Multi-faceted quantitative proteomics analysis of histone H2B isoforms and their modifications
© Molden et al.; licensee BioMed Central. 2015
Received: 19 November 2014
Accepted: 25 March 2015
Published: 22 April 2015
Histone isoforms and their post-translational modifications (PTMs) play an important role in the control of many chromatin-related processes including transcription and DNA damage. Variants of histones H2A and H3 have been studied in depth and have been found to have distinct functions. Although 13 somatic histone H2B isoforms have been identified by various biochemical and mass spectrometric (MS) approaches, the distinct roles of these isoforms within human cells are as yet unknown. Here, we have developed quantitative MS techniques to characterize isoform-specific H2B expression across the cell cycle, in differentiated myogenic cells, and in different cancer cell lines to illuminate potential functional roles.
Using the MS strategies that we developed, we identified differences in H2B isoform levels between different cancer cell types, suggesting cancer or tissue-specific H2B isoform regulation. In particular, we found large variations in the levels of isoforms H2B1B and H2B1M across the panel of cell lines. We also found that, while individual H2B isoforms do not differ in their acetylation levels, trends in the acetylation on all H2B isoforms correlated with acetylation on other histone family members in the cancer cell line panel. We also used the MS strategies to study H2B protein expression across the cell cycle and determined that H2B isoforms that are alternatively spliced to carry a polyadenylation signal rather than the standard histone downstream element are expressed independently of the cell cycle. However, the level of protein produced from the polyadenylated transcripts does not contribute significantly to the total pool of H2B isoforms translated across the cell cycle or in non-cycling myogenic cells.
Our results show that H2B isoforms are expressed at varying levels in different cells, suggesting isoform-specific, and possibly cell-type-specific, H2B gene regulation. The bottom-up mass spectrometry technique we developed for H2B quantification is compatible with the current standard histone H3 and H4 bottom-up ‘one-pot’ analysis platform so that H2B isoforms and their modifications can be studied in future experiments at the same time as histone H3 and H4 modifications. Therefore, we have expanded the histone landscape that can be interrogated in future experiments.
Histones and 147 base pairs of DNA are together the building blocks of nucleosomes, which form the basic repeating units of chromatin. Chromatin is further folded and packaged into higher order structures with increasing degrees of compaction . How tightly chromatin is packaged influences how accessible it is to DNA binding proteins, including transcription factors and DNA repair proteins . Histones play important roles in regulating chromatin structure and function through dynamic post-translational modifications (PTMs) and specialized histones called histone variants.
There are five types of histone proteins: H2A, H2B, H3, H4, and H1, and with the exception of H4, all have multiple isoforms . These isoforms are proteins encoded by separate genes that have the same core histone structure with some sequence variation . Variants differ from canonical histones in their protein-protein interactions, localization in chromatin, number and types of PTMs, nucleosome stability, and tissue-specific expression profiles . All of these differences lead to unique functions for histone variants compared to canonical histones. Histone variants are expressed in a DNA replication-independent fashion, are maintained at steady levels throughout the cell cycle, and are incorporated into nucleosomes outside of S-phase [3,6]. Some new histone variants have been discovered recently, but the most well-characterized core histone variants include H3.3, CENP-A, H2A.Z, H2A.X, and macroH2A . Loss of function of any of these histone variants leads to acute consequences in specific cellular and developmental processes, such as fertility, proper centromere formation, genome integrity, and the DNA damage response . Knockouts of variants H2A.Z and H3.3 are lethal in mice, and H2A.Z-knockdown embryonic stem cells have a decreased efficiency in self-renewal and differentiation [7-9].
Somatic human H2B isoforms are all encoded by genes located in histone cluster 1 on chromosome 6 (6p21-6p22), and in cluster 2 (1q21) and cluster 3 (1q42) on chromosome 1, and produce mRNA with characteristic 3′ stem-loop sequences and histone downstream elements . H2B genes in the gene clusters are arranged in pairs with divergent H2A sequences that have a shared promoter region . The histone genes inside of these clusters are all expressed at high levels during S-phase due to stabilization of transcripts with 3′ stem-loops . Although expression of genes within the histone clusters is coordinated during the cell cycle, individual H2B (and H2A) genes are regulated by different promoter sequences and are expressed at different levels [15,16]. Somatic human H2B isoforms have very little sequence variation - they differ by as few as two amino acids and up to five amino acids (Figure 1). Most of the changes in sequence are small, but a few could affect protein function. For instance, the proline-to-leucine conversion in H2B1L would change the structure of the H2B N-terminal tail. In addition, there are many serine-and-threonine-to-alanine conversions in H2B isoform sequences, including at S36 in H2B3B, a site that is important in cell stress response regulation . This serine-to-alanine difference occurs at the DNA-histone contact and could therefore potentially change nucleosome structure. In addition, differences in amino acid sequence could also be involved in isoform-specific protein-protein interactions. Small differences in known histone variants are responsible for variant-specific chaperone and chromatin remodeling enzyme recognition [19,20].
There is evidence for somatic H2B sequences having unique functions in other organisms. H2B variants in Trypanasoma brucei and in Plasmodium falciparum form dimers specifically with the H2A variant H2AZ and localize to active genes [21,22]. Most recently, it was discovered that the H2B variant H2BE is involved in controlling olfactory gene expression in mice, suggesting that some somatic H2B variants may have tissue-specific roles . This H2B variant has an analogous sequence in humans (H2B1L). In fact, H2B sequence variation is conserved among higher eukaryotes , and some of the variants characterized in other organisms may have counterparts with similar functions in human cells.
The sequence similarity between histone H2B isoforms makes it challenging to differentiate between isoforms at the protein level. Mass spectrometry (MS) methods for identifying histone proteins include top-down MS of intact proteins, middle-down MS of large peptides, and bottom-up MS of smaller peptides. Two previous investigations, by Siuti et al. and Bonenfant et al., used top-down and bottom-up MS to identify H2B isoforms in human cell lines [10,11]. Siuti et al. used top-down to identify 8/13 H2B isoforms; they also used l-methionine-methyl-13C-methyl-d 3 to label potential methylation sites and histone deacetylase inhibitors to increase levels of H2B acetylation. They found no evidence of H2B methylation but did identify two acetylation sites on the most abundant H2B isoform . Bonenfant et al. used a variety of protease digest methods, including Glu-C digests, to identify 11 histone H2B isoforms and 3 acetylation sites . These studies confirmed that many H2B isoforms are present at the protein level for the first time but failed to detect some lower abundance H2B isoforms.
