Genome defence in hypomethylated developmental contexts
Playfoot, Christopher James
Retrotransposons constitute around 40% of the mammalian genome and their aberrant activation can have wide ranging detrimental consequences, both throughout development and into somatic lineages. DNA methylation is one of the major epigenetic mechanisms in mammals, and is essential in repressing retrotransposons throughout mammalian development. Yet during normal mouse embryonic development some cell lineages become extensively DNA hypomethylated and it is not clear how these cells maintain retrotransposon silencing in a globally hypomethylated genomic context. In this thesis I determine that hypomethylation in multiple contexts results in the consistent activation of only one gene in the mouse genome - Tex19.1. Thus if a generic compensatory mechanism for loss of DNA methylation exists in mice, it must function through this gene. Tex19.1-/- mice de-repress retrotransposons in the hypomethylated component of the placenta and in the mouse germline, and have developmental defects in these tissues. In this thesis I examine the mechanism of TEX19.1 mediated genome defence and the developmental consequences upon its removal. I show that TEX19.1 functions in repressing retrotransposons, at least in part, through physically interacting with the transcriptional co-repressor, KAP1. Tex19.1-/- ES cells have reduced levels of KAP1 bound retrotransposon chromatin and reduced levels of the repressive H3K9me3 modification at these loci. Furthermore, these subsets of retrotransposon loci are de-repressed in Tex19.1-/- placentas. Thus, my data indicates that mouse cells respond to hypomethylation by activating expression of Tex19.1, which in turn augments compensatory, repressive histone modifications at retrotransposon sequences, thereby helping developmentally hypomethylated cells to maintain genome stability. I next aimed to further elucidate the role of Tex19.1 in the developing hypomethylated placenta. I determine that Tex19.1-/- placental defects precede intrauterine growth restriction of the embryo and that alterations in mRNA abundance in E12.5 Tex19.1-/- placentas is likely in part due to genic transcriptional changes. De-repression of LINE- 1 is evident in these placentas and elements of the de-repressed subfamily are associated with significantly downregulated genes. If retrotransposon de-repression is contributing to developmental defects by interfering with gene expression remains to be determined, however I identify a further possible mechanism leading to placental developmental defects. I determine that Tex19.1-/- placentas have an increased innate immune response and I propose that this is contributing to the developmental defects observed. Developmental defects and retrotransposon de-repression are also observed in spermatogenesis in Tex19.1-/- testes, the molecular basis for which is unclear. I therefore investigate the possibility that the TEX19.1 interacting partners, the E3 ubiquitin ligase proteins, may be contributing to the phenotypes observed in Tex19.1- /- testes. I show that repression of MMERVK10C in the testes is dependent on UBR2, alongside TEX19.1. Furthermore, I have identified a novel role for the TEX19.1 interacting partner, UBR5, in spermatogenesis, whose roles are distinct from those of TEX19.1. The work carried out during the course of this thesis provides mechanistic insights into TEX19.1 mediated genome defence and highlights the importance of protecting the genome from aberrant retrotransposon expression.