Mining candidate genes underlying seed oil content using BSA-seq in soybean

Attempts to alter seed oil content in soybean have been hampered by the complicated genetic architecture as being controlled by multiple genes. In order to investigate the genetic basis of oil content, bulk segregant analysis based on next generation resequencing (BSA-seq) was performed on a soybean natural population comprising 1551 acessions with diverse worldwide origins conferring wide diversity for seed oil content. Firstly, we identified a total of 2030 genes using the delta SNP-index method from the two phenotypic bulks showing extreme high and low oil content with significant differences. Secondly, using a stringent criterion of delta index greater than 0.9 or lower than −0.9, 130 candidate genes were further identified, including Glyma.04G047000, which encodes a lipid metabolism enzyme lecithin:cholesterol acyltransferase 1 (LCAT1). A 1-bp insertion was detected in LCAT1 gene leading to a truncated protein,and was found to be significantly correlating with seed oil content both in this natural population and also in another validation population containing 182 cultivars. Finally, high oil alleles for 130 candidate genes were shown to be more prevalent in cultivars than in landraces, and were particularly abundant in the accessions originated from the North ecoregion in China, which are recognized for their high oil content. This result demonstrates that these oil-associated alleles can be employed as selective markers for soybean breeding with high oil content. The BSA-seq strategy used in our study identified candidate genes for oil content and, therefore, may contribute to improving soybean seed oil content.