Contribution of Lsh to DNA methylation during neurodevelopment
Item statusRestricted Access
Embargo end date25/11/2020
Duthie, Leanne Beth
DNA methylation is considered one of the oldest epigenetic modifications and is associated with repression of transcriptional activity. Embryonic development sees dynamic changes in this mark. Initially, widespread demethylation occurs with the establishment of pluripotency in the developing zygote. This is then followed by a wave of de novo methylation in the implanting blastocyst, coincident with lineage specification, which continues until tissue specific patterns of DNA methylation are laid down. Disruptions in DNA methylation, and mutations in genes that form part of the DNA methylation machinery, are associated with neurodevelopmental disorders. Therefore, understanding the mechanisms of DNA methylation during neurodevelopment is key if we are to understand the pathology underlying these conditions. Investigating this has, thus far, proven difficult due to the lack of available models that survive into adulthood. Therefore the consequences of dysregulated DNA methylation during neurodevelopment on the mature brain are unknown. The DNA methyltransferase enzymes are responsible for the deposition of DNA methyl marks, although less is known about the cofactors required for their targeting and activity. Lsh (lymphoid specific helicase), a chromatin remodeller, has been described to play a key role in de novo methylation during development. The importance of this protein is highlighted by the fact that knockout mice die within a few hours of birth. Furthermore, whilst I was pursuing this project, Thijsen et al described mutations in this gene being causative for Immunodeficiency, Centromeric instability and Facial anomalies syndrome (ICF), in which a large proportion of patients suffer from intellectual disability (Thijssen et al., 2015). At the beginning of this project, nothing was known about the role of Lsh during neurodevelopment, nor why mutations in this gene should result in neurological defects. This thesis aims to determine the contribution of Lsh to DNA methylation during neurodevelopment and to investigate the consequences of its absence on the mature brain. In order to investigate the roles of Lsh at early stages of neurogenesis I made use of an in vitro neurodifferentiation system to investigate the differentiation of Lsh-/- mouse embryonic stem cells down neural lineage. This revealed an enhanced differentiation of these cells down neural lineage compared to wild type. Genome-wide methylome analysis uncovered a key role for Lsh in establishing appropriate DNA methylation at repetitive sequences during this developmental window. Previous papers, investigating selected loci, have suggested a role for Lsh in methylating promoters of single copy genes thereby controlling their transcription. By investigating transcription genome–wide, I noted a key role for Lsh in regulating genes associated with developmental processes. This regulation was not, however, related to their promoter methylation status. This led me to investigate the regulation of the Polycomb system in these cells, another key repressive epigenetic system involved in developmental gene transcriptional regulation. This analysis uncovered wide-scale redistribution of the polycomb mark H3K27me3 from target genes to hypomethylated repeats, similar to what is seen in other hypomethylated models. Whilst this could account for the dysregulation of developmental gene expression, I also discovered preliminary evidence of a role for Lsh in regulating transcription at these sites independent of its role in DNA methylation. To investigate the role of Lsh in vivo, I used a novel Nestin-Cre knockout mouse model. This new targeted mouse model, unlike previous models, survived to adulthood. This allowed investigation of the effects that disruption of DNA methylation during development has on the mature brain. This mouse displayed no gross changes in brain morphology or behaviour. DNA methylation analysis revealed gross hypomethylation at repeat sequences particularly in the knockout cerebellum. A consequence of this was aberrant repeat transcription. RNASeq revealed activation of innate immune response genes. This led me to propose a model by which aberrant repeat transcription results in an immune response due to cellular detection of cytoplasmic double stranded DNA generated via reverse transcription. This opens up exciting new avenues of research into the underlying pathology of the neurological deficits seen in ICF syndrome and potential therapeutic options.