Conserved but mechanistically diverse piRNA defence against gypsy LTR retrotransposons in insect ovaries

Abstract

Organisms have evolved highly adaptable defence systems to silence mobile genetic elements such as transposons. While individual transposons may emerge and vanish over time, the long-term evolutionary stability of the pathways that suppress them remains poorly understood. In the ovarian somatic cells of Drosophila, envelope-carrying gypsy long terminal repeat (LTR) retrotransposons are silenced by the Piwi-interacting RNA (piRNA) pathway, which depends exclusively on Piwi - an evolutionary derivative of an Aubergine/Piwi gene duplication unique to flies. Recent studies, however, have revealed that envelope-carrying gypsy elements are widespread across metazoans and particularly abundant in insect genomes, raising questions about the origin of ovarian somatic piRNA defence. Here we demonstrate that Aedes and Anopheles mosquitoes, as well as stingless bees and crickets, all produce piRNAs targeting gypsy elements in ovarian somatic cells, indicating a continuous arms race in this cellular niche for more than 400 million years of insect evolution. Notably, in Aedes aegypti, ovarian somatic cells express the same piRNA clusters as other somatic tissues, where they are known to target RNA viruses - suggesting a shared origin of anti-viral and anti-retrotransposon defences. Furthermore, we reveal striking lineage-specific differences in ovarian somatic piRNA biogenesis: slicing-independent phasing dominates in dipterans, ping-pong amplification in bees, and slicing-dependent phasing in crickets. Together, these findings indicate that distinct piRNA pathways have independently evolved at different timepoints to silence the same class of retrotransposons in insect evolution.