Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Apr 1.
Published in final edited form as: J Mol Biol. 2023 Dec 20;436(7):168413. doi: 10.1016/j.jmb.2023.168413

Hotspot cancer mutation impairs KAT8-mediated nucleosomal histone acetylation

Hongwen Xuan 1, Longxia Xu 1, Kuai Li 1, Fan Xuan 1, Tinghai Xu 2, Hong Wen 1, Xiaobing Shi 1,*
PMCID: PMC10957314  NIHMSID: NIHMS1955114  PMID: 38135180

Abstract

KAT8 is an evolutionarily conserved lysine acetyltransferase that catalyzes histone acetylation at H4K16 or H4K5 and H4K8 through distinct protein complexes. It plays a pivotal role in male X chromosome dosage compensation in Drosophila and is implicated in the regulation of diverse cellular processes in mammals. Mutations and dysregulation of KAT8 have been reported in human neurodevelopmental disorders and various cancers. However, the precise mechanisms by which these mutations disrupt KAT8’s normal function, leading to disease pathogenesis, remain largely unknown. In this study, we focus on a hotspot missense cancer mutation, the R98W point mutation within the Tudor-knot domain. Our study reveals that the R98W mutation leads to a reduction in global H4K16ac levels in cells and downregulates the expression of target genes. Mechanistically, we demonstrate that R98 is essential for KAT8-mediated acetylation of nucleosomal histones by modulating substrate accessibility.

Keywords: KAT8, R98W, R99W, Tudor-knot domain, cancer mutation

Graphical Abstract

graphic file with name nihms-1955114-f0001.jpg

Introduction

Lysine acetyltransferase 8 (KAT8), also known as males absent on the first (MOF), was initially identified as a gene required for dosage compensation in Drosophila [1]. KAT8 serves as the catalytic subunit of the male-specific lethal (MSL) complex [2, 3]. In Drosophila, the MSL complex binds to the male X chromosome and acetylates histone H4K16, thereby augmenting gene expression on the X chromosome in males [1, 4, 5]. In mammals, the MSL complex also coordinates with other chromatin regulators regulating gene expression beyond X chromosome [6, 7]. In addition to the MSL complex, KAT8 associates with eight other subunits composing the non-specific lethal (NSL) complex [7–10]. Genomic studies have shown that the MSL complex predominantly localizes to gene bodies and the 3’ ends of X-linked genes, depositing H4K16ac. In contrast, the NSL complex preferentially binds to the promoters and enhancers of targeted genes, acetylating histones H4K5 and H4K8 [11–13]. The depletion of KAT8 in cells leads to a global reduction in H4K16ac, but not H4K5ac or H4K8ac, likely due to compensatory actions by other H4-acetylating enzymes like TIP60 [13]. KAT8 belongs to the MYST (MOZ, YBF2/SAS3, SAS2, and TIP60) family of proteins, characterized by a MYST-type histone acetyltransferase (HAT) domain closely linked with a C2HC-type zinc finger motif [14, 15]. Additionally, it contains an N-terminal Tudor-knot domain, the precise function of which remains largely elusive. While recombinant KAT8 alone possesses HAT activity, its full catalytic capacity and substrate specificity necessitate integration into either the MSL or NSL complexes [16].

Numerous studies have unveiled the critical roles of KAT8 in regulating a wide array of cellular processes, including cell cycle progression, DNA damage response, and stem cell development [11, 17]. Like many other KATs, KAT8 regulates gene expression through the acetylation of both histones and non-histone proteins. For example, KAT8 acetylates p53 and NRF2, thereby regulating cellular responses to DNA damage and oxidative stress [10, 18, 19]. KAT8 is also recruited to DNA damage sites, facilitating DNA repair through various repair pathways [20–22]. Furthermore, KAT8 has been reported as a key regulator of stem cell self-renewal [23]. Given the importance of KAT8 in numerous cellular processes, it is not surprising that mutations or deregulation of KAT8 have been implicated in both human neurodevelopmental disorders and cancer. In human cancers, KAT8 exhibits amplification (490 cases) and depletion (321 cases) in various types of cancer [24]. Similarly, KAT8 mRNA expression can be either upregulated or downregulated in different cancers [17, 25, 26]. These observations suggest that KAT8 may function either as a tumor suppressor or an oncogene in a context-dependent manner.

In this study, we characterize a hotspot missense cancer mutation, the R98W point mutation, within the Tudor-knot domain. Our findings reveal that the R98W mutation reduces global H4K16ac levels and downregulates the expression of target genes. Notably, R98W does not affect KAT8 chromatin occupancy but impairs KAT8-mediated acetylation of nucleosomal histones. We propose a model in which the positively charged residue R98 in the Tudor-knot domain competes with histone tails for nucleosomal DNA binding, thus increasing accessibility of the histone tails for acetylation.

Results

The R98W cancer mutation of KAT8 leads to a global reduction in H4K16ac levels in cells

To investigate the impact of KAT8 mutations in cancer on its function, we analyzed the COSMIC database, which contains 394 curated cancer mutations of the KAT8 gene [24]. Among these, 188 are missense mutations, with notable hotspot mutations including R98W (14 cases), R140H/C/G (14 cases), and E350A/K (8 cases) (Figure 1a). E350 is a catalytic residue, and mutations at this site are known to abolish KAT8 HAT activity [27]. However, the effects of other hotspot mutations on KAT8 function have remained uncharacterized. R98 is located within the well-defined Tudor-knot domain of KAT8 (Figure 1a). While the precise function of the Tudor-knot domain remains elusive, it appears to be functionally important. Pan-cancer survival data from cBioPortal [28] reveals that patients with mutations in the Tudor-knot domain tend to have a worse prognosis compared to patients with wild-type (WT) KAT8 or mutations in other parts of KAT8 (Figure 1b and Supplementary Table S1).

Figure 1. The R98W cancer mutation of KAT8 reduces global H4K16ac in cells.

