Patricia E. Ganey

Modest Inflammation Enhances Diclofenac Hepatotoxicity in Rats: Role of Neutrophils and Bacterial Translocation

Idiosyncratic adverse drug reactions (IADRs) represent an important human health problem, yet animal models for preclinical prediction of these reactions are lacking. Recent evidence in animals suggests that some IADRs arise from drug interaction with an inflammatory episode that renders the liver sensitive to injury. Diclofenac (DCLF) is one of those drugs for which the clinical use is limited by idiosyncratic liver injury. We tested the hypothesis that modest inflammation triggered in rats by a small dose of lipopolysaccharide (LPS) renders a nonhepatotoxic dose of DCLF injurious to liver. Cotreatment of rats with nonhepatotoxic doses of LPS and DCLF resulted in elevated serum alanine aminotransferase activity and liver histopathologic changes 6 h after DCLF administration. Neither LPS nor DCLF alone had such an effect. Gene array analysis of livers revealed a unique gene expression pattern in the LPS/DCLF-cotreated group compared with groups given either agent alone. Antiserum-induced neutrophil (PMN) depletion in LPS/DCLF-cotreated rats protected against liver injury, demonstrating a role for PMNs in the pathogenesis of this LPS/DCLF interaction. Gut sterilization of LPS/DCLF-treated rats did not protect against liver injury. In contrast, gut sterilization did attenuate liver injury caused by a large, hepatotoxic dose of DCLF, suggesting that hepatotoxicity induced by large doses of DCLF is caused in part by its ability to increase intestinal permeability to endotoxin or other bacterial products. These results demonstrate that inflammation-DCLF interaction precipitates hepatotoxicity in rats and raise the possibility of creating animal models that predict human IADRs.

Microarray analysis of lipopolysaccharide potentiation of trovafloxacin-induced liver injury in rats suggests a role for proinflammatory chemokines and neutrophils

Idiosyncratic drug toxicity refers to toxic reactions occurring in a small subset of patients and usually cannot be predicted during preclinical or early phases of clinical trials. One hypothesis for the pathogenesis of hepatic idiosyncratic drug reactions is that, in certain individuals, underlying inflammation results in sensitization of the liver, such that injury occurs from an agent that typically would not cause hepatotoxicity at a therapeutic dose. We explored this possibility by cotreating rats with nonhepatotoxic doses of bacterial lipopolysaccharide (LPS) and trovafloxacin (TVX), a drug that caused idiosyncratic hepatotoxicity in humans. The combination of LPS and TVX resulted in hepatotoxicity in rats, as determined by increases in serum alanine aminotransferase activity and hepatocellular necrosis, which were not observed with either agent alone. In contrast, treatment with LPS and levofloxacin, a fluoroquinolone without human idiosyncratic liability, did not result in these changes. Liver gene expression analysis identified unique changes induced by the combination of TVX and LPS, including enhanced expression of chemokines, suggestive of liver neutrophil (PMN) accumulation and activation. Consistent with a role for PMN in the hepatotoxicity induced by LPS/TVX, prior depletion of PMN attenuated the liver injury. The results suggest that gene expression profiles predictive of idiosyncratic liability can be generated in rats cotreated with LPS and drug. Furthermore, they identify gene expression changes that could be explored as biomarkers for idiosyncratic toxicity and lead to enhanced understanding of the mechanism(s) underlying hepatotoxicity induced by TVX.

Concurrent inflammation as a determinant of susceptibility to toxicity from xenobiotic agents

Sensitivity to the toxic effects of xenobiotic agents is influenced by a number of factors. Recent evidence derived from studies using experimental animals suggests that inflammation is one of these factors. For example, induction of inflammation by coexposure to bacterial endotoxin, vitamin A or Corynebacterium parvum increases injury in response to a number of xenobiotic agents that target liver. These agents are diverse in chemical nature and in mechanism of hepatotoxic action. Factors critical to the augmentation of liver injury by inflammation include Kupffer cells, neutrophils, cytokines such as tumor necrosis factor-alpha (TNF-alpha) and lipid mediators such as prostaglandins, but these may vary depending on the xenobiotic agent and the mechanisms by which it alters hepatocellular homeostasis. In addition, the timing of inflammagen exposure can qualitatively alter the toxic response to chemicals. Inflammation-induced increases in susceptibility to toxicity are not limited to liver. Concurrent inflammation also sensitizes animals to the toxic effects of agents that damage the respiratory tract, kidney and lymphoid tissue. It is concluded that inflammation should be considered as a determinant of susceptibility to intoxication by xenobiotic exposure.

