Induction of MicroRNA-21 with Exogenous Hydrogen Sulfide Attenuates Myocardial Ischemic and Inflammatory Injury in Mice

Abstract

Background

Maintaining physiological levels of hydrogen sulfide (H₂S) during ischemia is necessary to limit injury to the heart. Due to the anti-inflammatory effects of H₂S, we proposed that the H₂S donor, Na₂S, would attenuate myocardial injury through upregulation of ‘protective’ microRNA (miR)-21 and suppression of the inflammasome, a macromolecular structure that amplifies inflammation and mediates further injury.

Methods and Results

Na₂S-induced miR-21 expression was measured by qPCR in adult primary rat cardiomyocytes and in the mouse heart. We measured inflammasome formation and activity in cardiomyocytes challenged with lipopolysaccharide (LPS) and adenosine-tri-phosphate (ATP) or simulated ischemia/reoxygenation; and in the heart following regional myocardial ischemia/reperfusion (I/R), in the presence or absence of Na₂S. To assess the direct anti-inflammatory effects of H₂S in vivo, we utilized a peritonitis model by way of intraperitoneal injection of zymosan A. Na₂S attenuated inflammasome formation and activity - measured by counting cytoplasmic aggregates of the scaffold protein Apoptosis Speck-like protein containing a Caspase-recruitment domain (ASC; −57%) and caspase-1 activity (−50%) in isolated cardiomyocytes and in the mouse heart (all P<0.05). Na₂S also inhibited apoptosis (−38%) and necrosis (−43%) in cardiomyocytes in vitro and reduced myocardial infarct size (−63%) following I/R injury in vivo (all P<0.05). These protective effects were absent in cells treated with antagomiR-21 and in miR-21 KO mice. Na₂S also limited the severity of inflammasome-dependent inflammation in the model of peritonitis (P<0.05) in wild-type but not in miR-21 KO mice.

Conclusions

Na₂S induces cardioprotective effects through miR-21-dependent attenuation of ischemic and inflammatory injury in cardiomyocytes.

Introduction

Ischemic injury to the heart is followed by an intense inflammatory response that begets further injury and loss of viable myocardium¹. Ischemia itself and the debris released during cell death impact infiltrating and resident cells by promoting the formation of the inflammasome, a macromolecular structure formed by apoptosis speck-like protein containing a caspase-recruitment domain (ASC), cryopyrin (or NLRP3) and caspase-1, which is responsible for the amplification of the inflammatory response². Caspase-1 is the effector enzyme of the inflammasome which is primarily responsible for the processing and release of IL-1β (as well as IL-18) and induction of inflammatory cell death². Formation of the inflammasome and increased caspase-1 activity during acute myocardial infarction (AMI) promote cell death, adverse cardiac remodeling and heart failure in mice^3,4,5,6,7.

Hydrogen sulfide (H₂S) is an endogenous gasotransmitter known to influence a multitude of physiological and pathological processes, including protection against ischemia^8,9,10,11, pressure overload¹², doxorubicin toxicity¹³, inflammation^14,15, arterial contraction¹⁶, blood vessel relaxation^17,18 and insulin release¹⁹. In addition to the benefits of exogenous administration of H₂S by way of donors, endogenous H₂S also seems to mediate the cardioprotective effects of cGMP regulating drugs including phosphodiesterase-5 inhibitors and the NO-independent guanylate cyclase activator, cinaciguat^20,21. While the mechanisms of cardioprotection by H₂S remain under investigation, a number of published studies suggest that opening of K_ATP channels (sarcolemmal and mitochondrial), activation of protein kinase C and Akt are considered as the potential targets^10,22,23. Nevertheless, the role of microRNA (miR) and regulation of inflammasome formation in mediating the cardioprotective effect of H₂S are currently unknown. To this context, we considered the possible role of miR-21 in cardioprotection. Prosurvival Akt signaling has long been established in mediating the cardioprotective effects of ischemic preconditioning (IPC) whereby IPC increased Akt phosphorylation in the heart and PI3K inhibitors abolished cardioprotection with IPC^24,25. Similarly, recent studies demonstrated that miR-21 expression signature is differentially expressed in the ischemic heart and it increases remarkably with IPC²⁶. Moreover, miR-21 expression in the infarcted area was significantly down-regulated, whereas IPC inhibited this down-regulation²⁷. We and others have recently demonstrated the cardioprotective role of miR-21 against I/R injury²⁸ whereby miR-21 levels were shown to decline in the setting of I/R, and that attempts to restore miR-21 in such stresses have proven to be beneficial in attenuating injury²⁹. Interestingly, Akt has been shown to positively regulate miR-21 in the heart³⁰. Additionally, H₂S^31,32 and miR-21^33,34,35 were indicated to have an anti-inflammatory effect in animal models. Based on this compelling rationale, we hypothesized that H₂S may provide a protective effect in the heart during myocardial I/R by inhibiting the formation and activation of the inflammasome in a miR-21-dependent mechanism. We therefore sought to examine whether miR-21 mediates the cardioprotective effect of H₂S against I/R injury.

