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 21;436(7):168414. doi: 10.1016/j.jmb.2023.168414

The Tudor-knot domain of KAT5 regulates nucleosomal substrate acetylation

Fan Xuan 1,#, Hongwen Xuan 1,#, Mengying Huang 1, He Wei 2, Han Xu 2,*, Xiaobing Shi 1,*, Hong Wen 1,*
PMCID: PMC10957329  NIHMSID: NIHMS1955323  PMID: 38141874

Abstract

The lysine acetyltransferase KAT5 is a pivotal enzyme responsible for catalyzing histone H4 acetylation in cells. In addition to its indispensable HAT domain, KAT5 also encompasses a conserved Tudor-knot domain at its N-terminus. However, the function of this domain remains elusive, with conflicting findings regarding its role as a histone reader. In our study, we have employed a CRISPR tiling array approach and unveiled the Tudor-knot motif as an essential domain for cell survival. The Tudor-knot domain does not bind to histone tails and is not required for KAT5’s chromatin occupancy. However, its absence leads to a global reduction in histone acetylation, accompanied with genome-wide alterations in gene expression that consequently result in diminished cell viability. Mechanistically, we find that the Tudor-knot domain regulates KAT5’s HAT activity on nucleosomes by fine-tuning substrate accessibility. In summary, our study uncovers the Tudor-knot motif as an essential domain for cell survival and reveals its critical role in modulating KAT5’s catalytic efficiency on nucleosome and KAT5-dependent transcriptional programs critical for cell viability.

Keywords: KAT5, TIP60, Tudor-knot domain, nucleosome, H4 acetylation

Graphical abstract

graphic file with name nihms-1955323-f0001.jpg

Introduction

The lysine acetyltransferase 5 (KAT5), also known as TIP60, was originally identified as a 60 kDa protein that interacts with HIV Tat [1]. KAT5 is highly conserved from yeast to human [2]. In humans, it assembles into a multi-mega Dalton protein complex alongside approximately 15 other proteins, with its yeast counterpart referred to as NuA4, short for nucleosome acetyltransferase of H4 [3–5]. KAT5 catalyzes acetylation primarily on histone H4, but also on other core histones like H2A and H3, as well as an array of non-histone proteins, including the pivotal p53 [6–12]. This multi-substrate adaptability positions KAT5 at the nexus of diverse cellular processes, including transcription [13–16], DNA damage responses [17–20], and signal transduction [9]. Its yeast counterpart, Esa1, is the only essential histone acetyltransferase in budding yeast [21, 22], and inactivation of KAT5 in mice results in embryonic lethality [23].

The KAT5 protein comprises two well-defined domains: a C-terminal MYST-type HAT domain and an N-terminal Tudor-knot domain. Recent cryo-EM analysis of the KAT5 complex has unveiled a conserved 3-dimensional structure of the HAT module, although the highly mobile Tudor-knot motif eluded observation in the structure [24]. The Tudor-knot domain is also referred to as the Chromo-Barrel domain and has been identified as a reader motif for histone methylation in some other proteins, such as RBP1 and MSL3 [25, 26]. However, its role as a histone reader within KAT5 remains a subject of debate [27–29]. Using peptide pulldown assays, Sun et al. reported its binding to H3K9 trimethylated peptides [27], while Jeong et al. showed its recognition of H3K4 monomethylation [28]. In contrast, a more extensive study by Zhang et al., employing fluorescence polarization and isothermal titration calorimetry assays, failed to detect interactions between the KAT5 Tudor-knot and modified histone peptides [29]. Therefore, the functional significance of this domain in human KAT5 has remained elusive. In this study, we present evidence indicating that while the Tudor-knot domain may not serve as a histone reader, its absence significantly diminishes KAT5’s HAT activity, both in vitro and in cells. Our findings suggest that the Tudor-knot domain of KAT5 plays a critical role in modulating the accessibility of HAT to the nucleosome substrate.

Results and Discussion

CRISPR tilling array identifies the KAT5 Tudor-knot motif as an essential domain for cell survival

To conduct a comprehensive analysis of the KAT5 protein, we took advantage of the CRISPR-Cas9 tiling array [30], a powerful tool for the evaluation of protein domain function in situ. The underlying rationale for tiling-array based CRISPR screen is that those sgRNAs targeting DNA sequences coding for essential protein domains result in more significant dropout phenotype, compared to those associated with amino acids outside the domains [31, 32]. In the screening process, a lentiviral pool of sgRNAs with high-density coverage across the entire coding sequences of genes of interest are delivered into target cells with the virus dose at a low multiplicity of infection (MOI of ~0.3), thus to ensure each cell expressed only a single sgRNA. These sgRNAs are subsequently assessed by the CRISPR-knockout hyper-sensitivity (CKHS) analysis using the ProTiler algorithm that we developed previously [30]. In this study, we conducted CRISPR tiling screens in three cell lines expressing wild-type KAT5: the acute myeloid leukemia (AML) cell lines MV4-11 and OCI-AML2, and the non-small cell lung cancer (NSCLC) cell line NCI-H23. KAT5 is an essential gene in these cell lines, relying on the catalytic activity of its HAT for function [2]. Indeed, our screens identified the HAT domain as an essential component in all three cell lines. Unexpectedly, we also observed sgRNAs with significant CKHS scores mapped to the N-terminal Tudor-knot domain (Figure 1A), indicating that the Tudor-knot domain is also indispensable for cell survival.

Figure 1. The Tudor-knot domain of KAT5 is essential for cell survival.

Figure 1.

