- Email: [email protected]
Abstinence-induced changes in plasma apolipoprotein levels of alcoholics
331
Drug and Alcohol Dependence, 8 (1981) 331 - 343 @ Elsevier Sequoia S.A., Lausanne - Printed in The Netherlands
ABSTINENCE-INDUCED CHANGES LEVELS OF ALCOHOLICS
JOSEPH
Service, Veterans of Pharmacology,
Administration The Medical
Medical Center, Wood, College of Wisconsin,
ALAUPOVIC
Laboratory Oklahoma PAUL
APOLIPOPROTEIN
J. BARBORIAK
Biochemistry Section, Research WI 53193, and the Department Milwaukee, WI 53226 (U.S.A.)
PETAR
IN PLASMA
of Lipid and Lipoprotein City, OK (U.S.A.)
Studies,
Oklahoma
Medical
Research
Foundation,
CUSHMAN
Departments Milwaukee, (Received
of Pharmacology, Medicine, and Psychiatry, Medical College of Wisconsin WI 53226 and DePaul Rehabilitation Hospital, WI 53221 (U.S.A.) October
19, 1981)
Summary Plasma lipids, including total cholesterol, plasma triglycerides, high density lipoprotein cholesterol and several apolipoproteins were measured in eight otherwise healthy male alcoholics at the beginning and during a 4-week inpatient abstinence treatment program. The initial triglyceride and apolipoprotein A-II levels were higher and apolipoprotein B was lower in alcoholics than in 12 nondrinking control subjects. During the alcohol abstinence there was a parallel fall in the levels of high density lipoprotein cholesterol, and apolipoproteins A-I and A-II and a rise in apolipoprotein B. The changes in apolipoproteins C-III and E were less consistent. These findings suggest that, even in otherwise healthy alcoholics, elevated levels of apolipoproteins A-I and A-II on admission are ameliorated during abstinence of several weeks. ____
Several investigators have reported that alcohol consumption, especially if excessive, is associated with increased levels of high density lipoproteins (HDL) [ 1 - 31 or HDL cholesterol [ 4 - 61. Little information is available on
Reprint requests should be addressed vice/151, Veterans Administration Medical
to Joseph J. Barboriak, ScD, Research Center, Wood, WI 53193, U.S.A.
Ser-
338
the corresponding levels of the individual apolipoproteins or on their abstinence-induced changes. To study the effect of abstinence on plasma apolipoproteins, we have measured the concentrations of apolipoproteins A-I, A-II, B, C-III and E in chronic alcoholics at the beginning, during and after a 4-week abstinence program. Results of this study show that abstinence had a marked effect on the levels of these apolipoproteins. Subjects
and methods
Eight male middle-class otherwise healthy alcoholics entering an inpatient abstinence-oriented alcoholism treatment center were studied. The average age of the patients was 28 f 7 years (mean + SD). Their preadmission alcohol intake over a 30-day period (expressed in absolute alcohol) was estimated as 250 f 100 ml/day. The average duration of excessive alcohol intake was 2 - 8 years. Seven of the eight patients were smokers, reportedly smoking 20 - 40 cigarettes daily, who continued to smoke during their hospital stay. Physical examination failed to demonstrate any overt sign of mental, nutritional and/or hepatic disease. Their weight on admission was 71.3 * 1.8 kg, with none reporting a loss of more than 2 kg in the month prior to admission. Laboratory tests indicated increased levels of gamma-glutamyltransferase and serum glutamic oxaloacetic transaminase in four patients. None of the patients admitted using other drugs, but benzodiazepines were detected in urine samples of two patients. Fasting (12 hours) blood samples were obtained on admission and again after 2 - 4 weeks of alcohol abstinence. All of the patients responded well to abstinence treatment; their mean weight gain was 0.5 kg during each 2-week interval. Plasma levels of total cholesterol and triglycerides were measured by automated procedures [ 7, 81 and the HDL cholesterol was determined by the heparin manganese precipitation method [ 91. Quality control of the cholesterol and triglyceride determinations was maintained by the World Health Organization Reference Center for Lipid Determination located at the Center for Disease Control, Atlanta, GA. The quantitative determination of plasma apolipoproteins A-I, A-II, B, C-III and E was carried out by electroimmunoassay in the Laboratory of Lipid and Lipoprotein Studies, Oklahoma Medical Research Foundation, according to already described procedures [ 10 - 131. Plasma samples were shipped on ice by airfreight from Milwaukee to Oklahoma City. All apolipoprotein assays were carried out in triplicate. Control values for HDL cholesterol and apolipoproteins were obtained from 12 age-matched nondrinking normolipidemic male subjects. Results Plasma HDL cholesterol levels were within normal limits on admission, but during abstinence treatment they fell from 54 f 3 to 41 f 4 mg/dl (p <
339
TG
TG
0 2 WEEKS
0
4
Fig. 1. Changes in plasma levels of total cholesterol, triglycerides, and high-density lipoprotein cholesterol during a 4-week alcohol abstinence. N denotes the values for 12 age-matched control subjects.
