Mechanisms of Histone Serotonylation and MLL5 in Coordinating the Epigenome during Neurodevelopment

Icahn School Of Medicine At Mount Sinai, New York NY

Investigators

Abstract

Project Summary A plethora of studies over the past 30+ years have demonstrated that lack of proficient epigenetic regulation contributes significantly to neurodevelopmental disorders (NDDs). Particularly, histone H3 lysine 4 tri-methylation (H3K4me3) regulation has been found to be aberrantly regulated in multiple NDDs, including autism spectrum disorders (ASD). Of note, dysregulation of H3K4me3 domains was found in post-mortem brain of ASD patients, with spreading of “broad” domains (>5 kb) noted. H3K4me3 broad domains have previously been linked to expression of cell type specific genes, such as those involved in synaptic signaling. This spreading has been linked to lysine methyltransferase 2e (KMT2E, or MLL5), which contains a plant homeobox domain (PHD) that binds to H3K4me3, as well as a SET domain (lacking methyltransferase activity). Over 60 individual patients have been identified with heterozygous mutations in Kmt2e, who have symptoms related to intellectual disability, developmental delay, and epilepsy. H3K4me3 is adjacent to H3 glutamine 5, which can be serotonylated (H3Q5ser), creating a dual modification H3K4me3Q5ser. This combinatorial mark has been shown to enhance permissive transcription of target genes compared to H3K4me3 alone. Preliminary data shows H3K4me3Q5ser enhances significantly the binding of MLL5PHD versus H3K4me3 alone. ChIP-sequencing of H3K4me3Q5ser in embryonic forebrain tissues identified ~4000 H3K4me3Q5ser broad domains which are enriched at genes involved in neurodevelopment-associated processes. Kmt2e-/- mice exhibit significant behavioral alterations, which match known developmental defect phenotypes of patients with KMT2E mutations. Immunoprecipitation- mass spectrometry of MLL5 revealed strong interactions with the nuclear co-receptor/histone deacetylase 3 (NCOR/HDAC3) complex, as well as lysine demethylase 5a (KDM5A, or JARID1A), which regulate H3/H4ac and H3K4me3 respectively. Thus, I hypothesize that H3K4me3Q5ser recruits MLL5 to key developmental loci during neurodevelopment, secondarily recruiting the NCOR/HDAC3 complex and JARID1A, regulating H3K4me3Q5ser broad domains, allowing for appropriate transcriptional programming in brain. First, I will characterize the molecular interactions between H3K4me3Q5ser, MLL5, NCOR/HDAC3, and JARID1A in cellulo, assessing key protein domains for direct protein:protein interactions and the requirement of MLL5 for co-targeting genome wide. Secondly, using our Kmt2e-/- mouse, I will assess the impact of MLL5 loss on histone modification landscapes in the developing brain, across both embryonic and postnatal development and intersecting this with transcriptomic profiling. Finally, I will assess the impact of Kmt2e on neurodevelopment by attempting to virally ‘rescue’ proper gene expression in Kmt2e-/- mice during postnatal development using viral expression, measuring changes in H3K4me3Q5ser levels and behavioral phenotypes versus control animals. Together, these aims will provide valuable insights into this novel epigenetic mechanism which regulates neurodevelopment, which is perturbed in patients with NDDs which have mutations in one of the aforementioned genes.