Extensive clonality in a Bahiagrass germplasm collection from Uruguay

SPERANZA PR; CATANZARO MP; REYNO R; RUA GH

Lake Buena Vista

Conferencia; International Forage and Turf Breeding Conference 2019 (IFTBC 2019); 2019

University of Florida, Institute of Food and Agricultural Sciences (UF/IFAS)

Populations of apomictic species are composed of single or a combination of clones. Individual plants in species like bahiagrass (P. notatum Flügge) may also propagate vegetatively by stolons and compete with one another resulting some populations dominated by individuals originating from a single seedling, a life form that is associated with extremely long lifespans. Consequently, long lived apomictic individuals composed of a very large number of ramets have ample opportunity to produce and disseminate a very high number of clonal seed further extending the range of a single clone to new locations. The rate of formation of new clones will depend on the reproductive system of each species and clone, its remaining ability to reproduce sexually or the availability of sympatric sexual biotypes. The resulting overall rate of origination of new clones is usually unknown and extensive germplasm collections may in extreme cases be majoritarily composed of a single widespread clone. In these cases germplasm evaluation, conservation, use and management in general may become extremely costly and inefficient. In order to assess the degree of variability and optimize the management and evaluation of the germplasm collection kept in INIA (Uruguay) we analyzed a total of 166 individuals from 74 different locations. Different individuals in a single collection point were chosen based on observable differences and distance. All individuals were analyzed with a set of four microsatellite markers specifically designed for the species which had previously been assessed for Mendelian behavior in a commercial sample of sexual diploid Pensacola bahiagrass. We identified a total of 29 multilocus genotypes (MLG) with an average 2.7 alleles per locus. Considering differences in individual alleles in otherwise identical MLG as putative mutations, a total of ten single genotypes and nine putative clones with more than one representative were identified. The most widespread clone contained 85 individuals from different locations in the whole country; the second most frequent one included 31 individuals mostly from eastern locations, and the rest of the putative clones included 2-7 individuals. Of the smaller clones the third most frequent one was also clustered in north-western locations. Especially in the case of highly apomictic grasses, specifically designing collection strategies that take into account the expected structure of variability of a wild species based on its reproductive biology can result in much a more efficient germplasm management strategy. Based on this analysis a whole collection can be reduced by almost ten times for evaluation, although the exploration of mutational variability within a promising widespread clone should not be ruled out. These results can also greatly simplify the choice of parents for crossing programs.