Interaction of Ethanol with p-Carotene: Delayed Blood Clearance and Enhanced Hepatotoxicity h!fARIA ANNA LEO, CHO-IL KIM, NANCY LOWE AND CHARLES s. LIEBER Section of Liver Disease and Nutrition and Alcohol Research and Treatment Center, Bronx Veterans Affairs Medical Center and Mount Sinai School of Medicine, New York, New York 10468
Because we had found that ethanol interacts with retinol, we investigated whether it also affects its precursor, p-carotene. In 14 baboons fed ethanol (50% of total energy) for 2 to 5 yr with a standard amount of p-carotene (one 200-gm carrot/day), levels of @-carotenewere much higher than in controls fed isocaloric carbohydrate, both in plasma (122.5 30.9 nmol/dl vs. 6.3 f 1.4 nmol/dl; p < 0.005) and in liver (7.9 +- 1.1 nmollgm vs. 1.8 2 0.5 nmol/gm; p < 0,001). Even 20 days after withdrawal of the carrots, plasma p-carotene levels remained higher in alcohol-fed baboons than in controls (10.1 2 3.8 nmolldl vs. < 0.1 nmol/dl). Next, the diet was supplemented with p-carotene beadlets: in four pairs of baboons given a low dose of &carotene (3mg/1,000 kcal), plasma levels were significantlyhigher in alcohol-fedanimals than in controls, even when expressed per cholesterol (although the latter increased with alcohol intake). Seven pairs of animals were given a higher dose (30mg/1,000 kcal) of p-carotene for 1mo, followed, in four pairs, by 45 mg for another month. On cessation of p-carotene treatment, plasma levels decreased more slowly in the alcohol-fed baboons than in the controls. Percutaneous liver biopsy specimens revealed that liver concentrations of p-carotenecorrelated with plasma levels but were higher in the alcohol-fedbaboons than in the control baboons, whereas the p-carotene-induced increase in liver retinoids was lower (p < 0.02). Furthermore, the ethanol-induced liver depletion of total retinoids (432 103 nmoUgm vs. 1,711 .t 103 in controls; p < 0.001) was not corrected (637 f 149 vs. 2,404 f 74; p < 0.001), despite the massive supplementation with p-carotene. Moreover, in the animals fed alcohol with p-carotene, multiple ultrastructural lesions appeared, with autophagic vacuoles, abundant myelin figures, degenerated mitochondria and increased blood levels of the mitochondrial enzyme glutamic dehydrogenase. The histological changes were either absent or much less prominent in the
*
*
Received September 5, 1991; accepted December 12, 1991. This work was presented in part at the Annual Meeting of the American Association for the Study of Liver Diseases, Chicago, IL, November 6 , 1990. This work was supported by Granb DK32810 and AA03508 and by the Department of Veterans Affairs. Address reprint requests to: Charles S. Lieber, M.D.,(151/G), Alcohol Research and Treatment Center, VA Medical Center, 130 West Kingsbridge Road, Bronx, NY 10468. 3U1135611
baboons given p-carotene with the control diet or in animals fed the ethanol or control diets without p-carotene. Thus the combination of an increase in plasma and liver p-carotene after ethanol and a relative lack of a corresponding rise in retinol suggests interference with the conversion of p-carotene to vitamin A. Because of an associated exacerbation of the liver toxicity, we conclude that Pcarotene must be administered cautiously in the presence of heavy alcohol consumption because the optimal human therapeutic dose remains to be defined. (HEPATOLOGY 1992;15:883-89 1.)
p-Carotene is a precursor of vitamin A and one of its major sources in human nutrition (1).In addition to its provitamin A activity, a selective anticancer effect has been attributed to p-carotene because its consumption was found to be associated with a lower risk of several epithelial neoplasms (2). Conversely, poor dietary intake or low blood levels have been associated with increased risk of malignancies (3, 4); carotene itself, rather than other components of vegetables and fruits, has been incriminated (5). Low circulating levels of total plasma carotenoids have been reported in alcoholics (6-9) and attributed to a poor intake. This investigation addressed the question whether, in addition to dietary factors, a more direct interaction occurs between p-carotene and ethanol. Furthermore, our previous studies had revealed that ethanol potentiates the hepatotoxicity of retinol (10, 11); whether this pertains t o its precursor, p-carotene, is unknown and was also studied here. MATERIALS AND METHODS Eleven baboons (Papio hamadryas) were given our regular, nutritionally adequate control liquid diet (12); 11 other animals were matched individually with these baboons according to body weight (10 to 20 kg) and approximate age (4 to 6 yr) and pair-fed with the same diet except for isocaloric replacement of carbohydrates (50% of total energy) with ethanol. Three additional baboons were also fed the same alcohol-containing liquid diet. All animals received humane care in compliance with the institution’s guidelines. The feeding techniques and the diet have been reported in detail previously (12, 13). This diet provides 1 kcal/ml and a daily supply of energy, proteins, vitamins and minerals well within the recommended daily allowances for the baboon (14). In addition, the animals received a daily carrot (approximately
883
884
LEO ET AL.