Our results show that H₂S reduces myocardial I/R injury as demonstrated by reduction of infarct size and preservation of left ventricular function. Moreover, H₂S increased miR-21 in the heart and cardiomyocytes, attenuated inflammasome formation and caspase-1 activity through a miR-21-dependent mechanism.

Materials and Methods

Animals

Adult male C57BL mice were supplied by The Jackson Laboratories (Bar Harbor, ME); the mean body weight was 32.4±0.9 g. MiR-21 knockout breeding pairs were purchased from Dr. Eric Olson at The University of Texas Southwestern Medical Center, Dallas, Texas. Adult male Wistar rats (300g) were purchased from Harlan Sprague-Dawley, Inc. (Indianapolis, IN). All animal experiments were conducted under the guidelines on humane use and care of laboratory animals for biomedical research published by the National Institutes of Health (No. 85-23, revised 1996).

Drugs and Chemicals

Sodium sulfide (Na₂S), triphenyltetrazolium chloride (TTC), Lipopolysaccharides from Escherichia coli 0111:B4 (LPS), ATP, Zymosan A from Saccharomyces cerevisiae and 4′,6-Diamidino-2-phenylindole dihydrochloride (DAPI) were purchased from Sigma-Aldrich (St. Loius, MO). Phthalo blue dye was purchased from Quantum Ink Company (Louisville, KY). AntagomiR-21 adenoviral vector³⁰ was obtained from Dr. Maha Abdellatif at the University of Medicine & Dentistry of New Jersey. Detailed information is included in Supplement.

To detect ASC localization, we used the combination of a rabbit anti-mouse-ASC (Sigma Aldrich) with a donkey anti-rabbit conjugated to an Alexa-fluor 594 (Life Technologies; Grand Island, NY) and a goat anti-mouse-cardiac actin (Sigma Aldrich) with a donkey anti-goat conjugated with an Alexa-fluor 488 (Life Technologies). For the in-vitro primary adult rat cardiomyocyte culture, a rabbit anti-rat-ASC (Santa Cruz) was used in combination with a goat anti-rabbit Alexa-fluor 594.

Adult Primary Cardiomyocyte Preparation

Adult rat ventricular cardiomyocytes were isolated using an enzymatic technique as previously reported³⁶. The simulated ischemia/reoxygenation protocol is described in detail in Supplement.

Real-Time PCR

Cardiomyocyte and myocardial miR-21 levels were assessed by real-time PCR as described in Supplement.

Experimental Groups (in vitro)

Six groups were used. 1- Control: Cardiomyocytes were isolated and subjected to SI/RO; 2- Na₂S: Cardiomyocytes were treated with 10 μM Na₂S 1 hour prior to SI/RO; 3- AntagomiR-21+Na₂S: Cardiomyocytes were infected with AntagomiR-21 (1.5×10⁹ pfu) for 24 hours prior to treatment with 10 μM Na₂S followed by SI/RO 1 hour later; 4- AntagomiR-21 Control: Cardiomyocytes were infected with antagomiR-21 for 24 hours prior to SI/RO; 5- Empty viral vector+Na₂S: Cardiomyocytes were infected with empty vector (1.5 × 10⁹ pfu) for 24 hours prior to treatment with 10 μM Na₂S followed by SI/RO 1 hour later; 6- Empty viral vector Control: Cardiomyocytes were infected with empty vector for 24 hours prior to SI/RO.

