- Email: [email protected]
α-Tocopherol deficiency fails to aggravate toxic liver injury but liver injury causes α-tocopherol retention
332
Journal of Hepatology, 1992; 16:332-337 01992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved. 0168-8278/92/$05.00
HEPAT 01161
-Tocopherol deficiency fails to aggravate toxic liver injury but liver injury causes -tocopherol retention Linda Barrow, H a s m u k h R. Patel and M.S. Tanner Department of Child Health, University of Leicester, Leicester, United Kingdom
Received10September 1991)
The possible aggravation of liver injury by impaired cellular antioxidant function was investigated. A vitamin E-deficient diet (0.5 mg/kg c~-tocopherol; control 100 mg/kg) significantly reduced rat liver ~-tocopherol concentrations after 4 weeks (!.8 +1.7 lag/g; control 34.4 + 2.4 lag/g, p < 0.001). The effects of copper loading (Cu, 3 g/kg diet); galactosamine (GAIN, 0.85 g/kg i.p.); or carbon tetrachloride (CC14, 10 mmol/kg i.p.) were examined. Serum aspartate transaminase activity was elevated slightly by vitamin E deficiency but not by hepatic copper accumulation. In vitamin E-replete (E+) and vitamin E-deficient (E-) rats, GaIN or CCI 4 caused a large and comparable elevation in serum AST and OCT activity. This effect on AST was markedly reduced by copper loading in vitamin E replete (E +) rats, but in E(-) rats copper had significantly less protective effect. Copper also diminished the OCT response to GaiN in E +, though not E-, rats. A significant rise in total hepatic ~-tocopherol content followed administration of GaIN or CC14 in both normocupric and copper-laden E(-) rats. Thus ct-tocopherol deficiency (a) was not hepatotoxic per se; (b) failed to potentiate the toxicity of copper, GaIN or CC14; but (c) partially abolished the protection by copper against toxininduced liver injury. Retention of hepatic ct-tocopherol after liver damage may partly explain low serum vitamin E levels seen in clinical liver disease. K e y words: Copper; Galactosamine; Carbon tetrachloride; Vitamin E
Low serum vitamin E levels in patients with cholestatic liver disease and alcoholic hepatitis have been attributed to malabsorption and to reduced dietary intake respectively (i-3). ct-Tocopherol, the biologically active homologue of vitamin E, forms an integral part of phospholipid membranes, affording both stability and a free radical scavenging action. Recent work has shown that there is a cooperative interaction of chain-breaking antioxidants in the protection of biological membranes against free radical damage. Glutathione and ct-tocopherol have a synergistic action (4) and ascorbate-mediated protection against lipid peroxidation of microsomal membranes is dependent on the presence of ct-tocopherol (5). A deficiency of vitamin E in the liver may therefore predispose the hepatocyte to injury by free radicals. The effect of vitamin E deficiency on the liver injury produced by copper, carbon tetrachloride (CC14), and galactosamine (GAIN) has therefore been studied. Hepatic copper overload is a feature of Indian child-
hood cirrhosis (ICC) (6), Wilson's disease (7) and chronic cholestatic disorders (8). The copper overload of ICC is of dietary origin (9) and evidence that the stored hepatic copper is toxic includes the prevention of ICC by reduction in dietary copper intake (10) and the therapeutic effect of D-penicillamine in pre-icteric ICC (11) and in Wiison's disease (7). By contrast, in primary biliary cirrhosis, where liver copper is raised as a secondary effect of cholestasis, there is no evidence that the stored copper is harmful (8). Furthermore, in both the lamb and the rat, copper loading alone fails to cause the hepatic features seen in children with ICC (12). Prior hepatic copper-loading in rats fails to act synergistically with subsequently administered acute hepatoxic insults, and indeed unequivocally ameliorates the liver damage induced by both GaIN and CCI4 (13,14). Copper is a transition element with a single unpaired electron in its outer shell, enabling it to participate in free radical-mediated reactions. Copper may potentiate
Correspondence to: Prof. M.S. Tanner, Department of Paediatrics, Children's Hospital, SheffieldS10 2TH, U.K.
HEPATICc~-TOCOPHEROLAND LIVER DAMAGE oxygen toxicity by promoting the formation of hydroxyl radicals which induce site-specific peroxidation of membrane lipids. Conversely, copper may scavenge free radicals and thus act as an antioxidant. The hepatotoxicity of CCI 4 depends on the microsomal production of toxic metabolites which trigger a chain reaction of peroxidative degradation of membrane phopholipids (15) leading ultimately to cellular necrosis. GAIN, in the presence of endogenous gut-derived or exogenous endotoxin, provokes a hepatitis-like injury resulting from a rapid and prolonged uridine nucleotide depletion within parenchyreal cells; the resulting impairment in nucleic acid and glycoprotein synthesis leads to the development of structural changes in hepatocellular membranes (16). GaIN is also reported to reduce superoxide dismutase and catalase activities within hepatocytes, enzymes responsible for scavenging free radical intermediates. This report describes the influence of ~-tocopherol deficiency in rats on the liver damage produced by hepatic copper loading, by acute hepatic insults (GAIN or CCh), or by these agents in conjunction. These experimental conditions were designed to model possible clinical situations in which an acute insult such as viral hepatitis occurred in a copper-loaded tocopheroldeficient liver. The influence of these hepatotoxins on the liver c~-tocopherol content is also described.
333
Normalcopper (0,01 glkg) n-18 I VitaminE-replete (100 mg/kg) a-36
~_~ Excesscopper (3 g~g) n=t8
Weanling male rats from an inbred strain (WAGS, Harlan Olac Ltd., Oxon, U.K.) were housed in groups of 6 in constant environmental conditions under veterinary supervision. Throughout the study, 36 rats were fed ad libitum a diet deficient in vitamin E (0.5 mg/kg) comprising edible casein 20 g, sucrose 60 g, groundnut oil 10 g, vitamins and minerals (SDS Ltd., Essex, U.K.). Thirty-six rats were fed the same diet supplemented with vitamin E to normal levels (100 mg/kg) (Fig. 1). These rats are referred to as E- and E +, respectively. After 4 weeks (T=4, Fig. 1) all rats were bled under halothane anaesthesia from a tail vein. Aliquots of serum were stored at -20°C for a maximum of 3 months. Seven days later (T=5), 24 E + and 24 E- rats underwent a liver lobectomy. Using buprenorphine, 0.15 mg/kg s.c. analgesia and halothane anaesthesia, a small section of the right median lobe of the liver was removed at laparotomy. Duplicate 100-mg wet weight pieces of liver were frozen at -20°C for up to 14 days, and duplicate pieces of 20-30 mg were freeze-dried. After a further 7 days (T--6), the two treatment groups were subdivided. 18 E- and 18 E + were given their respective diets
+ GaIN
n-6
+ CCI4
n-6
J
NOinjection n=61 +GAIN
n=6)
+ CCI4
n-61
8•_• 8•
NOinjection n=61
Normalcopper (0,01 g/kg] n=l I VitaminE-deficientdiet~ (0.5 mg/kg) n=36
+ GaiN
n=6~
+ CCI4
n=6~
NOinjection n-61
Excesscopper (3 g/kg) n=l
I 0
+ GaiN
n-6)
+ CCI4
n=6)
I Dietarycopper l normal/excess Vitamin E.repleteor Vitamn E-de c ent J 4 5 6 10 weeks Tailbleed Liver Iobectomy
Materials and Methods
~
No Injection n-6 ~
Injection TaiIbleedand sacrifice
Fig. I. Experimentaldesign. Details of dosages are as in the text.
