SEARCH

SEARCH BY CITATION

Abstract

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

Acetaminophen (N-acetyl-p-aminophenol [APAP]) is one of the leading causes of acute liver failure, and APAP hepatotoxicity is associated with coagulopathy in humans. We tested the hypothesis that activation of the coagulation system and downstream protease-activated receptor (PAR)-1 signaling contribute to APAP-induced liver injury. Fasted C57BL/J6 mice were treated with either saline or APAP (400 mg/kg intraperitoneally) and were euthanized 0.5-24 hours later. Hepatotoxicity and coagulation system activation occurred by 2 hours after administration of APAP. Treatment with APAP also caused a rapid and transient increase in liver procoagulant activity. In addition, significant deposition of fibrin was observed in the liver by 2 hours, and the concentration of plasminogen activator inhibitor-1 in plasma increased between 2 and 6 hours. Pretreatment with heparin attenuated the APAP-induced activation of the coagulation system and hepatocellular injury and diminished hepatic fibrin deposition at 6 hours. Loss of hepatocellular glutathione was similar in APAP-treated mice pretreated with saline or heparin, suggesting that heparin did not diminish bioactivation of APAP. In mice deficient in tissue factor, the principal cellular activator of coagulation, APAP-induced liver injury, activation of coagulation, and hepatic fibrin deposition were reduced at 6 hours. Formation of the tissue factor–factor VIIa complex leads to the generation of thrombin that can activate cells through cleavage of PAR-1. Mice lacking PAR-1 developed less injury and hepatic fibrin deposits at 6 hours in response to APAP than control mice. Conclusion: Activation of the coagulation system and PAR-1 signaling contribute significantly to APAP-induced liver injury. (HEPATOLOGY 2007.)

Acetaminophen (N-acetyl-p-aminophenol [APAP]) is the leading cause of drug-induced hepatic failure in humans. Early results in animal models revealed that APAP is bioactivated to a reactive metabolite that is responsible for the hepatotoxicity.1, 2 Many subsequent studies have identified numerous factors that appear to contribute to APAP-induced liver injury, including mitochondrial alterations, reactive oxygen and nitrogen species, and cytokines such as tumor necrosis factor-α.3–8 Despite extensive study, the factors and mechanisms involved in the initiation and progression of hepatocellular lesions during APAP hepatotoxicity are not fully understood.

Disturbances in the hemostatic system are well documented in human patients with APAP hepatotoxicity. During hemostasis, formation and lysis of clots is regulated by the balance among coagulant, anticoagulant and fibrinolytic pathways.9 The coagulation system is activated by tissue factor (TF), and this culminates in the generation of thrombin and formation of insoluble fibrin clots. Dissolution of fibrin clots is mediated by plasmin and is inhibited by plasminogen activator inhibitor-1 (PAI-1). In people who develop APAP-induced liver injury, prothrombin time increases, and this change correlates with the severity of toxicity.10–12 Moreover, the concentrations of several coagulation factors are decreased in APAP-poisoned patients,13 an effect that could be interpreted to be a consequence of decreased production of coagulation factors by the injured liver. An alternative interpretation, however, is that the decrease in coagulation factors reflects their consumption during activation of the coagulation system. Indeed, the concentration of plasma TF was elevated in APAP-poisoned patients, and this was associated with an increase in the plasma concentration of thrombin–antithrombin (TAT) complexes.13 All of these changes are consistent with activation of the coagulation system during APAP hepatotoxicity.

In APAP-induced liver injury in people and animal models, sinusoidal endothelial cells are damaged,14, 15 which would favor activation of the coagulation system. Moreover, production of proinflammatory cytokines, such as tumor necrosis factor-α, occurs both in animal models16 and during human APAP poisoning.13 Tumor necrosis factor-α induces the expression of TF and PAI-1 in cultured endothelial cells, both of which would promote thrombosis.

Activation of the coagulation cascade generates a series of serine proteases, including thrombin. Thrombin elicits diverse biological effects through intracellular signaling induced by cleavage of protease-activated receptor-1 (PAR-1).17 This receptor is expressed on a variety of liver cells.18, 19

Although both human and animal studies suggest that activation of the coagulation system occurs during APAP hepatotoxicity, its role in the pathogenesis is unclear. Using both pharmacological and genetic approaches, we tested the hypothesis that activation of the coagulation system and downstream PAR-1 signaling contribute to APAP-induced liver injury.

