Histone Tail Mutants: Versatile Tools for Decoding Chromatin, Development, and Disease

Abstract

Histone modifications have been associated with transcriptional regulation, development, and disease, yet their direct functional roles remain incompletely understood due to the redundancy and promiscuity of cognate histone-modifying enzymes. Experimental strategies based on cis- or trans-acting histone mutants, including lysine-to-methionine (K-to-M) substitutions of histone H3 and other cancer-associated onco-histone variants, enable direct interrogation of individual chromatin marks and circumvent potentially confounding effects of histone enzyme knockouts. Here, we review how histone mutants have been leveraged to uncover novel principles by which chromatin regulation governs physiological, pathological, and experimental cell fate transitions and discuss the ways in which these discoveries could be exploited in the future for therapeutic benefit.

Brief History of Histone Lysine Methylation Research

Discovery of histone-modifying enzymes and their functional complexity

The concept that DNA is packaged into chromatin (see Glossary) by histone proteins has long been appreciated, but the idea that histones themselves carry regulatory functions emerged more gradually. Since the 1960s, biochemical studies have identified covalent modifications on histones, including methylation, acetylation, phosphorylation, and ubiquitination [1, 2], with histone methylation (see Glossary) occurring on either lysine or arginine residues [3]. These early discoveries raised the possibility that posttranslational modifications (PTMs) of histones might directly influence gene expression, yet testing the functional significance of these modifications remained challenging [4]. The subsequent identification of site-specific methylation, most prominently on histone H3 lysine (K) 4, 9, 27, and 36, established the conceptual basis for understanding how histone methylation maintains active or repressed gene expression states [5, 6]. In this review, we will focus on histone lysine methylation, but we refer the reader to an excellent overview of arginine methylation [3, 7].

The discovery of enzymes that “write” and “erase”, and proteins that “read” specific types of histone lysine methylation provided another key conceptual advance and offered a crucial handle to probe the functional roles of individual chromatin marks [8–10]. For example, SET domains in histone methyltransferases were shown to write lysine methylation, while Jumonji C domains within demethylases were later shown to erase these marks, underscoring the dynamic and reversible nature of histone methylation [11, 12]. Similarly, reader proteins containing Tudor/PWWP domains and PHD fingers among others were subsequently shown to recognize methylated histones and recruit downstream effector complexes to regulate gene expression [13, 14].

Early knockout studies in mice demonstrated that many histone methyltransferases and demethylases are essential for mammalian development and tissue homeostasis, suggesting histone methylation is a physiologically critical mechanism to regulate cell identity and cell fate change [15, 16]. However, with increased maturity of the field, including common access to advanced genome modification tools such as CRISPR/Cas9, it became clear that the interpretation of phenotypes from histone-modifying enzyme (see Glossary) knockouts is more complex than initially assumed. This complexity stems from the growing insight that many chromatin regulators possess non-catalytic functions in addition to catalytic functions and these may differentially impact cellular function [17]. A case in point is the functional analysis of the mixed-lineage leukemia-1 (MLL1) enzyme associated with writing H3K4 trimethylation. While complete disruption of MLL1 results in profound hematopoietic defects in mice [18, 19], disruption of MLL1’s catalytic domain revealed its methyltransferase activity is in fact dispensable for hematopoiesis [20]. Indeed, recent evidence suggests that MLL1 protects cells against DNA damage via a methyltransferase-independent mechanism [21]. These and similar findings for other enzymes underscore the difficulty of attributing phenotypes solely to the loss of a specific histone modification.

Enzymatic redundancy is another factor that can complicate the functional interpretation of knockout phenotypes. In mammals, many histone methyltransferases and demethylases exist as families of proteins with overlapping substrate specificities. Enzyme redundancy reflects the increased complexity of mammalian tissues and is exemplified by the Polycomb Repressive Complex 2 (see Glossary) and its catalytic subunits Ezh1 and Ezh2, which both catalyze H3K27 methylation [22]. Accordingly, individual deletion of Ezh2 in the hematopoietic system produces relatively modest phenotypes while the combined deletion of Ezh1 and Ezh2 results in a fatal depletion of stem and progenitor cells [23–25]. Similarly, individual deletion of Suv39h1 or Suv39h2, which catalyze H3K9 trimethylation associated with heterochromatin, exerts only subtle defects in mice, whereas double-knockout animals exhibit impaired viability, chromosomal instability, and increased tumor susceptibility [26]. These examples highlight how redundancy of histone-modifying enzymes often masks the physiologic importance of specific methylation marks using conventional knockout approaches (Figure 1).

Fig.1: From Enzymes to Histone Mutants: A Conceptual Shift in Dissecting Chromatin Function.

Comparison of the advantages and disadvantages of histone-modifying enzyme knockout versus histone mutant approaches. Selected biological illustrations were created with BioRender.com.

Finally, accumulating evidence shows how chromatin-modifying enzymes “moonlight” by methylating or demethylating more than one histone substrate as well as non-histone substrates [17]. This promiscuity in substrate recognition can make it challenging to assign a given knockout phenotype to a specific histone modification or substrate without extensive follow-up experiments.

Mapping the chromatin landscape

Advances in next-generation sequencing technologies, coupled with reduced costs, have further transformed the field by enabling genome-wide chromatin mapping in rare cell populations. Early work using chromatin immunoprecipitation followed by sequencing (ChIP-seq) determined the distribution of classic histone marks in readily expandable cell lines and tissues. These studies revealed that trimethylation of histone H3 at lysine 4 (H3K4me3) is typically associated with active promoters, H3K27me3 with repressed promoters, H3K36me3 with active gene bodies, and H3K9me3 with constitutive heterochromatin, representing key canonical chromatin states across cell types [5, 15, 27–31].

