Attenuation of Alcohol-induced Hepatic Fibrosis by Polyunsaturated Lecithin CHARLES s. LIEBER,LEONORE M. DECARLI,KI M. W, CHO-IL KIM AND MARIA A. LEO Section of Liver Disease and Nutrition, Alcohol Research and Treatment Center, Bronx Veterans Affairs Medical Center and Mount Sinai School of Medicine (CUNY), New York, New York

urban areas (1).Treatment and prevention of the fibrotic process, however, are still elusive. Protein, methionine and choline deficiencieshave been incriminated in the pathogenesis of alcoholic liver injury for several decades because in growing rats, lack of dietary protein and lipotropic factors (choline and methionine) can produce a fatty liver (2). It also has been reported that ethanol increases choline requirements in the rat (3), possibly by enhancing choline oxidation (4). Primates, however, are far less susceptible to protein and lipotrope deficiency than rodents (5).Clinically, choline treatment of patients suffering from alcoholic liver injury has been found to be ineffective in the face of continued alcohol abuse (6-9). Furthermore, massive supplementation with choline failed to prevent the fatty liver produced by alcohol in volunteer subjects (10). Moreover, fatty liver and fibrosis (including cirrhosis) developed in baboons despite liberal amounts of methionine (11)and massive supplementation with choline, even to the point of toxicity (12). However, in addition to its possible role as a lipotrope, methionine plays an essential role in phospholipid metabolism and membrane structure and function (13). Methionine supplementation has been contemplated for the treatment of liver diseases, especially the alcoholic variety,. but some difficulties have been encountered. Indeed, excess methionine was shown to have some adverse effects (14) including a decrease in hepatic ATP (15). Furthermore, whereas in some patients with alcoholic liver disease, circulating methionine levels are normal (16), elevated levels have been reported in others (17-19). Moreover, Kinsell et al. (20) have observed a delay in the clearance of plasma methionine Alcoholic liver injury is a major public health problem after its systemic administration to patients with liver in Western countries, with cirrhosis being the fourth damage. Similarly, Horowitz et al. (21) reported that largest cause of death in the 25 to 64 yr age group in the blood clearance of methionine after an oral load of this amino acid was slowed in such subjects. Because about half of the methionine is metabolized by the liver, the above observations suggest impaired hepatic Received March 26, 1990; accepted June 27, 1990. metabolism of this amino acid in patients with alcoholThis study was supported by Department of Health and Human Services ic liver disease. Indeed, Duce et al. (22) reported degrants AA03508, DK 32810 and the Department of Veterans Affairs. Presented in part at the International Symposium on Phospholipids held creases of S-adenosylmethionine (AdoMet) synthetase December 4, 1989. in Cologne, FRG. (EC 2.5.1.6) and phosphatidylethanolamine-N-methylAddress reprint requests to: Charles S. Lieber, M.D. (151/G), Alcohol transferase (EC 2.1.1.17) activities in cirrhotic livers. Research and Treatment Center, Bronx Veterans Affaira Medical Center, Bronx. Thus, this double enzyme deficiency (decreased AdoMet NY 10468. synthetase and reduced methyltransferase activities) 31/1/24800

Characteristic features of alcoholic liver iqjury include fibrosis and striking membrane alterations, with associated phospholipid changes. To offset some of these abnormalities, a 10-yr study was conducted in baboons: 12 animals (eight females, four males) were fed a liquid diet supplemented with polyunsaturated lecithin (4.1 mg/kcal) for up to 8 yr, with either ethanol (50% of total energy) or isocaloric carbohydrate. They were compared with another group of 18 baboons fed an equivalent amount of the same diet (with or without ethanol), but devoid of lecithin. In the two groups, comparable increases in lipids developed in the ethanol-fed animals, but striking differences in the degree of fibrosis were eeen. Whereas at least septal fibrosis (with cirrhosis in two) and transformation of their lipocytes into transitional cells developed in seven of the nine baboons fed the regular diet with ethanol, septal fibrosis did not develop in any animals fed lecithin (p c 0.005). They did not progress beyond the stage of perivenular fibrosis (sometimes associated with pericellular and perisinusoidal fibrosis) and had a significantly lesser activation of lipocytes to transitional cells. Furthermore, when three of these animals were taken off lecithin, but continued on the same amount of the ethanol-containing diet, they rapidly (within 18 to 21 mo) progressed to cirrhosis, accompanied by an increased transformation of their lipocytes to transitional cells. These results indicate that some component of lecithin exerts a protective action against the fibrogenic effects of ethanol. Because we had previously found that choline, in amounts present in lecithin, has no comparable action, the polyunsaturated phospholipids themselves might be responsible for the 1990;12:1390-1398.) protective effect. (HEPATOLOGY

