Main

Human and animal copper-overload states are associated with severe hepatic injury, culminating in fulminant hepatic failure or cirrhosis(1, 2). In patients with Wilson's disease, a genetic defect in intracellular copper transport(3, 4) causes failure of hepatocyte excretion of copper into bile and impairment of secretion of ceruloplasmin, the major hepatic-derived circulating copper transport protein(1). The resultant accumulation of hepatocellular copper invariably leads to hepatic injury, fibrosis, cirrhosis, and death if not treated(1). In addition, other organs and tissues accumulate copper after the hepatic storage capacity has been exceeded, leading to basal ganglion, renal, corneal, and other lesions(1, 2). Hepatic copper toxicity also plays a role in Indian childhood cirrhosis(5) and in idiopathic North American childhood cirrhosis(6), although the etiology of copper overload in these disorders has not been fully elucidated. Recent evidence suggests an increased ingestion of copper-contaminated milk or water may be involved in the latter two disorders(7).

Treatment of human copper overload disorders is based on the chelating and cupriuretic effect of several agents(1, 2) and the impairment of copper absorption induced by oral zinc therapy(8). Despite general improvement in the hepatic injury if chelation therapy is started relatively early in the course of liver damage(2, 9), many patients fail to have a significant response to chelation therapy and eventually require liver transplantation to avoid death(10, 11). A similar lack of response to chelation therapy has been demonstrated in advanced Indian childhood cirrhosis(12). Moreover, in fulminant hepatic failure caused by Wilson's disease, copper chelation and zinc therapy is completely ineffective(1, 2, 13). Therefore, there is a continued need for the development of improved treatment for copper-overload conditions.

To develop treatment strategies for copper-overload diseases requires a thorough understanding of the underlying cellular and molecular mechanisms by which copper damages the liver. Previous studies in copper-overloaded rats(14, 15), dogs(16), and humans with Wilson's disease(16) have documented evidence of oxidant injury to hepatic mitochondria, suggesting that free radicals may be involved in copper hepatic toxicity. These studies led to the proposal that antioxidant therapy might be of value in states of copper overload. To test this hypothesis, we have attempted to develop an in vitro hepatocyte system in which copper toxicity and antioxidant status could be manipulated. Initial attempts were directed at incubating hepatocytes isolated from normal rats with toxic doses of cupric chloride and with antioxidants(17); however, the applicability of thisin vitro model to in vivo copper overload states is not clear. The purpose of the present study was to develop a model by whichin vivo copper overload of the rodent liver could be studied in an isolated hepatocyte preparation, to determine the role of oxidant stress in these in vivo copper-overloaded hepatocytes, and to compare the relative effects of copper chelation and antioxidant (vitamin E) treatment in this model.

METHODS

Animals. Weanling male Sprague-Dawley rats were obtained from Sasco Inc. (Omaha, NE) and housed in polyethylene cages with stainless steel wire tops on 12-h day-night light cycles. All rats received humane care under the guidelines of Animal Care Committee of the University of Colorado Health Sciences Center. To achieve two different states of copper nutritional status, rats were randomly assigned to receive an 8-wk course of one of two semisynthetic diets obtained from Dyets, Inc. (Bethlehem, PA): a high copper group received 1000 mg of copper (as cupric chloride) per kg of diet for the first 4 wk followed by 2000 mg/kg of diet for 4 wk, as previously described(14). This diet produces hepatic copper levels after 8 wk that are similar to those in humans with Wilson's disease, approximately 20-fold elevated compared with a control diet, and a portal triaditis with focal hepatocyte necrosis(14). A control diet had 6 mg of copper/kg of diet. Both diets contained 50 IU of α-tocopherol acetate/kg of diet (the recommended normal dietary content) and stripped corn oil as the dietary lipid source. Both diets were otherwise nutritionally complete and identical. Control rats were pair-fed to produce the same weight gain observed in the copper-overloaded rats.

Hepatocyte Isolation. After 8 wk of diet, rats were killed, and hepatocytes were isolated from each rat by a modification of the recirculatingin situ collagenase technique, as previously described(17). An important modification of this technique was the use of 9% O2/86% N2/5% CO2 as the gas bubbled through the isolation buffer and maintained in the atmosphere in flasks containing the isolated hepatocytes, to achieve the oxygen content normally present in the liver and thus avoid a hyperoxic stress. This modified technique results in over 92% viable hepatocytes based on trypan blue exclusion. After isolation, hepatocytes were maintained at 4°C until used for experiments within 2 h of isolation. Blood was obtained from the inferior vena cava before starting the liver perfusion and analyzed for serum aspartate aminotransferase, alanine amino-transferase, alkaline phosphatase, and bilirubin by an automated analyzer. An aliquot of freshly isolated hepatocytes was stored frozen at-20°C and analyzed for copper content by atomic absorbance spectroscopy(20) on a Perkin-Elmer 360 atomic absorption spectrophotometer and 2100-HGA controller and furnace (Perkin-Elmer Corp., Norwalk, CT). Hepatocyte copper content was expressed as micrograms/106 cells.

