Genome-wide association analysis in peanut accessions uncovers the genetic basis regulating oil and fatty acid variation

Abstract

Background: The cultivated peanut, Arachis hypogaea L., is a critical oil and food crop worldwide. Improving seed oil quality in peanut has long been an aim of breeders. However, our knowledge of the genetic basis of selecting for seed nutritional traits is limited. Based on AhFAD2A and AhFAD2B, scientists have now developed higher oleic acid (80-84%) in peanut. Decoding the genetic makeup behind natural variation in kernel oil and fatty acid concentrations is crucial for molecular breeding-based nutrient quantity and quality manipulation.

Results: Herein, we recognized 87 quantitative trait loci (QTLs) in 45 genomic regions for the concentrations of oil, oleic acid, and linoleic acid, as well as the oleic acid to linoleic acid (O/L) ratio via a genome-wide association study (GWAS) involving 499 peanut accessions. Eight QTLs explained more than 15% of the phenotypic variation in peanut accessions. Among the 45 potential genes significantly related to the four traits, only three genes displayed annotation to the fatty acid pathway. Furthermore, on the basis of pleiotropism or linkage data belonging to the identified singular QTLs, we generated a trait-locus axis to better elucidate the genetic background behind the observed oil and fatty acid concentration association. Expression analysis indicated that arahy.AV6GAN and arahy.NNA8KD have higher expressions in the seeds.

Conclusion: This natural population consisting of 499 peanut accessions combined with high-density SNPs will provide a better choice for identifying peanut QTLs/genes in the future. Together, our results provide strong evidence for the genetic mechanism behind oil biosynthesis in peanut, facilitating future advances in multiple fatty acid component generation via pyramiding of desirable QTLs.

Keywords: Fatty acid; GWAS; Genetic basis; Oil; Peanut.

Fig. 1

The phenotypic distribution and correlation among oil, linoleic, oleic, and O/L in peanut accessions. The diagonal area displayed the phenotypic variation of each trait. The lower left part showed scatter plots of every two traits. The upper right part was correlation coefficient between the composition traits, **, P < 0.01

Fig. 2

Single nucleotide polymorphisms (SNP) dispersion among the 20 chromosomes of cultivated peanut. Horizontal axis depicting chromosomal length (Mb), Green and red shades representing SNP density (SNPs quantity per window). Vertical axis depicting the 20 chromosomes. (a) Polymorphic SNPs apart from the scaffold biomarkers; (b) Polymorphic SNPs (apart from the scaffold biomarkers) following filter

Fig. 3

Overview of GWAS-identified SNPs for oil, linoleic, oleic and O/L traits. (a) Manhattan plot depicting GWAS data for 4 traits. Red dashed line representing the likelihood ratio test threshold (LRT = 4.04), subjected to log₂ transformation. The circles from the inside out are oil, oleic, linoleic and O/L, respectively. (b) Quantile-quantile plot for 4 traits. (c) The number of significant loci and total phenotypic variation explaining all identified loci by GWAS analysis. (d) The distribution of significant associated SNPs on 20 peanut chromosomes

Fig. 4

Trait–QTLs axis for 4 traits and 87 single QTLs. The trait–QTL axis generated according to the QTL findings involving 4 traits and co-localization data of all 87 QTLs. Traits and loci linked by solid lines within the axis were strongly linked to the traits. Loci joined by dashed lines within the axis if two QTLs co-localized at the same loci. Table 1 summarizes the trait names in detail

Fig. 5

Genetic impacts of the leading SNPs at the candidate genes of fatty acid pathway on linoleic, oleic content and O/L. (a-c) Genetic effects of AhFAD2A locus on linoleic, oleic content and O/L, respectively. (d-e) Genetic effects of FUS3 locus on linoleic, oleic content, respectively. (f) Genetic effects of KCS locus on O/L. The P-values were computed according to a two-tailed Student’s t-test

Fig. 6

Expression profiling of candidate genes at different development stages. (a) RNA-seq for the full development at 22 different tissue types and ontogenies, respectively [64]. (b-c) Peanut seeds at different developmental stages and RNA-seq for seed in high and low oleic acid content peanut (our group, unpublished). H57 is a high-oleic acid peanut, and L61 is a low-oleic acid peanut, both undergoing six seed developmental stages (1, 2, 3, 4, 5, 6)