• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of molcellbPermissionsJournals.ASM.orgJournalMCB ArticleJournal InfoAuthorsReviewers
Mol Cell Biol. Dec 2007; 27(24): 8466–8479.
Published online Oct 8, 2007. doi:  10.1128/MCB.00993-07
PMCID: PMC2169421

Mammalian ASH1L Is a Histone Methyltransferase That Occupies the Transcribed Region of Active Genes[down-pointing small open triangle]

Abstract

Histone lysine methylation regulates genomic functions, including gene transcription. Previous reports found various degrees of methylation at H3K4, H3K9, and H4K20 within the transcribed region of active mammalian genes. To identify the enzymes responsible for placing these modifications, we examined ASH1L, the mammalian homolog of the Drosophila melanogaster Trithorax group (TrxG) protein Ash1. Drosophila Ash1 has been reported to methylate H3K4, H3K9, and H4K20 at its target sites. Here we demonstrate that mammalian ASH1L associates with the transcribed region of all active genes examined, including Hox genes. The distribution of ASH1L in transcribed chromatin strongly resembles that of methylated H3K4 but not that of H3K9 or H4K20. Accordingly, the SET domain of ASH1L methylates H3K4 in vitro, and knockdown of ASH1L expression reduced H3K4 trimethylation at HoxA10 in vivo. Notably, prior methylation at H3K9 reduced ASH1L-mediated methylation at H3K4, suggesting cross-regulation among these marks. Drosophila ash1 and trithorax interact genetically, and the mammalian TrxG protein MLL1 and ASH1L display highly similar distributions and substrate specificities. However, by using MLL null cell lines we found that their recruitments occur independently of each other. Collectively, our data suggest that ASH1L occupies most, if not all, active genes and methylates histone H3 in a nonredundant fashion at a subset of genes.

Posttranslational histone modifications perform critical functions during many processes involving chromatin, including gene expression (27). Several enzymes responsible for acetylation, methylation, phosphorylation, and ubiquitylation of histone substrates have been identified, and in some cases, defects in these enzymes have been linked to cancer and developmental abnormalities in humans (17, 22). One of the most studied histone-modifying enzymes is the mixed-lineage leukemia 1 protein (MLL1), a histone H3 lysine 4 (H3K4)-specific histone methyltransferase (HMTase) frequently found mutated in acute myeloid leukemia (12, 32). MLL1 is related to the Drosophila melanogaster Trithorax protein and belongs to a family of mammalian proteins (MLL1 to MLL5), many of which have documented H3K4 methyltransferase activity and promote transcription via their localization to active genes (15, 26, 32). Thus, it is important to define the normal physiological function of histone-modifying enzymes and their substrates, as well as their role in the pathophysiology of disease.

Lysine methylation is perhaps the most complex among the histone modifications, as it can occur at multiple sites, including lysines 4, 9, 27, 36, and 79 of histone H3 and lysine 20 of histone H4 (H3K4, H3K9, H3K27, H3K36, H3K79, and H4K20, respectively). Moreover, lysine residues can be mono-, di-, or trimethylated, and both the site and the extent of lysine methylation impact the biological function by attracting or repelling specific effector proteins. For instance, trimethylation of H3K4 (H3K4me3), a mark associated with actively transcribed genes, can recruit and/or stabilize the chromatin-remodeling complex NURF (60). This occurs through the direct recognition of this mark by the plant homeodomain (PHD) finger of BPTF, a subunit of NURF. Thus, placement of this particular mark at the 5′ region of active genes may promote nucleosome remodeling and subsequent gene transcription. Another modification, H3K9 methylation, has been reported by several laboratories to be associated with transcriptionally active genes in addition to its well-defined role at heterochromatin (2, 18, 29, 35, 54), though the degrees of H3K9 methylation at active genes have varied in these studies. These findings suggest a dual role in gene activation and repression for this modification. While several enzymes responsible for methylating H3K9 during gene repression and heterochromatin formation have been described previously (9, 39, 40), an H3K9-specific methyltransferase that occupies active genes has not been identified.

With one known exception (DOT1L), HMTase activity is conferred by a conserved catalytic motif called the SET domain [Su(var)3-9, E(z), Trithorax]. Although the majority of HMTases target only one specific lysine residue, the SET domain of Drosophila Ash1 (absent, small, or homeotic discs 1) has been reported to methylate H3K4, H3K9, and H4K20 in vitro and in vivo at trithorax response elements (TRE) at active homeotic genes (3, 43). Although these findings have recently been challenged (4, 37, 38), we considered the hypothesis that the mammalian Ash1 homolog, ASH1-like (ASH1L), might methylate lysine residues at active mammalian genes. This possibility was raised by previous observations that methylated H3K9 and H4K20 can occur in the transcribed portion of active genes (2, 55), similarly to what has been observed for H3K4 (45). In this report, we show that ASH1L, like its Drosophila counterpart, occupies transcribed chromatin and can methylate histone tails in vitro. However, our data do not support a role for ASH1L in the methylation of H3K9 or H4K20 at transcribed regions. Specifically, ASH1L occupies the 5′-transcribed region of active genes, including Hox cluster genes, in a fashion virtually identical to that of MLL1, and its spatial distribution correlates tightly with H3K4me3. Moreover, the occupancy of ASH1L is lowest in portions of transcribed regions where H3K9 and H4K20 are most prevalent. In addition, the extended SET domain of ASH1L methylates H3K4 in vitro, and knockdown of ASH1L results in reduced H3K4me3 at the HOXA10 gene. Overall, our data suggest that ASH1L and MLL1 act in a parallel and partially redundant manner to maintain histone methylation at a surprisingly large number of active mammalian genes, further providing evidence that Trithorax-related HMTases are coupled to the activated transcription cycle.

MATERIALS AND METHODS

Cell culture.

HeLa, 293T, and K562 cells were cultured in Dulbecco modified Eagle medium (DMEM) with 10% fetal bovine serum, 2% penicillin-streptomycin, 1% glutamine, and 1% Na pyruvate. CCE embryonic stem cells were grown in leukemia inhibitory factor-containing DMEM with 10% fetal calf serum, 2% penicillin-streptomycin, 1% glutamine, 1% nonessential amino acids, and β-mercaptoethanol. Mll1+/+ and Mll1−/− mouse embryonic fibroblasts (MEFs) were kindly provided by J. Hess (University of Michigan) and grown in Iscove's DMEM containing 10% fetal bovine serum, 2% penicillin-streptomycin, 1% glutamine, and 1% nonessential amino acids. G1E cells were cultured as previously described (59). Differentiation of G1E cells was accomplished by treatment of cells with 1 μm estradiol for 21 h. HeLa cells were treated with 75 μM 5,6,-dichloro-1-β-d-ribofuranosylbenzimidazole (DRB) (Sigma) where indicated.

Antibodies.

Antibodies directed against human ASH1L and MLL1 were made as previously described (15). The anti-ASH1L 296 sera and affinity-purified 337 (337ap) antibodies were directed against residues spanning amino acids (aa) 1612 to 1767 and aa 8 to 146 of human ASH1L, respectively. MLL1 antibody (antibody 456) was made against residues spanning aa 2213 to 2593. Where indicated, the anti-ASH1L antibody ab4477 (Abcam), generated against residues 1650 to 1750, was used. All anti-ASH1L antibodies recognize human and mouse orthologs due to sequence homology. Anti-histone H3 (ab1791), anti-monomethyl H4K20 (ab9051), anti-monomethyl H3K36 (ab9048), and anti-trimethyl H3K36 (ab9050) antibodies were from Abcam. Anti-dimethyl H3K4 (07-030), anti-trimethyl H3K4 (07-473), anti-dimethyl H3K36 (07-369), and anti-trimethyl H3K9 (07-442) antibodies were from Upstate, and anti-RNA polymerase II (Pol II) antibody (sc-899) was from Santa Cruz.

ChIP assay.

Chromatin immunoprecipitation (ChIP) was performed as described previously (55). Cross-linking was performed with 1% formaldehyde. For ChIP experiments with antibodies against MLL1 and ASH1L, 1 × 107 cells in 45 ml phosphate-buffered saline were additionally cross-linked with 34.5 mg ethylene glycol bis(succinimidyl succinate) dissolved in 5 ml of dimethyl sulfoxide for 30 min prior to formaldehyde treatment (63). Quantification of immunoprecipitated samples was performed by real-time PCR using SYBR green dye on an ABI Prism 7900HT machine. Standards were made for each condition via serial dilution of unprecipitated input samples and used for each PCR. All reactions were performed in duplicate. Data shown are the averages of two or more independent experiments unless otherwise noted. Primers were designed using Primer Express software (Applied Biosystems), and sequences are available in the supplemental material.