Materials and Methods

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

Materials.

Heparin sodium salt with an activity of 180 U/mg (H-3149) and acetaminophen (A7302) were purchased from Sigma Chemical Co. (St. Louis, MO). Infinity ALT reagent was obtained from Thermo Electron Corp. (Louisville, CO), and the PAI-1 ELISA was obtained from Molecular Innovations (Southfield, MI). The kit to measure plasma fibrinogen concentration (B4233) and the ELISA for determination of TAT dimer concentration (OWMG15) were obtained from Dade-Behring Inc. (Deerfield, IL). Immunopure Peroxidase Supressor was purchased from Pierce Biotechnology, Inc. (Rockford, IL). For fibrin staining, the primary antibody, Dakocytomation rabbit anti-human/mouse fibrinogen, was obtained from Dako North America (Carpinteria, CA), and the secondary antibody, donkey anti-rabbit IgG conjugated to Alexa, was purchased from Molecular Probes/Invitrogen (Carlsbad, CA). EnVision+ System-HRP kit and Protein Block were obtained from Dako North America. Unless stated otherwise, reagents and chemicals were purchased from Sigma Chemical Company.

Animals.

C57BL/J6 mice (8 weeks old) or mice backcrossed at least 6 generations onto a C57BL/J6 background weighing approximately 20 g were used for these studies. Male mice were used for all studies except the TF studies. Animals were fed a standard chow (Rodent chow/Tek 8640, Harlan Teklad, Madison, WI) and allowed access to water ad libitum. They were allowed to acclimate for 1 week in a 12-hour light/dark cycle prior to use. All procedures on animals were performed according to the guidelines of the American Association for Laboratory Animal Science, the University Laboratory Animal Research Unit at Michigan State University and The Scripps Research Institute Animal Care and Use Committee.

To evaluate the role of TF in APAP-induced hepatotoxicity, mice deficient in TF were used. Disruption of the murine tissue factor (mTF) gene results in embryonic lethality due to a defect in the yolk sac vasculature.20 Expression of a low level of human tissue factor (hTF) (1% of wild-type levels) by an hTF minigene rescues the embryonic lethality of mTF−/− mice. The generation of these mTF−/−/hTF+ mice—or “low-TF” mice—was described previously.21 Female low-TF mice and heterozygous littermate controls (mTF+/−/hTF+) were backcrossed 6 times onto a C57BL/J6 background. Low-TF and littermate control mice between 8-16 weeks of age were used for all studies.

To evaluate the role of PAR-1 in APAP-induced hepatotoxicity, mice deficient in PAR-1 were used. PAR-1−/− mice were provided by Dr. S. Coughlin (University of California, San Francisco, CA). Male PAR-1−/− and PAR-1+/+ littermate controls were backcrossed 7 generations onto a C57Bl/J6 background and were used at 6-8 weeks of age.

Experimental Protocol.

Mice fasted for 15 hours were given 400 mg/kg APAP or its saline vehicle via intraperitoneal injection, and food was then returned. For time course studies, the mice were anesthetized with isoflurane 30 minutes and 1.0, 1.5, 2.0, and 6 hours after APAP administration, and, in a separate study, 24 hours after APAP administration. For all other studies, they were euthanized at 6 hours. Blood was drawn from the vena cava into a syringe and transferred to a tube containing sodium citrate (final concentration, 0.76%) for preparation of plasma. The left medial lobe of the liver was dissected and flash-frozen in liquid nitrogen for glutathione measurement; the left lateral lobe was fixed in 10% neutral buffered formalin for evaluation of histopathology. The right medial lobe was covered in Tissue-Tek O.C.T. (Optimal Cutting Temperature) Embedding Medium and frozen in liquid nitrogen–chilled isopentane for immunohistochemical analysis. For studies examining the effect of heparin, mice were treated with heparin (3,000 U/kg) or sterile saline intraperitoneally 1 hour before administration of APAP. Food was returned at the time of heparin administration.