The integration of diverse chromatin mark location data across distinct cell types, combined with base-resolution DNA methylation data, led to important insights regarding the collaborative and antagonistic relationships between individual epigenetic modifications [32]. For example, the integration of H3K4me3 and H3K27me3 binding data in pluripotent stem cells led to the identification of “bivalent domains” (see Glossary), which are regulatory regions simultaneously marked by both activation and repression associated histone modifications and thought to poise genes for gene expression during differentiation [33–35]. Although bivalent chromatin is now recognized to be more pervasive across cell types, these observations nevertheless provided a conceptual framework for understanding developmental plasticity and lineage commitment typically associated with stem cells [36]. The incorporation of chromatin accessibility datasets using Assay for Transposase-Accessible Chromatin using sequencing (ATAC-Seq) or Micrococcal Nuclease sequencing (MNase-Seq) further refined the functional interpretation of these chromatin states by revealing that open chromatin is positively associated with loci marked by active histone marks (e.g., H3K4me1/2/3, H3K27ac) as well as bivalent domains, leading to more precisely defined regulatory states across cell types and developmental transitions [32, 37]. Similarly, the comparison of DNA methylation and histone modification profiles led to the insight that certain histone modifications attract DNA methylation (e.g., H3K9me3, H3K36me3) while other modifications oppose DNA methylation (e.g., H3K4me3, H3K27me3), although exceptions to these rules have been observed in specific developmental and disease contexts such as the co-occurrence of H3K27me3 and DNA methylation in placental development and cancer [38, 39].

While the methylation of specific histone residues is generally associated with active or repressed transcriptional states, different methylation states within these residues provide additional nuance to gene regulation. For example, H3K4me1 is typically associated with active or poised enhancers (see Glossary), while H3K4me3 is associated with active or poised promoters [40]. Despite a body of literature correlating diverse chromatin states to gene expression or repression, the precise functional roles of individual and combinatorial histone modifications remain incompletely understood [36]. This is because conventional enzyme knockout strategies often lead to the disruption of multiple substrates, altered complex stability, and the activation of compensatory pathways, complicating the definition of how a single histone modification contributes to a specific cellular function. This conceptual gap in knowledge has motivated researchers to develop alternative strategies to directly examine histone residues.

Experimental histone mutants to study chromatin function

Experimental histone mutagenesis involves the replacement of modifiable amino acids such as lysines on the histone tail with residues that can no longer be modified such as arginines or alanines. This strategy has allowed researchers to uncouple the contribution of a specific histone modification from that of the cognate enzymes to specific cellular functions and phenotypes. While this approach has been extremely valuable, it is also quite cumbersome as it requires mutating in cis all copies of specific types of histones, which are typically encoded by multi-gene families.

Depending on the type of histone substitution, the effects can either be local by blocking methylation in cis at the mutated histones, or global by inhibiting methylation in cis and in trans (see section below). Moreover, different substitutions have distinct effects on the charge of the affected amino acid, which can affect nucleosome structure and the recruitment of chromatin-associated factors. For example, K-to-A mutations (e.g., H3K4A, H3K9A, H3K27A, H3K36A) eliminate both the modification site and positive charge [41, 42], while lysine-to-arginine (K-to-R) mutations (e.g., H3K27R) preserve the positive charge yet prevent the modification (acetylation and methylation) of histones [43].

Seminal studies in yeast used systematic lysine-to-alanine (K-to-A) histone mutagenesis to bypass enzymatic redundancy and directly test the function of specific histone residues [44, 45], while subsequent histone replacement approaches in Drosophila extended this approach to a metazoan model [46]. Substitution of lysine 27 with arginine across the ~100 canonical H3 gene copies per chromosome in Drosophila impaired Polycomb-mediated gene repression and developmental patterning, providing compelling evidence that disrupted H3K27 methylation itself underlies the developmental phenotype previously associated with enzymatic E(z) deficiency [47]. More recently, the Helin lab has in a tour de force extended this analysis to mammalian cells by replacing the lysines of all 28 histone H3 genes (H3.1, H3.2 and H3.3 variants) to non-methylatable arginines (K-to-R) using mouse embryonic stem cells (ESCs) [43]. Similar to Drosophila, this study confirmed that the loss of H3K27 methylation largely phenocopies PRC2 deficiency in a mammalian context. Together, these complementary experimental approaches have enabled the direct testing of the causal roles of histone modifications in cell identity, development, tissue maintenance, and disease.

In contrast to experimental K-to-A and K-to-R mutations, which act locally, experimental lysine-to-methionine (K-to-M) mutations (see Glossary) such as H3K9M dominantly inhibit cognate histone methyltransferases and thereby globally deplete the corresponding methylation mark, including at non-mutated histones [48]. Critically, K-to-M mutations have not only been leveraged as experimental tools, but they also occur naturally in certain types of cancer and therefore will be discussed in the subsequent section.

