Hydrogen Sulfide Attenuates Angiotensin II-Induced Cardiac Fibroblast Proliferation and Transverse Aortic Constriction-Induced Myocardial Fibrosis through Oxidative Stress Inhibition via Sirtuin 3

Sirtuin 3 (SIRT3) is critical in mitochondrial function and oxidative stress. Our present study investigates whether hydrogen sulfide (H2S) attenuated myocardial fibrosis and explores the possible role of SIRT3 on the protective effects. Neonatal rat cardiac fibroblasts were pretreated with NaHS followed by angiotensin II (Ang II) stimulation. SIRT3 was knocked down with siRNA technology. SIRT3 promoter activity and expression, as well as mitochondrial function, were measured. Male wild-type (WT) and SIRT3 knockout (KO) mice were intraperitoneally injected with NaHS followed by transverse aortic constriction (TAC). Myocardium sections were stained with Sirius red. Hydroxyproline content, collagen I and collagen III, α-smooth muscle actin (α-SMA), and dynamin-related protein 1 (DRP1) expression were measured both in vitro and in vivo. We found that NaHS enhanced SIRT3 promoter activity and increased SIRT3 mRNA expression. NaHS inhibited cell proliferation and hydroxyproline secretion, decreased collagen I, collagen III, α-SMA, and DRP1 expression, alleviated oxidative stress, and improved mitochondrial respiration function and membrane potential in Ang II-stimulated cardiac fibroblasts, which were unavailable after SIRT3 was silenced. In vivo, NaHS reduced hydroxyproline content, ameliorated perivascular and interstitial collagen deposition, and inhibited collagen I, collagen III, and DRP1 expression in the myocardium of WT mice but not SIRT3 KO mice with TAC. Altogether, NaHS attenuated myocardial fibrosis through oxidative stress inhibition via a SIRT3-dependent manner.


Introduction
Myocardial fibrosis is a cardiac interstitial remodeling characterized by excessive cardiac fibroblast proliferation, collagen deposition, and abnormal distribution [1,2]. It is closely related to hypertension, chronic heart failure, hypertrophic cardiomyopathy, dilated cardiomyopathy, viral myocarditis, and many other cardiovascular diseases, which is a potential risk factor of sudden cardiac death [3][4][5][6][7]. However, there is no ideal strategy for the treatment of myocardial fibrosis.
Silent information regulator 2 (SIR2), a highly conserved NAD-dependent family histone deacetylase, serves as a cell sensor for energy utilization and metabolism regulation [23,24]. SIR2 is widely found in mammals and consists of seven members including sirtuin 1 (SIRT1) to sirtuin 7 (SIRT7). The SIR2 family plays a role in metabolism, cancer, and other physiological and pathological processes [25,26]. Sirtuin 3 (SIRT3) is a member of histone deacetylase III to mediate redox signaling [27]. Previous research has demonstrated that H 2 S was capable of increasing SIRT3 to improve mitochondrial function and attenuate oxidative stress. We found that H 2 S improved endothelium-dependent relaxation of aortic and mesenteric arteries in paraquatadministrated wild-type (WT) mice but not SIRT3 knockout (KO) mice [18]. H 2 S reduced superoxide anion production in angiotensin II-(Ang II-) stimulated cardiomyocytes, but this effect is significantly weakened after SIRT3 was knocked down. H 2 S protected against myocardial hypertrophy in WT mice but not SIRT3 KO mice [16]. However, it is not known whether H 2 S protects against cardiac fibroblast proliferation and myocardial fibrosis via SIRT3 activation.
Therefore, our present study investigates whether H 2 S attenuated cardiac fibroblast proliferation in vitro and myocardial fibrosis in vivo and explores the possible role of SIRT3 on the protective effects.

