Article3 June 2002free access Essential function of histone deacetylase 1 in proliferation control and CDK inhibitor repression Gerda Lagger Gerda Lagger Institute of Medical Biochemistry, Division of Molecular Biology, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/2, A-1030 Vienna, Austria Search for more papers by this author Dónal O'Carroll Dónal O'Carroll Institute of Molecular Pathology, Vienna Biocenter, Dr Bohr-Gasse 7, A-1030 Vienna, Austria Present address: Laboratory for Lymphocyte Signaling, The Rockefeller University, 1230 York Avenue, New York, NY, 10021 USA Search for more papers by this author Martina Rembold Martina Rembold Institute of Medical Biochemistry, Division of Molecular Biology, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/2, A-1030 Vienna, Austria Search for more papers by this author Harald Khier Harald Khier Institute of Medical Biochemistry, Division of Molecular Biology, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/2, A-1030 Vienna, Austria Search for more papers by this author Julia Tischler Julia Tischler Institute of Medical Biochemistry, Division of Molecular Biology, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/2, A-1030 Vienna, Austria Search for more papers by this author Georg Weitzer Georg Weitzer Institute of Medical Biochemistry, Division of Biochemistry, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/3, A-1030 Vienna, Austria Search for more papers by this author Bernd Schuettengruber Bernd Schuettengruber Institute of Medical Biochemistry, Division of Molecular Biology, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/2, A-1030 Vienna, Austria Search for more papers by this author Christoph Hauser Christoph Hauser Institute of Medical Biochemistry, Division of Molecular Biology, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/2, A-1030 Vienna, Austria Search for more papers by this author Reinhard Brunmeir Reinhard Brunmeir Institute of Medical Biochemistry, Division of Molecular Biology, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/2, A-1030 Vienna, Austria Search for more papers by this author Thomas Jenuwein Thomas Jenuwein Institute of Molecular Pathology, Vienna Biocenter, Dr Bohr-Gasse 7, A-1030 Vienna, Austria Search for more papers by this author Christian Seiser Corresponding Author Christian Seiser Institute of Medical Biochemistry, Division of Molecular Biology, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/2, A-1030 Vienna, Austria Search for more papers by this author Gerda Lagger Gerda Lagger Institute of Medical Biochemistry, Division of Molecular Biology, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/2, A-1030 Vienna, Austria Search for more papers by this author Dónal O'Carroll Dónal O'Carroll Institute of Molecular Pathology, Vienna Biocenter, Dr Bohr-Gasse 7, A-1030 Vienna, Austria Present address: Laboratory for Lymphocyte Signaling, The Rockefeller University, 1230 York Avenue, New York, NY, 10021 USA Search for more papers by this author Martina Rembold Martina Rembold Institute of Medical Biochemistry, Division of Molecular Biology, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/2, A-1030 Vienna, Austria Search for more papers by this author Harald Khier Harald Khier Institute of Medical Biochemistry, Division of Molecular Biology, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/2, A-1030 Vienna, Austria Search for more papers by this author Julia Tischler Julia Tischler Institute of Medical Biochemistry, Division of Molecular Biology, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/2, A-1030 Vienna, Austria Search for more papers by this author Georg Weitzer Georg Weitzer Institute of Medical Biochemistry, Division of Biochemistry, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/3, A-1030 Vienna, Austria Search for more papers by this author Bernd Schuettengruber Bernd Schuettengruber Institute of Medical Biochemistry, Division of Molecular Biology, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/2, A-1030 Vienna, Austria Search for more papers by this author Christoph Hauser Christoph Hauser Institute of Medical Biochemistry, Division of Molecular