| A core subset captures most of the available abundant genetic diversity of the entire collection, which will lay the foundation for studying population differentiation, exploring elite gene resources, and accelerating peanut genetic improvement. Although great progress has been made on the development and evaluation of the core subset and mini core, the feature and relation of phenotype and genotypic variation is still unstated, as well as the genetic pattern of quantitative trait loci(QTLs) and different alleles per locus remains unclear.In this study, we compared the Coefficient of Variation(CV%) and the Shannon-Weaver index(H′) among five botanical types, seven ecological zones, and 36 subregions for 14 phenotypic traits to clarify the distribution of genetic diversity. Furthermore, peanut yield and quality are two important breeding targets in peanuts controlled by complex genetic factors and strongly influenced by environments. The yield-related traits and important quality traits of 169 peanut lines from the core subset were evaluated in six growing environments. A total of 139 SSR markers were used to genotype total lines for analyzing genetic diversity, linkage disequilibrium and population structure. The loci significantly associated with yield traits and important quality traits were obtained under multiple enviroments. Comparisons of phenotypic effect were also conducted to discriminate the different alleles at each locus. The main results were as follows:1. From the comparison of the Coefficient of Variation(CV%) and the Shannon-Weaver index(H′) of five botanical varieties in the core subset, the diversity of Virginia was the richest(CV%=219.26%, H′=20.96). Dragon also had extensive variation(CV%=216.71%, H′=19.97) despite of fewer landraces. The cultivated region EZ2, including five botanical varieties- Valencia, Spanish, Dragon, Virginia and Intermidate- had the richest diversity(CV%=215.27%, H′=23.53) compared with the other EZs. The primary diversity center of the peanut cultivars was on the border of Anhui, Jiangsu, Shandong, and Henan, namely sEZ6, between the north of the Huai River and the south of the Huang River. The sEZ10, sEZ17, and sEZ22, namely the Sichuan Basin, the Zhejiang coast and the Guangdong coast, gradually formed three secondary diversity centers of peanut cultivars.2. A PCA was used to analyze the genetic evolution among peanut cultivar populations. The first five independent components accounted for 76.10% of the observed variation. Using a scatter plot, 257 samples were plotted into two-dimension spaces determined by the first two PC scores. The accessions of Valencia were included into Spanish and those of Intermediate were included in Virginia, which indicated that the respective former two possibly derive from the latter two. Dragon mostly coincided with Virginia, showing the close affinity between them. The ranges of EZ1 to EZ7 were partially overlapping, indicating that the evolution of the peanut was a gradual spreading process from the center or sub-centers of diversity.3. On the basis of superior or equal performance over multiple environments, 60 excellent lines were selected with a near infrared reflectance spectroscopy model. There were 22 accessions whose protein content was above 31%, 20 accessions whose oil content was more than 53%, and 18 accessions whose oleic acid content exceeded 55%. The high protein accessions were mainly comprosed of Spanish and Dragon and came mainly from the Shandong peninsula. High oleic acid cultivars, whose botanical type was mainly Virginia, mostly originated from the mountains, especially the Taihang Mountains. High oil cultivars were from the mountains and the Han River plain, which were composed of Spanish and Virginia.4. The genetic diversity of the core subset was evaluated, and a total of 1571 alleles with an average of 11.08 on 139 polymorphic SSR loci were produced. The PIC values varied from 0.023 to 0.912, with an average of 0.778. The genetic distance ranged from 0.261 to 0.937, with an average of 0.805. A Cluster analysis and a model-based population structure analysis divided all accessions into three subpopulations. The Fst among subpopulations averaged 0.0195, and the Dsa averaged 0.5183. Additionally, 6.97% of SSR linear marker pairs showed significant LD(P<0.01). The r2 between nonlinear loci pairs averaged 0.011 and the LD extended 10.19 cM.5. A total of 169 loci were significantly associated(P≤0.001) with yield traits, and the 43 that were significantly associated(P≤0.001) with important quality traits involved 80 SSR markers. The R2 per associated locus ranged from 0.0604 to 0.6235. Among them, twenty-four loci were repeatedly associated with the same trait when analyzed by the phenotypic values in multiple environments or the average values, and 48 loci were associated with at least two traits simultaneously. Nineteen loci strongly(P≤5.88E-08) associated with the traits were major effect loci whose R2 was equal to or greater than 24.74%.6. The phenotypic effects of 24 stable loci were computed and were significantly associated with one trait in multiple environments. Some elite alleles were obtained, including of 40 dwarf alleles, 12 alleles increasing plant yield, 22 alleles increasing pod weight, and 14 high protein alleles, such as GA32-A373, PD59-A75, Ah3-A208, and PM358-A131. These results may provide useful information for marker-assisted selection in peanut breeding programs for yield traits and important quality traits. |
PDF Full Text Request