Materials and Methods
2.1. Cardiac Fibroblast Culture and Treatment. After anesthesia by isoflurane, neonatal Sprague Dawley rats aged 1 to 3 days old were killed by decapitation. Then, the left ventricular myocardium was cut into small pieces followed by digestion with trypsin in a constant temperature water bath of 37°C for 5-8 minutes. After about 10 times, all supernatants except the first time were mixed. Next, Dulbecco's modified Eagle's medium (DMEM, Wisent Inc., Montreal, QC, Canada) containing 10% fetal bovine serum (FBS, Wisent Inc., Montreal, QC, Canada) was timely added to stop excessive digestion. After centrifugation, fresh DMEM containing 10% FBS was used to resuspend the cell precipitate. After differential adhesion for 4 h, the cardiomyocytes in the medium were abandoned while confluent cardiac fibroblasts adhered to the walls. Cardiac fibroblasts of the 2 nd or 3 rd passage were used in our further experiments. Cells were then administrated with NaHS (50 μM, Sigma-Aldrich, St. Louis, MO), a common H 2 S donor, for 4 h followed by Ang II (100 nM, Sigma-Aldrich, St. Louis, MO) stimulation for an additional 24 h.

Cell Proliferation and Hydroxyproline Content
Measurement. Cell proliferation was represented as cardiac fibroblast number enhancement, which was determined by Cell Counting Kit-8 (CCK-8, Beyotime, Shanghai, China) as the previous study [4]. The content of hydroxyproline in the cell culture medium and in the myocardium was detected according to our previous research [4,22].

Luciferase Reporter Assay.
Neonatal rat cardiac fibroblasts were transfected with the SIRT3 promoter luciferase fusion plasmid using the Lipofectamine 3000 reagent (Invitrogen, Carlsbad, CA, USA), and the pRL-TK reporter plasmid served as a control reporter. After changing by the fresh medium with 10% FBS 24 h later, cells were incubated with NaHS (50 μM) for 4 h and subsequently stimulated by Ang II (100 nM) for 24 h. Next, the cell lysis buffer was added to harvest the cells. The luciferase activity was detected with a dual-luciferase reporter assay system (Promega, Madison, WI, USA). Activity of the SIRT3 promoter was assessed by the relative luciferase activities compared with pRL-TK, which was normalized to control in triplicate.
2.4. SIRT3 RNA Interference. After starvation for 24 h, the mixture of the SIRT3 siRNA or nonspecific control siRNA (NC siRNA) and the Lipofectamine 3000 reagent was added into cardiac fibroblasts. SIRT3 siRNA (5 ′ -CCAUCUUU-GAACUAGGCUUTT-3 ′ and 5 ′ -AAGCCUAGUUCAAA-GAUGGTT-3′) or NC siRNA with random sequences was commercially synthesized (GenePharma, Shanghai, China). The expression of SIRT3 mRNA and protein was measured by real-time PCR and western blot, respectively, to evaluate the efficiency of SIRT3 siRNA transfection after 48 h. Some other cells were stimulated with NaHS (50 μM) for 4 h followed by Ang II (100 nM) stimulation for 24 h after siRNA was transfected into the cardiac fibroblasts.
2.6. Measurement of Superoxide Formation. After treatment as the above description, the cardiac fibroblasts were incubated with dihydroethidium (DHE, 2 μM, in the Krebs-HEPES buffer) at 37°C without light for 30 min. The level of superoxide in cells, presented as DHE fluorescence intensity, was detected with a laser confocal microscope (Leica, Wetzlar, Germany). Oxidative Medicine and Cellular Longevity Santa Clara, CA, USA) [16]. After being plated in XF-96well plates, the cardiac fibroblasts were subjected to siRNA transfection and treatment as described above. Then, the medium was changed into unbuffered DMEM containing glucose (10 mM), pyruvate (10 mM), and GlutaMAX (2 mM, Invitrogen, Carlsbad, CA, USA). Basal respiration was first obtained. After ATP synthesis was inhibited by oligomycin (Oligo, 2 μg/mL), ATP generation during respiration was measured. Then, respiratory reserve capacity and maximal respiration were detected after uncoupler carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone (FCCP, 2 μM) was added into the cells. Finally, rotenone and antimycin A (rot & AA, 4 μM) were applied to completely block oxygen consumption. WT mice and SIRT3 KO mice were intraperitoneally administrated by NaHS (50 μmol·kg -1 ·day -1 ) once daily, and normal saline (NS) was given as a control. After 2 weeks, mice were intraperitoneally injected with a mixture of ketamine (100 mg·kg -1 ) and xylazine (5 mg·kg -1 ) to induce anesthesia which was confirmed by loss of reflex to toe pinching. Then, thoracotomy was performed to expose the aortic arch, and transverse aortic constriction (TAC) was carried out by tying a 6-0 nylon suture ligature against a 26-gauge needle. The needle was withdrawn immediately to form an incomplete constriction. The mice with the same operating procedures without constriction served as sham controls. After surgery, all mice went on being treated by NaHS or NS for another 2 weeks [28]. During the experiments, systolic blood pressure (SBP) was monitored by a noninvasive blood pressure analysis system (Visitech Systems, Apex, NC, USA) with the tail-cuff method every week.