Open in a new tab

(A) CRISPR-knockout hyper-sensitivity (CKHS) profile of KAT5 tilling sgRNA screen in MV4–11, OCI-AML2, and NCI-H23 cell lines. X-axis is amino acid location of KAT5 (aa1–513). Y-axis is Z-score of sgRNA count with negative values indicating dropout effect. Each dot represents an sgRNA mapped to the amino acid location in X-axis. Grey and blue dots represent filtered and remaining sgRNAs, respectively. Red lines show the segmented protein regions and sgRNAs dropout signal levels. (B) Cell growth curves of MV4–11, OCI-AML2, and H1299 cell lines. Error bars represent the standard deviation (SD) of 3 replicates. All p-values were calculated relative to the sgGFP cells. *, **, ***, **** indicate p-values <0.05, <0.01, <0.001, and 0.0001 respectively. N.S.: not significant (Two-way ANOVA). (C) Western blot analysis of the H4-PanAc, H4, and KAT5 protein levels of whole cell lysates from the indicated cell lines. sgGFP was used as a control sgRNA. FL: full length. ΔN : N-terminal Tudor-knot deletion.

To validate our CRISPR screen results, we knocked out (KO) endogenous KAT5 using two sgRNAs, followed by the reintroduction of sgRNA-resistant KAT5 in two forms: the full-length (FL) protein or the N-terminal Tudor-knot deletion mutant (ΔN). Utilizing this rescue system, we evaluated the essentiality of the KAT5 Tudor-knot domain in MV4-11 and OCI-AML2 cells by monitoring cell growth through CTG assays. As shown in Figure 1B, KO of KAT5 substantially inhibited cell growth, and this growth impairment could be rescued by stable overexpression of the FL KAT5 protein. In contrast, the KAT5-ΔN mutant failed to restore cell growth, demonstrating the essential nature of the Tudor-knot domain for cell survival. A similar pattern was observed in another NSCLC cell line, H1299 (Figure 1B). KAT5 is well-known for catalyzing histone H4 acetylation at multiple sites, primarily targeting residues near the N-terminal end of the tail, including H4K5, K8, and K12 [3–5]. Employing an H4-PanAc antibody, which specifically recognizes H4K5acK8ac (hereafter referred to as H4ac), we observed a marked reduction in global H4ac levels upon KAT5 KO in all three cell lines (Figure 1C). Notably, the reduction in H4ac was largely restored by stable expression of FL KAT5 but not the KAT5-ΔN mutant in the cells. Collectively, these findings indicate that the KAT5 Tudor-knot domain is critical for preserving cell viability and maintaining global H4ac levels in cells.

The Tudor-knot domain is not required for KAT5 chromatin recruitment.

The Tudor-knot domain of KAT5 has been previously reported to recognize H3K9me3 [27] and H3K4me1 [28]. These recognitions are likely achieved through a hydrophobic pocket formed by several aromatic residues including Y47 and F50 (Supplementary Figure 1A), as mutations of these residues were shown to disrupt histone bindings [27, 28]. However, the validity of these interactions has been subject to debate, as a more comprehensive study failed to detect any interactions between KAT5 Tudor-knot with these modified peptides [29]. To address this controversy and investigate whether the KAT5 Tudor-knot domain can potentially recognize methylation on other histone residues, we conducted an assessment using a modified histone peptide array. This array included peptides with methylation and acetylation modifications on all major lysine and arginine residues of core histones. Consistent with the findings of Zhang et al. [29], our analysis did not reveal binding of the Tudor-knot domain to any modified or unmodified histone peptide, as observed through both the histone peptide array and peptide pulldown assays (Supplementary Figure 1B and 1C). Our data thus support the conclusion that the KAT5 Tudor-knot domain does not function as a histone reader.

Despite the lack of in vitro histone binding by the Tudor-knot domain, its deletion has a notable impact on global histone H4 acetylation levels (Figure 1C). This raises the possibility that the Tudor-knot domain may be necessary for KAT5’s recruitment to chromatin. To explore this possibility, we generated stable cell lines expressing HA-tagged FL KAT5 and the ΔN mutant in the H1299 cell line, then conducted chromatin immunoprecipitation assays coupled with next-generation sequencing (ChIP-seq). Given that the mutation of the Y47 residue has previously been associated with impaired chromatin recruitment and functionality of KAT5 in cells [27, 28], we also included the KAT5-Y47A mutant in this experiment (Figure 2A). ChIP-seq analyses using anti-HA and anti-H4-PanAc antibodies showed a strong colocalization of KAT5 with H4ac around transcription start sites (Figure 2B and 2C, left panels). Notably, cells expressing the KAT5-ΔN or Y47A mutant displayed a marked reduction in H4ac levels (Figure 2A and 2C). However, surprisingly, the deletion of the Tudor-knot domain or the Y47A mutation did not seem to affect KAT5’s occupancy on chromatin (Figure 2B). The two mutants even exhibited slightly higher ChIP-seq signals, possibly due to their higher protein expression levels in the cells (Figure 2A). Taken together, these results suggest that the Tudor-knot domain of KAT5 plays a role in regulating global H4ac levels through a mechanism unrelated to KAT5 chromatin recruitment.

Figure 2. The Tudor-knot domain is required for KAT5-mediated H4 acetylation but not for KAT5 chromatin recruitment.

Figure 2.

Open in a new tab

(A) Western blot analysis with the indicated antibodies of whole cell lysates from H1299 cell lines stably expressing HA-KAT5-FL, HA-KAT5-Y47A, and HA-KAT5ΔN. (B) Heatmaps of HA-KAT5 ChIP-seq signal densities centered on transcription starting sites (TSS) across a ± 5kb window of all genes. (C) Heatmaps of H4-PanAc ChIP-seq densities centered on TSS across a ± 5kb window of all genes. Average profiles of ChIP-seq densities are shown on top of each column.

