• alcoholic liver disease;
  • hepatitis;
  • immunopathology;
  • steatosis


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

Whether immunological responses are involved in initiation and progression of alcoholic liver disease is unclear. We describe a mouse model of alcoholic liver injury characterized by steatosis and hepatic inflammation initiated by a recall immune response. Mice immune to Listeria monocytogenes fed a liquid diet containing ethanol and challenged with viable bacteria developed steatosis within 24 h and, at a later time, elevated serum alanine aminotransferase levels, indicating more liver damage in this group. Listeria antigen also induced steatosis and increased serum alanine aminotransferase levels in immune ethanol-consuming mice. The production of tumour necrosis factor by a recall immune response in this model is a major, but not the only, component in initiation of alcoholic liver disease.


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

Induction of pathological alterations in the liver of human beings related to chronic consumption of ethanol (ETOH) has been recognized for some time. Abuse of ETOH is also associated with immunological abnormalities, including alterations in immunoregulatory mechanisms that seem to play a role in autoimmune processes involved in alcoholic liver diseases. ETOH-induced liver injury is characterized by fatty liver, hepatomegaly, fibrosis and cirrhosis. Patients with severe alcoholic hepatitis and cirrhosis have increased levels of interleukin (IL)-1, IL-6, IL-8 and tumour necrosis factor (TNF)-alpha, as well as endotoxaemia [1–6]. Hypergammaglobulinaemia and antibodies against liver-specific proteins [7], liver membrane antigens [8] and acetaldehyde [9] have been associated with alcoholic liver disease. Up-regulation of expression of intercellular adhesion molecule-1 and lymphocyte function-associated molecule-3 on hepatocyte membranes of patients with alcoholic hepatitis may be responsible for leucocyte recruitment to the liver and (ultimately) cell-mediated immune damage [10]. Distribution and persistence of CD8+ and CD4+ lymphocytes in active alcoholic liver disease, together with enhanced expression of class I major histocompatibility antigens on hepatocytes and the relationship of lymphocytes to formation of hyaline and occurrence of hepatocellular necrosis, have led to the suggestion that cytotoxic T lymphocytes are involved in the development of alcoholic liver diseases [11]. Although immune system involvement in ETOH-associated liver diseases is suggested from these lines of evidence, it is still unclear whether ETOH-induced immunological disturbances are involved in initiation and progression of liver disease.

Fatty liver is one of the earliest consequences of alcohol abuse by human beings and may be a precursor of more severe liver lesions, such as acute inflammation or ‘alcoholic hepatitis’ and hepatocellular necrosis, fibrosis and cirrhosis [12–16]. Steatosis is the most consistent finding in animal models of alcoholic liver injury. Lieber and DeCarli [17] demonstrated the development of fatty liver after 1–2 months and fibrosis after 1–2 years in baboons fed an adequate diet with 50% of calories supplied by ETOH. Rats fed an ETOH-containing diet (36% of total calories) had an accumulation of fat in the liver after 10 days of feeding, with an even more pronounced accumulation after 24 days of ETOH consumption [18]. In a rat model of continuous intragastric infusion of ETOH Tsukamoto et al. [15] demonstrated steatosis in the livers of rats after 15 days, and significantly elevated liver triglyceride levels after 30 and 85 days of ETOH infusion. About one-third of animals in their study showed remarkably severe steatosis and focal necrosis with mononuclear cell infiltration. The model of Tsukamoto et al. requires constant feeding by infusion of an ETOH diet containing high levels of fat.

In models that involve mice as the experimental subject of ETOH consumption, fatty change in the liver is rarely seen. Results of two reports have shown that in mice fed an ETOH diet providing 30% or 36% of calories a mild microvesicular steatosis ultimately developed, but a relatively long feeding time (28–30 days) was required for its development [19,20]. In contrast to these study findings, it has been shown that mice provided an ETOH-containing diet given by a continuous, intragastric technique develop a severe steatosis and a modest increase in serum levels of alanine aminotransferase (ALT) after 30 days of feeding [21].

The relationship between liver steatosis and inflammation in alcoholic liver disease is still obscure. The hepatocyte damage associated with fatty liver induced by ETOH intake might be due to increased free fatty acid levels; free fatty acids are highly cytotoxic and can damage mitochondrial and lysosomal membranes, and it is clear that free-radical production associated with ETOH metabolism and lipid peroxidation is also pathogenic [12–14,16,22]. These increases in free fatty acid levels and lipid peroxidation, however, do not seem to occur in murine models of ETOH consumption. It is unclear why ETOH consumption does not in and of itself uniformly result in hepatitis, even when steatosis is present. Together, these observations lead us to suggest that factors in addition to ETOH consumption are necessary for development and progression of alcoholic liver disease.

Although a number of investigators have suggested a role for immune mechanisms in alcoholic liver diseases, no definitive experiments have been performed to support this suggestion. Studies performed in this laboratory to investigate pathological changes in the liver after infection of ETOH-consuming mice with a facultative intracellular bacterium (Listeria monocytogenes) have produced new data showing an obligatory role for an immune response in the rapid development of alcoholic fatty liver and hepatitis in a murine model of ETOH consumption. The findings presented in this paper provide for the first time a rodent model system of reproducible alcoholic liver disease without the need for an extraordinarily long exposure to ETOH or use of a high-fat diet.

Materials and methods

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

Mice and ETOH administration

Female, specific-pathogen-free, C57Bl/6 mice (6–8 weeks of age) were obtained from Charles Rivers through a contract with the National Cancer Institute (Frederick, MD, USA). They were allowed to acclimate in our facility for at least 1 week after arrival and were given a diet of laboratory chow and water ad libitum during this period. All animals received humane care in compliance with the criteria set forth in the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, Commission on Life Sciences, National Research Council, 1996). Animals were assigned randomly to one of two diet groups on the basis of feeding with a Lieber–DeCarli high-protein liquid diet (Dyets Inc., Bethlehem, PA, USA): one containing 36% ETOH-derived calories (E group) or one made isocaloric with dextrin and maltose with a pair-feeding protocol (PF group), as described previously [23]. The liquid diets provided 25% of calories as amino acids, 63% as carbohydrates and 12% as fat. Mice continued to receive their respective diets until they were studied.


