SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Acute liver failure induced by hepatotoxic drugs results from rapid progression of injury. Substantial research has shown that timely liver regeneration can prevent progression of injury leading to a favorable prognosis. However, the mechanism by which compensatory regeneration prevents progression of injury is not known. We have recently reported that calpain released from necrotic hepatocytes mediates progression of liver injury even after the hepatotoxic drug is cleared from the body. By examining expression of calpastatin (CAST), an endogenous inhibitor of calpain in three liver cell division models known to be resistant to hepatotoxicity, we tested the hypothesis that increased CAST in the dividing hepatocytes affords resistance against progression of injury. Liver regeneration that follows CCl4-induced liver injury, 70% partial hepatectomy, and postnatal liver development were used. In all three models, CAST was upregulated in the dividing/newly divided hepatocytes and declined to normal levels with the cessation of cell proliferation. To test whether CAST overexpression confers resistance against hepatotoxicity, CAST was overexpressed in the livers of normal SW mice using adenovirus before challenging them with acetaminophen (APAP) overdose. These mice exhibited markedly attenuated progression of liver injury and 57% survival. Whereas APAP-bioactivating enzymes and covalent binding of the APAP-derived reactive metabolites remained unaffected, degradation of calpain specific target substrates such as fodrin was significantly reduced in these mice. In conclusion, CAST overexpression could be used as a therapeutic strategy to prevent progression of liver injury where liver regeneration is severely hampered. (HEPATOLOGY 2006;44:379–388.)

High mortality associated with overdose of hepatotoxic drugs is a significant clinical problem. Overdose of commonly used analgesic drug acetaminophen (APAP) accounts for approximately 50% of the cases of acute liver failure and is the leading cause of liver transplantation in the United States.1 Because shortage of donors limits liver transplantation, regeneration of the native liver is being considered as an alternative method for treatment of patients with a failing liver.2, 3 Efficient and timely liver regeneration can prevent hepatotoxic drug-induced acute liver failure in experimental animal models as well as in humans.4–8 Monitoring the markers of liver regeneration in drug overdose victims has been proposed as a prognostic predictor.5 Under a variety of circumstances in which liver regeneration is stimulated experimentally either by chemical injury (e.g., low dose of thioacetamide, APAP)6–10 or by surgical resection (two-thirds partial hepatectomy),11–14 the liver is known to exhibit resistance toward the higher doses of the hepatotoxicants. Liver cell division during postnatal development in newborns is known to impart resistance against toxicity.11, 15–17 Hepatocytes isolated from the regenerating liver exhibit resistance against hepatotoxicants in vitro.18, 19 Although evidence suggests that the dividing cells in the regenerating liver are resistant to hepatotoxicity, the exact mechanism has remained elusive.

Even after the offending hepatotoxic drug is eliminated from the body, injury continues until the liver heals itself via liver regeneration and tissue repair.20, 21 A major hurdle in understanding the mechanism of resistance of newly divided cells is in part the incomplete understanding of how injury initiated by a toxicant expands even after the offending drug has been eliminated. On exposure to low doses of hepatotoxicants, liver mounts a proportional compensatory regeneration to overcome toxicity.20, 21 However, high doses of hepatotoxic drugs inhibit liver regeneration, which results in rapid and accelerated progression of injury and liver failure.20, 21 Acute liver injury manifests in two independent stages: initiation or stage I injury and progression or stage II injury.20 Liver regeneration primarily curtails stage II injury.21 Whereas mechanisms of initiation of injury or stage I injury induced by hepatotoxic drugs are widely known to be related to the formation of reactive metabolites, and generation of free radicals, the mechanism of progression of stage II injury remains unknown.

In a recent study, we demonstrated that expansion of toxicant-initiated injury is mediated by the degradative enzymes released from dying hepatocytes, among which calpain appears to play a major role.22 A series of experiments affirmed that calpain spills out from the necrosed hepatocytes and is further activated in the extracellular high Ca2+ environment. Activated calpain degrades the plasma membrane and cytoskeletal proteins from the surrounding hepatocytes, spreading the injury in a self-perpetuating manner.22 Numerous reports have linked activation of calpain in the pathogenesis of various ailments such as neurodegenerative disorders, including Alzheimer's disease,23 epilepsy, cerebral ischemia/excitotoxicity,24 demyelination,25 cataracts,26 ischemic liver27 and ischemic renal injury,28 toxic renal injury,29 and muscular dystrophy.30 Studies have also shown that the overexpression of calpastatin (CAST), an endogenous inhibitor of calpain, is protective against these disorders where calpain activation is pathogenic.30–34