The Tudor-knot domain regulates KAT5 HAT activity on nucleosomes.

The intriguing observation that alterations in the Tudor-knot domain of KAT5 affect global H4ac levels but not KAT5’s chromatin occupancy prompted us to investigate whether the Tudor-knot domain directly regulates KAT5’s HAT activity. To address this hypothesis, we purified the KAT5 complex from cells expressing FL KAT5, the Y47A mutant, and the KAT5-ΔN mutant, and conducted in vitro HAT assays using mono-nucleosomes as substrates (Figure 3A). The FL KAT5 complex exhibited robust nucleosome acetylation within a 30-minute reaction window. In contrast, the HAT activities of the KAT5 complexes associated with the ΔN and Y47A mutants were barely detectable (Figure 3B). Additionally, a dose-dependent HAT assay demonstrated that the ΔN and Y47A mutant-associated complexes had significantly reduced enzymatic activity acetylating histones in nucleosomes (Figure 3C).

Figure 3. The Tudor-knot domain is essential for the KAT5 complex to acetylate histones in nucleosomes.

Figure 3.

Open in a new tab

(A) Schematic of in vitro HAT assays using the indicated KAT5 complexes (Flag-KAT5-FL, Flag-KAT5-Y47A, and Flag-KAT5-ΔN) with mono-nucleosome as substrate. (B) Western blots analysis of the HAT assays with the indicated KAT complexes as in (A) and time courses. (C) Western blots analysis of the HAT assays with the indicated KAT complexes as in (A) and doses. (D) Western blots analysis of the indicated KAT5 complex components of Flag-IPs in cells expressing Flag-KAT5-FL, Flag-KAT5-Y47A, and Flag-KAT5-ΔN. Cells expressing an empty vector was used as a negative control. (E) Schematic of in vitro HAT assays using the indicated KAT5 complexes with histone octamer as substrate. (F) Western blots analysis of the HAT assays with the indicated KAT complexes as in (E) and time courses.

We reasoned that the Tudor-knot domain might regulate HAT activity through either an intra-domain interaction between the Tudor-knot and HAT domains within the KAT5 protein or through protein-protein interactions between KAT5 and other complex components. To investigate this, we first tested the effect of Tudor-knot mutations on KAT5 in isolation, without the presence of other complex components. We purified GST-tagged KAT5-FL, Y47A, and KAT5-ΔN from E. coli and performed HAT assays using the recombinant proteins. Surprisingly, the results revealed that deletion or mutation of the Tudor-knot domain did not affect the HAT activity of the recombinant KAT5 proteins (Supplementary Figure 2A and 2B). Next, we aimed to determine whether Tudor-knot alterations have impact on protein-protein interactions between KAT5 and other complex components. To this end, we conducted immunoprecipitation (IP) experiments using Flag M2 beads in cells expressing Flag-tagged FL KAT5, Y47A, and the KAT5-ΔN mutant, and assessed other complex subunits co-IPed with KAT5. However, no discernible differences were observed by Western blot analysis between FL KAT5 and the mutant KAT5 variants in their interactions with other complex subunits (Figure 3D). These findings suggest that deletion or mutation of the Tudor-knot domain does not compromise the integrity of the KAT5 complex.

In addition to testing HAT activity on nucleosomes, we also examined the HAT activity of the KAT5 complexes on histones using histone octamers as substrates (Figure 3E). To our surprise, both the FL KAT5- and the ΔN mutant-associated complexes exhibited robust acetylation activity on histones, with no significant differences observed between the two complexes (Figure 3F). The major difference between nucleosomes and histone octamers is the presence of DNA, which is exclusive to nucleosomes. Given its negatively charged nature, nucleosomal DNA interacts with positively charged residues on histones, causing histone tails to adopt a “closed” conformation. In contrast, histone tails in octamers are in a more “open” state and thus are readily accessible to histone-modifying enzymes. Indeed, a direct comparison using in vitro HAT assays with the KAT5 complex demonstrated that the octamer is a superior substrate compared to the nucleosome (Supplementary Figure 2C). In summary, these results suggest that the Tudor-knot domain regulates KAT5’s HAT activity on nucleosomes likely by fine-tuning substrate accessibility.

The Tudor-knot domain is important for KAT5-dependent transcriptional regulation.

The importance of the Tudor-knot domain in KAT5’s function is underscored by its necessity for cell survival. The deletion of the Tudor-knot domain or even a single-point mutation like Y47A failed to restore the growth of cells with endogenous KAT5 depleted (Figure 4A and 4B). KAT5 is a transcriptional coactivator known for its role in regulating gene expression through modulation of histone acetylation [2]. Given that alterations in the Tudor-knot domain led to a decrease in global histone H4 acetylation levels (Figure 4A), we posited that the Tudor-knot domain regulates cell viability by influencing KAT5-dependent transcriptional programs. To test this hypothesis, we conducted RNA-seq analysis in cells with KAT5 knocked out, and the KO cells rescued with FL KAT5, Y47A, and the KAT5-ΔN mutant. Depletion of KAT5 resulted in global gene expression changes, with 2376 genes upregulated and 512 genes downregulated (fold change > 1.5 and false discovery rate < 0.05) in the KAT5 KO cells (Figure 4C and Supplementary Table S1). Importantly, reintroducing FL KAT5 back into the KO cells largely restored the global gene expression to a pattern similar to that of control cells. In contrast, the gene expression patterns in the KO cells rescued with the Y47A or KAT5-ΔN mutant still remained similar to those of the KAT5 KO cells (Figure 4C). The KEGG pathway enrichments were also largely the same between these cells (Figure 4D and Supplementary Table S2). Finally, we assessed the correlation between H4 acetylation and gene expression changes in the KAT5 KO cells rescued with the Y47A and KAT5-ΔN mutants compared to the KO cells rescued with FL KAT5. The results demonstrated a strong correlation between changes in gene expression and changes in H4 acetylation levels (Figure 4E). Taken together, these findings suggest that the Tudor-knot domain regulates KAT5-dependent transcriptional programs through the modulation of histone H4 acetylation.