GST-protein purification.

ASH1L-CsetC (the SET domain of ASH1L with flanking sequences [aa 2040 to 2716]) was generated by PCR and inserted into pGEX-2TK (Amersham Biosciences). Glutathione S-transferase (GST) fusion proteins containing residues 1 to 46 of histone H3 [GST-H3(1-46)] and residues 1 to 34 of histone H4 [GST-H4(1-34)] were a kind gift of M. Grunstein (UCLA). Point mutations in the GST-H3 and -H4 constructs were generated using a QuickChange site-directed mutagenesis kit per the manufacturer's instructions (Stratagene). Constructs were expressed in bacteria (BL21 strain) and purified using S-trans glutathione-agarose beads according to the manufacturer's protocol (Invitrogen).

HMTase assays.

In vitro histone methylation reactions were carried out for 18 h at 30°C in a 20-μl reaction volume containing 10 mM Tris-HCl, pH 8.8, 50 mM NaCl, 1.5 mM dithiothreitol, 1 μl of [3H]methyl S-adenosyl-methionine (80 Ci/mmol; Amersham Biosciences), and protease inhibitor cocktail with 2 μg of GST-ASH1L-CsetC or GST alone. Oligonucleosomes (OGNS) were assembled according to the manufacturer's instructions (Active Motif), and 4 μl was used per 20-μl HMTase assay. Histone substrates (recombinant H3, core histones, and GST-H3 and -H4) were used at 1 to 2 μg per reaction, except histone peptides, which were used at 100 ng per reaction. Half of the reaction mixture was separated on a 15% sodium dodecyl sulfate (SDS)-polyacrylamide gel, stained with Colloidal blue (Invitrogen), and treated with Autofluor (National Diagnostics) prior to exposing to film. The other half of the reaction mixture was spotted on Whatman P81 phosphocellulose filter paper and washed three times with wash buffer (50 mM NaHCO2-Na2CO3, pH 9.2) and once with acetone before drying and quantification on a liquid scintillation counter. We note that the activities of GST-ASH1L-CsetC were highly variable among different preparations when assayed against peptides or GST-H3 as substrates.

Gene knockdown.

Generation of short-hairpin RNA (shRNA) expression constructs was performed per the manufacturer's protocol (Open Biosystems). Briefly, pSM2 plasmids containing shRNAs directed against human ASH1L (shASH1L) (clone identification no. V2HS_175879) were obtained from Open Biosystems and digested with XhoI/EcoRI. Digestion products were resolved on a 1.8% agarose gel, and an ~110-bp DNA fragment containing the shRNA was removed and cleaned. The fragment was then ligated into an XhoI/EcoRI-digested pLMP vector (pMSCV/LTRmiR30-PIG). Positive clones were identified via the presence of 985- and 7,019-bp fragments following SacII/XhoI digestion and confirmed by sequencing. For transfection, 293T cells plated at 60% confluence in 10-cm2 dishes were transfected by use of Fugene6 (Roche) with 2.1 μg of LMP vector that expresses shASH1L or shRNA directed against the control (shCTRL). Transfected cells were selected using 1 μg/ml puromycin, and cultures were used at 2 to 4 weeks posttransfection.

RESULTS

ASH1L is associated with the transcribed region of several highly active genes.

ASH1L is a large, multidomain protein containing motifs implicated in chromatin remodeling, including a SET domain, four AT hooks, a bromodomain, a bromoadjacent homology domain (BAH), a PHD finger, and MYND ligand domains (reference 31 and unpublished observations). In Drosophila, Ash1 is required for proper expression of the homeotic gene Ubx in haltere/third leg imaginal discs; however, the mechanism by which this occurs is controversial. In one study, Ash1 was reported to occupy the 5′ region of the actively transcribed Ubx gene (37). Another group reported that Ash1 was found at the promoter-proximal region of Ubx including the upstream TREs (43). In the latter study, the presence of Ash1 correlated with trimethylation of H3K4, H3K9, and H4K20, consistent with its described activity in vitro (3). While human ASH1L has been localized to scattered speckles throughout the nucleoplasm (31), the location of ASH1L at the resolution of individual genes has not been examined, i.e., mammalian ASH1L target genes have not been identified.

To begin to investigate ASH1L, we generated antibodies against distinct epitopes in the N terminus (aa 8 to 146, antibody 337ap) and middle portion (aa 1612 to 1767, antibody 296) of ASH1L. A commercially available antibody made against aa 1650 to 1750 was also used (ab4477), and all antibodies produced similar results in ChIP analyses. Although no mammalian target genes have been identified, we hypothesized that ASH1L may be present at active genes, consistent with putative H3K4, H3K9, and H4K20 methyltransferase activities. We therefore performed ChIP analyses with HeLa cells at several active (RPLP0, PPIA, RPS2, EEF1A1, and RPL41) and inactive (CD4 and MYT1) genes. All primers used for quantitative PCR in these experiments were designed to amplify the 5′-transcribed region (+0.5 kb downstream from the transcription start site [TSS]). Remarkably, ASH1L associated with the transcribed region of each active gene and was not detected at the transcribed region of the inactive genes CD4 and MYT1, similarly to RNA Pol II (Fig. 1A and B). These results were confirmed through the use of two additional independently derived antibodies (ab4477 [Fig. [Fig.1B]1B] and 337ap [data not shown]). Similar results were obtained with the human erythroleukemia cell line K562 and the murine embryonic stem cell line CCE (Fig. 1C to F). Together with our results from additional human (HEK-293T) and murine (G1E and MEFs [see below]) cell lines, this demonstrates that ASH1L is not employed in a cell type- or gene-specific manner. These findings suggest ASH1L recruitment might be linked to the general transcription machinery instead of recruitment via gene-specific transcription factors. It should be noted that this analysis was biased towards highly active genes and that ASH1L may not be detectable at genes expressed at low levels. While future ChIP-on-chip microarray analyses might allow for genome-wide profiling of ASH1L occupancy, our data indicate that ASH1L associates with many if not all active genes. It will be interesting to determine if Drosophila Ash1 similarly occupies a majority of active genes or if it is limited specifically to homeotic genes, such as Ubx.

FIG. 1.
ASH1L is associated with actively transcribed genes. (A) ChIP analysis, with HeLa cells, of RNA Pol II at the active genes RPLP0, PPIA, RPS2, EEF1A1, and RPL41 and the inactive genes CD4 and MYT1. (B) ChIP assay using anti-ASH1L antibodies 296 and ab4477 ...

Profile of ASH1L occupancy across a highly active mammalian gene.

To gain a better understanding of how ASH1L may function in mammalian cells, we profiled histone methylation and ASH1L occupancy across the human PABPC1 [poly(A) binding protein, cytoplasmic 1] gene in HeLa cells. The large transcribed region (>25 kb) and high expression level of the PABPC1 gene make it a suitable model to study by ChIP the pattern of histone modifications associated with transcription elongation (55). In addition, we previously mapped the locations of all six known sites of histone lysine methylation across this gene (55), thus allowing a direct comparison of ASH1L occupancy with underlying patterns of lysine methylation. For our quantitative ChIP analysis, we used primer pairs covering multiple regions across PABPC1, including sequences upstream of the TSS, near the promoter, the transcribed portion, and downstream of the poly(A) signal.

ChIP of ASH1L using the 296 and 337ap antibodies demonstrated that ASH1L associates primarily with the 5′-transcribed region of PABPC1 (Fig. (Fig.2A;2A; see also Fig. S1B in the supplemental material). This region encompasses +0.5 to 1.0 kb downstream of the TSS. With the ab4477 antibody, we observed a slight 5′ shift in ASH1L occupancy, which may be related to the accessibility of this particular epitope at distinct sites (Fig. (Fig.2B).2B). Specificities of the anti-ASH1L antibodies were confirmed in independent experiments using shRNA to knock down expression of ASH1L (see Fig. Fig.7).7). Consistent with prior findings, H3K4 trimethylation was restricted to the 5′-transcribed region of PABPC1 whereas H3K9 trimethylation was limited primarily to the 3′ portion of PABPC1 (Fig. (Fig.2C)2C) (55). Monomethylation of H4K20 was observed throughout the transcribed region of PABPC1 (Fig. (Fig.2D).2D). RNA Pol II peaks at the 5′-transcribed region but occupies the entire transcribed portion (Fig. (Fig.2E),2E), consistent with the high transcriptional activity of PABPC1 in HeLa cells.