Biology, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/2, A-1030 Vienna, Austria Search for more papers by this author Reinhard Brunmeir Reinhard Brunmeir Institute of Medical Biochemistry, Division of Molecular Biology, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/2, A-1030 Vienna, Austria Search for more papers by this author Thomas Jenuwein Thomas Jenuwein Institute of Molecular Pathology, Vienna Biocenter, Dr Bohr-Gasse 7, A-1030 Vienna, Austria Search for more papers by this author Christian Seiser Corresponding Author Christian Seiser Institute of Medical Biochemistry, Division of Molecular Biology, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/2, A-1030 Vienna, Austria Search for more papers by this author Author Information Gerda Lagger1, Dónal O'Carroll2,3, Martina Rembold1, Harald Khier1, Julia Tischler1, Georg Weitzer4, Bernd Schuettengruber1, Christoph Hauser1, Reinhard Brunmeir1, Thomas Jenuwein2 and Christian Seiser 1 1Institute of Medical Biochemistry, Division of Molecular Biology, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/2, A-1030 Vienna, Austria 2Institute of Molecular Pathology, Vienna Biocenter, Dr Bohr-Gasse 7, A-1030 Vienna, Austria 3Present address: Laboratory for Lymphocyte Signaling, The Rockefeller University, 1230 York Avenue, New York, NY, 10021 USA 4Institute of Medical Biochemistry, Division of Biochemistry, University of Vienna, Vienna Biocenter, Dr Bohr-Gasse 9/3, A-1030 Vienna, Austria *Corresponding author. E-mail: [email protected] The EMBO Journal (2002)21:2672-2681https://doi.org/10.1093/emboj/21.11.2672 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Histone deacetylases (HDACs) modulate chromatin structure and transcription, but little is known about their function in mammalian development. HDAC1 was implicated previously in the repression of genes required for cell proliferation and differentiation. Here we show that targeted disruption of both HDAC1 alleles results in embryonic lethality before E10.5 due to severe proliferation defects and retardation in development. HDAC1-deficient embryonic stem cells show reduced proliferation rates, which correlate with decreased cyclin-associated kinase activities and elevated levels of the cyclin-dependent kinase inhibitors p21WAF1/CIP1 and p27KIP1. Similarly, expression of p21 and p27 is up-regulated in HDAC1-null embryos. In addition, loss of HDAC1 leads to significantly reduced overall deacetylase activity, hyperacetylation of a subset of histones H3 and H4 and concomitant changes in other histone modifications. The expression of HDAC2 and HDAC3 is induced in HDAC1-deficient cells, but cannot compensate for loss of the enzyme, suggesting a unique function for HDAC1. Our study provides the first evidence that a histone deacetylase is essential for unrestricted cell proliferation by repressing the expression of selective cell cycle inhibitors. Introduction Controlled gene expression and cell proliferation are essential for all organisms to ensure their integrity and survival. During development from the fertilized egg to a multicellular organism, cell fate decisions are taken and cell lineage or tissue-specific gene expression patterns have to be established and maintained. For a long time, studies aimed at elucidating the mechanisms of transcriptional activation and repression sought to determine how DNA-binding proteins could influence the initiation and elongation of transcription by RNA polymerases. Recently, it has become clear that alteration of gene expression has to occur in the context of chromatin. The nucleosome as the basic unit of chromatin consists of ∼146 bp of DNA wrapped around a histone octamer made up of two copies each of four core histones H2A, H2B, H3 and H4 (van Holde, 1988). While the structure of the core nucleosome is well defined, the basic N-terminal histone tails protrude from the core nucleosome and show no defined structure (Luger et al., 1997). These tail domains are subject to post-translational modifications such as acetylation, phosphorylation, methylation and ADP-ribosylation (van Holde, 1988). Recent observations indicate that these modifications occur interdependently and create a pattern that might