1390

withdrawn from the study after 5 to 5.5 yr. One baboon fed ethanol and PUL for 3 yr. and two fed this regimen for 8 yr and three corresponding animals fed the PUL-control diet (all of which had maintained adequate food intake) underwent a second phase of the study in which PUL was withdrawn and the regular diet (with or without ethanol) was given for an additional 2 yr. All animals either maintained or slightly gained weight, and their general appearance remained normal throughout the MATERIALS AND METHODS study. Percutaneous liver biopsy specimens were obtained with the Thirty baboons (eight males, 22 females) (Pupioharnadryus 1 were studied in compliance with the Institution’s guidelines animals under ketamine anesthesia 6 hr after food withdrawal. for animal research and were gven four different diets. Nine Similar samples were taken from each of the groups. For light were fed our regular nutritionally adequate liquid diet con- microscopy and electron microscopy, specimens were coded for taining 100 mg cholineiL or per 1,000 kcal (11, 24) and blind reading and prepared and examined as described (12). modified as follows: the contents of the daily vitamin and The volume density of lipid droplets in lipocytes (It0 cells. mineral tablet given to the animals has now been incorporated fat-storing cells) in the perisinusoidal space and fibrous scars in the diet, with only one of those tablets gwen per week and, was estimated as reported before (28), using electron microin addition, these ingredients were changed to the following graphs of lipocytes (at x 12,500 or x 20,000 magnification). A concentrations: vitamin B,, (10 @L), menadione (250 pg/L), point grid with 0.5-pm point-to-point distance was superimascorbic acid (200 mg/L), manganese (15 mg/L). iron (100 posed on the lipocytes, and the number of points overlying the mg/L), copper (3 mg/L), zinc ( 5 mg/L). In addition, selenium lipid droplets was counted; this value was expressed as the (25 F ~ / L )chromium , I120 pg/L) and fluoride (6.25 m dL ) have percentage of the cell volume. As defined and validated before been added, and xanthan gum is now used as the stabilizer. (28).cells were considered to be lipocytes when the volume The oil mixture consisted of corn (5.1gmiL), olive (17.1 gmiLi density of lipid droplets was greater than 20% of the cell and safflower (1.6 gm/L) oils with the following fatty acid volume; conversely, cells were considered to be transitional composition (in percentage of total): myristic (14:0 ) 0.02%; cells when the volume of lipid droplets was less than 20% of the palmitic (16:O) 13.34; palmitoleic (16: 11, 1.2%;stearic (18:O) cell. After either 24 or 84 mo, liver total lipids and triglycerides 2.6%:oleic(18: 1152.7%;linoleic(18:2)28.4%;linolenic(18:31 0.7% and arachidonic ( 20: 4) 0.1%. Nine other animals were were determined (in each group) as described before (29) in matched individually with these baboons according to gender, those baboons in which sufficient tissue was available. Serum body weight ( 10 to 20 kg) and approximate age (4 to 6 yr) and AST activity was measured according to Karmen (30); serum pair-fed with the same diet except for isocaloric replacement of glutamate dehydrogenase activity was assessed by the method carbohydrate (50%of total energy I with ethanol. The feeding of Ellis and Goldberg (31); ALT was determined according technique has been reported in detail previously ( 11 1. The to Wroblewsky and LaDue (32). Measurements were also average daily intake was 69.9 t 6.3 mlkglday. This diet performed in the serum of total proteins (Bio-Rad Protein provided 1 kcal/ml and a daily supply of energy, proteins. Assay kit, Bio Rad, Rockville Center, NY),albumin, alkaline vitamins and minerals well within the recommended daily phosphatase and bilirubin (Sigma Diagnostics Kits, Sigma Chemical Co., St. Louis, MO). allowances for the baboon I 25 1. For statistical analysis, the chi-square test and Student’s Twelve animals (eight females. four males) were fed the same diets, but supplemented with PUL ( 4 . 1 gm/L of Lethi- t test were used (331,and the means were expressed with their con), a soybean extract kindly provided by the American standard error. Lecithin Co. (Woodside, NY).containing 60% phosphatidylcholine, a maximum of 30% of phosphatidylethanolamine, 6% RESULTS phosphatidic acid, 3% monophosphatidylinositol, 3% of lysophosphatidylcholine and the following fatty acid composition: All alcohol-fed baboons had fatty livers w h e n exmyristic (14:O) 0.01%; palmitic I 16:O) 15.4%; palmitoleic amined histologically after 6 o r 12 mo. Fat accumulation t 1 6 : l ) 0.2%; stearic ( 1 8 : O ) 3.4%;oleic (18:1) 10.8%;linoleic was found i n all 15 baboons fed ethanol, w h e th e r the (18:2) 61.5%; linolenic ( 1 8 : 3 ) 7.2% and arachidonic (20:4i animals had been fed a diet supplemented with PUL 0.2%. The PUL dose selected was comparable to the 3 to 4 @day of essential phospholipid used clinically (26) and also cn = 6 ) o r not ( n = 9).The degree of steatosis was provided 0.4 mg cholinekcal. Its energy contribution was less similar and comparable t o the fat accumulation dethan 4% of the total. The diet was again given with either scribed previously after ethanol feeding ( l l 1. This w a s ethanol (50% of total energy) or isocaloric carbohydrate for up confirmed by triglyceride measurements in 12 animals: to 8 yr. These animals were also matched and fed in pairs i n six baboons fed ethanol (w i t h PUL supplementation) isocaloric amounts of either the PUL-control or PUL-alcoholfor 2 yr, liver triglycerides averaged 160.1 5 39.2 mg/gm containing diets. The average daily intake was 71.0 5.6 (vs. 18.1 5 4.1 mg/gm i n the pair-fed controls fed PUL ml/kg/day. Thus diet and alcohol intake were comparable in alone); the corresponding mean values after 7 yr were both groups of baboons and resulted in blood ethanol levels 158.3 mg/gm (n = 2) in the alcohol-fed and 9.9 mg/gm (measured according to Korsten et al. 1271) fluctuating be- ( n = 2) in the correspondmg controls. In nine baboons tween 20 to 40 mmoVL. In both groups, the animals were fed up to 8 yr, except for two baboons of each alcohol-fed group, in fed the regular ethanol-containing diet for 2 yr, liver which the study was limited to 2.5 to 3 -yr because loss of triglycerides were 137.0 5 21.0 mg/gm in the ethanol appetite and associated insufficient food and alcohol intake fed baboons (vs. 10.3 t 0.3 mg/gm in the corresponding developed in the animals. For similar reasons, one baboon fed pair-fed controls). After 7 yr, the results averaged 94.1 ethanol and PUL and two animals fed ethanol alone were mg/gm ( n = 2) and 8.0 mg/gm ( n = 21, respectively.