Assessment of Mitochondrial Respiration
2.10. Determination of H 2 S Concentration in the Plasma and H 2 S Production in the Myocardium. The H 2 S level was determined using a H 2 S-specific microelectrode connected to a free radical analyzer (World Precision Instruments, Sarasota, FL, USA) in light of the previous description [16]. A standard curve was obtained by the current value in different concentrations (0, 0.5, 1, 2, 4, and 8 μM) of Na 2 S stock solution. H 2 S in the myocardium was determined in tissue homogenates. The homogenates were incubated in the assay mixture (500 μL) containing 460 μL tissue homogenates, 20 μL L-cysteine (10 mmol/L), and 20 μL pyridoxal 5 ′ -phosphate (2 mmol/L) at 37°C in tightly sealed vials for 30 min. Then, the plasma or homogenate was dropped into the reaction solution, and there was a significant change in the current value. The H 2 S level in the plasma or the myocardium can be calculated according to the standard curve.
2.11. Sirius Red Staining. After washing 3 times, myocardial paraffin sections about 5 μm thick were stained with saturated picric acid-Sirius red without light for 30 min. Then, the nuclei were lightly stained with Mayer's hematoxylin staining solution. Finally, the collagen deposition in the myocardium was observed and photographed with a microscope. The ratio of the perivascular collagen area (PVCA) to the luminal area (LA) was calculated to assess the degree of perivascular fibrosis in the myocardium. The collagen volume fraction (CVF), as the ratio of the interstitial collagen area to the myocardial area, was statistically analyzed to evaluate the degree of interstitial fibrosis.

Quantitative Real-Time PCR.
Expressions of mRNA were carried out by real-time PCR following the MIQE guidelines. Extracted RNA from cardiac fibroblasts or the myocardium with the TRIzol reagent (Takara, Kyoto, Japan) was subjected to reverse transcription according to the introduction of the PrimeScript™ RT Master Mix Kit (Takara, Kyoto, Japan). Amplification of cDNA was carried out three times in the PCR System (ABI 7500, ABI, Carlsbad, CA, USA) by SYBR (Takara, Kyoto, Japan) with 18S as the housekeeping gene. All sequences of the sense primers and antisense primers are listed below (Table 1).
2.14. Statistical Analysis. All data are shown as mean ± SEM and were analyzed by one-way ANOVA followed by the Bonferroni post hoc test (STATA 15.0). The values of P less than 0.05 were regarded as statistically significant.

NaHS Inhibits Cell Proliferation but Enhances SIRT3
Transcription in Ang II-Stimulated Cardiac Fibroblasts. After 3 Oxidative Medicine and Cellular Longevity NaHS (50 μM) administration for 4 h followed by Ang II (100 nM) stimulation for another 24 h, cell proliferation was evaluated by cell count analysis and hydroxyproline secretion. NaHS pretreatment significantly reduced cell numbers (Figure 1(a)) and reduced hydroxyproline content in the cell culture medium (Figure 1(b)) after Ang II stimulation. These data indicated that NaHS inhibited Ang IIinduced cardiac fibroblast proliferation. Next, we assessed the possible role of SIR2 family members in the attenuated effect of NaHS on cardiac fibroblast proliferation. The mRNA expressions of all seven members, SIRT1-SIRT7, were detected by real-time PCR. SIRT1 and SIRT3 were significantly decreased after Ang II stimulation, while SIRT2, SIRT4, SIRT5, SIRT6, and SIRT7 remained unchanged. Moreover, the decreased SIRT3 was corrected by NaHS preadministration for 4 h. However, there was no significant change in SIRT1 by NaHS, suggesting that SIRT1 was possibly not critical in the inhibitory effect of NaHS on Ang II-induced cardiac fibroblast proliferation. It is worth noting that there was no alteration on SIR2 mRNA of NaHS-treated cardiac fibroblasts without Ang II stimulation (Figure 1(c)). Moreover, NaHS also increased SIRT3 promoter activity in Ang II-stimulated cardiac fibroblasts (Figure 1(d)). Taken together, NaHS enhanced SIRT3 transcription in the inhibitory effect on Ang II-induced cardiac fibroblast proliferation.

