Skip to main content
NCBI home page
Elsevier Sponsored Documents logoLink to Elsevier Sponsored Documents
. 2016 Apr 21;62(2):169–180. doi: 10.1016/j.molcel.2016.03.014

Dynamic Competing Histone H4 K5K8 Acetylation and Butyrylation Are Hallmarks of Highly Active Gene Promoters

Afsaneh Goudarzi 1,8, Di Zhang 2,8, He Huang 2, Sophie Barral 1, Oh Kwang Kwon 2, Shankang Qi 2, Zhanyun Tang 3, Thierry Buchou 1, Anne-Laure Vitte 1, Tieming He 4, Zhongyi Cheng 4, Emilie Montellier 1, Jonathan Gaucher 1,5, Sandrine Curtet 1, Alexandra Debernardi 1, Guillaume Charbonnier 6, Denis Puthier 6, Carlo Petosa 7, Daniel Panne 5, Sophie Rousseaux 1, Robert G Roeder 3, Yingming Zhao 2,9,∗, Saadi Khochbin 1,9,∗∗
PMCID: PMC4850424  PMID: 27105113

Summary

Recently discovered histone lysine acylation marks increase the functional diversity of nucleosomes well beyond acetylation. Here, we focus on histone butyrylation in the context of sperm cell differentiation. Specifically, we investigate the butyrylation of histone H4 lysine 5 and 8 at gene promoters where acetylation guides the binding of Brdt, a bromodomain-containing protein, thereby mediating stage-specific gene expression programs and post-meiotic chromatin reorganization. Genome-wide mapping data show that highly active Brdt-bound gene promoters systematically harbor competing histone acetylation and butyrylation marks at H4 K5 and H4 K8. Despite acting as a direct stimulator of transcription, histone butyrylation competes with acetylation, especially at H4 K5, to prevent Brdt binding. Additionally, H4 K5K8 butyrylation also marks retarded histone removal during late spermatogenesis. Hence, alternating H4 acetylation and butyrylation, while sustaining direct gene activation and dynamic bromodomain binding, could impact the final male epigenome features.

Graphical Abstract

graphic file with name fx1.jpg

Open in a new tab

Highlights

  • •

    Active gene TSSs are marked by competing H4 K5K8 acetylation and butyrylation

  • •

    Histone butyrylation directly stimulates transcription

  • •

    H4K5 butyrylation prevents binding of the testis specific gene expression-driver Brdt

  • •

    H4K5K8 butyrylation is associated with delayed histone removal in spermatogenic cells

Histone butyrylation stimulates gene transcription while competing with acetylation at H4K5 to control Brdt bromodomain binding. Differential chromatin labeling with interchangeable H4 acylations is an important epigenetic regulatory mechanism controlling gene expression and chromatin reorganization.

Introduction

Besides lysine acetylation, we recently identified a variety of short-chain lysine acylations in core histones, including lysine propionylation, butyrylation, 2-hydroxyisobutyrylation, crotonylation, malonylation, succinylation, and glutarylation (Chen et al., 2007, Dai et al., 2014, Tan et al., 2011, Tan et al., 2014, Xie et al., 2012). Emerging data suggest that these new histone lysine acylations may have unique functions that depend not only on cell metabolism, but also on their ability to be deposited or removed by specific enzymes (Dai et al., 2014, Montellier et al., 2012, Rousseaux and Khochbin, 2015, Sabari et al., 2015, Sin et al., 2012, Tan et al., 2011). Nevertheless, the functional impact of differential histone acylation on chromatin recognition by specific factors has remained unexplored.

This study aims to understand the functional consequences of differential histone acylation. In particular, we decided to investigate histone butyrylation, because, in contrast to the acetyl (2-carbon) and propionyl (3-carbon) groups, the butyryl (4-carbon) group restricts the binding of bromodomains (Flynn et al., 2015). More specifically, we focused our attention on histone H4 at K5 and K8, whose acetylation is required to bind the first bromodomain of Brdt, a testis-specific member of the BET protein family (Morinière et al., 2009). Our previous work showed that Brdt stimulates the transcription of certain spermatogenic-specific genes by recruiting the P-TEFb complex and by directly binding to their transcriptional start sites (TSSs). Additionally, during late spermatogenesis, Brdt’s first bromodomain is necessary for the replacement of histones by non-histone sperm-specific transition proteins (TPs) and protamines (Prms) (Gaucher et al., 2012). Given the critical role of H4K5 and H4K8 acetylation in Brdt-driven activities, we hypothesized that other mutually exclusive histone marks at these two residues might have key regulatory roles in sperm cell genome programming.

Here, we identify major histone lysine butyrylation sites in cells from different species, including mouse spermatogenic cells. Using spermatogenesis as an integrated biological model system, in addition to in vitro experiments and targeted proteomic approaches, we demonstrate new characteristics of active gene TSSs. Our data indicate that interchangeable acetylation and butyrylation at H4K5 and H4K8 not only stimulates transcription, but could also underlie a highly dynamic interaction of histone post-translational modification (PTM)-binding factors such as Brdt. Additional data further show that stable differential use of acetylation and butyrylation could also durably affect genome organization in the maturing sperm. Altogether, these findings indicate how competition between histone acylation states could be an important epigenetic regulatory mechanism.

Results

Histone Lysine Butyrylation Is an Evolutionarily Conserved PTM

To identify histone butyryllysine (Kbu) sites and study their function, we first confirmed the presence of histone Kbu by western blotting. Our data suggest that histone Kbu is an evolutionarily conserved PTM in eukaryotic cells (Figure 1A). We then used mass spectrometry to identify possible Kbu sites in core histones from three species (Chen et al., 2007, Kim et al., 2006). Kbu sites were detected in the N-terminal tails of H3 (K9, K14, K18, K23, K27, K36, K37, K79, and K122), H4 (K5, K8, K12, and K16), and H2B (K5 and K20) (Figure 1B; Data S1).

Figure 1.

Figure 1

Open in a new tab

Histone Lysine Butyrylation Is an Evolutionarily Conserved PTM in Eukaryotic Cells

(A) The Kbu residues in core histones from the indicated species were detected by western blotting using a pan anti-Kbu antibody (upper). The corresponding Coomassie blue stained gel is shown (lower).

(B) Illustrations of histone Kac and Kbu sites in core histones identified by tandem mass spectrometry (MS/MS) (acetyl, Ac and butyryl, Bu). The annotated MS/MS spectra for histone Kbu peptides and the specific co-occurrence of K5bu and K8bu in H4 from spermatogenic cells are shown in Data S1.

Functional Significance of Histone Butyrylation

To examine the function of Kbu in chromatin biology, we used mammalian spermatogenesis as a model system. Spermatogenic cells can be roughly classified into three major types: proliferative progenitor cells (spermatogonia), meiotic cells (spermatocytes), and post-meiotic haploid cells (spermatids). During differentiation, highly specific gene expression programs are activated in both meiotic and early post-meiotic cells. Large-scale genome reorganization also takes place in spermatids, where a genome-wide replacement of histones by TPs and Prms occurs in post-meiotic cells known as elongating and condensing spermatids (Gaucher et al., 2010, Goudarzi et al., 2014, Govin et al., 2004).

