Moderate Drinking, Inflammation, and Liver Disease

Introduction 

Alcoholic liver disease (ALD) remains a major cause of morbidity and mortality worldwide. Alcohol-induced reactive oxygen and nitrogen species, as well as inflammatory and genetic factors, contribute to the development and progression of ALD. In contrast to the effects on coronary heart disease, moderate alcohol consumption has no benefit for liver disease and may even cause liver disease in susceptible individuals. Activation of inflammatory pathways by long-term heavy alcohol intake, as opposed to attenuation of inflammation by acute moderate alcohol consumption, likely contributes to some of the organ-specific effects of alcohol.

Alcohol Consumption and Liver Disease 

While multiple studies suggest a beneficial effect of moderate alcohol consumption on coronary heart disease, a linear correlation exists between the amount and duration of alcohol use and liver disease. In the Western world, up to 50% of cases of end-stage liver disease are related to alcohol use (1). The “threshold” of alcohol consumption associated with ALD depends on the daily amount and the duration of alcohol intake. Furthermore, gender difference is a major factor in ALD. In males, a daily intake of 40 to 80 g of alcohol over a period of 10 to 12 years can lead to ALD, while in females only 10 to 30 g of daily alcohol can lead to liver disease 2, 3. A large, population-based prospective study in Denmark found that the nadir of the estimated relative risk of developing liver disease was at an alcohol intake of 1 to 6 beverages per week, and above this level a steep increase in relative risk was observed (2). The level of alcohol intake above which the relative risk was significantly greater than one was observed at 7 to 13 beverages per week for women and 14 to 27 beverages per week for men. While the association with the amounts and duration of alcohol abuse is well established, at this time, there are insufficient data to evaluate whether different patterns of drinking would affect development of alcoholic liver disease.

Although the determination of a safe “threshold” of alcohol intake, with respect to dose and length of time, would be important for evaluating the global benefits and potential harms of moderate alcohol use, the amount of “safe” alcohol intake that would not cause ALD is difficult to determine. Long-term heavy alcohol use predictably results in the development of fatty liver in 90% to 100% of patients, while only 10% to 35% develop alcoholic hepatitis and 8% to 20% develop alcoholic cirrhosis (4). The severity of steatosis predicts the risk of progression to fibrosis or cirrhosis in alcoholic patients (5). Most individuals who habitually drink more than 60 g/d of alcohol will develop fatty liver and subjects with severe microvesicular/macrovesicular fat deposition are more likely to develop fibrosis and evolve to cirrhosis 5, 6. Some studies suggest that even moderate drinking can increase the risk of cirrhosis (7). A meta-analysis of 8 studies found a pooled relative risk estimate between 1.6 and 3.0 for moderate drinkers (<25 g/d) compared with nondrinkers in populations that included both men and women (8). Moderate drinking does not appear to have any beneficial effects on liver cancer. While liver cancer is associated with excessive and prolonged alcohol use, moderate doses of alcohol (25g/d) resulted only in a minimally elevated risk of liver cancer in both men and women (8).

It remains to be understood why only a subset of people with prolonged heavy alcohol use will develop ALD, and why cirrhosis occasionally develops even in “social drinkers” 9, 10, 11. Currently, there is a limited amount of knowledge regarding specific coexisting host and external factors that may interact with alcohol to cause and accelerate liver disease. Of those factors, female gender is a well-established risk factor for ALD, as lower doses of alcohol over a briefer period result in liver damage among females. Women have a significantly higher relative risk of developing alcohol-related liver disease than men for any given level of alcohol intake (2). The underlying mechanisms are not fully understood, though differences in the alcohol dehydrogenase enzyme and hormonal factors have been suggested. Importantly, ALD progresses more frequently in women even after cessation of alcohol use (4). The presence of other liver diseases, particularly chronic infection with hepatitis C (HCV) or B viruses (HBV), significantly increases the risk of progressive liver damage in combination with both moderate and excessive alcohol use (12). The prevalence of chronic HCV infection is increased in patients with long-term alcohol consumption, and alcohol use has been identified as an independent risk factor for progression of liver disease in chronic HCV infection (13). It remains to be determined whether the additive effects of alcohol and HCV infection are related to the effects of alcohol on HCV viral replication, host immune factors, or hepatocytes (14). Genetic factors, such as hereditary hemochromatosis, are also associated with an increased risk of alcohol-induced liver injury, as alcohol consumption greatly increases the prevalence of cirrhosis in these patients (15). Obesity and exposure to other substances/drugs present additional risks for alcoholic liver disease 16, 17. On the basis of these multiple factors, “safe” levels of alcohol consumption with regard to liver disease may significantly vary among individuals depending on the presence or absence of cofactors that may modulate the effects of alcohol on the liver.