NaHS Inhibits Ang II-Stimulated Cardiac Fibroblast
Proliferation via SIRT3. However, whether enhanced SIRT3 was crucial in the above effect remained to be elucidated. In our further study, SIRT3 was knocked down in cardiac fibroblasts by RNA interference technology. After specific SIRT3 siRNA transfection, SIRT3 was significantly reduced at both levels of mRNA and protein expressions (Figures 2(a) and 2(b)).
Then, cell numbers and hydroxyproline content of cardiac fibroblasts were assessed again after SIRT3 was silenced.
Our study showed that NaHS pretreatment significantly decreased cell numbers and reduced hydroxyproline content in Ang II-stimulated cardiac fibroblasts. However, the inhibitory effect was unavailable after SIRT3 was knocked down (Figures 2(c) and 2(d)). These results suggested that NaHS inhibited Ang II-stimulated cardiac fibroblast proliferation in a SIRT3-dependent manner.

NaHS Suppresses Collagen Expression in
Ang II-Stimulated Cardiac Fibroblasts via SIRT3. Next, the expression of collagen I and collagen III was also examined. We found that NaHS inhibited collagen I and collagen III expressions at both the mRNA and protein levels in Ang II-stimulated cardiac fibroblasts. However, the above protective effects by NaHS were absent after SIRT3 was silenced ( Figure 3). It suggested that NaHS suppressed Ang IIstimulated cardiac fibroblast collagen expression in a SIRT3-dependent manner.

NaHS Inhibits α-SMA Expression in Ang II-Stimulated
Cardiac Fibroblasts via SIRT3. The expression of α-SMA is commonly regarded as a robust and sensitive marker of cardiac fibroblast differentiation [4]. The present studies found that the α-SMA expression was enhanced after Ang II stimulation, which was attenuated by NaHS pretreatment. However, the above inhibitory effect of NaHS on α-SMA expression was obviously blunted if SIRT3 was knocked down (Figure 4(a)). It suggested that NaHS inhibited α-SMA expression in Ang II-stimulated cardiac fibroblasts via SIRT3.

NaHS Attenuates Oxidative Stress in Ang II-Stimulated
Cardiac Fibroblasts via SIRT3. Previous studies have verified that excessive oxidative stress was a critical pathophysiological mechanism in several cardiovascular diseases including myocardial remodeling, atherosclerosis, myocardial ischemia-reperfusion injury, and diabetic cardiomyopathy

Gene
Sense primer Antisense primer Oxidative Medicine and Cellular Longevity [29,30]. SIRT3, as a vital acetyl-lysine deacetylase, regulates mitochondrial function and reactive oxygen species (ROS) production [31]. Our present study found that NaHS alleviated DHE fluorescence intensity in Ang II-stimulated cardiac fibroblasts, suggesting that NaHS attenuated Ang IIstimulated ROS production. However, NaHS was no longer able to suppress DHE fluorescence intensity after SIRT3 was knocked down (Figure 4(b)). These data showed that NaHS attenuated oxidative stress in Ang II-stimulated cardiac fibroblasts via a SIRT3-dependent manner.