To date, methods to comprehensively, consistently, and quantitatively analyze histone H2B isoforms and their post-translational modifications have not been thoroughly compared and established. Here, we describe top-down and bottom-up mass spectrometry methods we developed to identify all 13 known H2B isoforms. The bottom-up method, unlike previous strategies, uses a well-established chemical derivatization method to create H2B N-terminal peptides of a similar and consistent length that can be resolved reproducibly using reversed-phase chromatography. As a result, this method can be used to quantify the 9 out of 13 somatic H2B isoforms that have unique N-terminal sequences and their modifications.
Since there is little research on potential biological differences between somatic H2B isoforms, we used the mass spectrometry methods developed to profile H2B isoform levels from mammalian cells. These top-down and bottom-up MS methods were utilized to profile the H2B isoform composition of different cancer cells. If H2B isoforms are regulated differently across cell types, then it is possible that some isoforms have unique cell-type-specific functions. We also used these methods to determine whether some H2B isoforms are expressed independently of DNA replication and to determine whether there are differences in the rates of synthesis of H2B variants between proliferating and terminally differentiated cells.
Top-down MS analysis of H2B
There are six main peaks in the intact mass spectrum of histone H2B (Figure 2A). The 5 most abundant masses match the predicted masses for 11 out of 13 known H2B isoforms (Figure 2A), as well as the monoacetylated forms of 3 H2B isoforms (H2B2F, H2B1D, and H2B1C) and the di-acetylated form of 2 isoforms (H2B1C and H2B1D). The low-mass peaks next to the main H2B peaks are artifacts caused by signal processing in the mass spectrometer. All of the H2B isoforms expressed in human somatic cells are the same length and have very few points of sequence variation. Consequently, many of them have the same intact mass or masses that are similar enough that they cannot be resolved based on the full-MS spectra alone (Additional file 1: Table S1). We identified these species at the MS/MS level using distinguishing c and z ions in their ETD fragmentation spectra. For example, histone H2B1K and histone H2B1H are proteins with masses that differ by only 2 Da and as a result have almost completely overlapping isotopic peak clusters. They differ in sequence by one amino acid at position 2 and at position 125 (Figure 1A). This sequence difference leads to unique c and z ion series for the two H2B isoforms (Figure 2B) in their combined fragmentation spectra. We did not detect two H2B isoforms (H2B3B, H2B1L) using top-down MS, despite prior evidence for these forms from real-time (RT)-PCR experiments . These isoforms have similar masses to the most abundant H2B isoforms, so it is possible that their signal was masked by the more abundant species. We estimated abundances of the isoforms and modified forms based on the protein ion relative ratios (PIRRs) using the five most abundant isotopic peaks for each isotopic cluster (Additional file 1: Table S1) as described previously .
Bottom-up MS analysis of H2B
Top-down MS is a powerful technique that can distinguish between the majority of H2B isoforms and their modifications; however, it is less sensitive and arguably less quantitative than other MS strategies. We developed bottom-up MS techniques to quantify and to identify H2B isoforms that were not abundant enough to be detected by top-down MS. As with other histone proteins, histone H2B is very lysine and arginine rich, making it difficult to digest with standard proteases and also making its digested fragments very hydrophilic and therefore not as compatible with standard reversed-phase C18-based chromatography as frequently utilized in proteomics experiments.
We analyzed H2B by derivatizing it first with propionic anhydride to cap primary amines on lysine residues with propionyl groups. This is a standard chemical derivatization technique that our lab uses to process histones , but we had not applied it yet to study H2B isoforms and their modifications. Capping the lysines prevents their digestion with trypsin so that the sequence is only digested after arginine in an Arg-C-like manner but with much more efficiency and specificity than using Arg-C itself. Propionylation also increases the hydrophobicity of histone peptides and thereby improves their retention and resolution on reversed-phase columns. Digesting after arginine creates a 29-amino-acid-long H2B N-terminal peptide encompassing the entire H2B N-terminal tail, the part of the H2B sequence with the most sequence diversity and modifications. It also creates a peptide encompassing the H2B C-terminal tail - the site of an important ubiquitination PTM .
CAD spectra were optimal for analyzing the N-terminal peptide sequence (Figure 3C). Fragmentation of both the +4 and +3 charge-state species created complete or nearly complete y-ion and b-ion series. There was a particularly strong y + 2 ion series for this peptide, with the most abundant fragment ions occurring on the N-terminal side of prolines. This proline-directed fragmentation was advantageous because it captures the sequence diversity of the H2B tail. This set of ions can be used to easily identify and distinguish between H2B isoforms. For example, the monoacetylated H2B1D N-terminal and the H2B1N N-terminal tail have different sequences but the same mass. The peaks can be distinguished from each other based on the characteristic proline-directed fragments in their MS/MS spectra (Additional file 2: Figure S1A and B).
After analyzing propionylated H2B samples, we identified H2B(1–29) N-terminal tail sequences representing seven unique H2B isoforms and two tail sequences that were shared between the remaining six H2B isoforms. In addition, each N-terminal tail sequence had up to a total of four lysines modified with acetyl groups (K5, K12, K15, K20). The most commonly acetylated residues, as determined by their unique fragment ion intensities, were lysine 12 and 15, which is in agreement with previous research which quantified H2B acetylation by Western blot . We also identified a potential new N-terminal H2B ubiquitination site at lysine 5 (Additional file 2: Figure S1), in addition to the more commonly studied C-terminal H2B(K123) ubiquitination site .
Profiling H2B isoform composition in commonly used cancer cell lines
Previous studies have demonstrated that there is variation in the expression of the mRNA of different H2B isoforms across cancer cell lines [12,31]. This suggests that there may be different abundances of H2B isoform proteins between cell lines and possibly cell-type-specific functions for the different H2B isoforms. Also, some H2B transcripts are frequently overexpressed in certain cancer types, so we would therefore expect to see some variations in their protein abundances as well [32,33]. We used top-down MS and bottom-up MS to confirm differences in H2B isoform protein levels across different cell lines, including a catalog of cancer cell lines that represent many different cell types and cancer types.
