Background
Interstitial fibrosis and fibrotic scar formation contribute to cardiac remodeling and loss of cardiac function in myocardial infarction (MI) and heart failure. In response to myocardial injury, the number of fibroblasts increased by replication of resident cardiac fibroblasts, recruitment of bone marrow cells and transformation of endothelial/epicardial cells to fibroblasts. Cardiac fibroblasts are activated leading to myofibroblasts, which are contractile, invasive, and high producers of extracellular matrix (ECM) proteins [
1‐
3]. While myofibroblasts serve an important role in wound healing and scar formation to repair areas of cardiomyocyte loss, further deposition of interstitial ECM causes myocardial stiffness and loss of ventricular function. Therefore, elucidating the mechanism(s) that regulates cardiac fibroblasts may provide a therapeutic strategy to reduce the amount of fibrosis in heart failure.
Epigenetic alterations, such as histone modifications, have been shown to be involved in tissue fibrosis in multiple organs including kidney, lung, and heart [
4]. Histone acetylation (by histone acetyl-transferases) relaxes normally tight chromatin super-coiling, enhancing accessibility of transcriptional regulatory proteins to promoter regions. On the other hand, histone deacetylases remove acetyl groups from lysine residues of histones and other proteins, which remodel chromatin, resulting in inhibition of gene expression. To date, 18 mammalian HDACs were identified and categorized into four classes. Class I HDACs (HDAC1, 2, 3, and 8) are widely expressed and have pro-hypertrophic function in heart disease. Among class I HDACs, HDAC1 and HDAC2 have been shown to play redundant roles in cardiac growth and function [
4,
5]. Several studies had reported that HDAC1 and 2 have pro-fibrotic roles in renal injury disease models [
6,
7]. Recently, an idiopathic pulmonary fibrosis study linked HDACs to myofibroblast differentiation and extracellular matrix deposition [
8]. Moreover, small molecule HDAC inhibitors, especially Class I and II inhibitors, were shown to effectively retard myocardial remodeling, decrease interstitial fibrosis, and improve cardiac function in pathological heart conditions [
9‐
13]. In most of these studies, HDAC inhibitors were applied in acute MI settings rather than after development of interstitial fibrosis in CHF. Thus, here we aim to investigate whether HDAC inhibition could be effective in reversing fibrosis in chronic conditions like ischemic heart failure. Here we explore the anti-fibrotic mechanisms of Mocetinostat, a highly specific HDAC1, 2, and 3 inhibitor, in cardiac fibroblasts.
In the present study, we demonstrated that HDAC 1/2 were upregulated in CHF infarcted areas and this upregulation progressed to non-infarcted myocardium including the left atrium when CHF advanced to a de-compensated stage. Using a small molecule HDAC inhibitor, Mocetinostat, we showed that inhibition of class I HDACs reversed interstitial collagen deposition and improved heart function in CHF. Moreover, Mocetinostat promoted reversal of myofibroblasts activation in vitro. In parallel, Mocetinostat treatment activated a p53/p21 axis and Caspase-3. Thus, the present study suggests that, in advanced heart failure settings, class I HDAC inhibition can reverse fibrosis. Anti-fibrotic activity of Mocetinostat includes reversal of myofibroblast phenotype and regulation of cell proliferation/apoptosis in cardiac fibroblasts.
Discussion
In this study, we demonstrated class I HDAC isoforms HDAC1 and HDAC2 are upregulated in cardiac fibroblasts as the animals progressed through heart failure. Interestingly, inhibition of Class I HDACs with Mocetinostat reversed the cardiac fibrosis in CHF animals suggesting an association between cardiac fibrosis and HDAC1/2 upregulation. Investigation of anti-fibrotic effects of HDAC inhibition on cardiac fibroblasts suggested several mechanisms including reversal of myofibroblast phenotype to fibroblast and induction of cell cycle arrest/apoptosis.
Analysis of HDAC1 and HDAC2 expression patterns in sham hearts showed that HDAC1 is mainly expressed by fibroblasts/interstitial cells while HDAC2 is more prominent in cardiomyocytes with low levels of expression in fibroblasts/interstitial cells (Figure
3). In addition, in CHF, both HDAC1 and HDAC2 showed strong staining in fibroblasts while HDAC2 maintained its expression in cardiomyocytes. Although our conclusions were based on confocal microscopy analysis, we cannot exclude the possibility of low level of expression of HDAC1 in cardiomyocytes. Altogether, these data suggest that in addition to their roles in cardiomyocyte regulation and hypertrophy [
22], HDAC1 and 2 are associated with regulation of cardiac fibroblast biology in heart.