In the current study, we tested the hypothesis that the proliferating and newly proliferated hepatocytes overexpress CAST and thereby prevent the calpain-mediated onslaught and progression of injury. Calpastatin is a specific endogenous inhibitor of calpain34–36 and is known to inhibit both a Ca2+-activated as well as a proenzyme form of calpain.35–40 Calpastatin is mainly localized in the cytoplasm and plasma membrane,41 and it is also present in the nucleus during cell division.42 In the current study, calpastatin expression was assessed in the livers of: (1) rats challenged with a nonlethal dose of CCl4, which stimulates compensatory liver tissue repair, allowing them to overcome toxicity; (2) rats that underwent 70% partial hepatectomy (PH); and (3) 20-day-old newborn rat pups known to have active liver growth. Finally, to test the role of overexpression of CAST in affording resistance against hepatotoxic drug overdose, CAST overexpression was induced in normal mouse liver by adenoviral transfection, and the resistance to progression of liver injury was tested against APAP-induced liver failure and mortality.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Animals and Treatments

Male Sprague-Dawley (S-D) rats (250-275 g) and 20-day-old male rat pups were obtained from ULM Central Animal Facilities. Male Swiss Webster (SW) mice (25-29 g) were purchased from Harlan Sprague-Dawley (Indianapolis, IN) and were acclimatized in the same animal facility for 1 week before their use. Rats and mice had unlimited access to rodent chow (Harlan Teklad Rat chow No. 7001, Madison, WI) and water. All animal husbandry and handling were in accord with NIH guidelines and approved by our Institutional Animal Care and Use Committee.

Models of Liver Tissue Repair or Liver Regeneration

Stimulation of Tissue Repair After CCl4-Induced Liver Damage.

A nonlethal dose of CCl4 (2 mL CCl4 in 2 mL corn oil/kg, intraperitoneally) was administered to male S-D rats (250-275 g) to initiate moderate liver injury that stimulates tissue repair.43 Rats were killed at 0, 6, 12, 24, 48, 72, and 150 hours after CCl4 administration (n = 4 for each time point).

Two-thirds PH.

To obtain a regenerating liver model containing a large number of dividing/divided cells after surgical resection, 70% PH was carried out in male S-D rats (250-275 g) under light ether anesthesia as previously described.44 Sham operations (SH) were conducted by making a midventral incision; the liver was exteriorized, but was inserted back in its place before suturing the incision. Rats were killed at 0 hours and 1, 2, 4, and 7 days after PH and SH (n = 3/each time point).

Postnatal Development of the Livers.

The 20-day-old male rat pups (NB) known to exhibit extensive liver growth during this early development14–16 were chosen for this study.

In all three models, after euthanasia blood was collected at indicated time points from the inferior vena cava for biochemical analysis, and the livers were excised for further analysis. A portion of the median liver lobe was fixed in 10% neutral buffered formalin and processed for paraffin embedding. The remaining portion of liver was quickly frozen in liquid N2 and was stored at −80°C until further analysis. Paraffin-embedded liver sections (4 μm) were cut for histopathological studies.

CAST and Proliferating Cell Nuclear Antigen Immunohistochemistry

Paraffin embedded liver sections obtained from CCl4-treated, PH, or NB rats were immunohistochemically stained independently for CAST and proliferating cell nuclear antigen (PCNA). Antigen retrieval was not required for CAST staining, while 1% zinc sulfate was used for PCNA antigen retrieval. Blocking was achieved using 0.5% casein for 20 minutes. The sections were further treated either with a goat polyclonal anti-calpastatin antibody (1:200, Santa Cruz Biotech, Santa Cruz, CA) for 3 hours or mouse monoclonal anti-PCNA antibody (1:5,000, PC.10, Dako Corporation, Carpinteria, CA) for 1 hour at room temperature. The appropriate biotinylated secondary antibody was linked to the primary antibody. The sections were labeled with strepatavidin-conjugated peroxidase followed by the substrate 1,2 diaminobenzidene, which gave a brown reaction product. The sections were counterstained with Meyers modified hematoxylin.

Assessment of Liver Injury

Liver injury was assessed by measuring plasma alanine aminotransferase activity (ALT; EC 2.6.1.2.) using reagent TR71121 (Thermo DMA, Waltham, MA). Paraffin-embedded liver sections (4 μm) were stained with hematoxylin-eosin for light microscopic histopathological examination.