Figure 4. The Tudor-knot domain is important for KAT5-dependent transcriptional regulation.

Figure 4.

Open in a new tab

(A) Western blot analysis with the indicated antibodies of whole cell lysates from H1299 cell lines treated with KAT5 sgRNAs and the KAT5 knockout cells stably expressing HA-KAT5-FL, HA-KAT5-Y47A, and HA-KAT5ΔN. Cells treated with sgGFP was used as a control. (B) Cell growth curves of the cells as indicated in (A). Error bars represent the standard deviation (SD) of 3 replicates. All p-values were calculated relative to the sgGFP cells. *, **, ***, **** indicate p-values <0.05, <0.01, <0.001, and 0.0001 respectively. N.S.: not significant (Two-way ANOVA). (C) Heatmap of differentially expressed genes (DEGs) of RNA-seq analysis in the cells as in (A). Red and blue indicate relatively high and low expression (FC > 1.5 and FDR < 0.05), respectively (details in Supplementary Table 1). (D) KEGG pathway analysis of GSEA in the indicated cells relative to sgGFP control. NES: normalized enrichment score. Positive and negative scores indicate terms enriched for up- and down-regulated genes, respectively (FDR < 0.05). (E) Correlation between fold changes of promoter H4ac and gene expression. X-axis is log2 fold change (FC) of RNA expression and Y-axis is log2 fold change of H4ac levels in cells expressing KAT5-Y47A (left) and KAT5-ΔN (right) compared to cells expressing the FL KAT5. Red and blue dots represent genes with increased or decreased H4ac and expression (FC>1.5), respectively. Orange lines indicate fitted linear models. PCC: Pearson correlation coefficient.

In this study, we have delved into the functional importance and potential mechanisms of the Tudor-knot domain in regulating KAT5-dependent chromatin and transcriptional regulation in human cancer cells. Utilizing an unbiased CRISPR tiling array screen, we identified the KAT5 Tudor-knot domain as an essential component for maintaining global histone H4 acetylation and cell viability. Our biochemical investigations indicate that the Tudor-knot domain is unlikely to function as a histone reader. Nonetheless, the conserved aromatic residues forming a hydrophobic pocket, crucial for histone methylation recognition in other domain-containing proteins, are indeed functionally important. A mutation in one such residue, Y47, led to a global reduction in histone H4 acetylation, subsequent gene expression changes, and a reduction in cell viability. While previous studies reported that the Tudor-knot domain recruits KAT5 to chromatin through binding to histone H3K9me3 [27] or H3K4me1 [28], our biochemical investigation and ChIP-seq results do not support these conclusions. However, we cannot rule out potential involvement of the Tudor-knot domain in KAT5 chromatin recruitment under specific conditions, such as in response to DNA damage [27] or hormone stimulation[28]. Nevertheless, while the precise molecular mechanism remains unclear, our study suggests that the Tudor-knot domain regulates accessibility of the HAT domain to histones in nucleosomes, the natural substrate of the enzyme in cells.

KAT5 is highly conserved from yeast to humans [2]. Esa1, the KAT5 homolog in budding yeast, also features a Tudor-knot domain. A prior study reported that the Tudor-knot domain of Esa1 is necessary for nucleosome substrate acetylation by the Piccolo NuA4 complex, a small complex composed of three essential subunits [33]. In particular, a mutation of the aromatic residue equivalent to Y47 in human KAT5 strongly hinders growth in yeast [33]. Since the Tudor-knot domain of Esa1 is not required for NuA4 binding to nucleosomes, it has been proposed that it may function after the enzyme binds nucleosomes to disengage substrate histone tails from nucleosomal DNA [33, 34]. Our results suggest that the Tudor-knot domain in human KAT5 likely operates through a similar mechanism regulating nucleosomal histone acetylation. However, this regulation may be more complex and involve other NuA4 subunits since the Tudor-knot domain does not exhibit such regulation when KAT5 is present in isolation. Moreover, it has been reported that the Tudor-knot domain of Esa1 may possess RNA binding activity [35], adding additional complexity to the scenario. Further studies, such as those employing improved techniques to obtain dynamic structures of the Tudor-knot domain associated with nucleosomes, are needed in the future.

Methods

Reagents

Human full-length (FL)(aa1–513) KAT5 cDNA, Tudor-knot (aa1–80), and dN (aa81–513) were cloned in pENTR3C and subsequently cloned into pCDH-3Flag-HA, p3Flag, and pGEX-6p-1 destination vectors using Gateway techniques (Invitrogen). The KAT5-Y47A (aa1–513), sgRNA-resistant KAT5-FL, and sgRNA-resistant KAT5-Y47A were generated by PCR mutagenesis. The primers used for mutagenesis in this study are available upon requests. sgRNA sequences were cloned into the lentiCRISPR V2 vector. sgRNAs used in this study include GFP sgRNA: GGGCGAGGAGCTGTTCACCG; KAT5 sgRNA1: GGTTCCGCCGCAGCACGGGT; KAT5 sgRNA2: CACGCTCAGGATCTCGGCCA. Antibodies used in this study include histone H4 pan-acetyl antibody (Active Motif, 39925), histone H4 antibody (Abcam, ab7311), KAT5 (Abcam, ab23886), HA tag (clone C29F4) (Cell Signaling Technology, 3724S), Flag M2 (Sigma, F3165), GAS41 (C-10) (Santa Cruz, sc-393708), p400 (Bethyl, A300-541A), MRG15 (Abcam, ab183663), GST (Santa Cruz, sc-459).

