Calpain Inhibition Attenuates iNOS Production and Midzonal Hepatic Necrosis in a Repeat Dose Model of Endotoxemia in Rats

Abstract

Systemic exposure to bacterial lipopolysaccharide (LPS, endotoxin) induces hypotension, disseminated intravascular coagulation and neutrophil infiltration in various organs including the lung, kidney and liver. A rat endotoxemic neutrophilic hepatitis model (repeat dose LPS, 10 mg/kg, i.v. 24 hours apart) was developed exhibiting hepatic neutrophil infiltration and mid-zonal hepatic necrosis. The goal of the study was to investigate the role of the intracellular enzyme calpain in the development of neutrophilic hepatitis with midzonal necrosis in this model. A second goal was to compare the observed protective effects of calpain inhibition with a relatively selective inducible nitric oxide synthase (iNOS) inhibitor aminoguanidine (AG) and an inhibitor of coagulation, heparin. When compared to rats administered LPS alone, administration of calpain 1 inhibitor prior to LPS significantly reduced hepatic iNOS expression, hepatic neutrophil infiltration and attenuated midzonal hepatic necrosis. Administration of AG or heparin prior to LPS also decreased liver iNOS expression, hepatic neutrophil infiltration and liver pathology comparable to calpain inhibition. Blood neutrophil activation, as measured by the neutrophil adhesion molecule CD11b integrin, was upregulated in all the LPS treated groups regardless of inhibitor administration. We conclude that amelioration of liver pathology via calpain inhibition is likely dependent on the down-regulation of iNOS expression in the rat model of LPS-mediated hepatitis.

Introduction

Lipopolysaccharide (LPS, endotoxin) is a component of the outer membrane of the cell wall of gram negative bacteria, which are often normal inhabitants of the mammalian intestinal tract (Uesugi et al., 2002). Enteric bacterium often serve as a source for systemic LPS exposure and under certain conditions, such as those associated with chronic alcohol consumption, increased gut-derived LPS is released into the portal circulation (Uesugi et al., 2002; Keshavarzian et al. 2003). LPS in the blood is bound by lipopolysaccharide binding protein, which interacts with CD14 receptors on hepatic Kupffer cells, setting off a series of pro-inflammatory events (Bautista, 2002; Limaye et al., 2003). First, CD14 mediated activation of the toll-like receptor 4 results in NF-κB regulated production of cytokines, including IL-1β, IL-6, IL-8, TNF-α and platelet-activating factor along with upregulation of the inducible isoform of nitric oxide synthase (iNOS) (Kamanaka et al., 2003; Luyendyk et al., 2003; Nagy, 2003). These strong mediators of inflammation leads to neutrophil activation, generation of reactive oxygen species (ROS) and hepatic injury (Enomoto et al. 2000).

Activated neutrophils express surface adhesion molecules such as CD11/18 integrins, which bind with complementary adhesion molecules, such as ICAM-1 and VCAM-1, on vascular endothelium and hepatocytes promoting hepatic transmigration (Jaeschke and Smith, 1997; Jilma et al., 1999; Picker, 1999). Concurrent with neutrophil activation, LPS mediated circulatory changes include reduced mean arterial pressure, microthrombi formation and altered microcirculation which impair hepatic perfusion and promotes leukocyte adhesion with subsequent tissue transmigration (Palmes et al., 2004; Asakura et al., 2005). Transmigrated neutrophils can elicit tissue damage through release of ROS, proteolytic enzymes and additional chemotactic factors further driving neutrophil mediated hepatitis (Ikeda et al., 2002).

Recent literature suggests that calpain may play an important role in liver injury mediated by carbon tetrachloride and acetaminophen (Limaye et al., 2003; Mehendale and Limaye, 2005). Limaye et al. (2003) reported that calpain inhibition substantially decreased the progression of liver injury and led to protection against carbon tetracholoride-induced lethality. Emerging evidence also suggests the role of calpain as a pro-inflammatory molecule by activation of nuclear factor kappa-B (NF- κ B) (McDonald et al., 2001; Riedemann et al., 2003) and the subsequent production of a variety of pro-inflammatory mediators and adhesion molecules (Eipel et al., 2004; Czermak et al., 1999). However, the protective role of calpain inhibition in endotoxin-mediated hepatitis is currently not known and is the basis of this investigation.