Top-down MS was used to confirm that there are differences in the levels of H2B isoforms across cancer cell lines (Additional file 4: Figure S3). Relative to the most abundant H2B1C/1 J/1O peak, we found the most variation in the H2B1H/H2B1K, H2B2E/2 F, and H2B1M levels. However, the top-down MS results were not as quantitative as the bottom-up results, and we were unable to deconvolute isobaric species contributions to the differences in abundances that we observed. So once we confirmed differences in H2B levels at the protein level using top-down MS, we also profiled H2B content in cancer cell lines using the bottom-up MS technique described in the previous section.
Similar to the top-down cancer cell line data, we observed differences in H2B isoform levels across the different cell lines (Figure 5). Cell lines from the same cell type clustered together. The only exceptions were the prostate cell lines PC3 and DU145 and the skin cell lines HaCAT (keratinocyte) and HFF (fibroblast), which had very different H2B isoform profiles. The most striking difference between cancer cell lines was in the levels of isoforms H2B1B and H2B1M. HFF and hES2 cells had higher levels of both isoforms compared to most of the cancer cell lines. In addition, one third of the cell lines profiled had significantly lower levels of H2B1B than the rest of the cell lines. Other H2B isoforms that varied in abundance between cell types, but did not show as striking differences as H2B1B and H2B1M, are H2B2F and H2B1L.
Most cell lines had very similar levels of H2B acetylation (approximately 11% of the total signal on average). The exceptions were hES2, Jurkat, and C33a cells, which had higher levels of acetylation (23%, 18%, and 22%, respectively, of the total signal), and SW480 cells, which had lower (6%) levels of acetylation. This result is consistent with a study from our lab that used the same cancer cell line data to analyze histone H3 and H4 modifications that also found high levels of histone acetylation in hESC and C33a cells . The low levels of H2B acetylation in SW480 cells are in agreement with a previous study, which found that overexpression of RGC-32 in SW480 suppressed H2BK5 acetylation .
Polyadenylated H2B mRNA and protein levels in the cell cycle and in terminally differentiated cells
Histone levels are very tightly regulated; in fact, excess histone synthesis is detrimental and leads to chromosome abnormalities and cell death [36,37]. Histone variants, such as macroH2A, H2A.Z, and H3.3, that are made outside of DNA replication have specific functions [3,7]. Polyadenylated H2B transcripts accumulate following p53-induced senescence and during terminal cell differentiation . Therefore, we wanted to determine whether the polyadenylated H2B mRNA is expressed in a cell-cycle-independent manner and whether this leads to an increase in the levels of these proteins outside of S-phase. If these H2B isoforms are produced outside of S-phase, then it is possible that they have unique functions as replacements for canonical H2B. In addition, it is possible that H2B isoforms produced outside of S-phase associate with other replication-independent variants when they are assembled into chromatin. We hypothesized that polyadenylated H2B sequences would be expressed in a cell-cycle-independent manner. In order to test this hypothesis, we designed primers against the histone H2B1D coding sequence and against the polyadenylation sequence. We synchronized cells and synthesized cDNA from mRNA collected during G1- and S-phases then compared the relative amount of mRNA across these cell cycle phases using quantitative real-time (q-RT)-PCR.
As expected, the known replication-independent histone variant H2A.Z continued to incorporate heavy arginine at the same rate after the thymidine block (Figure 7B) as before. Other replication-independent variants such as histone H1.x showed similar trends (data not shown). Heavy arginine incorporation for replication-dependent H2A (H2A2B) and H2B (H2B1C, H2B1L, H2B1M) histone isoforms with no known polyadenylation signal leveled off after the thymidine block close to the expected 1:1 heavy-to-light ratio. There was still some incorporation of heavy arginine because a portion of the cells was not completely synchronized after one thymidine block. The H2B isoform with a known polyadenylation signal, H2B1D, showed a trend that was indistinguishable from other replication-dependent histone isoforms. This indicates that the polyadenylated H2B1D mRNA did not contribute significantly to the H2B1D protein levels after the thymidine block.
Most histone mRNAs were produced at higher levels in the proliferating myoblasts compared to myotubes (Additional file 7: Figure S6A). This result was expected because the myoblasts are replicating, while myotubes have exited the cell cycle. Polyadenylated H2B1D mRNA, however, was produced at significantly (P < 0.05) higher levels in myotubes compared to myoblasts. This result is in line with other studies that found that polyadenylated H2B variants were produced at higher levels in non-cycling cells . Since polyadenylated H2B1D is synthesized at higher levels in myotubes, we would then expect that more heavy, newly synthesized H2B1D would be incorporated into the chromatin compared to non-polyadenylated variants. We would also expect that there would be a higher percentage of H2B1D in the overall H2B profile of myotubes compared to myoblasts.
The overall rate of heavy arginine incorporation was much lower in non-dividing myotubes than in the proliferating myoblasts (Figure 8A). The histone variants H2A.Z and H1.x were labeled at a faster rate than all of the other histones under both proliferating and differentiation conditions, indicating a higher turnover (Figure 8A and Additional file 8: Figure S7). All of the canonical histone variants, including histone H2B, incorporated heavy arginine at essentially the same rate (Figure 8A). In addition, the relative abundances of different histone H2B variants remained essentially unchanged between proliferating and differentiated cells (Figure 8B).
Through a combination of different mass spectrometry methods, we identified all 13 H2B isoforms. We also identified H2B sequences modified with up to four acetylations, the known C-terminal ubiquitination site, and one new ubiquitination site. We described a bottom-up method that can be used to consistently quantify the relative abundances of nine H2B isoform N-terminal tail sequences and their modifications from unfractionated histone samples. This same method can also be used to study the H2B C-terminal tail and its modifications in a very straightforward and rapid manner.
