Alcohol, Vol.9, pp. 341-348, 1992

0741-8329/92 $5.00 + .00 Copyright©1992PergamonPress Ltd.

Printed in the U.S.A. All rightsreserved.

Combined Effects of Ethanol and Protein Deficiency on Hepatic Iron, Zinc, Manganese, and Copper Contents A . C O N D E - M A R T E L , * E. G O N Z A L E Z - R E I M E R S , *l F. S A N T O L A R I A - F E R N A N D E Z , * V. C A S T R O - A L E M A N , t L. G A L I N D O - M A R T f N , I " F. R O D R f G U E Z - M O R E N O * AND A. MARTfNEZ-RIERA*

Hospital Universitario de Canarias, *Departmento de Medicina Interna, and tDepartmento de Qu[mica Anal[tica, Facultad de Medicina, Universidad de La Laguna, Canary Islands, Spain Received 4 O c t o b e r 1991; A c c e p t e d 11 M a r c h 1992 CONDE-MARTEL, A., E. GONZALEZ-REIMERS, F. SANTOLARIA-FERNANDEZ, V. CASTRO-ALEMAN, L. GALINDO-MARTfN, F. RODRfGUEZ-MORENO AND A. MARTfNEZ-RIERA. Combinedeffects of ethanoland protein deficiency on hepatic iron, zinc, manganese, and copper contents. ALCOHOL 9(5) 341-348, 1992.-The present study has been performed in order to establish the relative and combined roles of ethanol and malnutrition on liver Fe, Zn, Cu, and Mn alterations in alcoholic male adult Wistar rats, and also the relationships between these alterations and histomorphometricully determined hepatocyte and nuclear areas, perivenular fibrotic rim area, and total amount of fat present in the liver. Four groups of 8 animals each were fed: (1) a nutritionally adequate diet (C); (2) a 36070ethanol-containing (as percent of energy), isocaloric diet (A); (3) a 2eTeprotein-containing, isocaloric diet (PD); and (4) a 36tfe ethanol, 2°70 protein-containing, isocaloric diet (A-PD), respectively, following the Lieber-DeCarli model. Ethanol-fed, protein-deficient animals showed the highest liver Fe, and the lowest Zn and Cu values, although differences in liver Zn, Mn, and Cn values were not significantly different between PD and A-PD groups. Statistically significant differences of these parameters were observed between the A and the A-PD groups, and between the A and PD groups, except for liver iron. Except for liver Mn, differences between C and A groups were statisticallysignificant. These alterations correlated with liver fibrosis and steatosis, serum albumin, and weight loss, except for fiver Mn, which was not correlated with fibrosis or steatosis. Thus, protein deficiency seems to enhance ethanol-induced fiver Fe, Zn, and Cu alterations, whereas protein deficiency, but not ethanol, seems to play a major role on liver Mn alterations. Iron Zinc Copper Manganese Alcoholic fiver disease Fibrosis Steatosis Liver histomorphometry

SOME trace elements, especially iron and zinc, play important roles as cofactors of several enzymes involved in collagen synthesis (1-3) and other potentially hepatotoxic metabolic events (4). Hepatic iron overload, observed in 30% of alcoholics with chronic liver disease (5), and also in severe protein malnutrition (6), alters lysosomal membranes, and favors lipid peroxidation, both factors leading to hepatocyte necrosis (7,8). Regarding its role on hepatic fibrogenesis, it has been shown that iron activates transcription o f genes responsible for collagen synthesis (9), enhances hepatic prolylhydroxilase activity (10), and increases hepatic collagen fibrils content (11). Hepatic iron overload in chronic alcoholics is more closely related to steatosis than to fibrosis (5), although it has been shown that iron promotes collagen synthesis in cultured human fibro-

Malnutrition

Protein deficiency

blasts (12), and that fiver iron concentrations o f over 22,000 ppm are associated with liver fibrosis and cirrhosis in patients affected by hemochromatosis (13), even in the absence o f coexistent alcoholic fiver disease. Zinc depletion is a classical marker of protein wastage (14). Low hepatic zinc levels have been described in alcoholics (1517) and may be responsible for progressive fiver fibrosis. Indeed, Anttinen et al. (2) have shown that zinc supplementation hampers carbon tetrachloride-induced fiver fibrosis. Manganese depletion is commonly observed in malnutrition (18). Liver manganese has been reported to be elevated in alcoholic liver disease, probably because of impaired biliary excretion (16). Manganese acts as cofaetor of enzymes involved in collagen synthesis (l), and, in this way, manganese overload may

Requests for reprints should be addressed to E. GonzAlez-Reimers, Hospital Universitario de Canarias, La Laguna, Tenerife, Canary Islands, Spain. 341

342

CONDE-MARTEL ET AL.