FIG. 2.
Histone methylation and ASH1L occupancy at the PABPC1 gene. ChIP assays of HeLa cells with antibodies against (A) ASH1L (296), (B) ASH1L (ab4477), (C) H3K4me3 and H3K9me3, (D) H4K20me1, and (E) RNA Pol II. All error bars represent standard deviations. ...
FIG. 7.
ASH1L is required for full Hox gene expression and H3K4 trimethylation. (A) ChIP with anti-H3K4me3 and anti-ASH1L (296) antibodies at the active HoxA10 gene. (B) ChIP analysis of MLL1 (456) across the HoxA10 gene. (C) ChIP with anti-H3K4me2 across the ...

Strikingly, the occupancy profile of ASH1L across PABPC1 correlates specifically with trimethylation of H3K4. H4K20 monomethylation was also observed to overlap partially with ASH1L occupancy. However, the prominent distribution of this mark throughout the transcribed region suggests that it is established by an HMTase(s) other than ASH1L. Significantly, ASH1L and H3K9 trimethyl profiles had little to no overlap, suggesting that ASH1L is not responsible for H3K9 trimethylation at PABPC1.

While the manuscript was under review, Tanaka et al. reported that ASH1L can mono- and dimethylate H3K36 in vitro (52). Little is known about the function of H3K36me1/2 across active mammalian genes. H3K36me3, like H3K9me3 and H4K20me1, is found in the transcribed portion of active genes and is dependent upon transcription elongation (1, 20, 29, 55). To address the possibility that ASH1L occupancy also correlates with H3K36me1/2, we analyzed the methyl status of H3K36 across PABPC1. ChIP analysis demonstrated an enrichment of H3K36me3 across the transcribed region of PABPC1 (see Fig. S2C in the supplemental material), as previously shown (55). H3K36me3 was absent at the inactive genes CD4 and MYT1, consistent with this mark requiring transcription elongation to be placed (not shown). H3K36me1 and H3K3me2 were found to have a less defined pattern, as both marks were observed at the active PABPC1 locus and at the inactive CD4 gene but not at the inactive MYT1 gene. In our limited analysis, the highest detected levels of H3K36me1 and H3K36me2 were found 1.5 kb upstream of the PABPC1 TSS, followed by a precipitous drop in H3K36me1 and H3K36me2 throughout the transcribed portion of the gene (see Fig. S2A and B in the supplemental material). The latter finding likely reflects, in part, the conversion of H3K36me1/me2 to the trimethyl state. Consistent with these findings, Kim et al. observed H3K36me1/me2 at both intergenic and intragenic regions throughout the globin locus (18). Although the peak of ASH1L occupancy does not correlate with increased levels in H3K36 methylation, this result does not rule out ASH1L activity towards H3K36 in certain contexts in vivo.

ASH1L is recruited to gene loci upon transcription induction.

Our results thus far indicate that ASH1L may be a general factor associated with potentially all Pol II transcribed genes. To explore the temporal relationship between ASH1L occupancy and transcriptional activity, we investigated the recruitment of ASH1L during conditional activation. To do this, we studied the induction of an erythroid-specific gene in G1E cells. G1E cells are erythroid progenitors arrested in development due to a lack of the transcription factor GATA-1 (58). Estradiol treatment of G1E cells expressing GATA-1 fused to the ligand-binding domain of the human estrogen receptor (GATA-1-ER) triggers the activation of GATA-1 target genes, such as the β-major globin gene (Fig. (Fig.3A).3A). Previous reports used this system to demonstrate the inducibility of H3K4me3, H3K9me3, and H4K20me1 at the transcribed portion of the β-major globin gene (54, 55). We first analyzed the inducibility of Pol II occupancy across the entire β-major gene locus by ChIP in G1E cells exposed to estradiol for 21 h compared to that for untreated cells. Pol II was recruited primarily to the transcribed portion of the β-major globin gene after estradiol treatment (Fig. (Fig.3B).3B). Pol II was also induced at the locus control region (LCR) (hypersensitive sites 1, 2, 3, and 4), albeit at levels lower than that of the transcribed portion, consistent with a previous study with primary murine erythroid cells (44). We next evaluated ASH1L recruitment to the β-globin gene locus. Though little or no ASH1L was detected at the β-major globin gene open reading frame in undifferentiated G1E cells, upon GATA-1-ER-mediated activation, ASH1L was recruited to the transcribed portion of the β-major globin gene (Fig. (Fig.3C).3C). Moreover, ASH1L's occupancy correlated with the recruitment of Pol II and H3K4me3 (Fig. 3B and D). Localization of ASH1L to the transcribed region of the β-globin major gene indicates that it may facilitate transcription by acting during the early elongation stages of the transcription cycle. ASH1L was also present at the LCR in undifferentiated cells and induced further following estradiol treatment. Interestingly, experiments with G1E cells using only formaldehyde but not ethylene glycol bis(succinimidyl succinate) as the cross-linking reagent failed to detect significant levels of ASH1L at the LCR, although ASH1L was detected at the transcribed portion (data not shown). This suggests that the mechanism of recruitment of ASH1L is distinct at upstream cis-regulatory elements and transcribed regions.

FIG. 3.
Recruitment of ASH1L correlates with onset of gene expression. (A) Diagram of the murine β-globin gene locus (not drawn to scale). ChIP analysis of (B) RNA Pol II, (C) ASH1L (antibody ab4477), and (D) H3K4me3 in murine G1E cells before or after ...

ASH1L occupancy persists following a transcription elongation block.

H3K4 methylation is dependent upon transcription elongation (23). Considering ASH1L's colocalization with H3K4 trimethylation in the transcribed portion of active genes, we asked whether ASH1L occupancy with chromatin is coupled to transcription elongation. If so, it would be expected that ASH1L occupancy would decrease in a manner similar to H3K4 trimethylation. We approached this question by inhibiting transcription elongation in vivo with the inhibitor DRB. DRB prevents transcription elongation by inhibiting P-TEFb-mediated serine 2 phosphorylation of the C-terminal domain of Pol II. Treatment of HeLa cells with DRB for 6 h resulted in an approximately sixfold decrease in Pol II levels at +5.0 kb from the TSS, thus confirming the effectiveness of the drug (Fig. (Fig.4).4). While diminished Pol II occupancy at PABPC1 was also observed at +2.0 kb, sites closer to the TSS (i.e., +0.5 kb) displayed an increase in Pol II occupancy (Fig. (Fig.4).4). The elevated Pol II occupancy could be due to the combined effects of DRB-mediated Pol II stalling and continued initiation of Pol II. ChIP also confirmed that H3K4 trimethylation was sensitive to DRB treatment at all locations examined (Fig. (Fig.4).4). Surprisingly, despite diminished H3K4 trimethylation, ASH1L not only persisted but was also somewhat elevated between +0.5 and 2.0 kb of PABPC1 upon DRB treatment (Fig. (Fig.4).4). This suggests that while elongating Pol II might be involved in the initial ASH1L recruitment (Fig. (Fig.3B),3B), the continued presence of ASH1L at +2.0 kb upon lowering of Pol II might be mediated by other determinants, such as residual histone modifications. Similar results were obtained at the β-major globin gene in G1E cells treated for 18 h with estradiol followed by an additional 3 h in the presence of DRB (see Fig. S3 in the supplemental material). Consistent with these findings, Eissenberg et al. recently reported that the distribution of Ash1 on Drosophila polytene chromosomes was not affected by depletion of cyclin-dependent kinase 9, one of two components of Drosophila P-TEFb (10). However, cyclin-dependent kinase 9 depletion did significantly decrease H3K4 trimethylation and subsequent CHD1 recruitment. Moreover, these data, combined with the finding that ASH1L occupies the LCR of the β-major gene, suggest that the mere presence of ASH1L is insufficient for H3K4 methylation.

FIG. 4.
ASH1L occupancy persists following a transcription elongation block. ChIP with RNA Pol II, H3K4me3, and ASH1L (ab4477) antibodies at indicated sites at the PABPC1 gene in HeLa cells with (+ DRB) or without (no DRB) treatment with DRB for 6 h. ...

ASH1L methylates lysine 4 of histone H3 in vitro.

Drosophila Ash1 was reported to methylate H3K4, H3K9, and H4K20 at TREs (3); however, another report disagreed with some of these findings (3, 4). Our ChIP data are consistent with ASH1L targeting H3K4, as the occupancy of ASH1L correlates primarily with this mark. We therefore performed experiments to define the intrinsic histone-methylating activity of the mammalian ASH1L SET domain. The SET domain of ASH1L with flanking sequences (aa 2040 to 2716) was fused to GST (GST-ASH1L-CsetC), purified from bacteria, and assayed for HMTase activity in vitro. Initial experiments with recombinant histone H3 and core histones revealed that ASH1L's activity is quite low, ranging from 2 to 10% of the activity observed for other HMTases, such as SET7/9 and SUV39H1 (see Fig. S4 in the supplemental material; also, data not shown). Experiments using a minimal SET domain (aa 2138 to 2285) failed to demonstrate any measurable activity. In an attempt to improve activity of GST-ASH1L-CsetC, we varied experimental conditions, including pH, salt concentration, length of incubation time, and temperature. The activity we were able to detect under optimal conditions was still too low to obtain reliable results by using modification-specific antibodies or to determine target sites by mass spectrometry (not shown). Recombinant bacterially expressed Drosophila Ash1 was also reported to exhibit a very low activity (4).

