Background
Hepatic ischemia-reperfusion injury (I/Ri) is a serious and common adverse event during hepatic surgery that may ultimately lead to liver failure, systemic inflammatory response syndrome (SIRS) and even multiple organ failure syndrome (MOF) [
1‐
4]. Central to hepatic I/Ri is the generation of reaction oxygen species (ROS) by activated Kupffer cells or neutrophils upon the reintroduction of molecular oxygen to ischemic tissues. This pathogenic event triggers a series of deleterious effects that include oxidative modification of lipids and proteins, induction of apoptosis in hepatocytes, release of pro-inflammatory cytokines, increased expression of adhesion molecules, and infiltration of leukocytes, which together leads to massive tissue destruction [
5,
6].
To ameliorate the severity of liver I/Ri, several therapeutic strategies are currently being pursued, including the inhibition of apoptosis by decreasing cellular metabolism using the gas hydrogen sulphide (H
2S). Application of H
2S has shown promising activity in various pre-clinical I/Ri and transplantation models, including kidney and liver [
7,
8]. A second interesting strategy is to inhibit mitochondrial calcium overload, e.g. with 2-ABP [
9], and thus block the execution of mitochondrial apoptotic signaling [
10]. A third particularly appealing strategy is the use of anti-oxidants that directly counteract the deleterious effects of ROS. In this respect, dietary anti-oxidative supplements such as rutin and L-arginine have shown beneficial effects on severity of hepatic I/Ri [
11]. Moreover, carbon monoxide (CO) has raised particular therapeutic interest because of its potent anti-oxidant and anti-inflammatory activity.
CO is best known as an odorless and toxic gas which upon inhalation binds with high affinity to heme, thereby forming carboxyhemoglobin and severely impairing the respiratory system. However, CO is also produced by the protein heme oxygenase (HO) and as such functions as a potent endogenous antioxidant that counteracts toxic effects of ROS. HO-1 degrades heme into biliverdin, free iron, and CO [
12] and is one of the most prominent lines of cellular defense against damage induced by I/Ri. Indeed, during oxidative stress the expression of the inducible form of HO, (HO-1) is strongly up-regulated in the liver [
13].
Therapeutic up-regulation of CO tissue levels can be achieved via exogenous application of CO, for instance by direct inhalation of CO gas. In
ex vivo isolated liver perfusion and liver transplantation models exogenously added CO yields potent cytoprotective effects [
14‐
16]. Moreover, CO has anti-inflammatory activity e.g. by inhibiting the inflammatory response of donor Kupffer cells upon transplantation [
17], activation of anti-inflammatory mitogen-activated protein kinase signaling [
18], and up-regulation of the anti-inflammatory cytokine IL-10 [
19]. Furthermore, CO inhibits the acquisition of a pro-adhesive phenotype of vascular endothelial cells [
20].
Most studies on the role of CO in the amelioration of inflammatory responses have been performed by administration of gaseous CO [
21]. However, the applicability of gaseous CO is limited by its toxic effect on cellular respiration. Therefore, therapeutic use of CO as a cytoprotective agent clearly requires a pharmaceutical formulation that allows for the selective delivery and/or local release of CO from a non-toxic pro-drug. In this respect, transitional metal carbonyl-based compounds are of particular interest because of their capability for controlled release of CO in biological systems [
22]. These so-called CO-releasing molecules (CORMs) have been therapeutically tested in a variety of experimental inflammatory models with promising anti-inflammatory responses [
12,
23,
24]. Indeed, treatment of septic mice with tricarbonyldichlororuthenium (II) dimer (CORM-2) strongly attenuated liver inflammation, as evidenced by a decrease in serum liver damage markers and a reduced influx of inflammatory cells[
25]. In addition, CORM-2 was recently reported to improve outcome of ischemia reperfusion injury to the small bowel [
26]. Importantly, CORM treatment is associated with low or minimal formation of carboxyhemoglobin and is therefore considered a safer alternative to CO gas inhalation [
22].