supplemented with copper sulphate, from a normal copper concentration of l0 mg/kg to a level of 3 g/kg (Fig. 1). After 4 weeks ( T = 10), all rats were again bled from a tail vein under halothane anaesthesia, and aliquots of serum stored at -20°C. Each of the four groups of 18 rats was then further subdivided; 6 rats received GaIN (0.85 g/kg body wt., i.p.) in aft injection volume of approximately 0.5 ml (range 0.40-0.59 ml), 6 rats received CC14 (10 mmol/kg body wt., i.p.) in an injection volume of approximately 0.5ml (range 0.44-0.56ml), and 6 rats received no injection (Fig. I). Twenty-four h later blood samples were taken from the tail vein of all rats under haiothane anaesthesia and the rats sacrificed by cervical dislocation. The liver was removed, weighed and duplicate pieces of 100 mg wet wt. were frozen at -20°C, and of 20-30 mg were freeze-dried.
Assays Hepatic u-tocopherol concentrations were determined by H P L C separation on a C8 ultrasphere column using pure methanol as the mobile phase at a flow rate of
334
L. BARROW et al 76 + 12 g to 206+ 17 g respectively after 6 weeks. p<0.05). Copper supplementation caused both groups initially to lose weight, the E - rats to 183 + 23 g, and the E + to 202+18 g after 1 week (p < 0.01). Rats in the four dietary groups then gained weight at a comparable rate during the remaining 3 weeks of the study, resulting in final weights of: E + normal copper 235 +__21 g and E + excess copper 219+17 g, p<0.05; and Enormal copper 236 +17 g and E - excess copper 199___25 g, p<0.001. Three rats (2 E + , 1 E-) died following partial hepatic lobectomy. Within treatment groups, no difference in the measured parameters was seen between surviving rats who did and who did not undergo lobectomy. Vitamin E deficiency was documented by estimation of serum and hepatic c~-tocopherol concentration after 4 weeks (Table i). The dietary copper supplementation was proved to produce hepatic copper loading in both E + and E - rats (Table 2).
1.5 ml/min, followed by A28o spectrophotometry as previously described (17). Copper was determined in serum diluted 1 : 2 0 by electrothermal atomic absorption spectrophotometry (AAS) using a Pye Unicam SP9 with standards (0.01-0.10 ~tg/ml) prepared in 0.1% (v/v) nitric acid from stock (10 lag/ml) copper nitrate solution (Primar, BDH). Hepatic copper concentrations were determined in freeze-dried liver samples (approximately 10 mg dry wt.) digested in 2 ml 60% nitric acid for 4 h at 80°C. The samples were evaporated to dryness on a hotplate, redissolved in 2 ml of 0.1% nitric acid and analysed by AAS as above. Serum aspartate aminotransferase (AST) activity was determined spectrophotometrically in duplicate 25-~tl samples, according to Sigma technical bulletin 55UV. Serum ornithine transcarbamylase (OCT) activity was determined spectrophotometrically in 100-1al samples (18). Serum oxidase activity, a functional measure of caeruloplasmin, was determined by a modification of the p-phenylenediamine dihydrochloride (PPD) oxidation method (19). Duplicate 5-01 serum samples were added to 0.4 mol/l sodium acetate buffer (pH 5.5) and 3 mmol/1 P P D in a total volume of I ml, and incubated at 37°C for 1 h. Oxidation was terminated by adding sodium azide to a concentration of 0.05% (v/v). After a 30-min incubation at 4°C, the oxidase activity was quantified by A53o spectrophotometry and corrected for spontaneous oxidation of PPD in an azide-inhibited control.
Liver damage Assessed by serum enzymes, vitamin E deficiency produced no liver damage after 4 weeks (Table 1) but at 10 weeks it produced a small but significant rise in AST (Table 3). Copper loading did not elevate AST in either E + or E - rats. OCT was elevated in ED copper-loaded rats (Table 3). Both GaIN and CC14 produced the expected large increases in AST. Prior copper loading significantl~ reduced this effect for both agents in E + rats, an apparent amelioration of the liver injury. In E - rats, the rise in AST produced by GaIN and by CC14 was identical to that in E + rats but the apparent amelioration b3 copper of this effect was reduced. Similarly, serum OCT activity rose briskly after GaIN and CC14 in E + rats. For GAIN, though not for CC14, this rise was significantl~ less in copper-supplemented rats. In E - rats copper supplementation did not significantly reduce the rise in OCT produced by GaIN and CC14 (Table 3).
Statistical analysis Results are expressed as means and standard deviation and the significance of differences between nonparametric groups of data was determined by the MannWhitney U-test.
Results
All rats remained healthy. E - rats gained weight more slowly than E + (from 76+11 g to 195+21 g and from
TABLE I Serum tocopherol,copper, phenylenediamineoxidase,aspartate transaminase (AST),and ornithine transcarbamylase(OCT) and hepatic tocopherol and copper in rats fed either a vitamin E-replete(E +) or vitamin E-deficient(E-) diet for 4 weeks Dietary group E+ E-
Serum (n= 18)
Liver (n= 12)
Tocopherol (lagml) 8.8 (2.5) 1.1 (I.5)
Copper (lag ml) 1.00 I0.20) 1.09 (0.19)
P. oxidase (U) 0.28 (0.03) 0.28 (0.04)
AST (IUI) 53.3 (10.5) 59.6 (12.0)
OCT (mmol/min) 0,17 (0.02) 0.16 (0.02)
Tocopherol (lag/g) 34,3 (5.7) 4.8 (0.8)
Copper (lag/g) 48.3 110.3) 44.1 (6.0)
p < 0.001
n.s.
n.s.
n.s,
n.s.
p < 0.001
n.s.