0.02) (Fig. 1). The average initial total plasma cholesterol level of eight patients was 171 + 13 mg/dl (mean + SD), changing very little during the 4week period of alcohol abstinence. Although the average plasma triglyceride level decreased during the treatment from the initial value of 133 + 21 to 107 t 17 mg/dl, the difference was not statistically significant. Nevertheless, the initial triglyceride level of chronic alcoholics (133 + 21 mg/dl) was significantly higher @ < 0.05) than the average plasma triglyceride level of 85 * 28 mg/dl measured in the control group. The average initial values of both apo A-I and apo A-II from patients were higher than control subjects (Table 1); only the APO-II levels were significantly different 0, < 0.01). The
TABLE
1
Apolipoproteins
Admision 2 weeks 4 weeks Controlst
in eight male alcoholics
before
and in abstinence
(mg/dl)
HDLC
Apo A-I
A-I/ HDLC
Apo A-II
A-II/ HDLC
Apo B
54 * 8 47 * 8 41 + 8* 46i 8
123 f 7 104 t 3* 94 + 6* 115~ lO*
2.27 2.21 2.22 2.50
64? 3 48 i 4* 48 + 4* 52 i 8*
1.19 1.02 1.17 1.13
80? 83 t go+ 112 +
*Significantly different from admission tTwelve controls were used.
levels @ < 0.05).
APO A-I/B 5 6 7* 8*
1.56 1.28 1.08* 1.02*
340
ratios of Apo A-I/HDLC and Apo A-II/HDLC were unaffected either by alcohol or by abstinence, thus Apo A-I, Apo A-II and HDL cholesterol decreased in parallel (Table 1). There was a greater proportionate increase in Apo A-II than Apo A-I amongst the alcoholics. The initial average concentration of plasma Apo B was lower in the eight patients than in the control subjects (80 f 6 vs 112 + 8 mg/dl, p < 0.01). The Apo B level increased in five of eight patients during the abstinence, especially in the first 2-week period; the overall increase was not statistically significant. These changes in concentrations of major apolipoproteins were maximally reflected in the ratios of Apo A-I/APO B. A decrease from the initial value of 1.56 to a value of 1.08 at the end of the abstinence period represented a 31% change (p < 0.005). The latter value is very similar to the Apo A-I/APO B ratio in control subjects (1.02). The Apo C-III and Apo E concentrations were measured in six and four patients, respectively, and their average initial levels (9.7 t 2.7 and 8.1 + 3.7 mg/dl, respectively) did not differ markedly from the values for the control subjects (9.4 + 2.3 and 11.5 + 2.7 mg/dl, respectively). The abstinence-induced change in these apolipoproteins was less consistent, with Apo C-III showing a reduction in four of six patients and Apo E registering an increase in three of the four patients at the end of the 4-week abstinence period.