To confirm the existence of histone butyrylation in spermatogenic cells, we identified histone Kbu sites by mass spectrometry analysis of histones from mouse testis. We detected ten butyrylation sites including H4K5bu and H4K8bu (occurring separately or in combination), supporting the presence of these two histone marks in spermatogenic cells (Figure 1B; Data S1; see also Figure 7B). We then used highly specific anti-H4K5bu and anti-H4K8bu antibodies, along with the anti-H4K5ac and anti-H4K8ac antibodies, to investigate the stage-specific presence of these marks in spermatogenic cells. Immunohistochemistry analysis showed that H4K5 and K8 butyrylation is enhanced in elongating spermatids (Figure 2A), similar to earlier observations for histone H4K5 and H4K8 acetylation (Hazzouri et al., 2000).

Figure 7.

Figure 7

Open in a new tab

p300-Mediated Acetylation and Butyrylation of H4K5K8 and Functional Implications for Gene Promoter Activity

(A) In vitro reconstituted octamers were incubated with purified p300 in the presence of equal amounts of acetyl-CoA and butyryl-CoA, and the histones were subsequently analyzed by MS. The unique histone H4 peptides bearing multiple lysine modifications, which were specifically identified, are indicated along with their corresponding spectrum counts. The annotated MS/MS spectra for histone Kbu/ac peptides are presented in Data S1.

(B) Stoichiometry of H4K5bu and K8bu in sperm cells. The green letters indicate the in vitro chemical butyrylation, and the red letters indicate the endogenous modifications. The % represents the respective ratios of the peptide bearing H4K5bu and H4K8bu (summed peak areas) over the corresponding unmodified H4 (total germ cells: TGC; spermatocytes: Spc; round spermatids: RS; and elongating/condensing spermatids: E/CS). The values of peak area were used for calculations (Table S2; Figure S4A).

(C) Model of the functional interplay between acylation marks. We propose that at active gene TSSs, p300 (and possibly other HATs) randomly use acetyl- or butyryl-CoA to modify histone tails. HDACs and histone variants, also present at dynamic chromatin regions, accelerate the turnover of these histone marks. Alternate modifications of H4K5 with acetyl and butyryl marks, while maintaining the region competent for transcription, also lead to an oscillation in Brdt binding. The co-localization of H2A.Lap1 and the four H4 PTMs at gene TSSs is shown in Figure S4B. Dynamic Brdt binding could help the pre-initiation complex and RNA PolII to reassemble at TSSs and would favor successive cycles of gene transcription.

Figure 2.

Figure 2

Open in a new tab

Occurrence of H4K5ac, K5bu, K8ac, and K8bu Marks during Mouse Spermatogenesis

(A and B) Detection by immunohistochemistry on paraffin sections of staged testis tubules (A) and by immunofluorescence in elongating/condensing spermatids (B) using the indicated antibodies are shown. The tubule sections shown in (A) represent stage IX or X of spermatogenesis. The use of mouse monoclonal anti-H4K5ac or H4K8ac antibodies together with rabbit polyclonal anti-H4K5bu or H4K8bu antibodies allowed for the co-detection of the indicated marks. (B) Spermatids are presented as a function of their differentiation along spermiogenesis (upper to lower) as judged by their morphology and degree of genome compaction. The scale bars represent 5 μm. The images were acquired and treated under the same conditions.

(C) Total extracts from fractionated spermatogenic cells corresponding to cells enriched at the indicated stages along with extracts from the total (non-fractionated) spermatogenic cells were used to detect the indicated marks by western blotting. Compared to acetylation, longer exposure times were required for the butyrylation signal to be detected in all samples.

To study the dynamic changes of butyrylation versus acetylation at both H4K5 and H4K8 sites in spermatogenic cells, we examined the co-existence of H4K5ac and H4K5bu as well as that of H4K8ac and H4K8bu. In elongating spermatids, H4K5ac and H4K8ac are widely distributed, but their localization becomes biased toward the sub-acrosomal regions in later stages. In contrast, H4K5bu- and H4K8bu-containing nucleosomes are homogenously distributed in the same cells (Figure 2B). In late elongating spermatids, while acetylated histones are removed and degraded (Gaucher et al., 2012, Qian et al., 2013), butyrylated H4 species escape this wave of acetylation-dependent histone removal, to finally disappear in condensing spermatids (Figure 2B). Interestingly, immunoblotting experiments, using fractionated spermatogenic cells, confirmed that butyrylation persists longer than acetylation on H4K5 and K8 sites (Figure 2C). These results highlight a bimodal histone removal process, whereby the removal of H4 K5/K8 butyrylated nucleosome occurs after that of H4 K5/K8 acetylated nucleosomes.

Genome-wide Distribution of H4K5 and H4K8 Acetylation and Butyrylation

Our immunofluorescence analysis showed distinctive patterns of acetylation and butyrylation at H4K5 and H4K8, with both marks showing enhanced labeling in elongating spermatids, when cellular transcription dramatically decreases (Gaucher et al., 2010). This observation motivated us to study their genomic distributions and potential functions at earlier stages of spermatogenesis, when specific spermatogenic gene expression programs are activated (Dai et al., 2014, Gaucher et al., 2012, Tan et al., 2011). Toward this goal, mouse spermatogenic cells were fractionated into spermatocytes and post-meiotic round spermatids (Dai et al., 2014, Gaucher et al., 2012, Tan et al., 2011). The two pools of cells were subjected to chromatin immunoprecipitation followed by deep sequencing (ChIP-seq) using four anti-PTM-specific antibodies (anti-H4K5ac, -H4K8ac, -H4K5bu, and -H4K8bu). Analysis of the genomic distribution of these marks revealed that regions surrounding gene TSSs (located upstream of TSSs and at the 5′UTR) exhibit the highest coverage by the four PTMs in both spermatocytes and spermatids (Figure 3A). Further analyses showed that the four H4 PTMs are enriched at TSSs in a manner dependent on the transcriptional activity of the corresponding genes (Figure 3B).

Figure 3.

Figure 3

Open in a new tab

Genome-wide Mapping of H4K5ac, K5bu, K8ac, and K8bu Marks in Meiotic Spermatocytes and Post-meiotic Round Spermatids

(A) Bar diagrams showing the coverage of each of the genomic elements (as indicated) by different combinations of 4, 3, 2, 1, and 0 marks. 100% represents the total annotated genomic element in base pairs.

(B) Log2 ratio of ChIP-seq versus input signal corresponding to the four histone marks (with the indicated color code) is visualized over the gene TSS regions (± 5,000 bp). There are four gene categories that are ordered from left to right according to the level of their expression in the indicated cell types, spermatocytes, or round spermatids. To this aim, we used our transcriptomic analyses of these cells (Montellier et al., 2013). Antibody characterizations and ChIP-qPCR on selected regions are shown in Figure S1 and Table S1.

A critical concern in the above experiments is the possibility that histone sites subject to acetylation may be butyrylated at only low background levels, which would be misleadingly overestimated by the ChIP-seq analysis due to vastly different affinities of the antibodies used. To address this issue, we used surface plasmon resonance (SPR) to measure the affinities of antibodies for their respective targets. These measurements showed that all four antibodies have similar ranges of affinity (Figure S1A). Additionally, a ChIP-qPCR approach demonstrated that the four histone marks are significantly detected at selected genomic regions (Figures S1B and S1C). Together, these experiments confirmed that H4K5K8 butyrylation occurs at levels that largely exceed background noise.