We showed that class I HDAC inhibitor, Mocetinostat improved cardiac function in parameters of ventricular contractility and reduced total collagen amount in heart failure animals. Recent studies showed that HDAC inhibition retards ventricular and atrial fibrosis formation when applied immediately after injury [
23‐
27]. Moreover, in most of these studies pan-HDAC inhibitors such as TSA (Class I and Class II HDAC inhibitor) have been administrated for systematic HDAC inhibition. Our results highlight two important findings. First, we showed that selective inhibition of class I HDACs alone is effective in reducing fibrosis in CHF. Second, we started the treatments at 3w after MI, where animals already developed cardiac fibrosis. Thus, HDAC inhibition could effectively reverse interstitial fibrosis in heart failure and improve cardiac function.
Through HDAC inhibition with Mocetinostat
in vitro, we observed a reduction in α-SMA, a myofibroblast marker. We showed that Mocetinostat treatment downregulated collagen III in CD90+ cells, which is consistent with previous work where TSA blocked TGFβ-induced collagen synthesis in rat cardiac fibroblasts [
25]. Thus, our results suggest that the anti-fibrotic effect of Class I HDAC inhibition involves a reduction in myofibroblast activation and downregulation of one the most abundant ECM proteins, Collagen-III [
28].
As existing fibroblasts differentiate into myofibroblasts, epithelial, endothelial, and smooth muscle cells can also contribute to fibrosis via EMT, EndoMT, and mechanical tension. Recent study showed that HDAC inhibition blocked TGFβ-induced EMT via upregulation of E-cadherin and downregulation of α-SMA and ECM proteins in renal injury [
7]. Consistently, we showed that Mocetinostat application induced E-cadherin expression and surface translocation of β-catenin associated with morphological transformation of atrial CD90+ cells toward epithelial-like phenotype. The decrease in cell density in Mocetinostat-treated samples could induce changes in morphology of the cells as well. However, E-cadherin upregulation in these cells suggests that changes cell morphology does not solely depend on cell density. In addition, we demonstrated that in CD90+ fibroblasts, Mocetinostat reduced phosphorylation of Akt, an important EMT/EndoMT player [
29‐
33]. Thus, another possible anti-fibrotic effect of Class I HDAC inhibition could be reversal of EMT-induced fibrosis in cardiac fibroblasts.
Other effects of HDAC inhibitors are their ability to induce cell cycle arrest and apoptosis [
34]. Here, we showed a reduction in cell number, elevation of p21and p53 gene expression, and upregulation of Cleaved Caspase-3 protein with Mocetinostat treatment. p21 is a cyclin-dependent kinase inhibitor, which is involved in cell growth arrest [
35], while p53 induces apoptosis. Our data are supported by the recent studies where HDAC inhibition reduced cardiac fibroblast proliferation in isoproterenol induced heart failure [
24] and induced cell cycle arrest in cardiac fibroblast in model of angiotensin-II mediated fibrosis [
27]. On the other hand, we did not observe an increase in overall apoptosis in CHF myocardium with Mocetinostat treatment suggesting that systemic Mocetinostat administration does not induce excessive cell death in myocardium. However, rates of apoptosis in different cells types needs to be examined to determine whether Mocetinostat induces apoptosis selectively in certain cell types, that is, fibroblasts
in vivo. Altogether, these data suggest that regulation of fibroblast proliferation and apoptosis could be another possible mechanism of HDAC inhibition in control of fibrosis.
One of the limitations in our study is that we isolated cardiac fibroblast from atrial explants and ventricular tissue based on CD90 (Thy1) surface marker. Even though the majority of cardiac fibroblasts express CD90, this marker is not fibroblast-exclusive. Thus, other cells types such as blood originated cells could be included as well. However, a fibroblast specific marker representing majority of the fibroblasts in the heart is not currently available. In addition, CD90+ cells do not represent entire fibroblast population in heart; this marker is rather expressed on a subset of fibroblasts. Therefore, we used another marker, Vimentin (also not fibroblast specific), to confirm expression of HDAC1 and HDAC2 in cardiac fibroblasts. Thus, expression of HDAC1/2 in both CD90 and Vimentin + cells suggests that cardiac fibroblasts express HDAC1/2 in CHF hearts.