Mechanisms of Liver Injury and the Role of Inflammation in Alcoholic Liver Disease 

The mechanisms by which alcohol consumption causes liver damage involves a number of physiological and biochemical changes that lead to liver pathology. Alcohol dehydrogenase (ADH) is the dominant enzyme in alcohol oxidation producing acetaldehyde, which is further oxidized to acetate by aldehyde dehydrogenase (ALDH). When tissue alcohol levels exceed 10 mmol/L concentration, the microsomal enzyme cytochrome P450 2E1 (Cyp2E) is induced and participates in alcohol metabolism. It has been proposed that Cyp2E1 induction plays a role in ALD and that this enzyme is a source of oxidative stress that results in direct damage to hepatocytes, contributing to inflammation (18).

The natural history of ALD is characterized by chronic inflammation in the liver (Fig. 1). There are multiple lines of evidence for inflammatory cascade activation in ALD, including neutrophil and other inflammatory cell recruitment to the liver and increased circulating levels of proinflammatory cytokines such as tumor necrosis factor α (TNFα), interleukin 6 (IL-6), IL-8, and monocyte chemotactic protein 1 (MCP-1) 18, 19. This inflammation is most likely triggered by reactive oxygen species (ROS) and endotoxin (lipopolysaccharide, LPS). Alcohol consumption is associated with impaired function of the intestinal mucosa in patients with various stages of alcoholic liver injury, as well as in rodents after short-term alcohol administration 20, 21. Several studies have shown malabsorption of xylose in alcoholics without liver disease (22). In human alcoholics, LPS levels are increased in the blood and in the portal circulation (23). As it enters the liver, LPS is recognized by the Toll-like receptor 4 complex expressed on the resident macrophages, Kupffer cells, in the liver and induces downstream activation of intracellular pathways that result in activation of various proinflammatory genes including TNFα, IL-8, and MCP-1 (24). Kupffer cells isolated from mice after alcohol administration demonstrate increased spontaneous and stimulation-induced TNFα production (24). The increased production of TNFα by Kupffer cells is thought to be of particular importance in the pathogenesis of ALD, as TNFα is one of the principal mediators of the inflammatory response transducing differential signals that regulate cellular activation, proliferation, cytotoxicity, and apoptosis (25). LPS recognition by Toll like receptor 4 expressed on hepatic stellate cells and sinusoidal epithelial cells may also contribute to the progression of ALD 26, 27. The continued presence of increased inflammatory mediators and activation of Kupffer cells in the liver contribute to the activation of stellate cells by inflammatory mediators and induce collagen deposition, fibrosis, and progression to cirrhosis.

View Large Image Download to PowerPoint

Figure 1 Currently accepted model of the pathomechanisms of alcoholic hepatitis. Chronic alcohol use is associated with increased endotoxin levels in the portal blood presumably from increased gut permeability. Endotoxin in the liver activates Kupffer cells, the resident macrophages, to produce chemokines (IL-8, MCP-1), proinflammatory cytokines (TNFα, IL-1), reactive oxygen species (ROS), and transforming growth factor β (TGFß). The chemokines recruit other inflammatory cells, particularly polymorphonuclear leukocytes (PMNL) into the liver, which amplify the inflammation. ROS and TNFα can damage hepatocytes by inducing apoptosis. In addition, alcohol and its metabolites have direct negative effects on hepatocyte function and survival. Moreover, Kupffer cell–derived inflammatory mediators, TGFβ, and LPS activate stellate cells to proliferate and produce collagen, leading to fibrosis and progression of liver damage.