Based on the above, we hypothesized that CORM-2 may ameliorate damage occurring during hepatic I/Ri. Here, we investigated this hypothesis in a rat model of experimentally induced hepatic I/Ri.
Methods
Rats and experimental procedure
All animal experiments were performed in accordance with the experimental protocol approved by the Committee for Research and Animal Ethics of Harbin Medical University. For the experiments, healthy male Wistar rats (n = 40; body weight, 230-250 g) were purchased from the Central Animal Facility of the First Clinical Hospital of Harbin Medical University (Harbin, China). Rats were housed in cages under standard animal care conditions and fed with rat chow ad libitum. All surgical procedures were performed under general anesthesia using sodium pentobarbital (50 mg/kg i.p.). Normothermic partial hepatic ischemia was induced by performing a midline laparotomy, exposing the liver hilum and subsequent clamping of portal structures to the left and median lobes with a microvascular clip, yielding ~70% hepatic ischemia [
27]. The abdomen was covered during the ischemic period. After 60 min of partial hepatic ischemia, the clip was removed to initiate hepatic reperfusion and the abdominal cavity was closed with a 4-0 silk suture. Temperature was maintained at 37°C by a warming pad. Sham-operated rats underwent the same procedure without clamping the pedicle of the liver lobes.
Rats were randomly assigned into four groups with a sample size of 10. Sham group underwent a sham operation and received saline. I/R group underwent the hepatic I/R procedure and received saline. CORM-2 group underwent the same procedure and received 8 mg/kg of CORM-2 (Sigma-Aldrich, St. Louis, MO). iCORM-2 group underwent the same procedure and received iCORM-2 (8 mg/kg), which does not release CO. All treatments were administered prior to reperfusion. Rats were sacrificed by exsanguination at 6 hours post reperfusion upon which serum and liver samples were collected according to standard procedures. Rats in all experimental conditions survived the 6 hour reperfusion period.
Serum transaminase determination
At the end of reperfusion, 5 ml of blood was collected from the caval vein in heparinized tubes. Samples were centrifuges at 800 g for 10 minutes at room temperature. Plasma was then used to evaluate the extent of hepatic injury by measuring serum levels of ALT and AST using an Automated Chemical Analyzer (7600; Hitachi, Tokyo, Japan). Values were expressed as units per liter (U/L).
Liver histopathology score
Formalin-fixed liver samples were embedded in paraffin, sectioned into 5-μm thickness and stained with hematoxylin-eosin and microscopically inspected to assess inflammation and tissue damage. Histological examination of hepatic tissue damage was performed by two liver pathologists in a blinded fashion. Ten separated microscopic fields were scored on a scale from 0 to 3 as described before [
28]. The severity of tissue damage was defined as: grade 0, minimal or no evidence of injury; grade 1, mild injury consisting of cytoplasmic vacuolation to focal nuclear pyknosis; grade 2, moderate to severe injury with extensive nuclear pyknosis, cytoplasmic hypereosinophilia, and loss of intercellular borders; grade 3, severe necrosis with disintegration of hepatic cords, hemorrhage, and neutrophil infiltration.
Detection of apoptotic cells by terminal deoxyribonucleotide transferase (TdT)-mediated dUTP nick-end labeling (TUNEL) staining
TUNEL staining (Roche, Shanghai, China) was applied to paraffin-embedded 5-μm liver sections to detect DNA fragmentation as a measure for the number of apoptotic cells. Counterstaining was performed with 4',6-diamidine-2'-phenylindole dihydrochloride (DAPI) dye. Briefly, deparaffinized livers sections were incubated in permeabilization solution (0.1% TritonX-100 in 0.1% sodium citrate) for 2 min on ice, incubated with TUNEL reaction mixture for 60 min at 37°C in the dark and incubated for 4 min with DAPI dye in the dark. The numbers of apoptotic cells and total hepatic cells were counted in ten separated microscopic fields under 400× magnification. Numbers were then averaged and used to calculate the apoptosis index (AI) according to the previously reported formula: AI = (apoptotic cells/total hepatic cells) × 100% [
29].
