Hypoxia induces hypomethylation of the HMGB1 promoter via the MAPK/DNMT1/HMGB1 pathway in cardiac progenitor cells

Abstract

Apoptosis is involved in the death of cardiac progenitor cells (CPCs) after myocardial infarction (MI) in the heart. The loss of CPCs results in infarct scar and further deterioration of the heart function. Though stem cell-based therapy provides an effective approach for heart function recovery after MI, the retention of CPCs in the infarcted area of the heart is the main barrier that limits its promising therapy. Therefore, the underlying mechanisms of CPC apoptosis in hypoxia are important for the development of new therapeutic targets for MI patients. In this work, we found that the expression of high-mobility group box 1(HMGB1) was upregulated in CPCs under hypoxia conditions. Further study demonstrated that HMGB1 was regulated by DNA methyltransferases 1 (DNMT1) via changing the methylation state of CpGs in the promoter of HMGB1 in CPCs during hypoxia process. Additionally, mitogen-activated protein kinase (MAPK) signaling pathway was found to be involved in regulating DNMT1/HMGB1-mediated CPC apoptosis in hypoxia process. In conclusion, our findings demonstrate a novel regulatory mechanism for CPC apoptosis and proliferation under hypoxia conditions, which may provide a new therapeutic approach for MI patients.

CPCs, hypoxia, HMGB1, DNMT1, MAPKs

Introduction

Heart disease induced by ischemia and consequent heart failure is the main leading cause of death worldwide [1,2]. Recent studies revealed that apoptosis is recognized as the chief mechanism of cell loss during myocardial infarction (MI) [3]. Traditional therapies cannot produce functional contractile tissue to replace the cardiac scarring region. However, stem cell-based novel therapies can induce neovasculogenesis and cardiogenesis in the MI tissue and therefore may promote cardiac tissue regeneration and heart function recovery in the MI patients [4–6].

Cardiac progenitor cells (CPCs) are a population of cardiac resident stem cells [7,8]. CPCs have the ability to differentiate into myocytes, endothelial cells, as well as vascular smooth muscle cells (VSMCs) [9,10]. For the CPCs are the endogenous components of the heart, the CPCs might be suitable for repairing the injured myocardium after MI. Several reports have documented that the CPCs can effectively repair the cardiac scarring with functional contractile tissue and then improve the heart function [9–11]. Though the CPCs seem to be the best choice for stem cell-based cardiac repair in MI heart, the success rate of this therapy is still quite low. It has been reported that half of the resident CPCs are depleted just after one day of MI [12]. A previous idea believed that transplanting large amounts of CPCs using ex vivo expansion protocols to the ischemic heart might overcome this limitation. However, it is still only a small number of transplanted CPCs are retained in the injured myocardium. Therefore, exploring the molecular mechanism of CPC proliferation and retention in the ischemia myocardium is of great importance.

High-mobility group box 1 (HMGB1) is a highly conserved nuclear protein [13]. HMGB1 acts as a chromatin-binding factor through binding with DNA and promoting access to transcriptional protein complexes [14,15]. HMGB1 is closely associated with the inflammatory response, cell proliferation and apoptosis, angiogenesis, and cell growth [16]. It has been reported that HMGB1 can promote Jurkat cell apoptosis after treatment with chemical apoptosis inducers [17]. Additionally, HMGB1 is involved in hepatocyte apoptosis through a p38-dependent mitochondrial pathway [18]. However, whether HMGB1 contributes to CPC apoptosis after ischemia stimuli is poorly investigated.