Real-time Polymerase Chain Reaction of CAST

RNA was extracted from rat livers (n = 3) using the RNeasy Mini Columns (Qiagen, Valencia, CA) following the manufacturer's protocol. Taqman Two–step RTPCR system Assay on Demand® was used to assess the CAST mRNA expression (Applied Biosystems Inc., Foster City, CA). In the first step, cDNA was synthesized using the high-capacity cDNA archive kit. Assay-on-Demand gene expression system was used for polymerase chain reaction (PCR). Rat CAST primer used was TCACAATCAAGTGAGCCACCTGTGA (GenBank # NM053295). The primer is tagged with FAM (as the reporter dye) and a nonfluorescent quencher. The 18s primer was used as the internal control. Fifty nanograms of cDNA was used for each reaction, ran in duplicates (reaction conditions: 10 minutes at 95°C for initial set up, 15 seconds at 95°C for denaturation, and 1 minute at 60°C for annealing, total cycles = 40). The data were analyzed based on the comparative Ct (threshold cycle) method. The data were normalized to 18s expression (n = 3) and plotted as a fold difference of the mean Ct values (2 dCt) compared with 0 hour controls (n = 3).

CAST and Fodrin Western Blot

Liver cell lysates were prepared at 4°C using cell lysis buffer [1% Triton X-100, 150 mmol/L NaCl, 10 mmol/L Tris (pH 7.4), 1 mmol/L EDTA, 1 mmol/L EGTA, 2 mmol/L Na vanadate, 0.2 mL 200 mmol/L phenylmethylsulfonylfluoride, 1 mmol/L HEPES (pH 7.6), 1 μg/mL leupeptin, and 1 μg/mL aprotinin]. Polyacrylamide gels (7.5% for calpastatin and 6% for fodrin) were used for separation of proteins. After transferring the proteins for 1 hour on nitrocellulose membranes, the membranes were incubated with either anti-calpastatin antibody (Santa Cruz Biotech, CA, 1:100, overnight at 4°C) or anti-fodrin antibody (Chemicon, 1:2500, 3 hours at room temperature) with 1.5% non-fat dry milk in Tris-buffered saline with 0.1% Tween. Subsequently, the nitrocellulose membrane was incubated with the appropriate horseradish peroxidase–-linked secondary IgG with 1.5% non-fat dry milk in Tris-buffered saline with 0.1% Tween buffer for 2 hours. For visualization, the ECL kit from Pierce Biotechnology, Inc. (Rockford, IL) was used.

Preparation of Microsomes

Liver microsomes were prepared using the method described previously.45 Microsomal pellets were quick frozen and stored at −80°C for later use.

CYP2E1 and CYP1A1/1A2 Protein Estimation by Western Blot

Liver microsomal protein (20 μg) was separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (12% gel) and transferred to nitrocellulose membranes (Bio-Rad, Hercules, CA). Nitrocellulose membranes were incubated with either goat polyclonal anti-CYP2E1 or anti-CYP1A2 (known to cross-react to both CYP1A1 and 1A2) antibody (1:1,000, Gentest, Woburn, MA) for 3 hours. The blots were further incubated with horseradish peroxidase–linked anti-goat antibody (Sigma Chemicals, St. Louis, MO) for 1 hour and visualized using ECL reagent (Pierce Biochemicals, Rockford, IL).

Analysis of CYP2E1 Enzyme Activity

Microsomal CYP2E1-dependent hydroxylation of p-nitrophenol to p-nitrocatechol was used as a relatively specific marker of CYP2E1 enzyme activity according to the method described by Koop45 and as used previously.22

Covalent Binding of 14C-APAP–Derived Radiolabel to Liver Macromolecules

14C-APAP (Specific activity 5.3 mCi/mmol, Sigma Chemicals, St. Louis, MO) (600 mg/kg, approximately 4 μCi/mouse, in 20 mL/kg 0.45% NaCl, pH 8.0) was administered intraperitoneally to male SW mice (25-30 g) with and without pretreatment with Ad/CAST. Covalent binding was measured at 2 and 4 hours after APAP administration using the method previously described.22

