Transposable elements and host evolution: A case of transposon-mediated regulatory innovation during the evolution of pregnancy.

摘要

Organisms evolve by changes to coding and regulatory DNA. To understand how organismal phenotypes arise and diverge over time, therefore, it is necessary to elucidate the mechanisms underlying coding and regulatory evolution. Studies on regulatory evolution have provided insights into how regulatory elements change or are lost over time, but how new elements---such as promoters and enhancers---originate is an open question. One hypothesis is that new regulatory elements are generated de novo by the random gain of transcription factor binding sites (TFBSs) near genes. An alternative opinion, first proposed decades ago, is that new regulatory elements are co-opted from transposable elements as "ready-to-use" elements. Transposable elements (TEs), mobile DNA elements that invade and replicate within genomes, comprise about half of mammalian genomes and contain a variety of regulatory signals necessary for their own propagation. Thus, TEs have the potential to impact host regulatory evolution, and ultimately phenotypic evolution, in a significant way.;In this dissertation I investigate the role of transposable elements in the regulatory evolution of decidual prolactin (dPrl) in mammals, as the promoter in primates derives from a lineage-specific TE called MER39. To understand if and how MER39 has affected dPrl regulation and endometrial functioning in mammals and primates in particular, I surveyed its expression during pregnancy in various mammals and the location(s) of transcriptional initiation. I found that Prl was convergently recruited into uterine expression in primates, rodents, and elephants by the co-option of different transposable elements, highlighting the frequency at which TEs are used for regulatory functions by the host and their importance in regulatory innovation. I also traced the origin of the MER39-derived promoter in primates, showing that evolution of the strong dPrl promoter in apes was a multistep process that took millions of years. Strong promoter activity of MER39 evolved coincident with the origin of a novel reproductive character in apes, interstitial invasion; thus, transformation of this TE into a regulatory element likely played a role in the evolution of pregnancy in apes. Mechanistically, I show that strong promoter activity in apes involves epistatic interactions between TFBSs ancestral to MER39 and derived sites. I propose a novel mode of molecular evolution by which MER39 was transformed, called "epistatic capture," defined as the fixation of a TFBS that is ancestral but variable in outgroup lineages, and is fixed in the ingroup because of epistatic interactions with derived TFBSs. A review of the literature suggests that epistatic capture may be a common mechanism by which TEs are domesticated for regulatory functions in host tissues like the endometrium. Finally, since TEs have had a major impact on regulatory and other types of innovations in placental tissues, I argue that TEs have facilitated the rampant diversification of the placenta in eutherian mammals and potentially other fast-evolving tissues.