Values are mean_+(SD). n.s.= not significant
335
HEPATIC ~-TOCOPHEROL A N D LIVER D A M A G E TABLE 2
Liver weight, hepatic ct-tocopherol concentratoin, total or-hepatic tocopherol content, hepatic copper concentration and total copper content (mean+SD) in rats fed either a vitamin E-replete (E+) or vitamin E-deficient (E-) diet with normal or excess copper content, at 24 h following galactosamine (GAIN) 0.85 g/kg i.p. or carbon tetrachloride (CCI4) 10 mmol/kg i.p. and in controls Dietary group
Hepatotoxin
Liver weight (g)
Tocopherol concentration (I-tg/g)
Tocopherol content (~g)
Copper concentration (lag/g)
Copper content (lag)
E + , normal Cu
None GalN CCI 4 None GaiN CCI,, None GaiN CC14 None GaiN CCI,,
7.0 (I.0) 8.7 (1.7) 10.9 (1.2)*** 6.9 (I.5) 7.7 (0.9) 11.7 (1.3)*** 7.4 (0.1) 8.5 (0.3) 11.2 0.7)*** 6.9 (l.0) 7.2 (0.7) I 1.0 ( 1.6)* **
34.4 (2.4) 30.6 (3.1)** 21.0 (2.7)*** 33.5 (3.0) 26.8 (2.7)*** 25.6 0.5)*** 1.8 (I.7) 5.3 (1.6)* 3.2 (1.4) 1.9 (2.0) 3.4 (0.5) 4.0 ( 1.0)**
241 (27) 261 (46) 232 (39) 231 (43) 199 (32) 314 (21)*** 14 (13) 46 (14)*** 36 (20)** 13 (13) 25 (4)* 42 (7)***
50 47 42 3270 3312 1914 49 53 49 3498 3758 2043
0.35 0.41 0.46 22.88 24.85 21.86 0.37 0.45 0.55 23.95 26.71 22.17
E +, excess Cu E-, normal Cu E-, excess Cu
(4) (9) (9) (443) (Ill3) (773)** (13) (7) (5) (982) (1387) (346)* *
(0.07) (0.07) (0.12) (6.20) (5.98) (7.44) (0.13) (0.06) (0.08) (6.22) (8.99) (3.81 )
Values significantly different from those in rats of the same dietary group not receiving GaIN or CC14 are denoted by *p<0.05, **p<0.001. n as in Table 3. TABLE 3 Serum aspartate aminotransferase (AST) and ornithine transcarbamylase (OCT) activity (mean (SD)) in rats fed either a vitamin E-replete (E+) vitamin E-deficient (E-) diet with normal or excess copper content, at 24 h following a hepatotoxin, either galactosamine (GAIN) 0.85 g/kg i.p. or carbon tetrachloride (CC14) 10 mmol/kg i.p. and controls
or
Dietary group
Before hepatotoxin AST flU/I)
OCT (mmol/min)
E + , normal Cu In = 16)
70 (16)"
0.16 (0.03)
E + , excess Cu In= 18)
67 (19) b
0.17 (0.04) b
E-, normal Cu (n=17)
84 (16) c
0.16 (0.03)
E-, excess Cu In= 18)
85 (16) d
0.26 (0.09) d
Significance
0.02 'c
0.01 bd
Hepatotoxin (n)
None 5 GaIN 5 CCI,, 6 None 6 GaIN 6 CCl4 6 None 6 GaIN 6 CCI4 5 None 6 GaIN 6 CCI,, 6
After hepatotoxin AST flU/I)
OCT (mmol/min)
72 (7) 978 (97) 1922 (576) 77 (1 I) 322 (50) 1199 (142) 96 (7) 913 (118) 1926 (466) 159 (46) 644 (117) 1510 (371) see text
O. 17 (0.02) 8.46 (0.60) 8.41 (0.91) 0.20 (O.02) 3.12 (0.42) 7.59 (I.10) 0.17 (0.01) 7.48 (0.06) 7.49 (0. I 0) 0.31 (0.09) 7.68 (0.23) 6.59 (0.13) see text
0.01 bd
For further explanation, see Table 2.
:t- Tocopherol after liver injury Four weeks of dietary copper supplementation did not significantly alter the liver ~t-tocopherol concentration in either E + or E - rats (Table 2). Liver weights were increased following CC14 (Table 2); the hepatic ct-tocopherol results are therefore expressed both as a concentration and as a total hepatic content. In E + rats, with both normal and excess copper, total liver ct-tocopheroi content was unchanged following GaiN. In contrast in the E - rats both the concentration and the content of liver ~-tocopherol were significantly elevated following GAIN. Following CC14 administration, total hepatic ct-tocopherol content was significantly elevated in E + rats fed the high copper diet and in the E - rats fed both normal and excess copper.
Serum PPD oxidase activity Serum oxidase levels were similar in E + and E-rats after 4 weeks of diet (Table I). After 4 weeks Cu supplementation in E + rats ( T = I0), values rose to 0.31 +0.02 units (p<0.01). In E - rats lower values were seen at T = 10, being 0.26+0.06 (p=0.002) in non-Cu supplemented, and 0.27+0.03 in Cu supplemented (p<0.001). Discussion The following conclusions may be drawn from this study of the effect of vitamin E deficiency in rats on copper-and/or acute toxin-induced liver injury.
1. Copper alone did not cause liver injury A 60-fold elevation in hepatic copper concentration over four weeks did not produce elevation of serum AST
336 concentrations and produced only slight elevation in serum OCT activity in E - rats. If this finding can be extrapolated to man, copper loading alone does not provide an explanation for the severe hepatic damage seen in the copper storage diseases ICC and Wilson's disease.
2. Tocopherol deficiency caused minimal liver damage and did not exacerbate damage caused by GaIN or CC14 The AST elevation produced by vitamin E deficiency, though significant, was small. The large increases in serum AST and OCT activities produced by GaIN or CC14 were comparable in E + and E - rats. As in previous studies (I 3,14), the GAIN- or CCl¢-induced rise in AST was markedly less in copper-laden rats. This effect of copper was greater in E + than in E - copperladen rats. Thus vitamin E-deficiency failed to aggravate the hepatic injury although it partially reversed the protective effect of copper. The rise in OCT produced by GaIN was also diminished by copper in E + but not E - rats. The rise in OCT produced by CC1¢ was not diminished by copper in E + or E - rats. Electron-microscopic studies are needed to determine whether the different behaviour of these enzymes indicates differential organelle sensitivity. Serum copper oxidase activity was significantly lower in E - rats, a possible result of greater turnover of caeruloplasmin resulting from its antioxidant function. Whilst a deficiency of other antioxidant protective mechanisms has not been excluded, the present results provide no evidence that they are implicated in clinical copper storage disorders. Although copper does produce lipid peroxidation in isolated hepatocytes (20,21), Sokol et al. (20) postulate that the site of copper-induced hepatocyte injury may be thiol-rich cellular proteins. Iron-induced liver damage is similarly attributed to peroxidative decomposition of membrane phospholipids (22), but Bacon et al. (23) found no evidence of significant mitochondrial lipid peroxidation in ~-tocopherol-deficient rats with a normal hepatic iron content. These findings accord with our observation that ~-tocopherol deficiency in rats failed to aggravate copper- or toxin-induced hepatic injury.