The EGD strain of L. monocytogenes was obtained from Dr E. Havel (Trudeau Institute, Lake Sarnac, NY, USA). Bacteria for immunizations and challenges were grown in trypticase soy broth, and organisms in log phase of growth were frozen and stored at − 70°C in 1-ml aliquots. The number of bacteria in each preparation was determined by means of a colony-forming unit count in triplicate on trypticase soy agar containing 5% sheep blood. Tenfold dilutions were prepared in sterile, ice-cold saline, and 0·1-ml aliquots were spread on blood agar plates. After the plates were incubated at 37°C for 24–48 h, the number of bacteria was determined. To ensure appropriate numbers of bacteria were injected, the number of viable organisms in each bacterial dilution for injection was confirmed by culturing a sample of each dilution on blood agar plates, as above. This also ensured that the organisms were haemolytic and thus virulent. The median lethal dose (LD50) in this strain of mice was 5 × 104 bacteria. To prepare heat-killed L. monocytogenes (HKLM) antigen, bacteria were cultured in trypticase soy broth at 37°C overnight. Next, bacteria were pelleted by centrifugation (at 1000 g for 30 min) and the resulting organisms were washed once with phosphate-buffered saline and heated at 56°C for 120 min. After bacteria were washed with phosphate-buffered saline, they were resuspended in pyrogen-free water for injection, dried overnight in an evaporating centrifuge (SpeedVac, Jouan, Winchester, VA, USA) and, finally, resuspended in pyrogen-free saline to achieve an antigen concentration of 10 mg dry weight per milliliter.

Immunization and challenge of mice with viable bacteria

Mice were immunized intravenously with 2·5 × 103 colony-forming units of L. monocytogenes in 0·1 ml of pyrogen-free saline and then began receiving experimental diets 7–9 days after immunization. Immune and non-immune mice were challenged intravenously with 2·5 × 104 colony-forming units (1/2 LD50) in 0·1 ml of pyrogen-free saline.

Neutralization of tumour necrosis factor

To determine the effects of TNF in the steatosis and liver damage associated with administration of heat-killed L. monocytogenes antigen animals were administered antisera with specificity to TNF before antigen administration essentially as described by others [24,25], with anti-TNF and an isotype control sera generously provided by Dr Greg Bagby (Louisiana State University Medical Center, New Orleans). In experiments where steatosis and liver damage were induced in immune animals by injection of listeria antigen, ethanol-fed mice were injected with 0·1 ml of anti-TNF or an equal volume of the isotype control serum 24 h before administration of antigen.

Levels of TNF in sera were determined with the use of an ELISA (OptiELISA, Pharmingen, San Diego, CA, USA) according to the manufacturer's instructions.

Chemical assays for serum levels of ALT, triglycerides and total lipids

Serum ALT levels were determined by using a kit (Sigma ALT Kit [59-10]) obtained from Sigma Chemical Co. (St Louis, MO, USA). Animals were killed humanely by exsanguination after anaesthetization with CO2, and livers were removed. Liver homogenates (10% wt/vol) were prepared in sterile phosphate-buffered saline by using a motorized homogenizer (Pro 200 Homogenizer; Pro Scientific Inc., Monroe, CT, USA). To increase the efficiency of fat extraction from cells and obtain a homogeneous suspension of lipids, the homogenized specimens were sonicated for 30 s by using a VirSonic 300 Sonicator (Virtis Company, Inc., Gardiner, NY, USA). All preparatory steps were performed at 4°C. Triglyceride levels in the liver homogenates were determined by using a kit (Sigma Triglyceride Kit [339–20]) obtained from Sigma Chemical Co. Total lipid concentration of each liver homogenate was determined with the use of the sulfophosphovanillin reaction [26,27]. The total lipid concentration was calculated from the absorbance values of standards prepared by dissolving canola oil in ETOH. Triglyceride and lipid levels were expressed as milligrams per gram of liver tissue.

Tissue preparation

Liver tissue was fixed in 10% neutral buffered formalin for at least 24 h and embedded in JB-4 plastic (Polysciences, Inc., Warrington, PA, USA). Sections (3 µm) were stained with haematoxylin–eosin stain. Histological samples to determine the effects of anti-TNF were fixed in 10% buffered formalin, embedded in paraffin, and 5-µm sections cut. Sections were stained with haematoxylin–eosin stain. To prepare tissue to stain for fat, tissue was fixed in 2% paraformaldehyde, embedded in OCT compound (Miles Inc., Elkhart, IN, USA) and frozen with liquid nitrogen. Frozen sections were prepared and stained for fat with Oil Red O stain.

Statistical analysis

A one-way analysis of variance in association with the Student–Newmann–Keuls multiple comparison test was used to determine statistical significance of data. Differences at a P-value < 0·05 were considered statistically significant.


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

Mice immune to L. monocytogenes were challenged with 2·5 × 104 colony-forming units of live bacteria after 3 days of feeding with ETOH or control diet. This timing was chosen on the basis of previous studies that show the response of ETOH-fed mice to primary and recall infections [28]. Histological evaluation of livers obtained at various times after bacterial challenge of immune E group mice showed development of fatty liver. The steatosis was first evident 24 h after the challenge and progressed in severity through 3–5 days after infection (Fig. 1a–d). Macrovesicular fat droplets were predominant and evident in the midzonal area. Granulomas and focal areas of necrosis were also seen. Intracellular lipid deposits in E group mice were confirmed by staining with Oil Red O stain (Fig. 2). After 7 days of infection (10 days of ETOH feeding), less severe inflammation and fatty changes were observed in the livers of E group mice (Fig. 1d) than that seen at earlier times. The histological changes 3 and 5 days after infection were similar in both groups of mice, and only the photomicrographs obtained 3 days after infection are shown for convenience.


Figure 1. Effect of ethanol (ETOH) consumption by mice immune to Listeria monocytogenes on accumulation of fat and hepatic inflammation after a second challenge with L. monocytogenes. Female C57Bl/6 mice (6–8 weeks of age) were immunized by intravenous injection with 2·5 × 103 colony-forming units (CFU) of L. monocytogenes. After 7–9 days, animals were assigned randomly to one of two groups: one receiving an ETOH-containing liquid diet (E group) or one pair-fed an isocaloric control liquid diet (PF group). After 3 days of feeding, each animal was challenged with an intravenous injection of 2·5 × 104 CFU of L. monocytogenes, and the animals were maintained on the appropriate diet for the indicated times. At the indicated times after infection, animals in each group were killed humanely by exsanguination, and the livers were removed. Liver samples were fixed in 10% buffered formalin and embedded in plastic. Thin (3-µm) sections of the samples were prepared and stained. (a–d) Photomicrographs of sections of livers obtained from E group mice at the time of challenge, as well as after 3 days of feeding ETOH diet (a) or 1, 3 and 7 days after infection (b–d). (e–h) Photomicrographs of sections of livers obtained from PF group mice at the same time points. Photomicrographs presented in this figure are representative of five experiments (haematoxylin–eosin stain; final magnification ×170).