Recombinant Adenovirus Construction

AdEasy™ vector system (Qbiogene Inc., Carlsbad, CA) originally developed by He and associates46 was used to construct recombinant adenovirus for calpastatin (Ad/CAST). Full-length mouse calpastatin cDNA (a gift of Dr. Masatoshi Maki, Nagoya University, Japan) was first cloned into a transfer vector pAdTrack-CMV at EcoRV site using T4 DNA ligase (Promega, Madison, WI) and transformation was carried out in Escherichia coli strain JM109. The resulting plasmid was linearized with Pme I and co-transformed into E. coli strain BJ5183 with pAdEasy-1 by electroporation. The recombinant adenoviral construct was cleaved with Pac I (New England Biolabs, Beverly, MA) to expose its inverted terminal repeats and transfected into QBI-293A cells with LipofectAMINE2000 (Life Technologies, Inc., Rockville, MD) to produce viral particles. The virus was purified by CsCl banding and stored at −80°C. The control LacZ virus was a gift of Dr. W. El-Deiry (University of Pennsylvania, Philadelphia) and has been described previously.47 SW mice (28-34 g) were intravenously injected via tail vein in a volume of 100 μl with 0.5 × 1011 viral particles of Ad/CAST or Ad/LacZ.

Statistical Analysis

Group comparisons were performed using the independent t test.

A one-way analysis of variance (ANOVA) was used to determine statistically significant differences among more than two distributions or sample groups. Statistical analyses were made using SPSS 10.0 software (SPSS Inc., Chicago, IL). Statistical significance was set at P less than .05.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

CAST Expression After Induction of Tissue Repair by CCl4 Treatment

CAST Immunohistochemistry.

CAST expression was examined by immunohistochemistry during a time course after administration of CCl4 (2 mL/kg). CCl4 treatment produced centrilobular liver injury stimulating tissue repair in the perinecrotic hepatocytes. Tissue repair was apparent from 24 hours onwards, reaching a peak at 48 and 72 hours, and regressed thereafter (Fig. 1A). In the 0 hour control livers, where most of the hepatocytes are in the G0 phase [Fig. 1C(e)], CAST expression was very minimal, and its distribution was uniform, indicated by light brown cytoplasmic staining [Fig. 1C(a)]. After CCl4 administration, significant CAST expression was observed from as early as 6 hours only in the perinecrotic hepatocytes [Fig. 1C(b)], although no cell division was apparent at this time point [Fig. 1C(f)]. From 6 to 48 hours after CCl4 treatment, along with increased cell division [Fig. 1C(g)], intense CAST staining was observed (up to 8-9 cell layers) surrounding the necrotic area [Fig. 1C(c)]. By 72 hours, the entire liver lobule showed intense CAST staining [Fig. 1C(d)] coincident with extensive cell division [Fig. 1C(h)]. Calpastatin staining was evident in the cytoplasm, plasma membrane, and nucleus of the hepatocytes. At 150 hours, both cell division (Fig. 1A) and calpastatin expression returned to normal levels (data not shown). These findings indicated that the hepatocytes preparing for cell division, hepatocytes undergoing cell division, and the newly divided hepatocytes overexpress CAST. CAST protein expression data corroborated further by Western blot analysis showed increased CAST at 24, 48, and 72 hours after CCl4 administration that correlated well with liver regeneration (Fig. 1D).

thumbnail image

Figure 1. Liver cell division and calpastatin expression in rats undergoing tissue repair. (A) The number of cells in S-phase (1,000 cells/section/rat) were counted as a marker of cell division over a time course of 0 to 150 hours after administration of a nonlethal dose of CCl4 (2 mL/kg) in PCNA-stained liver sections. * = Values significantly higher than 0-hour control (n = 3) P < .05. (B) Real-time RT-PCR analysis of CAST mRNA in the liver over a time course after CCl4 (2 mL/kg, intraperitoneally) administration. The data were normalized by 18s internal control (n = 3) and then are plotted as fold difference of the mean Ct values (2 dCt) compared with 0-hour controls. (C) Representative photomicrographs of paraffin-embedded liver sections showing CAST immunohistochemical staining [(a) 0 hours, (b) 6 hours, (c) 48 hours, and (d) 72 hours] and corresponding PCNA immunohistochemical staining [(e) 0 hours, (f) 6 hours, (g) 48 hours, and (h) 72 hours] of livers obtained from rats treated with CCl4 (2 mL/kg). Brown color indicates positive staining for CAST (upper panel) and PCNA (lower panel), respectively. (D) Western blot analysis of CAST after CCl4 administration. Data represent pooled liver cell lysates from three separate rats for each time point.

Download figure to PowerPoint

Real-time PCR of CAST.

These observations were corroborated by CAST mRNA expression analysis by real-time PCR. Increased CAST mRNA was observed at 12 (approximately a 3.6-fold increase), 24 (approximately a 3-fold increase), and 48 hours (approximately a 5-fold increase) after CCl4 treatment (Fig. 1B), preceding the increased expression of CAST protein assessed by immunohistochemistry. CAST mRNA levels declined to normal levels at 72 hours, because both cell proliferation (Fig. 1A) and CAST protein expression declined after 72 hours (Fig. 1B).