HMTases can display increased activity when the histone substrate is packaged into OGNS arrays (34). To test this possibility, high-quality OGNS were assembled from HeLa core histones and DNA by use of the histone chaperone nucleosome assembly protein 1 (NAP-1) and ATP-utilizing chromatin assembly and remodeling factor (ACF). A small aliquot of OGNS was MNase digested to ensure proper assembly (data not shown), and the remaining OGNS were used in HMTase assays with GST alone, GST-ASH1L-CsetC, or GST-SUV39H1. Both GST-ASH1L-CsetC and GST-SUV39H1 reproducibly methylated OGNS to similar degrees (Fig. (Fig.5A).5A). Furthermore, autoradiography revealed a single band consistent with methylation of histone H3 (Fig. (Fig.5B5B).

FIG. 5.
The CsetC domain of ASH1L methylates H3K4 in vitro and is partially inhibited by preexisting H3K9 trimethylation. (A) HMTase activity of the SET domain of ASH1L (GST-ASH1L-CsetC), SUV39H1 (GST-SUV39H1), or GST alone with OGNS as the substrate. Four microliters ...

To better define the residue specificity of ASH1L, we performed HMTase assays using synthetic peptides containing the first 21 aa of histone H3 and residues 10 to 31 of histone H4. In addition, H3 peptides with preexisting trimethylation on K4, K9, or dually modified K4/K9 residues and monomethylated H4K20 were used. As seen in Fig. Fig.5A,5A, GST-ASH1L-CsetC methylated the H3 peptide but not the H4 peptide. Moreover, methylation of the H3 peptide was completely blocked when lysine 4 or lysine 4/9 was premethylated, indicating that ASH1L specifically targets lysine 4 in histone H3. Interestingly, preexisting methylation of H3K9 diminished methylation by GST-ASH1L-CsetC but not by the H3K4-specific HMTase SET7/9 (Fig. (Fig.5C;5C; see also Fig. S4C and D in the supplemental material), suggesting either that GST-ASH1L-CsetC can directly methylate H3K9 or that H3K9 methylation negatively impacts H3K4 methylation.

To distinguish between these possibilities, we used GST fusion proteins containing residues 1 to 46 of histone H3 [GST-H3(1-46)] and residues 1 to 34 of histone H4 [GST-H4(1-34)] along with mutant versions containing arginine substitutions at residue 4, 9, or 27 of histone H3 and at residue 20 of histone H4. As shown in Fig. Fig.5D,5D, GST-ASH1L-CsetC methylated GST-H3(1-46), H3K9R, and H3K27R. No activity was detected when H3K4R, GST-H4(1-34), or H4K20R was used as the substrate. In concert with the above-described results, this indicates that, under these conditions, ASH1L methylates H3K4 and that methylation at H3K9 impairs H3K4 methylation. The latter finding is intriguing as it suggests that cross-regulation among H3K9 and H3K4 methyl marks might contribute to their distinct distribution patterns along active genes (55). Similar cross talk between histone modifications has been described previously (39, 57). We observed no measurable activity towards H3K36, in contrast to what has been described recently (52). Ultimately, determining the exact target site(s) of ASH1L will require assaying the intact ASH1L complex.

Comparative analysis of ASH1L and MLL1 distribution across active genes in vivo.

The association of ASH1L with the 5′-transcribed region of all examined active genes and its intrinsic H3K4 HMTase activity are highly reminiscent of the TrxG-related H3K4-HMTase MLL1. Using ChIP-on-chip analyses, Guenther and colleagues found that MLL1 associates with the 5′ portion of approximately 90% of Pol II transcribed genes, correlating with H3K4 trimethylation (13). Together with our results, this suggests that MLL1 and ASH1L may cooccupy active genes to regulate their expression. Alternatively, ASH1L and MLL1 might occupy distinct subsets of transcribed regions in mammalian chromatin. To investigate this directly, we examined by ChIP the occupancy of MLL1 and ASH1L at the PABPC1 gene in HEK-293T cells. We found that MLL1 and ASH1L both not only associate with this gene but also show very similar patterns of occupancy that closely match the distribution of H3K4 trimethylation (Fig. (Fig.6A;6A; also, data not shown). Similar results were observed upon profiling of a second active gene, RPLP0 (data not shown). Furthermore, MLL1 and H3K4 trimethylation were found at all active genes examined (Fig. 6B and C), similarly to what we observed for ASH1L. This finding suggests that MLL1 and ASH1L, two Trithorax-related HMTases specific for H3K4 methylation, exist in parallel with the “on” state of transcription in mammalian cells, suggesting that these two enzymes may cooperate to methylate H3K4 at actively transcribed regions.

FIG. 6.
Overlapping chromatin occupancies of ASH1L and MLL1 at housekeeping genes. (A) ChIP with anti-ASH1L (296), anti-MLL1 (antibody 456), and control rabbit immunoglobulin G (rab Ig) at PABPC1 in 293T cells. (B and C) ChIP analysis, with 293T cells, of MLL1 ...

Critical downstream target genes of Ash1 and Trx in Drosophila are homeotic genes. For example, Ash1 and Trx genes are required individually for high-level expression of the homeotic gene Ubx (28, 42, 53). Interestingly, despite MLL1 being coupled in a generic fashion to the transcription apparatus (13), its critical downstream target genes in mammals appear to also be homeotic genes, particularly of the HoxA cluster (61, 62). Based on our findings that ASH1L and MLL1 seem to act in parallel in mammalian cells (Fig. (Fig.6A),6A), we hypothesized that mammalian HOXA genes might uniquely require ASH1L for their high-level expression. In addition, previous data suggest that Hox genes display greater sensitivity to levels of histone methylation for their expression than constitutively active genes (65). To develop this hypothesis further, we asked first whether parallel ASH1L/MLL1 occupancy occurs at active genes in the HoxA cluster and whether this distribution matches that of H3K4 methylation. Using reverse transcriptase PCR (RT-PCR), we identified HOXA10 as an actively expressed HoxA gene in HEK-293T cells (data not shown). We next examined, by using primers that amplify several regions across the gene, whether MLL1 and ASH1L associate with the transcribed region of HOXA10. We detected ASH1L occupancy across the entire ~4 kb of the transcribed region of HOXA10, with little occupancy upstream (−2.5 kb) or downstream (+8.6 kb) with respect to the TSS of the gene (Fig. (Fig.7A).7A). Signals at upstream and downstream regions of HOXA10 were similar to those at inactive CD4, as seen in Fig. Fig.6A.6A. Thus, the patterns of ASH1L occupancy at housekeeping versus Hox genes appear different; ASH1L distribution extends throughout the entire transcribed region of HOXA10 but is restricted to the 5′ portion of PABPC1. MLL1 also localizes to the transcribed region of HOXA10, but its pattern shows a higher similarity to H3K4 trimethylation than that of ASH1L (Fig. (Fig.7B).7B). Importantly, the association of both MLL1 and ASH1L across HOXA10 broadly encompasses regions containing H3K4 di- and trimethylation (Fig. 7A and C), consistent with these enzymes playing a role in maintaining steady-state levels of these modifications at this gene. These findings suggest that like PABPC1, active HoxA genes associate with both ASH1L and MLL1, representing dual activities capable of methylating H3K4 and correlating with the active state of transcription.