Download figure to PowerPoint


Figure 2. Histological demonstration of lipid in livers obtained from mice immune to Listeria monocytogenes after a second bacterial challenge. Immune mice either fed a diet containing ETOH or pair-fed control diet were challenged with L. monocytogenes, as described in legend for Fig. 1. Three days after the secondary infection, livers were obtained. Then to prepare tissue to stain for fat, tissue was fixed in 2% paraformaldehyde, embedded in OCT compound (Miles Inc., Elkhart, IN, USA) and frozen with liquid nitrogen. Frozen sections were prepared and stained for fat with Oil Red O stain. (a) and (b) Photomicrographs of a section of liver obtained from a PF group mouse and an E group mouse, respectively. (c) Photomicrograph of a section of liver obtained from an E group mouse but at a higher power view to show details of the stained droplets. Photomicrographs are representative of six experiments (final magnification [a and b]×200; and [c]×400).

Download figure to PowerPoint

The response of immune PF group mice was not associated with accumulation of obvious fat droplets in the liver at any time after the challenge infection with L. monocytogenes (Fig. 1e-h). A moderate depletion of glycogen was noted in liver sections obtained from some PF group mice. Granulomas and areas of necrosis were also noted in PF group mice. However, the granulomas were quantitatively fewer in number (data not shown) and appeared to be smaller.

The influence of the diets on hepatic damage associated with the bacterial challenge of immune mice was determined by measuring serum ALT levels in each group of experimental animals after various times of infection. The serum ALT levels (Fig. 3) were higher at all times in the animals consuming ETOH and statistically different from the values obtained from the PF group animals at 1, 3 and 5 days after infection.


Figure 3. Effect of ethanol consumption by mice immune to Listeria monocytogenes on serum alanine aminotransferase (ALT) levels after a second infection with L. monocytogenes. Sera samples were obtained from mice treated as described in the legend for Fig. 1 and assayed for ALT levels with the use of a kit (Sigma ALT Kit [59-10], Sigma Chemical Co., St Louis, MO, USA). Each point represents the arithmetic mean value of 3–7 mice ± s.e.m. of the E group (●) or the PF group (▪). Data presented in this figure are representative of three experiments. *E significantly different from PF at P ≤ 0·05; **E significantly different from PF at P ≤ 0·01.

In contrast to the response of immune mice, non-immune mice showed only mild steatosis 24 h after infection with a sublethal dose of L. monocytogenes; this was seen only in the non-immune E group mice (data not shown). Animals in the E group showed a progressive listerial infection of the liver that was associated with increased serum ALT levels (data not shown), which is consistent with earlier reported findings [28]. Steatosis was not seen in the E group after the first 24 h of infection, nor at any time in the PF group.

The formation of fatty liver and hepatic inflammation in the E group seemed to be associated with a recall immune response in the liver. To address this possibility, mice that were immunized as before with viable L. monocytogenes were assigned to either the E group or the PF group. After 7 days of diet consumption, mice were injected intravenously with 1 mg of HKLM antigen. Previous study results show that the maximal suppression of the immune system occurs after 7 days of feeding in this model [23], and it was believed that this time would provide a more reasonable model of immunopathological effects in the light of the well-known immunosuppressive effects of ETOH. Injection of antigen induced a rapid accumulation of fat in the liver of immune animals in the E group (Fig. 4). Quantitative determination of lipid levels in liver homogenates showed an increase in fat, and especially in triglycerides, in the E group animals at 2 and 6 h after antigen injection (Table 1). Serum ALT levels were higher in the E group animals, compared with levels in the PF group animals after HKLM antigen injection (Fig. 5). By 24 h after HKLM antigen injection in the E group, the liver steatosis was resolved. However, the serum ALT levels remained elevated. At this time, inflammatory cell infiltration and areas of hepatic necrosis were seen in sections of livers obtained from immune E group mice, but not in sections of livers obtained from PF group mice (Fig. 4). As controls, immune mice from E and PF groups were injected with phosphate-buffered saline, and non-immune E and PF group animals were injected with HKLM antigen. Neither steatosis (Table 1) nor increased serum ALT levels (Fig. 6) were seen in any of these control groups. Also, injection of immune E and PF group mice with lipopolysaccharide (10 µg per mouse) did not result in elevated serum ALT levels (Fig. 6).


Figure 4. Effect of ethanol consumption on accumulation of fat and inflammatory responses of mice immune to Listeria monocytogenes after injection of heat-killed L. monocytogenes (HKLM) antigen. Female C57Bl/6 mice were immunized with viable L. monocytogenes as described in the legend for Fig. 1. After 9 days, animals were assigned randomly to the E or the PF group. After 7 days of feeding, each animal was injected intravenously with 1 mg of HKLM antigen. At the indicated times after injection, animals were killed humanely by exsanguination. The livers were then removed and histological analyses were performed as described above. Photomicrographs presented in this figure are representative of five to seven experiments. (a–c, g, h) Photomicrographs of sections of livers obtained from E group mice before administration of HKLM (a) and 6 and 24 h after antigen administration (b, c). (g) Higher-power view of inflammatory cells noted 6 h after injection of HKLM, (h) higher-power view of a granuloma noted in a section obtained from a liver 24 h after injection of HKLM. (d–f) Representative sections of livers obtained from PF group mice at the same times as those shown for E group mice (final magnification [a–f]×180; [g]×400; and [h]×560.)

Download figure to PowerPoint

Table 1.  Effect of injection of heat-killed Listeria monocytogenes antigen on total lipids and triglyceride content in livers of E and PF groups
 Total lipids (mg/g of liver)Triglycerides (mg/g of liver)
  1.   Results are means ± s.e. of three independent experiments. E group denotes animals fed a Lieber–DeCarli high-protein diet containing 36% ethanol-derived calories; HKLM denotes heat-killed Listeria monocytogenes antigen; PBS denotes phosphate-buffered saline; PF group denotes animals fed a Lieber–DeCarli high-protein diet made isocaloric with dextrin and maltose (control diet) with a pair-feeding paradigm. *Intravenous injection with 1 mg of HKLM antigen. P < 0·01 E group versus PF group; ‡Intravenous injection with 0·1 ml of PBS as control; §P < 0·05 E group versus PF group.