CAST Expression After PH.

To examine whether PH-induced cell division/liver regeneration is accompanied by transcriptional upregulation of CAST, rat livers were examined on days 1, 2, 4, and 7 after partial hepatectomy by real-time PCR (Fig. 2A). CAST mRNA significantly increased with cell division from days 1 to 4 (approximately a 297- to 248-fold increase) after PH and declined rapidly by day 7 (approximately a 23-fold increase) as cell division regressed (Fig. 2A). CAST protein expression detected by immunohistochemistry on day 4 after PH when the cell division is robust [Fig. 2C(e)] was intense in the dividing and newly divided hepatocytes [Fig. 2C(b)], whereas on day 7 after PH, cell division and CAST staining regressed substantially [Fig. 2C(f)] and Fig. 2C(c)]. CAST staining did not show any zone specificity in PH rats. It was rather diffuse at both 4 and 7 days after PH. These results are concordant with CAST mRNA data (Fig. 1B). CAST staining was observed in the cytoplasm, plasma membrane, and nucleus of the hepatocytes. CAST protein expression data further corroborated by Western blot analysis showed increased CAST on day 4 and declined by day 7 after PH (Fig. 2B). Sham operated controls of 4 and 7 days after SH did not show any alteration in CAST expression from 0-hour controls (data not shown).

thumbnail image

Figure 2. Liver cell division and calpastatin expression in the partially hepatectomized rats. (A) Real-time RT-PCR analysis of calpastatin (CAST) mRNA in the liver over a time course after PH. The data were normalized by 18s internal control (n = 3) and then are plotted as fold difference of the mean Ct values (2 dCt) compared with 0-hour controls. (B) Western blot analysis of CAST in the liver cell lysates after PH over a time course. (C) Representative photomicrographs of paraffin-embedded liver sections obtained from partially hepatectomized rats stained for CAST [(a) 0 h, (b) 4 d, and (c) 7 d] and corresponding PCNA immunohistochemical staining [(d) 0 h, (e) 4 d, and (f) 7 d]. Brown color indicates positive staining either for CAST (upper panel) or PCNA (lower panel), respectively.

Download figure to PowerPoint

CAST Expression in Livers From Newborn Rats.

CAST mRNA expression in 20-day-old rat pups livers examined by real-time RT-PCR was approximately 29.7-fold higher in the NB livers than that of adults (Fig. 3A). Immunohistochemical analysis showed increased CAST staining in the NB livers [Fig. 3C(b)] along with extensive cell division [Fig. 3C(d)] compared with their 60-day adult controls [Fig. 3C(a) and 3C(c), respectively]. CAST staining was observed primarily in the cytoplasm and plasma membrane, and to some extent in the nucleus of the hepatocytes. Western blot analysis showed remarkably higher expression of CAST protein in the actively growing NB rat livers (Fig. 3B).

thumbnail image

Figure 3. Liver cell division and calpastatin expression in the newborn rats. (A) Real-time RT-PCR analysis of CAST mRNA in the liver of 60-day-old adults and 20-day-old NB rats (n = 3 each group). The data were normalized by 18s internal control (n = 3) and then are plotted as fold difference of the mean Ct values (2 dCt) compared with 0-hour controls (n = 3). (B) Western blot analysis of CAST in the liver cell lysates of 60-day-old adults and 20-day-old NB. (C) Representative photomicrographs of paraffin-embedded liver sections obtained from adult and NB rats stained for CAST [(a) Adult (b) NB] and corresponding PCNA immunohistochemical staining [(c) Adult (d) NB]. Brown color indicates positive staining for CAST (upper panel) and PCNA (lower panel), respectively.

Download figure to PowerPoint

CAST Overexpression Induced by Ad/CAST in the Normal Livers of SW Mice.