Figure 1.

Open in a new tab

a, Schematic of the KAT8 protein structure with missense mutations of cancer patients. Mutation information was generated by ProteinPaint (https://proteinpaint.stjude.org/) with data from COSMIC. Numbers of cases of mutations are shown in circles. b, Survival curves of patients with WT KAT8, mutations of the KAT8 Tudor-knot domain or other regions excluding Tudor-knot. Data were obtained from cBioPortal. Log-rank (Mantel-Cox) test was used to calculate p-values. c and d, Western blot analysis of H4K16ac in A549 (c) and H1703 (d) cells stably expressing Flag-tagged WT KAT8 and the indicated mutants. e, Western blot analysis of H4K16ac in A549 Tet-On cells with overexpression of WT KAT8 or R98W using pCDH vector. Total H3 is shown as a loading control. OE: overexpression.

KAT8 is primarily responsible for histone H4K16 acetylation (H4K16ac) in cells [13]. To assess the potential impact of the R98W mutation on KAT8’s function, we overexpressed the KAT8-R98W mutant alongside WT KAT8 in A549, a non-small cell lung cancer (NSCLC) cell line, and measured histone H4K16ac levels. Cells expressing the R98W mutant showed a clear reduction in H4K16ac levels compared to those expressing WT KAT8 (Figure 1c). To understand how this cancer mutation affects the molecular function of R98, we also mutated it to several other amino acids. The results indicated that mutation of R98 to a negatively charged amino acid (R98E) similarly reduced global H4K16ac levels, while mutations to alanine or lysine had minor or no effect (Figure 1c). Similar results were obtained in another NSCLC cell line H1703 (Figure 1d). These findings suggest that an increase in size or change in charge of the size chain affects the normal function of the arginine residue.

The R98W mutation is a somatic mutation that occurs on only one allele of the gene in cancers. In A549 and H1703 cells, the endogenous KAT8 is wild-type. Overexpression of KAT8 R98W in these cells results in a global reduction of H4K16ac levels, suggesting a dominant negative effect. To further test this, we used a Tet-On system to induce KAT8 expression (WT and R98W) at relatively low levels and we observed a mild reduction in H4K16ac in the R98W-expressing cells (Figure 1e, comparing lane 4 to lane 1). Importantly, overexpression of WT KAT8 (using the pCDH lentiviral vector) in the R98W-expressing cells can override the effect of the mutant, likely because WT KAT8, expressed at a much higher level, outcompetes the mutant protein (Figure 1e, comparing lane 5 to lane 4). These data suggest that the R98W mutation has a dominant negative effect, and heterozygous mutation of the KAT8 gene in cancer cells likely is sufficient to cause a global reduction in H4K16ac levels.

The KAT8 R98W mutation downregulates gene expression through modulating H4K16ac

To determine the specific genomic locations where H4K16ac is reduced, we conducted chromatin immunoprecipitation experiments coupled with next-generation sequencing (ChIP-seq) of H4K16ac in A549 cells expressing WT KAT8 and the R98W mutant. We assessed the distribution of H4K16ac on all genes in the genome. In agreement with previous reports [12, 13], H4K16ac was found to be enriched in gene bodies and peaked toward the 3’ ends of transcribed regions (Figure 2a). Importantly, the global reduction of H4K16ac observed in the R98W-expressing cells was evident on nearly all genes, with genes exhibiting higher basal levels of H4K16ac showing more significant fold-change reductions (Figure 2a). We speculated that the genome-wide reduction of H4K16ac might result from decreased KAT8 chromatin occupancy in the R98W-expressing cells. To explore this possibility, we conducted ChIP-seq experiments with an anti-HA antibody in the same cell lines as mentioned earlier to assess the chromatin occupancy of the ectopically expressed Flag-HA tagged KAT8 proteins. However, to our surprise, we did not observe a clear reduction in KAT8 chromatin occupancy in cells expressing the R98W mutant compared to those expressing WT KAT8 (Figure 2b). These results indicate that the global reduction of H4K16ac induced by KAT8 R98W mutation is due to a mechanism other than altered KAT8 chromatin occupancy.

Figure 2. The KAT8 R98W mutation downregulates gene expression through modulating H4K16ac but not KAT8 chromatin occupancy.

Figure 2.

Open in a new tab

a, Heatmaps of H4K16ac ChIP-seq densities (left panels) and log2 fold changes (FC, right panel) of all genes (58,639) from A549 cells expressing WT KAT8 and R98W. b, Heatmaps of HA-KAT8 ChIP-seq densities (left panels) and log2 fold changes (FC, right panel) from cells as in (a). In both (a) and (b), genes are ranked by H4K16ac densities in WT A549 cells from high to low. Signals are plotted from transcription start site (TTS) to transcription termination site (TSS) of all genes with 5kbp upstream and downstream flanking regions. c, Heatmap of differentially expressed genes (DEGs) in A549 cells expressing R98W compared to cells expressing WT KAT8. Values in heatmap represent z-scores normalized log2 CPM values from RNA-seq data. d, Enriched PANTHER GO-Slim Biological Process terms of the 629 downregulated genes as in (c). e and f, Boxplots of H4K16ac (e) and HA-KAT8 (f) ChIP-seq densities on up or down DEGs. Density values in cells expressing WT KAT8 are marked in grey, and density values in cells expressing R98W in red. Two-tail t-test was used to calculate p-values. g, Integrity plot for H4K16ac and HA-KAT8 ChIP-seq densities and RNA log2 FC in A549 cells expressing WT KAT8 and R98W. All genes (58,639) are ranked as in (a) and separated into 100 bins, with average signals in each bin are plotted from left to right. H4K16ac and HA-KAT8 ChIP-seq densities are scaled to the left Y-axis, and log2 FC of RNA (R98W vs. WT) is indicated with blue line and scaled to the right.

