Hydrogen sulfide attenuates hepatic ischemia-reperfusion injury: role of antioxidant and antiapoptotic signaling

Abstract

Hydrogen sulfide (H₂S) is an endogenously produced gaseous signaling molecule with diverse physiological activity. The potential protective effects of H₂S have not been evaluated in the liver. The purpose of the current study was to investigate if H₂S could afford hepatoprotection in a murine model of hepatic ischemia-reperfusion (I/R) injury. Hepatic injury was achieved by subjecting mice to 60 min of ischemia followed by 5 h of reperfusion. H₂S donor (IK1001) or vehicle were administered 5 min before reperfusion. H₂S attenuated the elevation in serum alanine aminotransferase (ALT) by 68.6% and aspartate aminotransferase (AST) by 70.8% compared with vehicle group. H₂S-mediated cytoprotection was associated with an improved balance between reduced glutathione (GSH) vs. oxidized glutathione (GSSG), an attenuated formation of lipid hydroperoxides, and an increased expression of thioredoxin-1 (Trx-1). Furthermore, H₂S inhibited the progression of apoptosis after I/R injury by increasing the protein expression of heat shock protein (HSP-90) and Bcl-2. These results indicate that H₂S protects the murine liver against I/R injury through an upregulation of intracellular antioxidant and antiapoptotic signaling pathways.

hepatic ischemia-reperfusion (I/R) injury occurs in diverse clinical settings, including liver transplantation, lobectomy (pringle maneuver), veno-occlusive disease including Budd-Chiari syndrome, congestive heart failure, shock states, and resuscitation (10, 14, 18, 20, 38). Liver injury induced by I/R represents a continuum of processes (26) that may produce profound hepatocellular injury and ultimately result in morbidity and mortality (25). The number of liver transplantations performed each year has been on the rise for last 20 years, but there exists a large imbalance between the number of available donors and patients waiting for transplantation. The noted increase in demand and shortage of supply has forced the consideration of cadaveric or steatotic grafts (19), which have a higher susceptibility to I/R injury and a much higher risk of primary nonfunction and mortality (34). Therefore, minimizing the adverse effects of hepatic I/R injury could increase the number of patients that may undergo a successful transplantation (11). However, at present there is no therapeutic treatment plan or strategy available to prevent hepatic I/R injury (26).

Experimental evidence suggests that liver I/R injury is biphasic: early I/R injury occurs with the initiation of an inflammatory cascade involving numerous reactive oxygen species, reactive nitrogen species (14, 38), chemokines, and cytokines (10, 28, 38), followed by neutrophil-mediated hepatic injury occurring at 6–24 h of reperfusion (18, 20). Additionally, a consequence of I/R injury is the disruption of intracellular energy metabolism and enzyme function, resulting in a depletion of ATP and an accumulation of intracellular sodium and edema (11), suggesting that the mitochondria play a role in the pathology of I/R injury (31). The role of necrosis was considered vital for I/R injury in the past, but it is now known that apoptosis also plays a significant role in cellular damage after I/R injury (12, 20).

Hydrogen sulfide (H₂S) has long been considered as a toxic environmental pollutant emerging from sewers, marshes, and volcanic eruptions. Recently, H₂S has been recognized alongside nitric oxide and carbon monoxide as an endogenously produced gaseous signaling molecule (29, 36, 39). In mammalian cells, H₂S is produced by two heme containing enzymes, cystathionine β-synthase and cystathionine γ-lyase, the activity of which depends on pyridoxal 5′-phosphate (36, 39). Cystathionine β-synthase is primarily responsible for production of H₂S in the central nervous system (8), while cystathionine γ-lyase is primarily expressed in peripheral tissues, including vascular and nonvascular smooth muscle (30, 36, 37, 48). Since the recent discovery that H₂S is a powerful physiological signaling molecule, experimental studies (30) have begun to characterize its biological profile. H₂S promotes vascular smooth muscle relaxation and induces vasodilation of isolated blood vessels (3, 27, 48). H₂S has also been shown to inhibit leukocyte-endothelial cell interactions in vivo (45) indicating an anti-inflammatory action. It has also become evident that H₂S is a potent antioxidant (24, 40, 44) and under more chronic conditions upregulates antioxidant defenses (23, 24). It has also been demonstrated that H₂S effectively inhibits apoptosis of a number of cell types (6, 32, 35), and this effect has been shown to promote cytoprotection.

