First International Symposium on the Molecular Pathology and Clinical Aspects of the Inflamed Liver: Alcohol and Cytokines

Manuela G. Neuman, PhD

Clin Invest Med 1998; 21(6): 283-6

The symposium was sponsored by the Medical Research Council of Canada, the Royal College of Physicians and Surgeons of Canada, the National Institute on Alcohol Abuse and Alcoholism, the University of Toronto, and many national and international societies.

Reprint requests to: Dr. Manuela Neuman, Division of Clinical Pharmacology, E 242, Sunnybrook Health Science Centre, 2075 Bayview Ave., Toronto ON M4N 3M5; fax 416 480-6025; manuela@fisher.sunnybrook.utoronto.ca

© 1998 Canadian Medical Association / Association médicale canadienne

Understanding of the mechanisms of injury to the liver has advanced rapidly in recent years, leading to new approaches to prevention and treatment of liver diseases. Among these exciting advances, the elucidation of the role of various cell types that constitute the functioning participants and targets of disease, and the role of cytokines, have been critical. Cytokines are inflammatory mediators that provide the interplay between cell types in the evolution of acute and chronic disease. The goal of the symposium was to provide an overview of the immunopathologic mechanisms of alcohol-, drug- and virus-induced liver injury, and their clinical implications, by focusing on the participation of each of the major cell types of the liver in injury and repair. The main unknowns in this area are the processes that contribute to chronic disease course, possible healing and transformation to malignancy.

Chronic drug abuse or viral infection activates cytotoxic T lymphocytes and initiates apoptotic processes. Tumour necrosis factor alpha (TNF-alpha), which is released by inflammatory cells, may also induce apoptosis through TNF receptors. Viral infection and replication commonly result both in apoptosis and necrosis. This may be responsible for many of the pathologic effects associated with hepatitis. Recent data suggest that hepatitis-B and -C viral proteins may modulate the immunologic profile of the host. Are the present therapies for the treatment of viral hepatitis changing the balance of the immunopathologic pattern in the patient by enhancing apoptosis or by reducing the patient's inflammatory responses? What is the role of drug therapy (with, for example, acetaminophen or phenobarbital) or of drugs of abuse (such as alcohol) in the molecular events that occur in this inflammation? A clearer understanding of the immunopathology of alcoholic and viral hepatitis, as well as the possible immunomodulatory effects of therapies on host-mediated response, may uncover novel targets for treatment.

The relation among drugs, alcohol and other hepatotoxins was presented by H.J. Zimmerman (Washington). The interplay of alcohol with drugs is complex and difficult to identify, but significant. Ethanol has been found to enhance the hepatotoxic effects of chloroalkanes, aflatoxin B₁, allyl alcohol, bromobenzene, cocaine, enflurane, galactosamine, halothane, isoniazid, nitrosamines, thioacetatnide, vinyl chloride, vinylidine chloride and vitamin A. The toxicity of several hepatotoxicants is unaffected, and at least 1 — amanitine, the alkaloid responsible for mushroom poisoning — is decreased by ethanol.

The effect of ethanol on the toxicity of carbon tetrachloride and acetaminophen has been studied most extensively. In vitro, my studies show that the combination of ethanol and acetaminophen produces an elevation of TNF-alpha levels that may contribute to hepatocytotoxicity. The combination induces TNF-dependent hepatocyte apoptosis resembling immune-mediated fulminant hepatic failure in humans. Intracellular pathways originating at the TNF receptor are either linked to apoptosis, nuclear factor (NF)-kappaB translocation, or jun kinase (JNK) activation.

Focusing on the advances in the understanding of alcohol-induced cell death by apoptosis versus by necrosis, I offered the view of the basic scientist investigating the molecular pathologic and subcellular morphologic mechanisms of alcohol-induced liver damage. I have established, in human primary hepatocytes in vitro, a model of ethanol-induced liver damage that mimics the biochemical and morphologic features obtained in vivo in patients with alcohol-induced liver damage. In biopsies of specimens from patients with alcoholic hepatitis, cell death by apoptosis has been observed. In vitro, in cells exposed repeatedly to high but clinically realistic doses of ethanol, the cytoplasm density is higher than normal, and nuclear chromatin condenses into compact masses. The integrity of organelles and plasma membrane remains intact, but cell size decreases (by 30%), and cell shape becomes elongated and fragmented into apoptotic bodies. The DNA fragments at the molecular level. The process of apoptosis appears to be multifactorial: both through induction of P450 2E1 and production of TNF. Based on these data, I have concluded that, in this in vitro model, ethanol-induced liver cell death is a controlled process, and the extent of apoptosis is regulated by both the ethanol dose and the frequency of exposure to ethanol.

