The silent guardian: unraveling the roles of H3K9me3 in genome maintenance

来源: 发布时间:2025-02-25 08:56:29 浏览次数: 【字体:

Review Article

Published: 02 July 2024

Zhiming Li & Zhiguo Zhang

Volume 5, pages 133–153, (2024)

Abstract

Genomic integrity is a fundamental prerequisite for all living organisms, ensuring the accurate transmission of genetic information and stability of cellular functions across generations. The maintenance of genome integrity relies on a meticulously orchestrated network encompassing a variety of DNA repair factors and pathways. In eukaryotic cells, genomic DNA is assembled into chromatin, a highly organized complex of proteins and DNA. Therefore, chromatin and epigenetic factors have emerged as crucial guardians against genotoxic stress. In this review, we provide a comprehensive summary of the multifaceted roles of H3K9me3, a histone mark associated with silent chromatin, in DNA damage repair and genome maintenance, elucidating its dynamic participation in chromatin organization, silencing of repetitive DNA sequences, and modulation of DNA repair pathways. Importantly, we discuss the potential contribution of H3K9me3 to epigenetic memory following DNA damage, which introduces an additional layer of complexity to our comprehension of genomic surveillance. Finally, we explore the implications of H3K9me3 dysregulation in human cancers and the promising therapeutic avenues that may arise from a nuanced understanding of its function in maintaining genomic stability.


Introduction

The integrity of the genetic material is constantly threatened by a myriad of endogenous and exogenous factors, such as ionizing radiation, ultraviolet light, chemical mutagens, and errors during DNA replication, leading to various forms of DNA damage and even cell death if left unattended. To counteract these assaults, cells have evolved highly sophisticated and tightly controlled DNA damage repair pathways, which form a dynamic network of hundreds of factors to safeguard the fidelity of the genome (Giglia-Mari et al., 2011; Sancar et al., 2004; Scully et al., 2019). The regulation of DNA damage repair is governed by cellular surveillance mechanisms, which can detect DNA lesions and activate DNA damage checkpoints that temporarily halt the cell cycle, providing time for repair before cell division (Branzei & Foiani, 2008; Cimprich & Cortez, 2008; Marechal & Zou, 2013; Sancar et al., 2004). Depending on the types of lesions, DNA damage is repaired through different pathways, including base excision repair (BER), nucleotide excision repair (NER), and double-strand break repair (DSBR) (Sancar et al., 2004). DSBs are the most cytotoxic due to their highly deleterious nature, which are typically repaired by homologous recombination (HR) and non-homologous end joining (NHEJ) pathways (Ceccaldi et al., 2016; Chapman et al., 2012; Scully et al., 2019; Shrivastav et al., 2008). Defects in these repair processes are associated with a range of human disorders, including cancer predisposition, aging and neurodegenerative diseases, underscoring the critical role of DNA damage repair in maintaining cellular health (Curtin, 2012; Hoeijmakers, 2009; Schumacher et al., 2021; Tubbs & Nussenzweig, 2017; Wilson et al., 2023; Zhao et al., 2023b).

In eukaryotic cells, the genomic DNA is packaged into chromatin, with the nucleosome consisting of 147 bp DNA wrapped around a histone octamer, as the basic repeating unit. Chromatin is further organized into different functional domains demarcated by various types of covalent histone post-translational modifications (PTMs). For example, histone acetylation, which creates a more relaxed chromatin architecture and facilitates the recruitment of transcriptional activators via acetyl-lysine recognizing bromodomains, promotes the assembly of the transcriptional machinery, thus leading to increased gene transcription (Grunstein, 1997; Struhl, 1998). Contrary to the permissive nature of histone acetylation, H3K27me3, a marker of facultative heterochromatin by the Polycomb complex PRC2, inhibits transcription by inducing chromatin compaction and promoting the loading of repressor complexes (Barski et al., 2007; Croce & Helin, 2013; Margueron & Reinberg, 2011; Mikkelsen et al., 2007).

It has become increasingly clear that chromatin regulators are critical players in coordinating cellular responses to DNA damage and maintaining genome stability (Hauer & Gasser, 2017; Lukas et al., 2011b; Soria et al., 2012; Yasuhara & Zou, 2021). Indeed, the dynamic and reversible nature of histone modifications extends their influence beyond transcription to encompass DNA damage repair and genome stability (Altaf et al., 2007; Cao et al., 2016; Kim et al., 2019). For example, one of the most prominent histone modifications involved in DNA damage repair is phosphorylation on serine 139 of H2AX, a histone H2A variant, namely γ-H2AX (Fernandez-Capetillo et al., 2004; Mah et al., 2010). In response to DSBs, γ-H2AX is rapidly installed by PI3K-like kinases, including ATM, ATR, and DNA-PK, which serves as an anchor point for the subsequent cascade of response and repair factors that are enlisted to resolve the DNA lesions, covering megabase (Mb)-sized genomic regions (Blackford & Jackson, 2017; Li et al., 2020a). H4K20me2, an abundant histone mark, also plays key roles in dictating cellular response to DSBs (Jorgensen et al., 2013; Kohi et al., 2022). In particular, H4K20me2 forms a bivalent chromatin docking site with H2AK15 ubiquitylation that recruits 53BP1 (p53 binding protein 1), a crucial component in DSB signaling and repair choices (Botuyan et al., 2006; Fradet-Turcotte et al., 2013; Mattiroli et al., 2012; Pei et al., 2011; Stewart et al., 2009; Wilson et al., 2016). Therefore, gaining a deeper comprehension of how different histone modifications participate in the response and repair of DNA damage holds the potential to enhance our knowledge of maintaining genome integrity and identifying possible targets for therapeutic interventions in diseases.