Assessment of Hepatotoxicity and Bioactivation.

Hepatocellular injury was estimated as an increase in plasma ALT activity and from histopathology. ALT activity was determined spectrophotometrically. Formalin-fixed liver samples were embedded in paraffin, sectioned, stained with hematoxylin-eosin, and examined with light microscopy.

The concentration of hepatic glutathione was used to estimate the bioactivation of APAP in heparin-treated, PAR-1−/−, and low-TF mice. Frozen liver samples (100 mg) were homogenized in 1 mL of cold buffer containing 0.2 M 2-N-morpholino ethanesulfonic acid, 50 mM phosphate, and 1 mM EDTA [pH 6.0]). Homogenates were spun in a centrifuge, and supernatants were collected. Total hepatic glutathione concentration in the supernatants was determined using a commercially available kit (Cayman Chemical Co., Ann Arbor, MI).

Evaluation of Changes in the Hemostatic System.

The plasma concentration of total PAI-1 was assessed using an ELISA that measures active PAI-1, inactive PAI-1, and PAI-1/plasminogen activator complexes. Plasma fibrinogen concentration was determined from thrombin clotting time of diluted samples using a fibrometer. TAT concentrations were determined via ELISA. Deposition of fibrin in liver was evaluated via immunohistochemistry. Staining and quantification of fibrin were performed as described previously.22

Procoagulant activity (PCA) of tissues provides a measure of functional TF activity. PCA of liver homogenates was determined using a single-stage clotting assay as described by Parry et al.21 Briefly, frozen liver samples were homogenized in 15 mM N-octyl-D-glucopyranoside in 25 mM HEPES/saline and incubated for 15 minutes at 37°C. A 50-μL aliquot of liver homogenate was mixed with 50 μL of pooled mouse plasma, and clotting was initiated by the addition of 25 mM CaCl2. Clotting time was measured using a START4 Coagulation Analyzer (Diagnostica Stago), and PCA was determined using a standard curve generated using isolated mouse brain thromboplastin. The PCA of liver homogenates was normalized to the total protein concentration of each sample measured using a Bio-Rad DC protein assay (Bio-Rad Laboratories, Hercules, CA).

Statistical Analysis.

Data were analyzed using 1-way (for time-course studies) or 2-way (for heparin studies) ANOVA with appropriate post hoc tests for multiple comparisons. Data from TF and PAR-1 studies were analyzed via Student's t test. Percentage data (e.g., PCA data) were transformed (arcsin square root) before analysis. A P value of less than 0.05 was the criterion for significance for all studies.

Results

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

Development of APAP-Induced Liver Injury.

ALT activity in plasma of APAP-treated mice was significantly greater than control mice at 2 hours and continued to increase through 6 hours (Fig. 1). Histological evaluation of liver sections revealed oncotic necrosis that was largely confined to pericentral regions of the liver lobules.

thumbnail image

Figure 1. Time course of APAP-induced hepatic injury. Mice were treated with 400 mg/kg APAP or sterile saline intraperitoneally at 0 hours. Liver injury was estimated from increases in plasma ALT activity. There were no significant differences among ALT activity values for controls over time, thus they were pooled and presented as time 0 hours (n = 3-9). aSignificantly different from vehicle-treated mice (0 hours).

Download figure to PowerPoint

Changes in the Hemostatic System in APAP-Induced Liver Injury.

Normal liver expresses a relatively small concentration of TF compared with other organs.23 APAP treatment significantly increased liver PCA within 30 minutes (Table 1). The increase in PCA was transient and decreased to near basal levels by 2 hours.

Table 1. Procoagulant Activity of Livers of APAP-Treated Mice
Hours After Administration of APAPPCA (% of Control)
  1. Mice were treated with 400 mg/kg APAP or sterile saline intraperitoneally at 0 hours. There were no significant differences among control values for procoagulant activity over time, thus they were pooled and presented as time 0 hours (n = 6-7). *Significantly different from vehicle-treated mice (0 hours).