2 h after injection
 Immune PF group – HKLM*29·4 ± 2·920·5 ± 1·7
 Immune E group – HKLM*66·8 ± 12·048·6 ± 7·0
 Immune PF group – PBS40·6 ± 2·114·9 ± 1·8
 Immune E group – PBS36·1 ± 5·319·8 ± 5·1
 Immune PF group – NONE39·1 ± 8·912·5 ± 1·4
 Immune E group – NONE40·0 ± 8·120·9 ± 7·5
6 h after injection
 Immune PF group – HKLM*23·9 ± 5·023·0 ± 3·0
 Immune E group – HKLM*54·0 ± 1·338·0 ± 2·4§
 Naive PF group – HKLM*32·4 ± 2·318·8 ± 0·5
 Naive E group – HKLM*31·3 ± 7·418·7 ± 2·2
24 h after injection
 Immune PF group – HKLM*30·5 ± 2·614·9 ± 2·1
 Immune E group – HKLM*30·8 ± 5·618·0 ± 2·5
 Naive PF group – HKLM*30·4 ± 3·812·3 ± 1·1
 Naive E group – HKLM*42·3 ± 1·918·5 ± 3·9

Figure 5. Effect of ethanol consumption by mice immune to Listeria monocytogenes on serum alanine aminotransferase (ALT) levels after injection with heat-killed L. monocytogenes (HKLM) antigen. Female C57Bl/6 mice were immunized with viable L. monocytogenes as described in the legend for Fig. 1. After 9 days, animals were assigned randomly to either the E or the PF group. After 7 days of feeding, each animal was injected intravenously with 1 mg of HKLM antigen. At the indicated times, sera samples were obtained from four to eight animals from the E group (●) and PF group (▪), and serum ALT levels were determined as described in the legend for Fig. 3. Each point represents the arithmetic mean ± s.e.m. *E group significantly different from PF group at P ≤ 0·001. Data presented in this figure are representative of three experiments.

Download figure to PowerPoint


Figure 6. Effect of ethanol consumption by mice immune to Listeria monocytogenes or by non-immune mice on serum alanine aminotransferase (ALT) levels after injection of heat-killed L. monocytogenes (HKLM) antigen or bacterial lipopolysaccharide (LPS). Female C57Bl/6 mice immunized as described in the legend for Fig. 1 or non-immune age-matched mice were assigned randomly to either the E or the PF group. After 7 days of feeding, animals in each group were injected intravenously with 1 mg of HKLM antigen, 10 µg of Escherichia coli LPS, or 0·1 ml of diluent (phosphate-buffered saline; PBS). Sera samples were obtained 24 h after injection and assayed for ALT levels as described in the legend for Fig. 3. Each point represents the arithmetic mean ± s.e.m. of 4–10 mice and is representative of two experiments.

It was found that injection of HKLM antigen induced detectable levels of TNF 4 h after injection determined with the use of a specific ELISA in the sera of all immune mice injected with HKLM antigen. Serum levels of TNF were higher in the ETOH-fed mice injected with HKLM antigen compared with the findings for pair-fed mice (857 ± 112 pg/ml vs. 650 ± 84 pg/ml, respectively). To investigate the possible role of TNF in the steatosis and liver damage associated with the recall immune response in ETOH-fed mice, neutralizing anti-TNF was injected 24 h before injection of antigen. It was found that pretreatment with anti-TNF markedly, but not completely, reduced steatosis compared with the effect of the control serum (Fig. 7a,c). Interestingly, administration of anti-TNF did not have an apparent effect on the infiltration of mononuclear cells into the liver (Fig. 7c). Anti-TNF treatment reduced serum levels of ALT by 43% measured 24 h after injection (1680 ± 210 U/l vs. 963 ± 127 U/l in the mice injected with control serum and anti-TNF, respectively). The levels of TNF in the sera obtained from the anti-TNF-treated mice determined 4 h after injection of antigen were below the levels of detection of the ELISA (15 pg/ml), and the ETOH-fed mice injected with the control serum had levels of TNF comparable to those of ETOH-fed animals injected with antigen (730 pg/ml). As expected, injection of the pair-fed control mice with anti-TNF had no detectable effect on the biochemical or histological effects of HKLM antigen injection (not shown)


Figure 7. Effect of administration of a neutralizing anti-TNF antibody on steatosis and inflammatory cell infiltration after injection of heat-killed L. monocytogenes antigen to immune C57Bl/6 mice fed ethanol. Female C57Bl/6 mice were immunized and fed the ethanol-containing diet for 7 days and administered heat-killed antigen as described in the legends for Figs 1 and 6. One hour before antigen injection one group of animals was injected with 0·1 ml of control serum by an intraperitoneal route (a,b) or injected with anti-TNF serum (c). Liver tissue was obtained from animals in each group 24 h after antigen injection, fixed in 10% buffered formalin and sections stained with haematoxilin and eosin. Large arrowheads indicate areas of macrovesicular steatosis (a) or microvesicular steatosis (c). Inflammatory cells are indicated by large arrows. Magnification in (a) and (c) = ×200 and in (b) = ×400.

Download figure to PowerPoint


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

We have presented evidence that the combination of a sustained recall immune response and ETOH consumption results in consistent production of fatty changes and inflammation in the liver of experimental animals. The protocols we have presented establish for the first time a murine model system of ETOH-induced fatty liver and hepatic inflammation. It is noteworthy that these changes occur after a relatively short period of ETOH consumption by an animal species that is resistant to changes in the liver due to ETOH consumption only.

Alcoholic liver disease is a well-described pathological consequence of ETOH abuse, but the mechanisms involved in the observed pathological changes are not well established. That obvious liver disease [29] develops in only a minority of alcoholics is difficult to understand in light of current knowledge regarding these mechanisms. One possible factor in the observed differences in individual susceptibility to alcoholic liver disease is the potential need for an immune response to a co-existing viral or bacterial infection.

Several investigators have proposed that hepatocyte damage can be mediated by a number of factors. These include toxic lipid peroxidation products [30] and proinflammatory cytokines such as TNF [6], as well as an influx of inflammatory cells induced by proinflammatory cytokines, including TNF, IL-6 and IL-8 [1,2,4–6,10,31]. The specific generation of an immune response to acetaldehyde–protein adducts or modified membrane proteins is another proposed mechanism [7–9,32]. The totality of the pathological condition known as alcoholic liver disease is likely to be complex and results from a combination of the above-listed mechanisms.

We believe the findings presented in this paper establish that at least one powerful mechanism of fatty liver and hepatic inflammation associated with alcoholic liver disease is a recall immune response to pathogenic microorganisms in conjunction with ETOH consumption. This suggestion is supported by data obtained from studies with the use of a transgenic strain of mice expressing the envelope proteins of the hepatitis B virus [33]. In this model, transfer of antigen-specific cytotoxic (CD8+) T cells resulted in severe liver damage and ultimately liver failure [33]. These findings demonstrate the potential of the immune system to initiate severe liver disease.

Further support of our hypothesis comes from study results that show that human beings who are infected with hepatitis C virus and abuse alcohol have a more severe hepatitis and a greater viraemia than non-alcoholic individuals [34], and these effects resolve during abstinence [34]. It is relatively clear that the infection with hepatitis C virus results in a population of antigen-specific CD8+ T cells and it has been proposed that the pathological effects of hepatitis C virus infection are the result of a cytotoxic T-cell response to the infected hepatocytes [35–39]. It is our contention that the liver damage associated with a recall immune response in the model described in this paper is the result of a similar activation of antigen-specific CD8+ T cells that ultimately migrate to the liver and through undefined mechanisms damage the liver.