In this review, we will focus on the various mechanisms employed by H3K9me3 to modulate DNA damage response/repair and maintain genome integrity (Fig. 1). Among the vast array of histone modifications that have been extensively delineated, H3K9me3 is a key repressive mark at constitutive heterochromatin that is composed of mainly repetitive DNA sequences. In addition, H3K9me3 is increasingly acknowledged for its versatile involvement in numerous cellular processes, such as embryonic development and stem cell differentiation, through both classic and alternative mechanisms. For example, H3K9me3 decorating a subset of cell-type specific genes, which are subjected to profound rearrangements and lose H3K9me3 signals during lineage commitment to allow for their transcriptional activation (Nicetto & Zaret, 2019; Nicetto et al., 2019), marks a pivotal role of H3K9me3 in cell fate determination. Ectopic H3K9me3 deposition by forced expression of SUV39h enzymes leads to compromised differentiation and epigenetic reprogramming (Burton et al., 2020). In addition, the revelation of bivalent domains containing H3K9me3 and other active marks, such as H3K4me3 and H3K36me3 (Barral et al., 2022; Matsumura et al., 2015), further indicates that H3K9me3 functions in a chromatin context-dependent manner. Our recent study indicate that parental H3K9me3 asymmetrically distribute at head-on LINE elements at the leading strands during DNA replication (Li et al., 2023). These findings underscore the multilayered roles entrusted to H3K9me3 as active defenders of the eukaryotic genome.

Fig. 1
figure 1

Simplified schematic models of the roles of H3K9me3 in maintaining genome stability. A H3K9me3 in the establishment and maintenance of heterochromatin. B An example of H3K9me3 in the repression of transposable elements (L1). C H3K9me3 in the regulation of cellular response to DNA damage. D Potential involvement of H3K9me3 in bookmarking DNA breaks and forming cellular memory of DNA damage


H3K9me3 writers, erasers and readers

H3K9me3 stands out as the unmistakable hallmark of heterochromatin. Like many other histone methylation marks, H3K9me3 is regulated by its writers, erasers and readers. Early studies in Drosophila identified factors involved in suppressing the position effect variegation (PEV) [Su(var)], which were the first known lysine methyltransferases (KMTs) for H3K9 in human/murine (SUV39H1/Suv39h1), Drosophila [Su(var)3–9] and yeast (clr4) (Nakayama et al., 2001; Rea et al., 2000). Several more conserved H3K9 KMTs were later identified, including ESET/SETDB1, G9a/GLP and Prdm3/Prdm16 (Pinheiro et al., 2012; Schultz et al., 2002; Tachibana et al., 2001). Although there are some redundancies among these KMTs in cells, they generally have different substrate preferences, targeting genomic regions and binding partners (Padeken et al., 2022). For example, while both SUV39h1 and SETDB1 can catalyze H3K9 from mono-methylation to tri-methylation status in vivo, G9a and GLP are the primary enzymes for H3K9me1 and H3K9me2, which play vital roles in transcription silencing and embryogenesis (Shinkai & Tachibana, 2011; Tachibana et al., 20052008). Prdm3 and Prdm16, on the other hand, are only responsible for cytosolic H3K9me1 that aids in the establishment and reinforcement of heterochromatin mediated by SUV39h enzymes (Pinheiro et al., 2012). Cellular levels of H3K9me3 are balanced by the actions of lysine demethylases (KDMs), including the KDM4 subfamily (rph1 in yeast) that targets H3K9me2/3 and KDM3 that targets H3K9me1/2 (Whetstine et al., 2006; Yamane et al., 2006). KDM1A/LSD1 and KDM7 (KDM7A, PHF8 and PHF2) can also demethylate H3K9me1/2, albeit with weaker specificity (Metzger et al., 2005; Tsukada et al., 2010). By reversing the histone methylation marks installed by KMTs, these KDMs counteract the repressive effects imposed by H3K9 methylation and generally facilitate a more permissive chromatin environment (Greer & Shi, 2012; Kooistra & Helin, 2012). For example, H3K9 demethylation mediated by LSD1 activates androgen-receptor-dependent transcription of target genes (Metzger et al., 2005; Shi et al., 2004). KDM4D-dependent demethylation of H3K9me3 promotes the formation of pre-initiative complex, which is required for efficient DNA replication (Wu et al., 2017). Together, by installing and removing H3K9 methylation in a highly spatiotemporal manner, these KMTs and KDMs ensure its precise and dynamic distribution throughout the genome (Fig. 1A).

H3K9me3 recruits a number of readers, a dedicated group of effector proteins harboring methyl-lysine binding domains to exert its downstream functions. Among the growing list of readers, heterochromatin protein 1 (HP1) was the first to be identified to recognize and bind H3K9me3 through its N-terminal chromodomain, which is essential for heterochromatin formation and transcription repression (Bannister et al., 2001; Jacobs et al., 2001; Lachner et al., 2001; Nakayama et al., 2001). HP1 in turn recruits SUV39h1, which spreads H3K9me3 to flanking nucleosomes, forming a positive feedback loop that consolidates the repressive chromatin domain over a long stretch of genomic DNA (Aagaard et al., 1999; Bannister et al., 2001). HP1 chromatin loading is further dynamically regulated by the mitotic H3K9/S10 ‘methyl/phospho switch’ (Fischle et al., 2005; Hirota et al., 2005). UHRF1 (ubiquitin-like PHD and RING finger domain-containing 1) is another well-characterized H3K9me3 reader through its tandem Tudor domain, which is required for the maintenance of DNA methylation by DNMT1 (DNA methyltransferase 1) (Bostick et al., 2007; Liu et al., 2013; Rothbart et al., 2012; Sharif et al., 2007). Later studies suggest that DNMT1 can directly read H3K9me3 through its RFTS (replication foci targeting sequence) domain, which boosts DNMT1 activity stimulated by H3 ubiquitylation and faithful inheritance of DNA methylation patterns (Ren et al., 2020). A recent study revealed that TNRC18 (trinucleotide-repeat-containing 18) reads H3K9me3 through its C-terminal BAH domain, which is required for silencing of several families of endogenous retroelements (Zhao et al., 2023a). In addition, H3K9 methylation has also been reportedly recognized by plant homeodomain (PHD) fingers (Mansfield et al., 2011; Tencer et al., 2017), WD40 domain (Chan & Zhang, 2012), MBT domain (Klymenko et al., 2006) and ankyrin repeats (Collins et al., 2008). Interestingly, many H3K9me3 writers also contain these functional domains (Zhang et al., 2015a). For example, SUV39h1/2 (Clr4 in yeast) contains a chromodomain at its N terminus that binds H3K9me3, which is required for SUV39h/Clr4 enzymatic activity, H3K9me3 propagation and stable inheritance (Al-Sady et al., 2013; Melcher et al., 2000; Shirai et al., 2017; Wang et al., 2012; Zhang et al., 2008). These results suggest a direct ‘read-write’ mechanism utilized by H3K9me3 writers for its precise restoration of H3K9me3 on nucleosomes formed with newly synthesized histones using parental and existing H3K9me3 as templates following DNA replication. Importantly, the interplay between H3K9me3 readers and enzymes that regulate other chromatin marks creates an opportunity for epigenetic crosstalk between H3K9me3 and other histone modifications (Lee et al., 2010; Suganuma & Workman, 2008). For example, H3K9me3-anchored HP1 recruits SUV4-20h enzymes, which generate H4K20me3, another characteristic heterochromatic mark (Schotta et al., 2004).