=
HEPATOLOGY
0 CONTROL BABOONS ETHANOL-FED
LIVER
PLASMA
-
500.
nil4
B
CONTROL BABOONS ETtWOL-FED
-200-
z k9 c
100.
50-
w W
ch
lo-
5-
a
I -
I---
< 0.005
+
0.2 J
I
0
20
40
60
80
100
120
TIME (DAYS)
FIG. 1. Effect of chronic ethanol consumption on p-carotene levels. A survey of our baboon colony fed liquid diets (with or without ethanol) and a daily carrot (200 gm corresponding to 30 mg p-carotene) revealed that the liver and plasma p-carotene levels, measured by HPLC, were significantly higher in ethanol-fed animals than in pair-fed controls.
FIG.3. Effect of ethanol consumption on the clearance of p-carotene from plasma (experiment 2). After daily administration of 30 mg p-caroteneb for 33 days, the clearance of plasma p-carotene was much slower in ethanol-fed baboons than in their controls.
reading, prepared and examined as described before (16), except that fixation was in 2% osmium tetroxide. A Plasma AST and glutamate dehydrogenase (GDH) activities were measured according to the methods of Karmen (17) and Ellis and Goldberg (181, respectively. Total plasma cholesterol was determined enzymatically with a modification of the -0) method of Allain et al. (19) using a Sigma Diagnostics kit 0 (Sigma Chemical Co., St. Louis, MO). Blood ethanol levels were I 20 measured according to the method of Korsten et al. (20); they U W fluctuated between 10 and 30 mmol/L. p-Carotene in plasma 1 and liver was measured by HPLC according to the method of Shapiro et al. (21). Vitamin A was determined in the same samples by simultaneous detection at 325 nm. All HPLC analyses were carried out with a HP-1090 liquid chromatograph equipped with a photo diode-array spectrophotometric detector and HPLC Chemstation (Hewlett Packard, Palo Alto, " CA). Results of plasma Q-carotene testing were expressed as 300 600 nmol/dl; they can be converted into Systeme International PLASMA nmolidl units with a multiplication factor of 10. Statistics. Data were analyzed by two-way ANOVA. FIG. 2. Correlation between liver and plasma p-carotene levels in alcohol-fed and control baboons. In both alcohol-fed and control One-way ANOVA and the Newman-Keuls' test were peranimals, the correlation was significant. formed to determine the significance of differences among individual groups. Linear regression analyses were performed to relate changes in liver and plasma p-carotene (22). All 200 gm) that provided 5,600 retinol equivalents, or about statistical analyses were carried out with a Stat-Pro statistical 30 mg p-carotene (15). After our initid survey of p-carotene package (Wadsworth Professional Software, Boston, MA) for levels, the carrots were withdrawn; 1 mo later, p-carotene the IBM PC computer (IBM Corp., Armonk, NY).To analyze (Solatene; Hoffman La Roche, 30-mg capsules) was adminis- p-carotene clearance, the regression lines were calculated tered as part of the liquid diet. In four pairs of baboons, a low according to a two-phase linear regression for a biexponential dose of p-carotene (3 mg/L) was given for 2 mo. Blood levels of model devised by Lee, Poon and Kingdon (23). p-carotene and retinol were determined at 31 and 61 days RESULTS (experiment 1). In addition, seven pairs of animals were given a higher dose (30to 45 mgL; experiments 2 and 3). The Effect of Chronic Ethanol Consumption on intake was comparable in ethanol-fed and control baboons. PCarotene Levels. A survey of our baboon colony fed The dose was selected to mimic commonly used therapeutic liquid diets (with or without ethanol) and a daily carrot amounts. Venous blood was sampled from the extremities, and revealed that the liver and plasma p-carotene levels were percutaneous liver biopsy specimens were obtained while significantly higher in ethanol-fed animals than in baboons were under ketamine anesthesia after a 6-hr food pair-fed controls (Fig. 1). A significant correlation was withdrawal. Similar samples were taken from each group. For found between liver and plasma levels (r = 0.662; light and electron microscopy, specimens were coded for blind p = 0.05; n = 9). When these results were combined fed
.
