Effect of Chronic Alcohol Use on Hepatic Testosterone 5a-A-Ring Reductase in the Baboon and in the Human Being

Transcription

1 GASTROENTEROLOGY 77: Effect of Chronic Alcohol Use on Hepatic Testosterone 5a-A-Ring Reductase in the Baboon and in the Human Being GARY G. GORDON, JOZEF VITTEK, RICKY HO, WILLIAM S. ROSENTHAL, A. LOUIS SOUTHREN, and CHARLES S. LIEBER Department of Medicine. Section of Endocrinology. New York Medical College. Alcoholism Research and Treatment Center. V. A. Medical Center. and Mt. Sinai School of Medicine. New York. New York Hepatic testosterone 5a-A-ring reductase (HTAR) activity was measured in open liver biopsies in eight alcohol-fed baboons and eight pair-fed controls. The animals were studied after at least 1 yr of alcohol feeding. In the alcohol-fed animals, a significant fall in enzyme activity was noted. This occurred whether the enzyme levels were related to soluble protein, to DNA, or to wet tissue weight, showing that the change was due to a decrease in the specific activity of the enzyme. In addition, aspiration liver biopsy specimens were obtained from 14 men and women with alcoholic liver disease. Again, there was a significant decrease in HT AR activity in these patients compared with a fl/;>rmal population. No relationship was found between hepatic histology and HT AR levels in either the baboon or human population with alcoholic liver disease, suggesting that the changes in enzyme activity were related to an alcohol effect rather than to liver disease per se. This study demonstrates that chronic alcohol use decreases the function of the enzyme which controls an important rate-limiting step in the metabolism of testosterone in the liver and that this effect may be due primarily to alcohol. The clearance of steroid hormones, such as cortisol,' aldosterone/ and testosterone,3 from the circulation is decreased in patients with cirrhosis of the liver. This decrease in rate of metabolism has been attributed to an increase in plasma protein binding and/or to a decrease in hepatic blood flow.'-3 The increase in protein binding is associated with a decrease in Received August Accepted February Address reprint requests to: Gary G. Gordon. M.D.. New York Medical College Fifth Avenue. N ew York. New York Supported by NIH grant Nos. AA02300, AA03508, AA and AM by the American Gastroenterological Association / 79/ $02.00 the free (easily diffusible) fraction of the circulating hormone.' This results in decreased delivery of the hormone to peripheral sites of metabolism and undoubtedly plays a major role in the control of the rate of hormone removal from the blood" However, for weakly bound hormones, such as aldosterone, changes in binding cannot explain decreased whole body clearance and/or hepatic extraction" For these steroids, the decreased hepatic blood flow and/or the shunting of blood through or around the liver in cirrhosis may be more important mechanisms for the decrease in clearance.'-3 Another mechanism that may account for a decreased hepatic extraction is an alteration in steroid A-ring reduction" This reaction is a significant and rate-limiting step in the hepatic intracellular metabolism of such steroids as cortisol, aldosterone, androstenedione, and testosterone,5 resulting in metabolic inactivation of these steroid hormones. s With regard to testosterone, induction of hepatic testosterone 5a-A-ring reductase (HT AR) activity with medroxyprogesterone acetate 6 and subacute use of alcohof has been associated with an increase in the metabolic clearance rate of testosterone. Although urinary metabolite studies have suggested a defect in 5a-reduction in c i r r h there s i s have, " been no direct measurements of hepatic steroid 5a-reductase activity reported to date after. chronic (more than 6 mo j use of alcohol in the experimental animal or in human alcoholic liver disease. This study presents such data in both the baboon and the human using hepatic biopsy material. Materials and Methods Baboon Studies Sixteen young adult female baboons were studied. Eight experimental animals were individually caged and

