Genome-Wide Association Study and Genomic Prediction Elucidate the Distinct Genetic Architecture of Aluminum and Proton Tolerance in Arabidopsis thaliana

Abstract

Under acid soil conditions, Al stress and proton stress can occur, reducing root growth and function. However, these stressors are distinct, and tolerance to each is governed by multiple physiological processes. To better understand the genes that underlie these coincidental but experimentally separable stresses, a genome-wide association study (GWAS) and genomic prediction (GP) models were created for approximately 200 diverse Arabidopsis thaliana accessions. GWAS and genomic prediction identified 140/160 SNPs associated with Al and proton tolerance, respectively, which explained approximately 70% of the variance observed. Reverse genetics of the genes in loci identified novel Al and proton tolerance genes, including TON1-RECRUITING MOTIF 28 (AtTRM28) and THIOREDOXIN H-TYPE 1 (AtTRX1), as well as genes known to be associated with tolerance, such as the Al-activated malate transporter, AtALMT1. Additionally, variation in Al tolerance was partially explained by expression level polymorphisms of AtALMT1 and AtTRX1 caused by cis-regulatory allelic variation. These results suggest that we successfully identified the loci that regulate Al and proton tolerance. Furthermore, very small numbers of loci were shared by Al and proton tolerance as determined by the GWAS. There were substantial differences between the phenotype predicted by genomic prediction and the observed phenotype for Al tolerance. This suggested that the GWAS-undetectable genetic factors (e.g., rare-allele mutations) contributing to the variation of tolerance were more important for Al tolerance than for proton tolerance. This study provides important new insights into the genetic architecture that produces variation in the tolerance of acid soil.

A single‐population GWAS identified AtMATE expression level polymorphism caused by promoter variants is associated with variation in aluminum tolerance in a local Arabidopsis population

“…In Arabidopsis, it has been reported that the ELP of AtALMT1 is associated with Al tolerance level among natural accessions (Hoekenga et al, 2006, Nakano et al, 2020. Besides, the Al-inducible AtMATE expression secondly contributes to Al tolerance in a manner similar to AtALMT1 (Liu et al, 2009).…”

“…The information of the position and description of the Arabidopsis genes located around the GWAS-detected SNPs was obtained from the Araport11 database (http://www.arapo rt.org/; Cheng et al, 2017). The information of the Al tolerance levels of the accessions belonging to the CA population was obtained from Nakano et al, (2020).…”

“…This suggests that the natural variation in Al tolerance occurs depending on the soil pH of the geographical location of the vegetation. Actually, Al-tolerant Arabidopsis accessions were found in acid soil areas (Nakano et al, 2020). Therefore, elucidating the alteration mecha- (Kochian et al, 2004), and this mechanism is substantially conserved across plant species (Ma et al, 2001).…”

“…The genetic factors determining the ELP of several stress tolerance genes have been identified by association mapping using a diversity panel; these factors are known to drive adaptation to environments (Baxter et al, 2010;Yang et al, 2013). Recently, we successfully identified the mechanisms underlying the ELP of the genes associated with H 2 O 2 and Al tolerance by expression GWAS (eGWAS) using Arabidopsis accessions (Sadhukhan et al, 2017;Nakano et al, 2020). In the current study, we analyzed the ELP of AtMATE among Arabidopsis natural accessions to evaluate the contribution of a secondary OA transporter in Al tolerance variation and to identify the genetic factors involved in the transcriptional regulation of AtMATE.…”

Genomic predictions and genome-wide association studies based on RAD-seq of quality-related metabolites for the genomics-assisted breeding of tea plants

“…To estimate how effective the SNPs were in explaining the variation and to validate the potential of the GWAS-associated SNPs in the GP model, a GP analysis by GBLUP (RR) was performed using the SNPs with the lowest p -values (top-ranked) in the GWAS, as described by Nakano et al (2020) with slight modifications 65 . The cumulative effects of the linked loci were estimated using 20–300 top-ranked SNPs (at 20 SNPs intervals).…”