H3K9me3 in the silencing of repetitive DNA sequences

A substantial portion of the eukaryotic genome is made up of repetitive DNA elements, which are generally organized into heterochromatin (Consortium, 2012; Lander et al., 2001; Nurk et al., 2022). The formation of heterochromatic domains is essential to prevent homologous recombination and abnormal expression of these DNA repeats (Allshire & Madhani, 2018; Janssen et al., 2018). As a critical regulator of the establishment and maintenance of heterochromatin states, H3K9me3 plays an important role in maintaining the stability of these repetitive elements in the genome (Allshire & Madhani, 2018; Grewal & Jia, 2007; Janssen et al., 2018; Padeken et al., 2022). Indeed, epigenomic profiling of H3K9me3 reveals its enrichment at repetitive DNA sequences (Barski et al., 2007; Gershman et al., 2022; Mikkelsen et al., 2007; Roadmap Epigenomics et al., 2015). Consistent with the importance of H3K9me3 at repetitive DNA sequences, double knockout of SUV39h1 and SUV39h2, which predominantly target satellite repeats at the pericentric heterochromatin, leads to weakened heterochromatin formation, reduced viability and developmental potential, spermatogenic failure and heightened risk of malignant transformation (Peters et al., 2001). Notably, the overall organization and clustering of heterochromatin remain largely intact in these cells (Peters et al., 2001), likely due to the compensating roles of other H3K9 KMTs and H3K9me3-independent mechanisms (Allshire et al., 2018; Janssen et al., 2018; Padeken et al., 2022). Besides its fundamental roles in silencing DNA repeats, H3K9me3 also suppresses transposable elements that scatter throughout the genome, including euchromatin. For example, together with histone variant H3.3, H3K9me3, deposited primarily by SETDB1, potentiates the silencing of endogenous retroviruses (ERVs) and evolutionarily young long interspersed elements (LINEs) that are widely dispersed in the open chromatin compartment (Elsasser et al., 2015; Karimi et al., 2011; Liu et al., 2018; Matsui et al., 2010; Robbez-Masson et al., 2018; Tchasovnikarova et al., 2015; Yang et al., 2015). SUV39h-dependent H3K9me3 also selectively targets a group of intact LINEs for their transcriptional silencing (Bulut-Karslioglu et al., 2014). G9a and GLP, on the other hand, play a more specific role in repressing class III endogenous retroviruses (MERVLs) and MERVL LTR-driven genic transcripts in two-cell stage embryos (Maksakova et al., 2013). Finally, progressive loss of all the six methyltransferases associated with H3K9me3, including G9a/GLP, SUV39h1/SUV39h2 and SETDB1/SETDB2, completely abolishes H3K9me3, resulting in activation of almost all families of repeat elements, dissolution of heterochromatin and massive genomic instability (Montavon et al., 2021).

Of note, while the activities of retrotransposons pose a threat to genome stability, they are also required for normal embryonic development (Dopkins & Nixon, 2024; Guo et al., 2024; Senft & Macfarlan, 2021). Mechanistically, transposable elements rewire the core transcription network of stem cells by providing transcription factor binding sites, cis-regulatory elements, alternative splicing sites and non-coding RNA that regulate pluripotency genes (Bourque et al., 2018; Fueyo et al., 2022). During early stages of mammalian embryo development, H3K9me3 undergoes dynamic reprogramming and large-scale re-allocation to selectively tune the stage-specific expression of lineage-specific genes, long terminal repeats (LTRs) and hominoid-specific SVA (SINE-VNTR-Alu) elements, which is critical for lineage commitment and embryogenesis (Wang et al., 2018; Xu et al., 2022a; Yu et al., 2022). As a result, disrupting H3K9me3 by deletion of SETDB1 and other KMTs, histone chaperones (e.g., CAF1), or certain transcription regulators (e.g., DUX and ZFP51) caused varying degrees of faulty development and embryonic lethality (Dodge et al., 2004; Nicetto et al., 2019; Wang et al., 2018; Xu et al., 2022a). In addition, genomic regions resistant to epigenetic reprogramming by iPS or somatic cell nuclear transfer (SCNT) are enriched with repeat elements, which are marked by H3K9me3 (Becker et al., 2016).