Previous studies (1, 6, 35, 47) have provided insights into the protective actions of H₂S in the setting of hypoxia-reoxygenation and I/R injury. However, to date the potential cytoprotective effects of H₂S have not been evaluated in hepatic I/R injury. The aim of present study was to investigate the effects of an exogenous administration of H₂S on the severity of I/R injury in an in vivo murine model of hepatic I/R injury.

MATERIALS AND METHODS

H₂S donor.

Sodium sulfide (Na₂S, IK1001) was produced by Ikaria (Seattle, WA) by using H₂S gas (Matheson, Newark, CA) as a starting material. Na₂S was formulated to pH neutrality and iso-osmolarity. Na₂S (stock solution at 0.55 mg/ml and 7.1 mM) was diluted in normal (0.9%) saline to the desired concentration in a rapid fashion, immediately before administration For acute hepatic I/R experiments, normal saline (100 μl) or H₂S donor (0.3, 1, and 2 mg/kg) in a final volume of 100 μl was injected intravenously into the inferior vena cava using a 32-gauge needle at 5 min before reperfusion.

Animals.

Mice (C57BL6/J) were used for the study. All mice were male, 8–10 wk of age, and were purchased from Jackson Laboratories. All experimental procedures were approved by the Institute for Animal Care and Use Committee at Albert Einstein College of Medicine and complied with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Hepatic I/R protocol.

The hepatic I/R protocol has been described previously (5). In the present study, mice were subjected to 60 min of hepatic ischemia and either 1, 5, or 24 h of reperfusion. Mice were anesthetized with xylazine (8 mg/kg) and ketamine (100 mg/kg), constituted in normal saline and administered intravenously. Core body temperature of the mice was monitored continuously during the entire period of surgery using a rectal probe, and a heat lamp was used to maintain the core body temperature at 37 ± 0.4°C. A midline laparotomy incision was performed to expose the liver. A microaneurysm clamp was applied to the hepatic artery and portal vein resulting in ischemia of the left lateral and median lobes of the liver. The procedure leads to segmental (70%) hepatic ischemia so as to prevent mesenteric venous congestion by allowing portal decompression through the caudate and right lobe of the liver. The liver was then repositioned back to its original position and kept moist throughout the period of ischemia (60 min) using gauze soaked with normal saline. Mice were injected with heparin (100 U/Kg) to prevent clotting before reperfusion. Sham surgeries were identical except liver ischemia was not performed using microaneurysm clamp. Study drug (Na₂S; IK1001) or vehicle was injected intravenously 5 min before reperfusion. After the ischemia period was completed, the clamp was removed and the liver was reperfused for different periods (1, 5, or 24 h) according to the protocol for the various experimental endpoints as described below. Liver tissue collected for Western blot analysis and glutathione assay was snap frozen in liquid nitrogen and stored at −80°C till further processing of the tissue.

Determination of liver transaminases.

At 5 h of reperfusion, mice were anesthetized again as described in Hepatic I/R protocol. Blood was collected from the inferior vena cava with a 20-gauge needle and placed in a microtainer serum separator tube. The blood sample was kept on ice for 15 min and then centrifuged at 14,000 g for 12 min. The serum samples were then analyzed for aspartate aminotransferase (AST) and alanine aminotransferase (ALT) using a commercially available kit (Thermo Electron, Melbourne, Australia) by spectrophotometric method. AST and ALT are released into blood circulation after hepatocellular injury and can be used to estimate extent of injury.

Lipid hydroperoxide assay.

