We have identified the repressive histone variant macroH2A1.2 as a critical modulator of BRCA1-dependent genome maintenance during DSB repair via homologous recombination (HR) (Khurana et al., Cell Reports, 2014). In this project, we aim at a detailed characterization of this novel epigenetic effector of genome maintenance in health and malignancy. Given that both HR and BRCA1 function are required for the efficient resolution of stalled and/or collapsed replication forks, we focused on macroH2A1.2 function in this process, invoking chromatin as a paradigm for the manipulation of the cellular response to replication stress. Using chromatin immunoprecipitation combined with deep sequencing (ChIP-Seq) in K562 erythroleukemia cells, in which both fragile sites and DNA replication patterns have been extensively characterized, we found that macroH2A1.2 preferentially localizes to sites of replication stress-induced DNA damage. Notably, macroH2A1.2 peak coverage was most prominent at common fragile sites and was further positively correlated with CFS susceptibility to DNA breaks. Consistent with an active role during replication stress, we observed a fragile-site specific increase in macroH2A1.2 beyond its basal level of enrichment, which required DNA damage signaling and concomitant H2AX phosphorylation to coordinate FACT histone chaperone-dependent deposition of macroH2A1.2. MacroH2A1.2, in turn, facilitates the accumulation of the tumor suppressor and HR effector BRCA1 at replication forks to protect from replication stress-induced DNA damage. The concomitant fragile site-associated chromatin reorganization is a driver of progressive epigenetic change, particularly in the context of replicative age. Specifically, we observed a robust, replication-dependent increase in macroH2A1.2 at CFSs in primary fibroblasts upon extended culture. Consistent with the notion that replication stress and the resulting DNA damage response (DDR) are important drivers of cellular senescence in primary cells, which in turn counteracts malignant transformation, we show that loss of macroH2A1.2 can cause a DDR-dependent, near complete cell cycle arrest in primary fibroblasts associated hallmarks of cellular senescence. Together this work establishes macroH2A1.2 as a bona fide epigenetic modulator of replication stress with implications for age-associated epigenetic change, genome integrity and malignant transformation. These findings have recently been published (Kim et al., Mol Cell 2018). A role for macroH2A1.2 during malignancy appears particularly likely in the context of tumors that rely on alternative lengthening of telomeres (ALT) for their growth. ALT is a homology-directed telomere maintenance pathway, which is tightly linked to increased telomeric replication stress as a source for the underlying DSBs. ALT telomeres furthermore exhibit a unique chromatin environment and generally lack the nucleosome remodeler ATRX, which was previously found to modulate macroH2A1.2 chromatin occupancy. As part of their unique chromatin composition, we found ALT telomeres to be highly enriched for macroH2A1.2, consistent with their inherent susceptibility to replication stress. However, in contrast to ATRX-proficient cells, ALT telomeres transiently lose macroH2A1.2 during acute fork stalling to facilitate DSB formation, a process that is almost completely prevented by ectopic ATRX expression. Telomeric macroH2A1.2 is re-deposited in a DNA damage response-dependent manner to promote HR-associated ALT pathways. Our findings thus identify the dynamic exchange of macroH2A1.2 on chromatin as an epigenetic link between ATRX loss, ALT initiation via replication stress and ALT execution via HR. MacroH2A1.2 may, thereby, provide a novel therapeutic target in ALT-dependent cancers. This work was performed in collaboration with Dr. Yie Liu at the National Institute on Aging and has been published this year (Kim et al., Nature Structural and Molecular Biology, 2019). Of importance for our understanding of macro-histone biology, macroH2A1.2 represents one of two alternative and mutually exclusive splice variants of the macroH2A1-endoding H2AFY gene. Based on data from us and others, these two variants, macroH2A1.1 and macroH2A1.2, have seemingly opposing roles during DNA repair and cell growth, and the elucidation of splice variant-specific macroH2A1 functions is, thus, an important aspect of our ongoing research program. As part of these efforts, we have generated a macroH2A1.2 variant-specific knockout mouse. Our initial analyses have uncovered an unexpected role for macroH2A1.2 in chromosome stability specifically at the inactive X chromosome (Xi). The Xi is inherently susceptible to genomic aberrations. Replication stress has been proposed as an underlying cause, but the mechanisms that protect from Xi instability remain unknown. Here, we show that the replication stress-protective macroH2A1.2 histone variant, which is enriched on the Xi, is required for Xi integrity and female survival. Mechanistically, macroH2A1.2 counteracts its equally Xi-enriched alternative splice variant, macroH2A1.1. Comparative proteomics revealed macroH2A1.1-specific association with alternative end-joining (alt-EJ) mediators that promote anaphase defects in the absence of macroH2A1.2. Depletion of macroH2A1.1 or alt-EJ factors reverses genomic instability upon macroH2A1.2 loss, and mice deficient for both variants harbor no overt female defects. Together, our findings uncover splicing-regulated genome maintenance as a long sought-after rationale for macroH2A1 accumulation on the Xi, with implications for aneuploidy-associated malignancies. This work is currently being submitted. Altogether, this project is designed to uncover the contribution of macro-histones to genome and epigenome integrity both in tumor cell lines and in animals. Given our published and ongoing research, we anticipate that these histone variants will emerge as critical and potentially druggable modulators of genome integrity, particularly in replicating cells, with consequences for both (stem) cell growth and malignant transformation.