In general, H3K9me3 is deposited to specific repetitive DNA sequences by its methyltransferases through sequence-specific regulators as well as local transcription. Particularly, low levels of transcription within heterochromatin, which produces small non-coding RNAs, such as small interfering RNAs (siRNAs) and PIWI-interacting RNAs (piRNAs), is crucial for the recruitment of silencing factors and subsequent heterochromatin formation (Castel & Martienssen, 2013; Motamedi et al., 2004; Reinhart & Bartel, 2002; Verdel et al., 2004). Importantly, these suppressors coordinate with H3K9me3 to achieve sequence-specific targeting and robust silencing. For example, KRAB-ZFP (Kruppel-associated box zinc finger protein) transcription factors can recognize the primer binding site (PBS) of ERVs and initiate their heterochromatic silencing by assembling the repressor complex containing HP1, SETDB1 and KAP1 (Geis & Goff, 2020; Yang et al., 2017a). One prominent example is ZFP809, which bridges KAP1/TRIM28 and the PBS sequences of integrated MLVs (murine leukaemia viruses) to mediate their transcriptional silencing in stem cells (Wolf & Goff, 20072009; Wolf et al., 2015). The HUSH (human silencing hub) complex, consisted of TASOR (transcription activation suppressor), MPP8 (M-phase phosphoprotein 8), and periphilin, is another active defender against repetitive DNA and plays an important role for the establishment of silencing at LINE elements (Muller & Helin, 2024; Seczynska & Lehner, 2023). MPP8 binds H3K9me3 through its chromodomain, which helps maintain HUSH at evolutionarily young and intact L1 retrotransposons, where HUSH tethers SETDB1 to deposit H3K9me3 and repress these L1s and intronless mobile elements (Douse et al., 2020; Liu et al., 2018; Prigozhin et al., 2020; Robbez-Masson et al., 2018; Seczynska et al., 2022; Tchasovnikarova et al., 2015). HUSH also recruits other cofactors, such as MORC2 (microrchidia CW-type zinc finger 2) and the NEXT (nuclear exosome targeting) complex to facilitate silencing of retroelements (Garland et al., 2022; Tchasovnikarova et al., 2017). As a result, HUSH depletion leads to widespread L1 activation and transcription deregulation of coding genes, which causes embryonic lethality and compromised stem cell self-renewal (Seczynska & Lehner, 2023). De-repression of L1s also generates double-stranded RNAs that induce interferon-stimulated genes and inflammatory reactions (Tunbak et al., 2020), which present novel therapeutic opportunities in anti-tumor immunotherapy (see more discussions below). Our recent results further suggest that HUSH cooperates with DNA polymerase Pol ε to mediate repression of L1 expression and retrotransposition during DNA replication (Li et al., 2023), revealing a role of the HUSH complex in the maintenance of L1 silencing. Importantly, the fact that only evolutionarily young and transcriptionally active L1s are targeted by HUSH suggests that HUSH represents an effective mechanism of genome surveillance against foreign nucleic acids (Liu et al., 2018; Robbez-Masson et al., 2018). Together, these studies underscore the central roles of H3K9me3 in defending against mobile DNA elements and maintaining genome integrity (Fig. 1B).

Direct involvement of H3K9me3 in DNA damage repair

One of the first pieces of evidence supporting the direct involvement of H3K9me3 in DNA damage response and repair comes from the observation that H3K9me3 is accumulated at the damage sites (Ayrapetov et al., 2014; Sun et al., 2009). Later studies further confirm the enrichment of several H3K9me3 methyltransferases, including SUV39h1, G9a, GLP and SETDB1, at damaged chromatin or DNA lesion sites (Alagoz et al., 2015; Ayrapetov et al., 2014; Ginjala et al., 2017; Li et al., 2020a; Lu et al., 2019; Yang et al., 2017b). By sensing the chromatin perturbations arising from DNA damage, H3K9me3 is either exposed by casein kinase 2 (CK2)-catalyzed phosphorylation and removal of HP1β or deposited to lesion sites within the first few seconds by SETDB1 and SUV39h1 to facilitate transient local chromatin compaction (Alagoz et al., 2015; Ayoub et al., 2008; Ayrapetov et al., 2014), which remodels the damaged chromatin to create an efficient template for the DNA repair machinery. Interestingly, such transient and dynamic H3K9me3 deposition is also recently reported as vital for the protection and restart of stalled replication forks during replication stress (Gaggioli et al., 2023), suggesting that H3K9me3-mediated heterochromatinization serves a more general purpose in safeguarding the genome from genotoxic stress.

While chromatin undergoes a swift cascade of dynamic epigenetic modifications upon DNA damage, H3K9me3 is regarded as a key signal to attract a multitude of factors essential for the initial response. For example, H3K9me3 recruits TIP60, a crucial histone acetyltransferase, through binding of its N-terminal chromodomain to DSB sites, by which TIP60 is supercharged to acetylate and activate ATM (Sun et al., 20052009), an apical kinase in DNA damage response. TIP60 also acetylates histones H4 and H2AX, promoting chromatin remodeling and γ-H2AX turnover, both of which are important for effective DNA repair (Ikura et al., 2007; Murr et al., 2006). In addition, a repressive complex comprising of KAP1, HP1 and/or SUV39h1 is loaded onto H3K9me3-marked DSB lesions through the chromodomain of HP1, which initiates a cycle that quickly propagates H3K9me3 across a long stretch of DNA flanking the damage sites (Ayoub et al., 2008; Ayrapetov et al., 2014; Baldeyron et al., 2011; Luijsterburg et al., 2009) (Fig. 1C). Intriguingly, efficient SUV39h1 recruitment and H3K9me3 deposition are later found to be contingent on TIP60 (Tu et al., 2020), or H4 monoufmylation by UFM1 specific ligase 1 (UFL1), a newly discovered histone PTM, and its reader serine/threonine kinase 38 (STK38) (Qin et al., 20192020). Upon the initial activation induced by chromatin compaction, ATM in turn phosphorylates KAP1, leading to the disassembly of the repressive KAP1/HP1/SUV39h1 complex, which restores DNA accessibility and facilitates the recruitment of additional repair factors, thereby establishing a negative feedback loop (Ayrapetov et al., 2014).

