Introduction

In eukaryotic cells, genomic DNA wraps around a histone octamer, consisting of two copies of each of the four core histones (H2A, H2B, H3, and H4), to form nucleosomes, the fundamental units of chromatin. Post-translational modifications (PTMs) of histones, such as phosphorylation, ubiquitination, methylation, and acetylation, are integral components of the epigenome, playing crucial roles in determining chromatin structure and function. The levels of histone PTMs are tightly regulated by histone-modifying enzymes (writers and erasers). For instance, histone acetyltransferases (HATs) catalyze the acetylation of histone lysine residues, while histone deacetylases (HDACs) remove these acetyl groups. Typically, euchromatin (open chromatin) is marked by histone acetylation and associated with active transcription and gene expression, whereas heterochromatin (closed chromatin) is marked by histone methylation and associated with transcriptional repression.

Recent advances in mass spectrometry-based proteomics have revealed a greater number and variety of histone PTMs than previously anticipated1. For example, at least 45 amino acids on histones have been identified as acetylation sites1, many of whose functions remain unknown. To elucidate the cellular functions of each histone PTM, methods for manipulating histone PTMs in living cells are essential. Small molecule inhibitors of histone-modifying enzymes, such as HDAC inhibitors, can modulate the levels of histone PTMs. However, since histone-modifying enzymes target multiple sites on histones and non-histone proteins, this approach lacks selectivity and may not be suitable for determining the function of a specific PTM.

An emerging alternative strategy involves directly promoting histone PTMs in living cells through artificial means, bypassing endogenous histone-modifying enzymes. These methods include genetic code expansion2,3, split intein ligation4,5, dCas9-writer fusion proteins6,7, and chemical catalysis8,9,10. Although genetic code expansion and split intein ligation can generate modified histones, they cannot alter endogenous histones. While dCas9-writer fusion proteins can modify the epigenome of target gene loci, installing a histone PTM at a specific site is challenging because writers often modify multiple sites on histones. Furthermore, these three methods require genetic engineering and thus have limited applications for therapeutic purposes11. In contrast, the chemical catalysis approach, where chemical catalysts directly promote histone PTMs without relying on histone-modifying enzymes, does not require genetic engineering and may be particularly suitable for studying the consequences of histone PTMs under physiological conditions. Moreover, given that an abnormal epigenome and dysregulated gene expression are hallmarks of cancer12,13,14,15, the chemical catalysis approach holds therapeutic potential.

Our group previously reported a histone acetylation catalyst, PEG-LANA-DSSMe, comprising a prodrug disulfide form of a thiol-tethered 4-dimethylaminopyridine (DSH) catalyst conjugated with a histone-binding LANA (latency-associated nuclear antigen) peptide and a poly(ethylene glycol) (PEG) moiety8. The PEG-LANA-DSH catalyst binds to the acidic patch of nucleosomes via LANA and efficiently promotes histone- and regio-selective acetylation of the proximal H2BK120 residue in living cells using thioester acetyl donors. However, efficient histone acetylation required a high concentration of the acetyl donor (30 mM) due to the catalyst’s moderate activity. In addition, since the PEG-LANA-DSSMe catalyst was not cell-permeable, this approach relied on a bead-loading method16, which transiently creates pores in the cell membrane to introduce the catalysts into cells. This physical cellular damage might affect the phenotypes of catalyst-promoted histone acetylation. Therefore, the PEG-LANA-DSSMe catalyst is not an optimal tool for studying the functions of histone PTMs in living cells.

In this study, we developed cell-permeable chemical catalysts that selectively acetylate H2BK120 of endogenous histones in a short reaction time. Using this minimally perturbating method with high time resolution, we investigated the function of H2BK120 acetylation in leukemia cells.

Results

Development of a cell-permeable histone acylation catalyst targeting H2BK120

We sought to design chemical catalysts that could selectively acylate H2BK120 by crossing the cell membrane, migrating to the nucleus, binding to the H2BK120-proximal acidic patch on the nucleosome, and activating a stable acyl donor molecule (Fig. 1a). We envisioned a chemical catalyst comprised of four motifs that could achieve catalytic histone acylation in living cells (Fig. 1a): a cell-penetrating peptide (CPP) to permeate the cell membrane, a LANA peptide to bind the nucleosome acidic patch17, PEG to shield LANA from peptidases18,19, and a catalytically active site (CAS) to activate an acyl donor for lysine acylation.

Fig. 1: Schematics of catalytic histone acetylation in living cells and chemical structures of catalysts and acetyl donor used in this study.
figure 1

a The BAHA-LANA-PEG-CPP44 catalyst penetrates the cell membrane via a cell penetrating peptide (CPP). Poly(ethylene glycol) (PEG) shields the LANA peptide from peptidases. The LANA peptide binds the acidic patch of nucleosomes, and a catalytically active site (CAS) promotes the acetylation of a proximal lysine residue in histones using an acetyl donor. b Structures of BAHA-LANA-PEG-CPP44 1, BAHA-mutLANA-PEG-CPP44 2, BAHA-LANA-PEG 3, and acetyl donor 4.

We began our study with catalyst structure optimization at the CAS motif. There are two CAS candidates functional in living cells; DSH, which activates thioesters, such as acyl-CoAs, through dynamic thiol-thioester exchange and intramolecular acyl group transfer to form a reactive acyl pyridinium intermediate20, and boronate-assisted hydroxamic acid (BAHA), which is composed of a hydroxamic acid moiety and a diol moiety21 (for the chemical structures and reaction mechanisms of CASs, see Supplementary Fig. 1). BAHA recruits boronic acid-containing acyl donors to the diol moiety and generates reactive acyl hydroxamate intermediates through the subsequent intramolecular acyl group transfer. We compared these CASs and found that BAHA showed higher histone acetylation activity than DSH, especially under BSO (an inhibitor of glutathione synthesis)-free conditions (Supplementary Fig. 2a–c), suggesting that BAHA is more resistant to intracellular glutathione than DSH. Next, we optimized LANA and PEG motifs. Previous studies suggested that conjugation of a PEG motif to the C-terminus of LANA decreased the affinity of LANA with nucleosomes8. Therefore, we screened the PEG-conjugation position by substituting each amino acid of the LANA5-15 peptide to an acetylated lysine, used as a model for PEG-conjugated lysine, and found that the 13th serine was the best position for PEG conjugation (Supplementary Fig. 3a–d). Finally, we optimized the CPP motif. We first conjugated oligo-arginine peptides22,23,24, such as 8r and 16r (an 8-mer and a 16-mer of D-arginine, respectively), to FITC-labeled LANA-PEG compounds and examined their membrane permeability for HeLa S3 cells (cervical cancer cell line). When FITC-LANA-PEG-8r (S22, Supplementary Fig. 4a, 50 μM) was added to cell culture media, fluorescent signals mainly localized in endosomes, suggesting that S22 failed to escape from endosomes (Supplementary Fig. 4b). For FITC-LANA-PEG-16r (S27, Supplementary Fig. 4c), 16r was conjugated with FITC-LANA-PEG via a disulfide bond, with the expectation that the disulfide bond would be cleaved by glutathione to release FITC-LANA-PEG in the cells. While S27 localized in nuclei at 5 μM, no nuclear localization was observed at 2 μM (Supplementary Fig. 4d). We also confirmed chromatin localization of S27 (5 μM) in HEK293T cells (embryonic kidney cell line) (Supplementary Fig. 4e). Taken together, S27 permeated through cell membranes and localized at chromatin in both HeLa S3 and HEK293T cells. We then conjugated cell type-selective CPPs, such as leukemia cell-selective CPP4425. FITC-LANA-PEG-CPP44 (S34, Supplementary Fig. 5a), but not FITC-LANA-PEG (S32, Supplementary Fig. 5a), efficiently entered THP-1 cells (MLL-rearrangement leukemia cell line) and localized at chromatin, while S34 did not enter HeLa S3 cells or PBMC (peripheral blood mononuclear cells) (Fig. 2a, b). We confirmed that the introduction of PEG prevents hydrolysis of the molecules in THP-1 cells (Supplementary Fig. 5b). S34 failed to enter THP-1 cells when a dynamin inhibitor was added or at 4 °C (Supplementary Fig. 5c, d), suggesting that the cell-penetrating mechanism of S34 into THP-1 cells was through a dynamin- and ATP-dependent endocytic pathway, as suggested by the previous report25. Based on these optimizations, we synthesized BAHA-LANA-PEG-CPP44 (1: Fig. 1b) as a leukemia cell-selective catalyst, and BAHA-LANA-PEG-16r (S30: Supplementary Fig. 4c), which may have no cell-type selectivity, as a catalyst for non-leukemia cells.