Because of the exclusive midzonal necrosis noted in our model and association of such lesions with LPS mediated hypotension (Imai et al., 1994; Landry et al., 2001) and microvascular disturbance (Asakura et al., 2005), the role of iNOS and coagulation was also investigated in this study. We thus compared the results of calpain inhibition with those of an iNOS inhibitor aminoguanidine and an inhibitor of coagulation and microthrombi formation, heparin in our rat model. Finally we evaluated the effect of calpain inhibition on cytochrome P450 2E1 (CYP2E1) enzyme in our LPS-mediated hepatitis. CYP2E1 was selected since it is known to be down-regulated in inflammatory conditions (Richardson and Morgan 2005) and this down-regulation has been reported to be mediated by nitric oxide (NO) (Udosen et al., 2003). The results indicate that the protection afforded by calpain inhibition is due to iNOS inhibition and likely restoration of liver microperfusion and not due to β₂ integrin-dependent neutrophil activation or CYP2E1 inhibition.

Materials and Methods

Induction of acute endotoxin–mediated neutrophilic hepatitis

Two doses of LPS (10 mg/kg, i.v. via the tail vein) were administered to male Sprague–Dawley rats (250–270 g body weight) 24 hours apart based on pilot studies performed in our laboratory. All sample collection times in this study occurred 6 hours after the second set of LPS injections. Samples from saline treated control animals were taken concurrent with treated animals. Inhibitors were administered 1 hour prior to each dose of LPS (Figure 1). All animal procedures were approved by the Texas A&M University Laboratory Animal Care Committee (ULACC) and conformed to the standards set forth in the Guide for the Care and Use of Laboratory Animals (NRC, 1996). All rats were barrier raised rodents obtained from a commercial vendor (Harlan, Indi-anapolis, IN) and were housed in an AAALAC accredited facility. Standard rodent chow (Teklad 8604, Madison WI) and reverse osmosis water were supplied ad libitum.

Calpain inhibitor study

This study was designed to assess the effect of calpain inhibition on neutrophilic infiltration and necrosis in a rat endotoxemic model of hepatitis. Male Sprague–Dawley rats (250–270 g body weight) were divided into three groups (N = 5). One group served as a saline vehicle control. A second group received LPS (E Coli serotype 0111: B4, Sigma Chemical Co., St Louis, MO) at a dose of 10 mg/kg iv in saline twice at 24-hour intervals. The third group was administered calpain 1 inhibitor (Sigma Chemical Co., St. Louis, MO.) at 10 mg/kg i.p., followed 1 hour later by LPS in saline at a dose of 10 mg/kg via tail vein injection with a second dose of calpain inhibitor and LPS treatment 24 hours later. All rats were sacrificed via CO₂ inhalation 6 hours after the second LPS injection.

Aminoguanidine and Heparin Study (Comparison Study)

The objective of this study was to compare the effect of calpain inhibition with iNOS and coagulation inhibition in a rat endotoxemic model of neutrophilic hepatitis and necrosis. Male Sprague–Dawley rats (250–270 g body weight) were divided into 4 groups (N = 6). One group served as a saline vehicle control with the 3 other groups administered LPS in saline twice, 24 hours apart, via tail vein injection at a dose of 10 mg/kg. One hour prior to each dose of LPS, one group received calpain 1 inhibitor as described above. Another group received the relatively selective iNOS inhibitor aminoguanidine (Sigma, St Louis, MO) in saline at a dose of 45 mg/kg intraperitoneally 1 hour prior to each LPS dose. The final group received heparin sodium (America Pharmaceutical Partners, Schaumburg, IL) at a dose of 1000 units/kg subcutaneously 1 hour prior to each LPS dose. The rats were sacrificed via CO₂ inhalation 6 hours after the second injection of LPS.

Sample Collection and Processing

Whole blood was collected via the abdominal vena cava and immediately transferred to a lithium heparin, EDTA or sodium citrate tube (Monoject, Sherwood Medical, St. Louis, MO). The samples in lithium heparin tubes were used for evaluation of neutrophil activation via flow cytometry. Samples in EDTA tubes were used for hematology evaluation and the samples in sodium citrate tubes were used to assess coagulation parameters. Representative liver sections were collected from different liver lobes and was consistently from the same lobe in different rats. The samples collected were placed in liquid nitrogen and stored at −80°C for further analysis. The remaining liver along with sections of the lung and kidney were collected and placed in 10% neutral-buffered formalin. The liver was processed and slides stained with hematoxylin and eosin (H&E) for histopathological evaluation or stained for choloracetate esterase to evaluate hepatic neutrophil infiltration. Kidney and lung sections were processed and stained with Mallory’s phosphotungstic acid hematoxylin (PTAH) for evaluation of fibrin.