0100 ± 10.3
0.5156 ± 17.4*
1150 ± 20.2
1.5136 ± 26
2112 ± 18.9

The concentration of TAT in plasma was determined as a marker of activation of the coagulation system. It was not different in APAP-treated and control mice for the first 90 minutes after APAP administration (Fig. 2A). By 2 hours after APAP treatment, plasma TAT concentration increased significantly, and it returned toward normal by 6 hours. PAI-1 is the major endogenous inhibitor of fibrinolysis. The concentration of PAI-1 in plasma increased significantly by 6 hours (Fig. 2B).

thumbnail image

Figure 2. Development of APAP-induced changes in plasma (A) TAT and (B) PAI-1 concentrations. Mice were treated with 400 mg/kg APAP or sterile saline intraperitoneally at 0 hours. There were no significant differences among control values for TAT or PAI-1 concentrations over time, thus they were pooled and presented as time 0 hours (n = 3-9). aSignificantly different from vehicle-treated mice (0 hours).

Download figure to PowerPoint

Deposition of fibrin occurs when the rate of fibrin formation exceeds the rate of fibrinolysis. Fibrin staining was not observed in livers from vehicle-treated mice (Fig. 3A). Immunohistochemical staining for fibrin revealed that no detectable deposition had occurred by 1.5 hours after treatment with APAP (Fig. 3B); however, by 2 hours significant fibrin was observed in pericentral areas of liver lobules (Fig. 3C). Fibrin deposition progressed between 2 and 6 hours (Fig. 3D). These changes were confirmed by quantification of fibrin staining (Fig. 3E).

thumbnail image

Figure 3. Development of hepatic fibrin deposition after treatment with APAP. Mice were treated with (A) sterile saline or (B-D) 400 mg/kg APAP intraperitoneally at time 0 hours. The animals were killed at (A) 0.5, (B) 1.5, (C) 2, or (D) 6 hours after treatment. Fibrin in liver sections was stained immunohistochemically and appears black in the photomicrographs. Fibrin staining was quantified (E) as described in Materials and Methods. There were no significant differences in fibrin staining among saline-treated mice over time, thus they were pooled and presented as time 0 hours in panel E (n = 3-9). aSignificantly different from vehicle-treated mice (0 hours). (Original magnification ×100.)

Download figure to PowerPoint

Effect of Anticoagulation With Heparin on APAP-Induced Liver Injury.

To assess the role of coagulation system activation in APAP-induced liver injury, mice were treated with the anticoagulant heparin, which inhibits coagulation by increasing the affinity of antithrombin for thrombin. Administration of APAP to mice resulted in a decrease in plasma fibrinogen concentration, indicating activation of the coagulation system (Fig. 4A). Pretreatment of animals with heparin 1 hour before administration of APAP inhibited fibrinogen consumption.

thumbnail image

Figure 4. Effect of heparin on APAP-induced activation of the coagulation system. Mice were treated with either 3,000 U/kg heparin or sterile saline intraperitoneally and 1 hour later with either 400 mg/kg APAP or sterile saline intraperitoneally. Plasma and liver samples were taken 6 hours after APAP administration. The concentration of fibrinogen in plasma was measured as a marker of coagulation system activation, and liver sections were stained immunohistochemically for fibrin. (A) Plasma fibrinogen concentration. (B) Quantification of fibrin deposition. (C-F) Representative sections of liver from mice treated with (C) saline and saline, (D) heparin and saline, (E) saline and APAP, or (F) heparin and APAP (n = 3-13). aSignificantly different from the respective groups not given APAP. bSignificantly different from the saline/APAP group.

Download figure to PowerPoint

Low-level fibrin staining was observed in livers of animals treated with saline alone or with heparin alone (Fig. 4B-D). In contrast, extensive deposition of fibrin was observed in centrilobular areas of livers 6 hours after treatment with APAP (Fig. 4B,E). Significantly less staining was seen in livers of APAP-treated mice that were pretreated with heparin (Fig. 4B,F).

