Genetic and epigenetic instabilities induced by tissue culture in wild barley (<Emphasis Type="Italic">Hordeum brevisubulatum</Emphasis> (Trin.) Link)

A simple tissue culture protocol was developed for efficient plant regeneration from young inflorescence-derived calli in wild barley, Hordeum brevisubulatum (Trin.) Link, an important pasturage grass. Genetic and epigenetic instabilities in the regenerated plants (regenerants) were assessed by three molecular markers AFLP, S-SAP and MSAP. Two pools of calli derived from young inflorescences of a single donor plant and 44 randomly chosen regenerants were subjected to AFLP analysis. Results showed that 74 out of 793 scored bands were polymorphic among the studied samples, giving rise to a genetic variation frequency of 9.3%. The number of variant bands as compared to the donor plant varied greatly among the regenerants, with a small number of regenerants accumulated a large number of variant bands (maximum 55), while the majority of regenerants showed only 2–3 variant bands. A subset of regenerants together with the two pools of calli were selected for S-SAP and MSAP analysis to detect possible retrotranspositional activity of a prominent retroelement family, BARE-1, in the genomes of Hordem species, and possible alterations in cytosine methylation. S-SAP analysis showed that of the 768 scored bands, 151 were polymorphic among the analyzed samples, giving rise to a genetic variation frequency of 19.7%, albeit no evidence for retrotranspositional event was obtained based on locus-specific PCR amplifications. MSAP analysis revealed that tissue culture has caused cytosine methylation alterations in both level and pattern compared with the donor plant. Sequencing of selected variant bands indicated that both protein-coding genes and transposon/retrotransposons were underlying the genetic and epigenetic variations. Correlation analysis of the genetic and epigenetic instabilities indicated that there existed a significant correlation between MSAP and S-SAP (r = 0.8118, 1,000 permutations, P < 0.05), whereas the correlation between MSAP and AFLP (r = 0.1048) is not statistically significant. Electronic supplementary material The online version of this article (doi:) contains supplementary material, which is available to authorized users. Xiaoling Li and Xiaoming Yu contributed equally to this work.