modulate the affinity of histone-binding proteins. These findings are the basis of the histone code hypothesis (Strahl and Allis, 2000; Turner, 2000). The best studied modification of core histones is the reversible acetylation of conserved lysine residues within the N-terminal tails. Two types of enzymes, the histone acetyltransferases (HATs) and the histone deacetylases (HDACs), control the acetylation of histones and other substrates. More than a dozen deacetylases have been identified during the last few years. According to their homology to yeast proteins, the HDACs are grouped into three classes (Gray and Ekström, 2001; Khochbin et al., 2001). The highly homologous class I enzymes HDAC1 and HDAC2 can heterodimerize and frequently are found in the same complexes. These deacetylases seem to be involved in more general cellular processes. In contrast, class II enzymes might have tissue-specific functions during the execution of developmental programs. According to its sequence, HDAC3 also belongs to the class I family. However, the protein can interact with both class I and class II enzymes, and might be a functional link between the two enzyme families (Grozinger et al., 1999; Fischle et al., 2001). The third mammalian HDAC class is formed of enzymes with homology to the NAD-dependent deacetylase Sir2. Human Sir2 was shown recently to be a p53 deacetylase involved in the control of cell survival (Luo et al., 2001; Vaziri et al., 2001). HDAC1, a homolog of the yeast protein Rpd3p, was the first protein shown to have histone deacetylase activity (Vidal and Gaber, 1991; Taunton et al., 1996). Subsequently, HDAC1 and its closest homolog HDAC2 (Yang et al., 1996) were found associated with several accessory proteins such as RbAp46/48, Sin3A/Sin3B, SAP18 and SAP30 (reviewed in Ahringer, 2000). These factors seem to be required as structural components of HDACI–HDACII complexes and for the interaction with core histones. A variety of transcription factors including Rb, Mad, unliganded nuclear hormone receptors and p53 have been shown to regulate target genes by association with HDAC1 (reviewed in Cress and Seto, 2000; Ng and Bird, 2000). Recruitment of HDAC1 by Rb and other pocket proteins is thought to be important for the growth-inhibitory effect of these proteins (Zhang and Dean, 2001). In addition, HDAC1 interacts with the methyl-binding proteins and DNA methyltransferases, strongly suggesting cooperativity between DNA methylation and histone deacetylation in gene silencing (Dobosy and Selker, 2001; Fuks et al., 2001). Despite the growing knowledge on the mechanisms of HDAC-dependent gene repression, surprisingly little is known about the biological function of individual mammalian deacetylases. Most data concerning the physiological role of HDACs were obtained by using deacetylase inhibitors, which block the majority of class I and class II enzymes. Here, we have disrupted the Hdac1 gene in mice to examine its role during development and proliferation. We show that HDAC1 is a major deacetylase in mouse embryonic stem (ES) cells and its loss results in a substantial reduction of cellular HDAC activity and specific changes in histone modifications. The related enzymes HDAC2 and HDAC3 are up-regulated in HDAC1-null cells, but cannot compensate for the loss of HDAC1 function. Targeting of both HDAC1 alleles leads to embryonic lethality before E10.5. HDAC1-deficient embryos and HDAC1-null ES cells have proliferation defects and display increased levels of a subset of cyclin-dependent kinase (CDK) inhibitors. Our data demonstrate that HDAC1 is essential for mouse embryonic development and that the enzyme ensures cell proliferation by repressing the expression of specific growth-inhibitory genes. Results Expression of HDAC1 during mouse embryogenesis Mouse HDAC1 was identified originally as a growth factor-inducible protein (Bartl et al., 1997). The enzyme is ubiquitously expressed in tissues of adult mice, with elevated levels in thymus, testis, ovary and placenta (Bartl et al., 1997). As a first step to investigate the function of HDAC1 for mammalian development, we analyzed the expression of the enzyme during different stages of mouse