Protein expression and purification

For GST-Tagged protein purification, Rosetta™2(DE3) pLysS competent cells (Novagen) were used. Cells were cultured at 37 °C to reach OD 0.6–0.8 in LB medium supplemented with 100 mM ZnCl2, then cells were transferred to a 16 °C incubator chilled down for 1h, and protein expression was induced with 0.4 mM IPTG overnight.

The next morning, cell pellets were collected, and resuspended in lysis buffer (50 mM Tris-HCl pH7.5, 300 mM NaCl, 1.5 mM MgCl2, 1 mM EDTA, 1% Triton100, 5 μM ZnCl2, 10% glycerol, 1 mM PMSF, and 1x cOmplete™ EDTA-free Protease Inhibitor Cocktail) with lysozyme to a final concentration of 0.5 mg/ml, and lysed on ice for 45min. Then break the cell by sonication using Branson Digital Sonifier at an output of 18% for 60 sec (1s/2s on/off). The lysate was centrifuged at 18,000 g for 15 min, and the supernatants were incubated with Glutathione Sepharose® 4B beads (Sigma) at 4° for 2 h. Then beads were washed with 10X beads volume lysis buffer three times. The GST-tagged protein that binds to beads was eluted with elution buffer (50 mM Tris-HCl pH7.5, 100 mM NaCl, 15% glycerol, and 15 mg/ml GSH).

For 3Flag-Tagged protein purification, the protein was stably expressed in HEK293T by lentivirus transduction. Cell pellets were collected and lysed in cell lysis buffer (50 mM Tris-HCl pH7.5, 300 mM NaCl, 1.5 mM MgCl2, 1 mM EDTA, 1% Triton100, 10% glycerol, 1 mM PMSF, and protease inhibitors) for 15min. Then, homogenate the samples by using Branson Digital Sonifier with a setting of 10% power, 0.5s/0.5s on/off for total 60s. The lysate was centrifuged at 4500 rpm for 10 min, and the supernatants were incubated with anti-Flag M2 beads (Sigma) at 4° overnight. The next day, the beads were washed 3 times with cell lysis buffer. The Flag-tagged protein binds to beads was eluted with elution buffer (50 mM Tris-HCl pH7.5, 100 mM NaCl, 15% glycerol, and 0.4mg/ml Flag peptide).

In vitro histone acetyltransferase (HAT) assay

Purified KAT5 protein was mixed with recombinant human mono-nucleosomes (100 nM, EpiCypher), and 100 mM Acetyl-CoA in 50 μL of HAT assay buffer (50 mM Tris pH8.0, 100 mM NaCl, 1 mM DTT, 10% glycerol and cOmplete™ EDTA-free Protease Inhibitor) on ice. The reaction was performed at 37 °C for the indicated time. Reactions were stopped by adding 15ul 5X SDS loading sample buffer and boiling at 95 °C for 5min.

Peptide microarray and peptide pull-down assay

Peptide microarray and peptide pull-down assays were performed as described previously in [36]. In brief, biotinylated histone peptides were printed in triplicate onto a streptavidin-coated slide (PolyAn) using a VersArray Compact Microarrayer (Bio-Rad). After a short blocking with biotin (Sigma), the slides were incubated with the GST-tagged KAT5 Tudor-knot domain in binding buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.1% NP-40, 1 mM PMSF, 20% FBS) overnight at 4 °C with gentle agitation. After being washed with the same buffer, the slides were probed with an anti-GST primary antibody and then a fluorescein-conjugated secondary antibody and visualized using a GenePix 4000 scanner (Molecular Devices).

For the peptide pull-down assays, 1 μg of biotinylated histone peptides with different modifications were incubated with 1 μg of GST-tagged protein in binding buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.05% NP-40, 1 mM PMSF) overnight. 5 μl streptavidin magnetic beads (Pierce) were added to the mixture, and the mixture was incubated for 1 h with rotation. The beads were then washed three times with binding buffer, Proteins that bind to beads were eluted by adding 60ul 2X SDS loading sample buffer and boiling at 95 °C for 5min.

Cell culture and virus transduction

MV4–11, OCI-AML2, NCI-H23, and H1299 cell lines were purchased from ATCC and cultured in RPMI1640 (Corning). HEK293T cell lines were purchased from ATCC and cultured in DMEM (Corning). All media were supplemented with 10% fetal bovine serum (Sigma), 1 mM sodium pyruvate (Corning), 1% non-essential amino acids (HyClone), and 1% penicillin–streptomycin (Corning). For the lentivirus package, HEK293T cells were used. lentiCRISPR V2 plasmids or pCDH-3Flag-HA plasmids were co-transfected together with psPAX2 and pMD2.G plasmids at a 2:2:1 ratio using the X-TremeGENE HP DNA transfection reagent (Sigma). 48h later, the medium containing the lentivirus was collected and filtered through a 0.45 μm Millex-HV Syringe Filter Unit (Millipore). For cell transduction, 2 million cells were seeded in a 60mm plate 24 hours before transduction. Cells were transduced in 1 ml lentivirus, 3 ml medium, and 4μg/ml polybrene. After 24 h culture at 37 °C, the medium containing the virus was removed and replaced with 4ml fresh medium. Cells were selected with 2 μg/ml puromycin or 10 μg/ml Blasticidin (Thermo Fisher Scientific) for at least 3 days.

