Introduction

Cardiac hypertrophy is regarded as an adaptive physiological response of the heart to external stress stimuli, such as pressure overload. However, prolonged cardiac hypertrophy leads to pathological ventricular remodeling and develops into heart failure (HF), which is the final stage of cardiac diseases with high incidence and mortality worldwide [1]. Pathological cardiac hypertrophy is accompanied with enlarged cardiomyocytes area and fetal genes program reactivation, resulting in adverse cardiac remodeling and declined myocardium function [2]. Currently, the combination of angiotensin converting enzyme inhibitors (ACEIs), β-blockers, and aldosterone receptor antagonists are used as the main treatment regimen, however, the overall curative effects are poor. Clearly, new intervention targets are urgently needed to attenuate the progress of pathological cardiac hypertrophy.

ER stress, caused by the accumulation of misfolded or unfolded proteins, is a common feature accompanied with the cardiovascular diseases progression [3]. The unfolded protein response (UPR), which is elicited by ER stress, activates three transmembrane sensors: Protein Kinase R-like ER Kinase (PERK), Inositol Requiring Enzyme 1α (IRE1α) and the Activating Transcription Factor 6 (ATF6) [4]. Among them, IRE1α is the most highly conserved and could splice 26 nucleotides from the un-spliced XBP1 (uXBP1) mRNA, resulting in frameshift and forming the spliced XBP1 (sXBP1), which functions via a variety of transcriptional targets and participates in numerous cellular stress responses [18]. Therefore, we were curious to determine whether PARP16 regulates GATA4 expression in response to TAC surgery or PE treatment. Here we found that GATA4 was significantly increased in both PE-treated NRCMs and TAC-operated heart tissue while downregulated by PARP16 deficiency (Fig. 8a–c). Similar patterns were observed with the immunofluorescence staining result (Fig. 8d). Furthermore, GATA4 was also activated directly by PARP16 OE, as evidenced by the high protein level and immunofluorescence intensity (Fig. 8e, f). To further confirm the influence of PARP16 on GATA4 localization, nuclear and cytoplasmic proteins were extracted and found PARP16 knockdown markedly decreased the expression of GATA4 in nucleus but not cytoplasm upon PE treatment (Fig. 8g). Collectively, these above results suggested that PARP16 regulates the expression of GATA4 in response to the cardiomyocyte hypertrophic growth.

Fig. 8
figure 8

PARP16 activated GATA4 via IRE1α–sXBP1 pathway in cardiac hypertrophy. a The protein expression of GATA4 in PE-treated NRCMs. b Knockdown of PARP16 abolished the activation of GATA4 in PE-treated NRCMs by western blot assay. c The protein expression of GATA4 was also regulated in heart sections from WT mice injected with AAV9-GFP or AAV9-shPARP16 and subsequently subjected to sham or TAC surgery by immunoblot analysis. d Representative immunofluorescence staining images of GATA4 in PARP16 knockdown followed by PE treatment. scale bar: 50 μm. e PARP16 OE directly activated the expression of GATA4 in NRCMs. f Representative immunofluorescence double staining images of PARP16 and GATA4 in PARP16 OE-transfected H9C2 cells. Scale bar: 50 μm. g The effects of PARP16 knockdown on the nuclear and cytoplasm distribution and expression of GATA4 by western blot. h, i The effects of the ER stress inhibitor (TUDCA) and the specific IRE1α inhibitor STF-083010 on the expressions of GATA4 as well as the hypertrophic markers ANP and BNP in response to PE treatment. j Knockdown of IRE1α abolished the activation of GATA4 or hypertrophic marker BNP in PE-treated NRCMs. k The effects of the specific IRE1α inhibitor STF-083010 on the protein expression of GATA4 as well as the hypertrophic marker ANP and BNP upon PARP16 OE treatment. l The effects of STF-083010 on the mRNA level of GATA4 upon PARP16 OE treatment. Data are presented as mean ± SD and analyzed using Student t test or one-way ANOVA followed by Tukey Post-hoc tests. n = 3. *p < 0.05, **p < 0.01, ***p < 0.001

Full size image

Interestingly, we have uncovered that PARP16 is an ER transmembrane protein that could interact with IRE1α and activated the IRE1α–sXBP1 pathway during PE-stimulated hypertrophic growth, and GATA4 was a transcription factor primarily located in the nucleus [19], also as evidenced in Fig. 8d, f, g. These outcomes led us to hypothesize that PARP16 may not directly mediate GATA4 activation and maybe through the IRE1α–sXBP1 pathway. To test this hypothesis, both tauroursodeoxycholic acid (TUDCA) (an ER stress inhibitor) and STF-083010 (a specific IRE1α inhibitor) were used to examine the relationship between IRE1α–sXBP1 pathway and GATA4. Indeed, treatment with either TUDCA or STF-083010 could blunt the expressions of GATA4 as well as the hypertrophic marker ANP and BNP in response to PE treatment (Fig. 8h, i). In addition, silencing IRE1α through siRNA upon PE stimulation decreased the protein expressions of GATA4 as well as the hypertrophic marker BNP in NRCMs (Fig. 8j). Importantly, the specific IRE1α inhibitor STF-083010 also attenuated PARP16 OE-induced upregulation of GATA4 both in protein and mRNA levels (Fig. 8k, l), suggesting that PARP16 mediated GATA4 upregulation maybe at least in part via the IRE1α–sXBP1 pathway.