Flow Cytometry Studies

To investigate treatment effects on neutrophil activation, heparinized whole blood samples from 3 rats per group were used to assess the neutrophil adhesion molecules Mac-1 (CD11b/18). Blood was divided into 250 uL aliquots in separate glass tubes and was kept on ice unless otherwise stated. Duplicate tubes of blood were designated to antibody stained or unstained groups. Prior to staining with the appropriate conjugated antibody, 10 μl of reagent grade mouse IgG (Sigma, St. Louis, MO) mixed at a concentration of 1 mg/ml was added to all but the unstained samples to reduce nonspecific binding of the conjugated antibodies. In the next step, neutrophils in designated tubes were labeled with fluorescein-conjugated HIS 48 (BD Biosciences, San Jose, CA). Mac 1 integrins were labeled with R-phycoerythrin conjugated CD11b (Serotec, Raleigh, NC). Following incubation with antibodies (20 min), the samples were treated with RBC lysis buffer for 10 minutes. The tubes were then centrifuged at 1000 rpm for 10 minutes at 3°C and resuspended several times with the lysis buffer until erythrocytes were removed. After lysing steps, the cells were washed with 250 μL of PBS and centrifuged at 1000 rpm for 10 minutes. Cells were re-suspended in 300 μL of PBS for analysis.

The cells were analyzed using a FACS Calibur (Becton Dickinson Immunocytometry Systems, San Jose, CA) flow cytometry, equipped with a 15mW air-cooled argon laser. Green fluorescence from fluorescein was collected through a 530/30-nm bandpass filter and orange fluorescence from R-phycoerythrin through a 585/42-nm bandpass filter. List mode data was acquired on a minimum of 10,000 events defined by light scatter gates to include the neutrophil population. Data analysis including spectral compensation was performed using FlowJo (Treestar Inc., Ashland, OR). A neutrophil gate was established based on side scatter and positive HIS48 staining, and the median fluorescence channel was determined for CD11b staining.

Hematology evaluation (calpain inhibitor study only)

Complete blood counts for all animals in the calpain inhibitor study were performed on a Cell-Dyn 3700 (Abbott Diagnostics, Abbott Park, IL) hematology analyzer. Wright-Geimsa stained blood smears were also evaluated microscopically to determine hematologic abnormalities.

Histopathology/Hepatic Neutrophil Infiltration

Cross-sections of liver collected from the left lateral and median lobes were fixed in 10% neutral buffered formalin, processed by standard histological methods, and embedded in paraffin. Then, 5-micron thick sections were stained with hematoxylin and eosin and evaluated for oncotic necrosis and apoptosis. Two liver sections per rat were evaluated and the microscopic lesions were scored using a semiquantitative scale as follows: 0-no evidence of lesions; 1–periportal infiltrate only; 2–periportal infiltrate with ≤2 foci of necrosis; 3–periportal infiltrate with 3–5 foci of necrosis; 4–periportal infiltrate with ≥6 foci of necrosis. The lesion distribution was scored as; P-periportal predominate; M-midzonal predominate and C-centrilobular predominate.

The H&E staining was also employed to identify the neutrophils, based on the segmented morphology of the nucleus. To quantitate the degree of neutrophilic inflammation, unstained paraffin embedded slides were deparaffinized then stained with a chloroactetate esterase kit (Sigma, St Louis, MO) according to the manufacturer’s recommendations. The number of neutrophils in 10 randomly chosen low power fields (20X) were recorded and the results expressed as the mean number per 20X field as described previously (Banerjee et al., 2006). To evaluate the effect of circulating microthrombi in this model, cross sections of the left lung lobe and longitudinal cross sections of the left kidney were fixed in neutral buffered formalin, processed by standard histological methods and embedded in paraffin. 5 micron sections were stained with phosphotungstic acid hematoxylin solution and evaluated for the presence of fibrin. The number of fibrin foci in 10 randomly chosen low power fields (10X) were recorded and the results expressed as the mean number per 10X field for kidney and lung sections.