Methods
Animals
This study was performed in an accredited facility by the American Association for Accreditation of Laboratory Animal Care and was approved by the Institutional Animal Care and Use Committee at Banner Sun Health Research Institute. Animals received humane care in compliance with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85–23, revised 1996).
Myocardial infarction and treatments
MI was created by ligation of the left coronary artery as previously performed by our laboratory [
36]. In brief, rats were anesthetized using 1 mL/kg of an MI cocktail composed of ketamine (50 mg/mL), xylazine (15 mg/mL), acepromazine (2 mg/mL), and atropine (1 mg/mL). Animals were intubated and ventilated using a small animal ventilator (Harvard Apparatus). A left thoracotomy was performed via the third intercostal rib, and the left coronary artery was ligated. In sham operated animals, the chest was closed without ligation of the artery. The rats were sacrificed at 3 days (Acute MI-AMI), 3 weeks, or 6 weeks (congestive heart failure (CHF)) post surgery. At 3 weeks or 6 weeks after successful infarction, rats exhibited CHF indicated by elevation of left ventricular end-diastolic pressure (LVEDP), LV remodeling, and fluid accumulation in the chest [
36,
37]. In addition, prominent fibrosis in the ventricles and left atrium was evident at 6 weeks post surgery. Closed-chest
in-vivo cardiac function was measured using a Millar pressure conductance catheter system (Millar instruments, Houston, TX, USA) as previously described [
38]. A transverse cut was made to postmortem hearts to assess scar size and LV remodeling. Only animals with visible scar, LV remodeling, and LVEDP greater than 20 mmHg were included in the 3-week or 6-week CHF group.
Three weeks post surgery, animals in the CHF group were further divided into two groups; the first group received 10 mg/kg Mocetinostat dissolved in 0.1 N HCl PBS solution daily for the duration of 3 weeks (n = 8). The second group received only vehicle for the same duration (n = 6). In addition, animals that underwent sham surgery were injected with vehicle only and served as control group (n = 5).
Cell isolation and culture
Atrial explant outgrowth was generated as previously described [
38,
39]. Briefly, tissue was cut into 1 to 2 mm
3 pieces and digested with 0.2% trypsin (Life Technologies, Carlsbad, CA, USA) and 0.1% collagenase IV (Life Technologies) for a total of 10 min. The remaining tissue fragments were cultured as ‘explants’ in explants medium (CEM), which was composed of IMDM supplemented with 10% fetal bovine serum (FBS, Lonza), 100 U/mL penicillin G, 100 μg/mL streptomycin, and 2 mmol/L L-glutamine (Sigma-Aldrich). After 21 days in culture, cells were collected by trypsinization. CD90+ cells were separated from the cell outgrowths using magnetic beads (MACS, Miltenyi Biotec) according to manufacturer protocol and analyzed by flow cytometry for purity assessment. CD90 cells from ventricles were isolated by enzymatic digestion with a Dispase II (2.4 mg/mL, Roche)/Collagenase II (0.05 mg/mL, Gibco) mix in PBS. Ventricles were cut into small pieces and digested for 10 min at 37°C with agitation. The supernatant spun at 1,200 g for 7 min to collect dissociated cells. The last two steps were repeated five times and cells were pooled. Dissociated cells were plated and media was changed after 2 h to discard non-attached cells and debris. Then, attached cells were trypsinized and collected for CD90 isolation as described above.
Isolated atrial or ventricular CD90+ cells were seeded on six well plates at a density of 0.2 × 106 cells/well in CEM. Cells were treated with 1 μM and 2 μM of Mocetinostat (SelleckChem). Control cells were treated with 0.005% DMSO.
Immunostaining
Heart tissue was embedded in tissue freezing media (Triangle Biomedical Science) snap-frozen in liquid nitrogen and sectioned in the coronal plane using Leica CM1900 cryostat (Leica Microsystems, Bannockburn, IL, USA). Coronal and axial tissue sections (5 to 7 μm thickness) were mounted on positively charged glass slides and fixed/permeabilized in a 1:1 mixture of acetone and ethanol. Sections were blocked with 3% BSA in PBS and incubated with primary antibodies against α-MHC (Abcam), HDAC1, HDAC2, α-SMA (Abcam), CD90 (BD), and Vimentin (Abcam). Specific staining was visualized using corresponding secondary antibodies conjugated with Alexa 488 or 568 (Molecular Probes). Nuclei were stained with DAPI, 4′ 6-diamidino-2-phenylindole (Life Technologies).
