Old world polyploids

Tragopogon has emerged as a model system for the study of recent polyploidy in natural populations. Both T. mirus and T. miscellus represent recent allopolyploids that formed in the United States after their three diploid progenitors (T. dubius, T. porrifolius and T. pratensis) were introduced from Europe in the early 1900s (Ownbey 1950; Soltis et al., 2012). These allopolyploids never formed in Europe due to their geographic/ecological isolation but have formed repeatedly since the diploids were brought into close proximity. These polyploids are estimated to be approximately 40 generations-old (~80-90 years for these biennials) (Soltis et al. 1995, 2004, 2012; Symonds et al. 2010).

Intensive study of these newly formed polyploids as well as synthetic lines of both polyploids has afforded the unique opportunity to elucidate the genetic and genomic changes that occur in the initial generations post polyploid formation. But what occurs deeper in evolutionary time? What is the fate of the genetic, chromosomal, and genomic changes that are observed in these young polyploids? To fill this crucial gap in our understanding of the evolutionary significance of the extensive genetic and chromosomal polymorphism in new polyploids, studies of related and older polyploids are needed. In an effort to fill this void, we established another, older model polyploid system in Tragopogon, T. castellanus.

Tragopogon castellanus is endemic to Spain and occurs only along the northern half of the Iberian Peninsula. Morphological, cytological and molecular phylogenetic analyses suggested that T. lamottei and T. crocifolius were the putative parents. Tragopogon castellanus is morphologically variable and somewhat similar to parental T. crocifolius; as a result T. castellanus was once considered a subspecies of T. crocifolius (Willkomm 1893; Díaz de la Guardia and Blanca 1989). To help resolve the ancestry of T. castellanus, phylogenetic analyses of multiple populations of the allotetraploid (2n = 24) and its putative diploid parents, T. crocifolius and T. lamottei, were conducted.  Sequence and cytogenetic data (FISH, GISH) confirmed that T. crocifolius and T. lamottei, are the diploid parents of T. castellanus, and that this polyploid formed at least three times ( Mavrodiev et al., 2008, 2015). Tragopogon castellanus may have formed before the last glacial maximum which would date the formation of this polyploid species to perhaps as long ago as 2.6 million years ( Mavrodiev et al., 2015).

The multiple, independent occurrences of Tragopogon allopolyploid formation in both young New World species and the older T. castellanus permits the assessment of the fate of chromosomal and genetic change through time, that is in both recent (neopolyploids) and older (mesopolyploids) polyploids. For example, of the three independent origins of T. castellanus, one group of populations is distinct in exhibiting two pairs of rearranged chromosomes—these changes are fixed in all plants. These data suggest that some of the chromosomal variants that originate in young polyploids may become fixed in populations, contributing to novelty in older polyploid lineages (Mavrodiev et al., 2015).  Patterns of gene loss and silencing were highly similar across origins of T. castellanus, mirroring the changes seen in the young polyploids (T. mirus and T. miscellus), although in the older polyploid T. castellanus the changes are fixed in populations (unlike the polymorphisms seen in the young polyploids)(Boatwright et al., 2018, 2021).