Besides its well-illustrated role in the TIP60-ATM axis, H3K9me3 is also instrumental in the pathway choices to resolve DSB, which are governed by 53BP1 and BRCA1, critical factors in DNA repair mediated by non-homologous end joining (NHEJ) and homologous recombination (HR), respectively (Panier & Boulton, 2014). While not a direct reader of H3K9me3, 53BP1 recruitment to the DNA lesions is indirectly regulated by H3K9me3. For example, TIP60, stimulated by H3K9me3, acetylates histone H2A and H4, which prevents 53BP1 loading to damaged chromatin through its bivalent recognition of histone modifications, thus favoring the HR pathway (Jacquet et al., 2016; Tang et al., 2013). However, later studies confirmed the interaction and co-localization between 53BP1 and H3K9me3 in cells (Zhang et al., 2022), suggesting a more intertwined relationship. In addition, chromatin condensation created by the cooperation between H3K9 methyltransferases, histone variant macroH2A1 and HP1 facilitates the recruitment and retention of BRCA1, which antagonizes the actions of 53BP1 and promotes DNA end resection (Khurana et al., 2014; Lee et al., 2013; Wu et al., 2015). These reports resonate with findings that heterochromatic DSBs, while complicated due to the limited accessibility and high repetitiveness, are largely repaired by HR (Chiolo et al., 2011). However, it should be noted that the NHEJ pathway is preferentially utilized for DSB repair in both heterochromatin and euchromatin when cells are mostly in G0/G1 and HR is inhibited (Janssen et al., 2016; Tsouroula et al., 2016). Moreover, the effects of H3K9me3-associated writes and readers in DSB repair choices can go both ways, exemplified by the different HP1 isoforms (HP1α/β promote HR while HP1γ is inhibitory) (Soria & Almouzni, 2013) and GLP/G9a (GLP promotes NHEJ while G9a favors HR) (Lu et al., 2019; Yang et al., 2017b), through varying mechanisms. These distinctions, likely stemming from the differences in genetic backgrounds, DNA damage types, local chromatin context of damage sites, availability of repair factors and/or cell cycle stages, highlight the importance to examine the role of H3K9me3 in DNA damage repair from a context-dependent perspective.

Heterochromatin constitutes a structural obstacle to efficient DNA repair by hindering the access of repair factors to damage sites (Hauer & Gasser, 2017; Soria et al., 2012), which necessitates the timely removal of H3K9me3 once it has accomplished its mission. In fact, most damage-elicited epigenetic marks are temporary and compromising their reversal generates sustained DNA damage response (Polo & Jackson, 2011), which is detrimental to cell survival. For example, deacetylation of ATM by histone deacetylase SIRT7 is required for ATM dephosphorylation and deactivation, facilitating an orderly exit as soon as DNA repair is finished (Tang et al., 2019). Therefore, persistent ATM acetylation by SIRT7 deletion or acetylation-mimic point mutation sequesters the repair apparatus at the damage sites, preventing the seamless completion of DNA damage repair (Tang et al., 2019). Analogously, demethylation of H3K9me3 by its erasers, KDM4/JMJD2/JHDM3, is also consequential for DNA repair. Both KDM4B and KDM4D have been reported to be mobilized to DNA damage sites, which depends on the activity of poly (ADP-ribose) polymerase 1 (PARP1) (Khoury-Haddad et al., 2014; Young et al., 2013), suggesting a rather early engagement of H3K9me3 demethylation in this process. KDM4D-mediated H3K9me3 demethylation is required for efficient DNA replication and repair (Khoury-Haddad et al., 2014; Wu et al., 2017), both of which are important aspects of guarding the genome. The reports regarding KDM4A in DNA damage repair, however, have been conflicting (Fnu et al., 2011; Mallette et al., 2012), likely due to its dual ability to erase H3K9/K36 methylation and read H4K20me2. Altogether, these research efforts support a model whereby the dynamics of H3K9me3 at DNA damage sites not only configurates a restrictive local chromatin microenvironment, but also creates a docking platform for a number of repair factors in a highly coordinated manner.

Inheritance of epigenetic information following DNA damage repair

Given the disruptive and short-lived nature of the epigenetic changes triggered by DNA damage, it is counterintuitive to imagine that cells would form any type of memory of DNA damage. Previous epidemiologic studies in the offsprings of people working in or living close to nuclear power plants and survivors of the atomic bombs have generated inconsistent results (Gardner et al., 1990; Kodaira et al., 19952010; Little et al., 2013; Morton et al., 2021; Yeager et al., 2021). However, both short-term and long-term epigenetic memory, that is, chromatin signatures that can be preserved throughout mitosis and passed on to progeny cells independent of the initial assault, have been documented. One of the first clues came from a phenomenon termed radiation-induced bystander effect (RIBE), which refers to the chromosomal instability observed in non-exposed cells that are in close proximity or even distant from the site of initial radiation (Mothersill & Seymour, 2004). Nonirradiated hemopoietic stem cells that were transplanted into mouse bone marrow sustained genomic instability over many generations for up to 24 months, conceivably transmitted from their irradiated counterparts (Watson et al., 2000). While the underlying mechanisms remain enigmatic, altered DNA methylation, histone modifications and non-coding RNAs have been implicated in this process (Ilnytskyy et al., 2009; Koturbash et al., 2007). γ-H2AX, a widely used marker of DSB repair efficiency, is found to linger for up to several months (Siddiqui et al., 2015), well beyond when DNA lesions are normally expected to be remedied. Other repair factors, such as 53BP1, have also been reported to persist through mitotic cell divisions (Markova et al., 2007). Although it's possible that low doses of unresolved or irreparable breaks tolerated by cells lead to the persistence of γ-H2AX and its associated factors (Banath et al., 2010; Fumagalli et al., 2012), they provide evidence of relatively long-lived epigenetic perturbations caused by DNA lesions. Functionally, unprogrammed DSBs induce transcriptional silencing, instigated by DNA damage sensing kinases (ATM and DNA-PK) and PARP1 (Kruhlak et al., 2007; Shanbhag et al., 2010), which can be stably inherited for more than 30 passages in certain groups of cells. Within the complex array of factors at play, altered epigenomic profiles have been increasingly recognized to consolidate the repressive chromatin organization and contribute significantly to the heritable gene silencing patterns (Machour & Ayoub, 2020). These studies, in which the chromatin landscape needs to be more closely monitored following DNA damage, do suggest that the epigenome may not always be restored identically after the completion of DNA repair, probably reflecting the delicate balance between epigenetic fidelity and plasticity. This heritable epigenetic scar hence serves as a fingerprint that likely facilitates continuous surveillance against secondary chromosomal defects or future genotoxic events, resembling the immunological memory.