C. Tiribelli (Trieste, Italy) presented work, conducted in collaboration with me, on the role of bile-acid therapy in alcohol-induced liver damage. Tiribelli described how chronic ethanol exposure leads to immunoregulatory and detoxification impairment. The experiments show that, in human hepatocytes, alcohol upregulates intracellular expression of cytokines, while bile acids downregulate it. Increased amounts of TNF-alpha and depletion of both mitochondrial and cytosolic glutathione contribute to ethanol cytotoxicity. Interestingly, bile acids prevent toxicity.

The role of Kupffer cells in alcohol-induced liver disease was discussed by C. McClain (Lexington, Kentucky). A TNF-promoter polymorphism has recently been linked with susceptibility to alcoholic steatohepatitis, suggesting that a subset of people who drink may be genetically predisposed to alcohol-induced liver disease. McClain reported elevated serum interleukin (IL)-6 concentrations in patients with alcoholic hepatitis or cirrhosis, and these levels correlated with markers of the acute phase response, with liver function and with clinical outcome. Increased plasma concentrations of chemo-attractant cytokines, such as IL-6, were observed in alcohol-induced liver disease, with the highest levels found in patients with acute alcoholic hepatitis. While there are elevated levels of pro-inflammatory cytokines in alcoholic liver disease, there also is under-production of the anti-inflammatory cytokine IL-8. As well, the decreased secretion of the anti-inflammatory cytokine IL-10 may play a role in the imbalance in cytokine levels in alcohol-induced liver disease.

The concept of anticytokine therapy is not unique to alcohol-induced liver disease. Successful human clinical trials involving other diseases, such as rheumatoid arthritis and inflammatory bowel disease, in which there are also increased levels of pro-inflammatory cytokines such as TNF, are already under way. However, with anticytokine therapy there is the potential to impair liver regeneration by completely inhibiting cytokine (TNF-alpha, IL-6) activity. Thus, approaches designed to inhibit only the toxic effects of TNF-alpha signalling may be a more effective strategy.

The regulation of transcriptional and translational factors in cytokine metabolism was discussed by D. Brenner (Chapel Hill, NC). The hepatic stellate cell, after exposure to a fibrogenic stimulus, is transformed from a quiescent to an activated state. Cytokines induce NF-kappaB activity in activated, but not in quiescent, stellate cells, with subsequent expression of NF-kappaB­responsive genes. Brenner's group investigated the effect of proteasome inhibitors and an IkappaB super-repressor on the cytokine-mediated activation of NF-kappaB and IL-6 in activated stellate cells. This inhibition is a potential target for anti-inflammatory therapy in the liver and might influence the activation of stellate cells after fibrogenic stimuli.

The next topic, presented by A.M. Diehl (Baltimore, Md.) concerned ethanol, cytokines and liver regeneration. Ethanol ingestion may interrupt the proregenerative signal transduction that is initiated by injury-related cytokines such as TNF-alpha and IL-6. Diehl's findings indicate that chronic ethanol ingestion arrests the regenerative process during the prereplicative period, and that increased levels of TNF, IL-6 and Stat-3 are not sufficient to assure hepatocyte proliferation after partial hepatectomy.

M. Rojkind (Bronx, NY) talked about the production of tumour growth factor ß (TGF-ß) and its role in hepatic fibrogenesis. TGF-ß is a member of a family of cytokines that modulate the expression of extracellular matrix components; thus, it plays a key role in wound healing and organ fibrosis. Several cells within the liver, including Kupffer cells and hepatic stellate cells (the main collagen-producing cells in normal and fibrotic livers), produce it. In chronic liver injury, when there are stellate cells already committed to produce type I collagen, nonspecific inflammatory stimuli, such as an acute phase response, contribute to liver fibrosis via production of IL-6. The same cytokines that modulate inflammatory and immunologic responses, and play a key role in hepatocyte proliferation during liver regeneration, have an important side effect: they contribute to liver fibrosis. Therefore, they should be considered possible targets for future antifibrogenic therapy.