2 July 1979 EFFECT OF ALCOHOL ON HEPATIC STEROID REDUCTASE 111 fed alcohol (50% of total calories) in a liquid diet. Pair-fed controls were maintained with carbohydrate substituted isocalorically for alcohol. Open liver biopsies were carried out under ketamine anesthesia, and the specimens were assayed for testosterone 5a-A-ring reductase (HT AR) activity. Patient Group The study population comprised 14 patients hospitalized at New York Medical College for active alcohol-related liver disease. Each patient was subjected to aspiration liver biopsy for diagnostic purposes. A portion of the biopsy material was used for the HT AR assay and the remainder for histologic study. No hepatic tissue was obtained solely for this project, and the study was approved by our committee for the protection of human subjects in research. The usual biochemical parameters of hepatic function were obtained in these patients. All patients had a history of a1cohol abuse for many years and were clinically and pathologically documented to have active cirrhosis or alcoholic hepatitis. Hepatic testosterone 5a-Aring reductase levels have previously been characterized in a normal population of similar age distribution. 6 7,9 Hepatic Testosterone 5a-A-Ring Reductase Assay The hepatic tissues were either frozen immediately at -70 C and assayed within 14 days or assayed immediately. No loss of activity was noted in specimens frozen for 14 days as compared with fresh tissue. The 5a-reductase assay has been r e p o rin t detail e previously.lo In brief, the liver specimen was homogenized in 2 volumes of 0.25 M sucrose using a glass microhomogenizer. The assay mixture contained 0.01 M potassium acetate buffer (ph 5.6), NADPH (7.2 X 10-4 M), testosterone (6 X 10-4 M), containing 7a- 3 H-testosterone (10 6 count/min), and liver enzyme preparation (whole homogenate with an average protein content of 8 mg per assay) in a total volume of 1 m!. The mixture was incubated at 37 C, and 0.2 ml aliquots were taken at lo-min intervals. The reaction was terminated by placing each aliquot in 4 ml of dichloromethane. Hepatic testosterone 5a-A-ring reductase activity was determined by the rate of formation of 5a-reduced metabolites of 3H-testosterone after isolation and identification. The enzyme activity was expressed as nanomoles per 10 minutes per milligram protein in the human material and per milligram protein, DNA, and wet weight of hepatic tissue in the baboon studies. DNA was measured by the method of Martin and Donohue." The tissues removed at biopsy were reviewed independently by the hospital pathologist and one of the authors (RH). The severity of the disease was graded according to the magnitude of changes in fatty metamorphosis, fibrosis, and necrosis on a scale of 0 to 4+. The clinical biochemical studies were performed in the hospital laboratory. Statistical analyses were carried out using the Student's t-test and Pearson correlation coefficients with a Programma 101 table top computer. Results Baboon Studies The baboons were fed alcohol as reported previously12 for periods ranging from 12 mo to 6 yr. The pathology ranged from normal histology (one animal), mild fatty changes (three animals), marked fatty changes (two animals), incomplete cirrhosis (one animal) to fully developed cirrhosis (one animal). All control animals had normal hepatic histology. Figure 1 shows the results of the HT AR data in the baboon. A significant decrease in enzyme activity was noted whether these data were expressed as 5a-metabolite formed per milligram of soluble protein, of DNA, or of wet weight of the liver. However, no relationship existed between the HT AR levels and hepatic histology. This was most evident in the a l c o h oanimal l f ethat d had a normal histologic appearance of the liver and yet showed a marked decrease in HT AR levels. Human Studies Table 1 shows the age and sex distribution of the human population studied, the clinical biochemical findings, and details of the light microscopy examination of the liver biopsy specimens. The patients ranged in clinical state at time of biopsy from severely ill to much improved over their condition on admission. None of the patients used alcohol (by history) within 1 wk of biopsy. The patients were all on multivitamins but were receiving no other medication in common. Figure 2 shows HT AR levels in patients with alcoholic liver disease and in healthy subjects. There was a significant decrease in the level of this hepatic enzyme both in men and women with alcoholic liver disease when compared with the control group, members of which were of similar age and sex distribution. Because no statistical difference existed between men and women in either the alcoholic liver disease group (P> 0.30) or normal subjects (P > 0.20), these data were pooled. Again, a significant decrease in enzyme levels was noted when the normal and the alcoholic liver d { s groups a s e w e compared without regard to gender. Furthermore, when e the patients with cirrhosis were compared with those with alcoholic hepatitis, no significant difference (P > 0.20) was noted; both groups shared a significant decrease in enzyme activity. The relationship between enzyme activity, liver function, and histologic changes was studied by regression methods and, again, as in the baboons, no significant correlation was established with any of these parameters.