embryogenesis. As detected by indirect immunofluorescence microscopy, fertilized oocytes and two-cell embryos show a rather strong cytosolic HDAC1 signal that might be due to maternal storage (Figure 1A). Concomitant with zygotic genome activation in the two-cell stage, a nuclear HDAC1 staining pattern became visible in mouse pre-implantation embryos. The expression of HDAC1 was constantly high during the following cleavages to the blastocyst stage (Figure 1A). Significant HDAC1 expression levels were also detected during post-implantation development, as shown by immunostaining of E7.5 and E8.5 embryos (Figure 1B). HDAC1 was highly abundant in most embryonic and extra-embryonic tissues, with specifically strong expression in trophoblast giant cells. These large embryonic cells have an important function for the nutrition of the embryo and contain an amplified genome. The high content of DNA in these cells correlates with elevated HDAC1 levels, but HDAC1 does not seem to be required for the endoreduplication of the trophoblast DNA (see below). Next, we compared the expression patterns of the homologous class I histone deacetylases HDAC1, HDAC2 and HDAC3 during mouse development by western blot analysis. All three enzymes were highly expressed from mid to late gestation in the mouse (Figure 1C). In contrast to the other HDACs analyzed, only HDAC1 is highly abundant in ES cells and embryoid bodies (Figure 1C, ES and EB). The RbAp48 protein was shown previously to associate with HDAC1, and this interaction might be required for its full enzymatic activity (Taunton et al., 1996). As shown in Figure 1C, RbAp48 has an expression pattern similar to HDAC1. Together, these data suggest a specific function for HDAC1 during early embryonic development. Figure 1.HDAC1 expression during mouse embryonic development. (A) HDAC1 expression in pre-implantation embryos from the fertilized oocyte to the blastocyst stage was monitored in indirect immunofluorescence experiments. HDAC1 was visualized using a polyclonal HDAC1 antiserum and a secondary Texas red-conjugated antibody. Genomic DNA was stained with DAPI. (B) Immunohistochemical analysis of HDAC1 expression in paraffin-embedded sections of mouse embryos. Adjacent sections were stained with hematoxylin/eosin (H+E) or immunostained for HDAC1 at E7.5 (upper panels) and E8.5 (lower panels). High HDAC1 expression is found in all embryonic and extra-embryonic tissues, in particular in trophoblast giant cells. am, amnion; ch, chorion; ec, ectoplacental cone; ee, embryonic ectoderm; em, embryonic mesoderm; en, endoderm; tgc, trophoblast giant cells; al, allantois; hf, headfold; so, somites. (C) Expression patterns of HDAC1, HDAC2, HDAC3 and RbAp48 during mouse embryogenesis. Western blot analysis of whole-cell extracts from ES cells, embryoid bodies (EB) and 10.5- to 17.5-day-old embryos (E10.5–E17.5). P1-3 indicates 1–3 days after birth. Exposure times for ECL detection were adjusted approximately to the individual sensitivities of the HDAC antibodies. Download figure Download PowerPoint Targeted disruption of the Hdac1 gene To evaluate the function of HDAC1 in mouse development, the Hdac1 locus was inactivated by a conventional targeting approach (Figure 2A). Part of exon 5 and exons 6–7 of the murine Hdac1 gene (Khier et al., 1999) were replaced by a β-galactosidase/neomycin phosphotransferase (lacZ/neo) cassette in E14.1 ES cells. The replaced sequence encodes the highly conserved deacetylase consensus motif (Hassig et al., 1998) (Figure 2A). Two independently targeted ES clones were injected into C57BL/6 mouse blastocysts and chimeras derived from both cell lines were used to generate heterozygous mice. The genotype of offspring derived from HDAC1 heterozygous intercrosses was determined by Southern blot analysis or PCR (Figure 2B, and data not shown). No HDAC1-null animals were observed in either a C57BL/6 × 129/Sv background or a pure 129/Sv background (Table I). Heterozygous animals were obtained with a frequency below the expected Mendelian ratio, but were viable and fertile and appeared to have a normal phenotype. Figure 2.Disruption of the murine HDAC1 locus. (A) Structure of the mouse Hdac1 