may promote mem br ane injury ( 2 3 )a n d ot her damage documented in alcohol-induced liver disease. The question arose wh e th er such deficiency could be att enuat ed, at least in part, by bypassing the enzym e defects t hrough phospholipid supplementation. T h i s was tested in baboons in the present s t udy by feeding alcohol i n a diet supplemented with polyunsaturated lecithin (PUL1.

+

1392

LIEBER ET AL.

Y

"

0

1

2

3

4

5

6

-f

7

8

YEARS

FIG.1. Sequential development of alcoholic liver injury in baboons fed ethanol with a normal diet for up to 8 yr. Liver morphology in pair-fed controls (not shown) remained normal.

The differences between the groups fed PUL and those that were not were not statistically significant. Thus, whereas ethanol had the expected effect in terms of lipid accumulation, PUL did not appear to be protective in that respect. As described before (12), the ethanol-induced steatosis was associated with mitochondrial alterations, particularly giant forms. These appeared as round, eosinophilic bodies by light microscopy and the mitochondrial nature of these bodies was confirmed by electron microscopy. Similar changes were observed in the baboons given ethanol with PUL. The one major difference between the groups was in terms of fibrosis. Whereas a t least septal fibrosis (with cirrhosis in two) (Fiz. 1)developed in all eight baboons fed our regular ethanol-containing diet for 6 yr or more, none of the animals fed PUL went beyond the stage of fatty liver (Figs. 2a and 2b) with perivenular fibrosis, sometimes associated with some interstitial fibrosis (pericellular and perisinusoidal) (Figs. 2b and 3).When analyzed in terms of the development of septal fibrosis, this difference was statistically significant (chi-square test; p < 0.005). Furthermore, when three of these ethanol-PUGtreated animals were taken off PUL but continued on the ethanol-containing diet, they rapidly progressed to more severe stages of fibrosis within 1yr, with septal fibrosis in one and cirrhosis in another (Fig. 4). After 18 to 21 mo, these three animals had all progressed to cirrhosis, as illustrated in Figures 2c and 2d. An associated striking alteration in lipocytes was seen. In the baboons that had septal fibrosis after alcohol feeding, an expected increase was noted in the appearance of transitional cells (81% vs. 8% in controls; Table 1)as described before (28). Whereas PUL by itself

HEPATOLOGY

had no effect on the occurrence of transitional cells (4% in PUL-fed baboons), when given in combination with ethanol, PUL prevented, at least in part, the increase in the number of transitional cells (55%vs. 81%in alcoholfed animals; p < 0.01 by chi-square test comparing two proportions). By contrast, on withdrawal of PUL, but continuation of ethanol, the number of transitional cells rapidly increased to 82% (p < 0.01; chi-square test) in association with the accelerated fibrosis, as demonstrated in the two baboons in which sufficient tissue was available to study this process (Table 1, Fig. 5). I n the corresponding PUL-fed nonalcohol controls, no changes in lipocytes were observed on PUL withdrawal. One baboon died 27 mo after PUL withdrawal with a nonobstructive cholestasis and obvious cirrhosis (Fig. 6). After PUL withdrawal, blood liver tests in the alcoholfed group changed as follows (average values for 21 mo before the switch vs. 21 mo thereafter): serum AST (54 L 10 to 108 2 20 IUL ; p < 0.051, ALT (69 2 5 to 153 19 IUL; p < 0.005), glutamate dehydrogenase (882 16 to 177 51 I U L at 37" C; NS), total proteins (7.52 L 0.16 to 6.86 k 0.15 gm/dl; p < 0.025), albumin (4.60 L 0.13 to 4.18 k 0.14 gm/dl; p < 0.051, bilirubin (0.098 0.040 to 0.161 2 0.053 mg/dl; p < 0.05) and alkaline phosphatase (122 k 18 to 245 2 23 I U L ; p < 0.001). The difference in blood liver tests between the conventional and PUL-supplemented alcohol-fed group was not significant except for serum albumin (3.98 ? 0.13 vs. 4.60 k 0.13 gm/dl; p < 0.005).

*

*

DISCUSSION

This study confirms our previous results that demonstrated that, in the baboon, feeding of ethanol results in hepatic fibrosis (including cirrhosis) even when associated with an adequate diet; we now show that this effect is attenuated by supplementing the diet with PUL and is exacerbated on PUL withdrawal. The capacity of ethanol to produce cirrhosis in the baboon despite an adequate diet has been challenged by Ainley et al. (34). The diet administered by Ainley et al. (34) however, was not fully defined and, in any event, differed from ours. Furthermore, some question exists concerning the amount of alcohol administered to Ainley's animals. Surprisingly, very high and most likely lethal amounts of alcohol (25 gm/kg of body weight/day) were reportedly administered, but the animals survived. Their blood ethanol levels were not higher than those we observed with four to five times less alcohol (namely, 5 to 6 gm/kg/day) (35, 36). In fact, as acknowledged by Ainley et al. (341, it is not known how much alcohol (or diet) was actually consumed by their animals. Moreover, as pointed out by French (371, the diet was administered only to two control animals, four ethanol-fed baboons and four animals fed ethanol with a zinc supplement, without pair-feeding and with mixed gender. It is therefore difficult to make significant comparisons between the experimental groups. In addition, only two of these baboons were given ethanol with the regular