Fig. 2: BAHA-LANA-PEG-CPP44 catalyst promoted H2BK120 acetylation in leukemia cells.
figure 2

a, b Cells were treated with S32 or S34 (50 µM) for 60 min. Representative microscopic images of FITC-LANAS13K-PEG550 S32 and FITC-LANAS13K-PEG550-CPP44 S34 in the indicated cell line in two independent experiments. DNA was stained with Hoechst 33342 to visualize chromatin distribution. In (b), DAPI-negative cells were selected for observation to examine compound internalization into living cells. Scale bar, 10 µm. c, d Cytotoxicity of catalyst 1 or acetyl donor 4. THP-1 cells were treated with 1 or 4 for 6 h. The bar graph shows the mean of two independent experiments. e Immunoblot showing H2BK120 acetylation (H2BK120ac) levels of recombinant nucleosomes treated with catalyst 1, 2, or 3 (2 µM) and acetyl donor 4 (0.1 mM) for 2 h. A representative blot in two independent experiments is shown. f Immunoblot showing H2BK120ac and H2BK120 ubiquitination (H2Bub) levels in THP-1 cells treated with catalyst 1, 2, or 3 (10 µM) and acetyl donor 4 (1 mM) for 4 h. Representative blots in two independent experiments are shown. g LC-MS/MS analysis for lysine acetylation levels of histone H2A, H2B, H3, or H4 in THP-1 cells treated with catalyst 1 or 2 (10 µM) and acetyl donor 4 (1 mM) for 4 h. The acetylation yield was calculated as the ratio of acetylated lysine to (acetylated lysine + unmodified lysine) detected by LC-MS/MS, and the difference of the yield with 1 from the yield with 2 is shown. Each number on the horizontal axis corresponds to the position of lysine residue in the histone protein. The bar graph shows the mean of two independent experiments. h Immunoblot showing H2BK120ac levels in the indicated cell lines treated with catalyst 1 or 2 (10 µM) and acetyl donor 4 (1 mM) for 2 h. The loading amounts of histones were visualized with Oriole staining in (eh). Representative blot in two independent experiments is shown.

In-cell H2BK120 acetylation by cell-permeable histone acylation catalysts

We examined the cytotoxicity of BAHA-LANA-PEG-16r catalyst (S30) in HeLa S3 cells. While 2 μM S30 did not show apparent cytotoxicity, greater than 5 μM S30 was cytotoxic (Supplementary Fig. 4f). We then evaluated the effects of the CPP motif by comparing the in-cell histone acetylation activity of S30 and CPP-non-conjugated BAHA-LANA-PEG catalyst (3) using HeLa S3 and HEK293T cells. After cells were treated with S30 or 3 (5 μM: without bead-loading) and acetyl donor 4 (0.4 mM), the cells were stained with 4′,6-diamidine-2-phenylindole dihydrochloride (DAPI) to exclude potential artifacts caused by contaminated dead cells. We sorted DAPI-negative cells (i.e., living cells) and examined the acetylation of H2BK120 by both immunoblot analysis using an H2BK120ac-specific antibody and LC-MS/MS analysis. Remarkably, S30, but not 3, promoted H2BK120 acetylation in both HeLa S3 and HEK293T cells (Supplementary Figs. 4g, h). However, the acetylation stoichiometry was low (~ 5% yield) even when using 5 µM catalyst S30, which exhibited high cytotoxicity (~ 50% viability) (Supplementary Fig. 4i). This limited the utility of S30 for analyzing the phenotypic effects of catalytic histone acetylation.

Next, we examined the cytotoxicity of BAHA-LANA-PEG-CPP44 catalyst (1) and found that less than 20 μM catalyst 1 did not show any cytotoxicity in THP-1 cells (Fig. 2c). We also confirmed that less than 2 mM acetyl donor 4 was not toxic in THP-1 cells (Fig. 2d). As a negative control catalyst, we synthesized BAHA-mutLANA-PEG-CPP44 (2), lacking nucleosome-binding ability (Fig. 1b and Supplementary Fig. 5e). We confirmed that 1 and 3, but not 2, had histone acetylation ability for recombinant nucleosomes in test tubes (Fig. 2e). The regioselectivity of catalysts 1 and BAHA-LANA S9 was consistent (Supplementary Fig. 3e), indicating that only BAHA and LANA moieties are essential for H2BK120-selective acetylation. We then assessed in-cell histone acetylation activity. We added 10 μM catalyst 1 and 1 mM acetyl donor 4 to the culture media of THP-1 cells and examined the acetylation of H2BK120 by immunoblot. H2BK120 acetylation was significantly promoted by catalyst 1 and acetyl donor 4 combined, but not by catalyst 1 or acetyl donor 4 alone (Fig. 2f and Supplementary Fig. 6a). H2BK120 ubiquitination was reduced by catalytic H2BK120 acetylation promoted by catalyst 1 and acetyl donor 4 (Fig. 2f), as was observed previously8. No significant changes were observed in the representative methylation marks of histones immediately or 24 h following catalytic histone acetylation for 4 h (Supplementary Fig. 6b, c). Catalyst 2 (lacking nucleosome-binding) or 3 (lacking CPP) with acetyl donor 4 did not promote H2BK120 acetylation in THP-1 cells, confirming that both LANA and CPP44 were essential for in-cell histone acetylation activity of the catalyst (Fig. 2f).

Characterization of the catalytic histone acetylation in leukemia cells

Since the catalytic histone acetylation by catalyst 1 inhibited Mg2+-promoted nucleosome self-association and the formation of phase-separated 12-mer nucleosome arrays in vitro (Supplementary Fig. 7a–c), we expected that H2BK120 acetylation may attenuate inter-nucleosomal interaction and affect chromatin structures as well as gene transcription in cells. Therefore, we characterized the catalytic histone acetylation by catalyst 1 in THP-1 cells in detail.

Catalyst 1 with acetyl donor 4 promoted histone-selective acetylation, and other proteins showed little to no acetylation (Supplementary Fig. 8a). To evaluate lysine residue selectivity, we quantified the acetylation stoichiometry of histone’s lysine residues by LC-MS/MS analysis. The result showed that the main acetylation site was H2BK120 (~ 38% yield, Fig. 2g). H2BK116 and H2AK15 were slightly acetylated (~ 5% yield), and other lysine residues, including those in H3 or H4 tail, were rarely acetylated (Fig. 2g). Catalyst 1 promoted acetylation of H2BK120 when H2BK120ac writer enzymes p300 (KAT3B) and CBP (KAT3A)26 were knocked down with RNAi (Supplementary Fig. 8b). Furthermore, treatment of THP-1 cells with an HDAC inhibitor SAHA did not affect catalyst-dependent H2BK120 acetylation (Supplementary Fig. 8c). These results indicate that H2BK120ac was directly introduced by catalyst 1 and the acetylation reaction did not depend on cellular HAT or HDAC activity.

The BAHA catalyst promotes not only acetylation but also other acylations, including non-natural ones, simply by changing acyl donors21. Therefore, we prepared acyl donors with pentynoyl S35, azidobutyryl S36, methyl succinyl S37, and butyryl S38 groups (Supplementary Fig. 9a) and conducted in-cell acylation reactions with catalyst 1. As expected, the addition of catalyst 1 and the acyl donors to THP-1 cells promoted histone acylation in all cases (Supplementary Fig. 9b, c). We showed that catalytically incorporated azidobutyrylated histones can be labeled by alkyne-conjugated TAMRA via click chemistry, visualizing endogenous histones in THP-1 cells (Supplementary Fig. 9d).

To address the cell-type selectivity of the catalytic histone acetylation reaction, we treated various leukemia cell lines (THP-1, MOLM-13, MV-4-11, Kasumi-1, and K562) and non-leukemia cell lines (HeLa S3, HEK293T, and PBMC) with 10 μM catalyst 1 and 1 mM acetyl donor 4 and examined acetylation of H2BK120 by immunoblot analysis (Fig. 2h). Consequently, H2BK120 acetylation was promoted by catalyst 1 with acetyl donor 4, but not by negative control catalyst 2 with acetyl donor 4, in all the leukemia cell lines. In contrast, it was hardly promoted in non-leukemia cell lines (Fig. 2h). Since the previous report indicated that CPP44 is also permeable to liver cancer cells21, we examined the catalytic histone acetylation in HepG2 cells (liver cancer cells) and HepaMN cells (normal liver cells). The results showed that catalyst 1 with acetyl donor 4 acetylated H2BK120 in HepG2 cells, but not in HepaMN cells (Supplementary Fig. 10a, b).