gene locus. In the targeting vector, part of exon 5 and exons 6 and 7 encoding the deacetylase consensus motif were replaced by the lacZ/neo cassette for G418 selection of positive clones. The diphtheria toxin gene (DTA) was inserted 3′ of the long arm to select against random integration of the targeting construct. (B) Southern blot analysis of tail DNA isolated from offspring of heterozygous intercrosses. Genomic DNA was digested with SacI and hybridized with an intron-specific probe, which recognizes a 7.3 kb fragment in the wild-type allele and a 4.3 kb SacI fragment in the targeted locus. (C) Yolk sac PCR analysis of E9.5 embryos. Download figure Download PowerPoint Table 1. Absence of HDAC1-null animals in offspring from HDAC1 heterozygous intercrosses in a mixed C57BL/6 × 129/Sv background and a pure 129/Sv background +/+ +/− −/− ND Total BL/6 × 129/Sv 466 718 0 17 1201 39% 60% 0% 1% 100% 129/Sv 114 137 0 0 251 45% 55% 0% 0% 100% The genotype was determined by Southern blot analysis or PCR. ND, not determined. To determine when the HDAC mutation produces a lethal phenotype, timed matings were set up and embryos were obtained from heterozygous intercrosses at E7.5–13.5 and genotyped by PCR analysis of DNA extracted from yolk sac or total embryos (Figure 2C). Examination of embryos isolated between E7.5 and E9.5 revealed all genotypes with a Mendelian ratio. All HDAC1 mutants appeared extremely growth retarded (see below). No homozygous null embryos could be detected after E9.5, indicating that lack of HDAC1 leads to embryonic lethality before day 10.5 of gestation. HDAC1 is indispensable for mouse embryonic development To reveal potential developmental defects, embryos from heterozygous intercrosses were analyzed by whole-mount in situ hybridization with an HDAC1 riboprobe. HDAC1 wild-type or heterozygous embryos showed HDAC1 expression throughout the embryo except the developing heart (Figure 3A–C). At E9.5, HDAC1-null embryos displayed numerous abnormalities including severely disturbed head and allantois formation (Figure 3D). HDAC1−/− embryos showed normal expression of the mesoderm marker brachyury (T) (Wilkinson et al., 1992), excluding the possibility of a gastrulation failure (Figure 3E). We also ruled out the possibility that the HDAC1 phenotype arises due to a defect in the specification of the placenta. Trophoblast giant cells showed high HDAC1 expression levels (Figure 1B). However, analysis of placental lactogen-1 (PL-1), a marker for trophoblast giant cells (Colosi et al., 1987), by whole-mount in situ hybridization revealed a staining pattern in the null embryo similar to wild-type (Figure 3F). To exclude, that increased apoptosis accounts for the reduced size of HDAC1−/− embryos, we analyzed wild-type and null embryos by TUNEL assay. At the onset of the HDAC1-null phenotype at E7.5, only a few apoptotic cells were observed in both null and wild-type embryos (Figure 3G and H). In addition, neither wild-type nor HDAC1-null embryos show signs of necrosis at E7.5 (Figure 3I and J). Figure 3.Phenotypic analysis of HDAC1 mutant embryos. (A–D) Whole-mount in situ hybridization of wild-type and HDAC1 mutant embryos with an HDAC1 riboprobe. (A) HDAC1 is highly expressed in the ectoplacental cone (ec), the extra-embryonic (ex) and embryonic (em) ectoderm at E7.5. (B) Elevated HDAC1 levels are detected in the head fold (hf) and the neural fold (nf) at E8.5. (C) HDAC1 is highly expressed throughout the E9.5 embryo except in the developing heart (he), with pronounced expression in the bronchial arches (ba) and the limb bud (lb). (D) Spectrum of phenotypes observed for HDAC1 mutant embryos at E9.5 with misformed allantois (al) and defects in head formation; hr, head region. (E) Brachyury expression at E9.5 is comparable in HDAC1 wild-type and HDAC1 mutant embryos. (F) E7.5 wild-type and mutant embryos within the maternal decidua are shown. Placental lactogen-1 expression in trophoblast giant cells (tgc) is similar in HDAC1 wild-type and HDAC1 null embryos. (G and H) TUNEL assay on paraffin-embedded sections of E7.5 wild-type (G) and mutant (H) embryos shows similar numbers of apoptotic cells in the embryonic ectoderm (ee) of