Vol. 12, No. 6, 1990

1393

POI.YYNSA'I'1:KATEI) LECITHIN A'T'IXNL:AI'ES ALCOHOLIC FIBROSIS

FIG.2. Baboons fed ethanol and Pl;L for 3 caJ and 8 tb) yr, respectively. In addition to fat, mild interstitial (a)or perivenular, pericellular and perisinusoidal (b) fibrosis is present. Same baboons as in ( a ) and rbb a t 18 mo Ic) and 24 mo rdl after PUL withdrawal. Cirrhosis is present (chromotrope aniline blue. magnification 150).

TABLE 1. Effects of PUL feeding on the appearance of transitional cells expressed in percentage of total lipocytes" Control with Control

PUL supplementation

Alcohol-fed

8% (26)

4Q 1261

Sl ?"

Alcohol-fed with PUL supplementation

Alcohol-fed after withdrawal from PUL

55% 153)

829kh

(49)

~~

"Numberof cells evaluated is indicated in parentheses. 'Significantly different from alcohol-fed with PUL supplementation ' p < 0.01 ).

(61)

1394

HEPATOLOGY

LIEBER ET AL. i

//i

I

I

Crrhows

I t

Cirrhosis

!

I I

I I

I septal Fibrosis

+++ Fot ++ +

i !i

Perlvffubr and/or

r

lntefSll1lDl

I1

Flbosls I

,

i

12

0 I

MONTHS

,

0

24

1

2

3

4

5

6

7

8

YEARS

FIG.3. Sequential development of alcoholic liver injury in baboons fed ethanol with a PUL-supplemented diet. Liver morphology in pair-fed controls given PUL without lecithin (not shown) remained normal.

alcohol-containing diet for a period exceeding 18 mo. Because the results of this study (Fig. 1)and those of our previous studies (11,381 showed that a longer period of treatment is required to produce septal fibrosis or cirrhosis, it is obvious that the number of animals and the duration of the study by Ainley et al. (34) do not allow for definitive conclusions. In the aggregate of the present and previous studies (12,35,38,39), production of cirrhosis was observed in 13 of a total of 63 baboons fed ethanol for 3 yr or more and septal fibrosis developed in an additional 13 animals. The present results also revealed that PUL exerts some protective action against the fibrogenic effects of ethanol. Indeed, although three of the four animals fed alcohol with PUL for 3 yr or more had perivenular and/or interstitial fibrosis, none of them progressed to a more severe stage (Fig. 3) such as septal fibrosis or cirrhosis, as observed in the unsupplemented baboons (Fig. 1). In view of the relatively small number of animals involved, however, the question might be raised whether such a protective action could have occurred only by chance, reflecting, for instance, animal variability and resistance to cirrhosis in those baboons. In terms of development of septal fibrosis, however, the protective effect of PUL was statistically significant (p < 0.005). Furthermore, on withdrawal of PUL in three of these baboons, cirrhosis did develop in all these animals (Fig. 4), although they had been refractory to ethanol before (Fig. 3). In fact, the progression appeared to be more rapid than in the regular alcohol-fed group (Fig. 11,suggesting that withdrawal of PUL unleashed a stimulatory effect of ethanol on fibrogenesis that had been kept in check by the PUL. The mechanism of the protective effect of PUL is not

FIG.4. Effect of PUL withdrawal on sequential development of alcoholic liver injury in baboons. Three of the animals shown in Figure 3 (and depicted with the same symbols) were studied here.

clear at present. Lecithin has a high choline content, to which some protective effect has been ascribed in the past, but this was restricted to rodents rather than primates, as reviewed in the introduction. Interest in choline was revived by Mezey et al. (40) when they failed to produce fibrosis in four monkeys given ethanol with a diet containing four times the amount of choline used in the baboon that had developed cirrhosis with ethanol in this or in previous studies (11, 38). This observation raised the possibility that choline might have exerted a protective effect. However, other explanations for the lack of fibrosis in the four monkeys studied by Mezey et al. (40) are more plausible. These include the small number of animals studied; in four monkeys, cirrhosis might have been missed because of our 63 baboons fed alcohol, only 13had cirrhosis (vide supra). Furthermore, because monkeys are smaller than baboons and have a higher metabolic rate, the same amount of alcohol is not comparable in the two species. Indeed, consistently elevated blood alcohol levels were not reported in the monkeys of Mezey et al. (40). Another possible explanation may be species difference in susceptibility to alcohol. The latter possibility was strengthened by the observation of Rogers, Fox and Gottlieb (411, who showed that even when fed diets with reduced choline levels, cirrhosis did not develop in monkeys with alcohol. Furthermore, by showing that fibrosis developed in the baboon, even when supplemented with large amounts of choline, a subsequent study clearly indicated that ethanol itself must be incriminated rather than a relative deficiency in choline induced by the ethanol (12). Moreover, because the amount of choline that had been supplemented before was equal to that contained in the PUL used in the present study, it is apparent that PUL must act by some mechanism other than providing free choline. It is conceivable, however, that PUL might represent a modality of choline supplementation with