Discussion The present study confirms the observations of the Scandinavian investigators [3, 41 that the high HDL cholesterol levels seen in alcohol abuse decline during a period of abstinence. Moreover, our data also show substantial reductions in HDL cholesterol levels when the initial values were only slightly elevated or within normal limits. In several of our patients, the HDL cholesterol levels (after a 4-week abstinence) approached the lower limit of normal range [ 141. This study also demonstrated that alcohol abstinence affects the levels of various apolipoproteins. The most novel findings were the reduction in levels of apolipoproteins A-I and A-II which paralleled the change in HDL cholesterol. The abstinence-induced changes in Apo A-I/APO B ratio, which decreased significantly in all eight patients, were striking. These data are consistent with other reports. Johansson and Laurel1 [l] reported that some of their alcoholic patients had reduced levels of electrophoretically separated /3-lipoproteins. Similarly, Castelli et al. [ 151 demonstrated that patients consuming larger amounts of alcohol had decreased levels of LDL cholesterol. Little definitive information is available on the possible mechanism(s) involved in the alcohol-induced increase of HDL cholesterol or apolipoproteins or in their changes following abstinence. Some investigators have reported that drugs, which either stimulate [16, 171 or reduce [18] the drug-metabolizing enzyme activity of the liver, also affect plasma HDL
341
cholesterol levels. An alcohol-induced increase in the activities of these enzymes has been previously reported [19, 201 and may be one of the contributing factors. The plasma or tissue lipoprotein lipase [ 21, 221 may be another factor regulating the HDL cholesterol levels. There is a strong positive correlation between the activity of adipose tissue lipoprotein lipase and HDL cholesterol levels [ 231. Several investigators have reported an increase in the activity of adipose tissue lipoprotein lipase after a prolonged alcohol intake [ 5, 241. Other possibilities which need careful study include an alcohol action on lecithin-cholesterolacyltransferase (LCAT), or rate of cholesterol removal from peripheral tissue, cholesterol excretion or synthesis. Studies from this and other laboratories [25] have indicated that lipoproteins represent a mixture of several discrete lipoprotein families, including lipoproteins A, B, C, D, E and F, which are secreted from the liver and/or intestine as the so-called nascent lipoproteins or are generated during the lipolysis of triglyceride-rich lipoproteins [ 261. Approximately 50% of HDL is generally thought to be protein, with phospholipid accounting for 25%, cholesterol 20%, and triglyceride 5% of this lipid moiety. Apolipoproteins in HDL are 90% APO-A, with A-I about 60% and A-II approximately 30%. Results of the present study indicate that the alcohol-induced increase in HDL cholesterol or its reduction during abstinence is accompanied by corresponding changes in the concentration of lipoproteins A and, to some extent, lipoproteins C. From the results of this study, it is not possible to predict whether these lipoprotein families were generated as nascent lipoproteins and/or as products of lipolysis. However, if high activity of lipoprotein lipase results in increased levels of HDL2 particles [22, 271, lipoproteins A and C may have been generated by both pathways. Danielsson et al. [ 31 reported that the chronic alcoholics exhibited either a moderate increase of HDLs or a four- to fivefold increase of HDL2. This may indicate that the nascent lipoproteins were the primary contributors to increased HDL levels in the former group of patients, while the lipoproteins generated during lipolysis represented the primary contributors in the latter group of chronic alcoholics. To further substantiate the role of lipolysis in chronic alcoholics, we have determined the HDL/VLDL ratio of Apo C-III. We have previously demonstrated that the high ratios (>1.5) of Apo C-III reflect an efficient and the low ratios an impaired catabolism of triglyceride-rich lipoproteins [ 281. Results of this analysis demonstrated that the Apo C-III ratio decreased significantly during the 4-week abstinence in two out of three patients. This finding suggests that the chronic alcohol intake may have a different effect on the qualitative and quantitative composition of high density lipoprotein families in individual patients. In some, the increased levels of HDL may be caused by increased secretion of nascent lipoprotein particles, and in others by increased lipolysis. The relevance of the above findings in chronic alcoholics to the general population is not clear. However, it has been reported by several investigators that even in the nonalcoholic subjects alcohol consumption is associated with an increase in plasma HDL cholesterol levels [6, 151. High density lipo-
