Rapid evolution of piRNA clusters in the Drosophila melanogaster ovary.

The piRNA pathway is conserved to repress transposable element (TE) activity in the animal germline via a specialized class of small RNAs called piwi-interacting RNAs (piRNAs). piRNAs are produced from discrete genomic regions called piRNA clusters (piCs). While the molecular processes by which piCs function are relatively well understood in Drosophila melanogaster , much less is known about the origin and evolution of piCs in this or any other species. To investigate piC origin and evolution, we use a population genomics approach to compare piC activity and sequence composition across 8 geographically distant strains of D. melanogaster with high quality long-read genome assemblies. We perform annotations of ovary piCs and genome-wide TE content in each strain. Our analysis uncovers extensive variation in piC activity across strains and signatures of rapid birth and death of piCs. Most TEs inferred to be recently active show an enrichment of insertions into old and large piCs, consistent with the previously proposed 'trap' model of piC evolution. By contrast, a small subset of active LTR families is enriched for the formation of new piCs, suggesting that these TEs have higher proclivity to form piCs. Thus, our findings uncover processes leading to the origin of piCs. We propose piC evolution begins with the emergence of piRNAs from a few specific LTR retrotransposon insertions that subsequently expand by accretion of other TE insertions during evolution to form larger 'trap' clusters. Our study shows that TEs themselves are the major force driving the rapid evolution of piCs.