Western Blot Analysis for iNOS and α-fodrin

α-Fodrin is a substrate for calpain and was used to confirm calpain inhibition. Liver lysates were prepared by grinding 300 mg frozen liver suspended in liquid nitrogen with a mortar and pestle in a lysis buffer containing 1% Triton X-100, 150 mM NaCl, 10 mM Tris (pH 7.4), 1 mM EDTA, 1 mM EGTA, 2 mM Na vanadate, 0.2 ml PMSF (200 mM PMSF in iso-propanol), 1 mM Hepes (pH 7.6), 1 μg/ml leupeptin, and 1 μg/ml aprotinin. Sample protein concentrations were determined using the method of Bradford (1976). Polyacrylamide gel (8%) electrophoresis was employed for α-fodrin analysis and 10% gel was employed to assess iNOS expression. After transferring the proteins onto nitrocellulose membranes, the membranes were blocked for 3 hours with 6% milk made with 1X TBS/Tween 20. After the blocking step, the membranes were incubated with primary rabbit polyclonal antibody for α-fodrin at 1:2500 dilution (Abcam, Cambridge, MA) and rabbit polyclonal antibody for iNOS at 1:2000 dilution (Cayman chemical company, Ann Arbor, MI). Subsequently, the nitrocellulose membrane was incubated with secondary goat anti rabbit antibody (Kirkgaard and Perry Labs, Gaithersburg, MD) in 2% milk made with 1X TBS/Tween 20 at a 1:4000 dilution. The blots were visualized using enhanced chemiluminescence kit (Pierce, Rockford, IL) and quantitated by densitometry.

Western Blot Analysis for Cytochrome P450 (CYP) 2E1

Preparation of Microsomes

Frozen liver tissue was homogenized (1:5 w/v) using an Ultra-Turrax 25 homogenizer (IKA, Wilmington, NC) in ice-cold Tris-acetate buffer (pH 7.4) containing 1.15% KCl. The homogenate was centrifuged at 10,000g for 30 minutes at 4°C. The supernatant was decanted and centrifuged at 100,000g for 60 minutes at 4°C. Microsomal pellets were resuspended, manually homogenized and again centrifuged at 100,000g for 60 min at 4°C. Microsomal pellets were again recovered, suspended in assay buffer (pH 7.4), manually homogenized then quick-frozen and stored at −70°C for later use. Sample protein concentrations were determined using the method of Bradford (Bradford, 1976).

Western Blot Analysis for CYP2E1

Microsomal protein from each rat was separated by SDS polyacrylamide gel (10%) electrophoresis and transferred to nitrocellulose membrane. After the blocking step, the membrane was incubated with a rabbit anti-human polyclonal anti-CYP2E1 primary antibody, 1:1000 dilution (generously gifted by Dr. Jerry Lasker, Hackensack University Medical Center, New Jersey) for 2 hours and further probed with HRP-linked goat anti-rabbit secondary antibody (Kirkgaard and Perry Labs, Gaithersburg, MD; 1;10,000 dilution) for 1 hour. Primary Rabbit polyclonal antibody for cytochrome 4A was supplied by Abcam (Cambridge, MA). The blots were visualized using enhanced chemiluminescence (Pierce, Rockford, IL).

Coagulation Parameters

To determine treatment effects and confirm efficacy of heparin administration, the values for prothrombin time (PT) and activated partial thromboplastin time (APTT) were determined from citrated blood samples collected from 3 rats per selected group (control, LPS only and LPS plus heparin) using the Beckman Coulter ACL 9000 coagulation analyzer (Fullerton, CA).

Statistics

To identify significant differences among groups, analysis of variance (ANOVA) with a post hoc Tukey’s HSD was performed using a computer software program (SPSS version 12.0, Claritas Inc.). Results were considered significant at the 0.05 confidence level.

Results

Clinical Signs

No mortalities were recorded. Rats administered LPS alone appeared more lethargic than rats in the other groups and developed diarrhea within 1–2 hours after LPS administration. Rats given LPS plus calpain1 inhibitor appeared more lethargic than untreated controls but were brighter subjectively than LPS only rats. Rats administered LPS plus calpain 1 inhibitor developed diarrhea but the time to onset was slower (4–6 hours) as compared to LPS only rats. Rats given LPS plus aminoguanidine or LPS plus heparin appeared similar to rats administered LPS plus calpain 1 inhibitor. Due to the short duration of these studies (a total of 30 hours from initial LPS injection), post treatment body weights, food consumption and water consumption were not measured.