342
protein is frequently viewed as a lipoprotein which is involved in the removal of free cholesterol from arterial sites. The LCAT enzyme may be considered a key step in this process. APO-I is generally associated with LCAT activity. The function of Apo A-II is less clear; however, it may stabilize the esterified cholesterol. While the enhancing effect of alcohol consumption on the HDL cholesterol levels and their reduction during alcohol abstinence have been reported by numerous investigators [ 1 - 61, the possible confounding effect of dietary changes, especially during abstinence studies, should also be considered. Although dietary intake in our patients was not measured, a general observation by the allied health personnel failed to observe any abnormal dietary pattern. Furthermore, their weight gain during the 4-week observation period (approximately 1 kg) was not sufficient to explain the observed marked reduction in plasma HDL cholesterol levels. In agreement with other findings [ 51 our data suggest that the observed lipoprotein and apolipoprotein changes were primarily due to alcohol abstinence. Acknowledgements The authors wish to thank Dr. John H. Kalbfleisch for his statistical advice and the assistance of J. Fesmire, R. Burns, E. Gehrke, R. Ballou, and J. Owenby for their valuable technical assistance. This work was supported in part by research grants HL 14378 and HI 23181 from the National Institute of Health, by NIDA grant DA 07181, by the Oklahoma Medical Research Foundation, and by the U.S. Veterans Administration. References B. G. Johansson and C. B. Lowell, Disorders of serum or-lipoproteins after alcohol intoxication. Stand. J. Clin. Lab. Inuest., 23 (1969) 231. B. G. Johansson and A. Medhus, Increase in plasma e-lipoproteins in chronic alcoholics after acute abuse. Acta Med. &and., 195 (1974) 273. B. Danielsson, R. Ekman and G. Fex, Changes in plasma high density lipoproteins in chronic male alcoholics during and after abuse. Stand. J. Clin. Lab. Invest., 38 (1978)
113. S. Wallerstedt,
A. Gustafson and R. Olsson, Serum lipids and lipoproteins during after heavy alcohol consumption in chronic alcoholics. Stand. J. Clin. Lab. Invest., 37 (1977) 599. B. B. Blefrange, I. Hagerstrand and N. Nilsson-Ehle, Alterations of lipid metabolism in healthy volunteers during long-term ethanol intake. Eur. J. Clin. Inuest., 7 (1977) abstinence
127. J. J. Barboriak, A. F. Anderson and R. G. Hoffmann, Interrelationship between coronary artery occlusion, high-density lipoprotein cholesterol, and alcohol intake. J. Lab. Clin. Med., 94 (1979) 348. W. D. Block, J. K. Jarrett and J. B. Levin, Use of a single color reagent to improve the automated determination of serum total cholesterol. In L. T. Skeggs Jr. (ed), Automation in Analytical Chemistry, Mediad, New York, 1965, pp. 345 - 347.
343
Drug and Alcohol Dependence, 8 (1981) 331 - 343 @ Elsevier Sequoia S.A., Lausanne - Printed in The Netherlands
ABSTINENCE-INDUCED CHANGES LEVELS OF ALCOHOLICS
JOSEPH
Service, Veterans of Pharmacology,
Administration The Medical
Medical Center, Wood, College of Wisconsin,
ALAUPOVIC
Laboratory Oklahoma PAUL
APOLIPOPROTEIN
J. BARBORIAK
Biochemistry Section, Research WI 53193, and the Department Milwaukee, WI 53226 (U.S.A.)
PETAR
IN PLASMA
of Lipid and Lipoprotein City, OK (U.S.A.)
Studies,
Oklahoma
Medical
Research
Foundation,
CUSHMAN
Departments Milwaukee, (Received
of Pharmacology, Medicine, and Psychiatry, Medical College of Wisconsin WI 53226 and DePaul Rehabilitation Hospital, WI 53221 (U.S.A.) October
19, 1981)
Summary Plasma lipids, including total cholesterol, plasma triglycerides, high density lipoprotein cholesterol and several apolipoproteins were measured in eight otherwise healthy male alcoholics at the beginning and during a 4-week inpatient abstinence treatment program. The initial triglyceride and apolipoprotein A-II levels were higher and apolipoprotein B was lower in alcoholics than in 12 nondrinking control subjects. During the alcohol abstinence there was a parallel fall in the levels of high density lipoprotein cholesterol, and apolipoproteins A-I and A-II and a rise in apolipoprotein B. The changes in apolipoproteins C-III and E were less consistent. These findings suggest that, even in otherwise healthy alcoholics, elevated levels of apolipoproteins A-I and A-II on admission are ameliorated during abstinence of several weeks. ____
Several investigators have reported that alcohol consumption, especially if excessive, is associated with increased levels of high density lipoproteins (HDL) [ 1 - 31 or HDL cholesterol [ 4 - 61. Little information is available on
Reprint requests should be addressed vice/151, Veterans Administration Medical
to Joseph J. Barboriak, ScD, Research Center, Wood, WI 53193, U.S.A.