In the preliminary studies, CAST expression was assessed on days 2, 4, and 7 after injection of adenovirus (Ad/CAST). Maximum CAST expression was evident on day 4 after Ad/CAST injection (Fig. 4A). This was also confirmed by CAST immunohistochemistry on the paraffin-embedded liver sections (Fig. 1C). We did not observe any apparent histological or biochemical damage [elevation in alanine aminotransferase (ALT)/aspartate aminotransferase (AST)] after Ad/LacZ or Ad/CAST treatment at 4 days after injection (data not shown). Therefore, we do not think that the CAST expression is associated with nonspecific liver damage after adenoviral injection.

thumbnail image

Figure 4. Calpastatin overexpression in Ad/CAST-treated mice. (A) Western blot analysis of CAST and loading control GAPDH at 4 days in the liver cell lysates obtained from SW mice treated with either Ad/CAST or Ad/LacZ (n = 3). (B) Densitometric analysis of the CAST Western blot. (C) Representative photomicrographs of paraffin-embedded liver sections obtained 4 days after Ad/CAST or Ad/LacZ treatment [(a) Ad/LacZ treated, (b) Ad/CAST treated].

Download figure to PowerPoint

Survival Study Using Ad/CAST-Pretreated Mice.

To assess whether CAST overexpression protects against APAP-induced acute liver failure, male SW mice pretreated with either Ad/CAST or Ad/LacZ were challenged with a lethal dose of APAP (600 mg/kg, intraperitoneally) at 90 hours after the virus injection. These mice were observed for 14 days for survival/mortality. Fifty-seven percent of the mice survived that received Ad/CAST, compared with 83% mice that died receiving APAP alone or Ad/LacZ + APAP (Fig. 5). The deaths occurred between 10 and 12 hours after APAP administration. Thereafter, all the remaining mice survived.

thumbnail image

Figure 5. Kaplan-Meyer survival plot after APAP administration. Male SW mice (25-29 g) were divided into three groups. Group I received PBS (100 μL), groups II received Ad/Calpastatin (CAST, 0.5 × 1011 v. p. in 100 μL PBS) while group III received Ad/LacZ (0.5 × 1011 v. p. in 100 μL PBS) by tail vein injection. All three groups were administered a single dose of acetaminophen (APAP, 600 mg/kg, intraperitoneally, in 0.45% NaCl, pH 8) 6 hours before the end of the 4th day after the respective pretreatments. All mice were observed six times on the first day and twice daily thereafter for 14 days. All deaths occurred between 10 and 12 hours after APAP administration. Thereafter, all remaining mice survived.

Download figure to PowerPoint

Assessment of Liver Injury in Ad/CAST + APAP–Treated Mice.

To assess the protection afforded by CAST overexpression against APAP toxicity in terms of liver injury; injury was assessed at 4 and 8 hours after APAP treatment. Plasma ALT (Fig. 6A) elevations were approximately twofold lower in Ad/CAST-pretreated mice as compared with their Ad/LacZ–treated controls. APAP administration induced centrilobular hepatic necrosis in both groups. However, the necrotic area as well as the number of necrotic foci appeared lower in the Ad/CAST–pretreated mice (Fig. 6B). The PCNA immunostaining analysis did not reveal any apparent staining in both the groups at either time point with Ad/CAST or Ad/LacZ receiving APAP (data not shown).

thumbnail image

Figure 6. Liver injury assessment after APAP administration with pretreatment with either Ad/CAST or Ad/LacZ. (A) Plasma ALT measured as an index of liver injury after the administration of APAP (600 mg/kg, intraperitoneally) with pretreatment with either Ad/CAST or Ad/LacZ. Zero-hour control values = 42.5 + 12.8. *Significantly higher than the zero-hour values of Ad/CAST or Ad/LacZ. #Values significantly lower than the group receiving Ad/LacZ + APAP alone at the same time point (n = 5). P < .05. (B) Representative photomicrographs of the paraffin-embedded liver sections stained with H-E after the administration of APAP (600 mg/kg, intraperitoneally) with pretreatment with either Ad/CAST [(b) 4 hours after APAP, and (d) 8 hours after APAP], or Ad/LacZ [(a) 4 hours after APAP, and (c) 8 hours after APAP]. Magnification = 200×. Arrows indicate necrotic area, CV represents central vein.

Download figure to PowerPoint

Assessment of CYP2E1 and CYP1A1/1A2 Protein, CYP2E1 Activity, and Covalent Binding of 14C-APAP–Derived Radiolabel to Liver After Ad/CAST Treatment.