There are variations in the levels of H2B isoforms and their modifications in different cancer cell lines. Variations in H2B isoform expression had been observed before in quantitative PCR (qPCR) assays and microarray studies at the mRNA level; however, this is the first investigation of variations at the protein level. H2B isoforms are all located in the histone gene clusters, facilitating their expression during S-phase, but individual H2B genes have different promoter sequences . This means that individual H2B genes may be regulated differently in cancer cell lines and possibly in a tissue-specific manner. We found that cancer cell lines from the same tissue exhibited similar H2B isoform profiles, so it is possible that H2B isoforms exhibit tissue-specific profiles. The differences in H2B cancer cell lines may alternatively result from chromosomal aberrations or just from other abnormal gene regulation in cancer cells. PANC cells consistently have an extra chromosome 6 (containing histone clusters 1 and 2), U2OS has rearrangements of chromosome 1 (containing histone cluster 3), and HeLa cells have extra copies of chromosomes 1 and 6 (ATCC.org). Regardless, differences in H2B isoform expression between cancer cell lines should be investigated further as controlling levels of H2B may be important for maintaining histone levels and maintaining a balance between histone family members.
The most striking difference between the cell lines was in the levels of H2B1B and H2B1M. Histone H2B1B only differs from other H2B isoforms at two positions with an alanine-to-serine difference and a valine-to-isoleucine difference, and it is unclear what affect these sequence variations would have on protein function. However, H2B1M has much larger differences in sequence across four amino acids in its N-terminal tail, including a threonine-to-asparagine difference that is not present in any other H2B isoforms. Differential expression of H2B1B had not been identified before, but low-H2B1M expression was identified as a biomarker for breast cancer and gastric cancer progression in microarray and proteomics studies [42,43]. Previous investigations of canonical H2A isoforms found that up-regulation of specific H2A sequences in chronic lymphocytic leukemia and in MCF-7 breast cancer cells caused increased proliferation and carcinogenesis in those cell types [44,45]. Similar studies should be performed by overexpressing H2B1B and H2B1M in some of the cancer cells where the levels of these variants are very low to see if this causes a decrease in cancer cell proliferation or progression.
We demonstrated that polyadenylated H2B mRNA is produced independent of DNA replication. In this manner, the H2B1D, H2B1K, and H2B2E isoforms that produce polyadenylated transcripts are similar to replication-independent variants. However, histone production is not independent of DNA replication at the protein level. After blocking DNA synthesis in S-phase using a double-thymidine block, there was very little synthesis of the H2B1D protein despite continued production of its transcript. It is possible that the polyadenylated transcript level is very low and does not contribute significantly to the total H2B protein level. In fact, a previous study on polyadenylated H2B2E found that polyadenylated H2B2E transcript levels were very low in normal cycling cells . If this is true, and the amount of protein being produced outside of DNA replication is very low, then we may not be able to measure the increase in H2B from replication-independent production. It is also possible that the polyadenylated H2B transcripts are not translated, and are instead the result of aberrant histone 3′ end processing as a result of missing cell-cycle-dependent histone mRNA processing proteins [39,46,47], and that most histone production in non-cycling cells comes from some basal histone transcription that occurs throughout the cell cycle. In non-cycling myotubes, all histone H2B isoforms turned over at the same rate, suggesting no preference for any sequence, regardless of whether it can be polyadenylated or not. On the other hand, histone variants with genes outside of the main clusters (for example H2A.Z and H1.x) had different turnover profiles compared to the replication-dependent variants in myoblasts and myotubes, suggesting that there may be specific regulation for these variants in myotubes.
Out of the three polyadenylated H2B isoforms, we were only able to distinguish H2B1D from other H2B isoforms using its N-terminal tail sequence and thus it was the only form that could be quantified independently using bottom-up MS. H2B1K and H2B2E have the same N-terminal tail sequence as H2B1C, and these variants can only be distinguished from other H2B by top-down mass spectrometry using instruments with a higher mass resolution than what is available in our lab, but it would be interesting to see if they show the same trends as H2B1D. In any case, H2B produced outside of the cell cycle in HeLa cells is not as abundant as known H2A variants and probably does not have the same set of important replication-independent functions.
We identified all 13 somatic H2B isoforms using a combination of different mass spectrometry methods in the most comprehensive study of H2B isoforms and their modifications to date. In addition, we developed a method to quantify nine of the H2B isoforms using middle-down MS. We applied this method to determine whether any H2B isoforms have variations in expression levels across cell types, are produced independently of cell cycle, or are associated with specific interacting partners. We found that the level of histone H2B isoforms, particularly H2B1B and H2B1M, varies across cancer cell lines, suggesting that H2B expression may be regulated differently across cell types or possibly misregulated in cancer. We also found that some H2B isoform genes can produce polyadenylated mRNA independent of the cell cycle, but these transcripts do to seem to be translated in significant quantities outside of S-phase.
HeLa S3 cells were maintained in suspension in minimal essential Joklik’s modification medium (Sigma Aldrich, St. Louis, MO, USA) with 10% newborn calf serum (Invitrogen, Carlsbad, CA, USA) and penicillin-streptomycin (pen-strep) solution diluted 1:100 (10,000 units penicillin G and 10 mg streptomycin per milliliter) (Thermo Fisher Scientific, Waltham, MA, USA) and supplemented with 1% Glutamax (Invitrogen, Carlsbad, CA, USA). Jurkat, and HL60, cells were maintained in RPMI-1640 (Mediatech, Inc., Manassas, VA, USA) medium supplemented with 10% fetal bovine serum and 1:100 pen-strep, and PANC cells and U2OS cells were grown in Dulbecco’s Modified Eagle’s Medium (DMEM; Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum and 1:100 pen-strep, and PC3 cells were grown in Ham’s F12-K media (Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum and 1:100 pen-strep. Data for the remaining cell lines were analyzed using RAW data from a previous study in our lab. Cell growth conditions, sample preparation, and mass spectrometer parameters for the remaining cell lines can be found in Leroy et al. (141).
For the cell cycle synchronization experiments, HeLa cells were blocked in S-phase by adding 2 mM thymidine to cells growing at a density of 3 × 105 cells/mL. The thymidine was removed for 9 h, followed by a second 17-h block in 2 mM thymidine. Following synchronization, cells were collected at 2, 4, 6, and 8 h and fixed in ethanol and stained with propidium iodide for cell cycle analysis by flow cytometry or snap-frozen and saved for q-RT-PCR analysis.