NaHS Improves Mitochondrial Respiration Function and
Membrane Potential in Ang II-Stimulated Cardiac Fibroblasts via SIRT3. The above attenuated effects of NaHS on cardiac fibroblast proliferation suggested that mitochondrial function might also be improved by NaHS via SIRT3.
To verify the hypothesis, the oxygen consumption rate (OCR) in Ang II-stimulated cardiac fibroblasts after NaHS administration with SIRT3 silencing was measured. Our study showed that NaHS significantly improved basal respi-ration, ATP generation, respiratory reserve capacity, and maximal respiration in Ang II-stimulated cardiac fibroblasts, which was unavailable after SIRT3 silencing (Figures 4(c) and 4(d)). These data suggested that NaHS improved mitochondrial respiration function in Ang II-stimulated cardiac fibroblasts via SIRT3. Next, the mitochondrial membrane potential of cardiac fibroblasts was measured by JC-1 staining. Cells with strong red fluorescence intensity indicated the normal mitochondrial membrane potential, while green fluorescence indicated a decreased one. Our results showed that NaHS increased red, but decreased green, fluorescence intensity in Ang II-stimulated cardiac fibroblasts, whereas the effect was significantly alleviated after SIRT3 was knocked down (Figure 4(e)). It suggested that NaHS improved mitochondrial membrane potential in Ang II-stimulated cardiac fibroblasts via a SIRT3-dependent manner. Relative promoter activity of SIRT3 was detected with a dual-luciferase reporter assay system. * * P < 0:01 as compared with untreated cells; ## P < 0:01 as compared with Ang II alone-stimulated cells. n = 6.

Oxidative Medicine and Cellular Longevity
showed that NaHS inhibited Ang II-induced cardiac fibroblast proliferation in a SIRT3-depended manner in vitro. Next, the role of SIRT3 in the protective effects of NaHS was investigated in mice. After TAC, SBP was significantly increased in WT and SIRT3 KO mice to the same extent in the next weeks. NaHS administration reduced SBP in both the WT and SIRT3 knockout mice ( Figure 5(a)). H 2 S concentration in the plasma and H 2 S production in the myocardium were impaired after TAC, which was restored by the exogenous NaHS supplement in both WT mice and SIRT3 KO mice (Figures 5(b) and 5(c)). SIRT3 expression was decreased in the myocardium of WT mice after TAC, which was elevated by NaHS. And there was no SIRT3 expression in the myocardium of SIRT3 KO mice ( Figure 5(d)). These data suggested that NaHS regulated blood pressure and H 2 S levels regardless of the presence or absence of the SIRT3 gene.

NaHS Ameliorates Perivascular and Interstitial Collagen
Deposition in the Myocardium of WT Mice but Not SIRT3 KO Mice with TAC. Perivascular fibrosis and interstitial fibrosis are the main manifestations of myocardial fibrosis [32]. Collagen, shown in red by Sirius red staining, was significantly suppressed by NaHS in WT mice but not SIRT3 KO mice with TAC ( Figure 6(a)). Statistical analysis showed that after TAC, both the ratio of PVCA to LA and the CVF in SIRT3 KO mice were higher than that in WT mice. NaHS decreased the ratio of PVCA to LA and reduced CVF in WT mice with TAC, which was unavailable in SIRT3 KO mice (Figures 6(b) and 6(c)). It indicated that NaHS ameliorated perivascular and interstitial collagen deposition in the myocardium of WT mice but not SIRT3 KO mice with TAC.  Figure 2: NaHS inhibits Ang II-stimulated cardiac fibroblast proliferation via SIRT3. (a, b) After SIRT3 siRNA or NC siRNA was transfected into neonatal rat myocardial fibroblasts for 48 h, expression of SIRT3 mRNA and protein was measured by real-time PCR and western blot, respectively. * * P < 0:01 as compared with cells with NC siRNA transfection. n = 6. (c) After SIRT3 siRNA or NC siRNA was transfected into cardiac fibroblasts for 24 h, the cells were pretreated with NaHS (50 μM) for 4 h followed by Ang II (100 nM) stimulation for another 24 h. The number of cardiac fibroblasts was detected with CCK-8. (d) The content of hydroxyproline in the cell culture medium was measured. * * P < 0:01 as compared with untreated cells with NC siRNA transfection; ## P < 0:01 as compared with Ang II alone-stimulated cells with NC siRNA transfection; && P < 0:01 as compared with untreated cells with SIRT3 siRNA transfection. n = 6.