Our ChIP analyses combined with our biochemical data suggest that ASH1L could be required for H3K4 methylation and gene transcription at a potentially large number of active genes, albeit with a significant degree of redundancy with MLL1. To examine the contribution of ASH1L to H3K4 methylation in vivo, we attempted to generate ASH1L knockdown cell lines by using shRNA. Briefly, HEK-293T cells were transfected with a shASH1L vector, which contains green fluorescent protein and puromycin resistance genes. To obtain pure populations of cells expressing shASH1L, cells were selected with puromycin for 2 weeks prior to analysis. Following selection, we measured ASH1L mRNA levels by real-time RT-PCR with two distinct primer pairs and found ASH1L transcripts to be consistently reduced approximately 10-fold compared to levels for control cells (Fig. (Fig.7D).7D). We also examined cell populations for ASH1L protein levels by Western blotting of nuclear extracts. However, this analysis was inconclusive since the three ASH1L antibodies used for ChIP analyses worked poorly in Western blots, perhaps due to the large predicted size of ASH1L (330 kDa). We therefore examined the protein knockdown efficiency by ChIP at the two housekeeping genes RPLP0 and PABPC1. Moreover, we examined HOXA6 and HOXA10 because of the previously reported sensitivity of HoxA genes to HMTase depletion (30, 61, 62). We consistently observed a reduction in ASH1L occupancy at HOXA10 and HOXA6 by ~50% and 35%, respectively (Fig. (Fig.7F).7F). Surprisingly, ASH1L occupancy at the highly transcribed PABPC1 and RPLP0 genes remained unchanged in the knockdown cell lines. This suggests that despite the substantial depletion of ASH1L mRNA, protein levels are not reduced at all genes to the same extent. Moreover, this indicates that ASH1L protein is limiting at some genes but not others. It also remains possible that changes in translation efficiency might contribute to compensation in protein levels. We observed a similar discrepancy between mRNA and protein levels when attempting to knock down ASH1L by using small interfering RNA oligonucleotides (not shown). Finally, a more dramatic depletion of ASH1L might be incompatible with growth or survival of the cell lines examined. Consistent with this, we found that in prolonged culture of shASH1L-transfected cells, but not shCTRL-transfected cells, green fluorescent protein expression and ASH1L depletion were lost (not shown).

We next measured by Western blotting with modification-specific antibodies whether ASH1L depletion leads to reduced histone methylation in bulk preparations of acid-extracted histones. No measurable differences in bulk H3K4me2/3, H3K9me2/3, H3K36me3, or H4K20me1/2/3 levels were observed (data not shown). While these findings contrast with those for Drosophila, where loss of Ash1 is associated with global reduction in H3K4 methylation at polytene chromosomes (4), they agree with a higher degree of redundancy among H3K4 HMTases in mammalian cells (11, 15, 30, 61).

To examine the effects of ASH1L reduction on gene expression, we performed quantitative real-time RT-PCR. The levels of both HOXA6 and HOXA10 mRNA were reduced modestly compared to levels of the GAPDH (glyceraldehyde-3-phosphate dehydrogenase) housekeeping gene (Fig. (Fig.7E).7E). Thus, ASH1L is required for maximal expression of select Hox genes. Reduced expression of HOXA6 and HOXA10 might result from diminished H3K4 methylation in ASH1L knockdown cells. Indeed, H3K4 trimethylation was reduced commensurately with the reduced expression of both Hox genes (in particular in HOXA10) while H3K4 trimethylation was unchanged at the other genes examined (Fig. (Fig.7G).7G). In contrast, dimethylation of H3K4 at these genes was not affected by the reduced ASH1L levels (see Fig. S5A in the supplemental material). The reduction in H3K4me3 was not a consequence of altered histone density, as total histone H3 levels were unchanged at these genes (see Fig. S5B in the supplemental material). In concert, these data suggest that ASH1L is required for maximal expression and H3K4 methylation of select Hox genes. To determine definitively whether ASH1L also plays an essential role at other genes requires a more dramatic knockdown or a knockout via gene targeting.

ASH1L's association with chromatin is MLL1 independent.

In Drosophila larvae containing a mutation in ash1, binding of Trx to chromosomes is reduced significantly (24). In addition, it has been observed that Drosophila Ash1 and Trx can physically interact with each other (42), suggesting a mutual requirement for their recruitment and/or activity. Combined with our finding that mammalian ASH1L and MLL1 have similar distribution profiles across active chromatin, we investigated the possible codependence of ASH1L and MLL1 association at active genes. To examine if MLL1 is required for ASH1L occupancy, we performed experiments with MEFs that are homozygous for a targeted deletion of Mll1 (62). Consistent with previous observations (30), the complete depletion of MLL1 protein leads to significantly reduced expression of HoxA9 mRNA (Fig. (Fig.8A).8A). Expression of an unrelated homeotic gene, Meis1, and that of the housekeeping gene TubA2 (tubulin, alpha 2) were not affected in Mll1−/− cells. We next measured by ChIP ASH1L and MLL1 occupancy at these genes. While MLL1 can be found at all three genes in wild-type MEFs, there was no detectable MLL1 at any of the analyzed genes in the Mll1−/− MEFs (Fig. 8B and C). In contrast, ASH1L was found at normal levels at the promoter and 5′-transcribed regions of Meis1 and TubA2 in Mll1−/− cells (Fig. (Fig.8B).8B). This demonstrates that despite the virtually identical distributions of MLL1 and ASH1L, their recruitments can occur independently of each other.

FIG. 8.
ASH1L occupancy is independent of MLL1. (A) Real-time RT-PCR analysis of HoxA9, Meis1, and TubA2 mRNA expression in Mll1+/+ and Mll1−/− MEFs. (B and C) ChIP analysis of ASH1L (296), MLL1 (456), H3K4me3, and H3K4me2 at (B) ...

Remarkably, in Mll1−/− MEFs, we observed only moderately reduced ASH1L levels at the MLL1-dependent gene HoxA9 (Fig. (Fig.8C).8C). Thus, despite the dramatic loss of HoxA9 transcription, ASH1L levels were maintained at an intermediate degree. This suggests that ASH1L can also be present, albeit at reduced levels, at inactive genes and that ASH1L by itself is not sufficient for gene activation. However, it remains possible that residual transcriptional activity accounts for the remaining ASH1L.

To explore the contribution of MLL1 and ASH1L to H3K4 di- and trimethylation, we performed ChIP at the TubA2, Meis1, and HoxA9 genes in wild-type and Mll1−/− MEFs. Two findings emerged from these experiments. First, in Mll1−/− cells both di-and trimethylation of H3K4 were moderately reduced at TubA2 and Meis1 despite normal levels of transcription (Fig. (Fig.8B,8B, bottom). Thus, MLL1 likely contributes to H3K4 methylation at these genes but compensation by other H3K4 HMTases, perhaps including ASH1L, is sufficient to allow for maximal gene expression. Moreover, the correlation between gene activity and H3K4 methylation is not absolute. Second, in Mll1−/− cells, there is a dramatic loss of H3K4 trimethylation at the MLL1-dependent gene HoxA9 and a more moderate reduction in H3K4 dimethylation (Fig. (Fig.8C,8C, bottom). Thus, despite the persistence of ASH1L at the HoxA9 gene in Mll1−/− cells, H3K4 methylation is lower, indicating that ASH1L cannot compensate for the loss of MLL1 but might be responsible for the residual H3K4 methylation.

DISCUSSION

Here we report that ASH1L is an HMTase that associates with all active genes examined. Association of ASH1L with active genes occurs irrespective of cell type. ASH1L is found at housekeeping genes and at tightly regulated tissue-specific genes, suggesting a general role in transcription. Its pattern of occupancy closely matches that of MLL1, but recruitments of these enzymes occur independently of each other. ASH1L is required for the full expression of select Hox genes, indicating that in spite of substantial overlap in function with MLL1 and presumably other H3K4 HMTases, ASH1L is required for normal gene expression.

ASH1L is likely recruited to active genes by mechanisms involving early elongating Pol II. This is supported by its peak localization to the 5′-transcribed portion of active genes and its dependence on active transcription. Moreover, at the β-major globin gene in erythroid cells, ASH1L is inducibly present at high levels near the active β-major globin gene transcribed region. Together, these findings indicate that the elongating form of Pol II plays a role in depositing ASH1L. Therefore, we found it surprising that in the presence of the elongation blocker DRB ASH1L binding persisted beyond the presence of Pol II and H3K4 methylation. In the absence of Pol II, ASH1L might be retained at the gene for extended periods of time by interacting with chromatin, perhaps via contacts with DRB-insensitive histone modifications, such as H3K9 acetylation (unpublished observation), through its bromodomain. The presence of ASH1L along with the loss of H3K4 methylation we interpret to suggest that the enzymatic activity of ASH1L is modulated by cofactors that are involved in transcription elongation or by additional histone modifications. ASH1L was also observed at the distal LCR, suggesting distinct modes of recruitment of this enzyme. However, full activity of ASH1L might require elongating Pol II since H3K4me3 levels are low at the LCR. In this regard, it is relevant that H3K4 methylation is coordinated by elongation factors, such as the Paf1 complex (23, 25, 33, 64), and is influenced by ubiquitylation of histone H2B (19, 46, 47, 51, 65). Furthermore, the LCR contains nonelongating Pol II (16) and would therefore not contain the required cofactors for placing H3K4me3.

