganic phosphate (RiP) in low phosphate conditions (a–d). The red circle represents the identified candidate gene. The different selection parameters used are shown in colored lines.