Transcriptional changes and anti-leukemia effects induced by H2BK120 acetylation

To investigate whether H2BK120 acetylation affected the transcription profile, we performed transcriptome analysis. RNA sequencing (RNA-seq) of acetylated THP-1 cells treated with catalyst 1 and acetyl donor 4 for 4 h identified 634 upregulated (UP) and 466 downregulated (DOWN) genes compared to unacetylated THP-1 cells treated with negative control catalyst 2 and acetyl donor 4 for 4 h (Fig. 3a and Supplementary Fig. 11a). Since the H2BK120 acetylation reduced ubiquitination of H2B (Fig. 2f), which is required for the expression of MLL-fusion target genes in MLL-rearranged leukemia cells27, we first assessed transcriptional changes of 330 MLL-fusion target genes. However, transcription of MLL-fusion target genes was largely unchanged by the catalytic histone acetylation (Fig. 3b), probably due to insufficient reduction of H2B ubiquitination. We then conducted Gene Ontology (GO) analysis to identify that UP genes were related to the regulation of apoptosis, cell cycle, or transcription (Supplementary Fig. 11b), although DOWN genes did not have such significant features (Supplementary Fig. 11c). We further confirmed this correlation by RT-qPCR analysis of selected genes related to these GO-terms, such as EGR1, JUN, or p21 (Supplementary Fig. 12a). Consistently, the catalytic histone acetylation induced cell cycle arrest at G1/S phase and apoptosis in THP-1 cells (Fig. 3c, d).

Fig. 3: Transcriptional changes and anti-leukemia effects induced by catalytic histone acetylation.
figure 3

Before analysis, cells were treated with catalyst 1 (acetylated) or 2 (control) (10 µM) and acetyl donor 4 (1 mM) for 4 h. a, b Correlation between catalytic histone acetylation and gene transcription. After acetylation for 4 h, THP-1 cells were analyzed by RNA-seq (n = 2). Red and blue dots represent significantly upregulated and downregulated genes, respectively (P < 0.05 in Student’s two-sided t test, log2FC > 1 or < 1). Gray dots represent other genes (NS). a Scatter plot of the FPKM + 1 value of genes. b Scatter plot of the FPKM + 1 value of MLL-fusion target genes (Human Gene Set: KUMAR_TARGETS_OF_MLL_AF9_FUSION). c, d Correlation between catalyst-promoted acetylation and cell cycles or apoptosis. After acetylation for 4 h, the medium was replaced with fresh medium. After 18 h, cells with different cell cycles (G0/G1, S, or G2/M) were classified by flow cytometry for DNA content (c). After 48 h, cells with early apoptosis (early apop.: high Annexin V and low 7-AAD), late apoptosis (late apop.: high Annexin V and high 7-AAD), or dead (low Annexin V and high 7-AAD) were classified by flow cytometry for Annexin V and 7-AAD staining (d). The bar graph shows the mean of three independent experiments. e, f Correlation between catalytic histone acetylation and viability of cell lines. After acetylation for 4 h, the medium was replaced with fresh medium. The numbers of trypan blue-negative cells at 0–6 days are shown, normalized to the cells treated with DMSO in (e), or the cells with 2 + 4 in (f). Data are shown as mean ± standard deviation of four independent experiments (*P < 0.05, one-way ANOVA with Tukey’s post hoc test to DMSO-treated cells in (e), or 4-treated cells in (f). NS: not significant). g, h Correlation between catalytic histone acetylation and the tumorigenic potential in mice. After acetylation for 4 h, Akaluc-expressing 3 × 106 MOLM-13 cells were intravenously injected into NOG mice (n = 5 for control, 6 for acetylated). g Images of bioluminescence signals in recipient mice at 7 and 14 days post-transplantation. h A Kaplan-Meier survival. Statistical significance was calculated with log-rank tests.

Next, we investigated whether H2BK120 acetylation affects the viability of leukemia cells. When histones in THP-1 cells were acetylated by catalyst 1 and acetyl donor 4, the viability of THP-1 cells decreased over several days (Fig. 3e). In contrast, treatment with negative control catalyst 2 and acetyl donor 4 did not reduce viability (Fig. 3e), indicating that the reduction in viability was a result of catalytic histone acetylation. The reduction in viability due to catalytic histone acetylation was observed not only in MLL-rearranged leukemia cells (THP-1, MOLM-13, MV-4-11) but also in leukemia cells without MLL translocations (Kasumi-1, K562) (Fig. 3f). In contrast, the viability of HeLa S3 cells was unaffected by treatment with catalyst 1 and acetyl donor 4 (Fig. 3f), confirming the leukemia cell-selectivity of the catalyst. We also tested the in vivo-tumorigenic potential of the catalytic histone acetylation in leukemia cells. The Akaluc-bearing MOLM-13 leukemia cells were acetylated by catalyst 1 and acetyl donor 4 and then intravenously injected into NOG mice. As a result, although cell viability was nearly identical to that of the negative control treated by catalyst 2 and acetyl donor 4 just before injection (Supplementary Fig. 13), mice treated with acetylated MOLM-13 cells showed significantly slower cancer progression and longer survival than the mice treated with unacetylated MOLM-13 cells (Fig. 3f, g). These results demonstrate that H2BK120 acetylation reduced the tumorigenic potential of leukemia cells both in vitro and in vivo.

Mechanistic insight into transcriptional changes by H2BK120 acetylation

To explore how H2BK120 acetylation affects gene transcription, more specifically how upregulated genes are differentiated from other genes, we studied the transcriptional changes immediately following the reaction. The time-course analysis of the reaction progress in THP-1 cells showed that H2BK120 acetylation proceeded within 30 min of the treatment with catalyst 1 and acetyl donor 4, and reached a plateau at ~ 60 min (Fig. 4a). This kinetic profile was similar to that of histone-tail acetylation after treatment with an HDAC inhibitor, SAHA (Fig. 4a). Thus, we analyzed transcriptional changes at 45 min after the treatment with catalyst 1 or negative control catalyst 2 and acetyl donor 4. RNA-seq of acetylated THP-1 cells with 1 and acetyl donor 4 identified 124 upregulated (UP) and 105 downregulated (DOWN) genes compared to unacetylated THP-1 cells treated with negative control catalyst 2 and acetyl donor 4 (Fig. 4b and Supplementary Fig. 11d). We examined chromatin binding of RNA polymerase II (Pol II) after the catalytic histone acetylation by chromatin immunoprecipitation sequencing (ChIP-seq) and found that chromatin binding of RNA Pol II was significantly increased in UP genes (Fig. 4c, e and Supplementary Fig. 12b). This result suggests that the catalytic histone acetylation promoted transcription of subset of genes by RNA Pol II.

Fig. 4: Mechanistic insights into transcriptional changes by catalytic histone acetylation.
figure 4

Before analysis, THP-1 cells were treated with catalyst 1 (acetylated) or 2 (control) (10 µM) and acetyl donor 4 (1 mM) for 0–120 min in (a), or 45 min in (bj). Red and blue dots or lines represent significantly upregulated (UP) and downregulated (DOWN) genes, respectively (P < 0.05 in Student’s two-sided t test, log2FC > 1 or < 1). Gray dots or lines represent other genes (NS). a Immunoblot showing histone modification levels after catalytic histone acetylation (1 + 4) or SAHA (2 µM) treatment. b Scatter plot of the FPKM + 1 value of genes in RNA-seq analysis (n = 2). c, d Metagene profiles in ChIP-seq analyses of RNA polymerase II (Pol II) in (c), or H2BK120ac in (d). Gray lines represent other genes (NS). e Genome browser track example of the ChIP-seq data over the GADD45B and CITED2 loci visualized by the Integrated Genomics Viewer (IGV). f Top-5 enriched clusters of ENCODE TF ChIP-seq 2015 library from enrichment analysis of 124 upregulated genes (log2FC > 1) The P-value was calculated by Fisher’s two-sided exact test. g Metagene profiles in ChIP-seq analyses of NELF. Venn diagram indicating overlapping upregulated (UP) genes and high NELFE genes, which had top-200 normalized tag count of NELFE ChIP-seq. The P-value was calculated by Fisher’s two-sided exact test. h Comparison of the change of normalized tag count of NELFE ChIP-seq after acetylation. Data are shown as mean ± SEM of the tag count for 13,710 genes. The P-value was calculated by the Student’s two-sided t test. i Immunoblot showing NELFE and H2BK120ac after catalytic histone acetylation. The loading amounts of proteins were visualized with Oriole staining. Representative blots in two independent experiments are shown. j Comparison of the fold change of mRNA levels after catalytic histone acetylation. Data are shown as mean ± SEM (n = 200 for high NELFE genes, 14,500 for low NELFE genes). The P-value was calculated by the Student’s two-sided t test. k A plausible mechanism of transcriptional activation of UP genes with high NELFE by the catalytic histone acetylation. See the Discussion section for details.