Assessment of Cell Death

Trypan blue exclusion assay³⁶ was used to assess loss of cell membrane integrity as seen in oncotic cell death (necrosis) or inflammatory cell death (pyroptosis) after 2 hours of RO. TUNEL³⁶ was used to detect nuclear DNA fragmentation as seen in apoptosis or pyroptosis after 18 hours of RO.

Study of the Inflammasome in Cardiomyocytes

Cardiomyocytes were challenged with lipopolysaccharide (LPS; 100 ng/ml, for 2 h) and ATP (5 mM, for 1 h), in absence or presence of Na₂S treatment (10 μM 5 min. before LPS challenge).

Cardiomyocytes were infected with antagomiR-21 (1×10³ pfu/cell) or the empty vector 1 hour following isolation. After 24 hours, the cells were incubated with Na₂S (10 μM) for 5 min. Subsequently, the cells were challenged with lipopolysaccharide (LPS; 100 ng/ml, for 2 h) and ATP (5 mM, for 1 h) as follows: 1- Empty viral vector control; 2- Empty viral vector+Na₂S; 3- Empty viral vector+LPS+ATP; 4- Empty viral vector+Na₂S and LPS+ATP; 5- AntagomiR-21 control; 6- AntagomiR-21+Na₂S; 7- AntagomiR-21+LPS+ATP; 8-AntagomiR-21+Na₂S and LPS+ATP. Where applicable, Na₂S was administered 5 minutes before LPS.

Two additional subsets of cells were subjected to SI/RO to study the inflammasome in this model: 1- Control: Cardiomyocytes were isolated and subjected to SI/RO; 2- Na₂S: Cardiomyocytes were treated with 10 μM Na₂S 1 hour prior to SI/RO.

All experiments were repeated three times and each single group was run in triplicates. Immunofluorescence was used to detect the presence of ASC aggregates, reflecting formation of the inflammasome, with a rabbit anti-rat-ASC antibody (1:200, overnight) used in combination with a goat anti-rabbit Alexa-fluor 594. In a similar experiment, caspase-1 activity, the effector enzyme in the active inflammasome, was measured using a FLICA fluorescent substrate (Axxora LLC, Farmingdale, NY) following the supplier’s instructions. In a parallel experiment, we used trypan blue staining to measure cell death as previously described³.

Myocardial Infarction Protocol and Infarct Size Measurement

The methodology of myocardial infarction was described previously²⁰. A brief summary of myocardial infarction and infarct size measurement is in Supplement.

Experimental Groups (in vivo)

Five groups were used. 1- Saline (Control): Each C57BL wild type mouse received 0.2 ml (i.p.) normal saline 1 h prior to I/R; 2- Na₂S: C57BL mice received 100 μg/kg Na₂S (i.p.) 1 h prior to I/R; 3- miR-21 knockout control: miR-21 knockout mice received 0.2 ml (i.p.) normal saline 1 h prior to I/R; 4- Na₂S+miR-21 knockout: miR-21 knockout mice received 100 μg/kg Na₂S (i.p.) 1 h prior to I/R. In all groups, infarct size was measured 24 hours after I/R. Prior to sacrifice, left ventricular (LV) function was analyzed using echocardiography. Six mice in each group were used for infarct size assessment and for functional analysis using echocardiography. The detailed experimental protocol is shown in Figure 1.

Figure 1.

Experimental protocol for in vivo experiments. Arrows indicate time points for treatment, performance of surgical procedures and measurement of various parameters.

Survival

Survival rate was determined based on the animals that survived the experimental protocol starting at recovery following surgery until 24 h after infarction.

Echocardiography

Echocardiography was performed using the Vevo770^™ imaging system (VisualSonics Inc., Toronto, Canada) prior to surgery (baseline) and 24 hours after surgery prior to sacrificing the animal. Pentobarbital (30 mg/kg; ip) was used for anesthesia and the procedure was carried out as previously described³⁷ to measure LV end-diastolic diameter (LVEDD) and end-systolic diameter (LVESD). LV fractional shortening (FS) was calculated as (LVEDD−LVESD)/LVEDD*100.

Western Blotting

Expression of phospho- and total Akt was assessed using Western blot analysis as described in Supplement.