Histone acetylation is normally associated with active transcription. To determine if the global reduction of H4K16ac impacts gene expression in R98W-expressing cells, we conducted RNA-seq analysis. With a fold-change >2 and an FDR<0.05, we identified 674 differentially expressed genes (DEGs) in R98W-expressing cells compared to those expressing WT KAT8. Notably, among all the DEGs, 93.3% (629) of genes were downregulated, while only 6.7% (45) were upregulated (Figure 2c and Supplementary Table S2). Gene ontology analysis using the Panther GO-slim biological process revealed that the downregulated genes were mainly enriched in cell projection organization and RNA Pol II-mediated transcription regulation, while the upregulated genes showed no enriched terms (Figure 2d and Supplementary Table S3).

Next, we examined the correlation between changes in gene expression and H4K16ac levels in these cells. We found that both downregulated and upregulated genes exhibited reduced H4K16ac in R98W-expressing cells (Figure 2e). However, in WT KAT8-expressing cells, downregulated genes displayed higher basal levels of H4K16ac than upregulated genes (Figure 2e). In contrast, there was no difference in KAT8 occupancy observed between these two groups of genes (Figure 2f). Analyzing all genes in the genome also revealed a strong correlation between the reduction in H4K16ac and decreased gene expression genome-wide (Figure 2g). In summary, these results suggest that the KAT8 R98W mutation downregulates gene expression by reducing H4K16ac levels without altering KAT8 chromatin occupancy.

The R98W mutation impairs KAT8-mediated acetylation of nucleosomal histones

Given that the R98W mutation does not alter KAT8’s chromatin occupancy, we hypothesized that the R98 residue might play a critical role in KAT8’s enzymatic activity. To investigate this, we first established stable cell lines expressing Flag-tagged WT KAT8 and the R98W mutant. We then purified the protein complexes associated with these two proteins through Flag IP and performed in vitro histone acetyltransferase (HAT) assays. We also included the R99W mutation in this assay since it’s another frequent cancer mutation within the Tudor-knot domain (Figure 1a). Using mono-nucleosomes as substrates, we discovered that both the R98W and R99W mutations abolished KAT8’s acetylation activity on H4K16 (Figure 3a). KAT8 is known to also acetylate histone H4 at K5 and K8 [8, 9]. Using a pan-H4ac antibody that recognizes both marks, we also observed a marked reduction in H4K5/K8 acetylation levels with both mutants compared to WT KAT8.

Figure 3. The hotspot cancer mutations impair acetylation of nucleosomal histone H4 by the KAT8 complexes.

Figure 3.

Open in a new tab

a, Western blots analysis of the HAT assays with the indicated KAT8 complexes purified from A549 cells and nucleosomes as substrates. b, Western blots analysis of the indicated KAT8 complex components of Flag-IPs in cells expressing Flag-tagged WT KAT8, R98W, and R99W. Cells expressing an empty vector was used as a negative control. c, Western blots analysis of the HAT assays with the indicated KAT8 complexes purified from A549 cells and histone octamers as substrates. d, KAT8 has higher HAT activity on histone octamer than nucleosome. Western blot analysis of HAT assays with the KAT8 complex as enzyme and nucleosomes and histone octamers as substrates.

KAT8 forms two distinct protein complexes, MSL and NSL, that are responsible for acetylating H4K16 and H4K5/K8, respectively [11]. Since KAT8’s catalytic activity heavily relies on its associated protein complexes, we examined whether the R98W and R99W mutations might affect protein-protein interactions between KAT8 and other components of the MSL or NSL complex. To do this, we immunoprecipitated (IPed) Flag-KAT8 using the M2 anti-Flag antibody and assessed the components of the MSL and NSL complexes co-IPed with WT and mutant KAT8. The results showed that both the R98W and R99W mutants efficiently co-IPed components of the MSL and NSL complexes, indicating that these mutations do not disrupt the integrity of the KAT8 complexes (Figure 3b).

In addition to assessing HAT activity on nucleosomes, we also conducted in vitro HAT assays using histone octamers as substrates. Surprisingly, we did not observe any significant differences between the mutants and WT KAT8 in catalyzing either H4K16ac or pan-H4ac (Figure 3c). The primary difference between nucleosomes and histone octamers is the presence of DNA in nucleosomes but not in histone octamers. DNA is negatively charged and naturally interacts with positively charged residues on histones, reducing the accessibility of the histone tails [29, 30]. In contrast, histone tails in octamers are relatively “free” and more accessible to histone-modifying enzymes. Indeed, a direct comparison using in vitro HAT assays with the KAT8 complex demonstrated that the octamer is a much better substrate than the nucleosome (Figure 3d). In summary, these results suggest that the R98 and R99 residues in the Tudor-knot domain regulate KAT8’s HAT activity on nucleosomes, likely by fine-tuning substrate accessibility.

The R98W mutation impedes the KAT8 Tudor-knot domain binding to nucleosome

Given that the R98W mutation impairs KAT8’s HAT activity on nucleosomes without affecting its interaction with other complex components, we hypothesized that the mutation might affect the catalytic activity of KAT8 in isolation, independently of its associated protein complexes. To investigate this, we expressed WT KAT8, R98W, and R99W in E. coli and purified these GST-tagged proteins via affinity purification. After cleaving the GST tag with protease, we further purified the recombinant proteins using gel filtration chromatography and conducted in vitro HAT assays with nucleosomes and histone octamers as substrates. The recombinant KAT8 proteins exhibited acetylation activity in vitro. Importantly, as observed earlier with the KAT8 complexes purified from cells, the recombinant R98W and R99W mutants displayed significantly weaker activity than WT KAT8 in acetylating H4K16 or H4K5/K8 on nucleosomes (Figure 4a). However, when using histone octamers as substrates, the mutants exhibited even higher HAT activities than WT KAT8 (Figure 4b). Compared to the HAT assays with purified KAT8 complexes, these results suggest that nucleosomal DNA has an even stronger impact on the enzymatic activity of KAT8 in isolation than in complexes. To eliminate the impact of DNA on KAT8’s catalytic activity, we compared WT and mutant KAT8 in the acetylation of a non-histone substrate, p53 [31], and found no difference between WT and mutants in acetylating p53K120 or in autoacetylation (Figure 4c and 4d).