Among the DNA damage-induced epigenetic changes, histone modifications can bookmark chromatin status and persist over many generations, in a process termed epigenetic inheritance (Du et al., 2022). In both yeast and mammalian cells, pre-existing parental histones retain the positional memory after its transfer to replicating DNA for the formation of nascent chromatin, at least in closed chromatin regions (Escobar et al., 2019; Schlissel & Rine, 2019), an imperative for the re-establishment of functional epigenomes in the daughter cells. (Escobar et al., 2021; Stewart-Morgan et al., 2020). During mitotic cell divisions, parental histones are recycled in a strand-specific fashion, guided by POLE3-POLE4 and MCM2-Pol α, to the leading and lagging strands, respectively (Gan et al., 2018; Li et al., 2020b; Petryk et al., 2018; Xu et al., 2022b; Yu et al., 2018a). Disruption of these pathways, not surprisingly, alters DNA damage response and repair. For example, in yeast, mutations of these two pathways invariably impaired the error-free repair by HR (Dolce et al., 2022; Gonzalez-Garrido & Prado, 2023; Karri et al., 2024). The impact on HR is likely a result of unbalanced parental histone distribution at the nascent strands or increased free histones that cannot be properly deposited (Gonzalez-Garrido & Prado, 2023; Karri et al., 2024). In human cells, POLE3 depletion impedes both HR and NHEJ (Mao et al., 2023). Together, these studies hinted a link between histone modifications and epigenetic memory of DNA damage.

Potential role of H3K9me3 in DNA damage-induced epigenetic memory

The intertwined relationship between DNA repair and factors involved in epigenetic inheritance raises the possibility that cellular memory of DNA damage is stored by histone modifications. Indeed, a recent breakthrough study demonstrated that faithful DNA repair gradually erodes the epigenetic landscape, which advances aging at both the molecular and physiological levels (Yang et al., 2023). Using the ICE (inducible changes to the epigenome) system, non-mutagenic double-strand breaks (DSBs) introduced by the I-PpoI endonuclease were quickly repaired both in vitro and in vivo. Over time, this process initiates the loss of epigenetic information and ultimately drives mammalian aging, supporting the existence of damage-induced epigenetic memory. However, contrary to the discussions above, this information theory of aging portraits a different outcome of such memory, whereby the aberrant chromatin features exacerbate epigenetic instability and render cells more vulnerable to DNA damage, setting up a positive feedback loop that accelerates the aging process (Lu et al., 2023; Yang et al., 2023). Alternatively, one can argue that this mechanism, caught up in the dilemma of a genotoxic catastrophe, likely prevents other more deleterious consequences, such as apoptosis and carcinogenesis. Importantly, H3K9me3 plays a major regulatory role in the aging process driven by erosion of epigenetic integrity. Epigenetic alterations, including global loss of core histones and repressive chromatin marks, have been considered a hallmark of aging (Kane & Sinclair, 2019; Lopez-Otin et al., 2023). Catalytically inactive mutation of SUV39h1 recapitulates the syndromes observed in Werner syndrome (WS), an accelerated aging disorder, which can be rescued by overexpression of HP1 (Zhang et al., 2015b). Another groundbreaking study in C. elegans provides more compelling evidence to support the role of H3K9me3 as a mediator of DNA damage memory (Wang et al., 2023). In this elegant design, DNA damage induced lethality is inherited to the next generation produced from irradiated male and healthy female worms. Mechanistically, DNA damage-induced heterochromatin formation redirects the germline cells to utilize a more error-prone repair pathway, effectively overcoming the biological checkpoints that would typically pause embryonic development. Removal of linker histone (HIS-24) or heterochromatin protein-1 (HPL-1) in the male worms reduces H3K9me3 and enables the high-fidelity HR pathway, thus reversing the transgenerational embryonic lethality (Wang et al., 2023). Taken together, these results lay the groundwork for future studies to further delineate the role of H3K9me3 in bookmarking DNA damage (Fig. 1D).

It remains an open question whether and how H3K9me3- and heterochromatin-marked epigenetic memory of DNA damage exists in the long-term or in other higher eukaryotic species. As possibly the most stable histone modification, H3K9me3 has been proven to perpetuate through mitosis and meiosis in many systems and organisms (Du et al., 2022). Despite the lack of definitive evidence, there are, however, a few indications that we can speculate on. Firstly, H3K9me3 is continuously engaged in DNA damage response and repair, from the very beginning to beyond the completion of DNA repair, a prerequisite for H3K9me3 to pass on any heritable information. Secondly, besides its role in cell fate determination during development, H3K9me3 also drives the long-term effects of stress response, such as therapy- or oncogene-induced senescence that alters tumor progression and treatment outcomes (Braig et al., 2005; Di Micco et al., 2011; Dorr et al., 2013; Milanovic et al., 2018; Schleich et al., 2020; Yu et al., 2018b). Thirdly, H3K9me3 extensively marks repetitive sequences, which are more susceptible to endogenous damage associated with replication stress (Janssen et al., 2018). This susceptibility could lead to repeated cycles of H3K9me3 deposition, which protects these sequences and creates a permanent long-term memory through millions of years of evolution. Alternatively, due to the unique challenges posed by heterochromatin organization, H3K9me3 bookmarks the damaged loci, which allows them to be amended in a more confined nuclear compartment, or in a slower yet more accurate fashion, thus serving as short-term memory of DNA damage. In line with this hypothesis, increased mobility and relocation of DSBs (Hauer & Gasser, 2017; Lamm et al., 2021) and chromatin compartmentalization (Arnould et al., 20212023), have been observed. H3K9me3-enriched chromatin domains also exhibit slower kinetics of DNA repair (Adar et al., 2016; Han et al., 2016). Paradoxically, H3K9me3-occupied regions are characterized with increased mutation burden and decreased DNA repair by MMR (mismatch repair) and NER (nucleotide excision repair) in cancers (Supek & Lehner, 2015; Zheng et al., 2014). Finally, H3K9me3 positions the repressive chromatin domains at the back end of replication timing, the deregulation of which disrupts the epigenome and causes replication stress (Klein et al., 2021), leading to further chromosomal aberrations and DNA damage that can be inherited to the unstressed daughter cells (Techer et al., 2017). While several pathways have been proposed to control DNA damage provoked by disordered replication timing (Lukas et al., 2011a; Spies et al., 2019; Wu et al., 2023), it would be tempting to investigate how H3K9me3 takes part in this process.