Quantification of lipid peroxidation was done to assess the extent of hepatic tissue oxidative injury. Lipid peroxidation results in the formation of highly unstable and reactive hydroperoxides of both saturated and unsaturated lipids. Hepatic tissue was collected at 1 and 5 h reperfusion. Lipid hydroperoxides were measured using a commercially available kit (Cayman Chemicals) according to the manufacturer's recommendations. The assay is based on the principle that hydroperoxides are highly unstable and react with ferrous ions readily to produce ferric ions. The resulting ferric ions are detected using thiocynate as the chromogen. Hepatic lipid hydroperoxide (LPO) is reported in micromolars.

Glutathione assay.

Hepatic tissue was collected for glutathione assay at 1, 5, or 24 h of reperfusion. Tissue levels of reduced glutathione (GSH) and oxidized glutathione (GSSG) were measured using a commercially available kit (Cayman Chemicals). The GSH-to-GSSG ratio was calculated for each sample according to manufacturer's recommendations.

Western blot analysis.

Western blot analysis was performed to evaluate the protein expression of cleaved caspase-3, Bcl-2, thioredoxin-1 (Trx-1), and 90-kDa heat shock protein (HSP-90) as described previously (16). Briefly, livers (ischemic lobes only) of sham, vehicle plus I/R and H₂S plus I/R from nondiabetic mice were excised after 5 and 24 h after hepatic I/R. Samples were homogenized in 1 ml of ice-cold RIPA lysis buffer. Homogenates were then centrifuged at 1,300 g to remove any cellular debris. The pellet was discarded, and the supernatant was again centrifuged at 16,000 g for 30 min at 4°C. The resultant supernatant (cytosolic fraction) was collected. The resulting pellet was resuspended in 300 μl of RIPA lysis buffer. Protein assay was done with the use of Bio-RAD DC protein assay as per recommendations from manufacturer. An equal amount of protein (25 μg) was loaded into each well and separated on 7–12% polyacryilamide gel (7% for HSP-90, 12% for cleaved caspase 3, Bcl-2, and Trx-1). Protein was transferred to immunoblot polyvinylidene difluoride membrane and then blocked in 5% milk in Tris-buffered saline Tween-20 at room temperature. Membranes were then incubated with primary antibodies (mouse anti-Trx-1, 1:2,000; mouse anti-HSP-90, 1:5,000; mouse anti-cleaved caspase 3, 1:500, and mouse anti-Bcl-2, 1:1,500) overnight at 4°C. Membranes were then washed and incubated with horseradish peroxidase-linked anti-rabbit IgG (1:10,000) at room temperature. Membranes were then washed, followed by incubation with ECL reagents (Amersham), and then exposed to film. Membranes were then stripped and incubated with either mouse anti-α-tubulin (1:10,000) or anti-cytochrome c oxidase (COX-IV; 1:20,000) overnight at 4°C. Membranes were then washed and incubated with horseradish peroxidase-linked anti-rabbit or anti-goat secondary (1:10,000), followed by incubation with ECL (Amersham) reagents, and then exposed to film. Anti-α-tubulin antibodies were purchased from Santa Cruz Biotechnology, and all other antibodies were purchased from Cell Signaling Technology. Densitometric analysis was performed using Image J software from the National Institutes of Health.

Statistical analysis.

Data were analyzed by one-way ANOVA with post-Turkey multiple comparison test and two-way ANOVA with post hoc Bonferroni analysis wherever appropriate using Prism software (San Diego, CA). Data are reported as means ± SE. P values <0.05 were considered to be statistically significant.

RESULTS

H₂S attenuated hepatic I/R injury.

Intravenous administration of H₂S donor at doses of 0.3 and 1.0 mg/kg before reperfusion limited serum elevations of the liver transaminases AST and ALT in a dose-dependent manner after 60 min of ischemia and 5 h of reperfusion (Fig. 1, A and B) . Specifically, the 0.3 mg/kg dose reduced serum AST by 51.6% (P < 0.05) and serum ALT by 47% (P < 0.05). The 1.0 mg/kg dose significantly reduced serum AST by 70.8% (P < 0.01) and serum ALT by 68.6% (P < 0.001). In sharp contrast to lower doses, 2.0 mg/kg of the H₂S donor increased serum AST by 151% and ALT by 124%, compared with the vehicle. Additionally, doses higher than 2 mg/kg were associated with significantly higher mortality (data not shown). Thus 1 mg/kg was considered the most effective dose for hepatocellular protection and was investigated in further experiments. Both vehicle- and H₂S-donor-treated mice had significantly higher liver transaminases as a consequence of I/R injury, compared with sham group. Sham values for AST were 58.4 ± 12.8 U/l, and ALT values in sham-operated controls were 38.3 ± 9.4 U/l.