Figure 4. The R98W mutation impedes the KAT8 Tudor-knot domain binding to nucleosome.

Figure 4.

Open in a new tab

a and b, Western blots analysis of the HAT assays with the indicated recombinant KAT8 proteins (WT, R98W, and R99W) and nucleosomes (a) or histone octamers (b) as substrates. c, Western blots analysis of the HAT assays with the indicated recombinant KAT8 proteins as in (a) and with recombinant p53 as substrate. d, Western blots analysis of autoacetylation of the indicated recombinant KAT8 proteins as in (a). e, KAT8 protein structure predicted by AlphaFold. Tudor-knot domain is shown in red, HAT domain in blue, and other regions in gray ribbons. The R98 and R99 residues are shown in red stick. f, Gel image (upper panel) and quantifications (lower panel) of EMSA assays with WT KAT8 or R98W and nucleosomes. Error bars represent SD from 3 biological replicates. Wilcoxon rank sum test was used to calculate p-value. g, Working model for KAT8 catalysis of nucleosomal histone acetylation: In normal cells, the Tudor-knot domain interacts with DNA through the positively charged loop comprising R98 and R99, rendering histone tails accessible to HAT for acetylation. In cancer, mutation of R98W impedes the interaction between Tudor-knot and nucleosomal DNA, the histone tails are thus not accessible to HAT, resulting in a global reduction of histone H4 acetylation. Figure is created with BioRender.com.

The results of the in vitro HAT assays suggest that nucleosomal DNA likely plays a crucial role in distinguishing WT and mutant KAT8. Since both R98 and R99 are positively charged, it is possible that these two residues may compete with histone tails for binding to nucleosomal DNA, thereby freeing up the histone tails for acetylation. AlphaFold predicts that the two arginine residues are located in a loop between the Tudor-knot and HAT domains (Figure 4e). If our hypothesis holds true, DNA binding may induce conformational changes in the Tudor-knot domain, thus opening up the catalytic site of HAT to substrate histone tails. To test this model, we determined the interaction between the KAT8 and nucleosomes using EMSA assays. With increasing protein-to-nucleosome ratios, WT KAT8 showed increased binding to nucleosomes (Figure 4f). However, compared to WT KAT8, the R98W mutant exhibited hindered binding to nucleosomes, as shown by a representative gel image of EMSA assays as well as quantifications (Figure 4f and Supplementary Table S4). In summary, our results support a model in which, upon KAT8 binding to nucleosomes, the Tudor-knot domain interacts with DNA through the positively charged loop comprising R98 and R99, rendering histone tails more accessible for substrate acetylation. Mutations of these essential residues in cancer reduce nucleosomal substrate accessibility to the HAT domain by impeding the interaction between the Tudor-knot domain and nucleosomal DNA, resulting in a global reduction of histone H4 acetylation (Figure 4g).

Discussion

In this study, we have focused on characterizing the most frequent mutation (R98W) of KAT8 found in human cancers and delved into how this mutation affects KAT8’s normal function. R98 is located within the Tudor-knot domain of the protein. This domain is also referred to as the Chromo-Barrel domain and has previously been identified as a reader motif for histone methylation in other proteins, such as RBP1 and MSL3 [32, 33]. However, our studies did not yield any evidence of interactions between the KAT8 Tudor-knot domain and either modified or unmodified histone tails (unpublished results), indicating that it may not function as a histone reader. Supporting this notion, our ChIP-seq experiments reveal that the R98W mutation does not affect the chromatin occupancy of KAT8. Instead, it impairs the catalytic activity of the HAT domain on nucleosomal histones. Our findings demonstrate that R98 is involved in the binding of KAT8 to nucleosomes, likely through electrostatic interactions between its positively charged side chain and negatively charged DNA. Intriguingly, the KAT8 orthologue in Drosophila has been reported to also bind RNA [34]. While RNA binding is also mediated through the Tudor-knot domain, it appears to depend on distinct residues (equivalent to Y90 and W103 in humans) [35]. It would be interesting to determine in future studies whether RNA and nucleosomal DNA synergize in regulating the catalytic activity of KAT8.

In human cancers, the mutation of R98 exclusively involves the substitution to tryptophan (W). Interestingly, this amino acid has also been reported to be mutated in patients with syndromic intellectual disability [36]. In contrast to cancer mutations, the arginine is mutated to glutamine (E) in patients with intellectual disability. The molecular basis for this mutation preference remains unclear. However, at the molecular level, both mutations affect the catalytic activity of KAT8 and lead to a global reduction in H4K16ac levels. It is worth noting that in human cancers, aside from R98, a few other hotspot mutations also involve arginine residues, including R99, R136, R140, and R224. Our findings demonstrate that the R99W mutation has a similar effect as R98W, impairing H4K16ac both in vitro and in cells. Unlike R98 and R99, which are located in the Tudor-knot domain, R136 and R140 reside in the unstructured linker region between the Tudor-knot and HAT domains, and they exhibit distinct mutation patterns (mutated to histidine or glycine rather than tryptophan). As histidine still retains a positive charge, it raises an interesting question of whether the R136 or R140 mutations also impact KAT8’s function through interactions with nucleosomes.

Another intriguing observation is that the R98W cancer mutation primarily affects KAT8’s activity toward histones in nucleosomes, while acetylation of non-histone substrates, such as p53, remains unaffected. Although further exploration is needed with additional non-histone substrates of KAT8, if this trend holds true, it may imply that R98W or other Tudor-knot domain mutations induce skewed acetylation patterns between histone and non-histone substrates. It is important to note that several non-histone substrates of KAT8 have been linked to tumor cell maintenance and anti-immune therapy [19, 37]. Consequently, the R98W mutation may possess unique features distinct from the simple loss-of-function HAT mutation of KAT8, such as the E350A mutation, in inducing global gene expression changes and tumorigenesis. Future studies are warranted to address these intriguing questions and provide a deeper understanding of the complex interplay between KAT8 and its substrates in cancer and disease pathogenesis.