Using in vitro HMTase assays, we demonstrated that the extended SET domain of mammalian ASH1L methylates H3K4. The activity required sequences flanking the core SET domain, similarly to what has been observed for SUV39H1 and SET7/9 (39, 57), but remained weak compared to that of other HMTases. In an attempt to boost activity and to better decipher the target sites and methylation states, we screened numerous conditions for the in vitro HMTase assays; however, specific activity remained too low for reliable Western blot analysis. Furthermore, overall activities varied considerably between different protein preparations for reasons that are unclear at this time. Nevertheless, using a combination of histone tail peptides with premethylated lysine residues and recombinant wild-type and mutant histones, we were able to pinpoint the major target site for ASH1L methylation to H3K4. The site specificity is consistent with the distribution pattern of ASH1L along active genes. Moreover, we found that premethylation of H3K9 reduced the activity of ASH1L towards H3K4, suggesting a degree of cross-regulation among these marks. Given that methylation of H3K4 and H3K9 can be resolved into distinct profiles at a large housekeeping gene (55), we speculate that modulation of ASH1L activity by H3K9 methylation might contribute to the correct partitioning of these marks along transcribed chromatin.

The in vivo target site specificity of Drosophila Ash1 has been the subject of controversy. Two reports using in vitro and in vivo experimentation suggested that Ash1 can trimethylate H3K4, H3K9, and H4K20 (3, 43). However, other reports did not support these conclusions, providing evidence that Ash1 is H3K4 selective in vivo (4, 37, 38). While our work agrees with the latter reports, we cannot rule out that in vivo or under conditions in vitro where ASH1L is complexed with other molecules ASH1L might target additional lysine residues. Several approaches to resolve this issue were unsuccessful in our hands. First, we were unable to obtain full-length tagged Ash1 for expression and purification in mammalian cells. Second, we attempted to determine ASH1L activity in vivo via the targeting of its extended SET domain to a stably integrated reporter gene by using the Gal4 DNA binding domain and also via recruitment to the endogenous VEGF gene by using engineered, sequence-specific zinc finger proteins (48). However, our ChIP experiments with modification-specific histone antibodies failed to detect histone methylation at the target genes despite successful expression of the fusion proteins (data not shown). It is clear from these experiments and from our in vitro data that ASH1L depends on additional cofactors for its activity. Similar cofactor requirements for increased activity or processivity were reported for other SET domain-containing enzymes, such as SET1, MLL1, ESET, and EZH2 (5, 7, 41, 49, 56). Finally, considering the weak activity of ASH1L towards histone H3, it is possible that it may also have nonhistone targets, as is the case for SET7/9 and SMYD2, which regulate the activity of the tumor suppressor p53 (6, 14) and, in the case of SET7/9, the general transcription factor TAF10 (21).

Of note, while the manuscript was under review, Tanaka and colleagues reported that both mammalian ASH1L and Drosophila Ash1 can mono- and dimethylate H3K36, but not H3K4, H3K9, or H4K20, in vitro (52). The reason for this discrepancy with our results is unknown but likely relates to the conditions used for the in vitro assay. To address a possible role for ASH1L in H3K36 methylation, we examined the spatial distribution of mono-, di-, and trimethylation of H3K36 across PABPC1 (see Fig. S2 in the supplemental material). We found that the distribution of ASH1L is generally dissimilar from that of methylated H3K36, although some overlap does exist. Of particular relevance is the significant drop in H3K36 dimethylation 0.5 kb downstream from the TSS of PABPC1, where ASH1L and H3K4me3 levels peak, suggesting that ASH1L is not the major H3K36 methyltransferase at these genes. Nevertheless, as stated above, our results do not exclude a role of ASH1L in H3K36 methylation.

MLL1 and ASH1L occupancies at 5′ regions of active genes are nearly indistinguishable, which suggests redundancy between ASH1L and MLL1 function. Yet, in vivo depletion of either enzyme results in diminished methylation of H3K4 at select active HoxA genes (this report and reference 30). This indicates that while both enzymes are present at most active genes, their activity is limiting only at a fraction of them. While the effects of reduced ASH1L levels were subtle, a broader role cannot be excluded at this time given the limitations of the knockdown experiments. In addition, a more dramatic role of ASH1L might be revealed in the context of knockout animals at stages of development where TrxG target genes are critical for maintenance of cell fate. Future studies that combine ChIP-on-chip microarray analysis with constitutive or conditional gene targeting might reveal the full complement of genes that require ASH1L for their expression.

In Drosophila, Trx fails to accumulate on polytene chromosomes in animals containing a mutation in ash1 (24), and Ash1 and Trx interact physically in Drosophila embryos (42), reflecting a tight interdependence of these molecules. Our results obtained using Mll1−/− MEFs suggest that ASH1L does not require MLL1 for its recruitment to MLL1-independent genes. Even at the MLL1-dependent HoxA9 gene, ASH1L levels were sustained at intermediate levels although HoxA9 expression was extinguished. Therefore, we conclude that ASH1L recruitment is independent of MLL1. In the converse experiment with ASH1L knockdown cells, we found that the MLL1 levels near active gene promoters were essentially normal as measured by ChIP (not shown), but a definitive answer will have to await the complete ablation of ASH1L. Nevertheless, together with the observations that none of the biochemical purifications of the MLL1 complex detected ASH1L (7, 8, 32, 36), the evidence weighs in favor of these two proteins being recruited and perhaps regulated independently of each other. In any case, the identification of the complete ASH1L complex will further our understanding of ASH1L's function(s).

In summary, while initially ASH1L might have been a suitable candidate enzyme for methylation of H3K4, H3K9, and H4K20, the work presented here suggests that ASH1L is an HMTase specific for H3K4. This is supported by its H3K4-specific in vitro activity, its colocalization with methylated H3K4 along transcribed genes, and its similarity to other MLL family members. A rewarding challenge will be to identify the enzymes that mediate methylation of H3K9, H3K36, and H4K20 in the coding region of active genes and to elucidate the mechanisms by which they place these marks in a spatially distinct manner.

Supplementary Material

[Supplemental material]

Acknowledgments

This work was supported by NIH grants DK58044 and DK54937 and an American Cancer Society institutional research grant (5P30CA016520-32) to G.A.B. and grants from the Israel Cancer Research Fund and US-Israel BSF to E.C. G.D.G. was supported by NIH training grant T32 HL07439-27. C.R.V. was supported by NIH training grant T32 HL07439-26.

We thank members of the Blobel laboratory for insightful discussion and Hongxin Wang and Keshet Ronen for their technical expertise.

Footnotes

[down-pointing small open triangle]Published ahead of print on 8 October 2007.

Supplemental material for this article may be found at http://mcb.asm.org/.

