| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Commentary |
From the Department of Pharmacology and Cell Biophysics, University of Cincinnati, Cincinnati, Ohio
Chronic alcohol use in humans results in functional changes and pathology in multiple organ systems. Particularly affected are the liver, the heart, and the pancreas. Alcohol-induced disease of these organs often is a significant clinical problem, resulting in serious or fatal illness, particularly in cases of chronic alcohol abuse (defined as greater than 90 g of alcohol per day for 5 years or more). In humans, such exposure to alcohol typically results in a form of dilated cardiomyopathy that is characterized by reduced contractility, ventricular dilatation, cardiomyocyte apoptosis, and fibrosis, often progressing to heart failure.1 Cardiac fibrosis involves fibroblast proliferation and transdifferentiation to myofibroblasts, extracellular matrix remodeling, and stiffening of the ventricular wall. Because fibrosis exacerbates the decline of ventricular function and contributes to dilatation and progression to failure, it represents an attractive therapeutic target.2 To date, features of alcoholic cardiomyopathy (ACM), notably fibrosis, are not reproducibly recapitulated in animal models. In this issue of The American Journal of Pathology, Wang and colleagues3 detail a novel model of ACM and investigate the effects of metallothionein (MT) and zinc on ACM and myocardial fibrosis.
Alcohol affects the heart both directly and indirectly. Direct effects involve perturbation of the structure/function and/or metabolism of cardiac constituent cells. Indirect effects are mediated by other organs and by neurohormonal factors. Although the liver is the primary site of alcohol metabolism, alcohol also acts on the myocardium via direct effects on cellular metabolism. Myocardial cells metabolize ethanol via alcohol dehydrogenase to produce acetaldehyde, which enhances release of catecholamines and directly impairs cardiac function.4 Alcohol also interferes with excitation-contraction coupling by affecting the calcium transit.5 Ethanol can combine with fatty acids to produce fatty acid ethyl esters that damage mitochondria and membranes, and alcohol enhances lipid peroxidation, which damages cellular membranes. Cardiac muscle from alcohol-fed animals has a reduced capacity for oxidative metabolism and increased glycolysis, resulting in a more acidic intracellular environment. Alcohol is known to alter mitochondrial metabolism and to enhance production of oxygen-derived free radicals, including superoxide (O2·–), and thus has direct effects on generation of reactive oxygen species.2,6 Acute alcohol exposure reduces the levels of the myofibrillar proteins and reversibly reduces cardiac contractile function while chronic exposure to alcohol leads to further irreversible reduction of myofibrillar protein.7 Chronic alcohol administration increases the probability of fibrillation5 and affects excretion and tissue levels of metal ions.8 Thus, alcohol exposure to the heart generally leads to decreased myocardial contractile function, cellular damage, and increased oxidative stress.
Effects of Alcohol on Minerals
In both the human liver and heart, alcohol use is associated with perturbation of metal ion homeostasis. It is not known whether this is a primary effect of alcohol or an indirect result of tissue damage, but serum zinc deficiency is common among patients with idiopathic cardiomyopathy.9,10 In both liver and heart, chronic alcohol use results in reduced levels of zinc.11 In the liver, where this is better studied, zinc depletion occurs in association with high tissue levels of copper, iron, and manganese and increased fibrosis.8 Numerous studies have reported that low zinc levels associate with alcohol abuse and correlate with the severity of liver cirrhosis and mortality.8,12-14 Anttinen and colleagues15 found that zinc supplementation reduces collagen accumulation and fibrosis in a model of carbon tetrachloride-induced liver injury. This result was confirmed by subsequent investigations.16,17 Elsherif and colleagues18 showed that liver fibrosis resulting from carbon tetrachloride treatment for 4 weeks was reversible by withdrawal of the carbon tetrachloride. However, damage resulting from an 8-week course of carbon tetrachloride treatment induced irreversible damage and fibrosis associated with reduced hepatic MT levels. These investigators went on to show that, in mice deficient for MT-I and -II (MT-KO), the 4-week treatment regimen induced irreversible liver damage and fibrosis. Importantly, adenoviral delivery of the human MT II gene to the liver reversed fibrosis and activated hepatocyte regeneration in both MT-knockout (KO) (4 weeks) and wild-type mice (8 weeks) with irreversible fibrosis. The increased MT activity was associated with increased collagenase activity, suggesting that the MT II gene therapy activated anti-fibrotic pathways in the damaged livers. This was the first study to implicate MT in the modulation of fibrosis, but whether this acts directly or via the reactive oxygen species scavenging activity of MT was not addressed. Until now, there has been no data bearing on a role for MT in cardiac fibrosis.