Methods

Reagents

Human full length KAT8 (1-458aa) cDNA were constructed in pCDH-3Flag-HA, pLenti-Tet-On, and pGEX-6p-1 vectors for mammalian and bacteria expression. Mutations were introduced into vectors through mutagenesis (NEB). KAT8 sgRNA (TTTGCCATCAACTTCGTACA) was cloned into pLenti-V2 vector. Antibodies and oligos used in this study are listed in Supplementary Table S5.

Protein expression and purification.

For KAT8 complex purification, 1:1000 doxycycline were added 2 days ahead of cell collection. A549 TetOn cells were washed twice with PBS, and lysed with cell lysis buffer (50 mM Tris-HCl pH=7.5, 250 mM NaCl, 0.5% Triton X-100, 10% glycerol, 1 mM DTT, 1 mM PMSF, and 1x cOmplete™ EDTA-free Protease Inhibitor Cocktail) in 4°C for 30min and sonicated briefly. The cell lysates were incubated with anti-Flag M2 beads (Sigma) in 4° overnight. The beads were then washed four times with cell lysis buffer and eluted with elution buffer (50mM Tris-HCl pH=8.0, 100mM NaCl, and 0.4 mg/ml 3xFlag peptide) on ice for 30min.

For prokaryotic expressed KAT8 protein purification, full length KAT8 was constructed in pGEX-6p-1 vector and transformed in Rosetta™2 (DE3) pLysS competent cells (Novagen). GST-tagged proteins were induced with 0.4 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) at 16°C for 20 h in 2YT medium supplemented with 100 mM ZnCl2. Cell pellets were resuspended in lysis buffer (50 mM Tris-HCl pH=8, 300 mM NaCl, 5 μM ZnCl2, 1 mM phenylmethane-sulphonyl fluoride (PMSF), and 1x cOmplete™ EDTA-free Protease Inhibitor Cocktail) and lysed by sonication. The lysates were centrifuged at 17,000g for 15min, and the supernatants were incubated with Glutathione Sepharose® 4B beads (Sigma) at 4° for 2h. Beads were then washed with lysis buffer twice, proteins were eluted with elution buffer containing 50 mM Tris-HCl pH=8, 300mM NaCl, and 15mg/ml GSH. PreScission Protease (Sigma) was used to cleave GST tag. The eluted solution were loaded onto HiLoad 16/600 Superdex 200 column (GE Healthcare). KAT8 protein without GST tag were collected in volume ~72ml with 1 ml/min speed.

The histone octamer were purified as previously described[38]. Brifely, pET-His6-Sumo-H3-H4 and pET- His6-Sumo-H2A-H2B were transfromed in E. coli BL21(DE3) pLysS cells. Protein expression were induced by 0.5 mM IPTG at 37 °C for 5h in LB medium. Cell pellets were then resuspended in histone lysis buffer (20 mM Tris-HCl, pH 8.0, 2.0 M NaCl, 25 mM imidazole, 10% glycerol, 1 mM PMSF and 0.5 mM tris(2-carboxyethyl)phosphine (TCEP)). Resuspended cells were lysed using a high-pressure homogenizer (APV), and the lysate was clarified by centrifugation at 38,700g at 4 °C for 1 hour. The resulting supernatant was collected and loaded onto a 5-ml HisTrap FF column (GE Healthcare). Bound proteins were eluted with Ni-buffer supplemented with 250 mM imidazole. His6-Sumo tags were cleaved by incubating the eluted proteins with purified ULP1 Sumo protease at a 1:1,000 ratio at 4°C for 4h. The histones were further purified by HiLoad 16/600 Superdex 200 column (GE Healthcare).

Full-length p53 constructed in the pET21a vector were transformed in Rosetta™2(DE3) pLysS competent cells and the His-tagged proteins were purified using Ni-NTA resins following the manufacture’s instruction. The eluted proteins were dialyzed in buffer containing 50 mM Tris-HCl pH7.5, 100 mM NaCl, 1mM DTT and 10% glycerol to remove imidazole.

In vitro histone acetyltransferase (HAT) assay.

KAT8 complex or bacteria purified KAT8 were mixed with recombinant human mono-nucleosomes (100 nM, EpiCypher), histone octamers, or His-p53 protein in 50 μL of HAT assay buffer containing 50 mM Tris-HCl pH=8.0, 100 mM NaCl, 100 mM Acetyl-CoA, 1 mM DTT, 10% glycerol and cOmplete™ EDTA-free Protease Inhibitor. The mixtures were incubated at 37°C together with substrates. Reactions were stopped by adding 2xSDS loading buffer and boiling at 95°C for 5 min.

Electrophoretic mobility shift assay (EMSA).

2 pmol nucleosome or 601 DNA were incubated with KAT8 protein in corresponding molar ratio in SDE buffer (5% sucrose and 1 mM DTT in water). After 20 min incubation in 25°C, samples were loaded into 6% TBE PAGE gel with 120V for 45min running. TBE PAGE gels were stained by 0.01% ethidium bromide buffer for 10min, results were exposed by Bio-Rad ChemiDoc MP.

Cell culture and virus transduction.