Cell growth assay

500 cells were seeded in 96-well plates and cultured for 6–7 days. Cell growth was measured using the CellTiter-Glo luminescent cell viability assay kit (Promega) according to the manufacturer’s instructions.

Protein immunoprecipitation assay

The 3Flag-tagged protein was expressed in HEK293T by transient transfection. Cell pellets were collected and lysed in cell lysis buffer (50 mM Tris-HCl pH7.5, 300 mM NaCl, 1.5 mM MgCl2, 1 mM EDTA, 1% Triton100, 10% glycerol, 1 mM PMSF, and protease inhibitors) for 15 minutes. Then, homogenate the samples by using Branson Digital Sonifier with a setting of 10% power, 0.5s/0.5s on/off for total 20s. The lysate was centrifuged at 4500 rpm for 10 min, and the supernatants were incubated with anti-Flag M2 beads (Sigma) at 4° overnight. The next day, the beads were washed 3 times with cell lysis buffer. Proteins that bind to beads were eluted by adding 100ul 2X SDS loading sample buffer and boiling at 95 °C for 5min.

Western blot analysis

For whole cell lysate, cell pellets were washed with PBS once, resuspended in pre-chilled cell lysis buffer (50 mM Tris-HCl pH7.4, 250 mM NaCl, 0.5% Triton X100, 10% glycerol, 1 mM DTT, PMSF, PI (Roche)) and incubated on ice for 15min. Then, homogenate the samples by using Branson Digital Sonifier with a setting of 10% power, 0.5s/0.5s on/off for total 10s. Protein concentrations were measured using Bradford assay and adjusted to 2ug/ul. 1 volume of whole cell lysate and 1 volume of 2X SDS loading sample buffer were mixed and boiled at 95 °C for 10 min.

All primary antibodies incubation was performed at 4 degrees overnight. All second antibodies incubation was performed at 25 degrees for 1–2h.

ChIP and ChIP-seq analysis

ChIP was performed as described previously in [37]. Briefly, 20 million cells were cross-linked by 1% formaldehyde in PBS for 10 min and stopped by adding 0.125 M glycine for 5 min at room temperature. Cell pellets were washed 3 times with ice-cold PBS, then lysed the cell using cell lysis buffer (5 mM PIPES pH 8.0, 85 mM KCl, 1% NP-40, and protease inhibitors) for 20 min on ice. The nuclei were pelleted by centrifuging, 1300 g for 5min. Then resuspended on nuclei lysis buffer (50 mM Tris, pH 8.0, 10 mM EDTA, and 1% SDS) and sonicated for 600 s using Covaris E220 Evo. 40 μg sonicated chromatin samples were incubated with antibodies at 4 °C overnight. Antibodies used for ChIP-seq were HA antibody (Cell Signaling Technology, 3724S, lot#8): 10 μl per ChIP; and H4 pan-acetyl antibody (Active Motif, 39925, lot# 12517004): 4 μg per ChIP. The next day, Dynabeads™ Protein G (Thermo Fisher Scientific) was added and incubated for 1 h. The beads were washed 2 times with low salt buffer, 2 times with high salt buffer, once with LiCl wash buffer, and once with TE buffer. Bound DNA was eluted using fresh 50 mM NaHCO3 and 1% SDS at Eppendorf Thermomixer R, gently shaken at 1200 rpm at 65 °C for 30min. Reverse crosslinks were performed at 67 °C for 18h and DNA was purified using PCR purification kit (Qiagen).

All ChIP-seq experiments were performed with three biological replicates. ChIP-seq libraries were constructed using the KAPA Hyper Prep Kit (Roche) and sequenced for 40M reads each (50bp, paired-end) using Illumina NovaSeq 6000 at the Van Andel Institute Genomics Core. Fastq reads were mapped to the hg38 human genome by HISAT2 (v2.1.0) with --no-spliced-alignment -k 1 -X 1000 [38]. All ChIP-seq signals were normalized to WT rep1 using quantitative siQ-ChIP [39]. Heatmaps and bigwig files were generated by Deeptools (v3.5.2) [40]. Pearson correlation coefficient (PCC) and p-values between fold change of promoter or genebody H4-PanAc and gene expression, and fitted linear models were calculated by R (v3.5.1).

RNA isolation and RNA-seq analysis

All RNA-seq experiments were performed with three replicates each group. Total RNA was extracted using RNeasy Plus Mini Kit (Qiagen). RNA-seq libraries were prepared using KAPA RNA HyperPrep Kit with RiboErase (HMR) (Roche) following the manufacturer’s instructions and sequenced (50bp, paired-end) using Illumina NovaSeq 6000 at the Van Andel Institute Genomics Core. Fastq reads were mapped to the hg38 human genome by HISAT2 (v2.1.0) [38], fold change values were calculated by HTSeq (v0.11.3)[41], and edgeR with Trimmed Mean of M-values (TMM) and Exact test model (v3.16.5)[42]. Differentially expressed genes (DEGs) were filtered by FDR < 0.05 and FC > 1.5. Heatmaps of RNA-seq data were visualized by Java TreeView (v1.2.0). Gene Set Enrichment Analysis (GSEA) [43] was employed for KEGG pathway analysis.