3 112 GORDON ET AL. GASTROENTEROLOGY Vol. 77, No.1 * '0; '2 300 ::J... c 'E Q ";;; 250 E..s 200 :> i= L&J 150 en u => o 100 * mg PROTE IN I 102 r'<'''l o CONTROL ALCOHOL T SEM l *mg DNA r'<o", *gm TISSUE (WET WEIGHT) r p<o-005 Figure 1. Hepatic 5a-A-ring reductase activity in control and alcohol-fed baboons. (!) z a: o Table 1. Clinical, Biochemical, and Histologic Data in the Study Population with Alcoholic Liver Disease Liver histology Hepatocellular necrosis Age Bilirubin SGOT Albumin Globulin Fatty meta- and cellular Patient (yr) (mg/dj) (IV/liter) (g/dj) (g/dj) morphosis Fibrosis infiltration Diagnosis MALE Cirrhosis, advanced Cirrhosis, early Hepatitis Cirrhosis, advanced Hepatitis Cirrhosis, moderate Cirrhosis, early Cirrhosis, moderate Cirrhosis, advanced Hepatitis Cirrhosis, advanqld Mean± S.E.M. 5.7 ± ± ± ± 0.3 FEMALE Hepatitis Cirrhosis, early Cirrhosis, advanced Mean± S.E.M. 1.4 ± ± ± ± 0.3 Normal range ( ) ( ) ( ) ( )

4 July 1979 EFFECT OF ALCOHOL ON HEPATIC STEROID REDUCTASE 113 r p< r p<0005 TSEM I" p<o OOI n=9 I I NORMAL A.L.D NORMAL ALD NORMAL A.L.D MALE MALE FEMALE FEMALE MALE AND FEMALE Figure 2. Hepatic Sa-A-ring reductase activity in normal men and women and in patients with alcoholic liver disease (A.L.D.). Discussion The present study demonstrates that chronic alcohol feeding is associated with a decreased level of activity of the hepatic testosterone 5a-reductase system in the baboon. It is also apparent that there is significant decrease in the level of this enzyme in humans with alcoholic liver disease. Because this enzyme system is rate limiting for the initial hepatic metabolic inactivation of testosterone and other A ring unsaturated steroids, a defect in this enzyme, in addition to protein binding and hemodynamic alterations, would be expected to contribute to the decreased removal rate of these hormones seen in patients with cirrhosis. 3 However, it should be noted that the total 'mass of hepatic enzyme (Le., specific enzyme activity X hepatic weight) may be of importance in determining the quantitative contribution of the liver to steroid metabolism, and this parameter could vary independently of enzyme levels. The observation in the baboon an'd man that HT AR levels varied independently of the alterations in liver histology suggests that alcohol per se can alter the functional state of this enzyme. We have previously reported that subacute exposure to alcohol (over a period of 4 wk) can increase the level of this enzyme in human volunteers' 13 and in the experimental animal,13 Similar findings have also been reported by others.14 It is now apparent that more prolonged exposure to alcohol, over a period of years, is associated with a decrease in the level of activity of this enzyme system. The fact that HT AR activity was significantly decreased in the baboon whether these changes were related to soluble protein, to DNA, or to wet weight shows that the enzyme activity per cell is decreased and that the fall in enzyme activity is not due to "dilution" of the specific activity of the enzyme protein by other proteins or by fatty infiltration or necrosis. The lack of correlation of the decrease in HT AR activity with hepatic histology also points to a specific decrease in enzyme activity rather than to a nonspecifc alteration due to increased fibrosis or necrotic elements. Studies in cirrhosis, using other methodology (Le., urinary radiometabolite patterns) are compatible with a decreased hepatic 5a-reductase activity for cortisol." Although our studies show decreased HT AR activity, each steroid hormone may have a separate and distinct isoenzyme 15 whose functional state could change independently, although a recent study suggests that the same enzyme may metabolize both testosterone and cortisol,16 Additional studies are necessary to determine whether a general reduction in hepatic steroid 5a-reductase activity is present in cirrhosis. The hepatic steroid 5a-reductase level has also been reported to be decreased in acute intermittent porphyria17 while cutaneous 5a-reductase activity was normal in this disorder."" No information exists concerning 5a-reductase activity in skin and other extrahepatic sites in alcoholic