In our model of ETOH-associated pathological changes in the liver there is a requirement for a recall immune response, which is provided by an immune response to the facultative intracellular bacterium L. monocytogenes. Together, the innate and antigen-specific immune responses induced by infection with L. monocytogenes kill the bacteria and resolve the infection relatively quickly. The specific immune response results in T-cell production of cytokines, including interferon-gamma, and these cytokines activate macrophages to a bactericidal state [40]. It has been shown that antigen-specific CD8+ T cells are the major cell type that mediate the ultimate clearance of bacteria from the host infected with L. monocytogenes[41,42], probably through a number of mechanisms to include production of interferon-gamma and lysis of the infected cells in an antigen-specific, MHC class I-restricted response [43]. Although granulocytes, T cells and macrophages are all important cells for resistance to a primary infection, antigen-specific memory T cells are critical for resistance to a secondary challenge with this organism [40,44,45]. Again, the predominant cell type involved in the recall immune response to this organism is the CD8+ T cell [41]. A number of cytokines are characteristically produced in response to a primary or secondary infection by L. monocytogenes. These cytokines are produced by Kupffer cells, inflammatory macrophages, T cells and other lymphoid cells, hepatocytes and endothelial cells [46]. The study results we present do not provide sufficient information to propose a complete mechanism of steatosis and hepatitis associated with a secondary stimulation of the immune system. However, it is probable, on the basis of our studies with anti-TNF (Fig. 7), that cytokines such as TNF produced during either the early non-specific inflammatory response or the specific immune response play a role.

Cytokines produced as a result of infection, inflammation, trauma or, importantly, a Th-1-type of T-cell-mediated immune response are required not only for co-ordinated immune responses but can also induce metabolic changes. Study findings have shown that severe liver steatosis does not occur in TNF receptor-1 knock-out mice [21], which supports the suggestion for a role of TNF in the regulation of lipogenesis/lipolysis or clearance of lipids from the liver. Also, results of in vivo studies have demonstrated that TNF and IL-6 stimulate lipolysis in adipose tissue and increase the free fatty acid supply to the liver. The cytokines IL-1, TNF and IL-6 induce hypertriglyceridaemia, as well as increase both the secretion of triglycerides by the liver and the de novo synthesis of free fatty acids in the liver [47–49]. At the same time, TNF increases the activity of lipoprotein lipase, which is the enzyme involved in the clearance of triglyceride-rich lipoproteins. Results of in vivo studies in rats fed ETOH [49] and in vitro studies with hepatocyte cultures [50] have shown the ability of ETOH to inhibit the secretion of triglycerides in very-low-density lipoprotein particles. In addition to an impaired secretion of triglycerides, ETOH could lead to decreased liver lipolysis [12,51]. Therefore, infection and the associated Th1-type cytokine production in animals consuming ETOH could induce an increase in triglyceride production. Because of the impaired export of triglycerides, decreased lipolysis or both, however, fat accumulates in the liver.

It has become increasingly clear that steatosis is associated with pathogenic changes in the liver that ultimately produce inflammatory changes that ultimately result in liver damage [12,16,52]. It is generally accepted that steatosis alone is insufficient for the progression to inflammation and that there is a requirement for another factor or ‘second hit’ for the induction of inflammation [16,52]. There is evidence that alcoholic liver disease, including hepatitis and fibrosis, is associated with lipid peroxidation that results from reactive oxygen species and other free radicals produced by metabolism of ETOH, and it has been suggested that this provides the second hit [13,52]. These free radicals are thought to be the result of P450 2E1 metabolism [53], changes in mitochondrial metabolism [14,22] or the production of hydrogen peroxide by β-oxidation of free-fatty acids [13].

There are also other potential co-factors that have been proposed that could produce inflammation in conditions that produce fatty changes in the liver. In the model of non-alcoholic steatohepatitis that has been developed by Diehl and colleagues that uses the genetically obese mouse strain (ob/ob) it has been shown that fatty liver is associated with changes in lymphocyte populations in the liver that are proposed to establish a situation where the liver is sensitized to damage by a number of stimuli [54] analogous to the stituation where the liver is sensitized to endotoxin by Propionibacterium acnes. Of importance to the interpretation of the data presented in this paper are the changes in cytokine regulatory networks in the liver in response to inflammatory stimuli that ultimately result in excess production of Th-1 inflammatory hepatotoxic cytokines [55–57]. It is possible, and perhaps probable, that similar changes are occurring in the ETOH-fed mice used in this study that resulted in an over-production of inflammatory cytokines in the liver by the activated CD8+ T cells. The changes in cell populations in P. acnes-treated mice also sensitize the hepatocytes to perforin and FAS lysis of hepatocytes [57–59], and FAS lysis of hepatocytes by activated T cells could explain the incomplete reduction in ALT in mice treated with anti-TNF in the study results presented in this paper. It appears that neutralization of TNF almost abrogates steatosis in immune mice challenged with specific antigen (Fig. 7) but did not completely abrogate liver damage, as assessed by serum levels of ALT, nor the infiltration of mononuclear cells into the liver (Fig. 7). As will be discussed in more detail in the following paragraphs, we believe the liver damage in this model of alcoholic liver disease is the result of an interaction of activated lymphocytes with hepatocytes. It is certainly possible that the steatosis that develops as a result of the recall immune response to listeria antigen in the ETOH-fed mice would in time result in hepatitis, but this would be relatively slow and require a ‘second hit’ of some kind. Our data would support the suggestion that steatosis that develops after a recall antigen administration to ETOH-fed mice is a very rapidly developing event induced by proinflammatory cytokines produced by activated T cells and that these cytokines are also, in part, involved in the hepatotoxicity noted in these studies.

It is clear from the observed serum ALT levels and histological findings (Figs 1 and 2) that a specific infectious challenge of immune E group mice was associated with a greater inflammatory response and more liver damage. This was also evident when immune E group mice were injected with HKLM antigen (Figs 4, 5, and 6). These findings support our suggestion that ETOH consumption is associated with an exaggerated hepatic inflammation after a recall immune response. The inability of endotoxin to produce changes in the liver supports this suggestion, and indicates that the profound changes in lymphocyte populations and Kupffer cell production of cytokines such as IL-12 and IL-18 associated with sensitization of the liver by P. acnes do not occur by ETOH feeding alone. On the basis of the data obtained after neutralization of TNF, it is likely that this inflammation is partially in response to the cytotoxic effects of TNF, as well as cytokines that induce the infiltration of inflammatory cells into the liver.