Onco-histones: Cancer-Associated Histone Mutations Redefine Chromatin Biology

Detection of onco-histones in human cancer

The discovery of recurrent histone mutations at residues associated with methylation in human cancer provided compelling support for the histone code hypothesis (see Glossary), which posits that combinatorial histone modifications instruct distinct gene expression states. In 2012/2013, genomic analyses of pediatric high-grade gliomas first identified lysine-to-methionine (K-to-M) substitutions at lysine 27 of the histone H3 (H3K27M) gene as the defining alteration in diffuse intrinsic pontine glioma (DIPG) patients (Figure 2A) [49–52]. Biochemical data revealed that H3K27M dominantly inhibits Polycomb Repressive Complex 2 (PRC2), resulting in a profound global reduction of H3K27me2/3 levels [49]. While the precise mechanisms by which specific onco-histones (see Glossary) suppress methylation are still being investigated and may vary for different mutations, existing data suggest that many onco-histones have higher affinity to cognate methyltransferases than non-mutated counterparts [49, 53], resulting in their sequestration and failure to spread methylation genome-wide. Following the initial discovery of H3K27M mutations in DIPG, sequencing efforts identified a host of additional recurrent histone mutations across a wide range of cancer types, including H3K4M in head and neck cancer, H3K36M in chondroblastoma and sarcomas, H3G34R/V in pediatric gliomas, and mutations in linker histone H1 variants in lymphoid malignancies (Figure 2A) [52, 54–60]. It is important to mention that not all onco-histones inhibit methylation in a dominant fashion, as exemplified by the H3G34R/V mutation, which locally disrupts H3K36 and DNA methylation [61]. Moreover, onco-histones are not confined to the classic K-to-M mutations at the histone tail but also comprise point mutations within the globular domain and acidic patch of histones, which perturb other aspects of chromatin signaling such as chromatin accessibility [56–58]. Overall, we still have a rather poor understanding of the scope of chromatin processes and cellular functions impacted by these methylation-independent mutants and most of our current knowledge stems from the study of K-to-M onco-histones discussed below.

Fig.2: Onco-histones and Epigenetic Rewiring in Disease.

(A) Representative onco-histones (H3K27M, H3K36M, and H3G34R/V/W/L) in cancer.

(B) H3K27M gliomas exhibit global DNA hypomethylation, which may cooperate with reduced H3K27me3 levels to activate oncogenic transcription. In H3K36M chondroblastoma, loss of H3K36 methylation is accompanied by redistribution of antagonistic H3K27me3 levels. G34 mutations sterically impair SETD2 binding, selectively blocking H3K36 methylation on mutant histones. Selected biological illustrations were created with BioRender.com.

Dominant inhibition of methylation by K-to-M histone mutations

A common principle underlying the function of onco-histones containing K-to-M substitutions is their dominant-negative effect on histone methylation (Figure 1). Methionine is structurally similar to lysine but cannot be methylated. When incorporated into chromatin, K-to-M mutant histones engage their cognate methyltransferases and inhibit catalytic activity through sequestration. In the case of H3K27M, the mutant tail binds to the catalytic PRC2 component EZH2, suppresses its enzymatic function, and thus prevents the spreading of methylation, leading to a global reduction of H3K27 methylation [49]. Analogously, H3K36M inhibits the SETD2 trimethyltransferase and the NSD family of dimethyltransferases, causing a widespread reduction of H3K36me2 and H3K36me3 levels and consequently a redistribution of the antagonistic mark H3K27me3 (see also section on epigenetic rewiring below) [54, 55]. Importantly, incorporation of only a small fraction of mutated histones (5–10%)–in the case of H3.3 K-to-M mutants–is sufficient to deplete methylation genome-wide including at unmutated H3.1, H3.2 and H3.3 copies of the histone pool [54]. Another intriguing effect of K-to-M mutations is that they diminish but do not fully erase histone methylation at specific sites. For instance, H3K27M expression globally reduces H3K27me3 levels yet preserves H3K27me3 deposition at a subset of PRC2 target loci containing the highest H3K27me3 levels in unperturbed cells [49, 50]. In fact, some studies even reported increased H3K27me3 levels at specific loci in H3K27M expressing cells [62–65], although it remains unclear whether these changes are direct or indirect effects of onco-histone expression [62]. Regardless, the retention and/or increase of H3K27me3 signal in H3K27M⁺ cancer cells likely explains why they remain sensitive to EZH2 inhibitor treatment [63, 64, 66].

Notably, the use of onco-histones has led to new biological insights not anticipated from previous knockout studies of the corresponding enzymes. For example, by leveraging a H3K4M mouse model, our lab discovered that the loss of H3K4 methylation leads to a fatal depletion of all mature blood cell types, even though upstream HSCs remain unaffected. Surprisingly however, committed progenitor cells were unable to differentiate in the absence of H3K4 methylation, revealing that H3K4 methylation is dispensable for the maintenance of the primitive stem cell pool but selectively required for progenitor cell maturation and consequently the survival of animals [16, 67]. This phenotype contrasts with previous MLL knockout studies [18, 19, 68–71], which reported depletion of HSCs upon loss of all forms of H3K4 methylation. Thus, histone mutations have not only redefined our understanding of cancer epigenetics but also provided a novel framework for decoding physiological and pathological roles of chromatin regulation in processes such as embryonic development, tissue homeostasis, regeneration, and malignant transformation (Figure 1). Previous excellent reviews on onco-histones have primarily focused on the four classic K-to-M mutations or the biochemistry of these mutants [58, 72]. In this review, we will instead discuss how histone mutants, including both engineered histones and onco-histones, have recently been leveraged as powerful perturbation tools to dissect the physiological roles of chromatin signaling in development, cell fate transitions, tissue maintenance, as well as summarize our current understanding of how onco-histones rewire the epigenome in a disease context.