Oxidative Medicine and Cellular Longevity
TAC. The data in vitro suggested that NaHS reduced collagen expression in Ang II-stimulated cardiac fibroblasts via a SIRT3-dependent manner. Next, the possible protective effect in vivo was further investigated. We found that after TAC, there was more hydroxyproline content and collagen expression in SIRT3 KO mice than in WT mice. NaHS reduced hydroxyproline content and collagen expression in the myocardium of WT mice with TAC. However, there was no inhibitory effect of NaHS on the above indexes in SIRT3 KO mice with TAC (Figures 7(a)-7(c)). Similarly, there was more α-SMA expression in SIRT3 KO mice with TAC than in WT mice. NaHS suppressed α-SMA expression in the myocardium of WT mice but not SIRT3 KO mice with TAC (Figure 7(d)). Taken together, NaHS alleviated myocardial fibrosis in mice with TAC via a SIRT3dependent manner.

NaHS Restores DRP1 Expression in the Cardiac
Fibroblasts with Ang II Stimulation and in the Myocardium of Mice with TAC via SIRT3. DRP1 is a protein associated with mitochondrial fission, which may induce structural damage and functional dysfunction of mitochondria [33].
Our present results showed that NaHS inhibited DRP1 expression in Ang II-stimulated cardiac fibroblasts. However, the inhibitory effect by NaHS was significantly weakened after SIRT3 was knocked down (Figures 8(a) and  8(b)). After TAC, there was more expression of DRP1 in the myocardium of SIRT3 KO mice than in WT mice. NaHS    Oxidative Medicine and Cellular Longevity suppressed DRP1 expression in the myocardium of WT mice but not SIRT3 KO mice with TAC (Figure 8(c)). All these data verified that NaHS restored DRP1 expression in the cardiac fibroblasts with Ang II stimulation and in the myocardium of mice with TAC via a SIRT3-dependent manner.

