Jumping the fine line between species: horizontal transfer and evolution of repetitive elements in eukaryotic species

Abstract

Transposable elements (TEs) are mobile DNA sequences, colloquially known as ‘jumping genes’ because of their ability to replicate to new genomic locations. Active TEs have the potential to transform genome structure by inserting into regulatory regions or accumulating within the genome. Mammals are particularly susceptible to TE expansion; TEs account for significant proportions of all eukaryotic genomes we see today. Horizontal transfer (HT) is the transmission of genetic material between non-mating species. HT is frequently observed in prokaryotes but rarely occurs in multicellular eukaryotes. As TEs are autonomous elements, they have the capability to move into another genome and immediately commence replicating, making them the perfect candidate for eukaryotic HT. Growing evidence indicates that this phenomenon is more widespread than current literature suggests, although questions still remain concerning the frequency of HT and whether all TEs are capable of moving between species. In this thesis, I describe large-scale phylogenomic analyses of eukaryotic species in order to identify and characterise TEs, particularly BovB and L1 (predominantly found in mammals). Past studies on this topic were limited by the scarce availability of genome sequences, which were mainly model organisms. I addressed this limitation by comprehensively screening more than 500 species, demonstrating the remarkable and overlooked diversity of L1s across the eukaryotic tree of life. The rapid explosion of L1s in mammals provides a striking contrast to the diverged L1 lineages found in other metazoans and plants. Even within individual genomes there are marked differences between ancient, degraded L1s and young, intact L1s that are potentially still active. L1s are only believed to vertically inherited; with my plethora of data, I challenged this perception by mining for L1 HT candidates. For comparison, I used BovB retrotransposons as an exemplar of obvious and rampant eukaryotic HT. I extended the current BovB paradigm to include more species, find new vectors of transfer, and refine the estimated times of insertion. Similarities between the distributions of L1 and BovB led me to postulate that the presence of L1s in therian mammals is due to an ancient HT event. Similar L1 HT events can be observed in plants. Given the extent of L1 colonisation in today’s mammals, the idea that L1s were initially introduced as foreign DNA has wide-reaching implications for our perception of genome evolution. Repetitive elements are often discarded from analyses because they are deemed ‘junk’ DNA. However, a genome’s junk is a bioinformatician’s treasure. Chapter 4 details a novel method for resolving species differences by using the repetitive intervals in a genome to identify binary variance (presence versus absence). We were able to infer the evolutionary relationships of 21 modern and ancient elephants and compare the results to an established phylogeny from single nucleotide polymorphisms (SNP). Repeats can thus be used as informative genetic markers, particularly useful for datasets with no known SNP variants. Altogether, this thesis presents in silico approaches for handling large and highly repetitive datasets. By characterising millions of repetitive elements from 503 eukaryotic species, we provide evidence of their impact and importance in eukaryotic evolution.