Following these control experiments, we investigated in more detail the relationship between gene expression and the co-occurrence of the four TSS-associated histone marks. Remarkably, the most active genes were found associated with all four marks at their TSS regions (Figures 3B and 4A). In contrast, genes lacking any one of these marks at their TSS showed significantly reduced expression (Figure 4A). Other genomic elements did not show such a direct relationship between the co-existence of the four histone marks and gene expression (Figure S2).

Figure 4.

Figure 4

Open in a new tab

Tight Relationship between TSSs Labeled with H4K5ac, K5bu, K8ac, and K8bu Marks and Gene Transcriptional Activity

(A) The transcriptional activities of genes whose TSS regions bear the indicated histone H4 modifications are shown as box plots in spermatocytes (upper) and round spermatids (lower). The absence or presence of a histone mark is represented by the numbers “0” and “1”, respectively, and the four histone marks are arranged from bottom to top as follows: H4K5ac, H4K5bu, H4K8ac, and H4K8bu. The gene transcriptional activities as a function of various combinations of the four H4 PTMs on different genomic elements are shown in Figure S2.

(B) The heatmaps and profiles (upper) show the peak intensities of the ChIP signal at the TSS regions of genes associated with each of the four histone marks (“0” is input; “1” is K5ac; “2” is K5bu; “3” is K8ac; and “4” is K8bu), in spermatocytes (“Spc”), and in round spermatids (“RS”). The genes were assigned to four groups corresponding to: (1) genes with no ChIP peaks (“no”), (2) genes with moderate intensity of TSS labeling by the four marks and no significant change between spermatocytes and round spermatids (=), (3) genes with variation in TSS labeling intensities by the four marks with high meiotic peak intensity and a decreased intensity in post-meiotic cells (>), and (4) genes with an increased peak intensity in post-meiotic cells compared to spermatocytes (<). The box plots (lower) show the expression of the corresponding genes in meiotic and in round spermatids. These data were obtained using transcriptomic data of control samples from our previous work (Montellier et al., 2013; GSE46136).

(C) Total peak counts (acetyl and butyryl H4s) per million base pair for each chromosome in spermatocytes (blue bars) and round spermatids (red bars) are shown. The expected random distributions of the ChIP-seq peaks are indicated as dashed lines (upper). The right panel shows the proportion of gene TSSs harboring 0, 1–3, or 4 of the studied H4 PTMs on autosomes and on the X chromosome in round spermatids (color coded). Of the X-linked genes that escape inactivation, the majority has either no H4K5K8 acetyl/butyryl marks or harbors one to three of these PTMs. A detailed consideration of these PTMs indicates that these TSSs are nearly always depleted in H4K8ac. The list of X-linked genes that are activated in post-meiotic cells was established based on our previous detailed post-meiotic transcriptomic analysis (Boussouar et al., 2014).

To further analyze the relationship between the four acylation histone marks and gene activity, we took advantage of the differences in gene expression programs between spermatocytes (meiotic cells) and the transcriptionally active haploid round spermatids (generated after meiosis). Using our ChIP-seq data from these two cell types, we divided genes into four categories according to the intensity of TSS labeling by the four H4 acylation marks, namely: (1) genes bearing none of the four marks (labeled “no”) and (2–4) genes bearing all four marks, either with (2) comparable (“=”), (3) higher (“>”), or (4) lower (“<”) peak intensities in spermatocytes compared to round spermatids (Figure 4B). We observed that the four groups of genes belong to different gene expression programs. Genes in category (1) are largely unexpressed, while those in category (2) exhibited no change in expression level between meiotic and post-meiotic cells. In striking contrast, genes in categories (3) and (4) exhibited differential expression in the two cell types, which was positively associated with the change in intensity of TSS labeling by the four histone marks (Figure 4B).

Additional support for a positive correlation between the presence of the four PTMs in the TSSs and gene expression is the observation that all four marks are depleted on the sex chromosomes compared to autosomes, consistent with the chromosome-wide meiotic transcriptional inactivation known to characterize sex chromosomes (Figure 4C). To further investigate this observation, we specifically considered the fraction of sex chromosome-linked genes that escapes meiotic sex chromosome inactivation (Namekawa et al., 2006). While all four PTMs were identified on a majority (73%) of TSSs associated with active autosomal genes, co-occurrence of the four PTMs was observed in only a minority (23%) of TSSs associated with sex-linked genes that are reactivated in post-meiotic cells. Hence, over three-quarters of this latter category of TSSs bear between zero and three of the four histone marks. Interestingly, most of the TSSs bearing 1–3 of the PTMs were depleted of H4K8ac, but not of H4K8bu (Figure 4C). This is consistent with other studies showing depletion of acetylation on the TSSs of sex-linked post-meiotic genes and their labeling with other acyl groups such as crotonyl (Sin et al., 2012, Tan et al., 2011).

Histone Butyrylation Directly Stimulates Transcription

Our ChIP-seq data showed that, like acetylation, histone butyrylation is associated with high levels of gene expression, suggesting the possibility that histone butyrylation directly stimulates gene expression. To test this hypothesis, we exploited a reconstituted activator-dependent in vitro transcription system. Our early studies had shown that in vitro, p300 and CREB binding protein (CBP) can catalyze lysine butyrylation by transferring the butyryl group from [14C] butyryl-CoA to core histone proteins (Chen et al., 2007). This activity was also observed in ex vivo transfection experiments (Chen et al., 2007). Moreover, in vitro butyrylation activity was confirmed both on a reconstituted chromatin template and on histone octamers by mass spectrometry analysis. Further analysis of this in vitro activity revealed that p300 efficiently butyrylates the sites of interest, H4K5 and H4K8, in histone octamers as well as in chromatin (Figures 5A and 5B; Data S1).

Figure 5.

Figure 5

Open in a new tab

p300 Uses Butyryl-CoA to Butyrylate Histones in Histone Octamers and in Chromatin and to Directly Stimulate Gene Transcription

(A) In vitro reconstituted histone octamers or a chromatin template (used in the assay shown in C) were incubated with butyryl-CoA in the presence of purified p300 and butyrylated histones were detected by MS.

(B) In another assay, histone octamers were incubated with butyryl-CoA in the presence or absence of p300 to measure the degree of non-enzymatic histone butyrylation. The histograms represent the spectra counts of peptides containing the indicated butyrylated lysines in two different experiments (1 and 2: color coded). The annotated MS/MS spectra for histone Kbu peptides are presented in Data S1.

(C) Schematic representation of the p53- and p300-dependent in vitro transcription assay showing the order of the added components (An and Roeder, 2004).

(D) Autoradiography of RNA products obtained by in vitro transcription under the conditions shown.

(E) The same reactions as in (D) were performed with chromatin templates generated by the use of either wild-type histone H3 and H4 or the indicated H3 and H4 K to R mutants.

After confirming that p300 is a histone butyryltransferase, we used a p300- and p53-dependent in vitro transcription system (Figure 5C) to test if histone butyrylation could stimulate transcription. We observed that p300-catalyzed histone butyrylation indeed directly stimulates transcription (Figure 5D). The mutation of lysine residues to arginine either on H3 or H4 tails eliminated acyl-CoA-stimulated transcription, indicating that acetyl/butyryl-CoA activates transcription through p300-catalyzed histone lysine acylation (Figure 5E). This experiment clearly demonstrates that, like acetylation, histone butyrylation can also directly stimulate gene transcriptional activity.