Fig. 1.

Serum transaminases were measured in mice after 60 min of hepatic ischemia and 5 h of reperfusion. A: aspartate aminotransferase (AST) levels (U/l); B: alanine aminotransferase (ALT levels; U/l). Hydrogen sulfide (H₂S) donor therapy (0.3 and 1.0 mg/kg) significantly reduced hepatic release of AST and ALT. *P < 0.05 vs. vehicle; **P < 0.01 vs. vehicle; ***P < 0.001 vs. vehicle; numbers inside bars represent number of animals investigated in each group.

H₂S increased hepatic tissue GSH-to-GSSG ratio.

The GSH-to-GSSG ratio (Fig. 2A) was calculated for sham and vehicle- and H₂S-treated mice (n = 6 in each group). There was a significant decrease (P < 0.001) in the GSH-to-GSSG ratio in both the vehicle and H₂S groups as consequence of I/R injury compared with sham. However, H₂S attenuated the I/R-induced reduction significantly (P < 0.05) when compared with the vehicle-treated mice at 5 and 24 h reperfusion. Additionally, similar trend was observed at 1 h of reperfusion. Specifically, the ratios in the H₂S-treated mice were 18.1, 35.8, and 63.3% higher than the vehicle-treated mice at 1, 5, and 24 h of reperfusion, respectively.

Fig. 2.

A: hepatic tissue reduced glutathione (GSH)-to-oxidized glutathione (GSSG) ratio at 1, 5, and 24 h after 60 min of ischemia. Data are presented for sham-operated controls, mice receiving vehicle (Veh), and mice treated with the H₂S donor (1.0 mg/kg). H₂S therapy significantly preserved the GSH-to-GSSG ratio compared with the vehicle group. B: representative immunoblot and densitometric analysis (bottom) of hepatic thioredoxin-1 (Trx-1) protein expression. Trx-1 protein levels remained significantly elevated in animals receiving H₂S at 24 h after reperfusion (Rep). **P < 0.01 vs. sham; ***P < 0.001 vs. sham; n = 6 in each group.

Trx-1 expression was preserved by H₂S after I/R injury.

Western blot analysis for protein expression of Trx-1 (Fig. 2B) revealed that Trx-1 levels decreased significantly (P < 0.05) in the vehicle group at 5 and at 24 h of reperfusion compared with the sham group. Conversely, Trx-1 levels in the H₂S-treated group (1.0 mg/kg) remained at levels similar to those of the sham group at both 5 and 24 h of reperfusion. The mean expression of Trx-1 was higher at both 5 and 24 h of reperfusion in H₂S group compared with vehicle-treated mice, but a statistical significance was only observed for at the 24 h time point (P < 0.01).

H₂S attenuated oxidative stress.

Lipid peroxidation was used as measure of hepatic oxidative stress after I/R injury. LPO levels (Fig. 3) were measured for sham and vehicle- and H₂S-treated mice (n = 8 in each group) after I/R injury. The vehicle group exhibited significantly higher levels of lipid hydroperoxides at 1 h postreperfusion and 5 h postreperfusion compared with sham-operated controls, as a consequence of I/R injury. Conversely, H₂S-treated mice displayed LPO levels similar to sham at both 1 and 5 h. H₂S-treated mice had significant reductions in lipid peroxidation by 36% at 1 h (P < 0.05) and 52% at 5 h (P < 0.001) reperfusion compared with vehicle.

Fig. 3.