Myeloperoxidase activity assay
Myeloperoxidase (MPO), an enzyme found predominantly in azurophilic granules of polymorphonuclear leukocytes, was measured as an index of neutrophil infiltration in the ischemic liver Briefly, liver tissue was homogenized, centrifuged at 12.000 g for 15 minutes at 4°C, after which MPO activity was measured using a commercial kit (NJJC Bio Inc., Nanjing, China) according to manufacturer's instructions. Absorbance was measured at 460 nm with a spectrophotometer. MPO activity was expressed as units per gram protein (U/g).
Cytokine Analysis
Serum samples were obtained from rats 6 h after the I/R procedure, at the time-point of termination, and stored at -20°C until further analysis. Serum samples were analyzed for TNF-α and IL-6 levels using commercially available enzyme-linked immunosorbent assay (ELISA) kits (R&D Systems, Minneapolis, MN) according to manufacturer's instructions. Cytokine levels were expressed as picogram per ml (pg/ml).
Western blot analysis
Liver whole cell homogenates and nuclear extracts were obtained by lysis of hepatic tissue with the Nuclear Extract Kit (Active Motif, Carlsbad, CA) according to manufacturer instructions. Protein aliquots (50 μg) were subsequently separated by SDS-PAGE on 5% or 10% acrylamide gels and proteins were electrotransferred to PVDF membrane. Protein levels were visualized by incubation with the following antibodies: bcl-2, caspase-3, ICAM-1, HO-1, NF-κB p65, β-actin, and Histone H1 (all from Santa Cruz Biotechnology, Inc. Santa Cruz, CA). Specific binding of antibodies was visualised using appropriate horseradish peroxidase-linked secondary antibody followed by detection of chemoluminescence using an enhanced chemoluminescence detection kit (Roche) according to manufacturer's instructions. Blots were stained with anti-β-actin or Histone H1 antibody to verify equal protein loading.
Liver nuclear factor-kappa B (NF-κB) activation
Activation of the transcription factor NF-kB was measured using a commercially available ELISA kit (Trans-AM™ NF-kB p65 Transcription Factor Assay Kit, Active Motif, Carlsbad, CA) according to manufacturer's instructions. Nuclear protein extract was obtained using Nuclear Extract Kit (Active Motif) according to manufacturer's instruction. Subsequently, 15 μg nuclear protein extract was used to assay for NF-kB activation. Values were represented as OD450 nm.
Statistical analysis
All values are expressed as the mean ± standard deviation (SD). Data were analyzed by one-way ANOVA followed by the Tukey-Kramer post test. For the semi-quantitative histopathological scores, statistical analysis was performed using the Kruskal-Wallis test followed by the Dunn's post test. P < 0.05 was considered to indicate statistical significance.
Discussion
Endogenous CO produced by HO-1 is an important cellular protective measure to prevent cytotoxic and pro-inflammatory effects during reperfusion injury. Here we show that exogenous CO released by CO-releasing molecule 2 (CORM-2) can be applied to reduce hepatic ischemia reperfusion injury (I/Ri), a common adverse event during liver surgery that is characterized by hepatocellular death and inflammatory cell influx. In our model we demonstrated that CORM-2 treatment reduced the extent of apoptosis and ameliorated the pro-inflammatory stress response as evidenced by a reduction in the expression of pro-inflammatory cytokines, vascular endothelial adhesion molecule and a markedly reduced influx of leukocytes
Importantly, therapeutic application of CO inhalation is severely hampered by the deleterious effects on the respiratory system due to carboxyhemoglobin formation. For instance, inhalation of 500 ppm gaseous CO in humans resulted in a peak carboxyhemoglobin level of 7%, whereas in animal studies levels of up to 25% were detected. In contrast, treatment with CO-releasing molecules such as CORM-2 does not lead to a dramatic increase in carboxyhemoglobin. Indeed, treatment with CORM-2 at doses up to 20 μmol/kg (10,25 mg/kg) had no negative impact on oxy-haemoglobin saturation [
48]. Thus, CORM-2 appears to be a potent inhibitor of negative effects of hepatic I/Ri, while at the same time having no appreciable negative effects on the respiratory system.