For the cell cycle SILAC-labeling experiments, HeLa cells were maintained in heavy arginine media with 10% dialyzed fetal bovine serum (FBS) and penicillin-streptomycin solution diluted 1:100. The media was made in-house using Joklik’s modification of Minimal Essential Media Eagle formulation, replacing arginine in the media with the same concentration of stable-isotope-labeled arginine (L-Arginine:HCl, 13C6,15 N4, Cambridge Isotope Laboratories, Andover, MA, USA). The cells were maintained in heavy media for 24 h and then blocked in S-phase by adding 2 mM thymidine to the growth medium. Cells were collected at time points before the thymidine block and up to 24 h after the block to monitor heavy arginine incorporation into histones.
Human myogenic cell line, LHCN M2, was a kind gift from Dr. Woodring Wright (UT Southwestern Medical Center at Dallas, Dallas, TX, USA). The immortalized myoblasts were cultured in proliferation medium and, once confluent, were differentiated in low-serum-containing medium as described previously . For heavy arginine labeling of myoblasts and myotubes, the media-contained DMEM was exchanged for one deficient in arginine and lysine (Thermo Scientific, Waltham, MA, USA) supplemented with [13C6] L-Arginine (98% purity) (Cambridge Isotope Laboratories, Tewksbury, MA, USA) and light lysine and proline (Sigma Aldrich, St. Louis, MO, USA). Similarly, dialyzed FBS (Invitrogen, Carlsbad, CA, USA) was used instead of FBS (Invitrogen, Carlsbad, CA, USA). Briefly, myoblasts grown in light proliferation medium (day 0) were divided into two batches, one expanded in heavy medium and the other in light medium. Cells grown in heavy proliferation medium were harvested on days 1, 2, 3, and 7. Cells that grew confluent in light proliferation medium (7 days) were switched to heavy differentiation medium and harvested on days 1, 2, 3, and 7 using trypsin after washing with sterile PBS. The cells were collected by centrifugation, washed in PBS, snap-frozen in liquid nitrogen, and stored at −80°C. Two biological replicates were used for histone analysis.
LHCN M2 cells were fixed 4 days after seeding for myoblasts and 6 days after differentiation for myotubes, and indirect immunofluorescence staining was performed using 1:20 dilution of myosin heavy chain supernatant (Developmental Studies Hybridoma Bank, cat# MF-20) as primary antibody and anti-mouse Alexa Fluor 488 as secondary antibody. The images were visualized at 20× in Nikon Inverted Microscope TE2000 and acquired with Image Pro Plus 7.0, (Media Cybernetics, Rockville, MD, USA).
RNA was extracted from synchronized cells using TRIzol (Invitrogen, Carlsbad, CA, USA). Five micrograms of RNA was then treated with DNase I (NEB, Ipswich, MA, USA) at 65°C for 1 h to remove contaminating genomic DNA. cDNA was synthesized using random hexamer primers and the Superscript III kit (Invitrogen, Carlsbad, CA, USA). H2B isoforms were then quantified in each sample using isoform-specific primers and Sybrgreen master mix (Invitrogen, Carlsbad, CA, USA). Two housekeeping genes (18S RNA and glyceraldehyde 3-phosphate dehydrogenase (GAPDH)) were included in cell cycle analyses, and three housekeeping genes (18S RNA, RORα, and GAPDH) were included for normalization (primer sequences available upon request). PCR was performed using 100 nM primers and the following cycling conditions: 50°C for 2 min, 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 59°C for 1 min. A no-RT control was also used to ensure that there was no contaminating DNA in the RT reaction; this is important because H2B genes do not contain introns and so have the same sequence as cDNA.
Cells were harvested and washed twice using phosphate-buffered saline. The cells were incubated in 10× the cell pellet volume of nuclear isolation buffer (15 mM Tris–HCl, 15 mM NaCl, 60 mM KCl, 5 mM MgCl2, 1 mM CaCl2, 250 mM sucrose, pH 7.5, and 0.5 mM AEBSF, 10 mM sodium butyrate, 2.5 μM microcystein, 1 mM DTT added fresh) with 0.3% NP-40 on ice for 5 min. The nuclei were collected by centrifuging in a table-top centrifuge at 500 × g at 4°C for 5 min. The resulting nuclear pellet was washed twice with nuclear isolation buffer without NP-40. After extracting the nuclei, histones were extracted by re-suspending the nuclei in 5× the nuclear pellet volume of 0.4 N sulfuric acid at 4°C with rotation. The insoluble nuclear debris were pelleted at 3600 × g at 4°C for 5 min, and the supernatant was retained. Finally, histone proteins were precipitated on ice for 1 h after adding 100% trichloroacetic acid to the acid extraction supernatant at a ratio of 1:5 (v/v). The precipitated proteins were washed once with 0.1% HCl in acetone (−20°C), and then twice with acetone (−20°C), then dried overnight on the bench-top.
RP-HPLC purification of histone H2B
Around 500 μg of acid-extracted histones were re-suspended in 200 μL RP-HPLC buffer A (5% HPLC-grade acetonitrile, 95% HPLC-grade water, 0.2% trifluoroacetic acid (TFA)) then loaded onto a Vydac 4.6 mm, 5 μm, 300 Å pore size column for 5 min at 2% RP-HPLC buffer B (95% HPLC-grade acetonitrile, 5% HPLC-grade water, 0.2% TFA) and resolved using a gradient from 30% buffer B to 60% buffer B at a flow rate of 0.8 μL/min over 100 min. Fractions corresponding to histone H2B were pooled and immediately dried in a speed-vac.
Propionic anhydride derivatization
Acid-extracted histones or HPLC-purified H2B were re-suspended in 20 μL 100 mM ammonium bicarbonate (pH 8), and the final pH was adjusted to 8 using ammonium hydroxide (28% to 30% ammonia basis, Sigma Aldrich, St. Louis, MO, USA). The starting pH for the propionic anhydride derivatization should be pH 8 and not higher (cause of unwanted serine propionylation) or lower (decreased reaction efficiency) to ensure that only primary amines (lysines and protein N-termini) are derivatized. After samples were set to pH 8, they were mixed with 10 μL of a freshly prepared propionic anhydride solution (75% isopropanol, 25% propionic anhydride) by vortexing, and then, 5 μL of ammonium hydroxide was immediately added to the mixture. More ammonium hydroxide was added if the pH needed to be adjusted to pH 8, then the samples were incubated at 37°C for 10 min. After incubation, the samples were dried in a speed-vac to <10 μL to remove propionic acid formed as a result of the reaction. The propionic anhydride derivatization reaction was repeated one more time. After the samples were dried to <10 μL, they were brought up to 50 μL using 10 mM ammonium bicarbonate (pH 8) and digested with trypsin at a ratio of 1 μg trypsin:20 μg sample for 6 to 8 h at 37°C. After the trypsin digest, the samples were dried to 20 μL and subjected to two more rounds of propionylation to cap peptide N-terminal amines. Following propionylation, the samples were de-salted using C18-StageTips  in preparation for MS analysis.