Supplementary Material

1
2

Research highlights:

  • The KAT5 Tudor-knot motif is an essential domain for cell survival

  • The Tudor-knot domain is important for KAT5-dependent transcriptional regulation

  • The Tudor-knot domain regulates KAT5 HAT activity on nucleosomes

  • The Tudor-knot domain may regulate accessibility of histone tails in nucleosomes

Acknowledgments

We thank J. Cote for the scientific discussion and sharing reagents. We thank Marie Adams and the Genomics Core at Van Andel Institute for NGS sequencing. This work was supported in part by grants from NIH/NCI (CA255506 and CA260666) to H.W., NIH/NCI (CA204020 and CA268440) to X.S., and NIH (GM137927) and CPRIT (RR160097) to H.X. H.X. is a CPRIT scholar of cancer research.

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.

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.

Declaration of Competing Interest

The authors declare no competing financial interests.

Data availability

The authors declare that the data supporting the findings of this study are available within the paper and its Supplementary Information, and available from the corresponding author upon request. All RNA-seq and ChIP-seq generated in this study have been deposited in the NCBI Gene Expression Omnibus (GEO) database and are accessible through the GEO SuperSeries accession number GSE245115 (RNA-seq, reviewer can access this by etwjsqciptwhhor) and GSE245116 (ChIP-seq, reviewer can access this by qtinuwkivbwtlcr).