Identification of factors from rat neutrophils responsible for cytotoxicity to isolated hepatocytes

Activated polymorphonuclear neutrophils (PMNs) have been shown to be cytotoxic to rat hepatic parenchymal cells in vitro. This cytotoxicity could be observed without direct cell‐cell contact, since the conditioned medium from PMNs activated with formyl‐Met‐Leu‐Phe (fMLP) was effective in hepatocyte killing. To identify the toxic factor(s) released by PMNs, degranulation was induced by fMLP in PMNs pretreated with cytochalasin B. The contents released from the phagocytes were subjected to gel filtration on a Sephadex G‐100 column. Resulting fractions were tested for cytotoxicity to isolated hepatocytes by using release of alanine aminotransferase as a marker for hepatocyte injury. Cytotoxicity was associated with fractions containing cathepsin G and elastase and not with other fractions, including those containing myeloperoxidase. The former two enzymes were purified to homogeneity with a carboxymethyl cellulose column. Each of these enzymes demonstrated concentration‐dependent cytotoxicity to hepatocytes at concentrations >2 μg/mL. Moreover, they exhibited an additive cytotoxic effect. Effective concentrations for the combined cathepsin C and elastase in the incubation mixture were similar to the concentrations of these enzymes in PMN‐conditioned medium that produced cytotoxicity to hepatocytes. Cytotoxicity of either purified enzyme or of conditioned medium could be prevented by plasma α‐1‐antitrypsin or soybean trypsin‐chymotrypsin inhibitor, which were also potent inhibitors of enzymic activity of both cathepsin G and elastase. By contrast, the serine protease inhibitors, aprotinin and 4‐(2‐aminoethyl)‐benzenesulfonyl fluoride, were less effective in inhibiting cathepsin G and elastase activities as well as cytotoxicity caused by the purified proteases or PMN‐conditioned medium. These results support the hypothesis that cathepsin G and elastase are important mediators of hepatic parenchymal cell killing produced by activated PMNs in vitro.

Development of morphologic, hemodynamic, and biochemical changes in lungs of rats given monocrotaline pyrrole

A single, intravenous administration of a low dose of monocrotaline pyrrole (MCTP), a derivative of the pyrrolizidine alkaloid monocrotaline (MCT), induces progressive pulmonary hypertension and right ventricular hypertrophy (RVH) in rats. The temporal relationship between morphologic alterations, biochemical markers of lung injury, and the development of pulmonary hypertension was determined during the developing pulmonary disease. Three days after a single iv injection of 3.5 mg/kg MCTP, small increases in bronchoalveolar lavage (BAL) fluid lactate dehydrogenase (LDH) activity and accumulation in the lungs of intravenously administered 125I-bovine serum albumin (BSA) were associated with minimal to mild interstitial edema around large airways and blood vessels. By Day 5, BAL fluid LDH activity and 125I-BSA accumulation had increased further, and lung weight/body weight ratio and BAL fluid protein concentration were greater than those of control. Interstitial edema was more pronounced and involved patches of alveolar septal walls. A mild increase in numbers of mononuclear cells, including hypertrophied interstitial cells, was evident in these areas. Walls of pulmonary arteries less than 60 microns in diameter were mildly thickened. By Day 8, scattered clusters of alveolar sacs contained serous exudate, and interstitial mononuclear infiltrates were more pronounced. Mild to moderate thickening of arterial walls was apparent in small and large vessels. By Day 14, pulmonary arterial pressure was elevated and RVH was evident. Arterial walls were thickened and had hypertrophy of medial smooth muscle cells and intercellular edema, which was particularly prominent in areas with perivascular interstitial inflammation. Large patches of lung interstitium and alveolar lumens contained serous or fibrinous exudate. In summary, a single, intravenous administration of MCTP induced a delayed and progressive pulmonary microvascular leak, interstitial inflammation, and alterations in muscular blood vessels which resulted in pulmonary hypertension within 14 days. These morphologic, biochemical, and hemodynamic changes are nearly identical to alterations induced by the parent alkaloid, MCT.