less toxicity and/or greater bioavailability than that of native choline. Indeed, it has been shown that in terms of central nervous effects, when choline is provided as lecithin, its biological activity is significantly increased (421, possibly because of bacterial degradation of free choline in the gut. Whether this also pertains to the alcohol-induced liver injury is uncertain and, in fact, unlikely. Indeed, if this mechanism were operative, it would imply that in animals fed alcohol, a functional state of choline deficiency exists despite an ample (and even excessive) dietary supply ( 12). Furthermore, the liver lesions produced by alcohol differ significantly from those resulting from choline deficiency, both chemically and ultrastructurally: orotic acid, which attenuates the choline deficiency fatty liver, has no such effect on the ethanol variety (43). Moreover, whereas choline deficiency is associated with a reduction in circulating lipoproteins, including high-density lipoproteins (441, the opposite is true of alcohol (45); the increased incorporation of palmitic acid-.'H into lipoproteins after alcohol contrasts with the decrease observed in choline deficiency (46). In addition, hepatic carnitine is decreased by choline deficiency (471, but increased after ethanol feeding (48).Ultrastructurally, the two types of fatty liver also differ (491. An alternate and more plausible hypothesis is that PUL exerts its beneficial effects not simply as a source of choline but by providing some other component. PUL is rich in linoleic acid but this fatty acid per se is probably not responsible for the protective effect because the basic diet was supplemented with linoleate and contained large amounts of corn oil (11) that is rich in linoleic acid. Furthermore, this fatty acid has been incriminated as a permissive rather than as a protective factor in alcoholic liver injury (50, 51). Therefore, it is possible that the phospholipids themselves might be responsible for some of the protective effects. Indeed, phospholipids rich in essential fatty acids have a high bioavailability. More than 50%of orally administered polyunsaturated phosphatidylcholine is made biologcally available for the organism either by intact absorption (lesser extent) or by reacylation of absorbed lysophosphatidylcholine (greater extent) (52). Pharmacokinetic studies in man using "HP4C-labeled phosphatidylcholine showed the absorption to exceed 90% (53).Similar observations were made in animals (54-56). Indeed, phosphatidylcholine in the diet is degraded by pancreatic phospholipase A, (57), and the products ( 1-lysophosphatidyl-cholineand fatty acids) are absorbed in the jejunum (58).Various authors (59-61) found persistent phosphatidylcholine accumulation in the liver during the first 24 to 48 hr after administration. Polyene phosphatidylcholine was taken up in toto by the liver cells and incorporated into the membrane-containing fractions (56). Thus phosphatidylcholine may directly influence membrane structures and provide a basis for some of the beneficial effects of essential phospholipids in the treatment of liver diseases (62). That ethanol administration does indeed alter cellular phospholipids has been shown in various species. In rat hepatic mitochondria, decreases in total

Fic. 5 Electron micrograph of 'i transitional cell f T C )in a baboon liver after PUL withdrawal The cell is seen in a small fibrous septum between two hepatocytes and has a lipid volume density of less than 20% of the cell volume P marks the cell process of an adjacent cell 1 magnification x 6,2501

phospholipid (63)and in the polyene phospholipid fatty acids have been described (64).In ethanol-fed baboons, the total phospholipid content of the mitochondrial membranes was diminished with a significant decrease in the levels of phosphatidylcholine (65). These alterations in the phospholipid composition of the mitochondrial membranes appeared responsible for some of the depression of cytochrome oxidase activity produced by chronic ethanol consumption (65).This, in turn, may be the cause, a t least in part, for the biochemical alterations of baboon hepatic mitochondria after chronic ethanol consumption (66). The mechanism whereby chronic ethanol consumption alters phospholipids has not been clarified but may be related to decreased phosphatidylethanolamine-N-methyltransferase (EC 2.1.1.17) activity described in cirrhotic liver (22). That this is not simply a consequence of the cirrhosis but may, in fact, be a primary defect related to alcohol is suggested by the observation that the enzyme activity is already decreased in baboons fed ethanol before the development of cirrhosis (Leo et al., Unpublished observation). Apart from their structural functions in cell membranes, the polyunsaturated fatty acids serve as precursors to the prostaglandins and related substances. The prostaglandins can, in general terms, be described

1396

LIEBER ET AL.

HEPATOLOGY

FIG.6. Gross appearance of the liver of one of the baboons shown in Figure 3 at time of death 27 mo after PUL withdrawal. Surface (a) and cross-section (b) show typical gross appearance of the nodular transformation of cirrhosis.

as a defensive regulatory system, but their postulated role in alcohol-induced liver damage (67) remains largely undefined. It is known, however, that chronic adminis tration of ethanol is associated with abnormalities in essential fatty acid levels in various lipid fractions (68-70). In addition to dietary factors, increased lipid peroxidation (69) by ethanol administration has been incriminated. The beneficial effect of PUL on the fibrosis induced experimentally by alcohol (as shown in the present study) may be of clinical relevance, in keeping with positive therapeutic effects described in human cirrhosis (71, 72). Histologically, alcohol-induced fat accumulation was decreased (731, whereas the present baboon study revealed no significant action in terms of fat accumulation after alcohol, possibly reflecting species difference. In nonalcoholic liver disease, beneficial effects have also been reported histologically in terms of recovery from kwashiorkor (74), in HBsAg-positive patients with cirrhosis (75) and in patients with chronic active hepatitis in terms of inflammatory parameters (76). Thus, in liver diseases of diverse origins, polyunsaturated lecithins appear to exert some relatively nonspecific beneficial effects (possibly through membrane stabilization) that might also affect alcohol-induced liver injury. However, the inhibition of alcohol-induced fibrosis, demonstrated in the present study, may also reflect a more specific action. Indeed, some selective ethanol-induced hepatic phospholipid abnormalities (vide supra) have been described; the effect of PUL on these specific defects, however, remains to be investigated. Alternatively or additionally, PUL could promote