References

  • [1].Kamine J, Elangovan B, Subramanian T, Coleman D, Chinnadurai G. Identification of a cellular protein that specifically interacts with the essential cysteine region of the HIV-1 Tat transactivator. Virology. 1996;216:357–66. [DOI] [PubMed] [Google Scholar]
  • [2].Steunou A-L, Rossetto D, Côté J. Regulating Chromatin by Histone Acetylation. Fundamentals of Chromatin 2014. p. 147–212. [Google Scholar]
  • [3].Cai Y, Jin J, Tomomori-Sato C, Sato S, Sorokina I, Parmely TJ, et al. Identification of new subunits of the multiprotein mammalian TRRAP/TIP60-containing histone acetyltransferase complex. J Biol Chem. 2003;278:42733–6. [DOI] [PubMed] [Google Scholar]
  • [4].Doyon Y, Selleck W, Lane WS, Tan S, Cote J. Structural and functional conservation of the NuA4 histone acetyltransferase complex from yeast to humans. Mol Cell Biol. 2004;24:1884–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [5].Kobor MS, Venkatasubrahmanyam S, Meneghini MD, Gin JW, Jennings JL, Link AJ, et al. A protein complex containing the conserved Swi2/Snf2-related ATPase Swr1p deposits histone variant H2A.Z into euchromatin. PLoS biology. 2004;2:E131. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [6].Kimura A, Horikoshi M. Tip60 acetylates six lysines of a specific class in core histones in vitro. Genes Cells. 1998;3:789–800. [DOI] [PubMed] [Google Scholar]
  • [7].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]
  • [8].Van Den Broeck A, Nissou D, Brambilla E, Eymin B, Gazzeri S. Activation of a Tip60/E2F1/ERCC1 network in human lung adenocarcinoma cells exposed to cisplatin. Carcinogenesis. 2012;33:320–5. [DOI] [PubMed] [Google Scholar]
  • [9].Lin SY, Li TY, Liu Q, Zhang C, Li X, Chen Y, et al. GSK3-TIP60-ULK1 signaling pathway links growth factor deprivation to autophagy. Science. 2012;336:477–81. [DOI] [PubMed] [Google Scholar]
  • [10].Jaiswal B, Gupta A. Modulation of Nuclear Receptor Function by Chromatin Modifying Factor TIP60. Endocrinology. 2018;159:2199–215. [DOI] [PubMed] [Google Scholar]
  • [11].Keogh MC, Mennella TA, Sawa C, Berthelet S, Krogan NJ, Wolek A, et al. The Saccharomyces cerevisiae histone H2A variant Htz1 is acetylated by NuA4. Genes Dev. 2006;20:660–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [12].Mitchell L, Lambert JP, Gerdes M, Al-Madhoun AS, Skerjanc IS, Figeys D, et al. Functional dissection of the NuA4 histone acetyltransferase reveals its role as a genetic hub and that Eaf1 is essential for complex integrity. Mol Cell Biol. 2008;28:2244–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [13].Brauns-Schubert P, Schubert F, Wissler M, Weiss M, Schlicher L, Bessler S, et al. CDK9-mediated phosphorylation controls the interaction of TIP60 with the transcriptional machinery. EMBO Rep. 2018;19:244–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [14].Acharya D, Hainer SJ, Yoon Y, Wang F, Bach I, Rivera-Perez JA, et al. KAT-Independent Gene Regulation by Tip60 Promotes ESC Self-Renewal but Not Pluripotency. Cell Rep. 2017;19:671–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [15].Yamagata K, Shino M, Aikawa Y, Fujita S, Kitabayashi I. Tip60 activates Hoxa9 and Meis1 expression through acetylation of H2A.Z, promoting MLL-AF10 and MLL-ENL acute myeloid leukemia. Leukemia. 2021;35:2840–53. [DOI] [PubMed] [Google Scholar]
  • [16].Janas JA, Zhang L, Luu JH, Demeter J, Meng L, Marro SG, et al. Tip60-mediated H2A.Z acetylation promotes neuronal fate specification and bivalent gene activation. Mol Cell. 2022;82:4627–46 e14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [17].Ikura T, Ogryzko VV, Grigoriev M, Groisman R, Wang J, Horikoshi M, et al. Involvement of the TIP60 histone acetylase complex in DNA repair and apoptosis. Cell. 2000;102:463–73. [DOI] [PubMed] [Google Scholar]
  • [18].Legube G, Linares LK, Tyteca S, Caron C, Scheffner M, Chevillard-Briet M, et al. Role of the histone acetyl transferase Tip60 in the p53 pathway. J Biol Chem. 2004;279:44825–33. [DOI] [PubMed] [Google Scholar]
  • [19].Liu N, Wang J, Wang J, Wang R, Liu Z, Yu Y, et al. ING5 is a Tip60 cofactor that acetylates p53 in response to DNA damage. Cancer Res. 2013;73:3749–60. [DOI] [PubMed] [Google Scholar]
  • [20].Pollina EA, Gilliam DT, Landau AT, Lin C, Pajarillo N, Davis CP, et al. A NPAS4-NuA4 complex couples synaptic activity to DNA repair. Nature. 2023;614:732–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [21].Allard S, Utley RT, Savard J, Clarke A, Grant P, Brandl CJ, et al. NuA4, an essential transcription adaptor/histone H4 acetyltransferase complex containing Esa1p and the ATM-related cofactor Tra1p. EMBO J. 1999;18:5108–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [22].Clarke AS, Lowell JE, Jacobson SJ, Pillus L. Esa1p is an essential histone acetyltransferase required for cell cycle progression. Mol Cell Biol. 1999;19:2515–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [23].Hu Y, Fisher JB, Koprowski S, McAllister D, Kim MS, Lough J. Homozygous disruption of the Tip60 gene causes early embryonic lethality. Dev Dyn. 2009;238:2912–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Qu K, Chen K, Wang H, Li X, Chen Z. Structure of the NuA4 acetyltransferase complex bound to the nucleosome. Nature. 2022;610:569–74. [DOI] [PubMed] [Google Scholar]
  • [25].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]
  • [26].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]
  • [27].Sun Y, Jiang X, Xu Y, Ayrapetov MK, Moreau LA, Whetstine JR, et al. Histone H3 methylation links DNA damage detection to activation of the tumour suppressor Tip60. Nat Cell Biol. 2009;11:1376–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [28].Jeong KW, Kim K, Situ AJ, Ulmer TS, An W, Stallcup MR. Recognition of enhancer element-specific histone methylation by TIP60 in transcriptional activation. Nat Struct Mol Biol. 2011;18:1358–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [29].Zhang Y, Lei M, Yang X, Feng Y, Yang Y, Loppnau P, et al. Structural and histone binding studies of the chromo barrel domain of TIP60. FEBS Lett. 2018;592:1221–32. [DOI] [PubMed] [Google Scholar]
  • [30].He W, Zhang L, Villarreal OD, Fu R, Bedford E, Dou J, et al. De novo identification of essential protein domains from CRISPR-Cas9 tiling-sgRNA knockout screens. Nat Commun. 2019;10:4541. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [31].Shi J, Wang E, Milazzo JP, Wang Z, Kinney JB, Vakoc CR. Discovery of cancer drug targets by CRISPR-Cas9 screening of protein domains. Nat Biotechnol. 2015;33:661–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [32].Munoz DM, Cassiani PJ, Li L, Billy E, Korn JM, Jones MD, et al. CRISPR Screens Provide a Comprehensive Assessment of Cancer Vulnerabilities but Generate False-Positive Hits for Highly Amplified Genomic Regions. Cancer Discov. 2016;6:900–13. [DOI] [PubMed] [Google Scholar]
  • [33].Selleck W, Fortin I, Sermwittayawong D, Cote J, Tan S. The Saccharomyces cerevisiae Piccolo NuA4 histone acetyltransferase complex requires the Enhancer of Polycomb A domain and chromodomain to acetylate nucleosomes. Mol Cell Biol. 2005;25:5535–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [34].Huang J, Tan S. Piccolo NuA4-catalyzed acetylation of nucleosomal histones: critical roles of an Esa1 Tudor/chromo barrel loop and an Epl1 enhancer of polycomb A (EPcA) basic region. Mol Cell Biol. 2013;33:159–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [35].Shimojo H, Sano N, Moriwaki Y, Okuda M, Horikoshi M, Nishimura Y. Novel structural and functional mode of a knot essential for RNA binding activity of the Esa1 presumed chromodomain. J Mol Biol. 2008;378:987–1001. [DOI] [PubMed] [Google Scholar]
  • [36].Wen H, Li Y, Xi Y, Jiang S, Stratton S, Peng D, et al. ZMYND11 links histone H3.3K36me3 to transcription elongation and tumour suppression. Nature. 2014;508:263–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [37].Wan L, Chong S, Xuan F, Liang A, Cui X, Gates L, et al. Impaired cell fate through gain-of-function mutations in a chromatin reader. Nature. 2020;577:121–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [38].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]
  • [39].Dickson BM, Tiedemann RL, Chomiak AA, Cornett EM, Vaughan RM, Rothbart SB. A physical basis for quantitative ChIP-sequencing. J Biol Chem. 2020;295:15826–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [40].Ramirez F, Ryan DP, Gruning B, Bhardwaj V, Kilpert F, Richter AS, et al. deepTools2: a next generation web server for deep-sequencing data analysis. Nucleic Acids Res. 2016;44:W160–5. [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].Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A. 2005;102:15545–50. [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
NIHMS1955323-supplement-1.xlsx (759.4KB, xlsx)
2
NIHMS1955323-supplement-2.pdf (868KB, pdf)

Data Availability Statement

The authors declare that the data supporting the findings of this study are available within the paper and its Supplementary Information, and available from the corresponding author upon request. All RNA-seq and ChIP-seq generated in this study have been deposited in the NCBI Gene Expression Omnibus (GEO) database and are accessible through the GEO SuperSeries accession number GSE245115 (RNA-seq, reviewer can access this by etwjsqciptwhhor) and GSE245116 (ChIP-seq, reviewer can access this by qtinuwkivbwtlcr).

RESOURCES