Zinc, copper, iron, and manganese play important roles as co-factors for several of the enzymes involved in collagen synthesis/degradation and extracellular matrix remodeling. Low zinc levels enhance prolylhydroxylase activity and inhibit collagenases, favoring collagen deposition.19,20 Also, low zinc levels impair superoxide dismutase activity21 that would normally counter increased generation of free radicals subsequent to alcohol exposure. Zinc and copper are required co-factors for the activity of matrix metalloproteinases (MMPs), which are critically involved in extracellular matrix remodeling. MMPs are grouped into four major categories; collagenases (eg, MMP1, MMP13), gelatinases (eg, MMP2, MMP9), stromelysin-specific (eg, MMP3), and membranous types (eg, MT1-MMP). MMPs have multiple roles in the events that underlie ventricular remodeling after infarction and development of ventricular failure in the heart. For instance, mice deficient for MMP9, a gelatinase, have reduced remodeling after infarct and improved function, suggesting that MMP9 activity contributes to fibrosis.22 On the other hand, MMPs with collagenase activity are anti-fibrotic.23 Dysregulation of MMP activity is generally associated with adverse myocardial remodeling.23,24 Thus, zinc potentially plays multiple regulatory roles relating to collagen deposition and degradation. The article by Wang and colleagues3 in this issue of The American Journal of Pathology confirms this by demonstrating that zinc deficiency alters the cardiomyopathic response to alcohol in the mouse.
Animal Models of ACM
Curiously, although alcohol induces hypertrophy and cardiac dysfunction in mice, fibrosis is not typically part of the pathophysiological picture. Thus, murine models of ACM have been problematic and of dubious clinical application. Investigators have turned to other animals, including the dog, rabbit, rat, and chicken,7,25-35 to develop more relevant models. Still, the features of ACM in these animal models do not exactly mirror those of human ACM and are completely reversible by withdrawal of alcohol, unlike the human condition.7 Nevertheless, use of these models has enabled a great deal of research and has revealed several important aspects of alcohol toxicity in the cardiovascular system. Although animal models have their limitations, many of the experimental approaches that have been used with these models could not be performed with humans due to obvious ethical dilemmas. However, the paucity of murine models in particular poses a severe limitation because one cannot take advantage of the powerful approaches available, including genetically engineered mice (transgenics and gene targeting), the mouse genome, gene expression and proteomic databases, the numerous inbred, outbred, mutant strains, and the inbred recombinant and congenic lines that have been engineered. Currently, many such resources for the mouse exist and their number and availability increase almost daily. Mouse models of cardiovascular disease have proven to be extremely powerful tools for the elucidation of pathophysiological mechanisms. Thus, developing mouse models of ACM that more closely mirror the human condition is paramount.
The Murine MT-KO Model of ACM
In their study, Wang and colleagues3 demonstrate that mice homozygous for a deficiency of both the MT I and II genes (MT-KO) present with an ACM more typical of the human disease because fibrosis occurs. In this model, alcohol feeding was accomplished using a Lieber and DeCarli liquid diet modified by replacing the alcoholic diet with the control diet on the last day of each week; this prevented the weight loss that otherwise is associated with long-term alcohol feeding regimens of this type. The content of alcohol was increased during the 2-month regimen, from 26 to 36% of total calories. Importantly, the blood alcohol levels in this model are close to those measured in alcoholic patients.3 At the 2-month time point, wild-type and MT-KO mice were euthanized, and the effects of alcohol on the morphological and histological features of cardiomyopathy were ascertained. Analysis of areas of collagen accumulation showed a significant twofold increase in fibrosis in MT-KO relative to wild-type mice treated with alcohol. This reflects increases in both perivascular and interstitial fibrosis. In MT-KO mice, this was associated with neutrophil infiltration and cardiomyocyte hypertrophy in regions of focal fibrotic lesions. This likely represents replacement fibrosis, which is quite characteristic of cardiomyopathy, including human ACM. This is the first important aspect of the article by Wang and colleagues,3 the development of the first mouse ACM model that recapitulates fibrosis, a critical aspect of pathophysiology associated with human ACM.