H3K9me3 as an anti-tumor therapeutic target

Epigenetic alterations have been considered a crucial driver in the onset and progression of a growing number of human diseases, such as cancer, neurodegeneration, auto-immune diseases and aging-related disorders. In fact, epigenetic reprogramming is regarded as one of the hallmarks of cancer and targeting the cancer epigenome has emerged as a promising anti-tumor approach (Cheng et al., 2019; Davalos & Esteller, 2023; Hanahan, 2022; Suva et al., 2013; Zhao et al., 2021). The first of such attempts, employing inhibitors of histone deacetylases and DNA methylation, alone or in combination, dubbed ‘epigenetic therapy’, effectively and selectively kills a variety of malignant cells (Cameron et al., 1999; Yoo & Jones, 2006; Zhu & Otterson, 2003), and has paved the way for a handful of FDA-approved drugs for tumor treatment. Later reports of histone demethylases put histone methylation on the map and invigorated the efforts to target these marks that were originally thought to be irreversible (Shi et al., 2004; Tsukada et al., 2006). Recent landmark discoveries of recurrent oncogenic histone (onco-histone) mutations in pediatric glioma, chondroblastoma, sarcoma and several other specific cancer types further highlight the driving roles of the epigenetic paradigm in tumorigenesis (Behjati et al., 2013; Chan et al., 2013a; Fang et al., 2016; Lewis et al., 2013; Lu et al., 2016; Nacev et al., 2019; Papillon-Cavanagh et al., 2017; Schwartzentruber et al., 2012; Wu et al., 2012). Importantly, the genetic vulnerabilities and dependencies on particular epigenetic factors conferred by these onco-histones, such as chromatin remodelers and modifiers (Grasso et al., 2015; Hashizume et al., 2014; Mo et al., 2022; Nagaraja et al., 2017; Panditharatna et al., 2022; Piunti et al., 2017), present a unique therapeutic opportunity to cure some of these deadly tumors. While lysine-to-methionine substitutions at K9 have not been documented in cancer cohorts, introduction of H3K9M reprograms the global epigenome in mammalian cells (Brumbaugh et al., 2019; Chan et al., 2013b; Lewis et al., 2013), fruit fly (Herz et al., 2014) and fission yeast (Shan et al., 2016), in a manner akin to the K27M and K36M mutations.

Genome instability, another fundamental cancer hallmark (Hanahan, 2022; Hanahan & Weinberg, 20002011), in which H3K9me3 plays a central regulatory role, has put the spotlight on H3K9me3 and its associated factors for devising novel anti-tumor strategies. Profiling of cancer epigenomes suggests a global loss of H3K9me3 in some cancers, such as prostate, bladder, breast and colorectal cancer (Ellinger et al., 20102014; Leszinski et al., 2012; Zhao et al., 2016; Zhou et al., 2022), yet elevated levels of H3K9me3 have been documented in others, including liver cancer and liposarcoma (Keung et al., 2015; Qian et al., 2020), suggesting a context-dependent relationship. H3K9me3 levels are also correlated with the TNM stage, prognosis and recurrence in various cancer types (Benard et al., 2014; Chen et al., 2013; Elsheikh et al., 2009; Lam-Ubol & Phattarataratip, 2022; Muller-Tidow et al., 2010; Park et al., 2008; Song et al., 2012; Xia et al., 2013), acting as a valuable theranostic marker in anti-tumor therapy. Interestingly, almost all enzymes regulating H3K9me3, including both methyltransferases and demethylases, are reportedly oncogenic and highly druggable (Bhat et al., 2021; Husmann & Gozani, 2019; Morera et al., 2016). For example, overexpression of G9a is observed in many malignancies (Casciello et al., 2015). Gain-of-function mutations of G9a drive melanoma formation through activation of the WNT signaling pathway (Kato et al., 2020). Similarly, ectopically overexpressed SUV39h1 drives tumor growth and metastasis (Yokoyama et al., 2013). Upon DNA damage, SET7/9-mediated methylation of SUV39h1 attenuated its activity, which led to genomic instability and specific killing of tumor cells (Wang et al., 2013), offering new ideas to target H3K9me3 and SUV39h1 in cancer. Surprisingly, in Ras-driven lymphoma models, SUV39h1 keeps lymphomagenesis in check through H3K9me3-related cellular senescence and functions as a tumor suppressor (Braig et al., 2005), underpinning the intricate roles of H3K9me3 in determining the fate of damaged cells between senescence and oncogenesis. On the other hand, all members of the KDM4 demethylases are overexpressed in multiple tumors (Berry & Janknecht, 2013; Lee et al., 2020). In ovarian cancer, KDM4A amplification promotes re-replication and copy gain of specific genomic regions, through interacting with the replication machinery (Black et al., 2013). Inhibition of JAK2 and KDM4C, both of which are frequently amplified in a subgroup of lymphoma, induces cooperative epigenetic changes, including upregulation of H3K9me3 and formation of repressive chromatin, leading to MYC silencing and elimination of lymphoma cells (Rui et al., 2010).