3. Liver injury caused hepatic tocopherol sequestration In E - rats, administration of GaIN or CCla caused liver u-tocopherol to rise. Acute liver injury was therefore
References 1 Sokol RJ, GuggenheimMA, Heubi JE, et al. Frequencyand clinical progression of the vitamin E neurologicdisorder in children with prolonged neonatal cholestasis. Am J Dis Child 1985; 139:121I-5. 2 Tanner AR, Bantock I, Hinks B, et al. Depressed selenium and
L. BARROW et al. associated with retention of ~-tocopheroi within the liver, presumably resulting from tissue redistribution since the diet was ~-tocopherol deficient. In E + rats this effect was also seen with CC1¢. This is an appropriate response to oxidant injury, but its mechanism is unknown. A possible mechanism would be impaired lipoprotein synthesis and/or secretion; however, preliminary evidence shows that accumulation of hepatic a-tocopherol was associated with a fall in the serum ct-tocopheroi/lipid ratio from i.00 in E + rats to 0.18 in E - rats (24). A mechanism similar to that in familial isolated vitamin E deficiency, in which there is failure to incorporate ~-tocopherol into lipoprotein, may obtain (25). Hepatic ct-tocopherol concentration increases after polychlorinated biphenyl administration in the rat (26), but unlike the present model this effect is more marked with high than with low ct-tocopherol dosage. Koremura et al. (26) showed that this was associated with increased ~x-tocopherol absorption rather than tissue redistribution. Two cytosolic ~-tocopherol-binding proteins are reported (27), a high molecular weight storage protein and a transfer protein which exists in two 30.5 kDa isoforms (28). The hepatic content of the transfer protein has been reported to increase in vitamin E deficient rats (29), posssibly as an adaptive response. A similar elevation of ct-tocopheroi concentration in gastric mucosal injury is suggested to be a mechanism of defense for compromised or repairing tissue (30). Clearly the redistribution of ct-tocopherol following acute hepatic insult has important implications for the interpretation of serum levels in patients with liver disease. The mechanism of hepatic cx-tocopherol sequestration requires elucidation.
Acknowledgements We thank Professor D.B. Morton and staff of the Unit of Biomedical Services for assistance with animal husbandry and procedures. L.B. was funded by a grant from the International Copper Research Association. Inc. vitamin E levels in an alcoholic population. Dig Dis Sci 1986; 31 1307-12. 3 JeffreyGP, Muller DPR, Burroughs AK, et al. Vitamin E deficienc~ and its clinical significancein adults with primary biliary cirrhosis and other forms of chronic liver disease. J Hepatol 1984;4: 307-17 4 BarclayLRC. The cooperativeantioxidant role of glutathione witl~
HEPATIC ~t-TOCOPHEROL AND LIVER DAMAGE
5
6
7
8 9
10
I1
12
13 14 15
16
17
a lipid-soluble and a water-soluble antioxidant during peroxidation of liposomes initiated in the aqueous phase and in the lipid phase. J Biol Chem 1988; 263: 16138-42. Wefers H, Sies H. The protection by ascorbate and glutathione against microsomal lipid peroxidation is dependent on vitamin E. Eur J Biochem 1988; 174: 353-7. Tanner MS. Indian childhood cirrhosis. In: Meadows SR, ed. Recent Advances in Paediatrics, Vol. 8. London: Churchill Livingstone, 1986; 103-20. Walshe JM, The liver in Wilson's disease, in: Schiff L, Schiff ER, eds. Diseases of the Liver, 6th Edn. Philadelphia: JB Lippincott Company, 1987; 1037-50. Epstein O, Arborgh B, Sagiv M, et al. Is copper hepatotoxic in primary biliary cirrhosis? J Clin Pathol 1981; 34: 1071-5. Bhave SA, Pandit AN, Tanner MS. Comparison of feeding history in children with Indian childhood cirrhosis and paired controls. J Pediatr Gastroenterol Nutr 1987; 6(4): 652-7. Bhave SA, Pandit AN, Singh S, Walia BNS, Tanner MS. The prevention of Indian childhood cirrhosis. Ann Trop Paediatr, in press. Tanner MS, Bhave SA, Pradhan AM, Pandit AN. Clinical trials of penicillamine in Indian childhood cirrhosis. Arch Dis Child 1987; 62: I 118-24. Kantarjian A, Ghosh A, Tanner MS. Species differences in copper toxicosis and their relevance to Indian childhood cirrhosis (ICCI. Gut 1982; 23: A909. Barrow L, Kantarjian A, Tanner MS. Copper protects against galactosamine-induced hepatitis. J Hepatol 1987; 5: 19-26. Barrow L, Tanner MS. The effect of carbon tetrachloride on the copper-laden rat liver. Br J Exp Pathol 1989; 70: 9-19. Link B, Durk H, Thiel D, Frank H. Binding of trichloromethyl radicals to lipids of the hepatic endoplasmic reticulum during tetrachloromethane metabolism. Biochem J 1984; 223: 577-86. Decker K, Keppler D. Galactosamine induced liver injury. In: Popper H, Schaffner F, eds. Progress in Liver Disease IV. New York: Grune and Stratton, 1972; 183-99. Patel HR, Ashmore SP, Barrow L, Tanner MS. Improved highperformance liquid chromatographic technique for the determination of hepatic ct-tocopherol. J Chromatogr 1989; 45: 269-74.
337 18 Ceriotti G. Ornithine transcarbamylase. In: Bergmeyer HU, ed. Methods of Enzymatic Analysis Vol. 2, 2nd Edit. New York: Academic Press Inc., 1974; 691-8. 19 Ravin HA. An improved colorimetric assay for ceruloplasmin. J Lab Clin Med 1961; 58: 161-8. 20 Sokol RJ, Devereaux MW, Traber MG, Shikes RH. Copper toxicity and lipid peroxidation in isolated rat hepatocytes: effect of vitamin E. Pediatr Res 1989; 25: 55-62. 21 Dillard C J, Tappel AL. Lipid peroxidation and copper toxicity in rats. Drug Chem Toxicol 1984; 7: 477-87. 22 Bacon BR, Tavill AS, Brittenham GM, et al. Hepatic lipid peroxidatioW in vivo in rats with chronic dietary iron overload. J. Clin Invest 1983; 71: 429-39. 23 Bacon BR, Britton RS, O'Neill R. Effects of vitamin E deficiency on hepatic mitochondrial lipid peroxidation and oxidative metabolism in rats with chronic dietary iron overload. Hepatology 1989; 9: 398-404. 24 Barrow L, Ashmore SP, Patel HR, Tanner MS. Do low serum ~t-tocopherol (0t-THI concentrations in liver damage result from hepatic ct-TH sequestration? Gut 1989 30:AI472-3 (Abstract). 25 Traber MG, Sokol RJ, Burton GW, lngold KU, Papas AM, Huffaker JE, Kayden HJ. Impaired ability of patients with familial isolated vitamin E deficiency to incorporate ct-tocopherol into lipoproteins secreted by the liver. J Clin Invest 1990; 85: 397-407. 26 Koremura N, Saito M, Hasegawa T, Innami S. Effect of polychlorihated biphenyls on tissue distribution and intestinal absorption of a-tocopherol in rats. J Nutr Sci Vitaminol 1990; 36: 457-65. 27 Murphy D J, Mavis RD. Membrane transfer of ~-tocopherol: influence of soluble ct-tocopherol-binding factors from the liver, lung, hart and brain of the rat. J Biol Chem 1981; 256: 10464-8. 28 Sato Y, Hagiwara K, Arai H, lnoue K. Purification and characterisation of the alpha-tocopherol transfer protein from rat liver. FEBS Lett 1991; 288: 41-5. 29 Verdon CP, Blumberg JB. Influence of dietary vitamin E on the membrane transfer of ~-tocopherol as mediated by an ~-tocopherol binding protein. Proc Soc Exp Biol Med 1988; 189: 52-60. 30 Yoshikawa T, Yasuda M, Ueda S, Naito Y, Tanigawa T, Oyamada H, Kondo M. Vitamin E in gastric mucosal injury induced by ischaemia-reperfusion. Am J Clin Nutr 1991; 53(1 Suppll: 210S-4S.