We initially hypothesized that the observed transcriptional changes induced by the catalytic histone acetylation were due to the difference in the H2BK120 acetylation level depending on the gene locus. Thus, we conducted ChIP-seq analyses of H2BK120ac. However, the H2BK120 acetylation level increased both in UP and DOWN genes (Fig. 4d, e). Because catalyst 1 consistently bound to nucleosomes irrespective of the gene loci, H2BK120 acetylation was not gene-selective. Thus, we alternatively assumed that a specific transcription factor (TF) was involved in the selective transcriptional changes. We searched for TFs correlated with UP or DOWN genes in the Encyclopedia of DNA Elements (ENCODE) Transcription Factor Targets dataset to identify that UP genes, but not DOWN genes, significantly correlated to the binding sites of negative elongation factor E (NELFE) in human leukemia cell line K562 (Fig. 4f and Supplementary Fig. 11e). NELFE is a component of a negative elongation factor (NELF) complex. ChIP-seq analyses of NELFE in unacetylated THP-1 cells treated with negative control catalyst 2 and acetyl donor 4 confirmed the localization of NELFE in significantly higher levels at UP genes than others (Fig. 4g). We then examined if the catalytic histone acetylation with catalyst 1 and acetyl donor 4 affects chromatin binding of NELFE. ChIP-seq analyses of NELFE in THP-1 cells showed that the chromatin localization of NELFE was notably reduced by the catalytic histone acetylation (Fig. 4e, h), while the protein level of NELFE in the whole cell was comparable between acetylated and unacetylated cells (Fig. 4i). The catalytic histone acetylation comprehensively reduced the localization of NELFE in gene loci with high and low levels of NELFE (Supplementary Fig. 12c). Interestingly, transcription of genes with high-level NELFE was significantly enhanced by the catalytic histone acetylation, whereas transcription of genes with low-level NELFE was not affected (Fig. 4j). This result suggests that high-level NELF inhibits gene transcription, while H2BK120 acetylation induces the release of NELF from chromatin and mitigates the inhibitory effects of high-level NELF (Fig. 4k). Partial knockdown of NELF enhanced the inhibition of cell viability by catalytic histone acetylation (Supplementary Fig. 14a, b). This is likely due to synergistic effects in reducing chromatin-bound NELF through catalytic histone acetylation combined with RNAi-mediated partial knockdown of NELF proteins. Overall, the observed gene selectivity in transcriptional upregulation by the catalytic histone acetylation is likely due to the distribution of high-level NELF on chromatin.

Discussion

Although previous reports showed chromatin localization of H2BK120ac on transcriptionally active genes26,28, functions of H2BK120ac have remained unclear. BAHA-LANA-PEG-CPP44 catalyst 1 selectively and rapidly acetylated H2BK120 in living cells and thus can be used to address the downstream consequences of H2BK120ac. The fact that chromatin localization of NELFE was reduced immediately after H2BK120 acetylation suggests the role of H2BK120ac to release NELF complex from chromatin in THP-1 leukemia cells. Further analysis is needed to elucidate the mechanisms of releasing NELF by H2BK120 acetylation. A reader protein of H2BK120ac might be involved in this process, but such a protein has not been identified yet. NELF was initially isolated from human nuclear extract as a complex that prevents RNA Pol II-dependent transcription on a naked DNA template29. Although in-vitro biochemical and structural studies indicated the role of NELF in RNA pol II pausing at the promoter-proximal region30,31,32, the function of NELF in cells is yet to be fully elucidated. While NELF is essential for regulating promoter-proximal RNA Pol II in the hsp70 gene in Drosophila33,34, several in vivo studies showed that not all promoter-proximal pausing is dependent on NELF35,36,37. Furthermore, a recent study suggested that NELF plays a positive role in transcription elongation by stabilizing 5’ end of transcripts35. Our data showed that transcription of genes with high-level NELFE was upregulated by the catalytic histone acetylation, which likely supports the negative role of NELF in transcriptional elongation. However, the fact that the catalytic histone acetylation also downregulated transcription of a subset of genes may reflect the positive roles of NELF in transcriptional elongation.

Several reports showed that knockout or knockdown of NELF inhibited the development of various cancers38,39,40,41, suggesting that NELF acts as an oncoprotein and is a potential target of cancer chemotherapy. Our observations that the catalytic histone acetylation furnished anti-leukemia effects are likely due to loss of chromatin localization of NELF. Catalyst 1 acetylates H2BK120, but not lysine residues in histone tails, the main targets of endogenous HAT and HDAC enzymes. Therefore, the catalytic histone acetylation is orthogonal to cancer epigenetic drugs targeting histone-modifying enzymes, such as HDAC inhibitors. The anti-cancer effects of our catalyst approach would be further enhanced by combining it with HDAC inhibitors. This strategy intervened in cancer epigenome and changing the transcription profile in living cells entirely relying on chemical catalysis independent of enzymatic processes. In this study, we demonstrated the anti-cancer potential of our method using ex vivo-treated cancer cells. However, our method did not completely inhibit cancer cell growth or prevent mortality in mice. This may be due to the synthetic histone acetylation introduced by our chemical catalyst being gradually diluted by cell division. To improve the in vivo efficacy of our approach, we will need to optimize both the number and dosage of catalyst applications. We attempted to introduce the catalyst molecule into mice via intravenous or intraperitoneal injection but found that the clearance rate was too rapid (t1/2 < 5 min). To improve the in vivo efficacy of our approach, we will need to optimize both the structure of the catalyst and the method of administration. One potential strategy is introducing an albumin-binding moiety to the catalyst molecule, which has been shown to improve pharmacokinetics in several types of drugs42. In addition, using targeted drug-delivery systems, such as nanoparticles, could improve the efficient delivery of the catalyst into target cancer cells in vivo43. Further in vivo studies with an optimized catalyst are needed as the next step to explore the potential clinical applications of our method.

The BAHA-LANA-PEG-CPP44 catalyst 1-acyl donor 4 system is highly modulable (Fig. 1b). For example, changing CPP44 to other CPPs, such as CPP225, may realize catalytic histone acetylation in cancer cells other than leukemia cells as CPP determines the cell-selectivity. Changing the position of the catalytically active BAHA motif on the LANA ligand can change the regioselectivity of histone acetylation from H2BK120 to H2BK108 or H2AK4344, since the BAHA catalyst selectively acetylated a proximal lysine. Changing the acyl group of 4 switches the acyl group introduced to H2BK120 and likely modulates the anti-cancer effects. Furthermore, other PTMs, such as lysine methylation or tyrosine phosphorylation, can be incorporated into histones when BAHA is changed to an alternate catalytically active motif promoting the intended PTM, which is under development in our group. We envision that the method for catalytic cancer epigenome manipulation presented here will lead to a modality of anti-cancer therapy in the future.

Methods

Ethical regulations

Animal experiments were performed in accordance with the institutional guidelines for the use of laboratory animals and approved by the Review Board for Animal Experiments of the University of Tokyo (approval ID, PA18-42). The IRB instructed that mice be euthanized immediately if animals showed more than 20% weight loss within 2–3 days or developed tumors exceeding 10% of body weight. No maximum tumor size/burden or weight loss was exceeded in this study.

Statistics and reproducibility

Statistical analyses were performed using Microsoft Excel (version 16.83) or R (version 4.3.2). P-values were determined by Student’s t tests, one-way ANOVA with Tukey’s post hoc tests, log-rank tests, or Fisher’s exact tests as appropriate and as listed in the figure legends. The statistical significances of differences are specified throughout the figures and legends. For in-test-tube experiments, each experiment was performed in duplicate, and only small variations between individual results were observed in all of the experiments. For in-cell experiments, each experiment was performed in n = 3–4, as indicated in the figure legends.