Onco-histones and Epigenetic Rewiring in Disease

H3K27M mutations are detected in up to ~80% of pediatric diffuse midline gliomas and ~60% of adult diffuse gliomas, suggesting they function as oncogenic drivers [51, 73, 74]. In addition to suppressing H3K27 methylation, H3K27M also promotes the aberrant deposition of H3K27ac at repetitive elements in both human and mouse high-grade gliomas, resulting in increased transcription of these loci and widespread redistribution of histone modifications [75]. H3K27M-mutant pediatric high-grade gliomas also exhibit global DNA hypomethylation (Figure 2B). However, it remains unclear whether this hypomethylation is a direct consequence of disrupted H3K27me or instead represents an indirect consequence of malignant transformation [76]. Functional studies demonstrate that H3K27M expression alone induces only a transient increase in the self-renewal potential of hindbrain neural stem cells cultured in vitro, without conferring immortality or delaying cellular senescence [77]. However, when combined with additional oncogenic drivers—such as PDGFRA activation and TP53 deletion—H3K27M cooperates to drive spontaneous brainstem glioma formation in mouse models [78]. These results suggest that H3K27M expression initiates oncogenesis but is by itself insufficient to malignantly transform cells (see Outstanding questions for details).

Outstanding Questions Box.

Are onco-histones drivers of tumorigenesis?

To what extent do histone mutants faithfully model the physiological roles of histone modifications?

How do histone modifications regulate epigenetic memory and cell identity during tissue maintenance and regeneration?

Can onco-histones be leveraged to uncover therapeutic vulnerabilities in cancer?

As discussed earlier, another common onco-histone mutation involves substitution of H3 lysine 36 with methionine (H3K36M). This mutation is present in approximately 95% of benign chondroblastomas [52] but has also been identified in undifferentiated soft-tissue sarcomas [54] as well as head and neck squamous cell carcinomas [79]. Loss of H3K36 methylation in H3K36M cells is typically accompanied by a redistribution of the antagonistic mark H3K27me3, particularly at intergenic regions, leading to altered PRC1 localization and gene expression (Figure 2B). On a functional level, H3K36M reportedly disrupts mesenchymal differentiation, thus promoting the formation of undifferentiated sarcomas [54, 55]. Interestingly, in chondroblastomas, the H3K36M mutation predominantly occurs in the histone variant H3.3 rather than in canonical H3.1 [52]. Because H3.3 and H3.1 are incorporated into distinct genomic regions, Zhang and Fang proposed that the chromatin deposition pattern of H3.3K36M may determine its preferential occurrence and oncogenic impact in chondroblastoma. Indeed, although both H3.3K36M and H3.1K36M mutations reduce global H3K36 methylation, only H3.3K36M produces cancer-associated phenotypes, suggesting that the genomic localization of the mutant histone influences epigenomic reprogramming and cellular outcomes in specific contexts [80].

In addition to its dominant-negative effect on H3K36 methylation and the rewiring of H3K27me3, H3.3K36M expression was recently shown to disrupt the maintenance of epigenetic memory initiated by the KRAB-induced recruitment of H3K9me3 to a synthetic reporter. The authors of this study suggested that H3K36me2 and H3K36me3 enable the recruitment of DNA methyltransferases, which cooperate with H3K9me3 to stably silence the reporter, and this mechanism is destabilized by H3K36M [81].

In contrast to K-to-M mutations, G34 mutations occur exclusively in the histone variant H3.3 [52]. H3.3G34R/V mutations are present in cortical high-grade gliomas [82], while H3.3G34W mutations are found in giant cell tumors of bone [52]. Unlike K-to-M mutations, G34 substitutions act in cis on the mutant histone tail, leading to local rather than global methylation changes. Structural analyses of the G34 mutant revealed that bulky amino acid substitutions at position 34 sterically interfere with binding of the H3K36 methyltransferase SETD2, preventing access of the catalytic domain, and hence blocking H3K36 methylation specifically on the mutated histones (Figure 2B) [83].

In the context of giant cell tumor of bone, H3.3G34W promotes tumorigenesis by sustaining a transformed state in osteoblast-like progenitors, which drive neoplastic growth, pathologic recruitment of giant osteoclasts, and bone destruction [84]. However, in high-grade gliomas, H3.3G34R/V mutations are only weakly tumorigenic on their own and may even be dispensable for tumor maintenance. Instead, oncogenic alterations such as mutant PDGFRA are co-opted to drive gliomagenesis [85]. Mouse models carrying H3.3G34R/V/W mutations confirm mutation-specific phenotypes. H3.3G34R animals display neurological abnormalities, whereas H3.3G34W animals exhibit mesenchymal defects accompanied by pronounced obesity, and H3.3G34V mice show intermediate phenotypes between these two extremes [61]. At the molecular level, the G34R mutation reduces H3K36me2 deposition in cis and impairs recruitment of the DNA methyltransferase DNMT3A, leading to a loss of CH methylation at intergenic regions and an aberrant gain of CG methylation at neuronal promoters. This epigenetic reprogramming ultimately results in transcriptional perturbations associated with neurodegeneration. Despite these insights, a unified mechanism explaining the diverse biological consequences of H3.3G34R/V/W/L/M mutations has not yet emerged. The distinct developmental and pathological outcomes associated with H3.3G34R/V/W/L/M mutations may reflect the unique biochemical properties of the substituted residues, which differentially affect post-translational modification of the adjacent K36 residue and consequently alter recognition by chromatin writers, erasers, and readers.

In summary, onco-histones broadly reprogram the epigenomic landscape through direct or indirect interactions with chromatin regulatory factors. These mutations not only perturb methylation of the mutated residues themselves but also the distribution of antagonistic chromatin marks and DNA methylation, thereby reshaping genome-wide epigenetic states that contribute to oncogenic transformation. Notably, the tissue-specific distribution of onco-histone mutations suggests a strong context dependence, the mechanistic basis of which remains to be elucidated.

