Diversity analysis of phenotypic characters in germplasm resources of hawthorn

Abstract:【Objective】Identification and evaluation of germplasm resources are essentially prerequisite for rational utilization. Diversity analysis of phenotypic characters is the most basic method for the evaluation of germplasm resources and the screening of excellent germplasm materials. The aim of this study was to fully understand the genetic diversity of hawthorn (Crataegus pinnatifida Bunge) germplasm resources and improve the efficiency of hawthorn breeding in China.【Methods】This study was conducted in 2019 and 2021 in the Hawthorn Germplasm Repository of Institute of Forestry and Fruit Tree Research, Beijing Academy of Agriculture and Forestry Sciences. 62 hawthorn germplasm resources were evaluated. All the 62 germplasm resources were ten years old and planted in the same plots with same soil and plant management practices. Three bearing-trees with similar growth state were selected for each resource as experimental subjects. 10 healthy, non-pest and disease mature fruits and 1- year twigs were randomly selected from the periphery of the crown in four directions of each sample for index determination. 22 descriptive phenotypic traits and twenty- six quantitative phenotypic traits were determined. The hawthorn germplasm resources description and evaluation system and the hawthorn germplasm resources description specification and data standards were referenced. 22 descriptive phenotypic traits of the 62 hawthorn germplasm resources were recorded by description and observa-tion method. The frequency distribution and diversity of descriptive phenotypic traits were analyzed. 26 quantitative characters of the 62 hawthorn germplasm resources were measured. The quantitative phenotypic traits were analyzed by correlation analysis, cluster analysis and principal component analysis. The frequency distribution of the descriptive phenotypic traits and analysis of diversity for quantitative phenotypic traits were performed using Pandas software v. 1.3.1 and Numpy software v. 1.19.5. The Shannon-Wiener information index H and Simpson genetic diversity index D of the descriptive phenotypic traits were analyzed using Vegan software v.2.5.7. The correlation analysis, cluster analysis and principal component analysis were performed using R software v. 4.1.4.【Results】A total of 92 variant types were observed for the 22 descriptive traits. Among them, the color of 2- years old twigs, stomata shape, stomata density, sepal state, pericarp color and pulp flavor had a large variation range. The Shannon-Wiener information index H and Simpson genetic diversity index D ranged from 0.28 5 to 1.608 7 and 0.202 7 to 0.888 7, respectively. The higher diversity indexes were observed in the color of 2-years old twigs (H: 1.5831, D: 0.7644), sepal state (H: 1.608 7, D: 0.744), pericarp color (H: 1.398 3, D: 0.709 3) and pulp flavor (H: 1.375 6, D: 0.675 4), indicating relatively rich genetic diversity in these traits. The average coefficient of variation of the 26 quantitative traits was 30.24%. Among them, the water content coefficient of variation was the smallest, (5.86% ); the kernel percentage coefficient of variation was the largest (79.91%). The correlation analysis showed that there were high correlations between the 26 quantitative traits, 56 of them were extremely significantly and positively correlated, and 28 were significantly and positively correlated, 35 were significantly and negatively correlated, and 16 were negatively correlated. The results of R-type cluster analysis showed that the 26 quantitative traits could be classified into four categories at Euclidean distance 12. The results of R-type cluster analysis were consistent with the results of correlation analysis, namely that the traits in each of the four categories were significantly or extremely significantly correlated with each other. According to cluster analysis, the 62 hawthorn germplasm resources were divided into 3 groups at the Euclidean distance of 20. The first group could be used for selecting high kernel percentage and high fruit setting rate cultivars, Group Ⅱ for selecting high yield, high edible rate and high soluble sugar content cultivars and Group Ⅲ for selecting large fruit and high Vc content cultivars. The Principal component analysis showed that the cumulative contribution rate of the first eight principal components reached to 76.825%, such as the max single fruit weight, fruit transverse diameter, fruit branch capacity and natural crossing rate and so on, which mainly reflected the fruit size and high yield trait factors. Combined to the correlation analysis results, 26 quantitative traits could be simplified into fourteen main indicators: the max single fruit weight, fruit transverse diameter, fruit branch capacity, average number of fruit setting, leaf width, leaf length, water content, average number of fruit branch, starch content, soluble sugar content, corolla size, pectin content, edible rate and middle fruit branch index, which improved the efficiency for evaluation and breeding selection of hawthorn.【Conclusion】The phenotypic traits of the 62 hawthorn germplasm resources had abundant genetic diversity. They could be divided into 3 groups based on the ward cluster analysis. The potential utilization value of each group for breeding was determined. The 26 quantitative traits could be simplified into fourteen indicators, which could be used as key indicators for the selection, identification and evaluation of hawthorn germplasm resources.