Phospholipase A2 is involved in the mechanism of activation of neutrophils by polychlorinated biphenyls

Aroclor 1242, a mixture of polychlorinated biphenyls (PCBs), activates neutrophils to produce superoxide anion (O2-) by a mechanism that involves phospholipase C-dependent hydrolysis of membrane phosphoinositides; however, subsequent signal transduction mechanisms are unknown. We undertook this study to determine whether phospholipase A2-dependent release of arachidonic acid is involved in PCB-induced O2- production. We measured O2- production in vitro in glycogen-elicited, rat neutrophils in the presence and absence of the inhibitors of phospholipase A2: quinacrine, 4-bromophenacyl bromide (BPB), and manoalide. All three agents significantly decreased the amount of O2- detected during stimulation of neutrophils with Aroclor 1242. Similar inhibition occurred when neutrophils were activated with the classical stimuli, formyl-methionyl-leucyl-phenylalanine (fMLP) or phorbol myristate acetate. The effects of BPB and manoalide were not a result of cytotoxicity or other nonspecific effects, although data suggest that quinacrine is an O2- scavenger. Significant release of 3H-arachidonic acid preceded O2- production in neutrophils stimulated with Aroclor 1242 or fMLP. Manoalide, at a concentration that abolished O2- production, also inhibited the release of 3H-arachidonate. Aspirin, zileuton, or WEB 2086 did not affect Aroclor 1242-induced O2- production, suggesting that eicosanoids and platelet-activating factor are not needed for neutrophil activation by PCBs. Activation of phospholipase A2 and O2- production do not appear to involve the Ah receptor because a congener with low affinity, but not one with high affinity for this receptor, stimulated the release of arachidonic acid and O2-. These data suggest that Aroclor 1242 stimulates neutrophils to produce O2- by a mechanism that involves phospholipase A2-dependent release of arachidonic acid. ImagesFigure 1. AFigure 1. BFigure 2.Figure 3.Figure 4. AFigure 4. BFigure 4. CFigure 5. AFigure 5. BFigure 5. CFigure 6. AFigure 6. BFigure 6. CFigure 6. D

IMPAIRMENT OF HUMAN NEUTROPHIL OXIDATIVE BURST BY POLYCHLORINATED BIPHENYLS : INHIBITION OF SUPEROXIDE DISMUTASE ACTIVITY

We report evidence of a novel mechanism by which polychlorinated biphenyls might act as potent inducers of inflammation. Aroclor 1242 (A1242), a polychlorinated biphenyl mixture, and 2,2′,4,4′‐tetrachlorobiphenyl (PCB47), a constituent of A1242 that produces the same patterns of effects, impaired the oxidative burst of human neutrophils by inhibiting the antioxidant enzyme superoxide dismutase, which converts O2 –to H2O2. Pre‐incubation of neutrophils with A1242 or PCB47 before stimulation with phorbol 12‐myristate 13‐acetate heightened the respiratory burst, producing a significant increase in intracellular O2 ‐ production along with a significant decrease in H2O2 production compared with unexposed agonist‐stimulated neutrophils. This was also evident in a physiologically relevant situation in which neutrophils pre‐incubated with A1242 were subsequently stimulated with a combination of N‐formyl‐L‐methionyl‐L‐leucyl‐L‐phenylalanine and tumor necrosis factor‐α. Incubation of bovine copper‐zinc superoxide dismutase (EC 1.15.1.1) with A1242 or PCB47 in a cell‐free system reversed the enzyme‐mediated inhibition of 6‐hydroxydopamine autoxidation, indicating that polychlorinated biphenyls inhibited superoxide dismutase activity. Low superoxide dismutase activity in neutrophils leads to imbalances between production of free radicals and antioxidant defense mechanisms, which can in turn induce tissue damage and hasten the onset of neutrophil apoptosis. J. Leukoc. Biol. 63: 216–224; 1998.

Gadolinium chloride pretreatment protects against hepatic injury but predisposes the lungs to alveolitis after lipopolysaccharide administration.

Exposure to lipopolysaccharide (LPS) can result in multi-organ failure and death. After an intravenous injection of LPS into rats, neutrophils (PMN) rapidly accumulate in the liver sinusoids and pulmonary vasculature, and PMN play a critical role in producing both hepatic and pulmonary injury. Kupffer cells (KC), the resident macrophages of the liver, phagocytose LPS and produce inflammatory mediators which may be chemotactic and stimulatory for PMN. The purpose of this study was to determine whether inhibition of KC function affects PMN accumulation and the development of parenchymal injury in the liver and lungs after systemic administration of LPS. Female, Sprague-Dawley rats (180–230 g) were pretreated with either gadolinium chloride-6H2O (GdCl3; 10 mg/kg, intravenously), to inactivate KC, or saline vehicle 24 h before receiving either LPS (4 mg/kg, intravenously) or saline vehicle. Rats were killed 1.5, 6, and 24 h after LPS administration. In a preliminary study, exposure to GdGl3 decreased uptake of carbon in the liver, indicating inhibition of phagocytosis by KC. Ninety minutes after administration of LPS, PMN accumulated in the livers of LPS-treated rats, and this effect was not altered by pretreatment with GdCl3. Similarly, exposure to LPS resulted in PMN accumulation in the pulmonary tissue, which was unaffected by GdCl3 pretreatment. Exposure to GdCl3 before LPS administration resulted in a significant increase in the number of PMN recovered by bronchoalveolar lavage at 24 h, indicating diffuse acute alveolitis. LPS-induced hepatic injury was prevented by pretreatment with GdCl3; however, the increased wet lung/body weight ratio observed after LPS administration was unaffected by GdCl3. These results confirm that inactivation of KC protects against hepatic injury and extend this finding by ruling out inhibition of hepatic PMN accumulation as a mechanism for this effect. The data also suggest that treatment with GdCl3 predisposes the lungs to alveolitis during systemic exposure to LPS.