Histone Mutants as Versatile Tools for Dissecting Chromatin Regulation of Cell Identity

Roles of histone methylation in maintaining cell identity

To explore whether histone mutants are useful to identify chromatin mechanisms that maintain cell identity, our lab has recently overexpressed the replication-independent histone variant H3.3 carrying distinct K-to-M mutations during the reprogramming of fibroblasts to induced pluripotent stem cells (iPSCs) [86]; the generation of iPSCs is typically inefficient due to epigenetic mechanisms that safeguard somatic cell identity and resist cell fate change [87, 88]. We found that H3K4M blocks while H3K9M slightly enhances reprogramming, indicating that H3K4 methylation is required for, while H3K9 methylation inhibits reprogramming consistent with prior studies that suppressed multiple redundant histone-modifying enzymes [89–92]. Strikingly, our study also uncovered a profound effect of H3K36M on iPSC reprogramming, with nearly all cells acquiring pluripotency within 7 days, suggesting that H3K36 methylation is a major barrier to iPSC reprogramming [86]. Importantly, this enhancement was specific to H3K36M expression and was not observed with overexpression of wild-type H3.3, revealing that it is not simply a consequence of increased H3.3 levels. Mechanistically, H3K36M drives the decommissioning of somatic enhancers downstream of TGFβ signaling leading to the repression of mesenchymal genes via PRC2-dependent H3K27me3 deposition and enabling the transition to an epithelial state. In parallel, H3K36M promotes the activation of pluripotency enhancers within H3K36me2-depleted domains, accompanied by increased H3K27ac deposition and chromatin accessibility. These chromatin changes facilitate the binding of pluripotency transcription factors, which have been shown to recruit Ten-Eleven-Translocation (Tet) enzymes to pluripotency regulatory elements and subsequently enable active DNA demethylation [93, 94]. Consistent with this model, inhibition of Tet activity impaired reprogramming even in the presence of H3K36M expression. Together, these results are consistent with a dual function of H3K36 methylation in the maintenance of cell identity, by integrating a crucial developmental pathway into sustained expression of cell type specific programs, and by antagonizing the expression of alternative lineage programs through enhancer regulation (Figure 3A).

Fig.3: Histone Mutants as Determinants of Cell Identity and Lineage Fidelity.

(A) H3K36M enhances iPSC reprogramming efficiency. Loss of H3K36 methylation promotes shutdown of somatic genes via PRC2-dependent H3K27me3 deposition and activation of pluripotency genes through TET-mediated enhancer DNA demethylation.

(B) Mutation of lysine 27 to arginine in all canonical histone H3 genes (H3K27R) causes global loss of H3K27 methylation and acetylation. H3K27R ESCs fail to differentiate properly, in part because of the derepression of PRC2 target genes.

(C) H3K36M-expressing mice develop hematopoietic defects including HSC loss and differentiation arrest in intestinal and epithelial tissues. Aberrant redistribution of PRC2-mediated H3K27me3 silences differentiation-associated genes.

(D) H3K4M expression induces a hematopoietic differentiation block by silencing fate-instructive bivalent genes through loss of H3K4me3 and ectopic gain of H3K27me3. H3K27M expression rescues the phenotype in H3K4M mice. Selected biological illustrations were created with BioRender.com.

Complementary results have been reported in mouse ESCs carrying mutations in endogenous histone genes. Briefly, Gehre et al. employed K-to-A mutations (e.g., H3.3K4A and H3.3K36A) to investigate the structural and functional requirement of K4 and K36 methylation at the histone H3.3 variant associated with promoter and enhancer regulation [41]. The authors found that homozygous H3.3K4A mutations markedly impair differentiation, accompanied by a loss of H3.3K4me3 levels, substantial depletion of H3.3 deposition at regulatory elements, altered transcriptional programs, and reduced recruitment of chromatin remodeling complexes. By contrast, homozygous H3.3K36A mutations selectively reduce H3.3K36me2 deposition, while largely preserving H3.3K36me3 and H3.3 deposition, with only mild effects on differentiation. This differential effect of H3.3K4A and H3.3K36A on histone methylation, deposition and differentiation likely reflects the enrichment of the H3.3 variant at promoters but its exclusion from gene bodies relative to canonical H3.1/H3.2. Together, these findings uncover a crucial role of H3.3K4 methylation in preserving the integrity of promoters critical for early lineage commitment.

Similar to the K-to-A approach for specific H3.3 residues, Sanker et al. used targeted histone H3K27 mutations in mouse ESCs to dissect Polycomb-mediated repression [43]. As we alluded to earlier in the text, the authors demonstrated that complete replacement of canonical H3K27 with arginine (H3K27R) abolishes PRC2-mediated H3K27me3 deposition genome-wide and results in inappropriate activation of Polycomb target genes, closely phenocopying PRC2 deficiency. Although H3K27 acetylation—associated with active promoters and enhancers—is depleted in the H3K27R mutant system, enhancer activation, RNA polymerase II recruitment, and occupancy of the transcriptional co-factor MED1 were largely preserved during exit from naïve pluripotency [43]. These findings indicate that H3K27me3 is the critical determinant of gene regulation in ESCs, while H3K27 acetylation is largely dispensable for these processes (Figure 3B).