Hepatomicrosomal CYP2E1 and CYP1A1/1A2 proteins remained unaltered after Ad/CAST treatment (Fig. 7A,C). P-nitrophenol hydrolxylase enzyme activity measured as a representative of CYP2E1-catalyzed reactions was also identical with and without Ad/CAST treatment at day 4 (Fig. 7B). A more direct measure of bioactivation of APAP estimated as the covalent binding of 14C-APAP–derived radiolabel to liver proteins was also not affected by Ad/CAST pretreatment (Fig. 7D). These findings suggest that Ad/CAST pretreatment does not inhibit APAP bioactivation and stage I injury.

thumbnail image

Figure 7. Assessment of acetaminophen bioactivating hepatic microsomal CYP2E1 and CYP1A1/1A2 at 0 and 4 days after treatment with Ad/CAST. (A) Representative Western blot and densitometric analysis of hepatic microsomal CYP2E1 protein (n = 3). (B) PNPH activity of hepatic microsomal CYP2E1 (n = 3). (C) Representative Western blot and densitometric analysis of hepatic microsomal CYP1A1/1A2 (n = 3). (D) Covalent binding of 14C-APAP–derived radiolabel to liver proteins (n = 3). P < .05.

Download figure to PowerPoint

Assessment of Breakdown of α-Fodrin.

To examine whether the lower injury observed as early as 4 hours after APAP administration in Ad/CAST–pretreated mice is attributable to lower progression of injury, breakdown of α-fodrin was assessed in the liver samples as a marker of calpain-mediated progression of injury. The 0-hour controls exhibit negligible calpain-mediated breakdown of α-fodrin (150 kDa) (Fig. 8). In Ad/LacZ + APAP–treated mice, significant breakdown of α-fodrin was apparent (Fig. 8, lanes 5 and 6). In contrast, mice pretreated with Ad/CAST before APAP administration experienced negligible breakdown of α-fodrin (Fig. 8, lanes 3 and 4). It should be noted that caspase-specific breakdown (120 kDa) was identical with and without Ad/CAST pretreatment. These findings indicate that Ad/CAST pretreatment–mediated overexpression of CAST prevents calpain-specific breakdown of its target substrates.

thumbnail image

Figure 8. Western blot analysis of α-fodrin, a protein substrate of calpain and its product of proteolysis in the liver. The 240-kDa band represents intact α-fodrin, whereas the 150-kDa band represents calpain-specific breakdown of α-fodrin. Caspase-specific breakdown is represented by the 120-kDa band (n = 3).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Our previous work indicated that on infliction of hepatic injury by hepatotoxic drugs, degradative enzymes spilt out from the necrosed hepatocytes can degrade the perinecrotic tissue that may mediate progressive damage and liver failure unless obtunded by liver regeneration.21 That same study revealed that calpain, a cytoplasmic cysteine protease, is the principal mediator of progression of injury after leakage around the neighboring hepatocytes. An important question was why calpain does not degrade the neighboring hepatocytes when they are in cell division cycle or when they divide anew. When the liver regeneration is impaired, as in hepatotoxicant overdose, therapeutic administration of either cell-permeable or impermeable calpain inhibitor curtailed the progression of injury and rescued the animals from acute liver failure and death.22, 48 We hypothesized that dividing hepatocytes escape calpain-mediated cell death by overexpressing its endogenous inhibitor, calpastatin.

In the current study, we observed a remarkable increase in CAST expression in all three models of liver cell division that were tested. Interestingly, CAST expression pattern coincided not only with the gradual increase/decrease in cell division but also with the time course of resistance/loss of resistance against hepatotoxicity in these models. In the chemical injury-regeneration model, CAST expression gradually increased from 6 hours and peaked at 72 hours after CCl4 administration. Although PCNA staining did not capture cell division at 6 hours, significant CAST expression at 6 hours and only in the perinecrotic area indicated that the cells preparing for division, that is, the cells between G0 to G1 phase were overexpressing CAST. At 72 hours, when CAST expression and cell division were maximal, there was a rapid and precipitous fall in the biochemical and histological markers of hepatic injury. These data strongly indicate that upregulation of CAST overexpression and recovery from progression of injury are closely and inversely related.22, 48

Partial hepatectomy, the second model of liver cell division studied in the current investigation, is known to protect the experimental animals from hepatotoxicity.11–13 However, the time of exposure of PH animals to the toxic insult is pivotal in determining whether the final outcome is favorable.12, 13 If the PH rats are exposed to model hepatotoxicant CCl4 at the time of peak liver regeneration (from 2 to 4 days after PH), progression of injury is substantially lower, leading to recovery from liver injury and animal survival. In contrast, if the rats are exposed to the same dose of CCl4, 7 days post-PH when cell division has tapered off, injury progresses rapidly, liver fails, and the rats die, signifying complete loss of protection.12, 13 CAST expression data obtained in the current study coincide very well with these survival versus death observations reported earlier.11 These findings are consistent with the notion that CAST overexpression may be the underlying mechanism of resiliency of the dividing/newly divided hepatocytes.