Flow Cytometry

The neutrophil adhesion molecule CD11b was significantly upregulated in the LPS, and LPS plus calpain 1 inhibitor as compared to untreated controls confirming neutrophil activation in both the calpain 1 inhibitor and the comparison study. CD 11b was also significantly upregulated in the LPS plus aminoguanidine and LPS plus heparin groups as compared to untreated controls in the comparison study (Figure 2). However, when comparison of CD11b levels was made between aminoguanidine and heparin with calpain inhibitor a mild decrease in neutrophil activation was noted.

Hematology

Calpain 1 inhibitor study-absolute neutrophil counts and neutrophil to lymphocyte ratios were elevated in LPS and LPS plus calpain 1 inhibitor groups as compared to untreated controls. There was no significant difference in absolute neutrophil numbers or neutrophil to lymphocyte ratio between LPS and LPS plus calpain 1 inhibitor groups (Table 1). Morphologically, toxic neutrophil changes such as Dohle body formation and toxic cytoplasmic granulation were present in both LPS and LPS plus calpain inhibitor groups. Nucleated red blood cells were a common feature of LPS and LPS plus calpain inhibitor dosed rats but were absent in controls.

Histopathology

Neutrophilic midzonal hepatocyte oncotic necrosis with apoptosis and periportal mononuclear inflammatory cell infiltrates were a major feature in the LPS only group as demonstrated by semiquantitative scoring (Tables 1 and 2) compared to the controls (Figures 3A–C). Significant decreases in necrosis and neutrophilic inflammation was identified in all of the inhibitor groups as demonstrated by semiquantitative scoring (Tables 1 and 2), although rare necrotic foci was identified in the iNOS inhibitor group (Figures 3D–F). There was reduction in fibrin deposition in the PTAH stained lung and kidney of only the heparin plus LPS group as compared to the LPS only group (Table 2). No significant difference in fibrin deposition was found in the lung or kidney among the calpain 1 inhibitor or aminoguanidine groups as compared to the LPS only group (Table 2).

Hepatic Neutrophil Scoring

Calpain 1 inhibitor study—There were significant differences in liver neutrophil counts among the 3 groups (Table 1). Control rats had significantly lower neutrophil counts as compared to all the other groups. LPS only rats had the highest neutrophil counts. However, administration of calpain 1 inhibitor significantly decreased the hepatic neutrophils counts. Notably, blood neutrophil count was not different between the LPS alone and the calpain inhibitor group.

Comparison Study

Control rats had significantly lower neutrophil counts as compared to all other groups (Table 2). LPS only rats had the highest hepatic neutrophil counts and were significantly higher than any other inhibitor group. The LPS plus calpain 1 inhibitor group had significantly reduced neutrophils (~70% reduction) as compared to LPS alone group. Hepatic neutrophil counts for the heparin (~60% reduction) and aminoguanidine (~65% reduction) groups were also significantly lower than the LPS only group but not different from the LPS plus calpain 1 inhibitor group (Table 2).

Western Blotting for α-Fodrin

To evaluate the degradative effect of released calpain on its substrate α-fodrin, a Western blotting technique was used. Calpain is known to degrade the 240-kDa α-subunit of fodrin at several sites to produce breakdown products including a 150 kDa, 120 kDa and 35 kDa product (Goll et al., 2003). In normal control rat livers almost all α-fodrin was in intact native form. The native form of α-fodrin was significantly lower in the LPS alone group suggesting calpain activation and fodrin degradation. However, calpain inhibition by calpain inhibitor 1 resulted in restoration of the intact form of fodrin to control levels confirming calpain inhibtion (Figure 4).

Western Blotting for iNOS

Densitometry results of iNOS protein expression revealed significantly lower levels in the calpain 1 inhibitor group as compared to the LPS only group and the levels were almost comparable to the untreated control group (Figure 5). iNOS expression in the aminoguanidine and heparin groups was also significantly lower than the LPS only group and was comparable to the calpain 1 inhibitor group (Figure 5). It has been reported that iNOS activity as measured in vitro based on the conversion of L-arginine conversion to L-citrulline strongly correlates with iNOS protein expression (Jinno-Oue et al., 2003). Based on this it was assumed that iNOS western blotting results correlate with the catalytic activity in this study.