In addition to mutational analyses of classic methylation sites such as H3K27 and H3K36, mutants affecting non-methylatable histone residues have provided valuable insight into chromatin mechanisms that preserve cell identity. Nacev et al. recently showed that H3R2C and H3R26C mutants reduce H3K27me3 levels through impaired PRC2 activity and disrupt differentiation in both mesenchymal progenitors and ESCs, illustrating how arginine mutations can uncover Polycomb-dependent mechanisms of lineage restriction beyond canonical lysine residues [95]. Library-scale histone mutant screening approaches have expanded this framework to unbiased discovery beyond histone H3 and lysine residues [57, 96, 97]. When applied to C3H10T1/2 mesenchymal stem cell-like cells, such screens uncovered unexpected roles for H2B mutants (e.g., E71K, E76K, and E113K) that selectively impair adipogenic differentiation, illustrating how scalable histone mutagenesis can reveal previously obscured chromatin mechanisms directing lineage specification [57]. Another recent overexpression screen encompassing >4,000 histone variants identified H4G4D as a mutant that disrupts heterochromatin maintenance in cis and trans while rewiring nuclear compartmentalization and transcription factor binding landscapes in T cells [96]. Future work is warranted to understand the precise molecular and physiological roles of these lesser studied histone mutants.

Collectively, these histone mutant studies exemplify how individual histone residues exert instructive roles in lineage commitment and cell identity, offering a powerful platform to directly interrogate chromatin function.

Leveraging Histone Mutants to Dissect Physiological Roles of Chromatin Signaling

Histone mutants reveal distinct functions of chromatin pathways in tissue homeostasis

In a pioneering study, Shilatifard and colleagues demonstrated that expression of H3K27M or H3K9M mutations in Drosophila embryos effectively depletes H3K27 and H3K9 methylation in vivo, leading to the de-repression of PRC2 target genes and defects in heterochromatic silencing, respectively [98]. Extending this paradigm, Chaouch et al. showed that H3K27M and H3K36M disrupt the balance of antagonistic H3K27 and H3K36 methylation landscapes in Drosophila, causing transcriptional dysregulation and developmental defects that could be rescued by rebalancing the opposing chromatin pathways [99]. Together, these studies established (dominantly acting) histone mutants as versatile tools to investigate the developmental roles of chromatin pathways in flies and motivated subsequent efforts to probe the physiological roles of specific histone modifications in a mammalian context using histone mutant transgenic tools.

Brumbaugh et al. generated the first transgenic mouse models enabling widespread and inducible expression of H3K9M or H3K36M [48]. Mice ubiquitously expressing H3K36M exhibited a global loss of H3K36me2/3, fatal anemia with defects in erythroid differentiation and a depletion of the hematopoietic stem cell (HSC) pool, as well as defective secretory differentiation in the intestine and spermatocyte differentiation in the testes (Figure 3C). By contrast, mice ubiquitously expressing H3K9M survived for up to a year but exhibited an expansion of short-term HSCs and an arrest of immature B cells, with some animals developing an aggressive form of T cell leukemia. These experiments highlight that the introduction of histone mutants that differ at a single amino acid (H3K9M vs H3K36M) into otherwise identical mice is sufficient to globally perturb two major chromatin pathways and exert specific and previously unknown effects on cell and tissue maintenance in mammals.

Following up on the secretory defect of H3K36M mice, Pashos et al. recently directed H3K36M expression specifically to the intestine to demonstrate that H3K36 methylation regulates both the maintenance of intestinal epithelial cell identity as well as epithelial plasticity during regeneration (Figure 3C) [100]. During intestinal homeostasis, H3K36me3 helps to maintain cell type-specific gene expression by restricting the spread of antagonistic H3K27me3. However, during regeneration, H3K36me3 levels are reduced, leading to a redistribution of H3K27me3 deposition patterns. Specifically, H3K27me3 levels accumulate over certain lineage-specific genes, facilitating their downregulation. As part of this chromatin reorganization, H3K27me3 levels decrease at a subset of repressed developmental and regenerative genes, resulting in their activation, increased epithelial plasticity and enhanced repair. In a complementary study by Ko et al., the authors investigated the role of H3K36 methylation in development and tissue homeostasis using a conditional mouse model expressing H3K36M in Krt14-expressing stratifying epithelial cells (Figure 3C) [101]. Disruption of H3K36 methylation in this epithelial context caused extensive tissue dysplasia with an aberrant increase in glandular cells in the tongue and skin of adult mice consistent with increased plasticity. Mechanistically, H3K36M expression led to a loss of H3K36me2 levels at differentiation-associated genes, which was accompanied by the redistribution of H3K27me3 away from repressed glandular genes to these differentiation genes, manifesting as increased plasticity. In line with its role as an onco-histone, persistent H3K36M expression in stratified epithelia predisposed animals to the development of squamous cell carcinomas, suggesting that normal homeostasis and increased plasticity need to be tightly regulated to prevent malignant transformation.

Physiological interrogation of enhancer and bivalent promoter regulation

To directly probe the role of H3K4 methylation in vivo, Jang et al. generated transgenic mice whereby H3K4M was targeted to skeletal muscle and preadipocyte lineages using the Myf5-Cre driver [102]. As predicted, expression of H3K4M resulted in a global reduction of H3K4me1, H3K4me2 and H3K4me3 levels. Mutant mice died shortly after birth due to respiratory failure caused by defective rib cage muscle groups, accompanied by moderate reduction of brown adipose tissue and back muscle mass. Mechanistically, the authors showed that H3K4M expression prevents enhancer activation during adipogenesis by destabilizing the enhancer-associated H3K4 methyltransferases MLL3 and MLL4. Indeed, deletion of the catalytic SET domain of MLL3/4 in adipogenic precursor cells phenocopied the defects observed in H3K4M mice. By contrast, H3K4M expression in differentiated adipocytes had little effect on adipose tissue maintenance. These findings suggest that H3K4 monomethylation is required for adipose development but dispensable for its maintenance.