the resorption of collagen, as suggested for a model of experimental toxic cirrhosis (77). Furthermore, the change of lipocytes into transitional cells induced by ethanol (28)was slowed down in the presence of PUL. In view of the putative role of those transitional cells in the process of fibrogenesis (78,79), it is conceivable that the slowing, by PUL, of the ethanol-induced fibrosis may be a consequence of a blocking effect of PUL on the transformation of the lipocytes. Whatever the mechanism, we may conclude that, in the baboon, feeding of lecithin rich in polyunsaturated fatty acids slows the ethanol-induced fibrogenesis and that its withdrawal is associated with a striking acceleration of the fibrotic process, leading to rapid development of cirrhosis.

Acknowledgments: We thank Ms. N. Lowe, Mr. C. Din and Mr. J. Saeli for expert technical assistance; Ms. P. Walker and Ms. R. Cabell for typing the manuscript and Mr. B. Seabrook and Ms. D. Copeland for providing baboon care. REFERENCES 1. Summary of Vital Statistics 1986. Department of Health, The city of New York. Bureau of Health Statistics and Analysis. 2. Best CH, Hartroft WS, Lucas CC, Ridout JH. Liver damage produced by feeding alcohol or sugar and its prevention by choline. Br Med J 1949;2:1001-1006. 3. Klatskin G, Krehl WA, Conn HO. The effect of alcohol on the choline requirement. I. Changes in the rat’s liver following prolonged ingestion of alcohol. J Exp Med 1954;100:605-614. 4. Thompson JA, Reitz RC. Studies on the acute and chronic effects of ethanol ingestion on choline oxidation. Ann NY Acad Sci 1976;273:194-204. 5. Hoffbauer FW,Zaki FG. Choline deficiency in baboon and rat compared. Arch Pathol 1965;79:364-369.

Vol 12. No 6. 1990

1’01,YL NSA‘I’rRA’I’ED LE(’1’I’HIN A‘I‘TKNrATKS .4IA’OHOLI(‘ FIBROSIS

6. Olson RE. Modern nutrition in health and disease. In: Wohl M G Goodhart WS, eds. Philadelphia Idea& Febiger. 1964:1037-1050 7 . Phillips GB. Davidson (3. Acute hepatic insufficiency of thc chronic alcoholic. Arch Intern Med 1954;94:585-603, 8 . Post JJ, Benton B. Breakstone R. Hoffman J. The effects of diet and choline on fatty infiltration of the human liver. Gastroenterology 1952;20:403-410. 9. Volwiler W, Jones CM. Mallory ‘I’B.(:riteria Ior the measurement of results of treatment in fatty cirrhosis. Gastroenterologv 1948;1 1 : 164- 182. 10. Rubin E. Lieber CS. Alcohol induced hepatic injury in nonalcoholic volunteers. N Engl J Med 1968:278:869-876. 1 1 . Lieber CS. DeCarli LM. An experimental model of alcohol feeding and liver injury in the baboon J Med Primatol 1974;3:153-163. 12. Lieber CS, Leo MA, Mak KM, DeCarli LM. Sat0 S. Choline fails to prevent liver fibrosis in ethanol-fed baboons but causes toxicity HEP~TOL(K:Y 1985:5:561-c575. 13. Mato JM. Progress in protein-lipid interactions. Vol. 2. New York Elsevier Scientific Publishing. 1986:267. 14. Finkelstein J D , Martin JJ. Methionine metabolism in mammals. adaptation to methioninr excess J Bin1 Chem 1986;261:15821587. 15. Hardwick DF. Applegarth UA. (’ockcroft DM. Ross PM, Cder RJ Pathogenesis of methionine-induced toxicity. Metabolism 1970. 19:381-391. 16. lob V, Coon WW. Sloan M. Free amino acids in liver, plasma and muscle of patients with cirrhosis of the liver. .I Surg Res 1967;7:41-43. 17. Iber FL, Rosen H, Stanley MA, Levenson SM. Chalmers TC The plasma amino acids in patients with liver failure. J Lab Clin Med 1957;50:417-425. 18. Fischer J E , Yoshimura N.Aguirre A. James J H . Cummings MG. Abel RN, Deindoerfer F Plasma amino acids in patients with hepatic encephalopathy. An1 J Surg 1974: 127:40-47. 19. Montanari A, Simoni I. Vallisa D. Trifiro A. Colla R. Abbiati R. Borghi L. et al. Free amino acids in plasma and skeletal muscle of patients with liver cirrhosis. Ht.iax I OI.O(;Y 1988;s:1034-1039. 20. l n s e l l L, Harper HA, Barton H(’, Michaels GD, Weiss HA. Rate of disappearance from plasma of intravenously administered methionine in patients with liver damage. Science 1947;106:589594. 21. Horowitz J H , Rypins EB. Henderson J M , Heymsfield SB. Moftitt SD, Bain RP. Chawla RK. et al Evidence for impairment of transsulfuration pathway in cirrhosis. Gastroenterology 1981:Xl. 668-675. 22. Duce AM, Ortiz P, Cabrero (:, Mato JM S-adenosyl-i -methimine synthetase and phospholipid methyltransferase are inhibited in human cirrhosis. Ht:PAroi.I 1988;8:65-68. 23. Yamada S, Mak KM. Lieber CS (’hronic ethanol consumption alters rat liver plasma membranes and potentiates release of alkaline phosphatase. Gastroenterology 1985;88:1799-1806. 24. Lieber CS, DeCarli LM. Animal models of ethanol dependence and liver injury. Fed Proc 1976:35.1232-1236. 25. Nicolosi W ,Hunt RD. Dietary allowances for nutrients in non-human primates. In- Ilayes KC’. ed. Primates in nutritional research. New York: Academic Press. Inc 1979:ll-37. 26. Knuchel F. Doppelblindstudie bei Patienten mit alkoholtoxisrher Fettleber. Med Welt 1979;:30:411-416 27. Korsten MA. Matsuzaki S. Feinman I.. Lieber CS. High blood acetaldehyde levels after ethanol administration: differences between alcoholic and non-alcoholic subjects. N Engl J Med 1975;292:386-389 28. Mak KM, Leo MA, Lieber CS. Alcoholic liver injury in baboons transformation of lipocytea to transitional cells. Gastroenterology 1984;87:188-200. 29. Lieber CS. Rubin E. Alcoholic fatty liver in man on a high protein and low fat diet. Am J Med 1968;44:200-206. 30. Karmen A. Transaminase activity in human blood. J Clin Invest 1955;34:131-133. 31. Ellis G, Goldberg DM. Optimal conditions for the kinetic assay of serum glutamate dehydrogenase activity a t 37 degrees C . Clin Chem 1972;18:523-527. 32. Wroblewsky F. LaDue JS Serum glutamic pyruvic transaminase ~