Discussion
Myocardial fibrosis is the process of extracellular matrix remodeling to significantly increase myocardial stiffness and eventually results in heart failure and even sudden cardiac death [34].    Figure 5: NaHS decreases blood pressure but restores H 2 S levels and SIRT3 expression in mice with TAC. After intraperitoneal injection by NaHS (50 μmol·kg -1 ·day -1 ) or normal saline (NS) for 2 weeks, male wild-type (WT) mice and SIRT3 knockout (SIRT3 KO) mice were subjected to transverse aortic constriction (TAC) surgery. NaHS or NS was administrated for another 2 weeks. (a) The level of SBP in WT mice and SIRT3 KO mice was monitored every week after NaHS administration. (b) H 2 S concentration in the plasma was measured. (c) H 2 S production in the myocardium was detected. (d) Expression of SIRT3 protein in the myocardium was measured by western blot. * * P < 0:01 as compared with WT+Sham; # P < 0:05 and ## P < 0:01 as compared with WT+TAC; && P < 0:01 as compared with SIRT3 KO+Sham; $$ P < 0:01 as compared with SIRT3 KO+TAC. n = 6.
effects on myocardial ischemia-reperfusion injury, cardiac infarction, arrhythmia, cardiac hypertrophy, heart failure, and diabetic cardiomyopathy [35][36][37]. It is noted that H 2 S suppressed cardiac fibroblast proliferation by inhibiting K + currents or channels and blocking the transformation of atrial fibroblasts into myoblasts [38]. NaHS administration ameliorated myocardial fibrosis via inhibiting cell aging in diabetic rats [39]. H 2 S alleviated myocardial fibrosis and restored cardiac function in both the CSE KO and WT mice with myocardial infarction [40]. In our present experiment, we verified that NaHS inhibited cell numbers, hydroxyproline content, α-SMA expression, and collagen production but enhanced SIRT3 expression in the cardiac fibroblasts and the myocardium. It suggested the inhibitory effect of H 2 S on both Ang II-induced cardiac fibroblast proliferation and TAC-induced myocardial fibrosis. Moreover, NaHS equally reduced blood pressure but enhanced the H 2 S level in all mice, suggesting that the failure of NaHS to inhibit myocardial fibrosis in SIRT3 KO mice is due to SIRT3 deficiency but not blood pressure.
The key role of SIRT3 in mitochondrial function has been intensively investigated. However, the possible role of SIRT3 on the protective effects of H 2 S against myocardial fibrosis was unknown. Our previous research studies have shown that H 2 S increased SIRT3 expression in human umbilical vein endothelial cells and cardiomyocytes [16,18,41]. Other groups found that H 2 S improved cardiac energy substrate metabolism via SIRT3 in db/db mice [42,43]. H 2 S activated SIRT3 through S-sulfhydration to attenuate cisplatin-induced acute kidney injury [44]. H 2 S also attenuated hydrogen peroxide-induced NLR family pyrin domain-containing protein 3 (NLRP3) inflammasome activation via SIRT3 in macrophages [45]. Moreover, we verified that SIRT3 deficiency exacerbated diabetic cardiomyopathy and delayed diabetic skin wound healing [46,47]. In our present experiment, mRNA expression of SIRT1 and SIRT3 was decreased, while other subtypes of the SIR2 family did not change significantly in cardiac fibroblasts with Ang II stimulation. Moreover, NaHS did increase SIRT3 transcription. Therefore, SIRT3 was focused on the possible pathway (c) Interstitial fibrosis of the myocardium was assessed by the collagen volume fraction (CVF). * * P < 0:01 as compared with WT+Sham; # P < 0:05 or @ P < 0:05 and ## P < 0:01 or @@ P < 0:01 as compared with WT+TAC; && P < 0:01 as compared with SIRT3 KO +Sham. n = 6.
11 Oxidative Medicine and Cellular Longevity of the protective effect on myocardial fibrosis next. We found that there was more serious fibrosis in SIRT3 KO mice after TAC, suggesting the protective role of SIRT3 in TACinduced myocardial fibrosis. Cardiac fibroblasts with SIRT3 knockdown and mice with SIRT3 deficiency further verified the protective effect on myocardial fibrosis by H 2 S via a SIRT3-depended manner. It is beneficial to clarify the mechanism of H 2 S against myocardial fibrosis.
It is worth noting that the researchers know little about the detailed mechanism of how H 2 S regulated SIRT3 transcription. Our previous study verified that H 2 S increased activator protein 1 (AP-1) binding activity with the SIRT3 promoter to enhance SIRT3 transcription in hydrogen peroxide-(H 2 O 2 -) stimulated EA.hy926 endothelial cells [18]. Others demonstrated that the protective effect of H 2 S in paraquat-induced liver injury was at least partly attributed to nuclear factor erythroid 2-related factor 2-(Nrf2-) dependent SIRT3 gene transcription [48]. H 2 S can induce Ssulfhydration on specific cysteine residues of target proteins to alter protein function and signal transduction [13]. Some studies found that H 2 S S-sulfhydrated c-Jun of AP-1 to enhance the transcriptional activity on SIRT3, which contributed to decrease ROS production in H 2 O 2 -stimulated macrophages [45]. However, the mechanism of how H 2 S  Figure 7: NaHS reduces collagen and α-SMA expressions in the myocardium of WT mice but not SIRT3 KO mice with TAC. After intraperitoneal injection by NaHS (50 μmol·kg -1 ·day -1 ) or NS for 2 weeks, male WT mice and SIRT3 KO mice were subjected to TAC surgery. NaHS or NS was administrated for another 2 weeks. (a) The content of hydroxyproline in the myocardium was measured. (b, c) Expression of collagen I and collagen III mRNA in the myocardium was measured by real-time PCR. (d) Expression of α-SMA protein in the myocardium was measured by western blot. * * P < 0:01 as compared with WT+Sham; ## P < 0:01 or @ P < 0:05 as compared with WT+TAC; && P < 0:01 as compared with SIRT3 KO+Sham. n = 8.