Measurement of the Inflammasome in the Heart

Following I/R protocol, tissue slides (3 μm thick slices) were prepared from formalin-fixed and paraffin-embedded hearts. The sections were deparaffinized and rehydrated using sequential washes in xylene and decreasing concentrations of ethanol. After antigen retrieval with 0.01 M citrate buffer (pH 6.0) for 20 minutes, the slides were blocked with 1% normal swine serum in TBS for 15 minutes. For characterization of cardiomyocyte-specific expression of the aggregates of the scaffold protein ASC, which is indicative of inflammasome formation; double immunofluorescence technique was used. After blocking, the slides were incubated with primary antibody for ASC (1:50, Sigma-Aldrich) overnight at 4°C. Anti-rabbit Alexa Fluor 594-conjugated secondary antibody (1:100) was applied for 4 hours at room temperature, then slides were incubated with a primary antibody for cardiac Actin (1:200, Sigma-Aldrich) overnight at 4°Cand Alexa Fluor 488-conjugated secondary antibody (1:100, Invitrogen) was applied for 4 hours at room temperature³. The cell counterstaining was performed with 4′,6-diamidino-2-phenylindole (DAPI) 1:1,000 for 5 minutes and the slides were coverslipped with SlowFade® Antifade (both Invitrogen). Negative controls with nonspecific IgG were run in parallel. Images were acquired with an IX70 microscope and MagnaFire 1.1 software (both Olympus) using a 40x objective (400x magnification). Color composite images were generated with ImageJ software.

Measurements of ASC in the infarct area was performed by two investigators who were blinded to treatment allocation and the expression was quantified using a semiquantitative scale ranging from 0 (no expression) to 1+ (minimal expression meaning either few aggregates [<1 per high power field] or mild diffuse stain without aggregates), 2+ (moderate expression meaning either 1–5 aggregates per high power field or diffuse stain with few aggregates), 3+ (diffuse intense staining with many cytoplasmic aggregates [>5 per high power field]).

Caspase-1 Activity in the Heart

An additional subset of mice was sacrificed 24 hours after surgery (n=4–6 per treatment group). The heart was removed, rinsed in PBS and snap frozen in liquid nitrogen. Caspase-1 activity was measured in the tissue using a fluorogenic substrate (Ac-YVAD-AMC) specific for caspase-1 (Enzo Life Sciences, Farmingdale, NY, USA)³. After homogenization using RIPA buffer containing a cocktail of protease inhibitors and centrifugation at 16,000 rpm for 20 minutes, 50 μg of protein from each sample were used for the assay as previously described³. Fluorescence was measured after 60 minutes and was expressed as arbitrary fluorescence units produced by one microgram of sample per minute (fluorescence/μg/min) and calculated as fold change compared to the caspase-1 activity in homogenates of the hearts of sham-operated mice, whereby oral intubation and a left thoracotomy were performed and a 7.0 silk suture was placed around the left coronary artery but not tightened.

Inflammasome-dependent Peritonitis Model

To determine the effects of H₂S on the inflammasome in vivo independent of its effects on ischemia or infarction, we used a model of peritonitis induced by the intra-peritoneal injection of zymosan A (1 mg/mouse), which induces the activation of the cryopyrin inflammasome³⁸. After 6 hours, peritoneal lavage was performed with sterile NaCl 0.9% (7 ml), and the fluid was assessed for leukocyte content using the Thoma chamber. The total amount of leukocytes was determined in the presence or absence of Na₂S (100 μg/kg). Glyburide (500 mg/kg), a specific cryopyrin inhibitor³⁹ or interleukin-1 receptor antagonist (IL-1Ra 20 mg/kg), which inhibits the IL-1 dependent peritoneal migration of leukocytes, were used as control agents of inflammasome-related activity. Equal volume of normal saline was used in the control group.

Statistics

All measurements are expressed as group means±SE. The data were analyzed by nonparametric Kruskal Wallis rank sum test which is a generalization of Wilcoxon rank sum test for more than two groups. We used Kruskal Wallis rank sum test in cases of two or more than two groups to remain consistent across all the experiments. All the data analyses were performed in statistical software R version 3.0.3 using functions in the base package.

The authors had full access to the data and take responsibility for its integrity. All authors have read and agree to the manuscript as written.

Results

H₂S donor, Na₂S, induces miR-21 in primary cardiomyocytes and the heart

To test the effect of H₂S on miR-21, the cardiomyocytes were incubated with Na₂S (10 μM) or control medium for 1 hour before measuring miR-21 expression. Figure 2A shows a significant increase in miR-21 in Na₂S-treated cardiomyocytes as compared to control. Similarly, miR-21 increased in the heart following treatment with Na₂S (100 μg/kg; i.p.) as shown in Figure 2B. This increase in miR-21 at 1 hour was not paralleled by an increase in Akt phosphorylation (Figure 2C), ruling out its potential role in the acute induction of miR-21.