The metabolism of ethanol impairs oxidation of fatty acids by liver cells. Because fatty acids are ligands for the peroxisome proliferator-activated receptor (PPAR), the effects of ethanol on the ability of transfected PPAR to activate a reporter gene element of the acyl-CoA oxidase gene have been studied (D. Crabb, Indiana University, Indianapolis, Ind.). It was suggested that oxidative stress-mediated mitochondrial injury inhibits transcriptional activation and DNA binding PPAR. S. Worral and M. Pennington (Queensland, Australia) discussed a possible role for acetaldehyde adducts in alcohol-induced liver disease as well as the importance of immune mediators and increased expression of intercellular adhesion molecule-1 (ICAM-1) on hepatocytes in experimental alcoholic hepatitis. Rats were fed either a normal diet (control) or an ethanol liquid diet for 6 weeks and injected with lipopolysaccharide (LPS). Hepatic necrosis with neutrophil infiltration was observed. LPS treatment induced expression of hepatic monocyte chemo-attractant protein-1, cytokine-induced neutrophil chemo-attractant and macrophage inflammatory protein mRNA. Hepatic ICAM-1 and vascular cell adhesion molecule-1 mRNA expression was elevated after LPS in both groups. The conclusion was that these adhesion molecules, in conjunction with elevated pro-inflammatory cytokine levels, may assist in the formation of a chemotactic gradient within the liver, resulting in the neutrophil infiltration seen both in this model and in human alcoholic hepatitis.

A interesting insight into the role of ethanol in adverse drug reactions was provided by N. Shear (Toronto, Ont.). Severe drug-mediated hepatotoxicity can occur in an unpredictable manner. Results from neither animal models nor initial clinical trials may help predict this. A drug-induced hypersensitivity syndrome is characterized by fever, lymphadenopathy, rash, nephritis, hematologic changes and hepatitis. This syndrome is not dependent on the dose of the drug, the frequency (number of doses per day) or duration (number of days) of its administration. Recently, Shear and I have demonstrated that in vitro models of toxicity, such as the lymphocyte toxicity assay, and hepatocyte culture systems using end points such as cytokine release and their messenger RNA expression, may help in understanding the pathogenesis of hypersensitivity reactions and in defining effective therapies for them.

J. Heathcote (Toronto, Ont.) spoke on cytokines in primary biliary cirrhosis (PBC). Since we know that cytokines are the "messengers" of the immune response, an alteration in the normal circadian rhythm of cytokines has been implicated in chronic fatigue syndrome. Heathcote's group examined whether there was an association between serum cytokine levels, fatigue and osteoporosis in subjects with PBC. Serum levels of IL-1 and IL-6, TNF-alpha and TGF-ß were assayed from patients with PBC. No significant differences in serum levels of any individual cytokine correlated with the presence or absence of fatigue or with osteoporosis in the patients with PBC studied.

The clinical theme continued with a review of hepatitis B and C as examples of virus-induced liver injury and an exploration of the possible ways in which alcohol may alter cytokine metabolism. D. Dhumeaux (Paris, France) and V. Feinman (Toronto, Ont.) made a presentation on the role of interferon therapy for chronic hepatitis B and C. Dhumeaux's new data on intracellular signal transduction by interferons may improve the clinical efficacy of interferon treatment. It may also become possible to modify resistance to interferon treatment. Introduction of recombinant technology has provided large amounts of recombinant interferon and made possible large-scale, randomized, multicentre trials of interferon treatment of viral diseases and cancer. The use of interferon in the treatment of chronic viral hepatitis B and C has been partially successful.

The symposium concluded with a poster session, which permitted young investigators and fellows from Mexico, Spain, Canada, Korea, the US, South Africa and Bangladesh to present their current basic and clinical research on virus- and drug-induced liver disease. The role of cytokines in the diagnosis and pathology of this damage was underlined.

I closed the symposium by noting the diversity of topics considered during the scientific sessions and by offering my view on the information that had emerged concerning the role of cytokines in both experimental and clinical liver inflammation.