A previous study indicated that epigenetic mechanism contributes to a variety of physiological and pathological processes [19]. These epigenetic mechanisms include DNA methylation, acetylation of histones and non-coding RNAs [19]. Aberrant DNA methylation is associated with gene silencing and therefore contributes to gene regulation [20,21]. DNA methylation which is catalyzed by DNA methyltransferases (DNMTs) mainly occurs at the position 5΄ of cytosines of CpG dinucleotides of the gene. DNMT1 is the major maintenance methyltransferase during DNA replication. DNMT3a and DNMT3b are the de novo methyltransferases [20,21]. It has been reported that the profiles of DNA methylation are highly dynamic during various ischemic stroke conditions, which plays an important role in neuroprotection [22]. In the heart, lower LINE-1 methylation is related to higher risk of incident ischemic heart disease and stroke [23]. Whether DNA methylation is involved in the apotosis of heart cells during MI is rarely reported.

In this study, we aimed to explore the mechanism of ischemia-induced apoptosis in the heart. We found that MAPKs/DNMT1/HMGB1 signaling axis plays an important role in regulating hypoxia-induced CPC apoptosis, which can facilitate the understanding and development of therapies for myocardium I/R injury disease.

Materials and Methods

Reagents

p-JNK inhibitor SP100625, p-p38 inhibitor SB203580, and p-ERK inhibitor PD98059 (Sellectchem, Houston, USA) were dissolved in dimethyl sulfoxide (DMSO). 5-Aza-2΄-deoxycytidine (Aza; Sigma, St Louis, USA) was dissolved in phosphate-buffered saline (PBS).

CPC isolation, culture, and identification

All the isolation and culture processes were performed under sterile conditions. Briefly, the hearts of C57BL/6 mice (Cyagen, Taicang, China) were chopped into small pieces in cold Hanks Balance Salts Solution (HBSS; Sigma). After collagenase (1 mg/ml in HBSS; Sigma) digestion, the digested small pieces of heart tissue were cultured in DMEM (Invitrogen, Carlsbad, USA) supplemented with 20% fetal bovine serum (Invitrogen) in Matrigel-coated six-well plates (Coring, Corning, USA). After 2 weeks of culture, the ‘small, round, and bright’ cells migrated on the edge of the heart tissue are mainly CPCs. c-Kit (Santa Cruz Biotech, Santa Cruz, USA)-coated beads (Thermo Fisher Scientific, Waltham, USA) were applied to purify the CPCs. After purification, the CPCs were cultured in a humidified cell incubator at 37°C and 5% CO₂. The purity of the CPCs was identified by flow cytometry. Briefly, after the CPCs were stained with anti-c-Kit antibody (Abcam, Cambridge, UK), anti-CD45 antibody (Biolegend, San Diego, USA), and anti-CD34 antibody (Abcam), the CPCs were analyzed with a flow cytometer equipped with CellQuest software (BD Biosciences, Franklin Lakes, USA).

Hypoxic preconditioning of CPCs

CPCs were grown under hypoxic conditions in a Modular Incubator Chamber (Billumps-Rothenberg, Del Mar, USA) according to the manufacturer’s protocol. In brief, CPCs were placed in the chamber and the chamber was filled with a mixture of 0.1% O₂, 5% CO₂, and 94.9% N₂. Then, the cells were maintained for 24 h in the chamber and used for further experiments.

Plasmid construction

The DNMT1 constructs were based on the pcDNA3.1 (Invitrogen, Carlsbad, USA). The full-length human DNMT1 complementary DNA (cDNA) sequence was amplified by quantitative real-time PCR (qRT-PCR) and ligated into the XhoI and EcoRI sites of the pcDNA3.1 vector. The primers used are as follows: the forward primer F, 5΄-ATGCCGGCGCGTACCGCCCCAGCC-3΄, and reverse primer R, 5΄-CTAGTCCTTAGCAGCTTCCTCCTCC-3΄.

RNA interference and cell transfection

HMGB1- and DNMT1-specific oligonucleotides were designed and synthesized by GenePharma (Shanghai, China). The sequences of siRNAs are as follows: HMGB1-siRNA1, 5΄-CUGCGAAGCUGA AGGAAAATT-3΄; HMGB1-siRNA2, 5΄-AUCACAGUGUUGUUA AUGUTT-3΄; HMGB1-siRNA3, 5΄-CACAAACTGCCATTCAACA GGTATT-3΄; DNMT1-siRNA1, 5΄-CAAUGAGACUGACAUCAAA TT-3΄; DNMT1-siRNA2, 5΄-GGAAGUGAAUGGACGUCUATT-3΄; DNMT1-siRNA3, 5΄-CACTGGTTCTGCGCTGGGA-3΄; Negative control, 5΄-CGUUCCACTCATUUCCACAGT-3΄.