Liver inflammation during monocrotaline hepatotoxicity.

Monocrotaline (MCT) is a pyrrolizidine alkaloid (PA) plant toxin that causes hepatotoxicity in humans and animals. Human exposure occurs from consumption of contaminated grains and herbal teas and medicines. Intraperitoneal injection (i.p.) of 300 mg/kg MCT in rats produced time-dependent hepatic parenchymal cell (HPC) injury beginning at 12 h. At this time, an inflammatory infiltrate consisting of neutrophils (PMNs) appeared in areas of hepatocellular injury, and activation of the coagulation system occurred. PMN accumulation was preceded by up-regulation of the PMN chemokines cytokine-induced neutrophil chemoattractant-1 (CINC-1) and macrophage inflammatory protein-2 (MIP-2) in the liver. The monocyte chemokine, monocyte chemoattractant protein-1 (MCP-1), was also upregulated. Inhibition of Kupffer cell function with gadolinium chloride (GdCl(3)) significantly reduced CINC-1 protein in plasma after MCT treatment but had no effect on hepatic PMN accumulation. Since inflammation can contribute to either pathogenesis or resolution of tissue injury, we explored inflammatory factors as a contributor to MCT hepatotoxicity. To test the hypothesis that PMNs contribute to MCT-induced HPC injury, rats were depleted of PMNs with a rabbit anti-PMN serum prior to MCT treatment. Anti-PMN treatment reduced hepatic PMN accumulation by 80% but had no effect on MCT-induced HPC injury or activation of the coagulation system. To test the hypothesis that Kupffer cells and/or tumor necrosis factor-alpha (TNF-alpha) are required for MCT-induced HPC injury, rats were treated with either GdCl(3) to inhibit Kupffer cell function or pentoxifylline (PTX) to prevent synthesis of TNF-alpha. Neither treatment prevented MCT-induced HPC injury. Results from these studies suggest that PMNs, Kupffer cells and TNF-alpha are not critical mediators of MCT hepatotoxicity. Accordingly, although inflammation occurs in the liver after MCT treatment, it is not required for HPC injury and possibly occurs secondary to hepatocellular injury.

Role of hepatic fibrin in idiosyncrasy-like liver injury from lipopolysaccharide-ranitidine coexposure in rats

Coadministration of nonhepatotoxic doses of the histamine 2‐receptor antagonist ranitidine (RAN) and bacterial lipopolysaccharide (LPS) results in hepatocellular injury in rats, the onset of which occurs in 3 to 6 hours. This reaction resembles RAN idiosyncratic hepatotoxicity in humans. Early fibrin deposition occurs in livers of rats cotreated with LPS/RAN. Accordingly, we tested the hypothesis that the hemostatic system contributes to liver injury in LPS/RAN‐treated rats. Rats were given either LPS (44.4 × 106 EU/kg) or its vehicle, then RAN (30 mg/kg) or its vehicle 2 hours later. They were killed 2, 3, 6, 12, or 24 hours after RAN treatment, and liver injury was estimated from serum alanine aminotransferase activity. A modest elevation in serum hyaluronic acid, which was most pronounced in LPS/RAN‐cotreated rats, suggested altered sinusoidal endothelial cell function. A decrease in plasma fibrinogen and increases in thrombin‐antithrombin dimers and in serum concentration of plasminogen activator inhibitor‐1 occurred before the onset of liver injury. Hepatic fibrin deposition was observed in livers from LPS/RAN‐cotreated rats 3 and 6 hours after RAN. Liver injury was abolished by the anticoagulant heparin and was significantly attenuated by the fibrinolytic agent streptokinase. Hypoxia, one potential consequence of sinusoidal fibrin deposition, was observed in livers of LPS/RAN‐treated rats. In conclusion, the results suggest that the hemostatic system is activated after LPS/RAN cotreatment and that fibrin deposition in liver is important for the genesis of hepatic parenchymal cell injury in this model. (HEPATOLOGY 2004;40:1342–1351.)