1397

i n cardiac and hepatic disease. Proc Soc Exp Biol Med 1956;91: 569-571 :X3. Snedecor GW, Cochran WG. Statistical methods. 7th ed. Ames. Iowa: Iowa State University Press, 1980. :{4. h n l e y CC, Senapati A. Brown IMH, Slavin BM, Davies DR. Keeling PWN, Thompson RPH. Is alcohol hepatotoxic in the baboon? J Hepatol 1988;7:85-92. 35. Lieber CS. DeCarli LM, Rubin E. Sequential production of fatty liver, hepatitis and cirrhosis in sub-human primates fed ethanol with adequate diets. Proc Natl Acad Sci USA 1975;72:437-441. 36 Jauhonen P, Baraona E. Miyakawa H, Lieber CS. Mechanism for selective perivenular hepatotoxicity of ethanol. Alcoholism: Clin Exp Res 1982;6:350-357. 37. French SW. Alcoholic hepatotoxicity. J Hepatol 1989;9:134-135. 38. Popper H. Lieber CS. Histogenesis of alcoholic fibrosis and cirrhosis in the baboon. Am J Pathol 1980;98:695-716. :$9.Lieber CS. Casini A, DeCarli LM, Kim C. Lowe N, Sasalu R, Leo MA. S-adenosyl-1.-methionine attenuates alcohol-induced liver injury in the baboon. H~~,vrcii.o(:u 1990:11:165-173. 40. Mezey E, Potter JJ. French SW. Tamura T, Halsted CH. Effect of chronic ethanol feeding on hepatic collagen in the monkey. HEPATOI.O(;Y 1983;3:41-44. 4 1 Rogers AE, Fox JG, Gottlieb LS. Effects of ethanol and malnutrition on nonhuman primate liver. In: Berk PD, Chalmers TC. eds. Frontiers in liver disease. Stuttgart: Georg Thieme Verlag. 1981:167-175. 42. Wurtman R J . Uses of choline and lecithin in brain disorders. In: Barbeau A, Growdon J. Wurtman R, eds. Nutrition and the brain. Vol. V. New York: Raven Press, 1979:73. 4 3 Edreira JG, Hirsch RL, Kennedy JA. Production of fatty liver with dietary ethanol despite orotic acid supplementation. Q J Stud Alcohol 1974;35:20-25. 14. Chalvardjian A. Mode of action of choline. V. Sequential changes in hepatic and serum lipids of choline-deficient rats. Can J Biochem 1970;48:1234-1240. 45. Baraona E, Lieber CS. Effects of chronic ethanol feeding on serum lipoprotein metabolism in the rat. J Clin Invest 1970:49: 769-778. 46. Haines DSM. The effects of choline deficiency and choline refeeding upon t h e metabolism of plasma and liver lipids. Can J Biochem 1966;44:45-57. 47. Corredor C, Mansbach C, Bressler R. Choline control of tissue carnitine content [Abstract I. Fed Proc 1967;26:278. 48. Kondrup J. Grunnet N. The effect of acute and prolonged ethanol treatment on the contents of coenzyme A, carnitine and their derivatives in rat liver. Biochem J 1973;132:373-379. 49. Iseri OA. Lieber CS, Gottlieb LS. The ultrastructure of fatty liver induced by prolonged ethanol ingestion. Am J Pathol 1966:48: 535-555. 50. Nanji AA, Mendenhall CL. French SW. Beef fat prevents alcoholic liver disease in the rat. Alcoholism: Clin Exp Res 1989;13:15-19. 51. Nanji AA. French SW. Dietary linoleic acid is required for development of experimentally induced alcoholic liver injury. Life Sci 1989;44:223-227. 52. Fox JM. Polyene phospbatidylcholine: pharmacokinetics after oral administration-a review. In: Avogaro P. Mancini M, Ricci G, Paoletti R, eds. Phospholipids and atherosclerosis. New York: Raven Press, 1983. 5 3 Zierenberg 0, Grundy SM. Intestinal absorption of polyenephosphatidylcholine in man. J Lipid Res 1982;23:1136-1142. 54. Parthasarathy S, Subbaiah PV, Ganguly J. The mechanism of intestinal absorption of phosphatidylcholine in rats. Biochem J 1974:140~503-508. 5 5 . Rodgers JB. O’Brien RJ, Balint JA. The absorption and subsequent utilization of lecithin by the rat jejunum. Am J Dig Dis 1975;20:208-211. 56. Lekim D. Graf E. Tierexperimentelle Studien zur Pharmakokinetik der “essentiellen” Phospholipide (EPL).Drug Res 1976:26: 1772-1782. 55. Akesson B. Content of phospholipids in human diets studied by the duplicate portion technique. Br J Nutr 1982;47:223-229. 58. Nilsson AKE.Intestinal absorption of lecithin and lysolecithin by lymph fistula rats. Biochim Biophys Acta 1968;152:379-390.