Figure 2.

Real-Time PCR analysis showing significant increase in miR-21 expression in adult primary rat cardiomyocytes (A) and in the mouse heart (B) 1 h following treatment with Na₂S as compared with control groups. (C) Western blots and densitometry showing no change in phospho-Akt to total Akt ratio at 1 hour following Na₂S treatment as compared to saline control. Horizontal lines represent Mean±SE.

Na₂S attenuates cardiomyocyte injury following ischemia/reoxygenation

Primary rat cardiomyocytes treated with Na₂S or control medium were exposed to 90 minutes of simulated ischemia followed by 2 (for necrosis) or 18 (for apoptosis) hours of reoxygenation. Trypan blue permeability and TUNEL positivity were measured as markers of necrotic cell death and DNA fragmentation as reported in Figures 3A & 3B. The treatment with Na₂S significantly reduced cell death and DNA fragmentation indicating its protective role against SI/RO injury. Interestingly, antagomiR-21 blunted the decrease in necrosis and apoptosis observed with Na₂S treatment.

Figure 3.

(A) Necrosis assessed by trypan blue staining following SI/RO in primary adult rat cardiomyocytes demonstrating a decrease with Na₂S treatment compared to control. AntagomiR-21 not only blocked the protective effect of Na₂S, but also exacerbated necrosis. (B) Representative TUNEL staining reflecting apoptosis in primary adult rat cardiomyocytes showing a decrease with Na₂S treatment compared to control. AntagomiR-21 abolished the anti-apoptotic effect of Na₂S.

Na₂S improves survival following I/R injury

A total of 134 mice were used in this study. Fifteen out of 18 C57BL mice survived with Na₂S (83%) as compared to 14 out of 25 with saline (56%). The survival rate was 100 % in sham-operated mice. Fourteen out of 29 miR-21 knockout mice survived in the Na₂S group (48%) and 14 out of 28 survived in the saline group (50%). This suggests that the survival benefits observed with Na₂S are dependent upon miR-21.

Na₂S protects against I/R injury through miR-21

To demonstrate the cardioprotective role of H₂S in vivo and to determine the role of miR-21 in protection, wild type and miR-21 KO mice were treated with Na₂S or saline prior to I/R injury (Figures 4A & 4B). Na₂S significantly reduced infarct size to 16.3±1.5% as compared to 44.4±1.6% in wild type mice, confirming the protective action of Na₂S observed in cardiomyocytes. Interestingly, Na₂S treatment in miR-21 KO mice failed to reduce infarct size, suggesting a central role of miR-21 in mediating cardioprotection with H₂S.

Figure 4.

(A) Representative heart sections stained with phthalo blue to demarcate the non-risk area and TTC to identify viable tissue. (B) Myocardial infarct size (% of RA) measured 24 h post-MI in the various groups. Note that Na₂S treatment exhibited a smaller infarct size following I/R compared to saline control. The infarct-sparing effect of Na₂S was abolished in miR-21 KO mice. The area-at-risk, expressed as percent of LV, was similar in all groups. (C) Cardiac function measured as LV fractional shortening was preserved following I/R injury with Na₂S as compared to saline control. This benefit was lost in miR-21 KO mice. (D) Western blots and densitometry showing a significant increase in myocardial phospho-Akt to total Akt ratio at 24 hour following I/R in Na₂S-treated mice as compared to saline-treated controls. Horizontal lines represent Mean±SE.

The beneficial effect of H₂S treatment was reflected also at the functional level, as demonstrated by preservation of left ventricular fractional shortening in wild type mice (Figure 4C). The absence of miR-21 per se did not alter baseline fractional shortening. However, Na₂S-treated miR-21 KO mice did not show improvement in ventricular function following I/R injury, confirming an indispensable role of miR-21 in mediating cardioprotection with H₂S.

Interestingly, Na₂S-treated mice exhibited a significant increase in myocardial Akt phosphorylation 24 hours following I/R injury as compared to saline-treated mice (Figure 4D).