Brdt Binds to Gene TSSs Harboring H4 K5/K8 Acetylation and Butyrylation

Association of the four histone acylation marks with the TSSs of most of the highly active genes in spermatogenic cells raises the possibility that the high transcriptional activity of these genes is mediated by the binding of Brdt to the acylation marks on their TSSs. To test this hypothesis, we compared previously identified Brdt-bound TSSs (Gaucher et al., 2012) with TSSs labeled with the four histone marks. This analysis showed that most of the Brdt-bound gene TSSs also bore high levels of acetylation and butyrylation marks at H4K5 and H4K8 (Figures 6A and 6B). To study if Brdt’s first bromodomain (BD1) mediates this interaction, we used spermatogenic cells from mice expressing a mutated form of Brdt lacking BD1 (Gaucher et al., 2012, Shang et al., 2007). The ChIP-seq analysis showed that, in both spermatocytes and round spermatids, the deletion of BD1 considerably weakens Brdt binding to TSSs bearing the four acylation marks (Figure 6A, Brdt ΔBD1). This result indicates that BD1 has a major role in targeting gene TSSs bearing H4K5K8 acetylation/butyrylation.

Figure 6.

Figure 6

Open in a new tab

Brdt Is Preferentially Recruited to Gene TSS Regions Enriched in H4K5ac, K5bu, K8ac, and K8bu Marks

(A) ChIP-seq data from chromatin immunoprecipitation of Brdt either from wild-type fractionated spermatogenic cells or the corresponding fractionated cells expressing a truncated Brdt lacking its first bromodomain (Brdt ΔBD1) were obtained and compared with ChIP-seq data from the four indicated histone H4 marks from wild-type spermatogenic cells. Seqminer software illustrates gene TSSs bound by the wild-type Brdt (Brdt wt), Brdt ΔBD1, and the occurrence of the four H4 marks on the same regions. The input corresponds to the sequencing of chromatin fragments before ChIP from wild-type cells. Exactly the same profile was obtained for the input chromatin fragments from Brdt ΔBD1 cells (data not shown).

(B) Gene TSS regions were divided into two categories, either bound or unbound by Brdt, in spermatocytes (upper) or round spermatids (lower). For each category, the proportion of genes whose TSS regions are enriched with none, 1, 2, 3, or 4 of the four histone PTMs are represented with the indicated colors.

(C) Mouse total testis extracts were prepared from wild-type mice (upper and lower left) or mice expressing the truncation mutant Brdt ΔBD1 (lower right) and incubated with the indicated peptides, Brdt was then visualized after pull-down using an anti-Brdt antibody (Gaucher et al., 2012). In some experiments, BET bromodomain inhibitor JQ1 was added to the extract prior to the peptide pull-down step, as indicated. In another experiment Brdt was identified by MS/MS (Figure S3B), and the annotated MS/MS spectra for Brdt peptides are presented in Data S1. The pull-down experiments were also performed on Brdt-expressing transfected cells and the data shown in Figure S3A.

(D and E) Modeling of Brdt-BD1 bound to H4 peptides bearing mixed butyryl and acetyl groups.

(D) Hypothetical model of an H4K5acK8bu peptide bound to Brdt-BD1. The model is based on the crystal structure of the Brdt/H4K5acK8ac complex and was built by substituting the K8 acetyl group with a butyryl group. The substitution allows all contacts between BD1 and the diacetylated peptide to be preserved, while the additional atoms of the butyryl group are accommodated within the space between residues Trp49 and Leu60.

(E) Comparison of the Brdt and Brd4 BD1 bromodomains bound to diacetylated and monobutyrylated peptides, respectively. The two structures were aligned via the pocket-defining BC and ZA loops. Brd4 is shown in green, and its associated water molecules and butyrylated peptide (for which only two residues are present in the crystal structure) are shown in magenta. Brdt is shown in light blue, and its water molecules and peptide ligand are shown in cyan. For clarity, most of the ZA loop is omitted. Brdt-BD1 residues 108 and 112–114 make direct and water-mediated hydrogen bonds to the peptide backbone atoms of H4 residues K5ac, G6, G7, and L10. The shift in peptide backbone position required to accommodate a butyryl group on K5 would disrupt some or all of these interactions. In the Brdt-BD1/H4K5acK8ac complex, hydrogen bonds mediated by water molecule w5 play a key structural role in linking the two acetylated lysine residues to each other and to the backbone carbonyl of BD1 residue Pro50. In the structure of Brd4-BD1/H3K14Bu, w5 is displaced by 1.1 Å toward Pro82, while the amide nitrogen of K14Bu is displaced in the opposite direction by 1.3 Å. Thus, compared to the diacetylated peptide, the presence of the butyryl group disrupts a water-mediated hydrogen bond between BD1 and the peptide.

To further characterize histone acylation at gene TSSs bound by Brdt, we considered the occurrence of the four studied H4 PTMs at TSSs as a function of Brdt-binding. Remarkably, the vast majority of Brdt-associated TSSs bore all four acylation marks (Figure 6B). These observations therefore led us to question the ability of Brdt to bind H4 bearing butyrylation at either K5 or K8 or both.

H4 K5 Butyrylation Inhibits Brdt Binding

To test the ability of Brdt’s bromodomains to bind a butyrylated H4 tail, we first carried out an in vitro binding assay using biotinylated H4 tail peptides bearing all four possible combinations of the K5ac, K5bu, K8ac, and K8bu acylation marks. We incubated the peptides with extracts from transfected Brdt-expressing Cos7 cells and performed pull-down experiments. As expected, Brdt efficiently interacts with H4 peptides that are either fully acetylated (i.e., on K5, K8, K12, and K16) or diacetylated on K5 and K8 (Figure S3A, upper). Strikingly, whereas Brdt binding was only slightly affected by the replacement of H4K8ac by H4K8bu, it was completely abolished by the substitution of H4K5ac by H4K5bu. Pull-down experiments using Brdt mutants bearing inactive bromodomains BD1 or BD2 demonstrated that Brdt binding to all the tested peptides depends on the integrity of its first bromodomain (Figure S3A, lower). These data clearly imply that the butyrylation of H4K5 inhibits the binding of Brdt to histone H4.

To validate this result, we performed the same experiment with nuclear extracts from mouse testis. Our results confirmed that butyrylation at H4K5 abolishes the binding of Brdt to H4 tails (Figure 6C, upper). As further confirmation, we repeated the pull-down assay on protein extracts from wild-type mouse testis using either fully acetylated or fully butyrylated immobilized H4 tail peptides and analyzed the bound fractions by mass spectrometry. Brdt was easily identified among the proteins affinity-isolated by the H4ac-containing peptide, but not by the H4bu-containing peptide (Figure S3B; Data S1). The use of extracts from mice testes expressing the Brdt ΔBD1 mutant confirmed that the Brdt-H4 tail interactions described above are primarily mediated by the BD1 domain (Figure 6C, lower right). This result was further corroborated by an experiment with JQ1, a BET bromodomain inhibitor, which abolished the binding of testis-derived Brdt to acetylated H4 and H4K5acK8bu peptides (Figure 6C, lower left). Taken together, these findings establish that H4K5bu, but not H4K8bu, abolishes the interaction between Brdt and the histone H4 tail.