affect hepatic fibrogenesis. On the contrary, ethanol-fed miniature pigs showed enhanced activity of manganese superoxide dismutase when compared with controls (19), with this metalloenzyme being considered as a scavenger mechanism against free radicals, exerting a protective effect against cell necrosis. Although there is general agreement in the observation of elevated liver copper levels in cholestatic syndromes (20) and in alcoholic cirrhosis (21), controversy exists about the effect of ethanol on liver copper changes (22-24), as well as regarding serum copper levels in alcoholic patients with or without liver damage (25-27). Both liver copper excess, by promoting necrosis (20) and enhancing lysyl-oxidase activity (1), and depletion-altering superoxide dismutase activity (28), may affect liver fibrogenesis. Malnutrition may cause copper deficiency (29). Because many alcoholic patients suffer variable degrees of malnutrition (30-33), it is important to establish the relative and combined roles of ethanol and malnutrition on hepatic trace element alterations. This is the aim of the present study. METHOD Thirty-two adult male Wistar rats were divided into four groups of 8 animals each, which were fed the following diets [Lieber-DeCarli diets (34,35), "Dyets," Bethlehem, PA]: 1. Nutritionally adequate, control diet (protein 18070, fat 35070, carbohydrates 47070, all of them as percent of energy). 2. Control, protein-deficient diet (protein 2070, fat 35070, carbohydrates 63070, as percent of energy). 3. Nutritionally adequate, alcohol diet (protein 18%, fat

TABLE 2 LIEBER-DECARL1 ETHANOL RAT DIET Ingredient Regular Casein (100-mesh) L-Cystine DL-Methionine Corn oil Olive oil Safflower oil Maltose dextrin Cellulose Salt mix Vitamin mix Choline bitartrate Xanthan gum Ethanol 95°70 2.0°70 caloric protein Casein (100-mesh) L-Cystine DL-Methionine Corn oil Olive oil Safflower oil Maltose dextrin Cellulose Salt mix Vitamin mix Choline bitartrate Xanthan gum Ethanol 950/o

g/1 of Diet 41.4 0.5 0.3 8.5 28.4 2.7 25.6 10.0 8.75 2.5 0.53 3.0 67 (ml) 4.6 0.06 0.03 8.5 28.4 2.7 64.8 10.0 8.75 2.5 0.53 3.0 67 (ml)

TABLE 1 LIEBER-DECARLI CONTROL RAT DIET Ingredient

g/I of Diet

Regular Casein (100-mesh) L-Cystine DL-Methionine Corn oil Olive oil Safflower oil Maltose dextrin Cellulose Salt mix Vitamin mix Choline bitartrate Xanthan gum

41.4 0.5 0.3 8.5 28.4 2.7 115.2 10.0 8.75 2.5 0.53 3.0

2.0% caloric protein Casein (100 mesh) L-Cystine r~L-Methionine Corn oil Olive oil Safflower oil Maltose dextrin Cellulose Salt mix Vitamin mix Choline bitartrate Xanthan gum

4.6 0.06 0.03 8.5 28.4 2.7 155.7 10.0 8.75 2.5 0.53 3.0

35%, carbohydrates 11070, ethanol 36070, as percent of energy). 4. Alcohol, protein-deficient diet (protein 2070, ethanol 36070, carbohydrates 27070, fat 35070). The detailed compositions of these diets are outlined in Tables 1, 2, and 3. All the diets were supplemented with identical amounts of regular salt and vitamin mixes, with vitamin mix being added at the moment at which the diet was prepared in the liquid suspension, as recommended elsewhere (34). Liquid suspension was prepared every 6-8 days. The rats receiving the alcohol, protein-deficient diet were limiting with respect to the dietary intake of the rats of the other three groups. The pair-feeding process was repeated every 2 days, so that the mean amount of diet consumed was nearly identical in the four groups (Table 4). Two months later the animals were anesthetized by pentobarbital and sacrificed. Liver specimens destined to light microscope examinations were fbted in formalin, embedded in paraffin, and cut into serial longitudinal sections. The latter were stained with hematoxylin-eosin and Van Giesson. The following parameters were histomorphometricaily determined using a W I D S II image analyzer: (a) Area of the perivenular fibrotic rim of at least 10 pericentral veins. In order to avoid errors due to differences in the diameter of the veins, we have calculated the ratio: area of the fibrotic rim/area of the vessel. (b) Total amount of fat: mean area of fat droplets x number

HEPATIC TRACE ELEMENTS, ALCOHOL, MALNUTRITION

343

TABLE 3 LIEBER-DECARLI LIQUID DIETS FOR VITAMIN AND SALT MIXES Lieber-DeCarli Liquid Diet Vitamin Mix (Use at 2.5 g/I Diet)

Salt Mix for Lieber-DeCarfiLiquid Diet (Use at 8.75 g/l Diet)