Unlike the other two models, where cell division is induced experimentally, in the third model of newborn animals the cell division is active as a physiological phenomenon; however, it is closely associated with resiliency against hepatotoxicants. At day 20 after birth, the rat liver reaches approximately 50% of the weight of an adult rat liver (60 days old) and during the intervening time the liver cell division is continuing.14, 15 Whereas 20-day-old newborns are completely resilient to a lethal combination of chlordecone + CCl4, 60-day-old adult rats exhibit 100% mortality.15, 16 Livers of 20-day-old rat pups examined in the current study showed remarkably high CAST expression as compared with the livers of 60-day-old adults (same controls used for the other two models of cell division investigated in the current study). These findings endorse the hypothesis that higher CAST in the dividing cells is essential to prevent progression of injury and acute liver failure on exposure to overdose of hepatotoxic drugs. A similar finding has been reported by Wingrave et al.49 in spinal cord injury. Twenty-one-day-old rats show resiliency against spinal cord injury due to upregulation of spinal cord calpastatin.49 Collectively, these observations suggest that calpastatin overexpression may be a generalized protective mechanism against destruction by proteases such as calpain in the newborns.

To further test our hypothesis that the new cells are resilient to progression of injury primarily due to the overexpression of CAST, CAST overexpression was induced in adult mice in vivo by employing recombinant adenoviral (Ad/CAST) transfection. When challenged with the lethal dose of APAP, known to inhibit liver regeneration, progression of injury in the mice overexpressing CAST was mitigated by approximately 50%, which led to 57% survival instead of the 83% mortality observed in the mice not overexpressing CAST. Protection afforded by Ad/CAST administration could be either by inhibition of APAP bioactivation (stage I injury) or by inhibition of calpain-mediated cell death (stage II injury). The former possibility was ruled out because APAP-bioactivating enzymes, CYP2E1 and CYP1A1/1A2, remained unchanged. Moreover, covalent binding of 14C-APAP–derived radiolabel, a direct measure of bioactivation, remained unaffected by Ad/CAST transfection. These findings clearly indicate that protection against the lethal dose of APAP cannot be attributed to compromised bioactivation of APAP.

To ascertain the latter possibility of protection by Ad/CAST transfection by inhibition of calpain-mediated progression of injury, breakdown of α-fodrin, a marker of calpain-mediated cellular degradation, was assessed. This study revealed that the decrease in progression of liver injury in the mice overexpressing calpastatin was associated with decreased degradation of α-fodrin. The protection offered by CAST overexpression may be due to inhibition of intracellular calpain activation after APAP administration. However, in the previous study, we demonstrated that a cell-impermeable calpain inhibitor, E64, affords equal protection from progression of CCl4-induced injury as that of cell permeable inhibitor, confirming that the extracellular calpain after its release from the necrosed hepatocytes mediates progression of injury.22, 48 These data support the notion that overexpression of calpastatin by Ad/CAST prevents cellular degradation mediated by calpain, released from hepatocytes necrosed due to APAP-initiated injury. Finally, the question that needs to be addressed is the mechanism by which CAST overexpressed in new cells prevents calpain-mediated cell death. In all three models we studied, CAST was found localized to the plasma membrane and cytoplasm of hepatocytes. This strategic location would be very effective in inhibiting the degradative action of extracellular calpain. Although nuclear localization of CAST was found to a certain extent, it may not be critical in protection against degradative action of extracellular calpain because nuclear CAST staining disappeared with time after CCl4 administration, and moreover, CAST was primarily localized to the cytoplasm and membranes of Ad/CAST transfected mice livers. The upregulated CAST may help in preventing activation of intracellular calpain by subsequent entry of extracellular Ca2+ inside the cells on initiation of extracellular calpain-mediated injury. However, more studies are needed to delineate the exact mechanism of this interaction.

Regardless of the mechanism of protection by upregulation of calpastatin, the finding that overexpression of CAST may rescue victims from APAP-induced liver failure is significant and could be used as a therapeutic strategy in averting an otherwise hopeless prognosis under the circumstances of APAP overdose, where liver regeneration is inhibited. This finding also suggests that strategies to promote liver regeneration after APAP overdose may be a promising approach worthy of investigation.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The authors thank Dr. Masatoshi Maki, Nagoya University, Nagoya, Japan for the gift of the mouse calpastatin cDNA. The authors also thank Dr. W. El-Deiry, University of Pennsylvania, Philadelphia for the gift of LacZ virus.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References