Structural Analysis of the Effect of Butyrylation on the Brdt BD1-H4 Tail Interaction

To understand the molecular basis of the inhibitory effect of H4K5 butyrylation on Brdt binding, we carried out a structural modeling analysis. In the crystal structure of Brdt-BD1 bound to H4K5acK8ac, residue K5ac is intimately recognized by BD1, whereas K8ac makes fewer contacts (Morinière et al., 2009). Modeling shows that replacing the K8 acetyl group with a butyryl group allows the additional atoms to be accommodated without compromising any of the interactions between BD1 and the peptide (Figure 6D). In contrast, replacing the K5 acetyl group by a butyryl group results in a steric clash with residues in the domain’s ZA loop, implying that some structural adjustments to the model are required to accommodate the butyryl group. Comparison with the published co-crystal structure of Brd4-BD1 bound to H3K14bu indicates the type of adjustments required to accommodate H4K5bu (Vollmuth and Geyer, 2010). Aligning the latter structure with that of the Brdt-BD1/H4K5acK8ac complex shows that H3K14bu occupies approximately the same position as K5ac (Figure 6E). However, the bulkier butyryl group results in the displacement of a water molecule within the ligand-binding pocket and causes the K14bu main-chain and side-chain atoms to be shifted relative to those of K5ac. In the context of an H4K5buK8ac peptide, such shifts would be predicted to disrupt several direct and indirect hydrogen bonds between BD1 and the peptide (see figure legend for details). Thus, the modeling approach provides a plausible structural basis for the poor affinity observed for the binding of Brdt to the H4K5buK8ac peptide compared with that to the H4K5acK8ac and H4K5acK8bu peptides.

Dynamic Mixed Labeling of H4K5K8 by Acetylation and Butyrylation at Active Chromatin Sites

Our pull-down and structural modeling data strongly suggest that Brdt is inhibited from binding TSS regions where histone H4 is modified by lysine butyrylation at the K5 position. This conclusion appears contradictory to the ChIP-seq data, where H4K5bu and Brdt were both associated with the same TSS regions. A hypothesis that would reconcile these observations is that acetylation and butyrylation of H4K5 exhibit a rapid turnover. Indeed, acyltransferases associated with highly active gene TSSs might feasibly drain cellular acetyl-CoA as well as butyryl-CoA toward these sites, leading to a mixture of histone H4 acetylation and butyrylation marks. To test this hypothesis, we performed in vitro assays by incubating reconstituted histone octamers with purified p300 and an equimolar mixture of acetyl-CoA and butyryl-CoA. Histones were then analyzed by mass spectrometry to detect acetylated and/or butyrylated peptides. Our results show that p300 can use acyl-CoAs to catalyze acetylation and butyrylation at both H4K5 and H4K8 sites (Figure 7A), as we detected H4 peptides bearing various combinations of acetylated or butyrylated H4K5 and H4K8 (Figure 7A; Data S1). Encouraged by this in vitro result, we then investigated whether histone H4 isolated from spermatogenic cells also contains diverse lysine acylation marks. Mass spectrometry analysis of these samples detected H4 peptides with various combinations of acetylation and butyrylation at H4K5 and H4K8.

In addition, depending on cell type, the stoichiometry of H4K5bu and H4K8bu could be higher than some of the widely studied histone marks such as H3K4me3 (Kulej et al., 2015) (Figures 7B and S4A; Table S2), but lower than those of H4 tail acetylation, such as that of K16, which can be as high as 20% of H4 species (Kulej et al., 2015).

Acetylation is known to have a high turnover rate at gene TSSs (Crump et al., 2011). Therefore, due to continuous acylation by histone acetyltransferase (HATs) such as p300 and the rapid turnover of these PTMs, it is feasible that H4K5 and K8 alternate between acetylated and butyrylated states. This model is supported by the detection of various combinations of H4K5K8 acetylation and butyrylation either in in vitro HAT assays (Figure 7A) or in vivo in spermatogenic cells (Figure 7B). A direct consequence of such alternating histone acetylation/butyrylation would be the dynamic binding of Brdt, which would oscillate between high- and low-affinity states depending on the acylation status of H4K5 (Figure 7C). Such dynamic histone H4 acylation could be facilitated by open nucleosomes on the corresponding gene TSSs by specific histone variants such as H2A.Lap1, a histone H2A variant capable of inducing unstable and open nucleosomes and known to associate with active gene TSSs in spermatogenic cells (Nekrasov et al., 2013, Soboleva et al., 2012). Indeed, analysis of ChIP-seq data for H2A.Lap1 revealed that the four H4 PTMs are particularly enriched on H2A.Lap1-associated TSSs (Figure S4B).

Taken together, the above results support a model whereby alternating competing histone acetylation and butyrylation underlie a dynamic interaction between the histone modifications and the cognate bromodomain.

Discussion

The present study reports our findings regarding the interplay between histone acetylation and butyrylation on the histone H4 tail during sperm cell differentiation. We found that in vitro, p300 uses available acetyl-CoA and butyryl-CoA sources to acylate the H4 tail at all the acceptor lysines in an indiscriminate manner. A proteomic approach also revealed the co-existence of the same combinations of H4 K5K8 acetyl and butyryl marks in different spermatogenic cell types. ChIP-seq analysis on fractionated spermatogenic cells further demonstrated that H4K5/K8 acetyl/butyryl are particularly enriched at the TSSs of the most active genes. Strikingly, however, functional and structural analysis revealed that the acylation state of H4K5 is a critical determinant of Brdt binding affinity, as Brdt binds the acetylated, but not the butyrylated state. In this context, a stable maintenance of differential acetylation and butyrylation could have important functional consequences for the genome reorganization observed during spermatogenesis. We found that, in contrast to earlier stages, in elongating spermatids, H4K5 and H4K8 acetylation and butyrylation become more markedly associated with specific regions of the genome. This could have a direct consequence on the action of Brdt in these cells. Indeed, we previously showed that in elongating spermatids, when histone hyperacetylation and a general transcriptional shut-down occur, BD1 is indispensable for the replacement of acetylated histones by TPs (Gaucher et al., 2012). Additionally, the hyperacetylated histones are known to be directly targeted for degradation by a PA200-containing specialized proteasome (Qian et al., 2013), suggesting that the Brdt-bound histone population enters this pathway. Here, we show that in elongating spermatids, butyrylated histones survive this wave of acetylation-dependent histone removal and degradation. This survival is perfectly consistent with the inability of Brdt to recognize butyrylated H4. These observations suggest that a stable differential labeling of H4 by acetylation and butyrylation may control the timing of histone removal. In this context, it is also possible to speculate that nucleosomes bearing butyrylated H4 could undergo a direct histone-to-Prm replacement due to the inability of Brdt to mediate the exchange of histones by TPs. Indeed, TP1-TP2 double KO cells can undergo direct histone-to-Prm replacement but, in this case, Prms are unable to tightly compact the genome (for review, see Gaucher et al., 2010)). The prediction is therefore that regions bearing butyrylated histones would evolve to a less compact structure in mature sperms than regions that are marked by acetylated histones before their removal. This regulatory mechanism could be an elegant way to introduce differences in genome compaction by Prms.

