Evolution of piRNA-mediated TE Repression in Drosophila

In metazoan germlines, the piRNA pathway acts as a genomic immune system, employing small RNA-mediated silencing to defend host DNA from the harmful effects of transposable elements (TEs). Expression of genomic TEs is proposed to initiate self regulation by increasing the production of repressive piRNAs, thereby “adapting” piRNA-mediated control to the most active TE families. However, the mutational and epigenetic processes that give rise to piRNA-mediated silencing when new TEs invade the host also remain poorly understood. In addition, piRNA pathway proteins, which execute piRNA biogenesis and enforce silencing of targeted sequences, evolve rapidly and adaptively in animals. If TE silencing is ensured through piRNA biogenesis, what necessitates changes in piRNA pathway proteins? In this dissertation, we first explored forces that underlie the adaptive evolution of piRNA pathway proteins by performing interspecific complementation and examined functional differences between Drosophila melanogaster and D. simulans alleles of three adaptively evolving piRNA pathway proteins: Armitage, Aubergine and Spindle-E. We discovered that D. simulans alleles of aub and armi exhibit enhanced off-target effects on host transcripts in a D. melanogaster background, as well as modest reductions in the efficiency of piRNA biogenesis, suggesting that promiscuous binding of D. simulans Aub and Armi proteins to host transcripts reduces their participation in piRNA production. Avoidance of genomic auto-immunity may therefore be a critical target of selection. P-elements have invaded the genomes of three Drosophila species within the last century, providing unique opportunities to study the evolution of piRNA mediated repression. In the second part of this dissertation, we introduced P-elements into a naive D. melanogaster strain through germline transformation, and chronicled their effects on the evolution of host TE repression in 10 laboratory populations at two different temperatures. We found that populations which evolved repression are associated with accumulation of active P-elements within their genomes. Furthermore, our observation that repression correlates with P-element insertions into multiple different piRNA clusters strongly supports the “trap” model that acquisition of TE repression is through TE insertions into piRNA clusters, and indicates that the evolution of repression occurs via simultaneous emergence of numerous different repressor alleles.