Daisies in the Desert: On Drivers of Extreme Chromosome Number Reduction in Plants

Abstract

Chromosome number is a key feature of the eukaryotic genome, both in terms of its taxonomic value and its potential to affect evolutionary dynamics. Flowering plants exhibit broad variation in chromosome number, from n = 2 to n ~320, but most taxa have numbers around n = 9-12. These low numbers exist in spite of multiple rounds of polyploidy, or whole genome duplication (WGD) experienced by most angiosperm lineages. That is, rather than successive genome doubling resulting in high chromosome numbers, most lineages experience dramatic chromosome number reduction following WGD during the diploidization process. The consistency of this pattern suggests a selective advantage to chromosome number reduction, but the ultimate drivers are unclear. This dissertation aims to provide insight into drivers of chromosome number reduction in angiosperms. Specifically, I first test a recent hypothesis that descending dysploidy following WGD is driven by selection to maintain gene flow between isolated sub-populations of a neo-polyploid lineage. Under this hypothesis, high chromosome numbers exacerbate the fitness effects of reciprocal gene loss, a potential mechanism by which reproductive isolation arises between lineages following WGD. I test this hypothesis using crossability data from congeneric pairs or groups of taxa, as well as chromosome numbers and estimates of shared WGD age in a comparative phylogenetic framework. My results provide no support for this hypothesis, as crossability showed no relationship with chromosome number or WGD age. I then examine patterns and drivers of chromosome number evolution in a group known for its low chromosome numbers, the Asteraceae subtribe Machaerantherinae. This clade contrasts with its closest relatives, the eurybioid asters, in having base chromosome numbers of x = 6 or lower, comprising a high proportion of annual species, and being adapted to arid environments. I first estimate rates of chromosome number evolution in this clade and its relatives, generating a time-calibrated phylogeny for the purpose. I demonstrate the existence of a distinct rate regime in the Machaerantherinae, characterized by rapid descending dysploidy and very low rates of chromosome number increase. This rapid descending dysploidy is shown to be attributable in part to the frequency of annuals, as they exhibit rates of descending dysploidy over ten times higher than perennials in this group. Finally, I examine biogeographical drivers of chromosome number variation in a test of the hypothesis that low chromosome numbers are favored in outcrossing annuals in unstable habitats. Path analyses provide support for the predicted relationship between aridity and low chromosome number and additionally suggest a role for UV-B radiation in facilitating this evolution. Taken together, these results provide some support for one hypothesis for chromosome number reduction while rejecting another. Namely, arid habitats and annual life history do appear to be associated with the evolution of low chromosome numbers.