REFERENCES

1. Bannister, A. J., R. Schneider, F. A. Myers, A. W. Thorne, C. Crane-Robinson, and T. Kouzarides. 2005. Spatial distribution of di- and tri-methyl lysine 36 of histone H3 at active genes. J. Biol. Chem. 280:17732-17736. [PubMed]
2. Barski, A., S. Cuddapah, K. Cui, T. Y. Roh, D. E. Schones, Z. Wang, G. Wei, I. Chepelev, and K. Zhao. 2007. High-resolution profiling of histone methylations in the human genome. Cell 129:823-837. [PubMed]
3. Beisel, C., A. Imhof, J. Greene, E. Kremmer, and F. Sauer. 2002. Histone methylation by the Drosophila epigenetic transcriptional regulator Ash1. Nature 419:857-862. [PubMed]
4. Byrd, K. N., and A. Shearn. 2003. ASH1, a Drosophila trithorax group protein, is required for methylation of lysine 4 residues on histone H3. Proc. Natl. Acad. Sci. USA 100:11535-11540. [PMC free article] [PubMed]
5. Cao, R., and Y. Zhang. 2004. SUZ12 is required for both the histone methyltransferase activity and the silencing function of the EED-EZH2 complex. Mol. Cell 15:57-67. [PubMed]
6. Chuikov, S., J. K. Kurash, J. R. Wilson, B. Xiao, N. Justin, G. S. Ivanov, K. McKinney, P. Tempst, C. Prives, S. J. Gamblin, N. A. Barlev, and D. Reinberg. 2004. Regulation of p53 activity through lysine methylation. Nature 432:353-360. [PubMed]
7. Dou, Y., T. A. Milne, A. J. Ruthenburg, S. Lee, J. W. Lee, G. L. Verdine, C. D. Allis, and R. G. Roeder. 2006. Regulation of MLL1 H3K4 methyltransferase activity by its core components. Nat. Struct. Mol. Biol. 13:713-719. [PubMed]
8. Dou, Y., T. A. Milne, A. J. Tackett, E. R. Smith, A. Fukuda, J. Wysocka, C. D. Allis, B. T. Chait, J. L. Hess, and R. G. Roeder. 2005. Physical association and coordinate function of the H3 K4 methyltransferase MLL1 and the H4 K16 acetyltransferase MOF. Cell 121:873-885. [PubMed]
9. Eissenberg, J. C., and A. Shilatifard. 2006. Leaving a mark: the many footprints of the elongating RNA polymerase II. Curr. Opin. Genet. Dev. 16:184-190. [PubMed]
10. Eissenberg, J. C., A. Shilatifard, N. Dorokhov, and D. E. Michener. 2007. Cdk9 is an essential kinase in Drosophila that is required for heat shock gene expression, histone methylation and elongation factor recruitment. Mol. Genet. Genomics 277:101-114. [PubMed]
11. Glaser, S., J. Schaft, S. Lubitz, K. Vintersten, F. van der Hoeven, K. R. Tufteland, R. Aasland, K. Anastassiadis, S. L. Ang, and A. F. Stewart. 2006. Multiple epigenetic maintenance factors implicated by the loss of Mll2 in mouse development. Development 133:1423-1432. [PubMed]
12. Gu, Y., T. Nakamura, H. Alder, R. Prasad, O. Canaani, G. Cimino, C. M. Croce, and E. Canaani. 1992. The t(4;11) chromosome translocation of human acute leukemias fuses the ALL-1 gene, related to Drosophila trithorax, to the AF-4 gene. Cell 71:701-708. [PubMed]
13. Guenther, M. G., R. G. Jenner, B. Chevalier, T. Nakamura, C. M. Croce, E. Canaani, and R. A. Young. 2005. Global and Hox-specific roles for the MLL1 methyltransferase. Proc. Natl. Acad. Sci. USA 102:8603-8608. [PMC free article] [PubMed]
14. Huang, J., L. Perez-Burgos, B. J. Placek, R. Sengupta, M. Richter, J. A. Dorsey, S. Kubicek, S. Opravil, T. Jenuwein, and S. L. Berger. 2006. Repression of p53 activity by Smyd2-mediated methylation. Nature 444:629-632. [PubMed]
15. Issaeva, I., Y. Zonis, T. Rozovskaia, K. Orlovsky, C. M. Croce, T. Nakamura, A. Mazo, L. Eisenbach, and E. Canaani. 2007. Knockdown of ALR (MLL2) reveals ALR target genes and leads to alterations in cell adhesion and growth. Mol. Cell. Biol. 27:1889-1903. [PMC free article] [PubMed]
16. Johnson, K. D., J. A. Grass, C. Park, H. Im, K. Choi, and E. H. Bresnick. 2003. Highly restricted localization of RNA polymerase II within a locus control region of a tissue-specific chromatin domain. Mol. Cell. Biol. 23:6484-6493. [PMC free article] [PubMed]
17. Jones, P. A., and S. B. Baylin. 2007. The epigenomics of cancer. Cell 128:683-692. [PMC free article] [PubMed]
18. Kim, A., C. M. Kiefer, and A. Dean. 2007. Distinctive signatures of histone methylation in transcribed coding and noncoding human beta-globin sequences. Mol. Cell. Biol. 27:1271-1279. [PMC free article] [PubMed]
19. Kim, J., S. B. Hake, and R. G. Roeder. 2005. The human homolog of yeast BRE1 functions as a transcriptional coactivator through direct activator interactions. Mol. Cell 20:759-770. [PubMed]
20. Kizer, K. O., H. P. Phatnani, Y. Shibata, H. Hall, A. L. Greenleaf, and B. D. Strahl. 2005. A novel domain in Set2 mediates RNA polymerase II interaction and couples histone H3 K36 methylation with transcript elongation. Mol. Cell. Biol. 25:3305-3316. [PMC free article] [PubMed]
21. Kouskouti, A., E. Scheer, A. Staub, L. Tora, and I. Talianidis. 2004. Gene-specific modulation of TAF10 function by SET9-mediated methylation. Mol. Cell 14:175-182. [PubMed]
22. Kouzarides, T. 2007. Chromatin modifications and their function. Cell 128:693-705. [PubMed]
23. Krogan, N. J., J. Dover, A. Wood, J. Schneider, J. Heidt, M. A. Boateng, K. Dean, O. W. Ryan, A. Golshani, M. Johnston, J. F. Greenblatt, and A. Shilatifard. 2003. The Paf1 complex is required for histone H3 methylation by COMPASS and Dot1p: linking transcriptional elongation to histone methylation. Mol. Cell 11:721-729. [PubMed]
24. Kuzin, B., S. Tillib, Y. Sedkov, L. Mizrokhi, and A. Mazo. 1994. The Drosophila trithorax gene encodes a chromosomal protein and directly regulates the region-specific homeotic gene fork head. Genes Dev. 8:2478-2490. [PubMed]
25. Laribee, R. N., N. J. Krogan, T. Xiao, Y. Shibata, T. R. Hughes, J. F. Greenblatt, and B. D. Strahl. 2005. BUR kinase selectively regulates H3 K4 trimethylation and H2B ubiquitylation through recruitment of the PAF elongation complex. Curr. Biol. 15:1487-1493. [PubMed]
26. Lee, S., D. K. Lee, Y. Dou, J. Lee, B. Lee, E. Kwak, Y. Y. Kong, S. K. Lee, R. G. Roeder, and J. W. Lee. 2006. Coactivator as a target gene specificity determinant for histone H3 lysine 4 methyltransferases. Proc. Natl. Acad. Sci. USA 103:15392-15397. [PMC free article] [PubMed]
27. Li, B., M. Carey, and J. L. Workman. 2007. The role of chromatin during transcription. Cell 128:707-719. [PubMed]
28. Mazo, A. M., D. H. Huang, B. A. Mozer, and I. B. Dawid. 1990. The trithorax gene, a trans-acting regulator of the bithorax complex in Drosophila, encodes a protein with zinc-binding domains. Proc. Natl. Acad. Sci. USA 87:2112-2116. [PMC free article] [PubMed]
29. Mikkelsen, T. S., M. Ku, D. B. Jaffe, B. Issac, E. Lieberman, G. Giannoukos, P. Alvarez, W. Brockman, T. K. Kim, R. P. Koche, W. Lee, E. Mendenhall, A. O'Donovan, A. Presser, C. Russ, X. Xie, A. Meissner, M. Wernig, R. Jaenisch, C. Nusbaum, E. S. Lander, and B. E. Bernstein. 2007. Genome-wide maps of chromatin state in pluripotent and lineage-committed cells. Nature 448:553-560. [PMC free article] [PubMed]
30. Milne, T. A., Y. Dou, M. E. Martin, H. W. Brock, R. G. Roeder, and J. L. Hess. 2005. MLL associates specifically with a subset of transcriptionally active target genes. Proc. Natl. Acad. Sci. USA 102:14765-14770. [PMC free article] [PubMed]
31. Nakamura, T., J. Blechman, S. Tada, T. Rozovskaia, T. Itoyama, F. Bullrich, A. Mazo, C. M. Croce, B. Geiger, and E. Canaani. 2000. huASH1 protein, a putative transcription factor encoded by a human homologue of the Drosophila ash1 gene, localizes to both nuclei and cell-cell tight junctions. Proc. Natl. Acad. Sci. USA 97:7284-7289. [PMC free article] [PubMed]
32. Nakamura, T., T. Mori, S. Tada, W. Krajewski, T. Rozovskaia, R. Wassell, G. Dubois, A. Mazo, C. M. Croce, and E. Canaani. 2002. ALL-1 is a histone methyltransferase that assembles a supercomplex of proteins involved in transcriptional regulation. Mol. Cell 10:1119-1128. [PubMed]