Experimental Design. Freshly isolated hepatocytes from both copper-overloaded and control rats were resuspended in Krebs-Henseleit buffer containing 2% BSA, 1% D-glucose, and a mixture of amino acids(18) at a concentration of 1.0 × 106 cells/mL. Aliquots of cells were incubated at 37°C in a shaking water bath with one of the following: D-α-tocopheryl succinate solubilized in ethanol (final concentration of 200 μM for α-tocopherol and 0.05% for ethanol); ethanol as a control (final concentration of 0.05%); or 2,3,2-tetramine, a potent copper chelator(19) (final concentration of 100 μM). All cells were incubated in sealed flasks under an atmosphere of 9% O2/86% N2/5% CO2 that was sparged every 15 min. Aliquots of hepatocytes, removed at baseline and each hour for 4 h, were analyzed for viability [trypan blue exclusion(17)] and lipid peroxidation by the TBARS method, as previously described(17). A standard curve for the TBARS assay was constructed daily using 1,1,3,3-tetraethoxypropane as the standard; results were expressed as nanomoles of TBARS/106 hepatocytes.

Statistical Analysis. Differences among experimental groups were tested for significance by the unpaired t test or the analysis of variance with the Scheffe F test. Correlations were examined by linear regression analysis (least squares technique). A p value of<0.05 was considered statistically significant. Data were expressed as mean± SEM.

RESULTS

Pair-fed rats maintained on the two dietary regimens were of equal weight at the time of sacrifice (Table 1). Serum aspartate aminotransferase and alanine aminotransferase concentrations were significantly increased in the copper-overloaded rats (Table 1), indicating chronic hepatocellular injury caused by excess dietary copper intake. Alkaline phosphatase and bilirubin did not vary between groups. Hepatocyte copper content (Table 1) reflected the dietary copper intake, with approximately a 15-fold copper increase in the copper-overloaded hepatocytes.

Table 1 Biochemical data for experimental groups
Full size table

Initial viability of hepatocytes isolated from copper-overloaded rats (95.9± 0.9%) did not differ from that of control rats (97.5 ± 0.4%). Likewise, initial TBARS values of hepatocytes isolated from copper-overloaded rats were undetectable in cells from all rats. Therefore, copper-loaded hepatocytes tolerated the isolation procedure without evidence of significant damage or oxidant injury occurring during the isolation.

Incubation of freshly isolated control hepatocytes over 4 h resulted in a small decrease of viability to 83 ± 2% at 4 h (Fig. 1A). Incubation of copper-loaded hepatocytes resulted in a progressive decrease in viability over 4 h to 60 ± 5% that was significantly greater than that observed in control hepatocytes (Fig. 1A). Incubation of copper-overloaded hepatocytes with 2,3,2-tetramine resulted in a significant but partial improvement in cell viability(Fig. 1A). However, incubation of copper-overloaded hepatocytes with α-tocopheryl succinate improved cell viability to equal that of the control hepatocytes (Fig. 2A).α-Tocopheryl succinate and 2,3,2-tetramine incubation had no significant effects on viability of control hepatocytes (data not shown).

Figure 1
figure 1

Trypan blue exclusion (A) and TBARS(B) of hepatocytes isolated from copper-overloaded (Cu) or control (Con) rats incubated in physiologic buffer and at physiologic oxygen tensions for 4 h. Hepatocytes isolated from copper-overloaded rats were also treated in vitro with 100 μM 2,3,2-tetramine (Cu + Tetra). *p < 0.05 vs control cells.

Full size image
Figure 2
figure 2

Effect of incubation of copper-overloaded hepatocytes with 200 μM α-tocopheryl succinate (Cu + Toc) on trypan blue exclusion (A) and TBARS levels (B). *p < 0.05 vs control (Con) and Cu + Toc.

Full size image

TBARS values increased progressively over 4 h in the copper-loaded hepatocytes and were significantly higher (p < 0.05) than those from control hepatocytes at time points beyond 1 h of incubation(Fig. 1B). Similar to the effects observed on cell viability, 2,3,2-tetramine incubation partially and α-tocopheryl succinate incubation completely ameliorated the rise of TBARS in copper-overloaded cells (Figs. 1B and2B). There was no significant effect of either α-tocopheryl succinate or 2,3,2-tetramine on TBARS values in control hepatocytes (data not shown).

Combining data from hepatocytes of all incubations, there was a strong correlation (p < 0.001) between the loss of cell viability(trypan blue exclusion) and the degree of oxidant damage to the hepatocytes(TBARS) after 1, 2, 3, and 4 h of incubation (r = 0.91, 0.88, 0.86, and 0.74, respectively).

