Camellia nitidissima var. phaeopubisperma is a variety in the section Chrysantha of the genus Camellia of the family Theaceae which is native to Fangchenggang, Guangxi, China. To date, the genetic diversity and structure of this variety remains to be understood. In the present study, eight simple sequence repeat (SSR) molecular markers previously screened were used to analyze the genetic diversity and structure of C. nitidissima var. phaeopubisperma natural populations from 14 growing areas in China, so as to determine the influence of environmental changes on genetic variations and provide the basis for introduction and selection of suitable growing sites of that variety. Our results show that, for the eight SSR loci, the observed numbers of alleles per locus (Na) and the effective numbers of alleles per locus (Ne) were nine and 3.206, respectively on average, and the Ne was lower than the Na for all loci; the observed heterozygosity (Ho) was lower than the expected heterozygosity (He). For all the eight loci, the fixation index (F) was greater than 0, and the intra-population inbreeding coefficient (Fis) for seven loci was positive. Three loci were moderately polymorphic (0.25 < polymorphism information content (PIC) <0.5), and five loci were highly polymorphic (PIC > 0.5); all eight loci had a moderate genetic differentiation level (0.05 < genetic differentiation coefficient (Fst) <0.25). As shown by the genetic diversity analysis, the He was bigger than the Ho for 10 populations, indicating the presence of a certain degree of intra-population inbreeding. The F had a negative value for four populations, suggesting that excessive random mating was present within each of them. Results of the analysis of molecular variance show that 19% of the total variation was attributed to among-individuals and 78% of the total variation originated from within-individuals. The adjusted Fst (F'st) was 0.073, indicative of a moderate level of genetic differentiation among the populations. The value of gene flow was greater than 1 (7.367), suggesting that genetic differentiation among populations was not caused by genetic drift. Results of the STRUCTURE analysis show that all the samples tested could be clustered into five ancestor groups. Results of the Unweighted Pair Group Method using Arithmetic Averages (UPGMA) clustering analyses show that the 84 plant samples could be divided into three clusters and natural populations from the 14 growing areas could be divided into two clusters. Clustering results of the populations were not affected by geographic distances, and gene flow occurred frequently among the populations, suggesting that the genetic variation among the natural populations of C. nitidissima var. phaeopubisperma from 14 growing areas was not influenced by environmental changes of these areas but mainly derived from the genetic variation present in pre-introduction populations.
Read full abstract