Cell culture

THP-1 (JCRB0112, purchased from JCRB Cell Bank), MOLM-13 (JCRB1810, purchased from JCRB Cell Bank), MV-4-11 (CRL-9591, purchased from ATCC), Kasumi-1 (JCRB1003, purchased from JCRB Cell Bank) and K562 (gifted from Kitagawa Lab) cells were grown in RPMI medium 1640 (Gibco, 11875) with 10% fetal bovine serum (FBS; Gibco), 100 µ/mL penicillin and 100 µg/mL streptomycin (Gibco) at 37 °C in an atmosphere of 5% CO2. HeLa S3 (gifted from Gotoh Lab), HEK293T (gifted from Hirota Lab) and HepG2 (gifted from Kitagawa Lab) cells were grown in Dulbecco’s modified Eagle medium (DMEM; Gibco, 12430) with 10% fetal bovine serum (FBS; Gibco), 1x GlutaMAX (Gibco) 100 µ/mL penicillin and 100 µg/mL streptomycin (Gibco) at 37 °C in an atmosphere of 5% CO2. HepaMN cells (JCRB1697, purchased from JCRB Cell Bank) were grown in EMUKK-15 medium (Mx), 100 µ/mL penicillin, and 100 µg/mL streptomycin (Gibco) at 37 °C in an atmosphere of 5% CO2. If necessary, 2 µM SAHA (Sigma, SML0061) was supplemented to the growth medium. Cryopreserved Peripheral Blood Mononuclear Cells (PBMC) were purchased from HemaCare (PB009C-1) and used for experiments soon after thawing.

Antibodies

Rabbit polyclonal antibodies against H3K9ac (Merck, 07–352), H3K18ac (abcam, ab1191), H3K4me3 (active motif, 39159), H3K9me3 (abcam, ab8898), H3K36me1 (abcam, ab9048), H3K79me2 (abcam, ab3594) and NELFE (proteintech, 10705-1-AP, for ChIP-seq), a rabbit monoclonal antibody against H2B ubiquitination (Cell signaling technology, 5546S), mouse monoclonal antibodies against H2BK120ac (Fujiwara et al. 8), H3K27me2/3 (abcam, ab6147), NELFE (Santa Cruz Biotechnology, sc-377052, for western blotting), unphosphorylated RNA Pol II45 were used for western blotting and ChIP-seq.

Fluorescent microscopy for living cells using CPP44-tethered compounds

To observe the subcellular distribution of fluorescein isothiocyanate (FITC)-labeled compounds in living cells, cells were washed twice with PBS and treated with the FITC-labeled compounds in a serum-free growth medium for 60 min. at 4 °C or 37 °C. Then, the cells were washed twice with PBS and then imaged by fluorescent microscopy (DMi8; Leica). For the mechanism study of catalyst internalization, the cells were pre-treated with 80 µM dynasore for 60 min.

Cell viability assay

After catalyst-promoted histone acetylation, the cells were washed once with PBS, resuspended in fresh growth medium, and mixed with CellTiter-Glo® 2.0 Reagent (Promega). After 10 min. incubation at r.t., the luminescent signal was measured by GloMax® Discover Microplate Reader (Promega). The cell viability was calculated by dividing the luminescent signal of the cells treated with compounds by the luminescent signal of the cells treated with DMSO.

Catalyst-promoted acetylation of recombinant mononucleosomes

Recombinant nucleosomes were prepared as described previously46. Recombinant nucleosomes (0.37 µM for DNA/histone octamer concentration) were mixed with catalyst, acetyl donor, TCEP (0 mM in Fig. 3 and 0.1 mM in Supplementary Figs. 2, 3) in 20 mM Tris-HCl (pH 7.5) and incubated for indicated time at 37 °C.

Western blotting analysis

After catalyst-promoted histone acetylation, the cells were washed once with PBS and lysed in pre-chilled CRB buffer (50 mM Tris-HCl (pH 7.5), 300 mM NaCl, 0.3% Triton X-100) supplemented with 2 mM MgCl2, 25 mµ/μL Benzonase, protease inhibitor cocktail (Sigma), and 1 mM PMSF on ice for 30 min. After centrifugation (21,130 rcf for 5 min. at 4 °C), the supernatants were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by oriole (Bio-Rad) or CBB (45 g citric acid monohydrate (Sigma), 15 g β-Cyclodextrin (Wako), 0.24 g CBB R-250 (Wako) in 3 L Milli-Q water) staining. After equalizing the protein concentration, the proteins were analyzed by western blotting.

Catalyst-promoted histone acetylation in living cells

Cells were washed twice with PBS and treated with catalyst and acetyl donor. For 2 h or 45 min. reaction, the cells were treated with catalyst and acetyl donor in serum-free growth medium for 2 h or 45 min. at 37 °C. For 4 h reaction, firstly, the cells were treated with catalyst and acetyl donor in serum-free growth medium for 2 h at 37 °C. Then, the cells were incubated in serum-containing growth medium for 1 h at 37 °C, followed by double PBS wash and another 2 h reaction by adding a new catalyst and acetyl donor.

LC-MS/MS analysis and quantification of the stoichiometry of acetylation

For recombinant nucleosomes, the acetylated nucleosomes were precipitated by trichloroacetic acid (16.6%). After DNA was digested by DNase I (Takara) for 30 min. at 37 °C, the samples were mixed with acetone (74%). After overnight incubation at − 30 °C, the proteins were collected by centrifugation, air-dried and dissolved in Milli-Q water. For acid-extracted histones from cells, the histones were dissolved in Milli-Q water and incubated for 10 min. at 25 °C.

To the solution of histones, 50 mM aqueous ammonium bicarbonate (NH4HCO3 aq.) and 25% propionic anhydride solution (methanol/propionic anhydride, 3:1 (vol/vol)) were added, and pH was adjusted to 8 by adding ammonia solution. After 30 min. incubation at 25 °C, the solvents were removed by Speed-Vac evaporator. The samples were resuspended in 50 mM NH4HCO3 aq. with 0.1% ProteaseMAX (Promega) and digested with 10 ng/μL Trypsin Gold (Promega), 10 ng/μL Glu-C (Promega) in 50 mM NH4HCO3 aq. with 0.02% ProteaseMAX at 37 °C for 3 h or overnight. Then, 5% aqueous formic acid (vol/vol) was added, and the solvents were removed by Speed-Vac evaporator to obtain dried digested samples, which were dissolved in 0.1% aqueous formic acid (vol/vol). After centrifugation (21,130 rcf, 5 min.), the supernatant was used for LC-MS/MS analysis (Triple TOF 6600 (AB Sciex) equipped with M5 micro LC (AB Sciex)) as previously reported (Fujiwara et al. 8). Data analysis was carried out using PeakView software (AB Sciex), and the average values were obtained from duplicated measurements. The stoichiometry of acetylated lysines was calculated as a percentage of the total peak area of the extracted ion chromatogram for acetylated peptides in the sum of those for acetylated peptides and propionylated peptides.

RNA-seq

After catalyst-promoted histone acetylation, RNA was extracted from the cells with NucleoSpin® RNA (Takara). TruSeq Stranded mRNA Sample Prep Kit (Illumina) was used to prepare the library for RNA-seq. Deep sequencing was performed on Illumina NovaSeq platform according to the manufacturer’s protocol.

ChIP-seq

After catalyst-promoted histone acetylation, 1.0 × 107 cells were fixed with 1% paraformaldehyde (freshly prepared from powder, filtered before use) in 1 mL PBS for 10 min. at 25 °C, which was quenched by the addition of 80 µL 1.25 M glycine in PBS. Then the cells were washed twice with ice-cold PBS and suspended in 156 µL SDS lysis buffer (50 mM Tris-HCl (pH 8.0), 1% SDS, 10 mM EDTA, protease inhibitor cocktail). After 10 min. incubation on ice, the cells were sonicated with BioRaptor® II (BM Equipment Co). The supernatants were diluted 10-fold with ChIP dilution buffer (16.7 mM Tris-HCl (pH 8.0), 0.01% SDS, 1.1% TritonX-100, 1.2 mM EDTA, 167 mM NaCl) and the DNA concentrations were adjusted by measurement with Nanodrop® lite (Thermo). If needed, spike-in chromatin (active motif) was added. Then, 1500 µL of extract was added to Dynabeads® Protein G (40 µL, Veritas), which was pre-incubated with antibodies (4 µg for anti-H2BK120ac or NELFE antibodies and 10 µg for anti-RNA Pol II antibodies) for ChIP and 2 µg spike-in antibodies (active motif) for 20 h at 4 °C with rotating. Then the mixture was incubated for 48 h at 4 °C with rotating. After incubation, the beads were washed with LS buffer (20 mM Tris-HCl (pH 8.0), 150 mM NaCl, 0.1% SDS, 1% Triton X-100, 2 mM EDTA), HS buffer (20 mM Tris-HCl (pH 8.0), 500 mM NaCl, 0.1% SDS, 1% Triton X-100, 2 mM EDTA), LiCl buffer (10 mM Tris-HCl (pH 8.0), 0.25 M LiCl, 1% NP-40, 1% sodium deoxycholate, 1 mM EDTA) and TE buffer (10 mM Tris-HCl (pH 8.0), 1 mM EDTA), followed by extraction with elution buffer (100 mM NaHCO3, 1% SDS, 10 mM DTT). The supernatant was incubated with 200 mM NaCl for 16 h at 65 °C, and treated with proteinase K for 1 h at 45 °C. DNA was purified from the supernatant of ChIP samples by PCR Clean-Up Mini Kit (Favorgen). ChIP libraries were constructed using the KAPA Hyper Prep Kit (Kapa Biosystems, Inc., Wilmington, MA, USA) according to the manufacturer’s instructions, quantified using Bioanalyzer (Aglient), and sequenced on the Illumina NovaSeq according to the manufacturer’s protocol.

