Format: ORAL

Ingrid E Jordon-Thaden1,2,3; Douglas E. Soltis3; Pamela S. Soltis 2,3

1 Department of Botany, University of Wisconsin Madison, Madison, Wisconsin, USA 2 Florida Museum of Natural History, University of Florida, Gainesville, Florida, USA 3 Department of Biology, University of Florida, Gainesville, Florida, USA

Recently formed allopolyploids Tragopogon mirus and T. miscellus (Asteraceae) and their diploid parental species, T. dubius, T. porrifolius, and T. pratensis, offer an opportunity to study the earliest stages of allopolyploidy. This study is the first large-scale common-garden and genotyping experiment to assess genomic and phenotypic trait variation in the North American Tragopogon allopolyploid complex from both natural and synthetic lines. We conclude that polyploidization contributes to larger progeny and therefore improved establishment of the young plant, as well as the already known gigas effect observed in mature polyploid plants, resulting in increased competitive ability. Natural populations as well as the three generations of resynthesized allopolyploid lines examined here generally showed minor phenotypic trait differences, indicating that any trait differences due to polyploidization were seen in the very first generation after allopolyploidy had occurred and did not change much in subsequent generations after polyploid formation. Likewise, we detected evidence of diploidization via gene loss and as well as gene silencing in the first generations after polyploid formation and repeated across different lineages. We observed increasing gene loss and silencing over several generations of synthetic polyploids and even more in the natural polyploids approximately 45-50 generations old. Both genomic and expression changes are maintained and continue to accumulate over subsequent generations at a steady rate over time, with gene loss occurring at a greater magnitude and rate than gene silencing over the same time period. Gene silencing does not seem to be a prerequisite for gene loss, and the two processes may operate independently of each other. Our findings confirm McClintocks hypothesis of genomic and transcriptomic shock following hybridization, but in this case following allopolyploidization. Together, these results further demonstrate the dynamic nature of polyploid genomes shortly after their formation.