Ser-
338
the corresponding levels of the individual apolipoproteins or on their abstinence-induced changes. To study the effect of abstinence on plasma apolipoproteins, we have measured the concentrations of apolipoproteins A-I, A-II, B, C-III and E in chronic alcoholics at the beginning, during and after a 4-week abstinence program. Results of this study show that abstinence had a marked effect on the levels of these apolipoproteins. Subjects
and methods
Eight male middle-class otherwise healthy alcoholics entering an inpatient abstinence-oriented alcoholism treatment center were studied. The average age of the patients was 28 f 7 years (mean + SD). Their preadmission alcohol intake over a 30-day period (expressed in absolute alcohol) was estimated as 250 f 100 ml/day. The average duration of excessive alcohol intake was 2 - 8 years. Seven of the eight patients were smokers, reportedly smoking 20 - 40 cigarettes daily, who continued to smoke during their hospital stay. Physical examination failed to demonstrate any overt sign of mental, nutritional and/or hepatic disease. Their weight on admission was 71.3 * 1.8 kg, with none reporting a loss of more than 2 kg in the month prior to admission. Laboratory tests indicated increased levels of gamma-glutamyltransferase and serum glutamic oxaloacetic transaminase in four patients. None of the patients admitted using other drugs, but benzodiazepines were detected in urine samples of two patients. Fasting (12 hours) blood samples were obtained on admission and again after 2 - 4 weeks of alcohol abstinence. All of the patients responded well to abstinence treatment; their mean weight gain was 0.5 kg during each 2-week interval. Plasma levels of total cholesterol and triglycerides were measured by automated procedures [ 7, 81 and the HDL cholesterol was determined by the heparin manganese precipitation method [ 91. Quality control of the cholesterol and triglyceride determinations was maintained by the World Health Organization Reference Center for Lipid Determination located at the Center for Disease Control, Atlanta, GA. The quantitative determination of plasma apolipoproteins A-I, A-II, B, C-III and E was carried out by electroimmunoassay in the Laboratory of Lipid and Lipoprotein Studies, Oklahoma Medical Research Foundation, according to already described procedures [ 10 - 131. Plasma samples were shipped on ice by airfreight from Milwaukee to Oklahoma City. All apolipoprotein assays were carried out in triplicate. Control values for HDL cholesterol and apolipoproteins were obtained from 12 age-matched nondrinking normolipidemic male subjects. Results Plasma HDL cholesterol levels were within normal limits on admission, but during abstinence treatment they fell from 54 f 3 to 41 f 4 mg/dl (p <
339
TG
TG
0 2 WEEKS
0
4
Fig. 1. Changes in plasma levels of total cholesterol, triglycerides, and high-density lipoprotein cholesterol during a 4-week alcohol abstinence. N denotes the values for 12 age-matched control subjects.
0.02) (Fig. 1). The average initial total plasma cholesterol level of eight patients was 171 + 13 mg/dl (mean + SD), changing very little during the 4week period of alcohol abstinence. Although the average plasma triglyceride level decreased during the treatment from the initial value of 133 + 21 to 107 t 17 mg/dl, the difference was not statistically significant. Nevertheless, the initial triglyceride level of chronic alcoholics (133 + 21 mg/dl) was significantly higher @ < 0.05) than the average plasma triglyceride level of 85 * 28 mg/dl measured in the control group. The average initial values of both apo A-I and apo A-II from patients were higher than control subjects (Table 1); only the APO-II levels were significantly different 0, < 0.01). The
TABLE
1
Apolipoproteins
Admision 2 weeks 4 weeks Controlst
in eight male alcoholics
before
and in abstinence
(mg/dl)
HDLC
Apo A-I
A-I/ HDLC
Apo A-II
A-II/ HDLC
Apo B
54 * 8 47 * 8 41 + 8* 46i 8
123 f 7 104 t 3* 94 + 6* 115~ lO*
2.27 2.21 2.22 2.50
64? 3 48 i 4* 48 + 4* 52 i 8*
1.19 1.02 1.17 1.13
80? 83 t go+ 112 +
*Significantly different from admission tTwelve controls were used.
levels @ < 0.05).
