Genome-Wide Association Insights into the Genomic Regions Controlling Oil Production Traits inAcrocomia aculeata(neotropical native palm)

Abstract

Macauba (Acrocomia aculeata) is a non-domesticated neotropical palm that has been attracting attention for economical use due to its great potential for oil production, comparable to the commercially used oil palm (Elaeis guineenses). This palm has an annual production, and its fruit harvest occurs without the destruction of the plant, allowing for sustainable management. Besides, it can be used in integrated Crop-Livestock-Forestry systems, preventing devastating effects on the environment. With genetic improvement, oil and biofuel production can increase and make this species more competitive. However, that requires an understanding of the architecture of the agronomic traits. For instance, the discovery of associations between quantitative trait loci and economically important traits represents an advance toward understanding its genetic architecture and can contribute to accelerating the macauba domestication. In this context, single-trait and multi-trait GWAS models were performed to identify candidate genes related to oil production traits in macauba. We randomly selected 201 palms from a natural population and collected phenotypic data in two years of production. Phenotypic measurements of 13 traits involved with fruit production, processing, and pulp oil content were obtained, and data analysis was performed using a mixed linear model. Genotyping was performed with SNP markers, following the protocol of genotyping-by-sequencing. The SNP calling was performed using three different strategies since macauba does not have a reference genome: using i) de novo sequencing, ii) the Elaeis guineenses Jacq. reference genome and iii) the macauba transcriptome sequences. The quality control of the SNP data was performed considering minor allele frequency ≥ 0.01, a maximum of 30% of missing data per locus, and 45% of missing data per individual. Missing data was inputted with the software Beagle 5.3. Single-trait analysis was fitted using five different models from GAPIT in the software R, while multi-trait analysis was fitted using a multivariate stepwise method implemented in the software TASSEL. Multi-trait analyses were conducted in all pairwise trait combinations. Results showed statistically significant differences in all phenotypic traits studied, and heritability values ranged from 63 to 95%. Genetic correlation between traits ranged from -0.47 to 0.99. Gene annotation in Blast2Go detected 15 candidate genes in seven phenotypic traits in the single-trait GWAS, ten in the oil palm genotypic dataset, four in the de novo dataset, and one in the transcriptome dataset, in the multi-trait GWAS, four candidate genes were detected in ten traits combination. Gene mapping for the candidate genes identified similarities with Elaeis guineensis and Phoenix dactylifera. Candidate genes are responsible for metal ion binding and transport, protein transportation, DNA repair, and other cell regulation biological process. We provide new insights on genomic regions that mapped candidate genes involved in macauba oil production phenotypes. These potential candidate genes need to be confirmed for future targeted functional analyses and multi-trait associations need to be scrutinized to investigate the presence of pleiotropic or linked genes. Associated markers to the traits of interest may be valuable resources for the development of marker-assisted selection in macauba for its domestication and pre-breeding.