Top-down H2B MS analysis
A linear triple quadrupole orbitrap (LTQ-Orbitrap) XL mass analyzer (Thermo Fisher Scientific, Waltham, MA, USA) was used for the initial top-down MS analyses of HeLa and Jurkat H2B. The LTQ-Orbitrap XL was set to have an automatic gain control (AGC) value of 5 × 105 for the FT full-MS scans and the FT MS/MS scans, the HCD gas and the ‘FT zero offset’ were turned off, and the detection delay was set to ‘low.’ Next, the LTQ ion optics were tuned on the myoglobin 848.64 m/z ion to improve ion transmission for large ions. The resolving power of the Orbitrap was set to 30,000.
Top-down H2B analysis of other cancer cell lines was performed on an Orbitrap Fusion instrument (Thermo Fisher Scientific, Waltham, MA, USA). The AGC target was set to 5 × 105, and the maximum injection time was 100 ms, for the Orbitrap when it was collecting both full-MS and MS/MS spectra. A resolving power of 60,000 was used for full-MS scans and 30,000 for MS/MS scans. The S lens voltage was set at 60%, and mild in-source fragmentation (around 40 V) was used to improve signal for the high-m/z range and to reduce non-covalent adducts.
Pure histone H2B at a concentration of 1 μg/μL was infused onto the mass spectrometer using a pulled, fused silica tip (360-μM outer diameter, 50-μm inner diameter) on a nano-ESI source at a flow rate of 0.7 μL/min using a 50-μL syringe. The source voltage was set to 2.3 V for both instruments. The histone H2B signal was averaged over 10 microscans, and the scan range was set to encompass the most abundant charge states of histone H2B. After recording the full-MS spectra for several minutes, H2B peaks were selected for fragmentation using an isolation window of 3 m/z with no fragmentation to record which species were being successfully isolated. Next, the ETD reaction time was increased from 5 ms until the optimal MS/MS signal and a wide range of fragment ion masses were achieved. ETD spectra were summed over 5 microscans, and the resulting spectra were averaged over several minutes of collection time to improve the signal.
Bottom-up LC-MS analysis
Samples were re-suspended in buffer A (0.1% formic acid in LC/MS-grade water). Chromatographic separation of propionylated, digested histone or pure H2B samples was performed on a 15-cm reversed-phase column packed with Magic C18 AQ resin (5 μm, 100 Å pore size, Michrom Bioresources, Auburn, CA, USA) using a flow rate of 0.25 μL/min on either an Easy-nanoLC (Thermo Fisher Scientific, Waltham, MA, USA) or an Eksigent nanoLC-Ultra 2D using a gradient from 1% buffer B (0.1% formic acid in LC/MS-grade acetonitrile) to 32% B over 42 min. The HPLC was coupled to either an LTQ-Orbitrap Velos or an LTQ-Orbitrap Elite. The mass spectrometer was operated in a semi-targeted mode over three segments of the run. During the first two segments, one full-MS spectrum was collected from 350 to 1,600 m/z, then five masses corresponding to histone H4 and H3 acetylation states were targeted, and the remaining scans were data-dependent CAD MS/MS scans of the top six most abundant masses from the full-MS scan. In the third segment, two full-MS scans were collected: one from 350 to 1,600 m/z and one from 900 to 1,300 m/z, followed by data-dependent scans of the top ten most abundant ions from the second, limited mass range full-MS scan. The AGC settings for the mass spectrometers were set as follows: 3 × 105 for the ion trap and 1 × 106 for the Orbitrap. The relative abundance of histone H2B isoforms and their acetylated forms was calculated based on the area under the extracted ion chromatograms from the limited mass-range MS1 scan for all of the detectable charge states of each H2B peptide.
RAW files of heavy-arginine-labeled histone samples were loaded into Proteome Discoverer (version 1.4) for analysis. The data was searched against a custom database of 82 known human histone H3, H4, H2A, H2B, and H1 and other chromatin-related protein sequences using Mascot (version 2.2, Matrix Science, Boston, MA, USA) in Proteome Discoverer. We set N-terminal propionylation as a static modification and lysine propionylation and acetylation as variable modifications. The enzyme was set as Arg-C, and up to two missed cleavages were allowed. Quantification was done through the event detector and precursor ion quantifier nodes of Proteome Discoverer. The search results and peak matches were validated manually for H2B spectra using MS/MS spectra and relative retention times as a guide, and some false-positive peptide identifications were excluded from the analysis. Results from each peptide were averaged. Only peptides that are unique to a particular histone sequence and that are known to not be heavily modified were used in our analyses.
automatic gain control
collisionally activated dissociation
electron transfer dissociation
higher energy collisional dissociation
human embryonic stem cell
human foreskin fibroblast
linear triple quadrupole orbitrap
stable isotope labeling of amino acids in cell culture
BG acknowledges funding from an NIH grant R01GM110174, an NIH Innovator grant (DP2OD007447) from the Office of the Director, and the National Science Foundation (NSF) Early Faculty CAREER award.