An emerging aspect of the implication of H3K9me3 in cancer therapeutics lies in its prospects for dictating tumor response to immunotherapy. As a promising frontier in the battle against cancer, anti-tumor immunotherapy is a breakthrough of the century, yet serious challenges remain, including limited efficacy and low response rates in certain tumors, lack of personalized targets and immune escape (Hegde & Chen, 2020; Mellman et al., 2011; Pardoll, 2012). Recently, targeting of H3K9me3 and its enzymes has shown potential to overcome some of these hurdles. Inhibition of SUV39h1 activates repetitive elements that are otherwise silenced by H3K9me3, which stimulates the accumulation of cytosolic dsRNA and dsDNA, activates the antiviral pathways and interferon signaling, and enhances cancer cell immunogenicity (Shen et al., 2021). In multiple cancer models, genetic ablation of SUV39h1 markedly improves the long-term persistence and overall anti-tumor efficacy of human chimeric antigen receptor (CAR) T cells, protecting mice from tumor relapse and rechallenges (Jain et al., 2024; Lopez-Cobo et al., 2024). In melanoma, SETDB1-mediated epigenetic silencing represents a strategy for tumors to evade immune surveillance (Griffin et al., 2021). Loss of SETDB1 derepresses latent retroelements, immunostimulatory genes, and TE-encoded viral antigens, which further triggers cytotoxic T cell responses and enhances the anti-tumor potency of immune checkpoint blockade (ICB) (Griffin et al., 2021; Zhang et al., 2021a). Inhibition of G9a, or simultaneous targeting of G9a and other epigenetic repressors have been reported to enhance tumor immune clearance by ICB in bladder cancer (Segovia et al., 2019), melanoma (Kelly et al., 2021) and ovarian serous carcinoma (Spiliopoulou et al., 2022). In squamous cell carcinoma, deletion of KDM4A potentiates the effects of PD1 neutralizing antibodies by inflicting replication stress and DNA damage, which activates tumor-cell-intrinsic immune response (Zhang et al., 2021b). In colorectal cancer, KDM4D co-activates transcription factor SP-1 to enhance the interferon pathway and PD-L1 expression, which promotes tumor immune escape (Chen et al., 2022). These reports underpin the critical roles played by H3K9me3 in reshaping the tumor immune microenvironment.

The fact that blocking either H3K9me3 methyltransferases or demethylases is beneficial for tumor patients suggests that a healthy dose of H3K9me3 is essential to keep cancers at bay. On the one hand, high levels of H3K9me3 drives chromatin compaction and epigenetic silencing of tumor suppressor genes or retrovirus-derived regulatory elements, which enables cancerous transformation and suppresses innate immune response. Such mechanisms are common strategies of the so-called ‘immune-cold” tumors to bypass immune surveillance, as discussed above, or to develop drug resistance during therapy-induced epigenetic reprogramming (Torrano et al., 2019). On the other hand, aberrant downregulation of H3K9me3 creates a permissive chromatin environment that allows for the accumulation of spontaneous oncogenic activation and chromosomal abnormalities. These mutations further grant growth advantage during successive clonal expansion and eventually result in tumorigenesis. The narrow window of H3K9me3 fluctuations underlines the tight control over its levels by a plethora of coordinating factors. Under pathological conditions, such duality agrees with the reversibility and plasticity of the epigenetic framework, which opens up avenues for developing targeted and more personalized disease-prevention tactics to treat cancers and beyond.

Concluding remarks

In summary, H3K9me3, a master regulator of heterochromatin, has also been proven multi-talented in DNA damage response and repair, which bears major implications for stress response, the etiology of human diseases and their clinical management. Within the cascade of molecular events upon DNA damage, H3K9me3 fine-tunes the response and repair process through an expanding array of reader proteins, in the context of chromatin microenvironment and potential crosstalk with other epigenetic marks. The spatiotemporal mechanisms underlying the dynamics of H3K9me3 and the ensuing chromatin rearrangements remain ambiguous, the understanding of which could yield conceptual advancement in the epigenetic regulation of genome integrity. The emerging perspectives of H3K9me3 in retaining and transmitting short-term or long-term cellular memory to stress response position it as a multifaceted molecular switch of cell fate transitions, including differentiation, programmed cell death, senescence, carcinogenesis, and others. These outstanding questions warrant continuous efforts to dissect the mechanistic details and functional significance in both physiological and pathological settings. Importantly, the translational and pre-clinical developments aimed at enzymes and effectors in connection with H3K9me3 have been promising, either alone or combining with other anti-tumor regimes. H3K9me3, therefore, plays the role of a silent guardian to preserve genetic and epigenetic stability against genotoxic stress.

References


Acknowledgements

The authors apologize for not being able to include all relevant literature due to space limitations. Z. L. and Z. Z. wrote the manuscript. This work is supported by NIH grants R35GM118015 (to Z. Z.).

Author information

Authors and Affiliations

  1. Institute for Cancer Genetics, Columbia University Irving Medical Center, New York, NY, 10032, USA

    Zhiming Li & Zhiguo Zhang

  2. Herbert Irving Comprehensive Cancer Center, Columbia University Irving Medical Center, New York, NY, 10032, USA

    Zhiming Li & Zhiguo Zhang

  3. Department of Pediatrics, Columbia University Irving Medical Center, New York, NY, 10032, USA

    Zhiming Li & Zhiguo Zhang

  4. Department of Genetics and Development, Columbia University Irving Medical Center, New York, NY, 10032, USA

    Zhiming Li & Zhiguo Zhang

Corresponding author

Correspondence to Zhiguo Zhang.

Ethics declarations

Conflict of interest

The authors declare no competing interests.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Li, Z., Zhang, Z. The silent guardian: unraveling the roles of H3K9me3 in genome maintenance. GENOME INSTAB. DIS. 5, 133–153 (2024). https://doi.org/10.1007/s42764-024-00131-x

Download citation

  • Received

  • Revised

  • Accepted

  • Published

  • Issue Date

  • DOIhttps://doi.org/10.1007/s42764-024-00131-x

Share this article

Anyone you share the following link with will be able to read this content:

Provided by the Springer Nature SharedIt content-sharing initiative

Keyords



About this journal

Electronic ISSN
2524-7662
Abstracted and indexed in
  1. CNKI
  2. Dimensions
  3. EBSCO Discovery Service
  4. Google Scholar
  5. Institute of Scientific and Technical Information of China
  6. Meta
  7. Naver
  8. OCLC WorldCat Discovery Service
  9. ProQuest-ExLibris Primo
  10. ProQuest-ExLibris Summon
  11. TD Net Discovery Service
Copyright information

Rights and permissions

Springer policies

© Shenzhen University School of Medicine; Fondazione Istituto FIRC di Oncologia Molecolare

×

用户登录