RNA-seq and ChIP-seq analysis

Sequenced reads in ChIP-seq experiment were mapped to UCSC human genome (hg19) using bowtie. Duplicated reads were removed with Picard tools. Peak count was performed by using HOMER software (http://homer.salk.edu/homer/index.html). Metagene profiles were produced with the use of deepTools (https://deeptools.readthedocs.io/en/develop/). Enrichment and Gene Ontology (GO) analysis was performed by using Enrichr (https://maayanlab.cloud/Enrichr/). For analysis for MLL-target genes, the gene set named “Human Gene Set: KUMAR_TARGETS_OF_MLL_AF9_FUSION” was used.

Flow cytometric analysis of cell cycle

After catalyst-promoted histone acetylation, the cells were washed once with PBS and resuspended in pre-chilled 70% ethanol. After 3 h of fixation at − 30 °C, the cells were washed once with PBS and mixed with Muse cell cycle reagent (Luminex). After 30 min. incubation at 25 °C, the cells were analyzed by Muse cell analyzer (Merck).

Flow cytometric analysis of apoptosis

After catalyst-promoted histone acetylation, the cells were washed once with PBS, resuspended in fresh growth medium and mixed with Muse annexin V & dead cell reagent (Luminex). After 20 min. incubation at 25 °C, the cells were analyzed by Muse cell analyzer (Merck).

Cell viability analysis

After catalyst-promoted histone acetylation, the cells were washed once with PBS and resuspended in fresh growth medium. Then, the cells were plated in 96-well plates at a density of 2 × 105 cells/mL in a final volume of 200 µL. The number of trypan-blue negative living cells were counted with TC20 automated cell counter (Bio-Rad), and the cells were replated at equal cell numbers in a fresh growth medium every 2 days.

Mice experiments

NOD.Cg-Prkdcscid Il2rgtm1Sug/ShiJic (NOG) were purchased from In-Vivo Science Inc. The AkaBLI system, composed of AkaLumine-HCl and Akaluc, provides a light source of sufficient strength to penetrate body walls, even in deep tissue areas. Akaluc was introduced into MOLM-13 cells using the CS-CDF-UbC-mScarlet-P2A-Akaluc-PRE lentivirus to confirm tumor burden in mice47. After catalyst-promoted histone acetylation, the cells were washed once with PBS and resuspended in PBS. Then, the cells (3 × 106 cells) were injected to 9-week-old NOG mice. To monitor the tumor burden, one hundred microliters of 5 mM AkaLumine-HCl (Wako) was injected intraperitoneally into mice immediately before image analysis, and mice under isoflurane anesthesia were imaged within 5–10 min after injection. The signal intensity was measured using an IVIS (Perkin Elmer).

Catalyst-promoted histone acetylation for living cells (cell impermeable catalyst)

HeLa S3 cells were treated with 100 μM BSO in a growth medium for 1 day at 37 °C. 0.5 mM catalyst was preincubated with acetyl donor, 2 mM TCEP, and 0.5 mg/mL Dextran-tetramethylrhodamine (Thermo) in PBS for 10 min. The preincubation mixture was incorporated into the cells by bead loading16, and the cells were incubated in growth medium containing acetyl donor and 100 μM BSO for 5 h at 37 °C. The cells harvested with accutase (Innovative Cell Technologies) were incubated with 1 μg/mL DAPI in growth medium for 15 min. at 25 °C to stain dead cells and sorted by FACS Arial III (BD Biosciences). The living cells (DAPI negative) were classified into three groups with their tetramethylrhodamine signals: low population have the same signal as nontreated cells, high population have the highest about 5% intensity of the sample, and remaining cells were classified as mid population. The histones were isolated from high-population cells with acid extraction48. The high population cells were washed once with PBS and incubated with 1 mL hypotonic lysis buffer (10 mM Tris-HCl (pH 8.0), 1 mM KCl, 1.5 mM MgCl2, 1 mM DTT, 1x cOmplete protease inhibitor mixture (Roche), 1 mM PMSF) on ice for 30 min. After centrifugation (10,000 × g for 10 min. at 4 °C), the precipitates were suspended in 200 µL 0.4 N H2SO4 aq. and incubated on ice for 30 min. After centrifugation (16,000 × g for 10 min. at 4 °C), the supernatants were mixed with 65 µL TCA and incubated on ice for 30 min. After centrifugation (16,000 × g for 10 min. at 4 °C), the precipitates were washed with acetone twice, and air dried.

Fluorescent microscopy for living cells using 8r/16r-tethered compounds

To observe the subcellular distribution of fluorescein isothiocyanate (FITC)-labeled compounds in living cells, cells were washed three times with PBS and treated with the FITC-labeled compounds in a serum-free growth medium for 60 min. at 37 °C. If necessary, LysoTracker Deep Red (Thermo) was supplemented to the medium. Then, the cells were washed three times with FluoroBrite DMEM (Gibco) containing 10% FBS, GlutaMAX, 100 µ/mL penicillin, 100 mg/mL streptomycin, and 1 mg/mL heparin sodium (Wako, 081–00136), and then imaged by fluorescent microscopy (DMi8; Leica).

Catalyst-promoted histone acetylation in HeLa S3 or HEK293T cells

Cells were washed three times with PBS and treated with 5 μM catalyst in serum-free growth medium for 1 h at 37 °C, then washed three times with 1 mg/mL heparin sodium in PBS, and treated with 400 µM acetyl donor in 10% serum-containing growth medium for 3 h at 37 °C. Then, the cells were harvested and stained by a DAPI-containing growth medium, and living cells were collected by FACS, followed by acid extraction of histones. The histones were analyzed by SDS-PAGE followed by oriole staining or LC-MS/MS analysis. After equalizing the protein concentration, the histones were analyzed by western blotting.

Electrophoretic mobility shift assays

Electrophoretic mobility shift assays of nucleosomes and catalysts were conducted as previously reported49. 50 nM recombinant nucleosomes were incubated in assay buffer (10 mM Tris-HCl (pH 7.5), 50 mM NaCl, 0.5 mM TCEP, 6% glycerol) in a 7 µL reaction mixture for 15 min. on ice. The mixture was resolved in a 0.2x TBE 5% 59:1 acrylamide:bis-acrylamide gel for 90 min. at 70 volts on ice. The gel was stained with ethidium bromide (Promega).

Preparation of histone octamer

Canonical human histones (H2A, H2B, H3.1, H4) were expressed in E. coli, purified, and reconstituted to histone octamer by salt gradient dialysis as previously described50.