Prior treatment with heparin significantly reduced the increase in ALT activity 6 hours after treatment with APAP (Fig. 5A). Histologically, extensive necrosis was observed in pericentral regions of the liver lobules in APAP-treated animals (Fig. 5B). In livers of animals treated with heparin before administration of APAP, lesions were smaller in size and less frequent (Fig. 5C). Heparin alone did not affect liver histology (data not shown).

thumbnail image

Figure 5. Effect of heparin on APAP-induced hepatocellular injury. Mice were treated as described in Fig. 4; plasma and liver samples were collected 6 hours later. ALT activity was measured in plasma, and liver sections were processed for histology and staining with hematoxylin-eosin. (A) Plasma ALT activity. (B) Representative section from a mouse treated with saline and APAP. (C) Representative section from a mouse treated with heparin and APAP (n = 3-13). In heparin-treated mice, livers had less CL necrosis. aSignificantly different from the respective groups not given APAP. bSignificantly different from the saline/APAP group. Abbreviations: CL, centrilobular; PP, periportal. (Original magnification ×200.)

Download figure to PowerPoint

Twenty-four hours after administration of APAP, plasma ALT activity was not significantly different in animals pretreated with saline (13,899 ± 4,286 U/L) or heparin (10,570 ± 1,708 U/L). Hepatic fibrin deposition was also similar in these 2 groups at 24 hours (0.096 ± 0.02 and 0.085 ± 0.02 fraction positive pixels, respectively).

APAP is bioactivated by cytochrome P450 to a reactive metabolite, N-acetyl-p-benzoquinoneimine (NAPQI), which is responsible for hepatotoxicity.1 NAPQI initially binds to glutathione, and it is reported that toxicity occurs when the hepatic concentration of glutathione falls below 20% of normal. Accordingly, the concentration of glutathione in liver has been used as a marker of production of NAPQI in APAP-treated animals. Mice treated only with saline had 5.6 ± 0.3 μmol of glutathione/g liver. One hour after administration of APAP, glutathione concentrations were 0.077 ± 0.007 and 0.117 ± 0.035 μmol/g liver in mice pretreated with saline and heparin, respectively. These latter 2 values were not significantly different.

Role of TF in APAP-Induced Coagulation and Liver Injury.

TF is the principal cellular initiator of blood coagulation. To test whether TF plays a role in APAP-induced coagulation and hepatotoxicity, low-TF mice and their heterozygous littermates (control mice) were treated with APAP. Low-TF mice had smaller plasma TAT concentrations 6 hours after treatment with APAP than did their heterozygous littermates (Fig. 6A). At this time, ALT activity was greater in plasma from control mice compared with low-TF mice (Fig. 6B), as was fibrin deposition in the liver (Fig. 6C). In contrast, plasma PAI-1 was not significantly different. Despite the lesser plasma ALT activity, hepatic glutathione concentrations were smaller in low-TF mice (0.063 ± 0.037 μmol/g liver) than in heterozygous littermates (0.24 ± 0.044 μmol/g liver) 1.5 hours after administration of APAP. Liver glutathione concentration was less in both groups compared with untreated control mice (7.9 μmol/g liver). Neither plasma ALT activity nor hepatic fibrin deposition was different between low-TF and control mice 24 hours after APAP administration (Table 2).

thumbnail image

Figure 6. APAP-induced coagulation and liver injury in low-TF mice. Female, low-TF mice (mTF−/−, hTF+) and their heterozygous littermates (mTF+/−, hTF+) were treated with 400 mg/kg APAP intraperitoneally. (A) TAT concentration in plasma, (B) ALT activity in plasma, (C) fibrin deposition in the liver, and (D) plasma PAI-1 concentration were determined 6 hours after APAP administration (n = 8-9). aSignificantly different from value for heterozygous littermates.

Download figure to PowerPoint

Table 2. Liver Injury and Hepatic Fibrin Deposition 24 Hours After Administration of APAP
Animal ModelPlasma ALT (U/L)Hepatic Fibrin (Fraction-Positive Pixels)
  1. Mice were treated with 400 mg/kg APAP or sterile saline intraperitoneally at 0 hours; plasma and liver samples were collected 24 hours later. Activity of ALT in plasma and deposition of fibrin in liver were determined (n = 5). No significant differences were observed between the low-TF (mTF−/−, hTF+) animals and their controls or between the PAR-1−/− animals and their controls.

mTF+/−, hTF+ (control)5,959 ± 1,0940.10 ± 0.03
mTF−/−, hTF+ (low-TF)3,119 ± 1,6010.11 ± 0.04
PAR-1+/+9,440 ± 7130.17 ± 0.03
PAR-1−/−6,553 ± 1,7660.13 ± 0.02

Role of PAR-1 in APAP-Induced Hepatotoxicity.

