The eco geography of sexual and apomictic polyploids in Potentilla puberula (Rosaceae)

Intraspecific ploidy variation is an evolutionarily important phenomenon in numerous plant species. Polyploidization is not only frequently accompanied by sudden changes in the reproductive system like the breakdown of self-incompatibility, but also with the evolution of apomixis (i. e. asexual reproduction via seeds). Variation in ploidy level and associated reproductive traits are of high relevance for the ecological and spatial distribution of cytotypes, and thus the eco-geography of a species. Sexual and apomictic ploidy cytotypes show various degrees of spatial separation, ranging from the geographic to the population scale, which is driven by three principal factors: migration, habitat preferences or tolerances, and reproductive interaction among cytotypes. In the proposed project we will study the eco-geography of reproductively differentiated cytotypes using the herbaceous cinquefoil Potentilla puberula (Rosaceae) comprising five ploidy levels as model organism. Preparatory studies carried out in the Eastern European Alps evidenced reproductive differentiation into sexual tetraploid and apomictic penta- to octoploids cytotypes which coincided with ecological differentiation and mutual spatial exclusion. Within a latitudinal transect in the Eastern Alps covering 150 populations we will (i) quantify the relative contributions of migration from Pleistocene refugia, ecological sites conditions, and co-occurrence patterns of cytotypes to the current distribution of cytotypes and reproductive modes by means of a variation partitioning algorithm. We will further (ii) identify processes underlying the observed strong spatial separation of reproductively differentiated cytotypes. To this end we will reconstruct the phylogeography and evolutionary origin of cytotypes (based on chloroplast DNA sequences and AFLP-fingerprinting), establish comprehensive measures of ecological site conditions, and test for non-random co-occurrence of cytotypes. As processes potentially causing spatial exclusion we will consider reproductive suppression (i.e., minority cytotype exclusion), reproductive transformation of sexuals by apomicts, and competitive replacement of cytotypes under contrasting ecological conditions using computer simulations parameterized by empirically and experimentally derived data pertaining to the reproductive system of sexual and apomictic individuals. The project thereby will be strongly interdisciplinary and combine the competence of scientists from the fields of phylogeography, vegetation ecology, biomathematics, and reproductive biology and will contribute to the general understanding of evolution and diversification in polyploid plants.

Variation in ploidy, i.e., the number of chromosome sets present per cell nucleus in an individual, is a frequent and evolutionary highly important phenomenon in plants. Potentilla puberula shows remarkable variation in ploidy: five ploidy levels (four to eight chromosome sets per cell nucleus, conditions called tetra-, penta-, hexa-, hepta- and octoploidy) are known which co-occur in many places. In addition the species shows differentiation in reproductive mode with ploidy: tetraploids are preferentially sexual, the higher ploids are usually asexual (also referred to as apomictic). In this project we addressed two main questions: (i) what is the relative importance of reproductive interactions, ecological preferences and colonization abilities in structuring the geographic and ecological distribution of sexual and asexual (i.e., apomictic) forms in Potentilla puberula? The study was conducted in a transect through the Eastern European Alps which covered both regions which potentially served as ice age refugia and such fully glaciated during the last ice age. We identified mutual exclusion of sexual and asexual forms mediated by reproductive interaction as the most important factor explaining the observed geographic distribution. Ecological differentiation among reproductive modes was also important and could be explained by reproductive mode, whereas the number of genomes explained only little of the ecological variation at the studied sites. Furthermore, we did not find evidence for differences in the colonization / migration abilities between reproductive modes. That could be either due to strong human disturbance of current distribution patterns, which has eradicated the natural phylogeographic signals, or because the ice age refugia of P. puberula, from which postglacial re-colonization started, are located outside of the study area. (ii) Which reproductive processes or traits explain the observed mutual spatial exclusion of sexuals and apomicts? In order to answer this question we studied five selected mixed populations along an elevation gradient. We modeled cytotype frequencies using a newly developed computer simulation tool based on the following demographic parameters: fertility (pollen and seed production), vitality (germination rates, growth rates) of individuals, and the strength of barriers preventing cross fertilization among reproductive modes (selfing, shifts in flowering time and pollen precedence). The simulations suggested no long-term co-existance of sexual and asexual individuals because one reproductive mode (in the majority of cases the sexuals) was displaced by the other. However, the process is considerably slow (typically taking several hundred years), and attenuated by the tendency of individuals to cluster by reproductive mode resulting in the patchy, mosaic-like distribution pattern that is observed throughout most of the study area. These results underline the importance of secondary contact zones in the generation of patterns of geographic parthenogenesis.