Na₂S inhibits formation of the inflammasome in cardiomyocytes in vitro

Na₂S attenuated the formation of the inflammasome (measured as ASC aggregates by 57%, P<0.05; Figures 5A & 5B) in cardiomyocytes challenged with LPS+ATP as compared to control cells and also in cardiomyocytes subjected to SI/RO (55% reduction, P<0.05; Figure 5C). The cytoprotective effect of Na₂S was abolished in cardiomyocytes infected with antagomiR-21. Moreover, Na₂S attenuated caspase-1 activity (50% reduction, P<0.05; Figures 6A & 6B) and cell death (51% reduction, P<0.05, Figure 6C) in cardiomyocytes challenged with LPS+ATP as compared to control cells.

Figure 5.

(A) Representative fields showing ASC aggregates (red) in adult primary cardiomyocytes. Adenoviral antagomiR-21 was utilized to test the role of miR-21 in mediating the anti-inflammasome effects of Na₂S whereas an empty virus was used as control. (B) LPS+ATP caused a significant increase in ASC aggregates which was attenuated by Na₂S. This anti-inflammatory effect of Na₂S was abolished by antagomiR-21. (C) Subjecting adult primary cardiomyocytes to SI/RO also caused a significant increase in ASC aggregates, which was attenuated by Na₂S treatment. Horizontal lines represent Mean±SE.

Figure 6.

(A) Representative fields illustrating caspase-1 activity (red) in adult primary cardiomyocytes. (B) LPS+ATP significantly increased caspase-1 activity which was attenuated by Na₂S. (C) Na₂S also significantly attenuated necrotic cell death caused by LPS+ATP. Horizontal lines represent Mean±SE.

Na₂S inhibits inflammasome formation in the heart following I/R injury

Treatment with Na₂S blunted inflammasome formation in the heart [ASC aggregates in the zone bordering the infarct (by 37%, P<0.05, Figures 7A & 7B) and caspase-1 activity (Figure 8A) following myocardial I/R injury. In contrast, miR-21 knockout mice exhibited a trend of elevated basal caspase-1 activity, which increased with I/R injury and was not attenuated with Na₂S treatment.

Figure 7.

(A) Representative slides showing ASC aggregates (red) in cardiomyocytes (in situ) in heart sections from infarct and remote zones stained with α-actinin (green) 24 h after I/R injury. (B) Na₂S significantly attenuated ASC aggregates following I/R injury as compared to saline control. Horizontal lines represent Mean±SE.

Figure 8.

(A) Caspase-1 activity was markedly increased at 24 h following I/R injury in the hearts of wild type mice and miR-21 KO mice. This increase in caspase-1 activity; however, was attenuated with Na₂S treatment only in wild type mice but not in miR-21 KO mice. (B) In the inflammasome-dependent peritonitis model, Na₂S attenuated leukocyte infiltration in wild type mice as compared to saline controls, but not in miR-21 KO mice where there was an increase in leukocyte infiltration with Na₂S treatment as compared to saline-treated controls. Horizontal lines represent Mean±SE.

Na₂S inhibits inflammasome-mediated peritonitis in the mouse

To further demonstrate the cause-and-effect relationship of miR-21 in H₂S-induced attenuation of inflammasome function in vivo, we used the zymosan A induced peritonitis model, an inflammatory model in which leukocyte migration into the peritoneal cavity is mediated by the inflammasome³⁸. Pretreatment with Na₂S significantly reduced the leukocyte infiltrate in the peritoneal lavage following challenge with zymosan A (45% reduction, Figure 8B). This effect was similarly reproduced by the inflammasome inhibitor glyburide and by the IL-1 blocker IL-1Ra (anakinra, not shown). However, Na₂S treatment did not prevent the leukocyte increase observed with zymosan A in miR-21 KO mice, supporting a pivotal role of miR-21 in mediating the anti-inflammatory effects of H₂S. Interestingly, glyburide and anakinra were still capable of reducing leukocyte infiltration following zymosan challenge in the miR-21 KO mouse (not shown).