Lipid hydroperoxide levels (μM) in hepatic tissue lyaste after 1 and 5 h of reperfusion after 60 min of ischemia. Treatment with H₂S (1.0 mg/kg) significantly attenuated hepatic formation of lipid hydroperoxides. **P < 0.01 vs. sham; ***P < 0.001 vs. sham; numbers inside bars represent number of animals investigated in each group.

HSP-90 mediated protective effects of H₂S.

After I/R injury, the hepatic HSP-90 protein levels in the vehicle-treated mice remained at levels similar to those observed in the sham mice (Fig. 4, A and B). Conversely, the administration of H₂S significantly increased the expression of HSP-90 compared with the levels observed in the sham (P < 0.01) and vehicle-treated mice (P < 0.05) at both 5 and 24 h of reperfusion.

Fig. 4.

A and B: 90-kDa heat shock protein (HSP-90) protein expression in hepatic tissue, representative immunoblots, and densitometric analysis are shown. H₂S therapy (1.0 mg/kg) significantly increased HSP-90 protein expression at both 5 and 24 h of repefusion compared with vehicle group. C and D: representative immunoblot and densitometric analysis of hepatic Bcl-2 protein expression in the mitochondrial fraction at 5 and 24 h of reperfusion. Bcl-2 protein expression decreased significantly in both vehicle and H₂S groups. However, Bcl-2 levels were relatively preserved in animals treated with H₂S donor therapy. COX-IV, cytochrome c oxidase. **P < 0.01 vs. sham; ***P < 0.001 vs. sham; n = 6 for each group.

H₂S increased mitochondrial Bcl-2 levels and inhibited apoptosis.

To gain insights into apoptotic signaling, we evaluated the protein expression of the antiapoptotoic protein Bcl-2 at 5 and 24 h after ischemia (Figs. 4, C and D). I/R injury led to a significant reduction in Bcl-2 expression in both the vehicle- and H₂S (P < 0.001)-treated groups compared with the sham group. However, Bcl-2 protein levels were significantly (P < 0.05) higher in H₂S-treated mice compared with vehicle-treated mice at 5 h reperfusion. However, at 24 h of reperfusion, Bcl-2 was not increased in the H₂S-treated group.

Western blot analysis of the hepatic tissue lysate (Fig. 5, A and B) obtained after 60 min of ischemia and 24 h of reperfusion revealed a significant (P < 0.01) increase in the expression of cleaved caspase-3 in the vehicle group compared with the sham group. In contrast, the expression of cleaved caspase-3 in the H₂S-treated group only increased slightly.

Fig. 5.

Representative hepatic tissue immunoblot (A) and densitometric analysis (B) show expression of cleaved caspase-3 at 24 h reperfusion in sham, vehicle, and H₂S donor groups. Cleaved caspase-3 was significantly elevated in vehicle group compared with sham-operated controls. The H₂S donor (1.0 mg/kg) attenuated the increase in hepatic cleaved caspase-3. **P < 0.01 vs. sham; numbers inside bars represent number of animals investigated in each group.

DISCUSSION

In the current study, administration of H₂S before reperfusion significantly attenuated I/R injury in the liver. The conferred hepatoprotection was characterized by a significant attenuation of serum transaminases level and lipid peroxidation by an upregulation of antioxidant and antiapoptotic signaling.

I/R injury involves the activation of the inflammatory cascade characterized by leukocyte infiltration as well as generation of reactive oxygen and nitrogen species (21, 38). H₂S possesses a number of signaling actions that are likely to attenuate the pathological aspects of I/R injury. Recently, Zanardo et. al. (45) observed that H₂S inhibits aspirin-induced leukocyte adherence in mesenteric venules and subsequent diapedesis suggesting that H₂S is a potent anti-inflammatory agent. H₂S also directly scavenges of reactive oxygen and nitrogen species preventing tissue damage (36, 40). Furthermore, H₂S reduces the protein carbonyl formation after burn and smoke injury, indicative of overall antioxidant effects (7). GSH is a potent intracellular antioxidant, and the GSH-to-GSSG ratio corresponds to the capacity of a cell to attenuate oxidative stress (9). Thioredoxin like glutathione is a regulator of cellular oxidation/reduction (redox) status and has been shown to protect against oxidative stress (41, 46) and to inhibit apoptosis (41). Moreover, mice over expressing the human thioredoxin gene (hTrx tg) are resistant to renal I/R injury (22). In our study, we observed that H₂S preserved the GSH levels and Trx-1 expression after hepatic I/R injury, suggesting that H₂S maintains the intracellular antioxidant capacity of hepatocytes. I/R injury involves oxidative damage not only to proteins and nucleic acids but also to lipids. Stress-induced lipid peroxidation has been associated with development of atherosclerosis, neurodegeneration, and even carcinogenesis (13). In the current study, H₂S significantly attenuated lipid peroxidation suggesting its protective effects against oxidative stress.