The siRNA and plasmid transfections in CPCs were performed using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s instruction. CPCs were seeded in six-well plates. Next day, the plasmids or siRNA oligonucleotides in opti-MEM (Invitrogen) were mixed with Lipofectamine 2000 and added into the plates. The cells were transfected with 50 nM of siRNA or negative control. The siRNA was dissolved in Rnase-free water. After 48 h, the transfected cells were harvested for the further experiments.

Total RNA isolation and qRT-PCR

Total RNA of CPCs with different treatments was obtained using Trizol reagent (Invitrogen). NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific) was used to detect the RNA concentration. EasyScript One-Step gDNA Removal and cDNA Synthesis SuperMix (Transgen biotech, Shanghai, China) was used to synthesize cDNA. cDNA (10 ng) was applied for qRT-PCR amplification using the SYBR Green reagent (Donghuan Biotech, Shanghai, China) and operated on the Prism 7500 SDS system (Applied Biosystems, Foster City, USA). The PCR was carried out with the following conditions: 95°C for 3 min, 40 cycles of 95°C for 30 s, 60°C for 30 s, and 72°C for 30 s. β-Actin was used as the housekeeping gene control. The primers for various genes are as follows: HMGB1 (forward: 5΄-GATGGGCAAAGGAGATCCTA-3΄; reverse: 5΄-CTTGGTCTCCCCTTTGGGGG-3΄); DNMT1 (forward: 5΄-CAG GACTACGCAAGGTTTGAG-3΄; reverse: 5΄-CGGATACAAGATA GGCAGAACT-3΄); DNMT3a (forward: 5΄-CTGAGAGTCAGGGA CTTGGC-3΄; reverse: 5΄-AGTCTAGCAATCGTTGGCGT-3΄); DNMT3b (forward: 5΄-CAGGGAAAACTGCAAAGCTC-3΄; reverse: 5΄-ATTTGTTACGTCGTGGCTCC-3΄); β-Actin (forward: 5΄-GCACCGTCAAGGCTGAGAAC-3΄; reverse: 5΄-ATGGTGGTGA AGACGCCAGT-3΄). The data were analyzed using the 2^−△△CT method and shown as fold change relative to β-Actin.

Western blot analysis

The CPCs were lysed using RIPA lysis buffer (Donghuan Biotech). Next, the lysates were gently agitated and centrifuged at 4°C. The concentration of the protein mixture samples were measured using BCA Protein Assay Kit (Donghuan Biotech). The protein mixture samples were separated on 12% SDS-PAGE and then examined by western blot analysis using the following primary antibodies: anti-Bax antibody (Abcam), anti-Bcl-2 antibody (Abcam), anti-C-cas3 antibody (Abcam), anti-HMGB1 antibody (Abcam), anti-DNMT1 antibody (Abcam), anti-DNMT3a antibody (Abcam), anti-DNMT3b antibody (Abcam), anti-p-JNK antibody (Abcam), anti-T-JNK antibody (Abcam), anti-p-ERK antibody (Abcam), anti-T-ERK antibody (Abcam), anti-p-p38 antibody (Abcam), anti-T-p38 antibody (Abcam), and anti-GAPDH antibody (Donghuan Biotech). Horseradish peroxidase-conjugated anti-IgG secondary antibody was used to detect the first antibody. The immune reactivity was visualized using the Chemiluminescence Detection Kit (Donghuan Biotech). And the data were analyzed using Gel-pro Analyzer software. Relative protein expression levels were normalized to GAPDH.