Discussion

H₂S is produced endogenously and maintained at physiologic concentrations in mammalian systems²¹. The concentration of H₂S has been shown to influence a wide range of physiological processes^8,16,17,40 and exogenous administration of H₂S by way of donors (i.e. Na₂S) has shown to modulate the course of several acute and chronic illnesses^9,14,15,41. In the current study, we show that administration of a H₂S donor during I/R in mice prevents the formation and the activation of the cryopyrin inflammasome, a macromolecular complex responsible for sensing tissue injury or ‘danger’, amplifying the inflammatory response, and inducing cell death^2,3,4. Na₂S prevented the formation of the inflammasome aggregates in cardiomyocytes in vitro and in the heart in vivo following both ischemic and non-ischemic injuries (i.e. LPS/ATP in vitro, peritonitis in vivo) thus demonstrating a specific anti-inflammasome effect of Na₂S that cannot be solely attributed to the reduction in ischemic injury.

The effects of Na₂S on inflammation have been widely reported. In a recent study, H₂S was used to treat arthritis in an inflammasome-mediated mouse model³². However, the mechanisms by which H₂S exerts its anti-inflammatory effects remain poorly understood. Due to its lipophilic structure as a gasotransmitter⁴², H₂S is likely to have many and variable targets, as it lacks a true receptor. Prior studies have shown that H₂S activates pro-survival Akt¹⁰, which is a key regulator of miR-21 expression³⁰. It is noteworthy that miR-21 has been implicated in the endogenous mechanisms of cardioprotection following ischemic or pharmacologic preconditioning²⁹ as well as in anti-inflammatory pathways^33,34,35. For this reason, we determined whether H₂S also positively regulates miR-21. Our results show that Na₂S significantly induced miR-21 expression in primary adult rat cardiomyocytes (2.3 fold increase) and in the intact mouse heart (2.7 fold increase). The same in vitro and in vivo doses of Na₂S caused a significant cytoprotective effect following ischemia. However, contrary to our initial hypothesis, Na₂S treatment did not alter the phosphorylation status of Akt at 1 hour after treatment, which rules out its role in acute miR-21 induction. Nonetheless, we did detect an increase in Akt phosphorylation at 24 hours following I/R in the Na₂S-treated group versus control. Other studies have reported that miR-21 itself can also increase Akt phosphorylation by inhibiting phosphatase and tensin homologue deleted on chromosome 10 (PTEN)³⁰, one of the known targets of miR-21. While it remains unclear how H₂S increases miR-21 expression acutely, we believe that increased Akt phosphorylation with Na₂S at 24 hours after I/R may be responsible for maintaining miR-21 levels in the H₂S group, which may initiate a self-propagating cycle of Akt activation and miR-21 induction. More in-depth studies involving multiple time points during and following ischemia are warranted to further examine the mechanism of miR-21 induction with Na₂S. MiR-21 has several known targets that support its protective role against ischemia, including PTEN, programmed cell death 4 (PDCD4), Toll-Interleukin Receptor, transforming growth factor-β receptor II and myeloid differentiation factor 88 (MyD88). However, investigations into the direct or indirect target(s) of miR-21 that are responsible for the infarct-sparing effect of Na₂S (which occurs within a few hours after MI) are also needed.

Considering the link between H₂S and miR-21, and the anti-inflammatory and cardioprotective effects of both H₂S^31,32,43 and miR-21^33,34,35, our data suggest that H₂S inhibits inflammasome in the heart following ischemic injury through induction of miR-21, especially since the protective effects of Na₂S were not observed in miR-21 KO mice. A recent study reported that miR-21 can suppress CSE expression⁴⁴ in de-differentiated human aorta SMCs and injured mouse carotid arteries. Although this may seem in apparent contrast with our signaling cascade, our study focuses on primary adult cardiomyocytes and the heart as a whole. However, even if these observations were to hold true in our model, we are not relying on endogenous production of H₂S but on H₂S donor therapy. In fact, this observation may represent a negative feedback loop whereby increasing H₂S levels by way of donors suppresses the need for its endogenous production.

Our results also show that H₂S interferes with the inflammasome activity not only following ischemic injury but also following canonical inducers of the inflammasome in vitro (LPS+ATP) and in vivo (Zymosan A). These data provide evidence that the anti-inflammatory effects cannot be solely attributed to the infarct-sparing effects of H₂S.