The second important aspect of this work is that fibrosis in the MT-KO/ACM model is reversible by zinc.3 The investigators used zinc supplementation with 100 mg of zinc per L liquid diet and found this sufficient to prevent the alcohol-induced decrease of zinc in the livers of experimental animals. In the MT-KO/ACM model, zinc supplementation significantly reduced fibrosis overall and nearly prevented interstitial fibrosis. The zinc supplementation did not affect the alcohol-induced cardiac hypertrophy measured by heart-to-body weight ratio. Importantly, hypertrophy was similar in wild-type and MT-KO mice subjected to the alcohol diet for 2 months. Zinc supplementation also suppressed some of the alcohol-induced mitochondrial dysmorphogenesis and was more effective in this respect in wild-type relative to MT-KO mice. In addition to delineating a role in fibrosis, these results suggest that zinc and MT may play important anti-oxidant roles in this model.
The seminal discovery by Wang and colleagues,3
that alcohol-induced fibrosis occurs in mice deficient for MT and is suppressed by zinc supplementation, demonstrates that zinc homeostasis is critical for development of fibrosis in murine ACM. Although reduced tissue zinc levels were previously associated with alcoholic pathophysiology in the heart and liver, the causal relationship was unknown. Fibrosis has been well studied in multiple organs and appears to be a critical component of wound healing after injury. As such, fibrosis has multiple critical components and is intimately linked to inflammatory signaling and cell-cell interactions between tissue-resident cells and infiltrating cells.36
In the heart, fibrosis is associated with increased proliferation of fibroblasts, increased transdifferentiation of fibroblasts to myofibroblasts, and increased deposition of collagen.37
In ACM, ethanol exposure tips the normal homeostatic balance between anti- and profibrotic signals in the direction of fibrosis (Figure 1)
. Critical alcohol-dependent profibrotic signals include increased catecholamine levels, oxidative stress, inflammatory cytokine levels, and as shown by Wang and colleagues,3
reduced zinc levels.6-8
|
. One would therefore predict that zinc supplementation would at least partially restore the decreased pool of zinc and improve the function of MT and collagenase MMPs, thereby reducing fibrosis (Figure 1)
. If zinc works partially through MT, this effect would be predicted to be stronger in wild-type than in MT-KO mice, exactly as observed by Wang and colleagues.3 Summary
The study by Wang and colleagues3 in this issue of The American Journal of Pathology describes a novel murine model of ACM. Results from experiments that use this model support the hypothesis that zinc availability critically affects fibrosis in ACM in vivo. In humans, chronic alcohol abuse is associated with irreversible fibrosis and dilated cardiomyopathy. In animal models, however, fibrosis can be reversed by removing alcohol from the diet or, as in this study, by zinc supplementation. As discussed above, zinc supplementation has been previously reported to reverse alcoholic liver fibrosis in the carbon tetrachloride liver injury model.15 It is currently unknown whether the reversibility of fibrosis and other features of ACM in animals, whether by alcohol withdrawal or zinc supplementation, is related to differences in basic mechanisms or to the fact that human disease is most often associated with much longer duration of alcohol usage.7 Despite this conundrum, studying the mechanisms that underlie the effects of alcohol on cardiac function and remodeling will likely result in significant new knowledge that is clinically relevant. Whether it proves efficacious to completely block the fibrotic component of remodeling remains to be seen; certain aspects of fibrosis and wound healing are likely required for the heart to sustain functional integrity in certain circumstances. Nevertheless, understanding the mechanisms that underlie reversal of fibrosis is of great interest. The existence of the MT-KO/ACM model will allow investigators to bring the power of mouse genetics to bear on the problems of ACM and cardiac fibrosis, increasing the probability that detailed aspects of the molecular mechanism will be forthcoming. These results will eventually have great impact in identifying novel therapeutic targets for limiting pathological fibrosis in the clinic.
Acknowledgements
I thank Maria Brown and Suiwen He for critical reading of the commentary.
Footnotes
Address reprint requests to W. Keith Jones, Department of Pharmacology and Cell Biophysics, 231 Albert Sabin Way ML0575, University of Cincinnati, Cincinnati, OH 45267-0575. E-mail: joneswk{at}uc.edu
Accepted for publication April 26, 2005.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