Preparation of 192 bp x 12 601 DNA

E.coli (DH5a) was transformed by pWM_12 × 601_45bpLinker51, gifted from Michael K. Rosen). The cells were cultured in 5 L of LB, and collected by centrifugation. The plasmids were purified by QIAfilter Plasmid Giga Kits (QIAGEN). The pellets of purified plasmids were dissolved in TE10:0.1 (10 mM Tris-HCl (pH 7.9), 0.1 mM EDTA) at 1 mg/mL, and 10x CutSmart Buffer (NEB) was added. 5000 U/21 mg-DNA of EcoRV-HF (NEB) was added to the solution, and the solution was incubated at 37 °C for 16 h. Reaction completion was checked by agarose gel electrophoresis, and quenched by adding 20 mM EDTA. The DNA solution was isolated by the addition of an equal volume of phenol: chloroform: isoamyl alcohol (25:24:1) suspension, vigorous shaking, centrifugation at 22,140 × g for 2 min., and collection of the supernatant; this procedure was repeated twice. To the resulting solution, equal volume of chloroform was added, vigorously shaken, and the mixture was centrifuged at 22,140 × g for 2 min. to remove residual phenol. After the collection of the supernatant, the DNA was precipitated by ethanol precipitation. The DNA was re-dissolved in TE10:0.1, and the concentration was adjusted to ca 2 mg/mL. To the DNA solution, x 0.154 (against the DNA solution before the addition of NaCl) of 5 M NaCl was added, and 40% PEG-6000 was added portion-wise until the solution became cloudy. The suspension was centrifuged at 15,000 × g at 4 °C for 20 min., and the supernatant was removed. The resulting pellet was re-dissolved in TE10:0.1, and the same procedure was repeated. The 12 × 601 DNA was precipitated by ethanol precipitation to remove residual PEG-6000 and re-dissolved in TE10:0.1. The solution was filtrated by 5 µm syringe filter and further purified by AKTA (GE health care) equipped with DEAE column (TSKgel DEAE-5PW, TOSOH). Combination of wash buffer (10 mM Tris-HCl (pH 7.5), 0.1 mM EDTA) and elution buffer (10 mM Tris-HCl (pH 7.5), 0.1 mM EDTA, 0.6 M NaCl) was used as a mobile phase, following the time program as follows: 50% elution buffer for 3 column volume (CV) followed by a linear gradient of 50–100% elution buffer over 10 CV. The fractions containing pure 12 × 601 DNA was collected, and the DNA was precipitated by ethanol precipitation. The resulting pellet was dissolved in TE10:0.1, flash frozen, and stored at − 80 °C.

Reconstitution of 12-mer recombinant nucleosome

400 µg of 12 × 601 DNA was mixed with 1.7, 1.9, or 2.1 times (in mole) of H3.1 histone octamer in 4 M NaCl solution. The solution was sealed in a dialysis tube, and the tube was immersed into 400 mL of dialysis buffer-high salt (10 mM Tris-HCl (pH 7.5), 1 mM EDTA, 2 M NaCl, 1 mM DTT) in the tall beaker. Dialysis buffer-low salt (10 mM Tris-HCl (pH 7.5), 1 mM EDTA, 2 M NaCl, 1 mM DTT) was transferred to the beaker by peristaltic pump at 4 °C over 32 h. The volume was maintained at 400 mL by discarding increased volume by peristaltic pump. The dialysis tube was transferred to 400 mL of storage buffer (10 mM Tris-HCl (pH 7.5), 1 mM DTT, 5% Glycerol) and incubated at 4 °C for at least 4 h with stirring. The sample was fractionated in a 10–40% sucrose gradient by ultracentrifugation (22,000 rpm, 4 °C, 14 h) using SW41Ti Swinging-Bucket Rotor (Beckman Colter), and all fractions were analyzed by agarose gel electrophoresis (0.7% Agarose, 0.2xTAE buffe). The fractions containing pure 12-mer nucleosome collected, transferred to dialysis tube, and dialyzed in 1 L of storage buffer at 4 °C for at least 4 h, twice. The solution was concentrated by iterative ultrafiltration using Amicon Ultra-15 100 kDa (Merck millipore). The resulting solution was dispensed, flash frozen by liquid N2, and sored at − 80 °C.

Quality check of 12-mer nucleosomal array

12-mer nucleosomal array was diluted to 100 ng/µL (DNA concentration) and digested by BstXI (NEB) or BsiWI (NEB) in low Mg buffer (10 mM Tris-HCl (pH 7.5), 50 mM NaCl, 0.5 mM MgCl2) for 16 h, at 30 °C and 50 °C, respectively. The reaction mixture of BstXI was analyzed by Native PAGE using 5% native gel in 1x TBE. The reaction mixture of BsiWI was analyzed by agarose gel electrophoresis using 0.7% agarose gel in 0.2x TBE. The gels were stained by ethidium bromide solution and imaged by UV transilluminator. The lot containing minimal digested fragments (in BsiWI digested samples) and minimal di- or tri-nucleosomes (in BstXI digested samples) was selected and used for the subsequent assays.

Catalyst-promoted acetylation of recombinant 12-mer nucleosome arrays

12-mer recombinant nucleosomes (0.37 µM for DNA/histone octamer concentration) were mixed with catalyst 1 or 2 (2.0 µM) and acetyl donor 4 (0.1 mM), in Nucleosome reaction buffer (20 mM HEPES-KOH (pH 7.5), 20 mM NaCl) and incubated for 2 h at 37 °C. The reaction mixture was quenched by the addition of 500 mM NaCl and the acetylated nucleosomes were purified by ultrafiltration using Amicon Ultra − 0.5 mL 100 kDa (Merck Millipore) with High salt buffer (20 mM HEPES-KOH (pH 7.5), 500 mM NaCl, 1 mM DTT, 0.005% Triton X-100) for 4 times to remove compounds, followed by No salt buffer (20 mM HEPES-KOH (pH 7.5), 1 mM DTT, 0.005% Triton X-100) for 2 times to remove NaCl.

Mg2+-promoted nucleosome sedimentation assay

The concentration of nucleosomes was adjusted to about 60 ng/µL dsDNA by No salt buffer. Then 7.5 µL of the acetylated nucleosomes in No salt buffer were mixed with 0.833 µL of 10x MgCl2 solution, and the mixture was incubated for 30 min. at 25 °C. After centrifugation at 21,130 × g for 10 min. at 25 °C, A260 of the supernatant was measured using NanoDrop Lite (Thermo).

Phase separation assay of 12-mer recombinant nucleosome

384 well glass bottom plate was passivated by 100 mg/mL BSA for 30 min. and washed vigorously with MQ. Using the 384 well plate, a phase separation assay was conducted based on the protocol previously described5. Briefly, in the 384 well plate, 5 µL of Chromatin dilution buffer (25 mM Tris-OAc (pH7.5), 0.1 mM EDTA, 5% Glycerol, 5 mM DTT, 0.1 mg/mL BSA, 750 nM (as mono-nucleosome) 12-mer recombinant nucleosome) and 5 µL of Phase separation buffer (25 mM Tris-OAc (pH7.5), 0.1 mM EDTA, 5% Glycerol, 5 mM DTT, 0.1 mg/mL BSA, 2x Mg(OAc)2, 2 µg/mL Glucose oxidase, 350 ng/mL Catalase, 4 mM Glucose) were mixed. After incubation for 2.5 h at 25 °C, the plate was subjected to microscopy.

Catalyst-promoted acetylation of nucleosomes in cell extract

THP-1 cells were washed once with PBS and lysed in pre-chilled CRB buffer (50 mM Tris-HCl (pH 7.5), 300 mM NaCl, 0.3% Triton X-100) supplemented with 2 mM MgCl2, 25 mU/μL Benzonase, protease inhibitor cocktail and 1 mM PMSF on ice for 30 min. After centrifugation (15,000 rpm for 5 min. at 4 °C), the supernatants were mixed with 2 µM catalyst and 0.1 mM acetyl donor and incubated for 2 h at 37 °C. After equalizing the protein concentration by SDS-PAGE and oriole staining, the cell extract was analyzed by western blotting.

siRNA treatment

In the 12-well plate, control siRNA-A (SantaCruz, sc-37007, 24 pmol for p300/CBP, 96 pmol for NELF-A), p300 siRNA (SantaCruz, sc-29431, 12 pmol) and CBP siRNA (SantaCruz, sc-29244, 12 pmol), or NELF-A siRNA (SantaCruz, sc-38095, 96 pmol) were mixed with Lipofectamine RNAiMax transfection reagent (Invitrogen, 13778075), followed by 10 min. incubation at r.t. Then, 2.8 × 105 THP-1 cells in 1 mL RPMI medium were added to the plate and incubated at 37 °C for 24 h (p300/CBP) or 72 h (NELF-A), followed by catalyst-promoted acetylation and western blotting as described above.

Catalyst-promoted histone acylation in living cells (SDS-PAGE or western blotting)

Cells were washed twice with PBS and treated with 10 µM catalyst and 0.1 mM acetyl donor in serum-free growth medium for 1 h at 37 °C, followed by acid extraction of histones.

For pentynoyl donor (S35) and azidobutyryl donor (S36), the histones were mixed with click reagent (50 µM 5-TAMRA-alkyne (Click Chemistry Tools) or Azide-fluor 545 (Sigma), 200 µM TBTA (TCI), 200 µM CuSO4 (Wako), 2.5 mM sodium ascorbate (Wako)) and incubated 18 h at 25 °C. Then the mixture was mixed with 14 µL TCA and incubated on ice for 30 min. After centrifugation (15,000 rpm for 5 min. at 4 °C), the precipitates were washed twice with acetone and air dried. The histones were analyzed by SDS-PAGE followed by CBB staining. After equalizing the protein concentration, the proteins were analyzed by SDS-PAGE.