1398

LIEBER ET AL.

59. Holzl J , Wagner H. Uber den Einbau von intraduodenal appliziertem ‘‘CP’P-Polyenphosphatidylcholin in die Leber von Ratten and seine Ausscheidung durch die Galle. Z Naturforsch 1971;26: 1151-1158. 60. Lekim D, Beting H, Stoffel W. Incorporation of complete phospholipid molecules in cellular membranes of rat liver after uptake from blood serum. Hoppe-Seyler’s Z Physiol Chem 1972;353S: 949-964. 61. Lekim D, Betzing H. The incorporation of essential phospholipids into the organs of intact and galactosamine intoxicated rats. Drug Res 1974;24:1217-1221. 62. Kuntz E. Pilotstudie mit Polyenylphosphatidylcholinbei schwerer Leberinsuffizienz. Med Welt 1989;40:1327-1329. 63. French SW, Ihrig TJ, Morin RJ. Lipid composition of RBC ghosts, liver mitochondria and microsomes of ethanol-fed rats. Q J Stud Alcohol 1970;31 Sol-809. 64. Thompson JA, Reitz RC. Effects of ethanol ingestion and dietary fat levels on mitochondrial lipids in male and female rats. Lipids 1978;13:540-550. 65. Arai M, Gordon ER, Lieber CS. Decreased cytochrome oxidase activity in hepatic mitochondria after chronic ethanol consumption and the possible role of decreased cytochrome aa:, content and changes in phospholipids. Biochim Biophys Acta 1984;797:320-327. 66. Arai M, Leo MA, Nakano M, Gordon ER, Lieber CS. Biochemical and morphological alterations of baboon hepatic mitochondria after chronic ethanol consumption. HEPATOLOGY 1984;4:165-174. 67. Anggard E. Ethanol, essential fatty acids and prostaglandins. Pharmacol Biochem Behav 1983;18:401-407. 68. Alling C, Aspenstrom G, Dencker S, Svennerholm L. Essential fatty acids in chronic alcoholism. Acta Med Scand Suppl 1979; 631:1-38. 69. Rouach H, Clement H, Orfanelli MT, Janvier B, Nordmann R. Fatty acid composition of rat liver mitochondrial phospholipids

HEPATOLDCY

during ethanol inhalation. Biochim Biophys Acta 1984;795:125129. 70. Horrobin DF. Essential fatty acids, prostaglandins, and alcoholism: an overview. Alcoholism: Clin Exp Res 1987;11:2-9. 71. Kalab M, Cervinka J. “Essential” phospholipids in the treatment of cirrhosis of the liver. Cas Lek Cesk 1983;122:266-269. 72. Petera V, Prokop V. Die kimpensierte Leberzirrhose: therapieerfahrungen mit Essential forte. Therapie Woche 1986;36:540-544. 73. SamochowiecVL,Wojcicki J , Dominiczak K. Effect of “essential” phospholipids on liver changes after chronic ethyl alcohol- and allylisosulfocyanate intoxication in albino rats. Arzneimittelforschung 1967;17:1374-1376. 74. Alliet J , Comlan G, Gourdier D. Steatosis of the human hepatocyte: an ultrastructure study-the reparative effects of polyenic fatty acid phosphatidylcholines. L’Ouest Med 1976;29:85-104. 75. Fassati P, Horejsi J , Fassati M, Spizek J , Jezkova Z. Essential choline phospholipids and theii effect on HBsAG and selected biochemical tests in cirrhosis of the liver. Cas Lek Cesk 1981;120: 56-60. 76. Jenkins PJ, Portmann BP, Edleston ALWF, Williams R. Use of polyunsaturated phosphatidyl choline in HBsAg negative chronic active hepatitis: results of a prospective double blind control trial. Liver 1982;2:77-81. 77. Maros T, Seres-Sturm L, Lakatos 0, Seres-Sturm MT, Blazsek V. Investigations regarding the possibility of resorption of the hepatic collagen in experimental toxic cirrhosis, under the effect of essential phospholipids. Arzneimittelforschung 1973;23:1538-1542. 78. Lieber CS, Leo MA. Interaction of alcohol and nutritional factors with hepatic fibrosis. In: Popper H, Schaffner F, eds. Progress in liver disease. Vol. 111. New York: Grune & Stratton, 1986:253272. 79. Mak KM, Lieber CS. Lipocytes and transitional cells in alcoholic 1988;8:1027liver disease: a morphometric study. HEPATOLOGY 1033.