Formation of the TF/factor VIIa complex leads to production of serine proteases, including thrombin, that initiate intracellular signaling pathways by activating PARs. In particular, PAR-1 is the primary cellular receptor for thrombin. Accordingly, we determined whether PAR-1 contributes to APAP-induced hepatocellular injury. There was no difference in plasma TAT concentration between wild-type mice and PAR-1−/− mice treated with APAP (Fig. 7A). By 6 hours after exposure to APAP, wild-type control mice (PAR-1+/+) developed liver damage (Fig. 7B). ALT activity was significantly reduced in PAR-1−/− mice relative to wild-type mice at this time. Hepatic fibrin deposition and plasma PAI-1 concentration were diminished in the PAR-1−/− mice relative to wild-type mice (Fig. 7C-D). No difference was observed in hepatic glutathione concentration between PAR-1+/+ and PAR-1−/− mice (0.19 ± 0.03 and 0.23 ± 0.009 μmol/g liver, respectively) 1.5 hours after administration of APAP, and values in both groups were small compared with untreated animals (8.5 μmol/g liver). By 24 hours after APAP administration, plasma ALT activity and hepatic fibrin deposition were similar in the 2 groups (Table 2).

thumbnail image

Figure 7. APAP-induced liver injury in PAR-1−/− mice. PAR-1−/− mice and their wild-type littermates (PAR-1+/+) were treated with 400 mg/kg APAP intraperitoneally. (A) TAT concentration in plasma, (B) ALT activity in plasma, (C) fibrin deposition in the liver, and (D) plasma PAI-1 concentration were determined 6 hours after APAP administration (n = 8-9). aSignificantly different from value for littermate control mice.

Download figure to PowerPoint

Discussion

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

Disturbances in the hemostatic system have been reported in individuals who develop APAP-induced liver injury. It has also been reported that APAP treatment of mice decreases antithrombin and protein C concentrations in plasma and elevates plasma concentrations of PAI-1 and von Willebrand factor,24 consistent with activation of coagulation. However, these biomarkers were measured only after hepatocellular injury had developed; accordingly, the findings do not address whether these changes contribute to initiation or progression of injury or are a consequence of hepatic damage. In the present study, we found that the coagulation system was activated within 2 hours of APAP administration, as reflected by increased plasma TAT concentrations (Fig. 2A) and by hepatic fibrin deposition (Fig. 3). This time was concurrent with the onset of liver injury as detected by an increase in plasma ALT activity (Fig. 1). Accordingly, our results raised the possibility that activation of the coagulation system participates in the pathogenesis of liver damage.

To test the hypothesis that thrombin contributes to APAP-induced liver injury, mice were pretreated with the anticoagulant heparin. Administration of heparin diminished activation of coagulation, as evidenced by inhibition of fibrinogen consumption and reduction of fibrin deposition in livers of APAP-treated mice (Fig. 4). Moreover, inhibition of coagulation with heparin significantly inhibited APAP-induced liver damage at 6 hours (Fig. 5). This result suggests that thrombin generation is involved in APAP-induced liver injury. Given that heparin has activities unrelated to inhibition of coagulation,25 a genetic approach was also taken to evaluate the role of the coagulation system.

TF is expressed constitutively on extravascular cells, and its expression is induced transcriptionally by various mediators in several cell types, including monocytes/macrophages and endothelial cells.23 During endothelial cell damage, TF initiates coagulation to limit hemorrhage. The TF/factor VIIa complex activates coagulation by generating factor Xa, which subsequently cleaves prothrombin to thrombin. TF is involved in the development of hepatocellular injury in other rodent models, such as hepatic ischemia–reperfusion and endotoxemia/partial hepatectomy.26–28 Although a causal role for TF in human APAP-induced liver injury has not been established, there is some evidence that TF mediates coagulation system activation in cases of APAP overdose. For example, the concentrations of several coagulation factors (II, V, VII, and X) were reduced in patients with acute liver injury from APAP overdose, and this was interpreted to reflect TF-initiated coagulation and consumption of these factors.13 The results presented here indicate that TF is an important component of APAP-induced liver injury in mice. Indeed, over the first 6 hours after administration of APAP, coagulation activation and hepatotoxicity were reduced in mice with low levels of TF (Fig. 6). The observation that low-TF mice were less sensitive to the hepatotoxic effects of APAP suggests that the mechanism by which APAP initiates coagulation and causes liver injury involves TF. At 24 hours, fibrin deposition and liver injury occurred from APAP in both strains of mice. TF activity in these low-TF mice is small but not completely lacking. Hence, fibrin deposition can occur with time. The concurrent return of both fibrin deposits and hepatocellular injury between 6 and 24 hours is consistent with the hypothesis that the hemostatic system is important for the progression of injury, perhaps acting through TF-independent means.