It is our working hypothesis that the CD8+ T cells that are activated in the recall immune response to either infection or in response to heat-killed antigen mediate, at least in part, the damage in the liver. In the situation where immune animals were infected with viable bacteria it is likely that the liver damage is caused in part by antigen-specific T cells that traffic to the liver and, as shown by others, recognize infected hepatocytes and lyse them [43]. As will be discussed in the next section, it is also likely that the CD8+ T cells that are activated in peripheral lymphoid organs, after injection of antigen, also traffic to the liver and also lyse hepatocytes similar to the finding of Russell et al. [60], when hepatitis could be produced in mice transgenic for ovalbumin. There have been a number of studies that show that activated CD8+ T cells migrate to the liver where, in normal conditions, they die through apoptosis [60–62]. It is clear that antigen-specific CD8+ T cells are responsible for long-term memory and resistance to reinfection in experimental listeriosis [40,41,44]. Therefore it is our contention that infection of immune mice activates the CD8+ T cells and these cells traffic to the liver and lyse hepatocytes. It appears that the hepatic inflammation associated with activation of memory T cells in the model system described in this paper is predominantly a mononuclear cell infiltration, which would support our contention that liver damage is mediated by activated lymphocytes, and perhaps activated macrophages. It must be noted that this is in contrast to alcoholic hepatitis in human beings, which is characterized by a predominant granulocyte inflammatory response. As discussed earlier, we propose that our study findings provide a model of CD8+ T-cell-mediated hepatotoxicity analogous to liver damage associated with hepatitis C virus infection of human beings who abuse alcohol. The mechanisms that activated CD8+ T cells use lyse hepatocytes include the production of cytotoxic cytokines such as TNF, as well as the more direct lytic mechanisms of perforin and FAS-mediated lysis through FAS-ligand expression on the activated CD8+ T cell [63–67]. From the data presented in this paper we suggest that the initial lysis of hepatocytes that are infected by relatively high doses of L. monocytogenes, as described by Saad, Domiati-Saad and Jerrells [28], by activated CD8+ T cells ultimately induces an inflammatory response that results in marked damage to the liver. A similar result occurs when immune mice are challenged with heat-killed antigen. However, in this situation the majority, if not all, of the antigen-responsive cells would likely be activated in peripheral lymphoid organs such as the spleen and then traffic to the liver where they directly kill hepatocytes regardless of whether they are infected or not. There are at least two concerns that must be addressed for this latter suggestion to be tenable. First, it has been shown that the predominant epitope for the antigen-responsive CD8+ T cells that arise in listeriosis is the secreted lysteriolysin O protein [68–70], which would not be present in the heat-killed antigen. However, there are at least three other epitopes of L. monocytogenes that activate CD8+ T cells [69] that might be present in the antigen preparation used for these studies. Further, it has also been shown that essentially any Th-1-type cell response can damage the liver [71] probably through direct effects of inflammatory cytokines, through metabolic effects of the inflammatory cytokines, or both, and the heat-killed antigen would activate this type of response. The possible role of activated CD4+ T cells in the noted phenomena requires further study. The second concern that must be considered is the restriction of obvious liver damage to the ethanol-fed mice regardless of how the CD8+ T cells are activated. One possible mechanism of this phenomenon is that the hepatocyte is somehow more sensitive to the lytic mechanisms active in this model. One explanation for this proposed sensitization of hepatocytes to immune lytic mechanisms is the presence of steatosis and the resulting mitochondrial changes associated with fatty changes in ETOH-fed animals [12,14,16,52,72,73]. There are, however, myriad other possible co-factors that need to be evaluated as well to include the possible role of corticosteroids and proinflammatory cytokines other than TNF, such as interferon-gamma, IL-12 and IL-18.

In summary, we have found that a recall immune response to viable L. monocytogenes or listeria antigen is associated with reproducible fatty liver and hepatitis in mice consuming a liquid diet containing ETOH. In this model, alcoholic fatty liver and hepatitis occur after consumption of an ETOH-containing diet and require a recall immune response. The pathological changes occur rapidly in an animal that is ordinarily resistant to liver damage by ETOH alone, and the model does not require the feeding of a high-fat diet. The production of TNF is at least partially responsible for the liver damage noted in the model system described in this paper and is a major factor in the production of the noted steatosis.