Both the cytoprotective and anti-inflammatory activity of CO appear to result, at least in part, from its ability to modulate the transcription factor NF-κB. For instance, CO-treatment of hepatocytes induces activation of NF-κB
in vitro and
in vivo [
33], which renders these cells more resistant to apoptosis by up-regulating the anti-apoptotic protein Bcl-xL and down-regulating the pro-apoptotic protein Bax [
49]. Similarly, we found a marked up-regulation of the anti-apoptotic protein Bcl-2 upon CORM-2 treatment. Since the balance between pro- and anti-apoptotic members of the Bcl-2 family is central to the control of the mitochondrial pathway of apoptosis, this increase in Bcl-2 expression is likely to inhibit execution of mitochondrial apoptosis. Of note, pre-treatment of LPS-stimulated human umbilical vein endothelial cells (HUVEC) with CO showed a reverse effect, namely inhibition of NF-κB activity [
50]. As a result, CO-treated endothelial cells showed a reduced expression of adhesion molecules, which may reduce pro-inflammatory processes such as leukocyte adhesion and tissue infiltration of inflammatory cells. Thus, CO can have opposite effects on NF-κB signaling depending on the particular cell type involved. Further detailed investigation, using e.g. laser dissection microscopy may yield insight into the effect of CO on hepatocytes and hepatic vascular endothelium
in vivo. However, from the above it is clear that these diverse effects on NF-κB cooperate to ameliorate cell damage and minimize inflammation.
In addition to NF-kB, protective effects of CO-released from CORM-2 may be related to the down-regulation of the iNOS/NO pathway in e.g. macrophages.
In vitro treatment of LPS-stimulated macrophages with CO indeed prevented expression of iNOS and blocked the pro-adhesive phenotype [
24,
51]. Furthermore, treatment of I/Ri in a rat liver transplantation model using gaseous CO was partly attributable to down-regulation of iNOS/NO [
16].
As anticipated, the induction of pro-inflammatory cytokines such as TNF-α during hepatic I/Ri is markedly decreased by treatment with CORM-2. Together with the accompanying decrease in expression of adhesion molecules these effects are likely accountable for the reduction in influx of inflammatory cells. The exact mechanism for down-regulation of TNF-α by CORM-2 treatment is still a matter of debate. Various reports have indicated that this effect might be attributable to direct CO-effects on vascular endothelium and circulating leukocytes. Indeed, CO has potent anti-inflammatory effects on LPS-stimulated HUVEC cells and macrophages [
24,
25,
51]. Another possible contributing factor to the reduction in TNF-α level upon CORM-2 treatment is the rescue of hepatocytes from apoptosis. Apoptosis of hepatocytes is a universal feature of liver inflammation and is associated with the production of various inflammatory cytokines. Thus, the marked reduction in apoptotic hepatocytes upon CORM-2 treatment might contribute to the downplaying of the inflammatory response.
Of note, exogenous application of CORM-2 had an augmenting effect on the expression levels of HO-1, indicating that the exogenous addition of one of the reaction products of HO-1 has a positive feed forward effect on HO-1 expression. Since activation of the HO system by an HO-1 inducer or by HO-1 gene therapy enhances hepatoprotection against warm and cold I/Ri in experimental animals [
46,
47], HO-1 upregulation upon treatment with CORM-2 may contribute to the beneficial effects on severity of I/Ri. Indeed, products of the HO-1 enzyme such as bilirubin have well-documented cytoprotective and anti-oxidative activity. Further experiments, e.g. using specific HO-1 inhibitors such as zink protoporhyrin or OB-14 [
52], may be used in conjunction with CORM-2 treatment to determine the relative contribution of these HO-1 products."
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
YW carried out acquisition of data. PC carried out acquisition of data and performed statistical analysis. MB carried out acquisition and interpretation of data. WZ designed the study. EB analyzed and interpreted data and wrote the manuscript. WH: participated in coordination and writing. All authors read and approved the final manuscript.