wild-type and mutant embryos. Arrows indicate apoptotic cells. (I and J) Hematoxylin/eosin staining of sections adjacent to the sections shown in (G) and (H). Download figure Download PowerPoint Therefore, the severely growth retarded appearance of HDAC1-deficient embryos is suggestive of a cellular proliferation defect. To test this assumption, we performed an immunohistochemical analysis of wild-type and HDAC1-deficient embryos at E7.5 for the expression of the proliferation marker Ki67 antigen (Schluter et al., 1993). The observed difference in size between HDAC1 wild-type and null embryos correlated well with a significantly reduced number of proliferating cells in HDAC1-null embryos (51% Ki67 positive) compared with wild-type embryos (65% Ki67 positive) (Figure 4). Taken together, these results strongly suggest that the reduced size of HDAC1-deficient embryos at E7.5 is due to a proliferation defect rather than to increased apoptosis. Figure 4.Reduced proliferation of HDAC1-null embryos. Immunohisto chemical analysis of HDAC1 and Ki67 antigen expression in mouse embryos at E7.5. Adjacent paraffin-embedded sagittal sections of mouse embryos were stained with hematoxylin/eosin (A and B) or immunostained for HDAC1 (C and D) or for the proliferation marker Ki67 nuclear antigen (E and F). The boxed areas show higher magnifications of the Ki67-stained embryo sections. Download figure Download PowerPoint Changed histone modifications in HDAC1-null cells To examine the effects of HDAC1 deficiency at the cellular level, we generated HDAC1-null ES cells by blastocyst outgrowth experiments (Hogan et al., 1994). Five independent HDAC1-deficient ES cell lines and corresponding wild-type and heterozygous control cell lines were analyzed and gave essentially the same results. Comparative analyses were performed with early passages of wild-type, heterozygous and null ES cells obtained from littermates. HDAC1 protein levels were decreased in heterozygous ES cells and undetectable in HDAC1-null cells (Figure 5A). Expression of the HDAC1-associated factor RbAp48 was unchanged in heterozygous and HDAC1-deficient cell lines. Levels for the HDAC1-related enzymes HDAC2 and HDAC3, in contrast, were found to be increased complementary to HDAC1 expression (Figure 5A). Similarly, HDAC2 protein was expressed at a higher level in various tissues of heterozygous mice (data not shown). Despite the up-regulation of HDAC2 and HDAC3, the total histone deacetylase activity was significantly decreased in HDAC1-null cells (Figure 5B). HDAC2 previously was found to be associated with HDAC1 (Hassig et al., 1998) and this association might be necessary for enzymatic activity of HDAC1–HDAC2-containing complexes. To examine whether HDAC2 activity is impaired in HDAC1-deficient ES cells, we analyzed HDAC1–HDAC2-containing complexes in fractionation experiments and HDAC activity assays. In wild-type cells, HDAC1 and HDAC2 were found in the same fractions of a 5–15% sucrose gradient (Figure 5C). In HDAC1-null cells, the corresponding fractions showed significantly reduced total histone deacetylase activity and HDAC2 displayed a slightly broader distribution profile, which might be due at least in part to the increased expression of HDAC2. HDAC2 co-sedimented in both cell lines with components of the NuRD and Sin3 complexes such as Sin3A, MTA1 and RbAp48, consistent with the presence of intact HDAC2 complexes (Figure 5C). Figure 5.HDAC1-deficient ES cells display decreased histone deacetylase activity and changes in histone modifications. (A) Western blot analysis of homologous HDAC proteins of total protein extracts prepared from wild-type, heterozygous and null ES cell lines. The blot was incubated sequentially with antibodies directed against HDAC1, HDAC2, HDAC3 and RbAp48, respectively. (B) Equal amounts of extracts described in (A) were analyzed for deacetylase activity with tritium acetate-labeled histones as substrates. Counted radioactivity corresponds to the amount of released acetyl moieties per hour and 10 μg protein and reflects the relative HDAC activity. Results are shown as mean values of three independent experiments. (C) Co-sedimentation analysis of