As alluded to previously, our lab has recently developed mouse models that allow for the inducible expression of H3K4M or H3K27M across diverse tissues [67]. Adult H3K4M mice treated with doxycycline became moribund within a few weeks due to a profound loss of all mature blood cells caused by a specific progenitor cell arrest. Strikingly, withdrawal of doxycycline readily restored H3K4 methylation, allowed progenitors to resume differentiation, and reversed morbidity, demonstrating that arrested progenitors remain fully functional upon normalization of H3K4 methylation and further highlighting the versatility of inducible histone mutants (see Glossary) to probe the reversibility of phenotypes caused by disrupted histone methylation. On a molecular level, the loss of H3K4me3 at promoters led to the aberrant silencing of fate-instructive genes caused by the aberrant deposition of antagonistic H3K27me3. Transcriptional and chromatin profiling further revealed that associated promoters are strongly enriched for bivalent (H3K4me3⁺/H3K27me3⁺) chromatin. Importantly, suppression of H3K27me3 levels using an H3K27M transgene restored gene activation, rescued hematopoietic differentiation, and significantly extended survival in H3K4M mice. Of relevance, H3K27M failed to restore enhancer-associated chromatin marks in H3K4M mice, indicating that the observed phenotypes are primarily driven by promoter dysregulation. Thus, the balance between promoter H3K4 trimethylation and H3K27 trimethylation, particularly at bivalent genes, is critical to safeguard mammalian hematopoietic differentiation (Figure 3D). More broadly, these experiments underscore the power of histone mutant tools to probe not only individual but also combinatorial roles of histone modifications during specific cell fate transitions.

Conceptual and Technical Limitations of Histone Mutant Tools

While histone mutant strategies have provided powerful tools to study chromatin biology, several conceptual and technical limitations should be considered when interpreting results from these systems. Histone mutants can influence multiple chromatin features simultaneously, providing potential challenges with attributing observed phenotypes to the loss of a single histone mark (Figure 1). For example, lysine residues can be subject to multiple types of modifications, including methylation, acetylation, or ubiquitination, depending on the chromatin context. Mutations that replace lysine residues therefore eliminate not only the targeted methylation mark but also other potential modifications at the same site. The effect of a histone mutant on these additional modifications also depends on the type of substitution and whether it acts in cis or in trans.

In addition, histone mutant strategies typically disrupt multiple methylation states such as mono-, di- and trimethylation at a given histone site as histone mutants often inhibit multiple enzymes that catalyze distinct methylation states. Because these methylation states are associated with distinct chromatin functions, it may be difficult to determine which specific methylation state is responsible for a given phenotype. Moreover, the sequestration model implies that K-to-M mutants may affect not only catalytic activity but also non-catalytic functions of cognate enzymes. Consequently, some phenotypes detected in histone mutant models may reflect the combined disruption of several regulatory networks rather than the specific loss of a single histone mark, necessitating follow-up experiments using pharmacological or genetic inhibition of individual enzymes. A case in point is the H3K4M-induced defect in adipocyte differentiation, which is driven by H3K4 monomethylation [102], compared to the H3K4M-induced defect in hematopoietic progenitor cell maturation, which is driven by H3K4 trimethylation [67].

Histone mutations can also alter the chemical properties of the histone tail itself, potentially affecting nucleosome interactions with DNA, histones, histone chaperones, or chromatin-associated proteins independently of modification status. Indeed, mutations within the globular domain disrupt histone-histone interactions and compromise nucleosome octamer stability, thereby perturbing transcriptional programs and cell identity [57, 60, 103]. For these reasons, complementary approaches remain essential for establishing causal relationships between histone modifications and biological phenotypes. Catalytic knockout or point-mutation models targeting histone-modifying enzymes provide valuable orthogonal evidence for the catalytic role of an enzyme. Although the interpretation of enzymatic knockout models is often complicated by enzyme redundancy and promiscuity, careful comparison between enzyme perturbations and histone mutant phenotypes can help disentangle catalytic versus non-catalytic and site-specific versus promiscuous roles of chromatin regulators. For instance, phenotypic similarities and differences between histone mutant models and catalytic inactivation of specific methyltransferases can strengthen the conclusion that the affected histone modification itself plays a causal role (see Outstanding questions for details).

Another potential limitation of certain histone mutants is that they may lead to global decreases but local increases in histone methylation with unclear consequences [50]. For example, the H3K27M mutation reportedly leads to elevated H3K27me3 peaks at PRC2 targets, although recent data suggest that these peaks may be an artifact caused by secondary changes in gene expression [62]. Regardless, it is well established that mutant histones actively reprogram chromatin regulatory networks [104]. Indeed, biochemical and proteomic studies have identified mutant-specific interactions with chromatin regulators. For example, Reinberg and colleagues used quantitative biochemical and mass spectrometry approaches to show that H3K27M nucleosomes can trap PRC2 and suppress its methyltransferase activity [105]. Such findings highlight the importance of integrating biochemical and proteomic analyses to understand the mechanistic underpinnings and functional consequences of histone mutations. A key remaining challenge is distinguishing primary from secondary chromatin effects following histone mutant expression. Although relatively few inducible time-course studies exist, doxycycline-inducible H3K27M systems show that H3K27me3 loss occurs rapidly within ~12 hours of induction, consistent with a direct effect on PRC2 activity [105]. In contrast, later chromatin and transcriptional changes are likely secondary consequences. Systematic time course chromatin profiling is warranted to clarify the temporal hierarchy of molecular changes.

Lastly, studies relying on the overexpression of histone mutants may introduce non-physiological artifacts compared with endogenous knock-in models, underscoring the importance of including wild-type histone overexpression controls. Looking forward, integrating histone mutant models with complementary genetic, biochemical, and genomic approaches will be critical for fully interpreting phenotypes. These approaches will refine the histone mutant paradigm and reveal how histone marks regulate development, tissue homeostasis, and disease.