- Andrews AJ, Luger K. Nucleosome structure(s) and stability: variations on a theme. Annu Rev Biophys. 2011;40:99–117.View ArticlePubMedGoogle Scholar
- Li G, Reinberg D. Chromatin higher-order structures and gene regulation. Curr Opin Genet Dev. 2011;21:175–86 [Chromosomes and Expression Mechanisms].View ArticlePubMed CentralPubMedGoogle Scholar
- Szenker E, Boyarchuk E, Almouzni G. Properties and functions of histone variants. In: Fundamentals of Chromatin. New York: Springer; 2014. p. 375–426.View ArticleGoogle Scholar
- Talbert PB, Henikoff S. Histone variants - ancient wrap artists of the epigenome. Nat Rev Mol Cell Biol. 2010;11:264–75.View ArticlePubMedGoogle Scholar
- Banaszynski LA, Allis CD, Lewis PW. Histone variants in metazoan development. Dev Cell. 2010;19:662–74.View ArticlePubMed CentralPubMedGoogle Scholar
- Henikoff S, Furuyama T, Ahmad K. Histone variants, nucleosome assembly and epigenetic inheritance. Trends Genet. 2005;20:320–6.View ArticleGoogle Scholar
- Maze I, Noh K-M, Soshnev AA, Allis CD. Every amino acid matters: essential contributions of histone variants to mammalian development and disease. Nat Rev Genet. 2014;15:259–71.View ArticlePubMed CentralPubMedGoogle Scholar
- Couldrey C, Carlton MBL, Nolan PM, Colledge WH, Evans MJ. A retroviral gene trap insertion into the histone 3.3A gene causes partial neonatal lethality, stunted growth, neuromuscular deficits and male sub-fertility in transgenic mice. Hum Mol Genet. 1999;8:2489–95.View ArticlePubMedGoogle Scholar
- Faast R, Thonglairoam V, Schulz TC, Beall J, Wells JR, Taylor H, et al. Histone variant H2A.Z is required for early mammalian development. Curr Biol. 2001;11:1183–7.View ArticlePubMedGoogle Scholar
- Bonenfant D, Coulot M, Towbin H, Schindler P, van Oostrum J. Characterization of histone H2A and H2B variants and their post-translational modifications by mass spectrometry. Mol Cell Proteomics. 2006;5:541–52.View ArticlePubMedGoogle Scholar
- Siuti N, Roth MJ, Mizzen CA, Kelleher NL, Pesavento JJ. Gene-specific characterization of human histone H2B by electron capture dissociation. J Proteome Res. 2006;5:233–9.View ArticlePubMedGoogle Scholar
- Kari V, Karpiuk O, Tieg B, Kriegs M, Dikomey E, Krebber H, et al. A subset of histone H2B genes produces polyadenylated mRNAs under a variety of cellular conditions. PLoS One. 2013;8, e63745.View ArticlePubMed CentralPubMedGoogle Scholar
- Zalensky AO, Siino JS, Gineitis AA, Zalenskaya IA, Tomilin NV, Yau P, et al. Human testis/sperm-specific histone H2B (hTSH2B) molecular cloning and characterization. J Biol Chem. 2002;277:43474–80.View ArticlePubMedGoogle Scholar
- Boulard M, Gautier T, Mbele GO, Gerson V, Hamiche A, Angelov D, et al. The NH2 tail of the novel histone variant H2BFWT exhibits properties distinct from conventional H2B with respect to the assembly of mitotic chromosomes. Mol Cell Biol. 2006;26:1518–26.View ArticlePubMed CentralPubMedGoogle Scholar
- Marzluff WF, Gongidi P, Woods KR, Jin J, Maltais LJ. The human and mouse replication-dependent histone genes. Genomics. 2002;80:487–98.View ArticlePubMedGoogle Scholar
- Albig W, Trappe R, Kardalinou E, Eick S, Doenecke D. The human H2A and H2B histone gene complement. Biol Chem. 1999;380:7–18.View ArticlePubMedGoogle Scholar
- Marzluff WF, Wagner EJ, Duronio RJ. Metabolism and regulation of canonical histone mRNAs: life without a poly (A) tail. Nat Rev Genet. 2008;9:843–54.View ArticlePubMed CentralPubMedGoogle Scholar
- Bungard D, Fuerth BJ, Zeng P-Y, Faubert B, Maas NL, Viollet B, et al. Signaling kinase AMPK activates stress-promoted transcription via histone H2B phosphorylation. Science. 2010;329:1201–5.View ArticlePubMed CentralPubMedGoogle Scholar
- Wen H, Li Y, Xi Y, Jiang S, Stratton S, Peng D, et al. ZMYND11 links histone H3. 3K36me3 to transcription elongation and tumour suppression. Nature. 2014;508:263–8.View ArticlePubMed CentralPubMedGoogle Scholar
- Mao Z, Pan L, Wang W, Sun J, Shan S, Dong Q, et al. Anp32e, a higher eukaryotic histone chaperone directs preferential recognition for H2A. Z Cell Res. 2014;24:389–99.View ArticleGoogle Scholar
- Lowell JE, Kaiser F, Janzen CJ, Cross GA. Histone H2AZ dimerizes with a novel variant H2B and is enriched at repetitive DNA in Trypanosoma brucei. J Cell Sci. 2005;118:5721–30.View ArticlePubMedGoogle Scholar
- Hoeijmakers WA, Salcedo Amaya AM, Smits AH, Françoijs K-J, Treeck M, Gilberger T-W, et al. H2A. Z/H2B. Z double‐variant nucleosomes inhabit the AT‐rich promoter regions of the Plasmodium falciparum genome. Mol Microbiol. 2013;87:1061–73.View ArticlePubMed CentralPubMedGoogle Scholar
- Santoro SW, Dulac C. The activity-dependent histone variant H2BE modulates the life span of olfactory neurons. Elife. 2012;1, e00070.View ArticlePubMed CentralPubMedGoogle Scholar
- González-Romero R, Rivera-Casas C, Ausió J, Méndez J, Eirín-López JM. Birth-and-death long-term evolution promotes histone H2B variant diversification in the male germinal cell line. Mol Biol Evol. 2010;27:1802–12.View ArticlePubMedGoogle Scholar
- Zamdborg L, LeDuc RD, Glowacz KJ, Kim Y-B, Viswanathan V, Spaulding IT, et al. ProSight PTM 2.0: improved protein identification and characterization for top down mass spectrometry. Nucleic Acids Res. 2007;35(2):W701–6.View ArticlePubMed CentralPubMedGoogle Scholar