Cell viability assay

After CPCs were cultured for 24 h, Cell Counting Kit-8 (CCK8) (Donghuan Biotech) was added to the 96-well plate and then the absorbance at 490 nm was detected using a microplate reader.

5-Ethynyl-2΄-deoxyuridine assay

Cells were planted in 48-well plates at a density of 2.5 × 10⁴ cells/well and grown for 18 h. The cell proliferation was measured using the 5-ethynyl-2΄-deoxyuridine (EdU) detection kit (Donghuan Biotech) according to the manufacturer’s protocol. The results were observed using a florescence microscope (Olympus, Tokyo, Japan).

Sodium bisulfite DNA sequencing for DNA methylation analysis

Sodium bisulfite DNA sequencing for DNA methylation analysis was performed as previously described [24]. Genomic DNA was isolated using a TIANamp Genomic DNA kit (Tiangen Biotech, Beijing, China) and was converted using the EZ DNA Methylation-Gold kit (ZYMO Research, Beijing, China). The primers for sodium bisulfite DNA sequencing were designed using online tools MethPrimer and the PCR products were cloned into the pEASY-T1 cloning vector (TransGen Biotech). The PCR primers were as follows: forward primer F, 5΄-GAGGGGGTTTTAAATTTATTTATTT G-3΄ and reverse primer R, 5΄-AAAACTCACCTCCTTTAATCC TATTC-3΄.

Statistical analysis

All experiments were performed at least three times. Data were analyzed by the statistical software SPSS (version13.0.0). All data are presented as the mean ± SEM. Statistical analysis was performed using Student’s t-test or one-way ANOVA. Results were considered statistically significant at P < 0.05, while a highly significant difference at P < 0.01.

Results

Hypoxia treatment induces apoptosis in CPCs

CPCs are c-Kit-positive and CD45 and CD34 receptor-negative [8,25]. As shown in Fig. 1A, flow cytometry results showed that the percentage of c-Kit single positive CPCs was about 88.94%, while the percentage of CD45 and CD34 leukocytes was only about 1.12% and 1.78%. These results suggested that the purity of c-Kit-positive CPCs that we isolated and purified was very high.

Figure 1.

Hypoxia treatment induces apoptosis in CPCs (A) Identification of the purity of CPCs by flow cytometry. (B) qRT-PCR and western blot analysis of mRNA and protein level of Hif-1α in CPCs after hypoxia treatment. (C) The cell viability of CPCs was decreased in CPCs after hypoxia stimuli. (D) EdU incorporation assay was used to detect the proliferation potential of CPCs after hypoxia stimuli. (E) FACS analysis of the apoptosis of CPCs after hypoxia stimuli. (F) Western blot analysis were used to detect the apoptosis protein in CPCs after hypoxia treatment. The CPCs were cultured under hypoxia conditions for 24 h. *P < 0.05 and **P < 0.01. All experiments were performed in three biological repeats. β-Actin was used as the control.

Given that hypoxia induced cell apoptosis in various cells [26], we wondered whether this phenomenon exists in CPCs. We cultured the CPCs in 0.5% O₂ for different time intervals and then reperfused the CPCs with normoxia (21% O₂). After hypoxia treatment, both the mRNA and protein levels of Hif-1α were increased in a time-dependent manner (Fig. 1B). Meanwhile, the viability of CPCs was significantly reduced (Fig. 1C). In addition, EdU incorporation assay demonstrated that the proliferation potential of CPCs was significantly decreased after hypoxia treatment (Fig. 1D). Furthermore, the FACS analysis demonstrated that hypoxia induced apoptosis in CPCs (Fig. 1E). In addition, the apoptosis-related proteins Bax and C-cas3 were increased proportionally after hypoxia treatment, while the antiapoptosis protein Bcl-2 was inhibited (Fig. 1F). These data clearly indicated that hypoxia can extensively induce CPC apoptosis.