The TF-positive cell type involved in APAP-induced coagulation in the liver has not yet been identified. One possibility is that de novo synthesis of TF by cells resident in hepatic sinusoids, such as endothelial cells or Kupffer cells, contributes to APAP-induced coagulation. APAP increased the hepatic expression of TF messenger RNA in C57BL/J6 mice, although induction did not occur until after the onset of injury29 (Supplementary gene expression data). The very rapid increase in liver PCA (Table 1) and the time course of coagulation system activation (that is, TAT generation) (Fig. 2) in APAP-treated mice presented here suggest additional mechanisms that might not be dependent on de novo synthesis. One possibility is that damage to sinusoidal endothelial cells, as has been observed in APAP-treated mice,30, 31 exposes extravascular TF to circulating coagulation factors. Another possibility is that inactive TF, possibly from blood, is oxidized to an active form at the site of liver injury.32 APAP treatment induces oxidative stress evident both in liver tissue and plasma.33, 34 Thus, APAP exposure might increase PCA without an accompanying increase in TF protein.

Results from studies using low-TF mice or pretreatment with heparin suggest that thrombin is an important mediator of APAP-induced liver injury. Thrombin could contribute to APAP-induced liver damage through the formation of fibrin clots or through PAR-1 signaling. PAR-1−/− mice developed less severe injury by 6 hours than their wild-type littermates (Fig. 7B), indicating that PAR-1 signaling is involved in APAP-induced liver injury. TF-dependent thrombin generation and the subsequent activation of PAR-1 play a role in other models of tissue injury.35 PAR-1 resides on various cell types, including hepatic endothelial cells and Kupffer cells.18, 36 Activation of PAR-1 by thrombin leads to intracellular signaling that culminates in a variety of effects, such as mitogenic responses, induction of cyclooxygenase-2 and neutrophil chemokines, and activation of platelets (reviewed by Coughlin,37 Uzonyi et al.,38 and Kawabata and Kawao39). Additionally, thrombin activation of PAR-1 can increase TF expression.40–43 Taken together, our results suggest that TF-dependent generation of thrombin activates PAR-1, and this contributes to liver injury in APAP-treated mice.

In addition to the activation of PAR-1 signaling, fibrin deposition resulting from thrombin generation could contribute to APAP-induced liver injury. For example, sinusoidal fibrin deposition could disrupt blood flow, resulting in localized hypoxia and hepatocellular necrosis. The observation that sinusoidal blood flow was reduced early after administration of a hepatotoxic dose of APAP to mice30 supports this possibility. In addition, administration of APAP to mice led to stabilization of hypoxia-inducible factor alpha,31 which is consistent with the development of hypoxia, although other mechanisms of stabilization occur. The results presented here demonstrate that hepatic fibrin deposits were localized to centrilobular areas of the liver that were selectively damaged. Furthermore, in each condition in which APAP hepatotoxicity was diminished (that is, after pretreatment with heparin, in low-TF mice, and in PAR-1−/− mice), hepatic fibrin deposition was also reduced. Furthermore, in each of these circumstances, by 24 hours liver injury was similar to that seen in the respective control groups, and fibrin deposition had increased to equivalent levels. Thus, fibrin deposition tracked well with liver damage. These data are consistent with the hypothesis that the hemostatic system is important in the progression of APAP-induced liver injury.