Ingredient

g/kg

Ingredient

g/kg

Thiamin HC1 Riboflavin Pyridoxine HC1 Niacin Calcium pantothenate Folic acid Biotin Vitamin BI2 (0.1%) Vitamin A acetate (500,000 IU/g) Vitamin D3 (400,000 IU/g) Vitamin E acetate (500 IU/g) Menadione sodium bisulfite p-amino benzoic acid Inositol Dextrose

0.6 0.6 0.7 3.0 1.6 0.2 0.02 10.0 4.8 0.4 24.0 0.08 5.0 10.0 939.0

Calcium phosphate, dibasic Sodium chloride Potassium citrate, monohydrate Potassium sulfate Magnesium oxide Manganous sulfate Ferrous sulfate Zinc carbonate Cupric carbonate Potassium iodate Sodium selenite Chromium potassium sulfate Sodium fluoride Sucrose, finely powdered

500.0 74.0 220.0 52.0 24.0 4.6 4.95 1.6 0.3 0.01 0.01 0.55 0.06 117.92

o f fat droplets in the liver section × lO0/total area o f the section). (c) Hepatocyte and nuclear areas o f at least 20 cells o f each three periportal and three pericentral areas. These measures (a-c) were m a d e at 400x. A part o f each o f the liver specimens was dehydrated, weighed, and solved in 65% nitric acid (Merck) and 10% hydrogen peroxide, and processed for trace element determination by atomic absorption spectrophotometry using a PerkinElmer 3030 B atomic absorption speetrophotometer. Liver, iron, copper, manganese, and zinc were also determined in this way. Blood samples were removed before sacrifice and destined to serum albumin determination. M e a n weight change o f the rats o f each o f the groups was also recorded; the first parameter serves as an indicator o f visceral proteins, and the second one, as a measure o f the intensity o f marasmus-type malnutrition.

Statistics The different parameters mentioned were compared between the four groups using analysis of variance (ANOVA), and, after, the Student-Newman-Keuls test. Liver iron, copper, manganese, and zinc were analyzed by two-way A N O V A to establish the separate and interactive effects o f ethanol and protein deficiency. Single correlation studies were performed to determine the significance o f the relation between two quantitative variables. RESULTS

Weight Change and Serum Albumin Weight changes experienced by the different experimental groups are shown in Table 4. A significant decrease in serum albumin was observed in rats fed diets II, III, and IV (Fig. 1). No differences were observed between ethanol-fed and pro-

TABLE 4 MEAN CONSUMPTION OF CALORIES, PROTEIN, ALCOHOL, AND TRACE ELEMENTS: WEIGHT AT THE BEGINNING AND AT THE END OF THE STUDY ( ~ + SD) Group Control(I) Protein deficient (II) Alcohol (III) Alcohol + prot. def. (IV)

Control (I) Protein deficient (II) Alcohol(III) Alcohol + p r o t . def. (IV)

Energy (kcal/d) 61.7 60.4 59.6 58.5

+ + ± ±

4.2 3.5 5 7

Protein (g/d)

Alcohol (g/d)

2.55 0.28 2.47 0.27

0 0 3.79 + 0.32 3.72 + 0.45

324 322 316 310

Cu O~g/d) Copm/d)

± + + +

0.17 0.02 0.21 0.04

Fe O,g/d) (ppm/d)

Mn (~s/d) (ppm/d)

Zn Og/d) (ppm/d)

985 964 951 933

904 885 873 857

449 440 434 426

± + + ±

67 56 80 112

± + + +

62 51 73 103

± + + +

31 25 36 51

Weight at Start (g)

84 82 81 79

+ + + +

± + + ±

18 17 20 29

6 5 7 9

Weightat End-Point (g) 392 259 313 204

+ + + +

11 16 32 33

344

C O N D E - M A R T E L ET AL.

ALBUMIN Oil 380 "

3.5 3

300"

i!ii ii ii! i

/--T-- 7

,f

2SO'

..............., +:¢.:.:t..:.Z

I ~

2.8 .'T/ i

2

":~'~:':'::':':

ililili!!i!!ii i!i i i i! ~:+:.:.:-:.:,:.:.:.:.:.:~.:.1

1.5

!i!i!!!

1 0.8

/

0

!

IIl

I

IV M

ALCOHOL ALC,PROT.OEF:.

100 SO' 0

!!!ii!C!ii!!iii! iiiiiil! i ~ii~iii~iii!ii!~?~iljil]

.:~,:,:,:.:.:

CON[ROL

11 ~-'~PROTEIN DEF.

1150'

iiiiiiii~iiiiiiiii!iiiiiiii}t

,~:.:,:+:....,

i

200'

PERICENTRAL

--:::::::::::::::::::::::::::::/-i!iiiiiiiiii!i!i!!ii!il,--;

pLO,O00[

PERIPORTAL TOTAL p=O.O04

p.£:')i:~:[

0.8 0

, CONTROL

! PROT-DEF

I

II

! ALCOHOL

IZI

f19.54 p