Differential histone tail acylation might also play important roles in the control of somatic cell gene expression, differentiation, and genome programming. Indeed, a recent study of p300-mediated histone H3K18 crotonylation revealed that an increase in cellular levels of crotonyl-CoA favors a more efficient de novo gene activation (Sabari et al., 2015). Although the precise mechanism underlying the role of H3K18cr in transcription is unknown, the differential affinity of a bromodomain protein toward crotonyl versus acetyl marks could conceivably account for the observed phenomenon.

Previous reports demonstrated a rapid turnover of histone acetylation on a sub-fraction of nucleosomes (Clayton et al., 2006, Waterborg, 2002), overlapping with active gene TSSs (Crump et al., 2011). However, the functional implications of the short half-life of histone acetylation on active chromatin regions, especially on highly transcribed gene TSSs, have remained elusive. The results described in our present study are consistent with a rapid turnover of both acetyl and butyryl marks on the H4 tail. Given the differential affinity of Brdt BD1 for the acetylated and butyrylated forms of H4K5, a rapid alternation of acylation states at H4K5 would result in a highly dynamic interaction between Brdt and chromatin. Thus, it is tempting to speculate that the generally observed rapid turnover of acetyl marks may be functionally significant because it enables rapid transitions between alternative states of lysine acylation. Such a mechanism, while maintaining histones permanently modified, would allow for a dynamic association with specific bromodomains, which might be important for sustaining successive cycles of transcription. Thus, either a change in the ratio of cellular acyl-CoAs (for instance caused by metabolic disorders; Pougovkina et al., 2014) or differential activities of histone deacetylase (HDACs) in removing acyl-groups (Rousseaux and Khochbin, 2015) could reprogram gene expression profiles.

In conclusion, we provide here the first demonstration that the interchangeable use of two closely related histone acylation marks at a specific site, H4K5, has important functional consequences by modulating the ability of a transcriptional regulator, Brdt, to recognize chromatin. This finding should improve our understanding of gene transcriptional regulation and its link to cell metabolism.

Experimental Procedures

Antibodies and Other Reagents

Pan anti-Kbu antibody, anti-histone site-specific Kbu, and acetyllysine antibodies used were purchased from PTM Biolabs and anti-histone H3 and anti-histone H4 antibodies were from Abcam. Mouse monoclonal antibodies against H4K5ac and H4K8ac were a generous gift from Dr. H. Kimura. Anti-Brdt is a homemade antibody previously described (Gaucher et al., 2012). Butyryl-CoA and acetyl-CoA were purchased from Sigma-Aldrich. The modified porcine trypsin was purchased from Promega. HPLC-grade acetonitrile, water, and ethanol were purchased from EMD Chemicals. Peptides bearing one or a few acetyl and Kbu residues were custom synthesized and were verified by HPLC and mass spectrometry. JQ1 was synthesized by Charles McKenna and Elena Ferri (University of Southern California) (Emadali et al., 2013) and used as described in this reference.

The antigen recognition capacities of the antibodies used in ChIP-seq and ChIP-qPCR were determined using SPR as described in the Supplemental Information.

Identification of Kbu Sites in Core Histones by Affinity Enrichment and Mass Spectrometry

Acetylation and butyrylation of histones were determined on 200 μg of core histones extracted from the different studied cell types. The detailed procedures are described in the Supplemental Information.

Quantification of H4K5K8 Butyrylation in Spermatogenic Cell Populations

Histone from spermatogenic cells underwent chemical butyrylation with deuterated (D5) butyryl anhydride and processed as described in the Supplemental Information.

Analysis of Mouse Spermatogenic Cells, ChIP, Bioinformatics, and In Vitro Transcription

All the experimental procedures, including immunostaining, cell fractionation, and ChIP-seq were carried out exactly as described previously (Dai et al., 2014, Gaucher et al., 2012, Tan et al., 2011). ChIP-qPCR was performed following our ChIP-seq protocol, but the recovered DNA was amplified using quantitative PCR and specific primers corresponding to regions indicated in Figure S1. Table S1 indicates the sequence of these primers and details of the ChIP experiments are described in Supplemental Information.

Brdt pull-down assays were performed as previously described (Huang et al., 2010, Pivot-Pajot et al., 2003). The in vitro transcription assay was also previously described (An et al., 2002, An and Roeder, 2004, Sabari et al., 2015, Tang et al., 2013). The bioinformatic analyses followed a pipeline similar to the one previously described (Dai et al., 2014, Gaucher et al., 2012, Montellier et al., 2013, Tan et al., 2011) and are detailed in the Supplemental Information.

p300 Expression and Purification for In Vitro Acetyl/Butyryl-Transferase Assays on Histone Octamer

In vitro acetyl/butyryl-transferase assays on histone octamer used recombinant p300 expressed and purified from baculovirus infected in Sf21 insect cells. Protein purification and the details of the HAT assays are described in the Supplemental Information.

Author Contributions

Y.Z. and S.K. designed and coordinated the whole project. A.G. performed most of the experiments on spermatogenic cells. S.B., J.G., and E.M. were involved in the ChIP setup and S.B. helped with the final experiments. A.-L.V. designed appropriate primers and performed qPCR after ChIP from total germ cells. J.G. and D.P. performed in vitro HAT assays. Z.T. and R.G.R. performed in vitro transcription assays. S.C. helped with pull-down assays. T.B. helped with spermatogenic cell fractionation. C.P. helped with structural analyses. A.D., G.C., and D.P. helped with the bioinformatic analyses under the supervision of S.R. D.P. D.Z. performed the IP/pull-down/MS/MS experiments, PTM quantifications, antibody characterization (SPR), and coordinated the MS analyses. H.H. and O.K. performed the HPLC-MS/MS experiments and data analysis. S.Q. constructed the ChIP-seq library for sequencing. T.H. and Z.C. were involved in antibody production and characterization.

Acknowledgments

The S.K. laboratory is supported by a grant from the Fondation pour la Recherche Medicale (FRM) “analyse bio-informatique pour la recherche en biologie” program as well as the ANR Episperm3 program (ANR-15-CE12-0005-02 to S. K. and D. P.) and by INCa (2013-082 program to S.K. and C. P.). D. P. laboratory is also supported by the World Wide Cancer Research foundation (16-0280). A.G. has been a recipient of a Marie Curie Initial Training Network funded by the European Commission (FP7-PEOPLE-2011-ITN, PITN-GA-289880) for three years and is now supported by “Fondation ARC.” G.C. is supported by the above-mentioned FRM grant. The Y.Z. laboratory is supported by the NIH (GM105933, DK107868, and GM115961). Y.Z. is a shareholder and a member of the scientific advisory board of PTM BioLabs, Co., Ltd. T.H. and Z.C. are PTM BioLabs employees. Work in the laboratory of R.G.R. was supported by grants from the NIH and the Starr Foundation Tri-Institutional Stem Cell Initiative (2014-034). We gratefully acknowledge the generous gift of mouse monoclonal antibodies against H4K5ac and H4K8ac by Dr. H. Kimura. JQ1 was synthesized and provided to us by Charles McKenna and Elena Ferri. We also acknowledge Miho Shimada for the preparation of histone mutants.