- Guner H, Close PL, Cai W, Zhang H, Peng Y, Gregorich ZR, et al. MASH suite: a user-friendly and versatile software interface for high-resolution mass spectrometry data interpretation and visualization. J Am Soc Mass Spectrom. 2014;25:464–70.View ArticlePubMed CentralPubMedGoogle Scholar
- Pesavento JJ, Mizzen CA, Kelleher NL. Quantitative analysis of modified proteins and their positional isomers by tandem mass spectrometry: human histone H4. Anal Chem. 2006;78:4271–80.View ArticlePubMedGoogle Scholar
- Plazas-Mayorca MD, Zee BM, Young NL, Fingerman IM, LeRoy G, Briggs SD, et al. One-pot shotgun quantitative mass spectrometry characterization of histones. J Proteome Res. 2009;8:5367–74.View ArticlePubMed CentralPubMedGoogle Scholar
- Chandrasekharan MB, Huang F, Sun Z-W. Histone H2B ubiquitination and beyond. Epigenetics. 2010;5:460–8.View ArticlePubMed CentralPubMedGoogle Scholar
- Beck HC, Nielsen EC, Matthiesen R, Jensen LH, Sehested M, Finn P, et al. Quantitative proteomic analysis of post-translational modifications of human histones. Mol Cell Proteomics. 2006;5:1314–25.View ArticlePubMedGoogle Scholar
- Doenecke D, Albig W, Bode C, Drabent B, Franke K, Gavenis K, et al. Histones: genetic diversity and tissue-specific gene expression. Histochem Cell Biol. 1997;107:1–10.View ArticlePubMedGoogle Scholar
- Van’t Veer LJ, Dai H, Van De Vijver MJ, He YD, Hart AA, Mao M, et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature. 2002;415:530–6.View ArticleGoogle Scholar
- Abba MC, Hu Y, Sun H, Drake JA, Gaddis S, Baggerly K, et al. Gene expression signature of estrogen receptor α status in breast cancer. BMC Genomics. 2005;6:37.View ArticlePubMed CentralPubMedGoogle Scholar
- LeRoy G, DiMaggio PA, Chan EY, Zee BM, Blanco MA, Bryant B, et al. A quantitative atlas of histone modification signatures from human cancer cells. Epigenetics Chromatin. 2013;6:1–14.View ArticleGoogle Scholar
- Vlaicu SI, Tegla CA, Cudrici CD, Fosbrink M, Nguyen V, Azimzadeh P, et al. Epigenetic modifications induced by RGC-32 in colon cancer. Exp Mol Pathol. 2010;88:67–76.View ArticlePubMed CentralPubMedGoogle Scholar
- Meeks-Wagner D, Hartwell LH. Normal stoichiometry of histone dimer sets is necessary for high fidelity of mitotic chromosome transmission. Cell. 1986;44:43–52.View ArticlePubMedGoogle Scholar
- Singh RK, Liang D, Gajjalaiahvari UR, Kabbaj MH, Paik J, Gunjan A. Excess histone levels mediate cytotoxicity via multiple mechanisms. Cell Cycle. 2010;9:4236–44.View ArticlePubMed CentralPubMedGoogle Scholar
- Collart D, Ramsey-Ewing A, Bortell R, Lian J, Stein J, Stein G. Isolation and characterization of a cDNA from a human histone H2B gene which is reciprocally expressed in relation to replication-dependent H2B histone genes during HL60 cell differentiation. Biochemistry (Mosc). 1991;30:1610–7.View ArticleGoogle Scholar
- Pirngruber J, Johnsen SA. Induced G1 cell-cycle arrest controls replication-dependent histone mRNA 3′ end processing through p21, NPAT and CDK9. Oncogene. 2010;29:2853–63.View ArticlePubMedGoogle Scholar
- Wunsch AM, Lough J. Histones synthesized at different stages of myogenesis are differentially degraded in myotube cells. J Cell Physiol. 1989;141:97–102.View ArticlePubMedGoogle Scholar
- Chowdhary R, Ali RA, Albig W, Doenecke D, Bajic VB. Promoter modeling: the case study of mammalian histone promoters. Bioinformatics. 2005;21:2623–8.View ArticlePubMedGoogle Scholar
- Cui J, Li F, Wang G, Fang X, Puett JD, Xu Y. Gene-expression signatures can distinguish gastric cancer grades and stages. PLoS One. 2011;6, e17819.View ArticlePubMed CentralPubMedGoogle Scholar
- Choong LY, Lim S, Chong PK, Wong CY, Shah N, Lim YP. Proteome-wide profiling of the MCF10AT breast cancer progression model. PLoS One. 2010;5, e11030.View ArticlePubMed CentralPubMedGoogle Scholar
- Su C-H, Tzeng T-Y, Cheng C, Hsu M-T. An H2A histone isotype regulates estrogen receptor target genes by mediating enhancer-promoter-3′-UTR interactions in breast cancer cells. Nucleic Acids Res. 2014;42:3073–88.View ArticlePubMed CentralPubMedGoogle Scholar
- Singh R, Mortazavi A, Telu KH, Nagarajan P, Lucas DM, Thomas-Ahner JM, et al. Increasing the complexity of chromatin: functionally distinct roles for replication-dependent histone H2A isoforms in cell proliferation and carcinogenesis. Nucleic Acids Res. 2013;41:9284–95.View ArticlePubMed CentralPubMedGoogle Scholar
- Romeo V, Griesbach E, Schümperli D. CstF64: cell cycle regulation and functional role in 3′ end processing of replication-dependent histone mRNAs. Mol Cell Biol. 2014;34:4272–84.View ArticlePubMedGoogle Scholar
- Youngblood BA, Grozdanov PN, MacDonald CC. CstF-64 supports pluripotency and regulates cell cycle progression in embryonic stem cells through histone 3′ end processing. Nucleic Acids Res. 2014;42:8330–42.View ArticlePubMed CentralPubMedGoogle Scholar
- Stadler G, Chen J, Wagner K, Robin J, Shay J, Emerson Jr C, et al. Establishment of clonal myogenic cell lines from severely affected dystrophic muscles - CDK4 maintains the myogenic population. Skelet Muscle. 2011;1:12.View ArticlePubMed CentralPubMedGoogle Scholar
- Rappsilber J, Ishihama Y, Mann M. Stop and go extraction tips for matrix-assisted laser desorption/ionization, nanoelectrospray, and LC/MS sample pretreatment in proteomics. Anal Chem. 2003;75:663–70.View ArticlePubMedGoogle Scholar
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.