Cells were fixed/permeabilized in a 1:1 mixture of acetone and ethanol. Cells were blocked with 3% BSA in PBS and stained with primary antibodies. Corresponding secondary antibodies were conjugated with Alexa-488 or Alexa 568 (Molecular Probes). Nuclei were stained with DAPI, 4′ 6-diamidino-2-phenylindole (Life Technologies).
Fluorescent images were captured using Leica TCS SPE confocal system configured with Leica DM 2500 microscope. Excitation maximums of 488 nm, 532 nm, and 405 nm, were used for image acquisition. Images were processed using LAS AF software (Leica Microsystems).
Scar size assessment and collagen assay
Heart tissue was processed as described above and sections of 7 or 20 μm were mounted on positively charged glass slides. Routine staining was performed with the Hematoxylin and Eosin kit (H&E, Sigma) according to manufacturer instructions. For scar size assessment, sections were stained with Masson’s Trichrome kit (Sigma-Aldrich) according to the manufacturer’s protocol. Transmitted light images of heart sections were processed using DP2-BSW software (Olympus Corp). Scar percentage was calculated as a ratio of collagen enriched scar area (blue staining) to the whole left ventricle area (red staining).
Total collagen amount was measured in 20 μm cross-sections using Sirus Red Fast Green Collagen staining kit (Chondrex) according to manufacturer instructions. Absorbance of collagen (540 nm) and non-collagenous protein (605 nm) was assessed using BioTek Synergy HT Microplate Reader. Collagen percentage was calculated as a ratio of OD540collagen to OD605non-collagenous protein.
RNA isolation and quantitative real-time RT-PCR
Total RNA was extracted from CD90+ cells using PureLink™ RNA Mini Kit (Life Technologies) according to the manufacturer’s protocol. RNA was then quantified with the Quanti-iT™ RiboGreen® RNA Assay Kit, and assessed using BioTek Synergy HT Microplate Reader (excitation/emission 480 nm/520 nm). Total RNA (200 ng) was reverse transcribed with QuantiTect Reverse Transcription kit (Qiagen). Real-time RT-PCR was conducted using the Rower SYBR Green Master Mix (Applied Biosystems) on a StepOnePlus Real-time PCR System (Applied Biosystems). Specific primers were synthesized by Life Technologies (sequences are available upon request). CYP A was used as a reference gene. Data analysis was performed on StepOne software version 2.1 (Applied Biosystems) using the comparative Ct (ΔΔCt) quantitation method.
Western blotting
Scar and LV tissue were dissected from AMI, CHF, and sham hearts, respectively. Dissected tissues and isolated cells were homogenized in lysis buffer (50 mM Tris- HCl pH7.5, 150 mM NaCl, 0.5% NP-40 (Sigma), 0.5% Triton-X (Sigma), 1 mM EDTA (Sigma), and complete mini protease inhibitor (Roche). Protein concentrations were determined by BCA assay (Thermo Scientific). Typically, 40 μg of protein was loaded on 4% to 12% Tris-Bis gels (Life Technologies), separated in MOPS running buffer, and transferred to a PVDF membrane (Millipore). After blocking with 5% non-fat dry milk in 1× TBS, membranes were probed with HDAC1, HDAC2 (Abcam, 1:1,000), α-SMA (Sigma, 1:5,000), E-cadherin, GSK3β (Santa Cruz, 1:500), p-GSK3β, Cleaved Caspase 3, p-Akt, Akt (Cell Signaling, 1:1,000), β-Catenin active (Millipore,1:1,000), and β-Actin (Sigma, 1:50,000) in TBS with 5% milk overnight at 4°C. Following three washes in TBS, membranes were incubated with HRP-conjugated secondary antibodies (Santa Cruz, 1:50,000) for 1 h at room temperature. An ECL (Millipore) system was used for detection of the bands and exposed to X-ray film (Thermo Scientific) in a dark room. Densitometry analysis was performed with Alpha Ease FC software.
Statistical analysis
All datasets are represented as mean ± S.E. Significance (P <0.05) was determined using Student’s t-test. Statistical analysis was conducted using Sigma Stat 3.5 software.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
HNG conceived and designed the experiments, performed western blotting, immunohistochemistry, and data analysis, and wrote the manuscript. LZ performed immunohistochemistry and contributed to writing of the manuscript. JN performed animal surgeries and cardiac function analysis. SP performed and participated in in vitro experiments. DM helped in design of the study. MG conceived and designed the experiments and helped in writing of the manuscript. All authors read and approved the final manuscript.