A549 and H1703 cell lines were purchased from ATCC. A549 cells are cultured in DMEM (Corning) and H1703 cells are cultured in RPMI1640 (Corning). Both media were supplemented with 10% fetal bovine serum (Sigma), 1mM sodium pyruvate (Corning), 1% non-essential amino acids (HyClone), and 1% penicillin–streptomycin (Corning). pCDH-KAT8-FL. pLenti-TetOn-KAT8-FL, and pLentiCRISPR V2 (KAT8 sgRNA: TTTGCCATCAACTTCGTACA) vectors were transfected into 293T cells together with psPAX2 and pMD2.G plasmids at a 2:2:1 ratio using the X-TremeGENE HP DNA transfection reagent (Sigma). Virus supernatant was collected and filtered through a 0.45 μm syringe filter 48h post transfection. For cell transduction, 4×106 cells were seeded in 10 cm dishes 24h before transduction. Cells were transduced with 2 ml lentivirus containing 4μg/ml polybrene. Cells were selected with 2 μg/ml puromycin or 10 μg/ml Blasticidin (Thermo Fisher Scientific) for 4–6 days.

Protein immunoprecipitation

For KAT8 complex immunoprecipitation, 1:1000 doxycycline were added 2 days ahead of cell collection. A549 TetOn cells were washed twice with PBS, and lysed with cell lysis buffer (50 mM Tris-HCl pH=7.5, 250 mM NaCl, 0.5% Triton X-100, 10% glycerol, 1 mM DTT, 1 mM PMSF, and 1x cOmplete™ EDTA-free Protease Inhibitor Cocktail) in 4°C for 30min and sonicated briefly. 5% supernatant were collected as 5% input. The cell lysates were incubated with anti-Flag M2 beads (Sigma) in 4° overnight. The beads were then washed four times with cell lysis buffer and boiled with SDS loading buffer (50 mM Tris-HCl pH=6.8, 2% SDS, and 6% glycerol) for western blot.

Western blot analysis.

Cells were washed with PBS and lysed with SDS loading buffer (50 mM Tris-HCl pH=6.8, 2% SDS, and 6% glycerol), cell lysis were boiled at 95°C for 20 min. In vitro HAT assay samples were mixed with equal volume 2x SDS loading buffer and boiled at 95°C for 5 min. 15% SDS-PAGE gels were used for histones and in vitro HAT assays, 7.5% gels were used for others. All the western blot in Figures have at least 2 biological replicates and show representative one.

ChIP and ChIP-seq analysis.

1×107 A549 cells were crosslinked by 1% formaldehyde in PBS for 10 min and stopped by adding 125 mM glycine for 5min at room temperature. Nuclei were isolated using cell lysis buffer (5 mM PIPES pH=8.0, 85 mM KCl, 1% NP-40, and protease inhibitors) for 20 min at 4°C. The isolated nuclei were resuspended in MNase buffer (50 mM Tris-HCl pH=8.0, 1 mM CaCl2, and 300 mM sucrose), 50U Miccrococal Nuclease (Worthington, Cat No. 4798) were added with 10 min incubation in 37°C, reaction was stopped by adding 500 mM EDTA. After brief sonication by Covaris E220 Evo, supernatant were collected with 13,000rpm at 4°C for 10 min. For H4K16ac ChIP, Drosophila chromatin spike-in were added to 1% of total chromatin. The chromatin samples were incubated with corresponding antibodies at 4°C overnight. Dynabeads™ Protein G (Thermo Fisher Scientific) were added and incubated for 1 h, and washed twice with low salt wash buffer (20 mM Tris-HCl pH=8, 150 mM NaCl, 2 mM EDTA, 1% Triton X-100, and 0.1% SDS), twice with high salt wash buffer (20 mM Tris-HCl pH=8, 500 mM NaCl, 2 mM EDTA, 1% Triton X-100, and 0.1% SDS), once with LiCl wash buffer (20 mM Tris-HCl pH=8, 250 mM LiCl, 1 mM EDTA, 1% NP-40, and 1% Na-Deoxycholate), and once with TE buffer (10 mM Tris-HCl pH=8, 1 mM EDTA). Bound DNA was eluted using fresh 50 mM NaHCO3 and 1% SDS, reverse crosslinked and purified using PCR purification kit (Qiagen).

ChIP-seq libraries were constructed using the KAPA Hyper Prep Kit (Roche) and sequenced by Illumina NovaSeq 6000 at the Van Andel Institute Genomics Core. Fastq reads were mapped to hg38 human genome and dmel_r6.24 drosophila genome (only for spike-in) by HISAT2 (v2.1.0)[39] with --no-spliced-alignment -k 1 -X 1000. Spike-in adjust ratios were calculated as (mapping ratioGFP sg)/(mapping ratioSample). Wig files were generated by Danpos2 (v2.2.2)[40]. BigWig files were transformed from wig files with wigToBigWig (v4). Heatmaps and profiles were generated by Danpos2. Heatmaps were visualized by TreeView.

RNA-seq analysis.

Total RNA was extracted from A549 cells using RNeasy Plus Mini Kit (Qiagen). RNA-seq libraries were prepared using KAPA RNA HyperPrep Kit with RiboErase (HMR) (Roche) or KAPA Stranded mRNA-Seq Kit (Roche) following the manufacturer’s instructions and sequenced by Illumina NovaSeq 6000. Fastq reads were mapped to hg38 human genome by HISAT2 (v2.1.0)[39] with -k 1. CPM (counts per million) and fold change values were calculated by HTSeq (v0.11.3)[41] with --stranded=no -a 0 and edgeR (v3.16.5)[42] with Trimmed Mean of M-values (TMM) and Exact test model. differentially expressed genes (DEGs) were filtered by FDR<0.05 and FC>2. Gene Ontology molecular function term enrichment was done by GeneOntology. Wig files were generated by Danpos2 (v2.2.2)[40]. BigWig files were transformed from wig files with wigToBigWig (v4). Tracks were visualized by IGV (v2.7.2)[43].

Supplementary Material

1

Research highlights:

  • R98W is a hotspot mutation in human cancers

  • The R98W mutation impairs KAT8-mediated acetylation of nucleosomal histones

  • The R98W mutation may impede accessibility of the histone tails to HAT

  • The R98W mutation leads to a reduction in global H4K16 acetylation level

Acknowledgements.