Apoptosis plays a significant role in the pathophysiology of I/R injury (12, 20). Previously, Rose et. al. (32) demonstrated that H₂S abolished β-phenyethyl isothiocyanate induced apoptotic cell death in the human adenocarcinoma cell line HCT116, suggesting its potential to be a potent inhibitor of apoptosis. We found that the protein expression of HSP-90 was elevated in H₂S-treated mice. HSP-90 is known to attenuate the injury caused by stress by maintaining the conformation of proteins, refolding proteins damaged by stressful stimuli, and rendering cells more resistant to apoptosis. HSP-90, the most abundant heat shock protein present in eukaryotic cells, is constitutively expressed but is induced under conditions of stress like I/R injury (15). HSP-90 interacts with a variety of intracellular pathways to evade apoptosis and promote cell survival, including the phospatidylinositol 3-kinase-AKT pathway, IGF-1, and inositol hexakisphosphate kinase-2 (2, 42). Additionally HSP-90 is known to mediate the upregulation of Bcl-2 expression induced by VEGF further contributing to apoptosis inhibition (4). Bcl-2 is an antiapoptotic protein vital for the integrity of the mitochondrial membrane, as it prevents the release of proapoptotic protein such as cytochrome c (33). Rat liver transfected with human Bcl-2 gene (hBcl-2) has been shown to be resistant to I/R injury (43). In the present study, H₂S therapy increased the protein expression of HSP-90 and Bcl-2 after I/R injury and also inhibited apoptosis, as revealed by decreased levels of cleaved caspse-3 in H₂S-treated mice. H₂S has previously been shown to protect the ischemic myocardium by preserving mitochondrial function (6). Therefore, it is possible that H₂S mediates preservation of mitochondrial function after I/R injury by upregulating Bcl-2 expression.

H₂S is a potent reversible inhibitor of cytochrome c oxidase and has been shown to inhibit mitochondrial respiration in a dose-dependent manner (6). It is possible that supratherapeutic concentrations of H₂S depress mitochondrial respiration irreversibly, affecting ATP synthesis and contributing further to I/R injury. However, to assess a complete side effect profile of H₂S is beyond the scope of current study. Pharmacological preconditioning with H₂S has been shown to protect the heart from I/R injury, and endogenous H₂S has been suggested to play a role in ischemic preconditioning (1, 17). Further studies are required to assess role of H₂S preconditioning in hepatic I/R injury. In the current study, H₂S has been shown to be protective in nondiabetic animals but its role in conditions with enhanced oxidative stress including diabetes and obesity remains unknown. Clearly, additional studies are warranted to further our understanding regarding the efficacy of H₂S under these pathological conditions.

In summary, our findings demonstrate that H₂S significantly attenuates hepatic I/R injury via preservation of intracellular redox balance and by inhibition of apoptosis during the evolution of I/R injury. These results suggest that H₂S is a promising therapeutic agent to protect against hepatic I/R injury.

GRANTS

The study was supported by National Institutes of Health (NIH) Grant 2RO1-HL-60849-08 and American Diabetes Association Grant 7-04-RA-59 to D. J. Lefer and by NIH Grant F32-DK-077380-01 to J. W. Calvert.

DISCLOSURES

The study was partially supported by a research grant from Ikaria. D. J. Lefer is a consultant for Ikaria.