The expression of PARP16 is mediated by JNK/ERK MAPK pathway in PE-induced cardiomyocyte hypertrophy

To further identify the upstream regulatory pathway of PE-induced PARP16 upregulation, we examined mitogen-activated protein kinase (MAPK) signaling pathway, a classical pathway involved in cardiac hypertrophy [20]. Our results showed that the phosphorylation of MAPK pathway members, including JNK, ERK1/2, except p38 MAPK were increased upon PE challenge in vitro (Fig. S6a). Then the specific inhibitors of JNK, ERK1/2 and p38 MAPK, namely SP600125, PD98059 and SB203580 were used and found the expression of PARP16 was inhibited by JNK and ERK1/2 inhibitors but not p38 MAPK inhibitors (Fig. S6b), suggesting that PE-induced PARP16 upregulation maybe mediated by JNK/ERK MAPK pathway.

Discussion

Pathological cardiac hypertrophy is regarded as an independent risk factor of HF, whose current main therapeutic interventions are still a bottleneck problem [1]. Thus, a deeper understanding of the mechanisms involved in pathological cardiac hypertrophy is necessary for exploring the potential drug targets. In the present study, we determined that PARP16 was a key driver of pathological cardiac hypertrophy. PARP16 deficiency rescued cardiac dysfunction and ameliorated TAC-induced cardiac hypertrophy and fibrosis in mice, as well as PE-induced cardiomyocyte hypertrophic responses. Whereas overexpression of PARP16 exacerbated the hypertrophic responses. Mechanistically, PARP16 interacted with IRE1α and ADP-ribosylated IRE1α and mediated the hypertrophic responses through activating the IRE1α–sXBP1–GATA4 pathway. Collectively, our results implicated that PARP16 may be a contributor of pathological cardiac hypertrophy at least partly via activating the IRE1α–sXBP1–GATA4 pathway, and may be regarded as a new potential target for exploring effective therapeutic interventions of pathological cardiac hypertrophy and HF.

The poly (ADP-ribose) polymerases (PARPs) family members such as PARP-1 and PARP-2 has been revealed to participate in the progression of cardiac hypertrophy due to different cellular localization and regulating mechanisms. PARP-1 contributes to caspase-independent myocyte cell death during heart failure [21] while knockdown of PARP-2 protects against cardiac hypertrophy via SIRT1 activation [22]. However, PARP16 is the only PARP family member to be located on the endoplasmic reticulum and its role in cardiovascular diseases remains to be clarified. Recently, emerging evidence has demonstrated that PARP16 is identified as a novel target for vascular diseases, such as vascular aging and neointimal hyperplasia related diseases [11, 12]. Consistent with this, we were curious about the effect of PARP16 on pathological cardiac hypertrophy and observed PARP16 was at a higher level in response to pressure overload-induced pathological cardiac hypertrophy. PARP16 deficiency rescued cardiac dysfunction and ameliorated TAC-induced cardiac hypertrophy and fibrosis as well as PE-induced cardiomyocyte hypertrophic responses. Whereas overexpression of PARP16 exacerbated the hypertrophic responses, implicating that PARP16 may be a contributor of pathological cardiac hypertrophy.

It is also well documented that GATA transcription factors have important roles to promote ER integrity and Gata4 and Gata6 knockout mice could lead to several ER stress signals activation, such as CHOP [23]. However, the relationship between ER stress and GATA4 hasn't been explained well in pathological cardiac hypertrophy. In our study, we provided evidence that PARP16 could regulate the transcription expression of GATA4 to some extent depending on the IRE1α–sXBP1 pathway by using two molecular inhibitors (TUDCA, an ER stress inhibitor and STF-083010, a specific IRE1α inhibitor) as well as IRE1α siRNA. Indeed, treatment with either TUDCA or STF-083010 could blunt the expression of GATA4 in response to PE or PARP16 OE treatment. Furthermore, knockdown of IRE1α with siRNA significantly decreased the protein expression of GATA4 and BNP in PE-induced NRCMs, suggesting that PARP16 mediated expression of GATA4 at least in part via the IRE1α–sXBP1 pathway. Moreover, GATA4 acts as a key zinc finger-containing transcription factor of numerous cardiac-specific genes including Nppa and Nppb [16, 24] and ablation of GATA4 by siRNAs decreased ANP expression [25], therefore PARP16 contribute to the upregulation of ANP and BNP at least partly via activating the IRE1α–sXBP1–GATA4 pathway, further promote the progress of cardiac hypertrophy.

Due to the limited clinic resources, further verifications were not carried out in human samples, which is the deficiency of this study. Another limitation of this study are that we did not construct cardiac-specific knockout mice or cardiac-specific targeting adenoviruses, which should be explored in future studies.