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

This work was supported by the National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health grants AA07731, AA12450 and KO2 AA00129. The authors thank Dr Wen Ma and Dr Robert Specian of the Histology Core Unit at Louisiana State University Medical Center – Shreveport for technical assistance; Jennifer Jerrells, MS, for assistance with the anti-TNF studies and Janice Jerrells, RN, BA, ELS, for editorial assistance with the manuscript.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    Bode C, Kugler V, Bode JC. Endotoxemia in patients with alcoholic and non-alcoholic cirrhosis and in subjects with no evidence of chronic liver disease following acute alcohol excess. J Hepatol 1987; 4:814.
  • 2
    Deviere J, Content J, Denys C et al. High interleukin-6 serum levels and increased production by leucocytes in alcoholic liver cirrhosis. Correlation with IgA serum levels and lymphokines production. Clin Exp Immunol 1989; 77:2215.
  • 3
    Hill DB, Marsano LS, McClain CJ. Increased plasma interleukin-8 concentrations in alcoholic hepatitis. Hepatology 1993; 18:57680.
  • 4
    Khoruts A, Stahnke L, McClain CJ, Logan G, Allen JI. Circulating tumor necrosis factor, interleukin-1 and interleukin-6 concentrations in chronic alcoholic patients. Hepatology 1991; 13:26776.
  • 5
    Ledesma Castano F, Echevarria Vierna S, Lozano Polo JL, Oloriz Rivas R, Alvarez Moreno C, Pons Romero F. Interleukin-1 in alcoholic cirrhosis of the liver: the influence of nutrition. Eur J Clin Nutr 1992; 46:52733.
  • 6
    McClain CJ & Cohen DA. Increased tumor necrosis factor production by monocytes in alcoholic hepatitis. Hepatology 1989; 9:34951.
  • 7
    Perperas A, Tsantoulas D, Portmann B, Eddleston AL, Williams R. Autoimmunity to a liver membrane lipoprotein and liver damage in alcoholic liver disease. Gut 1981; 22:14952.
  • 8
    Anthony RS, Farquharson M, MacSween RN. Liver membrane antibodies in alcoholic liver disease. II. Antibodies to ethanol-altered hepatocytes. J Clin Pathol 1983; 36:13028.
  • 9
    Hoerner M, Behrens UJ, Worner TM et al. The role of alcoholism and liver disease in the appearance of serum antibodies against acetaldehyde adducts. Hepatology 1988; 8:56974.
  • 10
    Adams DH. Leucocyte adhesion molecules and alcoholic liver disease. Alcohol Alcohol 1994; 29:24960.
  • 11
    Chedid A, Mendenhall CL, Moritz TE et al. Cell-mediated hepatic injury in alcoholic liver disease. Veterans Affairs Cooperative Study Group 275. Gastroenterology 1993; 105:25466.
  • 12
    Day CP & Yeaman SJ. The biochemistry of alcohol-induced fatty liver. Biochim Biophys Acta 1994; 1215:3348.
  • 13
    Misra UK, Bradford BU, Handler JA, Thurman RG. Chronic ethanol treatment induces H2O2 production selectively in pericentral regions of the liver lobule. Alcohol Clin Exp Res 1992; 16:83942.
  • 14
    Rashid A, Wu TC, Huang CC et al. Mitochondrial proteins that regulate apoptosis and necrosis are induced in mouse fatty liver. Hepatology 1999; 29:11318.
  • 15
    Tsukamoto H, French SW, Benson N et al. Severe and progressive steatosis and focal necrosis in rat liver induced by continuous intragastric infusion of ethanol and low fat diet. Hepatology 1985; 5:22432.
  • 16
    Tilg H & Diehl AM. Mechanisms of disease: cytokines in alcoholic and nonalcoholic steatohepatitis. N Engl J Med 2000; 343:146776.
  • 17
    Lieber CS & DeCarli LM. An experimental model of alcohol feeding and liver injury in the baboon. J Med Primatol 1974; 3:15363.
  • 18
    Lieber CS, Jones DP, DeCarli LM. Effects of prolonged ethanol intake: production of fatty liver despite adequate diets. J Clin Investigation 1965; 44:100921.
  • 19
    Holsapple MP, Eads M, Stevens WD et al. Immunosuppression in adult female B6C3F1 mice by chronic exposure to ethanol in a liquid diet. Immunopharmacology 1993; 26:3151.
  • 20
    Smith SM & Hoy WE. Ad libitum alcohol ingestion does not induce renal IgA deposition in mice. Alcohol Clin Exp Res 1990; 14:1846.
  • 21
    Yin M, Wheeler MD, Kono H et al. Essential role of tumor necrosis factor alpha in alcohol-induced liver injury in mice. Gastroenterology 1999; 117:94252.
  • 22
    Berson A, De Beco V, Letteron P et al. Steatohepatitis-inducing drugs cause mitochondrial dysfunction and lipid peroxidation in rat hepatocytes. Gastroenterology 1998; 114:76474.
  • 23
    Jerrells TR, Smith W, Eckardt MJ. Murine model of ethanol-induced immunosuppression. Alcohol Clin Exp Res 1990; 14:54650.
  • 24
    Khimenko PL, Bagby GJ, Fuseler J, Taylor AE. Tumor necrosis factor-alpha in ischemia and reperfusion injury in rat lungs. J Appl Physiol 1998; 85:200511.
  • 25
    Skerrett SJ, Bagby GJ, Schmidt RA, Nelson S. Antibody-mediated depletion of tumor necrosis factor-alpha impairs pulmonary host defenses to Legionella pneumophila. J Infect Dis 1997; 176:101928.
  • 26
    Knight JA, Anderson S, Rawle JM. Chemical basis of the sulfo-phospho-vanillin reaction for estimating total serum lipids. Clin Chem 1972; 18:199202.
  • 27
    Weidman SW & Schonfels G. Lipids and lipoproteins. In: Sonnenwirth AC, Jarett L, eds. Gradwohl's clinical laboratory methods and diagnosis, 8th edn. C. V. Mosby, St Louis, 1980.
  • 28
    Saad AJ, Domiati-Saad R, Jerrells TR. Ethanol ingestion increases susceptibility of mice to Listeria monocytogenes. Alcohol Clin Exp Res 1993; 17:7585.
  • 29
    Coates RA, Halliday ML, Rankin JG, Feinman SV, Fisher MM. Risk of fatty infiltration or cirrhosis of the liver in relation to ethanol consumption: a case–control study. Clin Invest Med 1986; 9:2632.
  • 30
    Kawase T, Kato S, Lieber CS. Lipid peroxidation and antioxidant defense systems in rat liver after chronic ethanol feeding. Hepatology 1989; 10:81521.
  • 31
    Goldin R. The pathogenesis of alcoholic liver disease. Int J Exp Pathol 1994; 75:718.
  • 32
    Israel Y, Hurwitz E, Niemela O, Arnon R. Monoclonal and polyclonal antibodies against acetaldehyde-containing epitopes in acetaldehyde-protein adducts. Proc Natl Acad Sci USA 1986; 83:79237.
  • 33
    Ando K, Moriyama T, Guidotti LG et al. Mechanisms of class I restricted immunopathology. A transgenic mouse model of fulminant hepatitis. J Exp Med 1993; 178:154154.
  • 34
    Nakano M, Maruyama K, Okuyama K et al. The characteristics of alcoholics with HCV infection: histopathologic comparison with alcoholics without HCV infection and chronic type C hepatitis. Alcohol Alcohol 1993; 1B(Suppl):3540.
  • 35
    Botarelli P, Brunetto M, Minutello M et al. T-lymphocyte response to hepatitis C virus in different clinical courses of infection. Gastroenterology 1993; 104:5807.
  • 36
    Erickson AL, Houghton M, Choo QL et al. Hepatitis C virus-specific CTL responses in the liver of chimpanzees with acute and chronic hepatitis C. J Immunol 1993; 151:418999.
  • 37