The inflammasome and the sterile inflammatory response following regional myocardial I/R injury have been identified as a target for intervention in order to prevent adverse cardiac remodeling secondary to AMI^1,3,45,46. While cellular injury triggers the formation of the inflammasome leading to the secretion of pro-inflammatory cytokines and promotion of cell death^47,48, inhibition of the components of the inflammasome [cryopyrin³ ASC⁶, and caspase-1^4,7] protects the heart from further injury. Therefore, we believe that mitigation of acute cardiac dysfunction observed with Na₂S is likely due to a combination of multiple mechanisms including its infarct-sparing effects and inhibition of the inflammasome, which were not observed in miR-21 knockout mice treated with Na₂S. The results of the current study confirm the essential role of miR-21 not only in mediating the infarct-sparing effects of H₂S, but also in attenuating myocardial inflammasome formation and its central role in post-infarction adverse remodeling, which might explain recent findings demonstrating the chronic benefits of Na₂S in preventing the progression to heart failure secondary to AMI^49,50. The importance of inflammasome attenuation is especially beneficial during the post infarction stages since these innate defense mechanisms stimulate the expression of multiple inflammatory mediators which orchestrate the recruitment of inflammatory cells and perpetuation of the inflammatory response⁴⁶.

The exact mechanisms by which H₂S inhibits the formation of the inflammasome and subsequent caspase-1 activation in the heart during AMI are unknown. Our results show that both formation of the inflammasome as well as activation of caspase-1 are impaired by H₂S, suggesting that it may be acting upstream of the aggregation of the multimers. This could be at the level of priming of the inflammasome (by inhibiting expression of key components) or at the level of triggering (by inducing structural changes in the sensor of the inflammasome). Of utmost importance is the fact that the inflammasome-inhibiting effects of H₂S were completely abolished with deletion of miR-21 by using two different and alternative approaches (genetically engineered mice and antagomir treatment). This suggests that miR-21 is essential for the protective effects of H₂S. MiRs are tight controllers of gene expression and may well explain the many and variable effects seen with H₂S. In particular, miR-21 has been shown to suppress toll-like receptor signaling and lung inflammation in mice³³ and in human peripheral blood mononuclear cells³⁴. Interestingly, a recent study demonstrated that hepatitis C virus-induced miR-21 contributes to evasion of host immune system by targeting MyD88 and interleukin-1 receptor-associated kinase 1 in human hepatoma and embryonic kidney cells³⁵. These reports are in agreement with our findings as they support the anti-inflammatory role of miR-21, especially by targeting upstream pathways that lead to inflammasome formation.

The use of adenoviral antagomiR-21 and genetically engineered miR-21 knockout mice are however also fraught with limitations. For instance, genetically engineered animals may have other alterations or compensatory mechanisms that are phenotypically masked and the use of antagomir may trigger off-target effects. In fact, Thum et al. demonstrated that miR-21 mediates cardiac fibrosis in a pressure-overload mouse model by augmenting extracellular-regulated kinase – mitogen-activated protein kinase (ERK-MAPK) activity through inhibition of sprouty homologue 1 in cardiac fibroblasts. Moreover, in vivo systemic delivery of antagomiR-21 reduced ERK-MAPK activity, inhibited interstitial fibrosis and attenuated cardiac dysfunction⁵¹. However, a more recent study in miR-21 KO mice revealed that stress-dependent adverse cardiac remodeling develops in the absence of miR-21, suggesting the lack of its role in pathological cardiac remodeling⁵². Although these findings are in apparent contrast, the question of antagomir specificity and compensatory modifications following genetic deletion remains unanswered. Moreover, the pressure overload model with a focus on cardiac fibroblasts is quite different from our model of myocardial ischemia/reperfusion. Nevertheless, the use of complementary approaches of knockout mice and antagomir provide conclusive evidence for the role of miR-21 in contributing to the anti-inflammatory effect of H₂S during ischemia. Moreover, other studies have illustrated that overexpression of miR-21 could also reduce fibrosis during ischemic heart disease or failure by decreasing myocyte death and, thus, inflammatory cell infiltration, and fibroblast proliferation³⁰.

In conclusion, H₂S exerts its cardioprotective effects and inhibits the formation and activation of the inflammasome through signaling pathways requiring miR-21. H₂S therapy or modulation of endogenous production of H₂S may be an attractive approach in reducing injury and inflammasome assembly/activation in the heart and other organs. Moreover, the important role that miR-21 plays in mediating cardioprotection with H₂S will enable us to design therapies using small molecules that would likely exhibit a more linear pharmacokinetic profile, as it has been shown for other miR targets⁵³.