HMGB1 is upregulated in CPCs in hypoxia conditions

HMGB1 was reported to promote Jurkat cell apoptosis after treatment with chemical apoptosis inducers [17]. We further investigated whether HMGB1 contributes to the apoptosis of CPCs after hypoxia stimuli. Both qRT-PCR and western blot analysis showed that the expression level of HMGB1 was elevated in CPCs after hypoxia treatment (Fig. 2A). To further investigate the role of HMGB1 in CPC function, we designed siRNAs specific to HMGB1 to repress the expression of HMGB1 in CPCs (Fig. 2B). HMGB1 knockdown significantly elevated the cell viability of CPCs after hypoxia treatment (Fig. 2C). Meanwhile, the proliferation potential of CPCs was increased after HMGB1 knockdown under hypoxia conditions (Fig. 2D). Additionally, the apoptosis-related proteins Bax and C-cas3 were decreased, but the level of Bcl-2 was increased, after HMGB1 knockdown under hypoxia conditions (Fig. 2E). These data clearly suggested that HMGB1 was involved in CPC apoptosis after hypoxia treatment.

Figure 2.

HMGB1is upregulated in CPCs under hypoxia conditions (A) qRT-PCR and western blot analysis of mRNA and protein level of HMGB1 in CPCs after hypoxia treatment. (B) qRT-PCR and western blot analysis identified the effectiveness of siRNA against HMGB1 in CPCs. (C) CCK8 assay analysis of the cell viability in CPCs transfected with siHMGB1 after hypoxia treatment. (D) EdU incorporation assay was used to detect the proliferation potential of CPCs after HMGB1 knockdown. (E) Western blot analysis of apoptosis protein level in CPCs after hypoxia treatment. siNC was the negative siRNA. The CPCs were cultured under hypoxia conditions for 24 h and then reperfused with normxia. *P < 0.05 and **P < 0.01. All experiments were performed in three biological repeats. β-Actin was used as the control.

DNMT1 inhibits HMGB1 expression in hypoxia-induced apoptosis in CPCs

As is known, epigenetic modifications can affect the balance between pro- and anti-apoptotic signaling pathways and then influence cell apoptosis and proliferation [27]. We thus investigated whether DNA methylation is involved in regulating HMGB1-mediated CPC apoptosis after hypoxia treatment. 5-Aza-2′-deoxycytidine (AZA) is a chemical epigenetic modifier which can inhibit DNA methyltransferase activity and then results in DNA demethylation. As showed in Fig. 3A, AZA treatment significantly increased HMGB1 expression level both at mRNA and protein levels. In addition, the expression of DNMT1, but not DNMT3a or DNMT3b, was inhibited by AZA, suggesting that it is DNMT1 that may regulate HMGB1 expression. Indeed, qRT-PCR and western blot analysis results showed that overexpressed DNMT1 can down-regulate HMGB1 expression (Fig. 3B,C). To confirm the regulation of HMGB1 by DNMT1, we designed siRNAs specific to DNMT1. qPCR and western blot analysis demonstrated that the DNMT1 siRNAs could repress the DNMT1 expression (Fig. 3D). Then, the qPCR and western blot analysis also showed that the HMGB1 protein levels were elevated in cells transfected with DNMT1 siRNAs (Fig. 3E). These observations suggested that HMGB1 was regulated by DNMT1.

Figure 3.

DNMT1inhibitsHMGB1expression in CPCs (A) qRT-PCR and western blot analysis of mRNA and protein levels of HMGB1, DNMT1, DNMT3a and DNMT3b in CPCs after AZA treatment. (B) qRT-PCR and western blot analysis of mRNA and protein level of DNMT1 in CPCs transfected with pcDNA3.1-DNMT1 plasmid. (C) qRT-PCR and western blot analysis of mRNA and protein level of HMGB1 in CPCs transfected with pcDNA3.1-DNMT1 plasmid. (D) qRT-PCR and western blot analysis of mRNA and protein level of DNMT1 in CPCs transfected with siDNMTs and siNC. (E) qRT-PCR and western blot analysis of mRNA and protein level of HMGB1 in CPCs transfected with siDNMTs and siNC. *P < 0.05 and **P < 0.01. All experiments were performed in three biological repeats. β-Actin was used as the control.

