Histone deacetylases 1 and 2 redundantly regulate cardiac morphogenesis, growth, and contractility

Abstract

Histone deacetylases (HDACs) tighten chromatin structure and repress gene expression through the removal of acetyl groups from histone tails. The class I HDACs, HDAC1 and HDAC2, are expressed ubiquitously, but their potential roles in tissue-specific gene expression and organogenesis have not been defined. To explore the functions of HDAC1 and HDAC2 in vivo, we generated mice with conditional null alleles of both genes. Whereas global deletion of HDAC1 results in death by embryonic day 9.5, mice lacking HDAC2 survive until the perinatal period, when they succumb to a spectrum of cardiac defects, including obliteration of the lumen of the right ventricle, excessive hyperplasia and apoptosis of cardiomyocytes, and bradycardia. Cardiac-specific deletion of either HDAC1 or HDAC2 does not evoke a phenotype, whereas cardiac-specific deletion of both genes results in neonatal lethality, accompanied by cardiac arrhythmias, dilated cardiomyopathy, and up-regulation of genes encoding skeletal muscle-specific contractile proteins and calcium channels. Our results reveal cell-autonomous and non-cell-autonomous functions for HDAC1 and HDAC2 in the control of myocardial growth, morphogenesis, and contractility, which reflect partially redundant roles of these enzymes in tissue-specific transcriptional repression.

Figure 1.

Generation of a conditional HDAC1 allele. (A) Strategy to generate a conditional HDAC1 allele. Protein, corresponding exonic structure, targeting vector, and targeted allele are shown. loxP sites were inserted into introns 4 and 7 through homologous recombination. The neomycin resistance cassette, flanked by FRT sites, was removed by crossing to transgenic animals expressing hACTB∷FLPe in the germline. Cre-mediated excision results in one loxP site in the place of exons 5–7. (B) Southern blot analysis of agouti offspring from chimera/C57BL/6 intercrosses from targeted ES cells. Tail DNA was digested with EcoRI, and the corresponding wild-type (∼14 kb) and targeted (∼9.5 kb) bands are indicated for the 3′ probe. (C) Genotyping of HDAC1^floxed mice by genomic PCR. Primer set includes three primers. One primer set flanks the 5′ loxP site with the third primer downstream from the 3′ loxP site. Global deletion by CAG-Cre removes the primer within the loxP sites, resulting in one ∼550-bp product for the HDAC1^{floxed/floxed} animals. (D) Wild-type and HDAC1^loxP/loxP; CAG-Cre mutant embryos at E9.5. Recapitulation of the previously reported global KO is shown.

Figure 2.

Cardiac-specific deletion of HDAC1. (A) Wild-type and HDAC1^loxP/loxP; αMHC-Cre mice at 6 wk of age. Cardiac-specific deletion of HDAC1 was achieved by crossing HDAC1^loxP/loxP mice to mice harboring a transgene for αMHC-Cre. Hematoxylin and eosin (H&E)-stained hearts at 6 wk of age show no gross abnormalities. (B) Immunostaining for HDAC1 (green) and α-actinin (red) on cardiomyocytes isolated from wild-type and HDAC1^loxP/loxP; αMHC-Cre neonatal mice. Note cardiac fibroblasts positive for HDAC1 that are negative for α-actinin. Immunostaining for HDAC1 in Cre-positive myocytes shows no positive staining for HDAC1. (C) Transcript analysis of cardiac-specific deletion of HDAC1. Transcript levels were analyzed by semiquantitative RT–PCR for HDACs 1–5 and HDACs 7–9. GAPDH levels were detected as a control. HDAC1 transcript levels are not entirely lost due to high expression in contaminating cardiac fibroblasts, as seen in B.

Figure 3.

Generation of a conditional HDAC2 allele. (A) Strategy for targeting of HDAC2. Exonic structure, targeting vector, and targeted allele for HDAC2 are shown. loxP sites were introduced upstream of exon 2 and downstream from exon 4. Flanking FRT sites allowed for the neomycin resistance cassette to be removed by crossing to FLPe recombinase transgenic animals. Cre excision results in one loxP site in the place of exons 2–4. Probes for Southern analysis and primer positions for RT–PCR are shown. (B) Southern blot analysis for HDAC2. Tail DNA from agouti offspring was digested with SacI and probed with the indicated 3′ probe. Wild-type (7.5 kb) and targeted (5 kb) bands indicate proper transmission of targeted allele. (C) Table of offspring from HDAC2 heterozygous intercrosses. HDAC2-null mice were born at near Mendelian ratios, but showed 100% lethality within the first 24 h. (D) Transcript analysis of HDAC2 by RT–PCR. Genotypes of animals are shown at top, and primer positions from A are shown to the left. GAPDH levels were detected as a control. (E) Western blot analysis to show deletion of HDAC2. Western blot analysis on heart and lung tissue isolated from P1 wild-type, heterozygous, and HDAC2 homozygous-null neonates. eIF5 was detected as a loading control. (F) Transcript analysis on HDAC2-null mice. mRNA transcript levels were analyzed by semiquantitative RT–PCR for HDACs 1, 3–5, and 7–9. GAPDH was detected as a control.

Figure 4.