Journal of Hepatology, 1992; 16:332-337 01992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved. 0168-8278/92/$05.00
HEPAT 01161
-Tocopherol deficiency fails to aggravate toxic liver injury but liver injury causes -tocopherol retention Linda Barrow, H a s m u k h R. Patel and M.S. Tanner Department of Child Health, University of Leicester, Leicester, United Kingdom
Received10September 1991)
The possible aggravation of liver injury by impaired cellular antioxidant function was investigated. A vitamin E-deficient diet (0.5 mg/kg c~-tocopherol; control 100 mg/kg) significantly reduced rat liver ~-tocopherol concentrations after 4 weeks (!.8 +1.7 lag/g; control 34.4 + 2.4 lag/g, p < 0.001). The effects of copper loading (Cu, 3 g/kg diet); galactosamine (GAIN, 0.85 g/kg i.p.); or carbon tetrachloride (CC14, 10 mmol/kg i.p.) were examined. Serum aspartate transaminase activity was elevated slightly by vitamin E deficiency but not by hepatic copper accumulation. In vitamin E-replete (E+) and vitamin E-deficient (E-) rats, GaIN or CCI 4 caused a large and comparable elevation in serum AST and OCT activity. This effect on AST was markedly reduced by copper loading in vitamin E replete (E +) rats, but in E(-) rats copper had significantly less protective effect. Copper also diminished the OCT response to GaiN in E +, though not E-, rats. A significant rise in total hepatic ~-tocopherol content followed administration of GaIN or CC14 in both normocupric and copper-laden E(-) rats. Thus ct-tocopherol deficiency (a) was not hepatotoxic per se; (b) failed to potentiate the toxicity of copper, GaIN or CC14; but (c) partially abolished the protection by copper against toxininduced liver injury. Retention of hepatic ct-tocopherol after liver damage may partly explain low serum vitamin E levels seen in clinical liver disease. K e y words: Copper; Galactosamine; Carbon tetrachloride; Vitamin E
Low serum vitamin E levels in patients with cholestatic liver disease and alcoholic hepatitis have been attributed to malabsorption and to reduced dietary intake respectively (i-3). ct-Tocopherol, the biologically active homologue of vitamin E, forms an integral part of phospholipid membranes, affording both stability and a free radical scavenging action. Recent work has shown that there is a cooperative interaction of chain-breaking antioxidants in the protection of biological membranes against free radical damage. Glutathione and ct-tocopherol have a synergistic action (4) and ascorbate-mediated protection against lipid peroxidation of microsomal membranes is dependent on the presence of ct-tocopherol (5). A deficiency of vitamin E in the liver may therefore predispose the hepatocyte to injury by free radicals. The effect of vitamin E deficiency on the liver injury produced by copper, carbon tetrachloride (CC14), and galactosamine (GAIN) has therefore been studied. Hepatic copper overload is a feature of Indian child-
hood cirrhosis (ICC) (6), Wilson's disease (7) and chronic cholestatic disorders (8). The copper overload of ICC is of dietary origin (9) and evidence that the stored hepatic copper is toxic includes the prevention of ICC by reduction in dietary copper intake (10) and the therapeutic effect of D-penicillamine in pre-icteric ICC (11) and in Wiison's disease (7). By contrast, in primary biliary cirrhosis, where liver copper is raised as a secondary effect of cholestasis, there is no evidence that the stored copper is harmful (8). Furthermore, in both the lamb and the rat, copper loading alone fails to cause the hepatic features seen in children with ICC (12). Prior hepatic copper-loading in rats fails to act synergistically with subsequently administered acute hepatoxic insults, and indeed unequivocally ameliorates the liver damage induced by both GaIN and CCI4 (13,14). Copper is a transition element with a single unpaired electron in its outer shell, enabling it to participate in free radical-mediated reactions. Copper may potentiate
Correspondence to: Prof. M.S. Tanner, Department of Paediatrics, Children's Hospital, SheffieldS10 2TH, U.K.
HEPATICc~-TOCOPHEROLAND LIVER DAMAGE oxygen toxicity by promoting the formation of hydroxyl radicals which induce site-specific peroxidation of membrane lipids. Conversely, copper may scavenge free radicals and thus act as an antioxidant. The hepatotoxicity of CCI 4 depends on the microsomal production of toxic metabolites which trigger a chain reaction of peroxidative degradation of membrane phopholipids (15) leading ultimately to cellular necrosis. GAIN, in the presence of endogenous gut-derived or exogenous endotoxin, provokes a hepatitis-like injury resulting from a rapid and prolonged uridine nucleotide depletion within parenchyreal cells; the resulting impairment in nucleic acid and glycoprotein synthesis leads to the development of structural changes in hepatocellular membranes (16). GaIN is also reported to reduce superoxide dismutase and catalase activities within hepatocytes, enzymes responsible for scavenging free radical intermediates. This report describes the influence of ~-tocopherol deficiency in rats on the liver damage produced by hepatic copper loading, by acute hepatic insults (GAIN or CCh), or by these agents in conjunction. These experimental conditions were designed to model possible clinical situations in which an acute insult such as viral hepatitis occurred in a copper-loaded tocopheroldeficient liver. The influence of these hepatotoxins on the liver c~-tocopherol content is also described.
333
Normalcopper (0,01 glkg) n-18 I VitaminE-replete (100 mg/kg) a-36
~_~ Excesscopper (3 g~g) n=t8
Weanling male rats from an inbred strain (WAGS, Harlan Olac Ltd., Oxon, U.K.) were housed in groups of 6 in constant environmental conditions under veterinary supervision. Throughout the study, 36 rats were fed ad libitum a diet deficient in vitamin E (0.5 mg/kg) comprising edible casein 20 g, sucrose 60 g, groundnut oil 10 g, vitamins and minerals (SDS Ltd., Essex, U.K.). Thirty-six rats were fed the same diet supplemented with vitamin E to normal levels (100 mg/kg) (Fig. 1). These rats are referred to as E- and E +, respectively. After 4 weeks (T=4, Fig. 1) all rats were bled under halothane anaesthesia from a tail vein. Aliquots of serum were stored at -20°C for a maximum of 3 months. Seven days later (T=5), 24 E + and 24 E- rats underwent a liver lobectomy. Using buprenorphine, 0.15 mg/kg s.c. analgesia and halothane anaesthesia, a small section of the right median lobe of the liver was removed at laparotomy. Duplicate 100-mg wet weight pieces of liver were frozen at -20°C for up to 14 days, and duplicate pieces of 20-30 mg were freeze-dried. After a further 7 days (T--6), the two treatment groups were subdivided. 18 E- and 18 E + were given their respective diets
+ GaIN
n-6
+ CCI4
n-6
J
NOinjection n=61 +GAIN
n=6)
+ CCI4
n-61
8•_• 8•
NOinjection n=61
Normalcopper (0,01 g/kg] n=l I VitaminE-deficientdiet~ (0.5 mg/kg) n=36
+ GaiN
n=6~
+ CCI4
n=6~
NOinjection n-61
Excesscopper (3 g/kg) n=l
I 0
+ GaiN
n-6)
+ CCI4
n=6)
I Dietarycopper l normal/excess Vitamin E.repleteor Vitamn E-de c ent J 4 5 6 10 weeks Tailbleed Liver Iobectomy
Materials and Methods
~
No Injection n-6 ~
Injection TaiIbleedand sacrifice
Fig. I. Experimentaldesign. Details of dosages are as in the text.