33. Ng, H. H., F. Robert, R. A. Young, and K. Struhl. 2003. Targeted recruitment of Set1 histone methylase by elongating Pol II provides a localized mark and memory of recent transcriptional activity. Mol. Cell 11:709-719. [PubMed]
34. Nishioka, K., and D. Reinberg. 2003. Methods and tips for the purification of human histone methyltransferases. Methods 31:49-58. [PubMed]
35. O'Geen, H., S. L. Squazzo, S. Iyengar, K. Blahnik, J. L. Rinn, H. Y. Chang, R. Green, and P. J. Farnham. 2007. Genome-wide analysis of KAP1 binding suggests autoregulation of KRAB-ZNFs. PLoS Genet. 3:e89. [PMC free article] [PubMed]
36. Papoulas, O., S. J. Beek, S. L. Moseley, C. M. McCallum, M. Sarte, A. Shearn, and J. W. Tamkun. 1998. The Drosophila trithorax group proteins BRM, ASH1 and ASH2 are subunits of distinct protein complexes. Development 125:3955-3966. [PubMed]
37. Papp, B., and J. Muller. 2006. Histone trimethylation and the maintenance of transcriptional ON and OFF states by trxG and PcG proteins. Genes Dev. 20:2041-2054. [PMC free article] [PubMed]
38. Petruk, S., Y. Sedkov, K. M. Riley, J. Hodgson, F. Schweisguth, S. Hirose, J. B. Jaynes, H. W. Brock, and A. Mazo. 2006. Transcription of bxd noncoding RNAs promoted by trithorax represses Ubx in cis by transcriptional interference. Cell 127:1209-1221. [PMC free article] [PubMed]
39. Rea, S., F. Eisenhaber, D. O'Carroll, B. D. Strahl, Z. W. Sun, M. Schmid, S. Opravil, K. Mechtler, C. P. Ponting, C. D. Allis, and T. Jenuwein. 2000. Regulation of chromatin structure by site-specific histone H3 methyltransferases. Nature 406:593-599. [PubMed]
40. Rice, J. C., S. D. Briggs, B. Ueberheide, C. M. Barber, J. Shabanowitz, D. F. Hunt, Y. Shinkai, and C. D. Allis. 2003. Histone methyltransferases direct different degrees of methylation to define distinct chromatin domains. Mol. Cell 12:1591-1598. [PubMed]
41. Roguev, A., D. Schaft, A. Shevchenko, W. W. Pijnappel, M. Wilm, R. Aasland, and A. F. Stewart. 2001. The Saccharomyces cerevisiae Set1 complex includes an Ash2 homologue and methylates histone 3 lysine 4. EMBO J. 20:7137-7148. [PMC free article] [PubMed]
42. Rozovskaia, T., S. Tillib, S. Smith, Y. Sedkov, O. Rozenblatt-Rosen, S. Petruk, T. Yano, T. Nakamura, L. Ben-Simchon, J. Gildea, C. M. Croce, A. Shearn, E. Canaani, and A. Mazo. 1999. Trithorax and ASH1 interact directly and associate with the trithorax group-responsive bxd region of the Ultrabithorax promoter. Mol. Cell. Biol. 19:6441-6447. [PMC free article] [PubMed]
43. Sanchez-Elsner, T., D. Gou, E. Kremmer, and F. Sauer. 2006. Noncoding RNAs of trithorax response elements recruit Drosophila Ash1 to Ultrabithorax. Science 311:1118-1123. [PubMed]
44. Sawado, T., J. Halow, M. A. Bender, and M. Groudine. 2003. The beta-globin locus control region (LCR) functions primarily by enhancing the transition from transcription initiation to elongation. Genes Dev. 17:1009-1018. [PMC free article] [PubMed]
45. Schneider, R., A. J. Bannister, F. A. Myers, A. W. Thorne, C. Crane-Robinson, and T. Kouzarides. 2004. Histone H3 lysine 4 methylation patterns in higher eukaryotic genes. Nat. Cell Biol. 6:73-77. [PubMed]
46. Shahbazian, M. D., K. Zhang, and M. Grunstein. 2005. Histone H2B ubiquitylation controls processive methylation but not monomethylation by Dot1 and Set1. Mol. Cell 19:271-277. [PubMed]
47. Shilatifard, A. 2006. Chromatin modifications by methylation and ubiquitination: implications in the regulation of gene expression. Annu. Rev. Biochem. 75:243-269. [PubMed]
48. Snowden, A. W., P. D. Gregory, C. C. Case, and C. O. Pabo. 2002. Gene-specific targeting of H3K9 methylation is sufficient for initiating repression in vivo. Curr. Biol. 12:2159-2166. [PubMed]
49. Steward, M. M., J. S. Lee, A. O'Donovan, M. Wyatt, B. E. Bernstein, and A. Shilatifard. 2006. Molecular regulation of H3K4 trimethylation by ASH2L, a shared subunit of MLL complexes. Nat. Struct. Mol. Biol. 13:852-854. [PubMed]
50. Su, A. I., M. P. Cooke, K. A. Ching, Y. Hakak, J. R. Walker, T. Wiltshire, A. P. Orth, R. G. Vega, L. M. Sapinoso, A. Moqrich, A. Patapoutian, G. M. Hampton, P. G. Schultz, and J. B. Hogenesch. 2002. Large-scale analysis of the human and mouse transcriptomes. Proc. Natl. Acad. Sci. USA 99:4465-4470. [PMC free article] [PubMed]
51. Sun, Z. W., and C. D. Allis. 2002. Ubiquitination of histone H2B regulates H3 methylation and gene silencing in yeast. Nature 418:104-108. [PubMed]
52. Tanaka, Y., Z. Katagiri, K. Kawahashi, D. Kioussis, and S. Kitajima. 2007. Trithorax-group protein ASH1 methylates histone H3 lysine 36. Gene 397:161-168. [PubMed]
53. Tripoulas, N. A., E. Hersperger, D. La Jeunesse, and A. Shearn. 1994. Molecular genetic analysis of the Drosophila melanogaster gene absent, small or homeotic discs1 (ash1). Genetics 137:1027-1038. [PMC free article] [PubMed]
54. Vakoc, C. R., S. A. Mandat, B. A. Olenchock, and G. A. Blobel. 2005. Histone H3 lysine 9 methylation and HP1gamma are associated with transcription elongation through mammalian chromatin. Mol. Cell 19:381-391. [PubMed]
55. Vakoc, C. R., M. M. Sachdeva, H. Wang, and G. A. Blobel. 2006. Profile of histone lysine methylation across transcribed mammalian chromatin. Mol. Cell. Biol. 26:9185-9195. [PMC free article] [PubMed]
56. Wang, H., W. An, R. Cao, L. Xia, H. Erdjument-Bromage, B. Chatton, P. Tempst, R. G. Roeder, and Y. Zhang. 2003. mAM facilitates conversion by ESET of dimethyl to trimethyl lysine 9 of histone H3 to cause transcriptional repression. Mol. Cell 12:475-487. [PubMed]
57. Wang, H., R. Cao, L. Xia, H. Erdjument-Bromage, C. Borchers, P. Tempst, and Y. Zhang. 2001. Purification and functional characterization of a histone H3-lysine 4-specific methyltransferase. Mol. Cell 8:1207-1217. [PubMed]
58. Weiss, M. J., C. Yu, and S. H. Orkin. 1997. Erythroid-cell-specific properties of transcription factor GATA-1 revealed by phenotypic rescue of a gene-targeted cell line. Mol. Cell. Biol. 17:1642-1651. [PMC free article] [PubMed]
59. Welch, J. J., J. A. Watts, C. R. Vakoc, Y. Yao, H. Wang, R. C. Hardison, G. A. Blobel, L. A. Chodosh, and M. J. Weiss. 2004. Global regulation of erythroid gene expression by transcription factor GATA-1. Blood 104:3136-3147. [PubMed]
60. Wysocka, J., T. Swigut, H. Xiao, T. A. Milne, S. Y. Kwon, J. Landry, M. Kauer, A. J. Tackett, B. T. Chait, P. Badenhorst, C. Wu, and C. D. Allis. 2006. A PHD finger of NURF couples histone H3 lysine 4 trimethylation with chromatin remodelling. Nature 442:86-90. [PubMed]
61. Yu, B. D., R. D. Hanson, J. L. Hess, S. E. Horning, and S. J. Korsmeyer. 1998. MLL, a mammalian trithorax-group gene, functions as a transcriptional maintenance factor in morphogenesis. Proc. Natl. Acad. Sci. USA 95:10632-10636. [PMC free article] [PubMed]
62. Yu, B. D., J. L. Hess, S. E. Horning, G. A. Brown, and S. J. Korsmeyer. 1995. Altered Hox expression and segmental identity in Mll-mutant mice. Nature 378:505-508. [PubMed]
63. Zeng, P. Y., C. R. Vakoc, Z. C. Chen, G. A. Blobel, and S. L. Berger. 2006. In vivo dual cross-linking for identification of indirect DNA-associated proteins by chromatin immunoprecipitation. BioTechniques 41:694-698. [PubMed]
64. Zhu, B., S. S. Mandal, A. D. Pham, Y. Zheng, H. Erdjument-Bromage, S. K. Batra, P. Tempst, and D. Reinberg. 2005. The human PAF complex coordinates transcription with events downstream of RNA synthesis. Genes Dev. 19:1668-1673. [PMC free article] [PubMed]
65. Zhu, B., Y. Zheng, A. D. Pham, S. S. Mandal, H. Erdjument-Bromage, P. Tempst, and D. Reinberg. 2005. Monoubiquitination of human histone H2B: the factors involved and their roles in HOX gene regulation. Mol. Cell 20:601-611. [PubMed]

Articles from Molecular and Cellular Biology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...