Population diversity and analysis of cytogenetic affinities between different Silybum marianum L. populations from Iran

Abstract

Cytogenetic diversity on ten different Silybum marianum populations was evaluated in this study by using karyotype decomposition technique. The obtained results from karyologic studies of root tip samples showed that the base number of chromosomes was x = 17 in all the considered populations and the number of chromosomes were 2n = 2x = 34. Regarding the ploidy level, all the populations were diploid. According to Stebbin’s classification of karytype symmetry, all the populations were placed in “2B” category, indicating the similar evolutionary situations of the populations. The obtained results from the variance analysis of the resulted data from measuring the characteristics showed that there is a significant difference; as for the probability of 0.1, between the populations, regarding the total chromosome lengths (TL), sum of long arms (Sla), coefficient of variability of centromic index (CVci) and the asymmetric index (AI) and there is a significant difference (p < 0.05) regarding the longest and shortest chromosomes (LC, SC), sum of short arms (SSa), difference in relative lengths (DRL) and the ratio of long to short arms (AR). This indicated the diversity in the size and symmetry of chromosomes among the considered populations. Factor analysis provided three factors that justified nearly 96 % of the total variations between the populations, such that in the first factor, the difference in the relative length and the total chromosome length had the greatest role in creating variations in the populations by having the highest specific vector coefficients. For the 2nd factor, the asymmetric index and variation coefficient of centromic index had a great importance for the variance between the populations. For the 3rd factor, the relative length difference had the most important role in establishing this factor. Cluster analysis with dendrogram sectioning at the distance of 10.59 categorized populations in 4 groups. The least Euclidean distance was observed between the populations in Ahvaz and Sari and the most Euclidean distance was between the populations of Ahvaz and Rasht. The results from cluster variances analyses showed that all the characteristics, apart from DRL had significant diversity between the groups. Regarding the characteristics related to genome length, the populations of Ahvaz, Sari and Esfahan had the highest amounts and Kurdestan population in the 4th group had the highest value regarding AI, CVci and AR characteristics.