supplemented with copper sulphate, from a normal copper concentration of l0 mg/kg to a level of 3 g/kg (Fig. 1). After 4 weeks ( T = 10), all rats were again bled from a tail vein under halothane anaesthesia, and aliquots of serum stored at -20°C. Each of the four groups of 18 rats was then further subdivided; 6 rats received GaIN (0.85 g/kg body wt., i.p.) in aft injection volume of approximately 0.5 ml (range 0.40-0.59 ml), 6 rats received CC14 (10 mmol/kg body wt., i.p.) in an injection volume of approximately 0.5ml (range 0.44-0.56ml), and 6 rats received no injection (Fig. I). Twenty-four h later blood samples were taken from the tail vein of all rats under haiothane anaesthesia and the rats sacrificed by cervical dislocation. The liver was removed, weighed and duplicate pieces of 100 mg wet wt. were frozen at -20°C, and of 20-30 mg were freeze-dried.
Assays Hepatic u-tocopherol concentrations were determined by H P L C separation on a C8 ultrasphere column using pure methanol as the mobile phase at a flow rate of
334
L. BARROW et al 76 + 12 g to 206+ 17 g respectively after 6 weeks. p<0.05). Copper supplementation caused both groups initially to lose weight, the E - rats to 183 + 23 g, and the E + to 202+18 g after 1 week (p < 0.01). Rats in the four dietary groups then gained weight at a comparable rate during the remaining 3 weeks of the study, resulting in final weights of: E + normal copper 235 +__21 g and E + excess copper 219+17 g, p<0.05; and Enormal copper 236 +17 g and E - excess copper 199___25 g, p<0.001. Three rats (2 E + , 1 E-) died following partial hepatic lobectomy. Within treatment groups, no difference in the measured parameters was seen between surviving rats who did and who did not undergo lobectomy. Vitamin E deficiency was documented by estimation of serum and hepatic c~-tocopherol concentration after 4 weeks (Table i). The dietary copper supplementation was proved to produce hepatic copper loading in both E + and E - rats (Table 2).
1.5 ml/min, followed by A28o spectrophotometry as previously described (17). Copper was determined in serum diluted 1 : 2 0 by electrothermal atomic absorption spectrophotometry (AAS) using a Pye Unicam SP9 with standards (0.01-0.10 ~tg/ml) prepared in 0.1% (v/v) nitric acid from stock (10 lag/ml) copper nitrate solution (Primar, BDH). Hepatic copper concentrations were determined in freeze-dried liver samples (approximately 10 mg dry wt.) digested in 2 ml 60% nitric acid for 4 h at 80°C. The samples were evaporated to dryness on a hotplate, redissolved in 2 ml of 0.1% nitric acid and analysed by AAS as above. Serum aspartate aminotransferase (AST) activity was determined spectrophotometrically in duplicate 25-~tl samples, according to Sigma technical bulletin 55UV. Serum ornithine transcarbamylase (OCT) activity was determined spectrophotometrically in 100-1al samples (18). Serum oxidase activity, a functional measure of caeruloplasmin, was determined by a modification of the p-phenylenediamine dihydrochloride (PPD) oxidation method (19). Duplicate 5-01 serum samples were added to 0.4 mol/l sodium acetate buffer (pH 5.5) and 3 mmol/1 P P D in a total volume of I ml, and incubated at 37°C for 1 h. Oxidation was terminated by adding sodium azide to a concentration of 0.05% (v/v). After a 30-min incubation at 4°C, the oxidase activity was quantified by A53o spectrophotometry and corrected for spontaneous oxidation of PPD in an azide-inhibited control.
Liver damage Assessed by serum enzymes, vitamin E deficiency produced no liver damage after 4 weeks (Table 1) but at 10 weeks it produced a small but significant rise in AST (Table 3). Copper loading did not elevate AST in either E + or E - rats. OCT was elevated in ED copper-loaded rats (Table 3). Both GaIN and CC14 produced the expected large increases in AST. Prior copper loading significantl~ reduced this effect for both agents in E + rats, an apparent amelioration of the liver injury. In E - rats, the rise in AST produced by GaIN and by CC14 was identical to that in E + rats but the apparent amelioration b3 copper of this effect was reduced. Similarly, serum OCT activity rose briskly after GaIN and CC14 in E + rats. For GAIN, though not for CC14, this rise was significantl~ less in copper-supplemented rats. In E - rats copper supplementation did not significantly reduce the rise in OCT produced by GaIN and CC14 (Table 3).
Statistical analysis Results are expressed as means and standard deviation and the significance of differences between nonparametric groups of data was determined by the MannWhitney U-test.
Results
All rats remained healthy. E - rats gained weight more slowly than E + (from 76+11 g to 195+21 g and from
TABLE I Serum tocopherol,copper, phenylenediamineoxidase,aspartate transaminase (AST),and ornithine transcarbamylase(OCT) and hepatic tocopherol and copper in rats fed either a vitamin E-replete(E +) or vitamin E-deficient(E-) diet for 4 weeks Dietary group E+ E-
Serum (n= 18)
Liver (n= 12)
Tocopherol (lagml) 8.8 (2.5) 1.1 (I.5)
Copper (lag ml) 1.00 I0.20) 1.09 (0.19)
P. oxidase (U) 0.28 (0.03) 0.28 (0.04)
AST (IUI) 53.3 (10.5) 59.6 (12.0)
OCT (mmol/min) 0,17 (0.02) 0.16 (0.02)
Tocopherol (lag/g) 34,3 (5.7) 4.8 (0.8)
Copper (lag/g) 48.3 110.3) 44.1 (6.0)
p < 0.001
n.s.
n.s.
n.s,
n.s.
p < 0.001
n.s.