liver disease. However, a decreased 5a-reductase activity in male accessory sexual glands, i.e., prostate and seminal vesicles, could lead to a decreased conv'ersion of testostm-one to the biologically active dihydrotestosterone (DHT)"" and contribute to the hypotestosteronemic hypogonadism seen in alcoholic cirrhosis in the male. The significance bf the observed alteration in hepatic metabolism of testosterone in alcoholic liver disease is not clear, although an alteration of one major pathway of testosterone metabolism may divert the steroid to other pathways, such as to the estrogens. The present study offers additional support to the concept that alcohol use is characterized initially by an adaptive response (i.e., increased enzyme activity) and later, with more chronic exposure, to a pat-

5 114 GORDON ET AL. GASTROENTEROLOGY Vol. 77, No.1 tern (i.e., decreased activity) of microsomal injury References 1. Peterson RE: Adrenocortical steroid metabolism and adrenal cortical function in liver disease. J Clin Invest 39: , Coppage WS Jr, Island DP, Conner AE, Liddle GW: The metabolism of aldosterone in normal subjects and in patients with hepatic cirrhosis. J Clin Invest 41: , Southren AL, Gordon GG, Olivo J, Rafii F, Rosenthal WS: Androgen metabolism in cirrhosis of the liver. Metabolism 22: , Tait JF, Burstein S: In vivo studies of steroid dynamics in man. In The Hormones. Vol V. Edited by G. Pincus, KV Thimann EB Astwood. New York, Academic Press, 1964, p Tomkins GM: Enzymatic mechanisms of hormone metabolism I: oxidation-reduction of the steroid nucleus. In Recent Progress in Hormone Research. Vol XII. Edited by G. Pincus. New York, Academic Press, 1956, p Gordon GG, Altman K, Southren AL, Olivo J: Human hepatic testosterone A-ring reductase activity: effect of medroxyprogesterone acetate. J Clin Endocrinol Metab 32: , Gordon GG, Altman K, Southren AL, Rubin E, Lieber CS: Effect of alcohol (ethanol) administration on sex-hormone metabolism in normal men. N Engl J Med 295: , Zumoff B, Bradlow HL, Gallagher TF, Hellman L: Cortisol metabolism in cirrhosis. J Clin Invest 46: , Gordon GG, Southren AL, Tochimoto S, Olivo J, Altman K, Rand J, Lemberger L: Effect of medroxyprogesterone acetate (provera) on the metabolism and biological activity of testosterone. J Clin Endocrinol Metab 30: , Altman K, Gordon GG, Southren AL, Vittek J, Wilker S: Induction of hepatic testosterone A-ring reductase by medr- oxyprogesterone acetate. Endocrinology 90: , Martin, RF, Donohue DC: New analytical procedure for the estimation of DNA with p-nitrophenylhydrazine. Anal Biochem 47: , Lieber CS, DeCarli LM: An experimental model of alcohol feeding liver injury in the baboon. J Med Prim 3: , Rubin E, Lieber CS, Altman K, Gordon GG, Southren AL: Prolonged ethanol consumption increases testosterone metabolism in the liver. Science 191: , Bode C, Martini GA, Bode JC: Effect of alcohol on microsomal cortisol 4 en-5a-reductase in the liver of rats fed on a standard or low protein diet. Horm Metab Res 10:62-64, Peterson RE: Metabolism of adrenal cortical steroids. In The Human Adrenal Cortex. Edited by NP Christy. New York, Harper and Row, 1971, p Fisher LK, Ogut MD, Moore RJ, Goebelsman U, Weitzman n, Isaacs H Jr, Griffin JE, Wilson JD: Clinical endocrinological and enzymatic characterization of two patients with 5a-reductase deficiency. Evidence that a single enzyme is responsible for the 5a-reduction of cortisol and testosterone. J Clin Endocrinol Metab 47: , Kappas A, Bradlow HL, Gillette PN, Gallagher TF: Studies in prophyria I: a defect in the reductive transformation of natural steroid hormones in the hereditary liver disease acute intermittent porphyria. J Exp Med 136: , Bradlow HL, Anderson K, Kappas A: Differences between cutaneous and hepatic steroid ""-5a-reductase in patients with acute intermittent porphyria. In Porphyrins in Human Diseases. Edited by M Doss. Basel, S. Karger AG, 1976, P Bruchovsky N, Wilson JD: The conversion of testosterone to 5a-androstan-17,B-ol-3-one by rat prostate in vivo and in vitro. J BioI Chem 243: , Lieber CS: Liver adaptation and injury in alcoholism. N Engl J Med 288: , 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 72: , 1975