Origin of Highly Polyploid Species: Different Pathways of Auto- and Allopolyploidy in 12–18x Species ofAvenula(Poaceae)

Abstract

Polyploidy belongs to the most striking factors in plant evolution. To address the role of allo- versus autopolyploid speciation in natural polyploids, we studied species of the oatlike genus Avenula, which contains diploids and polyploids with up to 22x valence. We conducted DNA sequence analyses of plasmid-cloned ITS repeats of nuclear ribosomal (nr) 35S DNA, complemented by cytogenetic studies including fluorescence in situ hybridization (FISH) with nr and satellite (sat) DNAs in 12–18x polyploid species from SE Europe and Asia Minor. Altogether, four different ITS repeat types are involved in their parentage. One is confined to Avenula armeniaca and the second shared between A. armeniaca and Avenula blaui, none of which is known from extant diploid species. The latter repeat type is characteristic of A. blaui but prevails also in one of the A. armeniaca accessions, as a result of hybridization and/or genetic introgression with an ongoing process of sequence change in favor of A. blaui repeats. The two other repeat types are encountered in Avenula adsurgens and match that of extant diploid species. Auto- and allopolyploidy are corroborated by cytogenetic data, and both types of polyploid evolution thus coexist within the genus Avenula. As exemplified by A. armeniaca, auto- and allopolyploidy are intergrading in highly polyploids, making transitional forms difficult to treat under the classical concepts of autopolyploidy, allopolyploidy, or segmental allopolyploidy. Results point to a pronounced capability of highly polyploids to incorporate further genomes via hybridization or introgression rather easily. This is considered an important evolutionary mechanism for the transfer and/or de novo origin of traits related to ecological adaptations and occupation of new or changing niches, underlining that high polyploidy is not a dead end in evolution. It might explain also the well-documented success of highly polyploids in Avenula and other grass genera relative to the diploids in terms of overall geographical distribution and abundance. Some ITS repeat types with non-Avenula identity belong to the genera Sesleria and Agropyron. All evidence, including cytogenetic data, points to an artifact caused by contamination of DNA with that of alien grass pollen, thus excluding actual hybridization between these three genera.