Low-density SNP marker sets for genetic variation analysis and variety identification in cultivated citrus

The Citrus species are major fruit crops cultivated in the world and have complex genetic relationships due to sexual comparability between Citrus and related genera. Of these, satsuma mandarin (C. unshiu (Mak.) Marc.) and sweet orange (C. sinensis (L.) Osb.) are widely grown diploid species. In this study, genotyping by sequencing (GBS) was conducted to identify single nucleotide polymorphisms (SNPs) for investigating genetic variation in a citrus collection.

A total of 26,903 high-quality SNPs were detected across nine chromosomes in the 144 citrus varieties, consisting of 70 C. unshiu, 40 C. sinensis, 22 interspecific hybrids, and 12 others. Of these, a core set of 481 SNPs was filtered based on polymorphism information content and genome distribution. Both principal component analysis (PCA) and model-based clustering showed genetic differentiation between C. unshiu and C. sinensis. For interspecific hybrids, these were separated from two species in PCA, but were mixed with each species in model-based clustering. Significant genetic differentiations between three populations were also found using the pairwise F_st. In addition, interspecific hybrids showed higher level of genetic diversity relative to the C. unshiu and C. sinensis populations. With the 481 SNPs, four subsets (192, 96, 48, and 24 SNPs) were generated to evaluate their performance for variety identification. Both 192 and 96 SNP sets distinguished all 144 varieties, while the 48 and 24 SNP sets separated 134 (93.1%) and 110 (76.4%), respectively.

The GBS-based SNP discovery led to robust and cost-effective molecular marker sets to assess genetic variation in the cultivated citrus species with narrow genetic bases. The resulting SNP sets are a resource to enhance the phenotype-based DUS testing by developing a DNA barcode system and thus facilitate new variety breeding and protection in citrus.