Other tetraploid species and conspecific diploids as sources of genetic variation for an autotetraploid

PREMISE OF THE STUDY: Most plants are polyploid and have more than two copies of the genome. The evolutionary success of polyploids is often attributed to their potential to harbor increased genetic variation, but it is poorly understood how polyploids can attain such variation. Because of their formation bottleneck, newly formed tetraploids start out with little variation. Tetraploids may attain genetic variation through a combination of new mutations, recurrent formation, and gene exchange with diploid ancestors or related tetraploid species. We explore the role of gene exchange and introgression in autotetraploid Rorippa amphibia, a species that harbors more genetic variation than its diploid ancestors. • METHODS: We crossed autotetraploid R. amphibia to diploid conspecifics and tetraploid R. sylvestris and backcrossed resulting F(1) hybrids. We used flow cytometry to determine the ploidy of all progeny. • KEY RESULTS: Tetraploids of R. amphibia and R. sylvestris were interfertile; F(1) hybrids were fertile and could backcross. Crosses between diploids and tetraploids yielded a small number of viable, often tetraploid progeny. This indicates that unreduced gametes can facilitate gene flow from diploids to tetraploids. We detected a frequency of unreduced gametes of around 2.7 per 1000, which was comparable between diploids and tetraploids. • CONCLUSIONS: Introgression from tetraploid R. sylvestris provides a realistic source of variation in autotetraploid R. amphibia. Only in a scenario where other compatible partners are absent, for example immediately after tetraploidization, gene flow through unreduced gametes from diploids could be an important source of genetic variation for tetraploids.