Concluding Remarks and Future Perspectives

Histone mutant strategies have recently emerged as powerful tools that complement traditional genetic knockout approaches of histone-modifying enzymes. By selectively perturbing individual chromatin marks, these tools have revealed causal roles of specific modifications in transcriptional regulation, cell identity, lineage commitment, tissue homeostasis, and cancer. Unlike conventional histone methyltransferase knockouts, which are complicated by catalytic and non-catalytic functions of enzymes and enzymatic redundancy, histone mutants provide a dominant approach to interrogate chromatin function. We therefore conclude that histone mutants serve as powerful complementary strategies to enzymatic knockout approaches in defining the functional roles of chromatin modifications, with each strategy having advantages and disadvantages.

Inducible, reversible, and tissue-specific mutant systems enable temporal and context-specific dissection of epigenetic regulation, allowing researchers to uncover dynamic roles of histone modifications in development, regeneration, and disease progression. Integrating histone mutant models with complementary genomic, biochemical, and proteomic approaches will be essential to dissect the mechanistic interplay between histone modifications, chromatin regulators, and transcriptional programs. In addition, combining histone mutant tools with recent lineage tracing technologies, such as CRISPR-based DNA barcoding systems (e.g., CARLIN [106] or DARLIN [107]), will enable tracking of clonal dynamics, stem cell heterogeneity, and cellular origins of disease in vivo.

We anticipate that histone mutant strategies will illuminate the principles of epigenetic memory and plasticity and reveal how specific chromatin states can be modulated to restore tissue function or reverse malignancy (see Outstanding questions for details). From a translational perspective, defining vulnerabilities associated with specific histone modifications may open avenues for targeted therapy. Importantly, the epigenetic plasticity of histone mutant driven chromatin states suggests that transient chromatin modulation could reprogram aged or malignant cells toward normal tissue function or enhance tumor sensitivity to combination therapies (see Outstanding questions for details). Ultimately, onco-histones and engineered histone mutants provide a versatile platform not only for decoding chromatin regulation and cell identity but also for translating mechanistic insights into strategies for tissue rejuvenation and cancer therapy.

HIGHLIGHTS.

• Histone tail mutants inhibit methylation locally or globally at specific sites, circumventing the redundancy and promiscuity of histone-modifying enzymes.

• Beyond engineered histone tail mutants, cancer-associated histone mutations (onco-histones) are thought to drive oncogenesis.

• Histone mutant tools allow mechanistic dissection of the role individual and combinatorial chromatin marks play in transcription and cell identity control.

• Transgenic models enable studying the physiological consequences and reversibility of disrupted histone methylation in vivo.

• Histone-mutant screening strategies have the potential to reveal new drug targets and inform future strategies in regenerative medicine and cancer treatment.

Glossary

Bivalent Domains

Promoter regions marked simultaneously by H3K4me3 associated with active chromatin and H3K27me3 associated with repressed chromatin. Bivalent domains are prevalent in pluripotent and adult stem cells where they are thought to keep developmental genes transcriptionally poised for activation or repression during differentiation. However, bivalently marked genes have also been detected in differentiated cells.

Chromatin

Complex composed of DNA and proteins, including histone proteins that package the genome within the nucleus.

Enhancer

Distal regulatory DNA element that enhances transcription of target genes in a cell type-specific manner. Enhancers are frequently associated with histone modifications such as H3K4me1 and H3K27ac.

Histone Code Hypothesis

Concept proposing that distinct combinations of histone modifications encode regulatory information that instructs chromatin states and transcriptional outcomes through interactions with reader and effector proteins.

Histone Methylation

A type of post-translational modification (PTM) whereby methyl groups are added to specific lysine residues on histone tails. Depending on the lysine residue and methylation state (-me1, -me2, -me3), PTMs either enhance or suppress gene expression. H3K4 and H3K36 methylation are associated with active chromatin, with H3K4 methylation being enriched at promoters (H3K4me3) or enhancers (H3K4me1) and H3K36 being enriched along transcribed gene bodies (H3K36me3) and in intergenic regions (H3K36me2). H3K27 and H3K9 methylation are associated with repressed chromatin, with H3K27 methylation being linked to Polycomb-mediated gene repression and H3K9 methylation being associated with constitutive heterochromatin.

Histone-Modifying Enzymes

Enzymes that catalyze histone modifications. These are grouped into “writers” that add modifications (e.g., methyltransferases), “erasers” that remove modifications (e.g., demethylases), and “readers” that recognize modified histone residues to recruit downstream regulatory complexes.

Inducible Histone Mutants

Experimental systems in which mutant histone proteins are expressed in a controlled temporal or tissue-specific manner, allowing researchers to investigate the causal roles of individual histone residues in development, tissue maintenance, and disease without disrupting the respective histone-modifying enzymes.

K-to-M mutations

Lysine-to-methionine substitutions of histone genes, most commonly histone H3, at sites normally modified by histone methyltransferases. K-to-M mutations dominantly inhibit the function of histone methyltransferases, resulting in a global suppression of histone methylation at specific sites.

Onco-histones

Recurrent histone mutations identified in diverse human cancers and thought to function as oncogenic drivers. Depending on the type of mutated histone and residue, onco-histones can lead to local or global changes in chromatin signaling.

Polycomb Repressive Complex 2 (PRC2)

A multiprotein complex responsible for catalyzing H3K27 methylation. PRC2 mediates transcriptional repression of developmental genes and plays essential roles in cell fate regulation, development and cancer.