Western Blotting for CYP2E1

CYP2E1 expression was down-regulated in the LPS only group as compared to calpain 1 inhibitor and controls. There was no significant difference between CYP2E1 expression among the aminoguanidine and heparin groups as compared to calpain 1 inhibitor and controls (Figures 6a and 6b). A recent study from our lab and others have shown that CYP2E1 protein expression strongly correlates with CYP2E1 catalytic activity as evaluated by chlorzoxozone 6-hydroxylase and p-nitrophenol hydroxylase assay. (Turner et al., 1988; Amato et al., 1998; Banerjee, et al., 2006). Because of this strong correlation, we did not measure CYP2E1 activity in this study.

Coagulation Parameters

The mean PT and APTT values were employed to detect inhibition of major pathways of coagulation. PT and APTT values for the LPS plus heparin group were elevated as compared to the control and LPS only groups confirming inhibition of coagulation by heparin (Table 2). APTT values were elevated by LPS alone compared to the controls with further prolongation of APTT values by heparin.

Discussion

The above experiments investigated the protective role of calpain in a LPS mediated neutrophilic hepatitis model in the rat. The interesting observation noted in this model was the exclusive midzonal distribution of necrosis. Although, midzonal lesions have been described by others during states of endotoxemia, these lesions were generally accompanied by periportal necrosis (Pirisi et al., 2000; Katsushige et al., 2005). However, the necrosis in the present study originated primarily in the midzonal region. One question that can be potentially raised is whether the neutrophils are arriving in response to midzonal necrosis or the necrosis noted the result of neutrophil infiltration. Both scenarios may be operating in this study. However, based on the absence of necrosis in the single LPS dose study in the presence of midzonal neutrophil infiltration (unpublished findings), it is likely that hepatic neutrophil infiltration may be preceding the necrosis in this study.

Primary midzonal hepatic lesions are uncommon but have been associated with viral yellow fever and shock associated hypotension (de la Monte et al., 1984). Yee and coworkers have also described midzonal necrosis following large doses of LPS in rats (Yee et al., 2003). It is reported that excessive production of nitric oxide (NO) as a consequence of macrophage release of iNOS contributes to the hypotension which is a feature of endotoxemia (Imai et al., 1994; Landry and Oliver, 2001). NO is known to play a dual role in the liver during inflammatory states with low levels of NO, produced consequent to activation of constitutive eNOS, shown to be protective and higher levels of NO, as a result of iNOS release, demonstrated to be cytotoxic (Rocky and Shah, 2004). Based on these reports, the midzonal distribution of hepatic lesions seen in this study is speculated to be the result of a combination of nitric oxide-mediated hypotension by iNOS and subsequent oxidative stress secondary to circulatory impairment, and parenchymal migration and proteolytic enzyme release by neutrophils, resulting in hepatocyte injury (Jaeschke and Smith, 1997; Lawson et al., 2000).

The occurrence of midzonal necrosis combined with the fact that calpain inhibition decreased liver injury and neutrophil infiltration led us to investigate the relationship between calpain inhibition and iNOS. One explanation for decreased hepatic neutrophil infiltration in the calpain 1 inhibitor group may simply be the result of decreased neutrophil numbers or decreased transmigration consequent to decreased neutrophil activation in the vasculature. However, hematology and flow cytometry results revealed that neutrophilia and activation of neutrophils as measured by CD11b was not different between the LPS and LPS plus calpain 1 inhibitor groups. Based on these results, it can be interpreted that the afforded protection by calpain inhibition is not solely due to decrease in peripheral neutrophil count or decreased neutrophil activation. Although the neutrophil activation was decreased in the aminoguanidine and heparin group compared to the calpain 1 inhibitor group, there was enough neutrophil/leukocyte activation to account for hepatic neutrophil infiltration as evidenced by hepatic neutrophil counts and periportal infiltrates on histology. In light of the presence of adequate neutrophil activation and periportal neutrophil infiltrate in all treated groups, mechanisms other than neutrophil and vascular endothelial cell activation seem to be playing a role in the observed protection.