Baboons A Pre-load 0 Post-load (3Omg/L) 0 Post-load (45mgiL)
Vol. 15, No. 5, 1992
6oo
885
INTERACTION OF ETHANOL WITH @-CAROTENE 600 -
1
a
5 5a
Q 1-
05-
.
0.5 0.2J
I
0
Because we had found that ethanol interacts with retinol, we investigated whether it also affects its precursor, p-carotene. In 14 baboons fed ethanol (50% of total energy) for 2 to 5 yr with a standard amount of p-carotene (one 200-gm carrot/day), levels of @-carotenewere much higher than in controls fed isocaloric carbohydrate, both in plasma (122.5 30.9 nmol/dl vs. 6.3 f 1.4 nmol/dl; p < 0.005) and in liver (7.9 +- 1.1 nmollgm vs. 1.8 2 0.5 nmol/gm; p < 0,001). Even 20 days after withdrawal of the carrots, plasma p-carotene levels remained higher in alcohol-fed baboons than in controls (10.1 2 3.8 nmolldl vs. < 0.1 nmol/dl). Next, the diet was supplemented with p-carotene beadlets: in four pairs of baboons given a low dose of &carotene (3mg/1,000 kcal), plasma levels were significantlyhigher in alcohol-fedanimals than in controls, even when expressed per cholesterol (although the latter increased with alcohol intake). Seven pairs of animals were given a higher dose (30mg/1,000 kcal) of p-carotene for 1mo, followed, in four pairs, by 45 mg for another month. On cessation of p-carotene treatment, plasma levels decreased more slowly in the alcohol-fed baboons than in the controls. Percutaneous liver biopsy specimens revealed that liver concentrations of p-carotenecorrelated with plasma levels but were higher in the alcohol-fedbaboons than in the control baboons, whereas the p-carotene-induced increase in liver retinoids was lower (p < 0.02). Furthermore, the ethanol-induced liver depletion of total retinoids (432 103 nmoUgm vs. 1,711 .t 103 in controls; p < 0.001) was not corrected (637 f 149 vs. 2,404 f 74; p < 0.001), despite the massive supplementation with p-carotene. Moreover, in the animals fed alcohol with p-carotene, multiple ultrastructural lesions appeared, with autophagic vacuoles, abundant myelin figures, degenerated mitochondria and increased blood levels of the mitochondrial enzyme glutamic dehydrogenase. The histological changes were either absent or much less prominent in the
*
*
Received September 5, 1991; accepted December 12, 1991. This work was presented in part at the Annual Meeting of the American Association for the Study of Liver Diseases, Chicago, IL, November 6 , 1990. This work was supported by Granb DK32810 and AA03508 and by the Department of Veterans Affairs. Address reprint requests to: Charles S. Lieber, M.D.,(151/G), Alcohol Research and Treatment Center, VA Medical Center, 130 West Kingsbridge Road, Bronx, NY 10468. 3U1135611
baboons given p-carotene with the control diet or in animals fed the ethanol or control diets without p-carotene. Thus the combination of an increase in plasma and liver p-carotene after ethanol and a relative lack of a corresponding rise in retinol suggests interference with the conversion of p-carotene to vitamin A. Because of an associated exacerbation of the liver toxicity, we conclude that Pcarotene must be administered cautiously in the presence of heavy alcohol consumption because the optimal human therapeutic dose remains to be defined. (HEPATOLOGY 1992;15:883-89 1.)
p-Carotene is a precursor of vitamin A and one of its major sources in human nutrition (1).In addition to its provitamin A activity, a selective anticancer effect has been attributed to p-carotene because its consumption was found to be associated with a lower risk of several epithelial neoplasms (2). Conversely, poor dietary intake or low blood levels have been associated with increased risk of malignancies (3, 4); carotene itself, rather than other components of vegetables and fruits, has been incriminated (5). Low circulating levels of total plasma carotenoids have been reported in alcoholics (6-9) and attributed to a poor intake. This investigation addressed the question whether, in addition to dietary factors, a more direct interaction occurs between p-carotene and ethanol. Furthermore, our previous studies had revealed that ethanol potentiates the hepatotoxicity of retinol (10, 11); whether this pertains t o its precursor, p-carotene, is unknown and was also studied here. MATERIALS AND METHODS Eleven baboons (Papio hamadryas) were given our regular, nutritionally adequate control liquid diet (12); 11 other animals were matched individually with these baboons according to body weight (10 to 20 kg) and approximate age (4 to 6 yr) and pair-fed with the same diet except for isocaloric replacement of carbohydrates (50% of total energy) with ethanol. Three additional baboons were also fed the same alcohol-containing liquid diet. All animals received humane care in compliance with the institution’s guidelines. The feeding techniques and the diet have been reported in detail previously (12, 13). This diet provides 1 kcal/ml and a daily supply of energy, proteins, vitamins and minerals well within the recommended daily allowances for the baboon (14). In addition, the animals received a daily carrot (approximately
883
884
LEO ET AL.
