Search

Previous studies have suggested that the H haplome in Elymus could originate from different diploid Hordeum species, however, which diploid species best represent the parental species remains unanswered. The focus of this study seeks to pinpoint the origin of the H genome in Elymus. Allopolyploid Elymus species that contain the StH genome were analyzed together with diploid Hordeum species and a broad sample of diploid genera in the tribe Triticeae using DMC1 sequences. Both parsimony and maximum likelihood analyses well separated the American Hordeum species, except Hordeum brachyantherum subsp. californicum, from the H genome of polyploid Elymus species. The Elymus H-genomic sequences were formed into different groups. Our data suggested that the American Horedeum species, except H. brachyantherum subsp. californicum, are not the H-genomic donor to the Elymus species. Hordeum brevisubulatum subsp. violaceum was the progenitor species to Elymus virescens, Elymus confusus, Elymus lanceolatus, Elymus wawawaiensis, and Elymus caninus. Furthermore, North American H. brachyantherum subsp. californicum was a progenitor of the H genome to Elymus hystrix and Elymus cordilleranus. The H genomes in Elymus canadensis, Elymus sibiricus, and Elymus multisetus were highly differentiated from the H genome in Hordeum and other Elymus species. The H genome in both North American and Eurasian Elymus species was contributed by different Hordeum species.

It has been hypothesized from isozymic and cytological studies of Elymus species that the Old and New World taxa may be of separate origin of the H genome in the StH genome species. To test this hypothesis, and estimate the phylogenetic relationships of polyploid Elymus species within the Triticeae, the second largest subunit of RNA polymerase II (RPB2) sequence of 36 Elymus accessions containing StH or StY genomes was analyzed with those of Pseudoroegneria (St), Hordeum (H), Agropyron (P), Australopyrum (W), Lophopyrum(Ee), Thinopyrum(Eb) and Dasypyrum (V). Our data indicated that the H genome in Elymus species is differentiated in accordance with geographical origin, and that the Eurasian and American StH genome species have independent alloploid origins with different H-genome donors. Phylogenetic analysis of Y genome sequences with other genome donors (St, H, P, W) of Elymus revealed that W and P genomes are sister to Y genome with a 87% bootstrap support, and that StY and StH species group might have acquired their RPB2 St sequences from distinct Pseudoroegneria gene pools. Our data did not support the suggestion that the St and Y genomes have the same origin as put forward in a previous study using ITS data. Our result provides some insight on the origin of Y genome and its relationship to other genomes in Elymus.

The morphological similarity of Elymus trachycaulus to the Eurasian E. caninus has often been noted. This has lead to controversial and contradicting taxonomic treatments. Nevertheless, there has been no systematic investigation on molecular genetic similarity between E. trachycaulus and E. caninus. In this study, random amplified polymorphic DNA (RAPD) analysis was used to study the similarity between the two species. RAPD analysis of 38 samples representing E. caninus and E. trachycaulus complex yielded 111 interpretable RAPD bands. The Jaccard's similarity values for E. caninus ranged from 0.38 between accessions H10345 and H10353 to 0.97 between accessions H8745 and H10096, with an average of 0.67. The Jaccard's similarity values for E. trachycaulus complex ranged from 0.09 between E. trachycaulus ssp. subsecundus (PI 537321) and E. trachycaulus ssp. violaceus (PI 272612) to 0.78 between accessions PI 315368 and PI 372644, with an average of 0.43. The results from different analyses (NJ and PCA) were similar but not identical. The molecular genetic separation between E. caninus and E. trachycaulus was consistent. The PCA analysis clearly separated all E. caninus accessions from E. trachycaulus and its subspecies. The NJ analysis also showed separation between most accessions of E. caninus and E. trachycaulus. Further analysis excluding E. trachycaulus ssp. subsecundus and ssp. violaceus revealed that E. caninus species and E. trachycaulus species were clearly separated into two distinct groups. The RAPD data thus support the treatment of E. caninus and E. trachycaulus as distinct species. The analyses further indicate that E. violaceus is nested within E. trachycaulus, and more related to E. trachycaulus complex rather than to E. caninus.