Published: April 21, 2016

Footnotes

Supplemental Information includes Supplemental Experimental Procedures, four figures, two tables, and one data file and can be found with this article online at http://dx.doi.org/10.1016/j.molcel.2016.03.014.

Contributor Information

Yingming Zhao, Email: yzhao2@bsd.uchicago.edu.

Saadi Khochbin, Email: saadi.khochbin@univ-grenoble-alpes.fr.

Accession Numbers

The accession numbers for the H4 K5 and K8 acetylation and butyrylation ChIP-seq data and Brdt ChIP-seq data reported in this paper are GEO: GSE77277 and GSE39910, respectively.

Supplemental Information

Document S1. Supplemental Experimental Procedures, Figures S1–S4, and Tables S1 and S2
mmc1.pdf (1.7MB, pdf)
Data S1. Encompasses All the Annotated Tandem Mass Spectrometry, MS/MS, Spectra, Related to Figures 1B, 5A, 5B, S3B, and 7A

The annotated MS/MS spectra of the histone Kbu sites from different species related to Figure 1B (indicated in each spectrum); the annotated MS/MS spectra for the histone Kbu sites from the experiments shown in Figures 5A and 5B (indicated in each spectrum); the annotated MS/MS spectra for the Brdt protein after the testis-extract pull-down shown in Figure S3B (indicated in each spectrum); the annotated MS/MS spectra for the H4 with the multiple acetylation/butyrylation shown in Figure 7A (indicated in each spectrum).

mmc2.zip (4MB, zip)
Document S2. Article plus Supplemental Information
mmc3.pdf (5.2MB, pdf)