HDAC1–HDAC2-containing complexes in wild-type and HDAC1-null cells. One aliquot of each fraction was analyzed on western blots for the presence of components of HDAC1–HDAC2 complexes. A second aliquot of each fraction was tested for total HDAC activity. (D) Comparison of deacetylase activities associated with components of the HDAC1–HDAC2 complexes. HDAC1, HDAC2, Sin3A and MTA1 were immunoprecipitated from whole-cell extracts prepared from wild-type or HDAC1-null ES cells and analyzed for associated HDAC activity as described in (B). The data shown are representative of three independent experiments. (E) Changes in core histone modifications in HDAC1 heterozygous and homozygous ES cells. Lack of HDAC1 resulted in H3 and H4 hyperacetylation, increased S10/K14 phosphoacetylation and reduced K9 methylation of histone H3. Histones were extracted and analyzed on western blots with antibodies recognizing acetyl-histone H4 (AcH4), acetyl-histone H3 (AcH3), histone H3-phosphoS10-acetylK14 (P/AcH3) and histone H3-methylK9 (MetH3). Equal loading was controlled by probing with an H3 antibody (H3). Download figure Download PowerPoint To test directly a requirement for HDAC1 for HDAC2 enzymatic activity, we analyzed the deacetylase activity of immunoprecipitated HDAC1 and HDAC2. As expected, immunoprecipitates obtained with HDAC1 antibodies from null ES cell extracts showed only background activity, while HDAC2 activity was significantly increased in HDAC1-deficient cells (Figure 5D). In contrast, enzymatic activity associated with Sin3A and MTA1 immunoprecipitates was clearly reduced in HDAC1-null cells. Taken together, these findings argue against a strict requirement for HDAC1 for HDAC2 enzyme function, but indicate a potential role for HDAC1 for a subset of deacetylase complexes that contain both enzymes (Humphrey et al., 2001). Our data show that HDAC1 is a major histone deacetylase in mouse ES cells. Next, we analyzed the effects of HDAC1 deficiency on modifications of histone tails. While the acetylation of bulk histones as detected on acidic Triton–urea gels was not significantly affected by the loss of HDAC1 (data not shown), western blot analysis with modification-specific antibodies revealed increased acetylation levels of a subset of histones H3 and H4 in HDAC1-null cells (Figure 5E). The difference in the results obtained by the two methods is due to the higher sensitivity of the western blot analysis and indicates that only a relatively small portion of core histones are hyperacetylated in HDAC1-deficient cells (see Discussion). Different modifications of N-terminal histone tails such as phosphorylation, methylation and acetylation have been shown to be interdependent (Cheung et al., 2000b; Clayton et al., 2000; Lo et al., 2000; Rea et al., 2000; Nakayama et al., 2001). In good agreement with these findings, we observed an increase in histone H3 phosphorylation on Ser10 and phosphoacetylation on Ser10 and Lys14 in HDAC1-null ES cells (Figure 5E and data not shown). Phosphoacetylation of histone H3 was shown to play an important role for transcriptional activation (Cheung et al., 2000a; Clayton et al., 2000). In contrast, histone H3 methylation on Lys9 was implicated in transcriptional silencing (Jenuwein, 2001). As shown in Figure 5E, histone H3 methylation was slightly reduced in HDAC1−/− ES cells. Our results support the idea of cooperation between histone acetylating and phosphorylating enzymes and suggest that deacetylation by HDAC1 might be linked to efficient histone methylation. HDAC1 is essential for unrestricted cell proliferation of mouse ES cells ES cells, when cultured under defined conditions and in the absence of leukemia inhibitory factor (LIF), differentiate into embryoid bodies (Desbaillets et al., 2000). Embryoid bodies were generated from HDAC1 wild-type, heterozygous or null ES cell lines. ES cells showed an HDAC1 dose-dependent size of the inner cell mass, suggesting a prolonged generation time for HDAC1-deficient ES cells (Figure 6A). The differentiation of ES cells, however, was not affected by the absence of HDAC1 as judged by parietal endoderm formation (Figure 6A) and the spectrum of differentiated cells that appea