We thank Yali Dou and Scott Rothbart for sharing reagents. We thank Marie Adams and the Genomics Core at Van Andel Institute for NGS sequencing. Computation for the work described in this paper was supported by the High-Performance Cluster and Cloud Computing (HPC3) Resource at the Van Andel Research Institute. This work was supported in part by grants from NIH/NCI (CA255506 and CA260666) to H.W. and NIH/NCI (CA204020 and CA268440) to X.S.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Competing interests. The authors declare no competing interests.

Declaration of generative AI and AI-assisted technologies in the writing process. During the preparation of this work the authors used ChatGPT in order to edit the grammar. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Data availability.

The authors declare that the data supporting the findings of this study are available within the paper and its Supplementary Information. All RNA-seq and ChIP-seq data described in the manuscript have been deposited in the NCBI Gene Expression Omnibus (GEO) database and are accessible through the GEO SuperSeries accession number GSE245009 (Reviewers’ token: ahyhoqgyflepfgl).

References

  • [1].Hilfiker A, Hilfiker-Kleiner D, Pannuti A, Lucchesi JC. mof, a putative acetyl transferase gene related to the Tip60 and MOZ human genes and to the SAS genes of yeast, is required for dosage compensation in Drosophila. EMBO J. 1997;16:2054–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [2].Akhtar A, Becker PB. The histone H4 acetyltransferase MOF uses a C2HC zinc finger for substrate recognition. EMBO reports. 2001;2:113–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [3].Smith ER, Pannuti A, Gu W, Steurnagel A, Cook RG, Allis CD, et al. The drosophila MSL complex acetylates histone H4 at lysine 16, a chromatin modification linked to dosage compensation. Mol Cell Biol. 2000;20:312–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [4].Bone JR, Kuroda MI. Dosage compensation regulatory proteins and the evolution of sex chromosomes in Drosophila. Genetics. 1996;144:705–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [5].Bone JR, Lavender J, Richman R, Palmer MJ, Turner BM, Kuroda MI. Acetylated histone H4 on the male X chromosome is associated with dosage compensation in Drosophila. Genes Dev. 1994;8:96–104. [DOI] [PubMed] [Google Scholar]
  • [6].Taipale M, Rea S, Richter K, Vilar A, Lichter P, Imhof A, et al. hMOF histone acetyltransferase is required for histone H4 lysine 16 acetylation in mammalian cells. Mol Cell Biol. 2005;25:6798–810. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [7].Dou Y, Milne TA, Tackett AJ, Smith ER, Fukuda A, Wysocka J, et al. Physical association and coordinate function of the H3 K4 methyltransferase MLL1 and the H4 K16 acetyltransferase MOF. Cell. 2005;121:873–85. [DOI] [PubMed] [Google Scholar]
  • [8].Raja SJ, Charapitsa I, Conrad T, Vaquerizas JM, Gebhardt P, Holz H, et al. The nonspecific lethal complex is a transcriptional regulator in Drosophila. Molecular cell. 2010;38:827–41. [DOI] [PubMed] [Google Scholar]
  • [9].Mendjan S, Taipale M, Kind J, Holz H, Gebhardt P, Schelder M, et al. Nuclear pore components are involved in the transcriptional regulation of dosage compensation in Drosophila. Molecular cell. 2006;21:811–23. [DOI] [PubMed] [Google Scholar]
  • [10].Li X, Wu L, Corsa CA, Kunkel S, Dou Y. Two mammalian MOF complexes regulate transcription activation by distinct mechanisms. Molecular cell. 2009;36:290–301. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [11].Sheikh BN, Guhathakurta S, Akhtar A. The non-specific lethal (NSL) complex at the crossroads of transcriptional control and cellular homeostasis. EMBO reports. 2019;20:e47630. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [12].Kind J, Vaquerizas JM, Gebhardt P, Gentzel M, Luscombe NM, Bertone P, et al. Genome-wide analysis reveals MOF as a key regulator of dosage compensation and gene expression in Drosophila. Cell. 2008;133:813–28. [DOI] [PubMed] [Google Scholar]
  • [13].Radzisheuskaya A, Shliaha PV, Grinev VV, Shlyueva D, Damhofer H, Koche R, et al. Complex-dependent histone acetyltransferase activity of KAT8 determines its role in transcription and cellular homeostasis. Molecular cell. 2021;81:1749–65 e8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [14].Neuwald AF, Landsman D. GCN5-related histone N-acetyltransferases belong to a diverse superfamily that includes the yeast SPT10 protein. Trends Biochem Sci. 1997;22:154–5. [DOI] [PubMed] [Google Scholar]
  • [15].Borrow J, Stanton VP Jr., Andresen JM, Becher R, Behm FG, Chaganti RS, et al. The translocation t(8;16)(p11;p13) of acute myeloid leukaemia fuses a putative acetyltransferase to the CREB-binding protein. Nat Genet. 1996;14:33–41. [DOI] [PubMed] [Google Scholar]
  • [16].Cai Y, Jin J, Swanson SK, Cole MD, Choi SH, Florens L, et al. Subunit composition and substrate specificity of a MOF-containing histone acetyltransferase distinct from the male-specific lethal (MSL) complex. J Biol Chem. 2010;285:4268–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [17].Singh M, Bacolla A, Chaudhary S, Hunt CR, Pandita S, Chauhan R, et al. Histone Acetyltransferase MOF Orchestrates Outcomes at the Crossroad of Oncogenesis, DNA Damage Response, Proliferation, and Stem Cell Development. Mol Cell Biol. 2020;40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [18].Sykes SM, Stanek TJ, Frank A, Murphy ME, McMahon SB. Acetylation of the DNA binding domain regulates transcription-independent apoptosis by p53. J Biol Chem. 2009;284:20197–205. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [19].Chen Z, Ye X, Tang N, Shen S, Li Z, Niu X, et al. The histone acetylranseferase hMOF acetylates Nrf2 and regulates anti-drug responses in human non-small cell lung cancer. Br J Pharmacol. 