Noteworthy in our studies is the reduced expression of hepatic iNOS in the calpain 1 inhibitor group which coincided with reduced hepatic pathology. Inhibition of calpain is reported to block production of iNOS through an NF-κB dependent mechanism (Lancaster et al., 2001; Constantin et al., 2004; Kabori, 2006). Asakura et al. (2005) also reported attenuation of liver and kidney pathology after administration of the iNOS inhibitor L-NIL in an LPS mediated rat model of DIC. Similarly Guler et al. (2004) have demonstrated protection from liver damage, as measured by histopathology, in iNOS deficient mice infected with mycobacteria after challenge with LPS.

Similar to inhibition of iNOS by calpain inhibition, iNOS inhibition by aminoguanidine also resulted in significantly reduced neutrophils and liver pathology. These results are consistent with reports of Suliburk et al. (2005) who noted improvement in LPS mediated liver injury in rats, as measured by ALT and AST levels, after administration of aminoguanidine 1 hour prior to administration of LPS. Although we do not have cause and effect evidence (for example using iNOS knock out mice), the association of decreased iNOS with decreased neutrophil infiltration and necrosis suggests that iNOS inhibition is the possible reason for the observed protection of calpain inhibition noted in our model.

Interestingly, heparin administration prior to LPS also reduced microscopic liver pathology and iNOS expression comparable to calpain 1 inhibition and aminoguanidine. Reduced microthrombi deposition was demonstrated in the lung and kidney of heparin administered rats in this study. Moulin et al. (1996) reported that heparin administration prior to LPS attenuates hepatocellular injury in rats as measured by ALT and histopathology in a thrombin-dependent manner. Also, in a model of liver transplantation, Vajdova et al. (2000) reported co-administration of heparin minimized endotoxin mediated preservation-reperfusion injury by prolonging coagulation and limiting microthrombi deposition. Microthrombi inhibition and preservation of microcirculation by heparin is reported to result in reduced hepatocyte oxidative stress that subsequently decreases iNOS expression in LPS mediated injury (Guler et al., 2004; Suliburk et al., 2005). Based on our findings and reports in the literature, it appears that the protective effect of heparin noted in the present study is mostly due to the result of decreased microthrombi deposition resulting in improved microperfusion. The heparin-mediated decrease in iNOS noted is likely a secondary effect from reduced oxidative stress.

Although the precise mechanism behind decreased iNOS and protection noted in the current study is not investigated, the following mechanism may have some underlying mechanistic influence. Attenuating NO dependent hypotension via iNOS inhibition in the present study is likely resulting in preserved liver microperfusion and reduced oxidative stress secondary to reduced tissue oxygenation (Figure 7). Indeed, using intravital microscopy, Horie and co-workers (2000) reported reduced sinusoidal blood flow and increased endothelial-leukocyte adhesion predominately in the mid-zonal region after LPS administration in rats.

Since CYP2E1 is known to be altered with endotoxin injection, we measured CYP2E1 protein to investigate if the observed protection mediated by calpain 1 inhibition is because of CYP2E1 inhibition. Based on our results, it appears that it is less likely involved. Finally emerging evidence suggests NO plays a role in epoxygenase dependent vascular responses via inhibition of CYP P450 enzymes (Udosen et al., 2003). Inhibition of NO production subsequent to LPS exposure might preserve epoxygenase dependent vasodilation that plays a role in maintaining microcirculation (Fleming, 2001). We employed CYP2E1 as a model for LPS mediated CYP downregulation as this cytochrome’s response to LPS has been well documented (Renton and Nicholson, 2000; Morgan, 2001). Administration of calpain 1 inhibitor preserved CYP2E1 expression which may suggest a role of iNOS inhibition and cytochrome expression in this model. There is debate in the literature concerning NO down-regulation of cytochromes and further mechanistic work is needed (Sewer et al., 1997, 1998; Sewer and Morgan, 1998; Vuppugalla and Mehvar, 2004).

In summary, inhibition of calpain attenuated LPS induced midzonal hepatic necrosis in our two-dose model of endotoxemia. Lowered hepatic neutrophil migration along with attenuated midzonal necrosis consequent to calpain inhibition accompanied reduced iNOS expression as compared to LPS only administered rats. Based on these findings and comparisons of the effect of calpain inhibition with those of aminoguanidine and heparin, which also decreased hepatic iNOS and neutrophil infiltration, we suggest that calpain inhibition reduces liver pathology via maintenance of hepatic microcirculation most likely in an iNOS dependent fashion.