APO A-I/B 5 6 7* 8*
1.56 1.28 1.08* 1.02*
340
ratios of Apo A-I/HDLC and Apo A-II/HDLC were unaffected either by alcohol or by abstinence, thus Apo A-I, Apo A-II and HDL cholesterol decreased in parallel (Table 1). There was a greater proportionate increase in Apo A-II than Apo A-I amongst the alcoholics. The initial average concentration of plasma Apo B was lower in the eight patients than in the control subjects (80 f 6 vs 112 + 8 mg/dl, p < 0.01). The Apo B level increased in five of eight patients during the abstinence, especially in the first 2-week period; the overall increase was not statistically significant. These changes in concentrations of major apolipoproteins were maximally reflected in the ratios of Apo A-I/APO B. A decrease from the initial value of 1.56 to a value of 1.08 at the end of the abstinence period represented a 31% change (p < 0.005). The latter value is very similar to the Apo A-I/APO B ratio in control subjects (1.02). The Apo C-III and Apo E concentrations were measured in six and four patients, respectively, and their average initial levels (9.7 t 2.7 and 8.1 + 3.7 mg/dl, respectively) did not differ markedly from the values for the control subjects (9.4 + 2.3 and 11.5 + 2.7 mg/dl, respectively). The abstinence-induced change in these apolipoproteins was less consistent, with Apo C-III showing a reduction in four of six patients and Apo E registering an increase in three of the four patients at the end of the 4-week abstinence period.
Discussion The present study confirms the observations of the Scandinavian investigators [3, 41 that the high HDL cholesterol levels seen in alcohol abuse decline during a period of abstinence. Moreover, our data also show substantial reductions in HDL cholesterol levels when the initial values were only slightly elevated or within normal limits. In several of our patients, the HDL cholesterol levels (after a 4-week abstinence) approached the lower limit of normal range [ 141. This study also demonstrated that alcohol abstinence affects the levels of various apolipoproteins. The most novel findings were the reduction in levels of apolipoproteins A-I and A-II which paralleled the change in HDL cholesterol. The abstinence-induced changes in Apo A-I/APO B ratio, which decreased significantly in all eight patients, were striking. These data are consistent with other reports. Johansson and Laurel1 [l] reported that some of their alcoholic patients had reduced levels of electrophoretically separated /3-lipoproteins. Similarly, Castelli et al. [ 151 demonstrated that patients consuming larger amounts of alcohol had decreased levels of LDL cholesterol. Little definitive information is available on the possible mechanism(s) involved in the alcohol-induced increase of HDL cholesterol or apolipoproteins or in their changes following abstinence. Some investigators have reported that drugs, which either stimulate [16, 171 or reduce [18] the drug-metabolizing enzyme activity of the liver, also affect plasma HDL
341
cholesterol levels. An alcohol-induced increase in the activities of these enzymes has been previously reported [19, 201 and may be one of the contributing factors. The plasma or tissue lipoprotein lipase [ 21, 221 may be another factor regulating the HDL cholesterol levels. There is a strong positive correlation between the activity of adipose tissue lipoprotein lipase and HDL cholesterol levels [ 231. Several investigators have reported an increase in the activity of adipose tissue lipoprotein lipase after a prolonged alcohol intake [ 5, 241. Other possibilities which need careful study include an alcohol action on lecithin-cholesterolacyltransferase (LCAT), or rate of cholesterol removal from peripheral tissue, cholesterol excretion or synthesis. Studies from this and other laboratories [25] have indicated that lipoproteins represent a mixture of several discrete lipoprotein families, including lipoproteins A, B, C, D, E and F, which are secreted from the liver and/or intestine as the so-called nascent lipoproteins or are generated during the lipolysis of triglyceride-rich lipoproteins [ 261. Approximately 50% of HDL is generally thought to be protein, with phospholipid accounting for 25%, cholesterol 20%, and triglyceride 5% of this lipid moiety. Apolipoproteins in HDL are 90% APO-A, with A-I about 60% and A-II approximately 30%. Results of the present study indicate that the alcohol-induced increase in HDL cholesterol or its reduction during abstinence is accompanied by corresponding changes in the concentration of lipoproteins A and, to some extent, lipoproteins C. From the results of this