    Fiore G, Angarano I, Caccetta L et al. In-situ immunophenotyping study of hepatic-infiltrating cytotoxic cells in chronic active hepatitis C. Eur J Gastroenterol Hepatol 1997; 9:4916.
  • 38
    Koziel MJ, Dudley D, Wong JT et al. Intrahepatic cytotoxic T lymphocytes specific for hepatitis C virus in persons with chronic hepatitis. J Immunol 1992; 149:333944.
  • 39
    Koziel MJ, Dudley D, Afdhal N et al. Hepatitis C virus (HCV)-specific cytotoxic T lymphocytes recognize epitopes in the core and envelope proteins of HCV. J Virol 1993; 67:752232.
  • 40
    Kaufmann SH. Immunity to intracellular bacteria. Annu Rev Immunol 1993; 11:12963.
  • 41
    Harty JT & Bevan MJ. CD8+ T cells specific for a single nonamer epitope of Listeria monocytogenes are protective in vivo. J Exp Med 1992; 175:15318.
  • 42
    Harty JT, Schreiber RD, Bevan MJ. CD8 T cells can protect against an intracellular bacterium in an interferon gamma-independent fashion. Proc Natl Acad Sci USA 1992; 89:116126.
  • 43
    Jiang X, Gregory SH, Wing EJ. Immune CD8+ T lymphocytes lyse Listeria monocytogenes-infected hepatocytes by a classical MHC class I-restricted mechanism. J Immunol 1997; 158:28793.
  • 44
    Czuprynski CJ & Brown JF. Effects of purified anti-Lyt-2 mAb treatment on murine listeriosis: comparative roles of Lyt-2+ and L3T4+ cells in resistance to primary and secondary infection, delayed-type hypersensitivity and adoptive transfer of resistance. Immunology 1990; 71:10712.
  • 45
    Sasaki T, Mieno M, Udono H et al. Roles of CD4+ and CD8+ cells, and the effect of administration of recombinant murine interferon gamma in listerial infection. J Exp Med 1990; 171:114154.
  • 46
    Ehlers S, Mielke ME, Blankenstein T, Hahn H. Kinetic analysis of cytokine gene expression in the livers of naive and immune mice infected with Listeria monocytogenes. The immediate early phase in innate resistance and acquired immunity. J Immunol 1992; 149:301622.
  • 47
    Feingold KR, Serio MK, Adi S, Moser AH, Grunfeld C. Tumor necrosis factor stimulates hepatic lipid synthesis and secretion. Endocrinology 1989; 124:233642.
  • 48
    Feingold KR, Soued M, Adi S et al. Effect of interleukin-1 on lipid metabolism in the rat. Similarities to and differences from tumor necrosis factor. Arterioscler Thromb 1991; 11:495500.
  • 49
    Nonogaki K, Fuller GM, Fuentes NL et al. Interleukin-6 stimulates hepatic triglyceride secretion in rats. Endocrinology 1995; 136:21439.
  • 50
    Venkatesan S, Ward RJ, Peters TJ. Effect of chronic ethanol feeding on the hepatic secretion of very-low-density lipoprotein. Biochim Biophys Acta 1988; 960:616.
  • 51
    Takeuchi N, Ito M, Yamamura Y. Esterification of cholesterol and hydrolysis of cholesteryl ester in alcohol induced fatty liver of rats. Lipids 1974; 9:3537.
  • 52
    Day CP & James OF. Steatohepatitis: a tale of two ‘hits’? (Editorial). Gastroenterology 1998; 114:8425.
  • 53
    Albano E, Clot P, Morimoto M, Tomasi A, Ingelman-Sundberg M, French SW. Role of cytochrome P4502E1-dependent formation of hydroxyethyl free radical in the development of liver damage in rats intragastrically fed with ethanol. Hepatology 1996; 23:15563.
  • 54
    Guebre-Xabier M, Yang S, Lin HZ, Schwenk R, Krzych U, Diehl AM. Altered hepatic lymphocyte subpopulations in obesity-related murine fatty livers: potential mechanism for sensitization to liver damage. Hepatology 2000; 31:63340.
  • 55
    Tanaka Y, Takahashi A, Kobayashi K et al. Establishment of a T cell-dependent nude mouse liver injury model induced by Propionibacterium acnes and LPS. J Immunol Methods 1995; 182:218.DOI: 10.1016/0022-1759(95)00014-2
  • 56
    Tanaka Y, Takahashi A, Watanabe K et al. A pivotal role of IL-12 in Th1-dependent mouse liver injury. Int Immunol 1996; 8:56976.
  • 57
    Tsutsui H, Matsui K, Kawada N et al. IL-18 accounts for both TNF-alpha- and Fas ligand-mediated hepatotoxic pathways in endotoxin-induced liver injury in mice. J Immunol 1997; 159:39617.
  • 58
    Dao T, Mehal WZ, Crispe IN. IL-18 augments perforin-dependent cytotoxicity of liver NK-T cells. J Immunol 1998; 161:221722.
  • 59
    Ohtsuki T, Micallef MJ, Kohno K, Tanimoto T, Ikeda M, Kurimoto M. Interleukin 18 enhances Fas ligand expression and induces apoptosis in Fas-expressing human myelomonocytic KG-1 cells. Anticancer Res 1997; 17:32538.
  • 60
    Russell JQ, Morrissette GJ, Weidner M, Vyas C, Aleman-Hoey D, Budd RC. Liver damage preferentially results from CD8 (+) T cells triggered by high affinity peptide antigens. J Exp Med 1998; 188:114757.
  • 61
    Huang L, Soldevila G, Leeker M, Flavell R, Crispe IN. The liver eliminates T cells undergoing antigen-triggered apoptosis in vivo. Immunity 1994; 1:7419.
  • 62
    Luettig B, Pape L, Bode U et al. Naive and memory T lymphocytes migrate in comparable numbers through normal rat liver: activated T cells accumulate in the periportal field. J Immunol 1999; 163:43007.
  • 63
    French LE & Tschopp J. Thyroiditis and hepatitis: Fas on the road to disease (news, comment). Nat Med 1997; 3:3878.
  • 64
    Jones RA, Johnson VL, Buck NR et al. Fas-mediated apoptosis in mouse hepatocytes involves the processing and activation of caspases. Hepatology 1998; 27:163242.
  • 65
    Kondo T, Suda T, Fukuyama H, Adachi M, Nagata S. Essential roles of the Fas ligand in the development of hepatitis. Nat Med 1997; 3:40913.
  • 66
    Li J, Rosen D, Ronen D et al. The regulation of CD95 ligand expression and function in CTL. J Immunol 1998; 161:39439.
  • 67
    Ni R, Tomita Y, Matsuda K et al. Fas-mediated apoptosis in primary cultured mouse hepatocytes. Exp Cell Res 1994; 215:3327.DOI: 10.1006/excr.1994.1349
  • 68
    Busch DH & Pamer EG. MHC class I/peptide stability: implications for immunodominance, in vitro proliferation, and diversity of responding CTL. J Immunol 1998; 160:44418.
  • 69
    Busch DH, Pilip I, Pamer EG. Evolution of a complex T cell receptor repertoire during primary and recall bacterial infection. J Exp Med 1998; 188:6170.
  • 70
    Busch DH, Pilip IM, Vijh S, Pamer EG. Coordinate regulation of complex T cell populations responding to bacterial infection. Immunity 1998; 8:35362.
  • 71
    Nishimura T & Ohta A. A critical role for antigen-specific Th1 cells in acute liver injury in mice. J Immunol 1999; 162:65039.
  • 72
    Cortez-Pinto H, Yang SQ, Lin HZ, Costa S, Hwang CS, Lane MD, Bagby G, Diehl AM. Bacterial lipopolysaccharide induces uncoupling protein-2 expression in hepatocytes by a tumor necrosis factor-alpha-dependent mechanism. Biochem Biophys Res Commun 1998; 251:3139.DOI: 10.1006/bbrc.1998.9473
  • 73
    Pastorino JG & Hoek JB. Ethanol potentiates tumor necrosis factor-alpha cytotoxicity in hepatoma cells and primary rat hepatocytes by promoting induction of the mitochondrial permeability transition. Hepatology 2000; 31:114152.
  1. §Current address: Dr Igor I. Slukvin, Wisconsin Regional Primate Center, University of Wisconsin, Madison, WI, USA.