Values are mean_+(SD). n.s.= not significant
335
HEPATIC ~-TOCOPHEROL A N D LIVER D A M A G E TABLE 2
Liver weight, hepatic ct-tocopherol concentratoin, total or-hepatic tocopherol content, hepatic copper concentration and total copper content (mean+SD) in rats fed either a vitamin E-replete (E+) or vitamin E-deficient (E-) diet with normal or excess copper content, at 24 h following galactosamine (GAIN) 0.85 g/kg i.p. or carbon tetrachloride (CCI4) 10 mmol/kg i.p. and in controls Dietary group
Hepatotoxin
Liver weight (g)
Tocopherol concentration (I-tg/g)
Tocopherol content (~g)
Copper concentration (lag/g)
Copper content (lag)
E + , normal Cu
None GalN CCI 4 None GaiN CCI,, None GaiN CC14 None GaiN CCI,,
7.0 (I.0) 8.7 (1.7) 10.9 (1.2)*** 6.9 (I.5) 7.7 (0.9) 11.7 (1.3)*** 7.4 (0.1) 8.5 (0.3) 11.2 0.7)*** 6.9 (l.0) 7.2 (0.7) I 1.0 ( 1.6)* **
34.4 (2.4) 30.6 (3.1)** 21.0 (2.7)*** 33.5 (3.0) 26.8 (2.7)*** 25.6 0.5)*** 1.8 (I.7) 5.3 (1.6)* 3.2 (1.4) 1.9 (2.0) 3.4 (0.5) 4.0 ( 1.0)**
241 (27) 261 (46) 232 (39) 231 (43) 199 (32) 314 (21)*** 14 (13) 46 (14)*** 36 (20)** 13 (13) 25 (4)* 42 (7)***
50 47 42 3270 3312 1914 49 53 49 3498 3758 2043
0.35 0.41 0.46 22.88 24.85 21.86 0.37 0.45 0.55 23.95 26.71 22.17
E +, excess Cu E-, normal Cu E-, excess Cu
(4) (9) (9) (443) (Ill3) (773)** (13) (7) (5) (982) (1387) (346)* *
(0.07) (0.07) (0.12) (6.20) (5.98) (7.44) (0.13) (0.06) (0.08) (6.22) (8.99) (3.81 )
Values significantly different from those in rats of the same dietary group not receiving GaIN or CC14 are denoted by *p<0.05, **p<0.001. n as in Table 3. TABLE 3 Serum aspartate aminotransferase (AST) and ornithine transcarbamylase (OCT) activity (mean (SD)) in rats fed either a vitamin E-replete (E+) vitamin E-deficient (E-) diet with normal or excess copper content, at 24 h following a hepatotoxin, either galactosamine (GAIN) 0.85 g/kg i.p. or carbon tetrachloride (CC14) 10 mmol/kg i.p. and controls
or
Dietary group
Before hepatotoxin AST flU/I)
OCT (mmol/min)
E + , normal Cu In = 16)
70 (16)"
0.16 (0.03)
E + , excess Cu In= 18)
67 (19) b
0.17 (0.04) b
E-, normal Cu (n=17)
84 (16) c
0.16 (0.03)
E-, excess Cu In= 18)
85 (16) d
0.26 (0.09) d
Significance
0.02 'c
0.01 bd
Hepatotoxin (n)
None 5 GaIN 5 CCI,, 6 None 6 GaIN 6 CCl4 6 None 6 GaIN 6 CCI4 5 None 6 GaIN 6 CCI,, 6
After hepatotoxin AST flU/I)
OCT (mmol/min)
72 (7) 978 (97) 1922 (576) 77 (1 I) 322 (50) 1199 (142) 96 (7) 913 (118) 1926 (466) 159 (46) 644 (117) 1510 (371) see text
O. 17 (0.02) 8.46 (0.60) 8.41 (0.91) 0.20 (O.02) 3.12 (0.42) 7.59 (I.10) 0.17 (0.01) 7.48 (0.06) 7.49 (0. I 0) 0.31 (0.09) 7.68 (0.23) 6.59 (0.13) see text
0.01 bd
For further explanation, see Table 2.
:t- Tocopherol after liver injury Four weeks of dietary copper supplementation did not significantly alter the liver ~t-tocopherol concentration in either E + or E - rats (Table 2). Liver weights were increased following CC14 (Table 2); the hepatic ct-tocopherol results are therefore expressed both as a concentration and as a total hepatic content. In E + rats, with both normal and excess copper, total liver ct-tocopheroi content was unchanged following GaiN. In contrast in the E - rats both the concentration and the content of liver ~-tocopherol were significantly elevated following GAIN. Following CC14 administration, total hepatic ct-tocopherol content was significantly elevated in E + rats fed the high copper diet and in the E - rats fed both normal and excess copper.
Serum PPD oxidase activity Serum oxidase levels were similar in E + and E-rats after 4 weeks of diet (Table I). After 4 weeks Cu supplementation in E + rats ( T = I0), values rose to 0.31 +0.02 units (p<0.01). In E - rats lower values were seen at T = 10, being 0.26+0.06 (p=0.002) in non-Cu supplemented, and 0.27+0.03 in Cu supplemented (p<0.001). Discussion The following conclusions may be drawn from this study of the effect of vitamin E deficiency in rats on copper-and/or acute toxin-induced liver injury.
1. Copper alone did not cause liver injury A 60-fold elevation in hepatic copper concentration over four weeks did not produce elevation of serum AST
336 concentrations and produced only slight elevation in serum OCT activity in E - rats. If this finding can be extrapolated to man, copper loading alone does not provide an explanation for the severe hepatic damage seen in the copper storage diseases ICC and Wilson's disease.
2. Tocopherol deficiency caused minimal liver damage and did not exacerbate damage caused by GaIN or CC14 The AST elevation produced by vitamin E deficiency, though significant, was small. The large increases in serum AST and OCT activities produced by GaIN or CC14 were comparable in E + and E - rats. As in previous studies (I 3,14), the GAIN- or CCl¢-induced rise in AST was markedly less in copper-laden rats. This effect of copper was greater in E + than in E - copperladen rats. Thus vitamin E-deficiency failed to aggravate the hepatic injury although it partially reversed the protective effect of copper. The rise in OCT produced by GaIN was also diminished by copper in E + but not E - rats. The rise in OCT produced by CC1¢ was not diminished by copper in E + or E - rats. Electron-microscopic studies are needed to determine whether the different behaviour of these enzymes indicates differential organelle sensitivity. Serum copper oxidase activity was significantly lower in E - rats, a possible result of greater turnover of caeruloplasmin resulting from its antioxidant function. Whilst a deficiency of other antioxidant protective mechanisms has not been excluded, the present results provide no evidence that they are implicated in clinical copper storage disorders. Although copper does produce lipid peroxidation in isolated hepatocytes (20,21), Sokol et al. (20) postulate that the site of copper-induced hepatocyte injury may be thiol-rich cellular proteins. Iron-induced liver damage is similarly attributed to peroxidative decomposition of membrane phospholipids (22), but Bacon et al. (23) found no evidence of significant mitochondrial lipid peroxidation in ~-tocopherol-deficient rats with a normal hepatic iron content. These findings accord with our observation that ~-tocopherol deficiency in rats failed to aggravate copper- or toxin-induced hepatic injury.