DISCUSSION

The cellular mechanisms involved in the pathogenesis of copper-induced liver injury are poorly characterized. Postulated mechanisms include inhibition of sulfhydryl-containing enzymes, polymerization of proteins, alteration of charge of the plasma membrane, inhibition of tubulin polymerization, and perturbations in lysosomal function(1). The role of oxidant damage to the liver during states of excess hepatic copper has received recent attention. Copper is a potent catalyst of the Fenton reaction by which O˙2 reacts with H2O2 yielding the extremely reactive hydroxyl free radical(OH·) (reactions 1–3). The resulting hydroxyl radical readily reacts with polyunsaturated fatty acids to initiate lipid peroxidation, thiol-containing proteins, and nucleic acids, resulting in cellular injury(21). Excess copper may also cause oxidant damage by the direct oxidation of susceptible proteins and lipids. Theoretically, during copper overload, if the copper chelation capacity of intracellular metallothionein in hepatocytes is exceeded, copper could become available for these reactions.

In this study we have validated the utility of using hepatocytes isolated from dietary copper-loaded rats to study the role of oxidant stress in copper hepatic toxicity. The results show that these in vivo copper-overloaded hepatocytes undergo a time-dependent loss in cell viability compared with control hepatocytes when incubated in a physiologic buffer and at physiologic oxygen concentrations. This decrease in viability was closely paralleled by increased lipid peroxidation compared with control values. Moreover, the lipid-soluble, membrane-localized antioxidant,α-tocopherol, which is taken up well by isolated hepatocytes in suspension when delivered as the succinate ester(17), completely ameliorated both hepatocyte injury and lipid peroxidation. Although hepatocyte α-to-copherol concentrations were not measured in this study, we have previously shown a 6-fold increase in hepatocyte α-tocopherol content after a 30-min incubation with 200 μM α-to-copherol under similar experimental conditions(22). Because these incubations were carried out at the relatively low physiologic oxygen concentrations found in liver (9% O2), oxygen toxicity to the hepatocytes does not explain these findings. The potent copper chelating agent, 2,3,2-tetramine(19), had only a partial protective effect in our model, most likely because insufficient time was available in our system for this chelator to be taken up well by hepatocytes and chelate the excess copper that was present in multiple intracellular compartments. It is also possible that a portion of the intracellular copper was sequestered and inaccessible to the 2,3,2-tetramine. Nevertheless, the 2,3,2-tetramine was shown to reduce injury and lipid peroxidation to a significant, albeit partial, extent, confirming that the excess copper played a role in the cell injury observed in this model. The protective effect ofα-tocopherol and partial protection of 2,3,2-tetramine are consistent with the postulated role of oxidant damage to the hepatocyte in the genesis of copper-induced hepatocyte injury.

This isolated copper-loaded hepatocyte model confirms on a cellular basis the prior observations of oxidant injury in livers removed from mammalian models of copper-overload and from humans with Wilson's disease(14–16, 23). By using two independent techniques to measure lipid peroxidation (lipid-conjugated dienes and TBARS), we have reported markedly increased peroxidation of mitochondrial lipids from dietary copper-overloaded rats(14, 15); from the Long-Evans Cinnamon rat(23), a naturally occurring inbred model of Wilson's disease(24); from Bedlington terriers(25) with copper toxicosis(16); and from patients with Wilson's disease undergoing liver transplantation(16). Combined with the morphologic changes of hepatic mitochondria observed in rats(14) and humans with copper overload(1, 2), these data suggest that the hepatic mitochondrion is an important target organelle during copper toxicity. Indeed, abnormal respiratory function of hepatic mitochondria, as demonstrated by perturbations of state 3 mitochondrial respiration and low activity of the respiratory chain protein complex, cytochrome C oxidase, have been demonstrated in mitochondria isolated from copper-overloaded rats(15). Although in this current study we did not determine the intracellular site of increased lipid peroxidation, the demonstrated close relationship between the degree of oxidative damage and loss of cell viability supports the role of oxidant injury suggested by the mitochondrial studies.

The results of this study suggest that antioxidants might be hepatoprotective during copper overload states, such as Wilson's disease and Indian childhood cirrhosis. It should, however, be noted that the levels of hepatocyte α-tocopherol achieved by in vitro incubation(6-fold increase in α-tocopherol content) may not be duplicated by oral supplementation of vitamin E in the intact rat(14). The limited absorption of ingested vitamin E supplements(26) and the avid transfer by the liver of absorbed vitamin E into nascent very low density lipoproteins(27) appear to limit increases of hepaticα-tocopherol content in vivo. It still remains to be determined if the achievable hepatic levels of α-tocopherol, other antioxidants alone or combined antioxidant treatment would be of value in human copper overload disorders. Thus, the possible therapeutic use of antioxidants in copper-overload states merits investigation. Finally, this isolated rat hepatocyte model of copper overload appears to be an excellent system in which to investigate the cellular pathogenesis of copper-induced hepatotoxicity.