For methyl succinyl donor (S37) and butyryl donor (S38), the histones were analyzed by western blotting.

Catalyst-promoted histone acylation in living cells (microscope)

Cells were washed twice with PBS and treated with 10 µM catalyst and 0.1 mM acetyl donor in serum-free growth medium for 1 h at 37 °C, followed by PBS wash and fixation with 4% PFA in PBS for 10 min. at 25 °C. The cells were washed twice with TBS-Tx (25 mM Tris-HCl (pH 7.4), 137 mM NaCl, 2.7 mM KCl, 0.1% Triton X-100) and incubated in 3% BSA (Roche) in TBST (20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 0.1% Tween20) for 30 min. at 25 °C, followed by TBST wash. The cells were mixed with click reagent (50 µM 5-TAMRA-alkyne (Click Chemistry Tools) or Azide-fluor 545 (Sigma), 200 µM TBTA (TCI), 200 µM CuSO4 (Wako), 2.5 mM sodium ascorbate (Wako)) and incubated 18 h at 25 °C. Then the cells were washed five times with TBST and imaged by fluorescent microscopy (DMi8) in TBST containing 1 µg/mL DAPI (Dojindo Molecular Technologies).

RT-qPCR

After catalyst-promoted histone acetylation, RNA was extracted from the cells with NucleoSpin® RNA and reverse transcribed using SuperScript III First-Strand Synthesis System for RT-PCR (Invitrogen). cDNA was quantified by real-time PCR (LightCycler 480 II, Roche) with cycling conditions of 2 min. at 50 °C, 10 min. at 95 °C, 40 cycles for 15 s at 95 °C and 1 min. at 60 °C. The target mRNA level was normalized to 18S rRNA.

Primers for RT-qPCR

 

Forward primers

Reverse primers

18S rRNA52

CGACGACCCATTCGAACGTCT

CTCTCCGGAATCGAACCCTGA

EGR153

AGCCCTACGAGCACCTGAC

GGGCAGTCGAGTGGTTTG

BTG254

AAGATGGACCCCATCATCAG

AGCACTTGGTTCTTGCAGGT

JUN55

GAGCTGGAGCGCCTGATAAT

CCCTCCTGCTCATCTGTCAC

MYC56

TGAGGAGACACCGCCCAC

CAACATCGATTTCTTCCTCATCTTC

P2157

CATGCCAGCTACTTCCTCCT

CAGGTCTGAGTGTCCAGGAA

HOXA1058

CTGGGCAATTCCAAAGGTGA

ACTCTTTGCCGTGAGCCAGT

MEIS159

GATATAGCCGTGTTCGCCAAA

CGGTGGCAGAAATTGTCACAT

NELF-A39

GTCGGCAGTGAAGCTCAAGT

TTCACACTCACCCACCTTTTCT

General comments for chemical synthesis

MALDI-TOF-MS spectra were measured on a Shimadzu Biotech Axima-ToF2 TM spectrometer (Shimadzu, Co., Kyoto, Japan) using α-cyano-4-hydroxy cinnamic acid (Sigma-Aldrich, Inc., St. Louis, MO, U. S. A.) as a matrix. Milli-Q water means distilled water purified with a Millipore Milli-Q water purification system (Merck KGa, Co., Darmstadt, Germany). All protected α-amino acids were purchased from Watanabe Chemical Industries, Ltd. (Hiroshima, Japan). Acetyl donor 4, pentynoyl donor S35, azidobutyryl donor S36, methyl succinyl donor S37, and butyryl donor S38 were synthesized as previously reported21. DSH-LANA5-15 S13 was synthesized as previously reported20. NAC-Ac S12 was synthesized as previously reported60. DSSMe-LANA5-22-PEG0 S19, DSSMe-LANA5-22-PEG550 S20 were synthesized as previously reported8. Other reagents were used as received from commercial sources (Aldrich, TCI, or Wako) unless otherwise stated.

Analytical HPLC

Analytical HPLC was conducted by using a JASCO HPLC system equipped with a UV-2075 spectrometer, PU-4180 pumps, a DG-2080-32 degasser, and an MX-2080-32 mixer (JASCO International, Co., Ltd.). Eluent was linear gradient of MeCN in 0.1% TFA aq. (LC-MS grade solvent). The C18 reverse phase column was used at 40 °C at a flow rate of 1 mL/min.

Preparative HPLC

Preparative HPLC was conducted by using a JASCO HPLC system equipped with a UV-2075 spectrometer, PU-2086 pumps, a DG-2080-53 degasser, and an MX-2080-32 mixer (JASCO International, Co., Ltd.). Eluent was linear gradient of MeCN in 0.1% TFA aq. (LC-MS grade solvent). The C18 reverse phase column was used at 40 °C at a flow rate of 3 or 10 mL/min.

LC-MS

LC-MS analysis was conducted by using an Agilent Technologies LC-MS (ESI) system equipped with a 1260 Infinity High-Performance Degasser, an Agilent 1260 Infinity Binary Pump, a 1260 Infinity Standard Autosampler, a 1290 Infinity Thermostatted Column Compartment, a 1260 Infinity Variable Wavelength Detector, and an Agilent 6120 Single Quadrupole LC-MS (Agilent Technologies, Inc., Santa Clara, CA, U. S. A.). Retention times (Rt/min.) were recorded using a gradient elution method of 2 − 90% B over 13 min., solution A consisted of water (buffered with 0.1% HCO2H) and solution B consisted of acetonitrile (LC-MS grade solvents). C18 reverse phase column (2.0 × 50 mm; YMC-Triart C18; YMC Co., Ltd.) at 40 °C at a flow rate of 0.2 mL/min. The eluent was monitored by absorbance at 254 nm.

General SPPS

Peptides were synthesized on a solid phase using Rink Amide AM resin (Novabiochem). Fmoc-amino acid (3 eq.) was sequentially coupled using HATU (2.7 eq.) and DIEA (6 eq.) in 3 mL DMF for 60 min. at 25 °C. after removal of each Fmoc group with 20% piperidine-DMF for 10 min. N-terminus was capped by DSSMe-CO2H8 using HATU (2 eq.) and DIEA (5 eq.), DSSTrt-CO2H8 using HATU (2 eq.) and DIEA (5 eq.) or acetic anhydride in DCM, if necessary. The peptide was cleaved from the resin by treatment with TFA in the presence of TIPS and H2O (95:2.5:2.5) for 3 h at 25 °C, concentrated under reduced pressure, and precipitated with ether to afford crude peptides, which were purified with preparative HPLC to afford product peptides as solids after lyophilization. For the one-pot synthesis of peptide containing 2-pyridinesulfenyl-protected Cys, 2,2’-dipyridyl disulfide (20 eq.) was added in TFA/TIPS/H2O. FITC- or maleimide-labeled peptide was synthesized by SPPS using Fmoc-Lys(Mtt)-OH. The Mtt group on the lysine was removed by treatment with TFA and TIPS in DCM (3:5:92) for 5 min. x 4 times at r.t. The partially deprotected compound on the resin was treated with fluorescein isothiocyanate isomer I (3 eq.) and DIEA (6 eq.) in DMF or 6-maleimidohexanoic acid (3 eq.), HATU (2.7 eq.) and DIEA (6 eq.) in DMF for 4 h at r.t.

HPLC Methods

Method A: YMC-Triart C18 (150 × 4.6 mm I.D.), linear gradient of 10–100% acetonitrile in 0.1% aqueous TFA over 3–18 min. (HPLC analysis at 230 nm).

Method B: YMC-Triart C18 (50 × 2 mm I.D.), linear gradient of 30–70% acetonitrile in 0.1% aqueous HCO2H over 3–8 min. (HPLC analysis at 230 nm).

Method C: YMC-Triart C18 (150 × 4.6 mm I.D.), linear gradient of 2–90% acetonitrile in 0.1% aqueous TFA over 3–18 min (HPLC analysis at 230 nm)

Method D: YMC-Triart PFP (150 × 4.6 mm I.D.), linear gradient of 10–100% acetonitrile in 0.1% aqueous TFA over 8–20 min (HPLC analysis at 230 nm)

Method E: YMC-Triart C18 (50 × 2 mm I.D.), linear gradient of 30–70% acetonitrile in 0.1% aqueous HCO2H over 3–8 min, and 70–98% over 8–9 min (HPLC analysis at 254 nm).

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.