Cardiac defects in HDAC2^−/− neonates. (A) Structural abnormalities in the hearts of HDAC2-null neonates. Hearts from mice of the indicated genotypes at P1 are shown at the left. H&E-stained sections at four successive levels of wild-type and HDAC2-null hearts are shown. The myocardium of HDAC2-null animals displays a thicker septum and diminished or dislocated right ventricular lumen. (B) Heart rates of HDAC2-null neonates. EKGs show reduced heart rate in the HDAC2-null mice. (C) Increased proliferation of cardiomyocytes in HDAC2-null mice. Immunohistochemistry for phospho-histone H3 was performed on heart sections from P1 mice. Quantification of phospho-histone H3-positive cells was performed on six sections from three individual hearts and averaged. (D) Immunohistochemistry of heterozygous and HDAC2-null hearts at 20× magnification. DAPI (blue) and phospho-histone H3 (green) staining show increased proliferation in both the right and left ventricles.

Figure 5.

Cardiac defects resulting from cardiac deletion of HDAC1 and HDAC2. (A) Kaplan-Meier survival curves for cardiac deletion of HDAC1 and HDAC2 by αMHC-Cre. Note that one copy of HDAC2 is sufficient for 100% viability. (B) H&E-stained sections of wild-type and dCKO mice at P8, P11, and P13. Deletion of both HDAC1 and HDAC2 results in severe dilated cardiomyopathy by P11. (C) ECGs performed on wild-type and dCKO mice at P10. Deletion of HDAC1 and HDAC2 in cardiomyocytes leads to cardiac arrhythmia by P10. (D) Apoptosis in dCKO mice. TUNEL staining of wild-type and dCKO ventricular sections showed enhanced apoptosis in dCKO mice at P10. (E) Quantification of apoptosis in dCKO hearts. TUNEL-positive cells were quantified from six individually stained sections at 40× and averaged; P < 0.0001. (F) Expression of cardiac stress markers in dCKO mice at P11. mRNA transcript levels were detected by real-time RT–PCR and normalized to 18S RNA.

Figure 6.

Aberrant cardiac gene expression resulting from cardiac deletion of HDAC1 and HDAC2. (A) Gene ontology analysis was performed with PANTHER. Significantly (P < 0.05) enriched biological processes are shown. Plotted is the −log (P value) with the threshold set to 1.3 [log(0.05)]. (B) Molecular functions were assigned to the genes that fall in the most significantly enriched process, “cell structure and motility” (P < 0.00000005). (C) Calcium channel subunit dysregulation in the hearts of dCKO mice. Real-time RT–PCR analysis of transcript levels from wild-type, cardiac-specific HDAC1 KO, cardiac-specific HDAC2 KO, and dCKO hearts at P11. L-type and T-type calcium subunits were dysregulated in the dCKO hearts. Error bars indicate standard deviation. (D) Ca_V3.2 expression in dCKO hearts. Western blot analysis was performed on hearts from P11 wild-type (lane 1) and dCKO (lanes 2,3) mice. dCKO mice show increased expression of Ca_V3.2. Tubulin was detected as a loading control. (E) ChIP assays were performed from neonatal rat ventricular myocytes. Chromatin was immunoprecipitated with antibodies against HA as a negative control, HDAC1, HDAC2, or NRSF. Primers were designed around the NRSE within intron 1 of CACNA2D2, and precipitated DNA was analyzed by PCR. PCR was also performed from a nonimmunoprecipitated sample as an input control. (F) Specific dysregulation of skeletal myofibrillar proteins in the hearts of dCKO mice. Real-time RT–PCR analysis of transcript levels from wild-type, cardiac-specific HDAC1 KO, cardiac-specific HDAC2 KO, and dCKO hearts at P11. Error bars indicate standard deviation. (G) RNA in situ hybridization of Tnni2 transcripts on wild-type and dCKO hearts. (H) Troponin isoform expression in dCKO heats. Western blot analysis was performed on hearts from P11 wild-type (lane 1) and dCKO (lanes 2,3) mice. dCKO mice show heterogeneity in troponin isoform expression. Lane 4 is wild-type skeletal muscle as a control for Tnni2. (I) ChIP assays were performed on the IRE of Tnni2 in neonatal rat ventricular myocytes. Chromatin was immunoprecipitated with antibodies against HA, HDAC1, HDAC2, and NRSF. Precipitated DNA was analyzed by PCR using primers flanking the IRE. HDAC1 and HDAC2, but not NRSF, mediate repression of Tnni2 at basal levels. PCR was performed prior to immunoprecipitation as an input control.

Figure 7.

Stress-dependent cardiac hypertrophy in mice lacking cardiac expression of HDAC1 and HDAC2. (A) Wild-type, HDAC1^loxP/loxP; αMHC-Cre, HDAC2^loxP/loxP; αMHC-Cre, and HDAC1^loxP/+; HDAC2^loxP/loxP; αMHC-Cre mice at 8 wk of age were subjected to isoproterenol or saline infusion. Mice were sacrificed after 7 d, and cardiac hypertrophy was evaluated by heart weight/body weight ratios. (B) Wild-type and HDAC1^loxP/loxP; αMHC-Cre mice were subjected to TAC or sham operation. Heart weight/body weight ratios and heart weight/tibia length ratios were determined after 21 d. (C) Histological sections of representative hearts from B are shown stained with Masson Trichome. Bottom panels are 40× magnifications of the above hearts to show fibrosis.