=
HEPATOLOGY
0 CONTROL BABOONS ETHANOL-FED
LIVER
PLASMA
-
500.
nil4
B
CONTROL BABOONS ETtWOL-FED
-200-
z k9 c
100.
50-
w W
ch
lo-
5-
a
I -
I---
< 0.005
+
0.2 J
I
0
20
40
60
80
100
120
TIME (DAYS)
FIG. 1. Effect of chronic ethanol consumption on p-carotene levels. A survey of our baboon colony fed liquid diets (with or without ethanol) and a daily carrot (200 gm corresponding to 30 mg p-carotene) revealed that the liver and plasma p-carotene levels, measured by HPLC, were significantly higher in ethanol-fed animals than in pair-fed controls.
FIG.3. Effect of ethanol consumption on the clearance of p-carotene from plasma (experiment 2). After daily administration of 30 mg p-caroteneb for 33 days, the clearance of plasma p-carotene was much slower in ethanol-fed baboons than in their controls.
reading, prepared and examined as described before (16), except that fixation was in 2% osmium tetroxide. A Plasma AST and glutamate dehydrogenase (GDH) activities were measured according to the methods of Karmen (17) and Ellis and Goldberg (181, respectively. Total plasma cholesterol was determined enzymatically with a modification of the -0) method of Allain et al. (19) using a Sigma Diagnostics kit 0 (Sigma Chemical Co., St. Louis, MO). Blood ethanol levels were I 20 measured according to the method of Korsten et al. (20); they U W fluctuated between 10 and 30 mmol/L. p-Carotene in plasma 1 and liver was measured by HPLC according to the method of Shapiro et al. (21). Vitamin A was determined in the same samples by simultaneous detection at 325 nm. All HPLC analyses were carried out with a HP-1090 liquid chromatograph equipped with a photo diode-array spectrophotometric detector and HPLC Chemstation (Hewlett Packard, Palo Alto, " CA). Results of plasma Q-carotene testing were expressed as 300 600 nmol/dl; they can be converted into Systeme International PLASMA nmolidl units with a multiplication factor of 10. Statistics. Data were analyzed by two-way ANOVA. FIG. 2. Correlation between liver and plasma p-carotene levels in alcohol-fed and control baboons. In both alcohol-fed and control One-way ANOVA and the Newman-Keuls' test were peranimals, the correlation was significant. formed to determine the significance of differences among individual groups. Linear regression analyses were performed to relate changes in liver and plasma p-carotene (22). All 200 gm) that provided 5,600 retinol equivalents, or about statistical analyses were carried out with a Stat-Pro statistical 30 mg p-carotene (15). After our initid survey of p-carotene package (Wadsworth Professional Software, Boston, MA) for levels, the carrots were withdrawn; 1 mo later, p-carotene the IBM PC computer (IBM Corp., Armonk, NY).To analyze (Solatene; Hoffman La Roche, 30-mg capsules) was adminis- p-carotene clearance, the regression lines were calculated tered as part of the liquid diet. In four pairs of baboons, a low according to a two-phase linear regression for a biexponential dose of p-carotene (3 mg/L) was given for 2 mo. Blood levels of model devised by Lee, Poon and Kingdon (23). p-carotene and retinol were determined at 31 and 61 days RESULTS (experiment 1). In addition, seven pairs of animals were given a higher dose (30to 45 mgL; experiments 2 and 3). The Effect of Chronic Ethanol Consumption on intake was comparable in ethanol-fed and control baboons. PCarotene Levels. A survey of our baboon colony fed The dose was selected to mimic commonly used therapeutic liquid diets (with or without ethanol) and a daily carrot amounts. Venous blood was sampled from the extremities, and revealed that the liver and plasma p-carotene levels were percutaneous liver biopsy specimens were obtained while significantly higher in ethanol-fed animals than in baboons were under ketamine anesthesia after a 6-hr food pair-fed controls (Fig. 1). A significant correlation was withdrawal. Similar samples were taken from each group. For found between liver and plasma levels (r = 0.662; light and electron microscopy, specimens were coded for blind p = 0.05; n = 9). When these results were combined fed
.
Baboons A Pre-load 0 Post-load (3Omg/L) 0 Post-load (45mgiL)
Vol. 15, No. 5, 1992
6oo
885
INTERACTION OF ETHANOL WITH @-CAROTENE 600 -
1
a
5 5a
Q 1-
05-
.
0.5 0.2J
I
0