study, it is not possible to predict whether these lipoprotein families were generated as nascent lipoproteins and/or as products of lipolysis. However, if high activity of lipoprotein lipase results in increased levels of HDL2 particles [22, 271, lipoproteins A and C may have been generated by both pathways. Danielsson et al. [ 31 reported that the chronic alcoholics exhibited either a moderate increase of HDLs or a four- to fivefold increase of HDL2. This may indicate that the nascent lipoproteins were the primary contributors to increased HDL levels in the former group of patients, while the lipoproteins generated during lipolysis represented the primary contributors in the latter group of chronic alcoholics. To further substantiate the role of lipolysis in chronic alcoholics, we have determined the HDL/VLDL ratio of Apo C-III. We have previously demonstrated that the high ratios (>1.5) of Apo C-III reflect an efficient and the low ratios an impaired catabolism of triglyceride-rich lipoproteins [ 281. Results of this analysis demonstrated that the Apo C-III ratio decreased significantly during the 4-week abstinence in two out of three patients. This finding suggests that the chronic alcohol intake may have a different effect on the qualitative and quantitative composition of high density lipoprotein families in individual patients. In some, the increased levels of HDL may be caused by increased secretion of nascent lipoprotein particles, and in others by increased lipolysis. The relevance of the above findings in chronic alcoholics to the general population is not clear. However, it has been reported by several investigators that even in the nonalcoholic subjects alcohol consumption is associated with an increase in plasma HDL cholesterol levels [6, 151. High density lipo-
342
protein is frequently viewed as a lipoprotein which is involved in the removal of free cholesterol from arterial sites. The LCAT enzyme may be considered a key step in this process. APO-I is generally associated with LCAT activity. The function of Apo A-II is less clear; however, it may stabilize the esterified cholesterol. While the enhancing effect of alcohol consumption on the HDL cholesterol levels and their reduction during alcohol abstinence have been reported by numerous investigators [ 1 - 61, the possible confounding effect of dietary changes, especially during abstinence studies, should also be considered. Although dietary intake in our patients was not measured, a general observation by the allied health personnel failed to observe any abnormal dietary pattern. Furthermore, their weight gain during the 4-week observation period (approximately 1 kg) was not sufficient to explain the observed marked reduction in plasma HDL cholesterol levels. In agreement with other findings [ 51 our data suggest that the observed lipoprotein and apolipoprotein changes were primarily due to alcohol abstinence. Acknowledgements The authors wish to thank Dr. John H. Kalbfleisch for his statistical advice and the assistance of J. Fesmire, R. Burns, E. Gehrke, R. Ballou, and J. Owenby for their valuable technical assistance. This work was supported in part by research grants HL 14378 and HI 23181 from the National Institute of Health, by NIDA grant DA 07181, by the Oklahoma Medical Research Foundation, and by the U.S. Veterans Administration. References B. G. Johansson and C. B. Lowell, Disorders of serum or-lipoproteins after alcohol intoxication. Stand. J. Clin. Lab. Inuest., 23 (1969) 231. B. G. Johansson and A. Medhus, Increase in plasma e-lipoproteins in chronic alcoholics after acute abuse. Acta Med. &and., 195 (1974) 273. B. Danielsson, R. Ekman and G. Fex, Changes in plasma high density lipoproteins in chronic male alcoholics during and after abuse. Stand. J. Clin. Lab. Invest., 38 (1978)
113. S. Wallerstedt,
A. Gustafson and R. Olsson, Serum lipids and lipoproteins during after heavy alcohol consumption in chronic alcoholics. Stand. J. Clin. Lab. Invest., 37 (1977) 599. B. B. Blefrange, I. Hagerstrand and N. Nilsson-Ehle, Alterations of lipid metabolism in healthy volunteers during long-term ethanol intake. Eur. J. Clin. Inuest., 7 (1977) abstinence
127. J. J. Barboriak, A. F. Anderson and R. G. Hoffmann, Interrelationship between coronary artery occlusion, high-density lipoprotein cholesterol, and alcohol intake. J. Lab. Clin. Med., 94 (1979) 348. W. D. Block, J. K. Jarrett and J. B. Levin, Use of a single color reagent to improve the automated determination of serum total cholesterol. In L. T. Skeggs Jr. (ed), Automation in Analytical Chemistry, Mediad, New York, 1965, pp. 345 - 347.
343