3. Liver injury caused hepatic tocopherol sequestration In E - rats, administration of GaIN or CCla caused liver u-tocopherol to rise. Acute liver injury was therefore
References 1 Sokol RJ, GuggenheimMA, Heubi JE, et al. Frequencyand clinical progression of the vitamin E neurologicdisorder in children with prolonged neonatal cholestasis. Am J Dis Child 1985; 139:121I-5. 2 Tanner AR, Bantock I, Hinks B, et al. Depressed selenium and
L. BARROW et al. associated with retention of ~-tocopheroi within the liver, presumably resulting from tissue redistribution since the diet was ~-tocopherol deficient. In E + rats this effect was also seen with CC1¢. This is an appropriate response to oxidant injury, but its mechanism is unknown. A possible mechanism would be impaired lipoprotein synthesis and/or secretion; however, preliminary evidence shows that accumulation of hepatic a-tocopherol was associated with a fall in the serum ct-tocopheroi/lipid ratio from i.00 in E + rats to 0.18 in E - rats (24). A mechanism similar to that in familial isolated vitamin E deficiency, in which there is failure to incorporate ~-tocopherol into lipoprotein, may obtain (25). Hepatic ct-tocopherol concentration increases after polychlorinated biphenyl administration in the rat (26), but unlike the present model this effect is more marked with high than with low ct-tocopherol dosage. Koremura et al. (26) showed that this was associated with increased ~x-tocopherol absorption rather than tissue redistribution. Two cytosolic ~-tocopherol-binding proteins are reported (27), a high molecular weight storage protein and a transfer protein which exists in two 30.5 kDa isoforms (28). The hepatic content of the transfer protein has been reported to increase in vitamin E deficient rats (29), posssibly as an adaptive response. A similar elevation of ct-tocopheroi concentration in gastric mucosal injury is suggested to be a mechanism of defense for compromised or repairing tissue (30). Clearly the redistribution of ct-tocopherol following acute hepatic insult has important implications for the interpretation of serum levels in patients with liver disease. The mechanism of hepatic cx-tocopherol sequestration requires elucidation.
Acknowledgements We thank Professor D.B. Morton and staff of the Unit of Biomedical Services for assistance with animal husbandry and procedures. L.B. was funded by a grant from the International Copper Research Association. Inc. vitamin E levels in an alcoholic population. Dig Dis Sci 1986; 31 1307-12. 3 JeffreyGP, Muller DPR, Burroughs AK, et al. Vitamin E deficienc~ and its clinical significancein adults with primary biliary cirrhosis and other forms of chronic liver disease. J Hepatol 1984;4: 307-17 4 BarclayLRC. The cooperativeantioxidant role of glutathione witl~
HEPATIC ~t-TOCOPHEROL AND LIVER DAMAGE
5
6
7
8 9
10
I1
12
13 14 15
16
17
a lipid-soluble and a water-soluble antioxidant during peroxidation of liposomes initiated in the aqueous phase and in the lipid phase. J Biol Chem 1988; 263: 16138-42. Wefers H, Sies H. The protection by ascorbate and glutathione against microsomal lipid peroxidation is dependent on vitamin E. Eur J Biochem 1988; 174: 353-7. Tanner MS. Indian childhood cirrhosis. In: Meadows SR, ed. Recent Advances in Paediatrics, Vol. 8. London: Churchill Livingstone, 1986; 103-20. Walshe JM, The liver in Wilson's disease, in: Schiff L, Schiff ER, eds. Diseases of the Liver, 6th Edn. Philadelphia: JB Lippincott Company, 1987; 1037-50. Epstein O, Arborgh B, Sagiv M, et al. Is copper hepatotoxic in primary biliary cirrhosis? J Clin Pathol 1981; 34: 1071-5. Bhave SA, Pandit AN, Tanner MS. Comparison of feeding history in children with Indian childhood cirrhosis and paired controls. J Pediatr Gastroenterol Nutr 1987; 6(4): 652-7. Bhave SA, Pandit AN, Singh S, Walia BNS, Tanner MS. The prevention of Indian childhood cirrhosis. Ann Trop Paediatr, in press. Tanner MS, Bhave SA, Pradhan AM, Pandit AN. Clinical trials of penicillamine in Indian childhood cirrhosis. Arch Dis Child 1987; 62: I 118-24. Kantarjian A, Ghosh A, Tanner MS. Species differences in copper toxicosis and their relevance to Indian childhood cirrhosis (ICCI. Gut 1982; 23: A909. Barrow L, Kantarjian A, Tanner MS. Copper protects against galactosamine-induced hepatitis. J Hepatol 1987; 5: 19-26. Barrow L, Tanner MS. The effect of carbon tetrachloride on the copper-laden rat liver. Br J Exp Pathol 1989; 70: 9-19. Link B, Durk H, Thiel D, Frank H. Binding of trichloromethyl radicals to lipids of the hepatic endoplasmic reticulum during tetrachloromethane metabolism. Biochem J 1984; 223: 577-86. Decker K, Keppler D. Galactosamine induced liver injury. In: Popper H, Schaffner F, eds. Progress in Liver Disease IV. New York: Grune and Stratton, 1972; 183-99. Patel HR, Ashmore SP, Barrow L, Tanner MS. Improved highperformance liquid chromatographic technique for the determination of hepatic ct-tocopherol. J Chromatogr 1989; 45: 269-74.
337 18 Ceriotti G. Ornithine transcarbamylase. In: Bergmeyer HU, ed. Methods of Enzymatic Analysis Vol. 2, 2nd Edit. New York: Academic Press Inc., 1974; 691-8. 19 Ravin HA. An improved colorimetric assay for ceruloplasmin. J Lab Clin Med 1961; 58: 161-8. 20 Sokol RJ, Devereaux MW, Traber MG, Shikes RH. Copper toxicity and lipid peroxidation in isolated rat hepatocytes: effect of vitamin E. Pediatr Res 1989; 25: 55-62. 21 Dillard C J, Tappel AL. Lipid peroxidation and copper toxicity in rats. Drug Chem Toxicol 1984; 7: 477-87. 22 Bacon BR, Tavill AS, Brittenham GM, et al. Hepatic lipid peroxidatioW in vivo in rats with chronic dietary iron overload. J. Clin Invest 1983; 71: 429-39. 23 Bacon BR, Britton RS, O'Neill R. Effects of vitamin E deficiency on hepatic mitochondrial lipid peroxidation and oxidative metabolism in rats with chronic dietary iron overload. Hepatology 1989; 9: 398-404. 24 Barrow L, Ashmore SP, Patel HR, Tanner MS. Do low serum ~t-tocopherol (0t-THI concentrations in liver damage result from hepatic ct-TH sequestration? Gut 1989 30:AI472-3 (Abstract). 25 Traber MG, Sokol RJ, Burton GW, lngold KU, Papas AM, Huffaker JE, Kayden HJ. Impaired ability of patients with familial isolated vitamin E deficiency to incorporate ct-tocopherol into lipoproteins secreted by the liver. J Clin Invest 1990; 85: 397-407. 26 Koremura N, Saito M, Hasegawa T, Innami S. Effect of polychlorihated biphenyls on tissue distribution and intestinal absorption of a-tocopherol in rats. J Nutr Sci Vitaminol 1990; 36: 457-65. 27 Murphy D J, Mavis RD. Membrane transfer of ~-tocopherol: influence of soluble ct-tocopherol-binding factors from the liver, lung, hart and brain of the rat. J Biol Chem 1981; 256: 10464-8. 28 Sato Y, Hagiwara K, Arai H, lnoue K. Purification and characterisation of the alpha-tocopherol transfer protein from rat liver. FEBS Lett 1991; 288: 41-5. 29 Verdon CP, Blumberg JB. Influence of dietary vitamin E on the membrane transfer of ~-tocopherol as mediated by an ~-tocopherol binding protein. Proc Soc Exp Biol Med 1988; 189: 52-60. 30 Yoshikawa T, Yasuda M, Ueda S, Naito Y, Tanigawa T, Oyamada H, Kondo M. Vitamin E in gastric mucosal injury induced by ischaemia-reperfusion. Am J Clin Nutr 1991; 53(1 Suppll: 210S-4S.