Then, we further explored whether the regulation of HMGB1 by DNMT1 exists in regulating CPC apoptosis and viability after hypoxia stimuli. qRT-PCR and western blot analysis results illustrated that DNMT1 expression was reduced in hypoxia-induced CPC apoptosis model (Fig. 4A). Though HMGB1 expression was promoted by hypoxia stimuli, it was reduced when DNMT1 was overexpressed (Fig. 4B). DNA methylation leads to transcriptional silencing in variety of physiological and pathological progresses. We therefore proposed that the up-regulation of HMGB1 might be due to the alteration of DNA methylation induced by inhibition of DNMT1 in apoptotic CPCs after hypoxia stimuli. By detecting the methylation state of CpGs in the upstream HMGB1 promoter region, we found that the CpG methylation level in CPCs was reduced after hypoxia treatment, while DNMT1 overexpression significantly restored the CpG methylation level (Fig. 4C). These data indicated that DNMT1 can modulate the HMGB1 expression by affecting the CpG methylation in HMGB1 promoter region in CPCs under hypoxia stimuli.

Figure 4.

DNMT1inhibitsHMGB1expression in hypoxia-induced apoptosis in CPCs (A) qRT-PCR and western blot analysis of mRNA and protein level of DNMT1 in CPCs after hypoxia treatment. (B) qRT-PCR and western blot analysis of mRNA and protein level of HMGB1 in CPCs transfected with pcDNA3.1-DNMT1 plasmid after hypoxia treatment. (C) Bisulfite pyrosequencing analysis of the DNA methylation status of the HMGB1 promoter region in CPCs treated with hypoxia or transfected with pcDNA3.1-DNMT1 plasmid. The CPCs were cultured under hypoxia conditions for 24 h and then reperfused with normxia. *P < 0.05 and **P < 0.01. All experiments were performed in three biological repeats. β-Actin was used as the control.

MAPK signaling pathway is involved in DNMT1/HMGB1-mediated hypoxia-induced apoptosis in CPCs

Mitogen-activated protein kinases (MAPKs) have been reported to involve in modulating cell apoptosis and proliferation [28]. Therefore, further study was carried out to identify the possible role of MAPKs in DNMT1/HMGB1-mediated CPC apoptosis under hypoxia treatment. Since JNK, ERK, and p38 are three main MAPK subfamilies, we detected the protein levels and phosphorylation statuses of JNK, ERK, and p38. Western blot analysis results showed that p-JNK, p-ERK and p-p38 levels were increased in CPCs under hypoxia conditions (Fig. 5A). To further explore whether p-JNK, p-ERK and p-p38 affect DNMT1 and HMGB1 expression, we treated the CPCs with SD100625 (p-JNK inhibitor), PD98059 (p-ERK inhibitor) and SB203580 (p-p38 inhibitor). Indeed, the phosphorylation levels of p-JNK, p-ERK and p-p38 were significantly decreased in CPCs after treatment with these inhibitors (Fig. 5B). It was also found that all these three inhibitors can reduce HMGB1 expression but increase DNMT1 expression (Fig. 5C). Additionally, the cell viability and proliferation potential were significantly elevated by these three inhibitors (Fig. 5D,E). Meanwhile, the apoptosis protein Bax was decreased, while the antiapoptosis protein Bcl-2 was increased (Fig. 5F). Taken together, these data suggested that MAPK signaling pathway was involved in the regulation of DNMT1/HMGB1-mediated CPC apoptosis induced by hypoxia stimuli.

Figure 5.