References

  1. An W., Roeder R.G. Reconstitution and transcriptional analysis of chromatin in vitro. Methods Enzymol. 2004;377:460–474. doi: 10.1016/S0076-6879(03)77030-X. [DOI] [PubMed] [Google Scholar]
  2. An W., Palhan V.B., Karymov M.A., Leuba S.H., Roeder R.G. Selective requirements for histone H3 and H4 N termini in p300-dependent transcriptional activation from chromatin. Mol. Cell. 2002;9:811–821. doi: 10.1016/s1097-2765(02)00497-5. [DOI] [PubMed] [Google Scholar]
  3. Boussouar F., Goudarzi A., Buchou T., Shiota H., Barral S., Debernardi A., Guardiola P., Brindle P., Martinez G., Arnoult C. A specific CBP/p300-dependent gene expression programme drives the metabolic remodelling in late stages of spermatogenesis. Andrology. 2014;2:351–359. doi: 10.1111/j.2047-2927.2014.00184.x. [DOI] [PubMed] [Google Scholar]
  4. Chen Y., Sprung R., Tang Y., Ball H., Sangras B., Kim S.C., Falck J.R., Peng J., Gu W., Zhao Y. Lysine propionylation and butyrylation are novel post-translational modifications in histones. Mol. Cell. Proteomics. 2007;6:812–819. doi: 10.1074/mcp.M700021-MCP200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Clayton A.L., Hazzalin C.A., Mahadevan L.C. Enhanced histone acetylation and transcription: a dynamic perspective. Mol. Cell. 2006;23:289–296. doi: 10.1016/j.molcel.2006.06.017. [DOI] [PubMed] [Google Scholar]
  6. Crump N.T., Hazzalin C.A., Bowers E.M., Alani R.M., Cole P.A., Mahadevan L.C. Dynamic acetylation of all lysine-4 trimethylated histone H3 is evolutionarily conserved and mediated by p300/CBP. Proc. Natl. Acad. Sci. USA. 2011;108:7814–7819. doi: 10.1073/pnas.1100099108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Dai L., Peng C., Montellier E., Lu Z., Chen Y., Ishii H., Debernardi A., Buchou T., Rousseaux S., Jin F. Lysine 2-hydroxyisobutyrylation is a widely distributed active histone mark. Nat. Chem. Biol. 2014;10:365–370. doi: 10.1038/nchembio.1497. [DOI] [PubMed] [Google Scholar]
  8. Emadali A., Rousseaux S., Bruder-Costa J., Rome C., Duley S., Hamaidia S., Betton P., Debernardi A., Leroux D., Bernay B. Identification of a novel BET bromodomain inhibitor-sensitive, gene regulatory circuit that controls Rituximab response and tumour growth in aggressive lymphoid cancers. EMBO Mol. Med. 2013;5:1180–1195. doi: 10.1002/emmm.201202034. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Flynn E.M., Huang O.W., Poy F., Oppikofer M., Bellon S.F., Tang Y., Cochran A.G. A subset of human bromodomains recognizes butyryllysine and crotonyllysine histone peptide modifications. Structure. 2015;23:1801–1814. doi: 10.1016/j.str.2015.08.004. [DOI] [PubMed] [Google Scholar]
  10. Gaucher J., Reynoird N., Montellier E., Boussouar F., Rousseaux S., Khochbin S. From meiosis to postmeiotic events: the secrets of histone disappearance. FEBS J. 2010;277:599–604. doi: 10.1111/j.1742-4658.2009.07504.x. [DOI] [PubMed] [Google Scholar]
  11. Gaucher J., Boussouar F., Montellier E., Curtet S., Buchou T., Bertrand S., Hery P., Jounier S., Depaux A., Vitte A.L. Bromodomain-dependent stage-specific male genome programming by Brdt. EMBO J. 2012;31:3809–3820. doi: 10.1038/emboj.2012.233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Goudarzi A., Shiota H., Rousseaux S., Khochbin S. Genome-scale acetylation-dependent histone eviction during spermatogenesis. J. Mol. Biol. 2014;426:3342–3349. doi: 10.1016/j.jmb.2014.02.023. [DOI] [PubMed] [Google Scholar]
  13. Govin J., Caron C., Lestrat C., Rousseaux S., Khochbin S. The role of histones in chromatin remodelling during mammalian spermiogenesis. Eur. J. Biochem. 2004;271:3459–3469. doi: 10.1111/j.1432-1033.2004.04266.x. [DOI] [PubMed] [Google Scholar]
  14. Hazzouri M., Pivot-Pajot C., Faure A.K., Usson Y., Pelletier R., Sèle B., Khochbin S., Rousseaux S. Regulated hyperacetylation of core histones during mouse spermatogenesis: involvement of histone deacetylases. Eur. J. Cell Biol. 2000;79:950–960. doi: 10.1078/0171-9335-00123. [DOI] [PubMed] [Google Scholar]
  15. Huang R., Holbert M.A., Tarrant M.K., Curtet S., Colquhoun D.R., Dancy B.M., Dancy B.C., Hwang Y., Tang Y., Meeth K. Site-specific introduction of an acetyl-lysine mimic into peptides and proteins by cysteine alkylation. J. Am. Chem. Soc. 2010;132:9986–9987. doi: 10.1021/ja103954u. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kim S.C., Sprung R., Chen Y., Xu Y., Ball H., Pei J., Cheng T., Kho Y., Xiao H., Xiao L. Substrate and functional diversity of lysine acetylation revealed by a proteomics survey. Mol. Cell. 2006;23:607–618. doi: 10.1016/j.molcel.2006.06.026. [DOI] [PubMed] [Google Scholar]
  17. Kulej K., Avgousti D.C., Weitzman M.D., Garcia B.A. Characterization of histone post-translational modifications during virus infection using mass spectrometry-based proteomics. Methods. 2015;90:8–20. doi: 10.1016/j.ymeth.2015.06.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Montellier E., Rousseaux S., Zhao Y., Khochbin S. Histone crotonylation specifically marks the haploid male germ cell gene expression program: post-meiotic male-specific gene expression. BioEssays. 2012;34:187–193. doi: 10.1002/bies.201100141. [DOI] [PubMed] [Google Scholar]
  19. Montellier E., Boussouar F., Rousseaux S., Zhang K., Buchou T., Fenaille F., Shiota H., Debernardi A., Héry P., Curtet S. Chromatin-to-nucleoprotamine transition is controlled by the histone H2B variant TH2B. Genes Dev. 2013;27:1680–1692. doi: 10.1101/gad.220095.113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Morinière J., Rousseaux S., Steuerwald U., Soler-López M., Curtet S., Vitte A.L., Govin J., Gaucher J., Sadoul K., Hart D.J. Cooperative binding of two acetylation marks on a histone tail by a single bromodomain. Nature. 2009;461:664–668. doi: 10.1038/nature08397. [DOI] [PubMed] [Google Scholar]
  21. Namekawa S.H., Park P.J., Zhang L.F., Shima J.E., McCarrey J.R., Griswold M.D., Lee J.T. Postmeiotic sex chromatin in the male germline of mice. Curr. Biol. 2006;16:660–667. doi: 10.1016/j.cub.2006.01.066. [DOI] [PubMed] [Google Scholar]
  22. Nekrasov M., Soboleva T.A., Jack C., Tremethick D.J. Histone variant selectivity at the transcription start site: H2A.Z or H2A.Lap1. Nucleus. 2013;4:431–438. doi: 10.4161/nucl.26862. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Pivot-Pajot C., Caron C., Govin J., Vion A., Rousseaux S., Khochbin S. Acetylation-dependent chromatin reorganization by BRDT, a testis-specific bromodomain-containing protein. Mol. Cell. Biol. 2003;23:5354–5365. doi: 10.1128/MCB.23.15.5354-5365.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Pougovkina O., Te Brinke H., Wanders R.J., Houten S.M., de Boer V.C. Aberrant protein acylation is a common observation in inborn errors of acyl-CoA metabolism. J. Inherit. Metab. Dis. 2014;37:709–714. doi: 10.1007/s10545-014-9684-9. [DOI] [PubMed] [Google Scholar]
  25. Qian M.X., Pang Y., Liu C.H., Haratake K., Du B.Y., Ji D.Y., Wang G.F., Zhu Q.Q., Song W., Yu Y. Acetylation-mediated proteasomal degradation of core histones during DNA repair and spermatogenesis. Cell. 2013;153:1012–1024. doi: 10.1016/j.cell.2013.04.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Rousseaux S., Khochbin S. Histone acylation beyond acetylation: Terra incognita in chromatin biology. Cell J. 2015;17:1–6. doi: 10.22074/cellj.2015.506. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Sabari B.R., Tang Z., Huang H., Yong-Gonzalez V., Molina H., Kong H.E., Dai L., Shimada M., Cross J.R., Zhao Y. Intracellular crotonyl-CoA stimulates transcription through p300-catalyzed histone crotonylation. Mol. Cell. 2015;58:203–215. doi: 10.1016/j.molcel.2015.02.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Shang E., Nickerson H.D., Wen D., Wang X., Wolgemuth D.J. The first bromodomain of Brdt, a testis-specific member of the BET sub-family of double-bromodomain-containing proteins, is essential for male germ cell differentiation. Development. 2007;134:3507–3515. doi: 10.1242/dev.004481. [DOI] [PubMed] [Google Scholar]
  29. Sin H.S., Barski A., Zhang F., Kartashov A.V., Nussenzweig A., Chen J., Andreassen P.R., Namekawa S.H. RNF8 regulates active epigenetic modifications and escape gene activation from inactive sex chromosomes in post-meiotic spermatids. Genes Dev. 2012;26:2737–2748. doi: 10.1101/gad.202713.112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Soboleva T.A., Nekrasov M., Pahwa A., Williams R., Huttley G.A., Tremethick D.J. A unique H2A histone variant occupies the transcriptional start site of active genes. Nat. Struct. Mol. Biol. 2012;19:25–30. doi: 10.1038/nsmb.2161. [DOI] [PubMed] [Google Scholar]
  31. Tan M., Luo H., Lee S., Jin F., Yang J.S., Montellier E., Buchou T., Cheng Z., Rousseaux S., Rajagopal N. Identification of 67 histone marks and histone lysine crotonylation as a new type of histone modification. Cell. 2011;146:1016–1028. doi: 10.1016/j.cell.2011.08.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Tan M., Peng C., Anderson K.A., Chhoy P., Xie Z., Dai L., Park J., Chen Y., Huang H., Zhang Y. Lysine glutarylation is a protein posttranslational modification regulated by SIRT5. Cell Metab. 2014;19:605–617. doi: 10.1016/j.cmet.2014.03.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Tang Z., Chen W.Y., Shimada M., Nguyen U.T., Kim J., Sun X.J., Sengoku T., McGinty R.K., Fernandez J.P., Muir T.W., Roeder R.G. SET1 and p300 act synergistically, through coupled histone modifications, in transcriptional activation by p53. Cell. 2013;154:297–310. doi: 10.1016/j.cell.2013.06.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Vollmuth F., Geyer M. Interaction of propionylated and butyrylated histone H3 lysine marks with Brd4 bromodomains. Angew. Chem. Int. Ed. Engl. 2010;49:6768–6772. doi: 10.1002/anie.201002724. [DOI] [PubMed] [Google Scholar]
  35. Waterborg J.H. Dynamics of histone acetylation in vivo. A function for acetylation turnover? Biochem. Cell Biol. 2002;80:363–378. doi: 10.1139/o02-080. [DOI] [PubMed] [Google Scholar]
  36. Xie Z., Dai J., Dai L., Tan M., Cheng Z., Wu Y., Boeke J.D., Zhao Y. Lysine succinylation and lysine malonylation in histones. Mol. Cell. Proteomics. 2012;11:100–107. doi: 10.1074/mcp.M111.015875. [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

Document S1. Supplemental Experimental Procedures, Figures S1–S4, and Tables S1 and S2
mmc1.pdf (1.7MB, pdf)
Data S1. Encompasses All the Annotated Tandem Mass Spectrometry, MS/MS, Spectra, Related to Figures 1B, 5A, 5B, S3B, and 7A

The annotated MS/MS spectra of the histone Kbu sites from different species related to Figure 1B (indicated in each spectrum); the annotated MS/MS spectra for the histone Kbu sites from the experiments shown in Figures 5A and 5B (indicated in each spectrum); the annotated MS/MS spectra for the Brdt protein after the testis-extract pull-down shown in Figure S3B (indicated in each spectrum); the annotated MS/MS spectra for the H4 with the multiple acetylation/butyrylation shown in Figure 7A (indicated in each spectrum).

mmc2.zip (4MB, zip)
Document S2. Article plus Supplemental Information
mmc3.pdf (5.2MB, pdf)

RESOURCES