Evolution of repetitive DNA in angiosperms: examples from Nicotiana allopolyploids

Abstract

Allopolyploidy, interspecific hybridisation coupled with genome multiplication, is a prevailing force in the evolution of angiosperms. This thesis examines the consequences of allopolyploidy at the genomic level. The genus Nicotiana is an ideal model system for such studies as it includes allopolyploids formed over widely different time frames (recent to millions of years). The global genome composition of several diploid and allopolyploid species was analysed using a graph-based clustering approach, grouping next generation sequencing reads into clusters (families) of repetitive DNA. Such analysis enables examination of genome size change and diploidisation processes postallopolyploidy. I compared the abundance of >14,000 repeats in the young allopolyploid N. tabacum (less than 0.2 million years old) with relatives of the diploid progenitors, N. tomentosiformis (paternal genome donor) and N. sylvestris (maternal genome donor). Repetitive DNA from the paternal genome tends to be eliminated, whereas DNA from the maternal line remains largely unchanged. A newly described tandem repeat (NicCL3) paternally inherited in N. tabacum, is a striking example. Despite a predicted abundance of ~1% NicCL3 now accounts for only 0.1% of the genome in the allopolyploid, a loss repeated in some synthetic lines of N. tabacum after only four generations. Nicotiana section Repandae formed from a single hybridisation event between relatives of N. sylvestris and N. obtusifolia c 5 million years ago. Subsequent 6 diversification has produced four species where genome size varies by 33%; N. repanda showing genome upsizing and N. nudicaulis showing genome downsizing compared with the expected genome size. There was evidence for the erosion of low copy-number repetitive DNA in both allopolyploids. However in N. repanda genome downsizing has been counteracted by the expansion of a few repeat types. Notably these processes are concurrent with the failure to distinguish progenitor chromosome sets, which I argue is part of the diploidisation process.