12
Oxidative Medicine and Cellular Longevity regulates SIRT3 transcriptional activity during myocardial fibrosis is not known at present and needs to be confirmed in further studies.
SIRT3 plays a vital role in mitochondrial biosynthesis and oxidative stress, which might be critical in the process of myocardial fibrosis [49]. Several studies also indicated  Figure 8: NaHS restores DRP1 expression in the cardiac fibroblasts with Ang II stimulation and in the myocardium of mice with TAC via SIRT3. (a) After SIRT3 siRNA or NC siRNA was transfected into neonatal rat cardiac fibroblasts for 24 h, the cells were pretreated with NaHS (50 μM) for 4 h followed by Ang II (100 nM) stimulation for another 24 h. DRP1 expression in cardiac fibroblasts was detected by immunofluorescence with Cy3 (red)-conjugated IgG. The nuclei were stained using DAPI (blue). Bar = 25 μm. (b) Expression of DRP1 protein was measured by western blot. * * P < 0:01 as compared with untreated cells with NC siRNA transfection; ## P < 0:01 as compared with Ang II alone-stimulated cells with NC siRNA transfection; & P < 0:05 as compared with untreated cells with SIRT3 siRNA transfection. n = 6. (c) After intraperitoneal injection by NaHS (50 μmol·kg -1 ·day -1 ) or NS for 2 weeks, male WT mice and SIRT3 KO mice were subjected to TAC surgery. NaHS or NS was administrated for another 2 weeks. Expression of DRP1 protein in the myocardium was measured by western blot. * * P < 0:01 as compared with WT+Sham; # P < 0:05 or @@ P < 0:01 as compared with WT +TAC; && P < 0:01 as compared with SIRT3 KO+Sham. n = 6.
that H 2 S was capable of attenuating oxidative stress and mitochondrial dysfunction [16,50]. H 2 S inhibited oxidative stress in the myocardium of chronic heart failure [51]. Studies have shown that impaired mitochondrial permeability transition pores reduced mitochondrial membrane potential depolarization and inhibited the activation of proapoptotic proteins [52]. It has been reported that H 2 S effectively improved mitochondrial membrane potential in high glucose-stimulated human umbilical vein endothelial cells [53]. This may be one of the important mechanisms for protection against mitochondrial function by H 2 S [54]. We found that NaHS significantly inhibited Ang II-induced ROS production, improved mitochondrial respiration function, and restored membrane potential in a SIRT3dependent manner. However, the detailed mechanism of the SIRT3-mediated inhibitory effect on oxidative stress by H 2 S was not known well now. Mitochondria are highly dynamic organelles undergoing cycles of fusion and fission to modulate the morphology, distribution, and function. DRP1, a key protein to regulate mitochondrial fission, is related to clearing the damaged mitochondria and maintaining the process of cellular and organ dynamics [55]. However, an excessive increase of DRP1 will break the balance between mitochondrial fusion and fission to cause mitochondrial dysfunction. Therefore, mitochondrial fission manipulated by targeting DRP1 has been an appealing therapeutic strategy for cytoprotection [56]. We found that NaHS corrected the excessive enhancement of DRP1 in both the cardiac fibroblasts with Ang II stimulation and the myocardium with TAC via a SIRT3mediated pathway. It clarified a novel mechanism for H 2 S and proposed an alternative approach for mitochondrial function protection.
In conclusion, NaHS, a H 2 S donor, enhanced SIRT3 transcription and decreased the DRP1 level to possibly ameliorate mitochondrial membrane rupture, suppress oxidative stress, and alleviate Ang II-induced cardiac fibroblast proliferation and TAC-induced myocardial fibrosis. An illustration of the mechanism is outlined in Figure 9. However, these protective effects of H 2 S were unavailable if SIRT3 was silenced in cells or deficient in mice. These results have shed new light on the molecular mechanism responsible for the cardioprotective effect of H 2 S against myocardial fibrosis through SIRT3 activation, which might propose a novel strategy for myocardial fibrosis prevention and treatment.

Data Availability
The data used to support the finding of this study are available from the corresponding author upon request.

SIRT3 SIRT3
Myocardial fibrosis H 2 S H 2 S Ang II / TAC Figure 9: Illustration of the mechanism of protective effects on myocardial fibrosis by H 2 S. H 2 S enhanced SIRT3 transcription, decreased the DRP1 level, ameliorated mitochondrial membrane rupture, suppressed oxidative stress, and alleviated Ang IIinduced cardiac fibroblast proliferation and TAC-induced myocardial fibrosis. However, these protective effects of H 2 S were unavailable if SIRT3 was silenced in cells or deficient in mice. It suggested that H 2 S attenuated myocardial fibrosis through oxidative stress inhibition via a SIRT3-dependent manner.
14 Oxidative Medicine and Cellular Longevity