2014;171:3196–211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [20].Gupta A, Sharma GG, Young CS, Agarwal M, Smith ER, Paull TT, et al. Involvement of human MOF in ATM function. Mol Cell Biol. 2005;25:5292–305. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [21].Gupta A, Hunt CR, Hegde ML, Chakraborty S, Chakraborty S, Udayakumar D, et al. MOF phosphorylation by ATM regulates 53BP1-mediated double-strand break repair pathway choice. Cell reports. 2014;8:177–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [22].Khoa LTP, Tsan YC, Mao F, Kremer DM, Sajjakulnukit P, Zhang L, et al. Histone Acetyltransferase MOF Blocks Acquisition of Quiescence in Ground-State ESCs through Activating Fatty Acid Oxidation. Cell Stem Cell. 2020;27:441–58 e10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [23].Li X, Li L, Pandey R, Byun JS, Gardner K, Qin Z, et al. The histone acetyltransferase MOF is a key regulator of the embryonic stem cell core transcriptional network. Cell Stem Cell. 2012;11:163–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Tate JG, Bamford S, Jubb HC, Sondka Z, Beare DM, Bindal N, et al. COSMIC: the Catalogue Of Somatic Mutations In Cancer. Nucleic Acids Res. 2019;47:D941–D7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [25].Kim JY, Yu J, Abdulkadir SA, Chakravarti D. KAT8 Regulates Androgen Signaling in Prostate Cancer Cells. Mol Endocrinol. 2016;30:925–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [26].Zhang J, Liu H, Pan H, Yang Y, Huang G, Yang Y, et al. The histone acetyltransferase hMOF suppresses hepatocellular carcinoma growth. Biochem Biophys Res Commun. 2014;452:575–80. [DOI] [PubMed] [Google Scholar]
  • [27].Kadlec J, Hallacli E, Lipp M, Holz H, Sanchez-Weatherby J, Cusack S, et al. Structural basis for MOF and MSL3 recruitment into the dosage compensation complex by MSL1. Nat Struct Mol Biol. 2011;18:142–9. [DOI] [PubMed] [Google Scholar]
  • [28].Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, et al. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal. 2013;6:pl1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [29].Karch KR, Coradin M, Zandarashvili L, Kan ZY, Gerace M, Englander SW, et al. Hydrogen-Deuterium Exchange Coupled to Top- and Middle-Down Mass Spectrometry Reveals Histone Tail Dynamics before and after Nucleosome Assembly. Structure. 2018;26:1651–63 e3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [30].Murphy KJ, Cutter AR, Fang H, Postnikov YV, Bustin M, Hayes JJ. HMGN1 and 2 remodel core and linker histone tail domains within chromatin. Nucleic Acids Res. 2017;45:9917–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [31].Sykes SM, Mellert HS, Holbert MA, Li K, Marmorstein R, Lane WS, et al. Acetylation of the p53 DNA-binding domain regulates apoptosis induction. Mol Cell. 2006;24:841–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [32].Gong W, Zhou T, Mo J, Perrett S, Wang J, Feng Y. Structural insight into recognition of methylated histone tails by retinoblastoma-binding protein 1. J Biol Chem. 2012;287:8531–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [33].Moore SA, Ferhatoglu Y, Jia Y, Al-Jiab RA, Scott MJ. Structural and biochemical studies on the chromo-barrel domain of male specific lethal 3 (MSL3) reveal a binding preference for mono- or dimethyllysine 20 on histone H4. J Biol Chem. 2010;285:40879–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [34].Akhtar A, Zink D, Becker PB. Chromodomains are protein-RNA interaction modules. Nature. 2000;407:405–9. [DOI] [PubMed] [Google Scholar]
  • [35].Conrad T, Cavalli FM, Holz H, Hallacli E, Kind J, Ilik I, et al. The MOF chromobarrel domain controls genome-wide H4K16 acetylation and spreading of the MSL complex. Dev Cell. 2012;22:610–24. [DOI] [PubMed] [Google Scholar]
  • [36].Li L, Ghorbani M, Weisz-Hubshman M, Rousseau J, Thiffault I, Schnur RE, et al. Lysine acetyltransferase 8 is involved in cerebral development and syndromic intellectual disability. J Clin Invest. 2020;130:1431–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [37].Wu Y, Zhou L, Zou Y, Zhang Y, Zhang M, Xu L, et al. Disrupting the phase separation of KAT8-IRF1 diminishes PD-L1 expression and promotes antitumor immunity. Nat Cancer. 2023;4:382–400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [38].Xu TH, Liu M, Zhou XE, Liang G, Zhao G, Xu HE, et al. Structure of nucleosome-bound DNA methyltransferases DNMT3A and DNMT3B. Nature. 2020;586:151–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [39].Kim D, Paggi JM, Park C, Bennett C, Salzberg SL. Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotype. Nat Biotechnol. 2019;37:907–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [40].Chen K, Chen Z, Wu D, Zhang L, Lin X, Su J, et al. Broad H3K4me3 is associated with increased transcription elongation and enhancer activity at tumor-suppressor genes. Nat Genet. 2015;47:1149–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [41].Anders S, Pyl PT, Huber W. HTSeq--a Python framework to work with high-throughput sequencing data. Bioinformatics. 2015;31:166–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [42].Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010;26:139–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [43].Robinson JT, Thorvaldsdottir H, Winckler W, Guttman M, Lander ES, Getz G, et al. Integrative genomics viewer. Nat Biotechnol. 2011;29:24–6. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1955114-supplement-1.xlsx (753.2KB, xlsx)

Data Availability Statement

The authors declare that the data supporting the findings of this study are available within the paper and its Supplementary Information. All RNA-seq and ChIP-seq data described in the manuscript have been deposited in the NCBI Gene Expression Omnibus (GEO) database and are accessible through the GEO SuperSeries accession number GSE245009 (Reviewers’ token: ahyhoqgyflepfgl).

RESOURCES