Inhibition of fibrinolysis by PAI-1 might contribute to APAP-induced hepatic fibrin deposition and liver damage. Hepatic expression of PAI-1 messenger RNA increased by 6 hours after administration of APAP to mice.44 In addition, the plasma concentration of PAI-1 increased dramatically in 2 patients during the first 12 hours after admission to the hospital for APAP overdose.45 In the present study, the concentration of PAI-1 in plasma increased between 2 and 6 hours after treatment with APAP (Fig. 2). Although liver injury and hepatic fibrin deposition progressed during this period, the concentration of TAT in plasma decreased. These results are consistent with the hypothesis that the coagulation system plays a major role in the initiation of fibrin deposition after APAP treatment whereas impaired fibrinolysis contributes to its progression at later times. This might explain why fibrin deposition in low-TF and littermate controls is not different at 24 hours.

Recently, activation of c-Jun NH2-terminal kinase was found to be important for APAP-induced hepatotoxicity in mice.46 PAR-1 activation by thrombin can lead to PAI-1 expression through activation of c-Jun NH2-terminal kinase.47, 48 This would favor inhibition of fibrinolysis and promote the deposition of fibrin, and it could explain why APAP-induced fibrin deposition was smaller in PAR-1−/− mice compared with controls, despite a similar TAT response (Fig. 7). Indeed, the plasma concentration of PAI-1 was reduced in PAR-1−/− mice, suggesting that PAR-1 signaling contributes to APAP-induced PAI-1 expression (Fig. 7).

APAP toxicity depends on metabolism to NAPQI, its toxic metabolite that binds to cellular nucleophiles and glutathione.1 Toxicity only occurs when glutathione is depleted, and the degree of depletion has served as a biomarker of APAP activation. The decrease in hepatic glutathione concentration we observed in APAP-treated animals was pronounced. The protection from toxicity observed in our studies could be due to diminished bioactivation to NAPQI; however, the decrease in liver glutathione concentration from APAP exposure was no different after heparin treatment or in PAR-1−/− mice relative to their APAP-treated controls. In addition, TF-deficient mice had slightly smaller hepatic glutathione concentrations after APAP administration than did their controls, despite having less hepatocellular injury. These results suggest that the protection afforded by these manipulations cannot be explained by reduced APAP bioactivation.

In summary, coagulation system activation, increased PAI-1 concentration and hepatic fibrin deposition occurred during the development of hepatocellular injury in mice treated with a toxic dose of APAP. Anticoagulation with heparin reduced APAP hepatotoxicity, and both coagulation and hepatocellular damage were reduced in mice deficient in TF. Furthermore, APAP-induced liver injury was less severe in PAR-1−/− mice compared with wild-type mice. These results are consistent with evidence of coagulation system activation in APAP-poisoned human patients and suggest an important role for the coagulation system in APAP hepatotoxicity.

References

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

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information
FilenameFormatSizeDescription
jws-hep.21779.fig1.pdf40KSuppl Fig 1: OT-I CD8 T-cells were purified from the lymph nodes of OT-I mice. 4 million cells were CFSE-labelled and transferred intravenously into ASBT-OVA or TF-OVA mice. Experiments in ASBT-OVA and TF-OVA mice were carried out simultaneously and mice received cells from the same batch of OT-I T-cells. 20 hours after transfer, cells were isolated form the livers and stained for CD8 and Va2. In (A), expression levels of CD8 and Va2 within the population of CFSE-positive (transgenic) cells are depicted. In (B), numbers of CD8+Va2+ and CFSE+ cells isolated from the livers of ASBT-OVA and TF-OVA mice are compared (n=6 each).
jws-hep.21779.fig2.pdf34KSuppl Fig 2: OT-I CD8 T-cells were purified from the lymph nodes of OT-I or OT-I RAG-/- mice. Cells isolated from OT-I mice were also selected for CD62L expression. 4 million cells were CFSE-labelled and transferred intravenously into TF-OVA (A) or ASBT-OVA mice (B). T-cells were isolated from the indicated tissues (Ing. LN. Inguinal lymph node; Liv. LN. Liver lymph node; Intrahep. Intrahepatic) at 44 hours and analyzed for the presence of proliferating transgenic T-cells by flow cytometry. Plots display data gated on CD8+Va2+ cells.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.