Fig. 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.

Fig. 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.

Fig. 6. Loss of HDAC1 in ES cells leads to impaired proliferation. (A) Three ES cell lines of different genotype were subjected to differentiation. The HDAC1-null embryoid bodies show a reduced inner cell mass (ICM), but normal differentiation. pe, parietal endoderm. (B) Growth curves of HDAC1 wild-type, heterozygous and null ES cell lines. Equal numbers of cells (5 × 10⁵) were seeded in triplicate and aliquots were counted daily during a time period of 6 days. (C) Colony formation assay with wild-type and HDAC1 mutant ES cells. A total of 1.5 × 10³ cells were seeded on SNL feeder layers and cultivated for 8 days. Cells were fixed and stained with methylene blue and colonies were counted. The average clone numbers of four plates are indicated with the respective standard deviations.

Fig. 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.

Fig. 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).

Fig. 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).

Fig. 7. The CDK inhibitors p21 and p27 are up-regulated in HDAC1-deficient ES cells. (A) Western blot analysis of whole-cell extracts prepared from wild-type, heterozygous and null ES cell lines. Western blots were probed sequentially with the indicated antibodies. (B) HDAC1-deficient ES cells show reduced cyclin A- and cyclin E-associated kinase activity. Cyclin A or cyclin E complexes were immunoprecipitated from ES cell extracts and incubated with histone H1 in the presence of [γ-³²P]ATP. Phospholabeled histone H1 was quantified and is shown for cyclin A (black bars) and cyclin E (white bars) relative to the kinase activity of wild-type cells arbitrarily set as 100. The results are representative of three independent experiments. (C) p21 and p27 expression is increased in E9.5 mutant embryos. HPRT was used as a control. The levels of p21, p27 and HPRT mRNA were determined by semiquantitative RT–PCR analysis. (D) Hyperacetylation of specific target promoters in HDAC1-null cells. Chromatin isolated from wild-type or HDAC1-deficient cells was precipitated in the absence of specific antibodies (no AB) or with antibodies specific for acetylated histone H3 (AcH3) or acetylated histone H4 (AcH4). Total input DNA (1×, 0.25× and 0.0625×) and DNA from the antibody-bound fractions were analyzed by quantitative PCR. Amounts of amplified DNA were quantified and are indicated relative to the signals from wild-type cells (set to 1).