MAPK signaling pathway is involved inDNMT1/HMGB1-mediated hypoxia-induced apoptosis in CPCs (A) Western blot analysis of protein levels and phosphorylation statuses of JNK, ERK, and p38 in CPCs after hypoxia treatment. (B) Western blot analysis of protein levels and phosphorylation statuses of JNK, ERK, and p38 in CPCs after p-JNK, p-ERK, and p-p38 inhibitors treatments in CPCs under hypoxia conditions. (C) Western blot analysis of HMGB1 and DNMT1 expression levels in CPCs after treatment with p-JNK, p-ERK, and p-p38 inhibitors under hypoxia conditions. (D) CCK8 assay analysis of the cell viability in CPCs after treatment with p-JNK, p-ERK, and p-p38 inhibitor. (E) EdU incorporation assay was used to detect the proliferation potential of CPCs after treatment with p-JNK, p-ERK, and p-p38 inhibitors under hypoxia conditions. (F) Western blot analysis of apoptosis protein level in CPCs after treatment with p-JNK, p-ERK, and p-p38 inhibitors. The CPCs were cultured under hypoxia conditions for 24 h and then reperfused with normoxia. *P < 0.05 and **P < 0.01. All experiments were performed in three biological repeats. β-Actin was used as the control.

Discussion

Recently, the process of programmed cell death has been extensively studied in various cardiovascular diseases [13,29,30]. In our study, in order to develop new therapy targets for MI patients, we explored the underlying mechanism of CPC apoptosis after stress stimuli, especially ischemia. We monitored the in vitro ischemia injury in the CPCs by culturing the CPCs under 0.5% O₂ conditions and then re-perfusing the CPCs with normxia (21% O₂ concentration) conditions. The hypoxia treatment can effectively induce the apoptosis of CPCs. Meanwhile, the viability of CPCs, as well as the expressions of the apoptosis-related proteins were reduced. These results are consistent with a previous study showing that hypoxia stimuli are the inducer of apoptosis in various cell types and tissues.

It has been reported that HMGB1 is involved in inflammation and cell death after liver I/R stimuli, which is mediated by MAP kinase and NF-κB signaling pathways [31]. In the heart, the patients with acute coronary syndrome have higher levels of serum HMGB1 [32]. Additionally, the HMGB1 level was found to be immediately elevated in mouse heart after hypoxia in vitro or in vivo ischemic injury, which leads to the activation of pro-inflammatory pathways and then promotes myocardial injury [33]. We demonstrated that HMGB1 was significantly elevated in CPCs after hypoxia treatment. Meanwhile, knockdown of HMGB1 in CPCs can attenuate hypoxia-induced apoptosis, suggesting that HMGB1 plays an important role in the apoptosis and proliferation of CPCs.

Methylation of cytosines in CpG dinucleotides results in the transcriptional inactivity in mammalian genomes [34]. In our work, we further proved that DNMT1 can modulate HMGB1 expression by affecting the CpG methylation of its promoter region in CPCs after hypoxia treatment. Meanwhile, the regulation of HMGB1 by DNMT1 was mediated by the MAPK signaling pathway. Our data clearly confirmed that epigenetic modifications can affect HMGB1 expression and then regulate the biological property of the CPCs.

In summary, we found that the expression level of HMGB1 was upregulated in CPCs under hypoxia conditions. Further study demonstrated that HMGB1 was regulated by DNMT1 via changing the methylation state of CpGs in the promoter of HMGB1 in CPCs during the hypoxia process. Additionally, MAPK signaling pathway was involved in regulating DNMT1/HMGB1-mediated CPC apoptosis in hypoxia process. Our findings provide a novel mechanism for the apoptosis and proliferation of CPCs, which is important for the development of effective therapeutic targets for MI patients.

Funding

This work was supported by the grants from the Shanghai Pudong New Area Health Planning Commission Industry Special Project (No. PW2013E-2), the Shanghai Municipal Health Planning Commission Key Specialist Project (No. ZK2015A13), the Shanghai Pudong New Area Science Committee Project (No. PKJ2017-Y42), and the National Natural Science Foundation of China (No. 81560077).

Author notes

Jinwen Su and Ming Fang contributed equally to this work.