Different Regulation of Inflammatory Pathways by Acute and Chronic Alcohol Use 

Increasing evidence suggests that in contrast to the proinflammatory activation in alcoholic hepatitis by chronic excessive alcohol consumption, acute alcohol administration has anti-inflammatory effects. Consistent with this notion, peripheral blood monocytes obtained from patients with alcoholic hepatitis spontaneously produced increased amounts of TNFα and responded to ex vivo LPS stimulation with increased TNFα levels 19, 28, while acute alcohol consumption in normal volunteers results in the attenuation of TNFα production by various stimulants (29). Acute alcohol intake in humans also attenuated monocyte production of other proinflammatory cytokines, including IL-1β and the chemokines, IL-8 and MCP-1 29, 30, suggesting that common pathways in the regulation of these cytokines may be affected by alcohol. Indeed, activation of the nuclear regulatory factor κB (NF-κB), a common element in the promoter region of these proinflammatory genes, is inhibited by acute alcohol intake. NF-κB nuclear translocation, which is associated with proinflammatory cytokine induction, is decreased in monocytes of individuals after short-term alcohol intake (29). Similar inhibition of NF-κB activation has been reported in other studies using in vitro alcohol administration and other cell types (31). In contrast to the studies above, increased NF-κB activation has been reported in monocytes of chronic alcoholics with alcoholic hepatitis (28).

In animals, administration of an alcohol-containing liquid diet to mice for 4 weeks results in fat deposition in the liver, similar to changes seen in humans after excessive alcohol use (Fig. 2). Furthermore, alcohol feeding causes liver damage, indicated by increased serum levels of alanine aminotransferase in alcohol-fed animals as compared with those fed control liquid or chow diet. Activation of the proinflammatory cascade is demonstrated by the increased baseline NF-κB activation in the livers of alcohol-fed mice and not in the livers of control mice (Fig. 2). In contrast, acute alcohol administration in mice attenuates LPS-induced NF-κB activation in the liver and serum TNFα induction (Fig. 3). Together, these results indicate that acute ingestion of alcohol and chronic heavy use have opposite effects on inflammatory pathways, inflammatory cell activation in the liver, and in isolated monocytes (Fig. 4).

View Large Image Download to PowerPoint

Figure 2 Chronic alcohol feeding results in liver injury and inflammatory activation in mice. Mice (3–4 mice per group) were fed with a Lieber–de-Carli liquid diet containing 4.5% vol/vol alcohol, pair-fed liquid or chow diet. Top panels: Liver sections were stained with hematoxylin-eosin to demonstrate steatohepatitis. Bottom right: Serum ALT was determined using a commercially available kit demonstrating an increase in alcohol-fed mice. Bottom left: Nuclear extracts from livers were tested in an electromobility gel shift assay for NF-κB binding to a consensus NF-κB oligonucleotide as described (31), demonstrating an increase in NF-κB in alcohol-fed animals.

View Large Image Download to PowerPoint

Figure 3 Acute alcohol administration inhibits NF-κB activation and LPS-induced TNFα production in mice. Mice (4–5 per group) were given acute alcohol (2 mL in intraperitoneal injection), followed by an LPS challenge. Serum and livers were collected 2 hours after the LPS administration. Right panel: Serum TNFα levels were determined by enzyme-linked immunosorbent assay demonstrating decreased TNFα in acute alcohol-treated mice. Left panel: Nuclear extracts from livers were evaluated for NF-κB activation in an electrophoretic mobility shift as described (31), demonstrating the decreased NF-κB in acute alcohol-treated mice.

View Large Image Download to PowerPoint

Figure 4 Opposite effects of acute and chronic alcohol consumption on inflammation. Acute alcohol consumption has an anti-inflammatory effect through its effects on macrophages by inhibiting stimulation-induced proinflammatory pathways, while chronic alcohol use is associated with up-regulation of the same pathways in inflammatory cells.

Conclusions 

Moderate alcohol use in the amount of 20 to 30 g/d for men and 10 to 15 g for women is likely to be safe for most individuals. Nevertheless, an individual's threshold for alcohol-induced liver disease depends on multiple susceptibility factors and coexisting liver conditions. The presumed “safe” limits of alcohol use, therefore, cannot apply to people with cofactors such as chronic viral hepatitis, obesity, hemochromatosis, and other factors that increase the susceptibility of the liver to alcohol-induced injury. Given the frequency of chronic HCV infection and the increasing epidemic of obesity in Western countries, universal limits of “safe” alcohol use are difficult to determine for the liver.

There is an obvious anti-inflammatory effect of acute, moderate alcohol intake on inflammatory cell functions that may be a critical component of cardioprotective effects of alcohol use. To date, there is no evidence for a beneficial role of the anti-inflammatory effect of acute moderate alcohol consumption on the liver, but the effects on inflammation may contribute to the effects of alcohol on other organ systems.