- Email: [email protected]
Impaired androgen 16α-hydroxylation in hepatic microsomes from carbon tetrachloride-cirrhotic male rats
GASTROENTEROLOGY
Impaired Androgen 16wHydroxylation Hepatic Microsomes From Carbon Tetrachloride-Cirrhotic Male Rats MICHAEL
MURRAY,
LOUISE
ZALUZNY,
1987:93:141-7
in
and
GEOFFREY C. FARRELL Department
of Medicine,
University
of Sydney,
Hepatic cirrhosis produced by repeated inhalation of carbon tetrachloride is associated with reduced levels of microsomal cytochrome P450. In this study the C,,-steroids androstenedione and testosterone were used as specific probes of the functional activity of several forms of cytochrome PdSOin microsomal fractions from control and cirrhotic rat liver. The 16~ principal finding, that androstenedione hydroxylation and testosterone Za-, 16~, and 17~ hydroxylation were reduced to 14%--38% of control activity, strongly suggests that levels of the male sexually differentiated cytochrome PdsO (P450,,,] are decreased in hepatic cirrhosis. The activity of other cytochrome P,,,-mediated &-steroid hydroxylases, with the exception of androstenedione Go-hydroxylase, appeared essentially unaltered in microsomes from cirrhotic rats. Cirrhosis induced by carbon tetrachloride was also associated with greatly decreased activity of the microsomal cytochrome PdSOindependent 17P-oxidoreductase, an enzyme that catalyzes the conversion of androstenedione to testosterone. Consequently, and in view of the impaired activity of cytochrome P4,,,-mediated testosterone 17a-hydroxylation, the capacity of cirrhotic microsomes to catalyze the interconversion of androstenedione and testosterone was much lower than that of control microsomes. The present data confirm and extend earlier observations that selective impairment of drug oxidation pathways occurs Received August 5, 1986. Accepted November 10, 1986. Address requests for reprints to: Michael Murray, Ph.D., Department of Medicine, Westmead Hospital, Westmead NSW 2145, Australia. This work was supported by a grant from the Australian National Health and Medical Research Council. M. Murray is a National Health and Medical Research Council Research Fellow. The authors thank Diane West for secretarial assistance in the preparation of the manuscript and the Department of Anatomical Pathology at Westmead Hospital for histologic studies. 0 1987 by the American Gastroenterological Association 0016-5085/87/$3.50
Westmead
Hospital,
Westmead,
Australia
in hepatic cirrhosis. These changes are unrelated to the acute toxicity produced by carbon tetrachioride exposure. The available evidence supports the assertion that specific forms of cytochrome Pd5,, are subject to altered regulation in cirrhosis. Cytochrome PJSO (P4& is a family of isozymes, situated in the hepatic endoplasmic reticulum, that is active in the oxidation of a variety of endogenous and exogenous compounds, including drugs and steroid hormones. Oxidative metabolism of many drugs is impaired in patients with liver disease (l-3). In vitro studies of microsomal P450-mediated mixed-function oxidase activity (1)and in vivo studies of hepatic drug clearance (2) have led to the suggestion that selective changes in individual forms of PJSOmay occur in chronic liver disease. Studies of experimental cirrhosis produced in rats by the repeated inhalation of carbon tetrachloride (CC14) (4) have demonstrated that hepatic microsomes from cirrhotic rats have a decreased capacity to bind (5) and to catalyze the oxidative metabolism (6,7)of drug substrates. In this model hepatic heme turnover was normal (8), and the capacity for microsomal enzyme induction was retained (7). From these findings it was proposed that impaired drug oxidation may be a consequence of altered regulation of certain forms of PJ5,, (8). Recent studies, undertaken in the choline-deficient rat model of hepatic cirrhosis, suggested that the suppression of P 450 levels in chronic liver disease is selective for some forms of cytochrome (9). A large body of evidence now exists that individual PdsOs catalyze the position-specific hydroxylation of C19-steroids, including androst-4-ene-3,17dione (androstenedione) and 17P-hydroxyandrost4-ene-3-one (testosterone) (10-12). The present study used these steroid substrates as probes to Abbreviation
used
in this paper:
P,,,,
cytochrome
P,,,.
GASTROENTEROLOGYVol. 93, No. 1
142 MURRAY ET AL.
pathways in microsomal fracassess hydroxylation tions from Ccl,-cirrhotic and control rat liver. From these studies it is apparent that selective changes in individual forms of P4sa occur in the Ccl, rat model of chronic liver disease and that these changes are attributable to the pathophysiologic state of cirrhosis itself.
amide adenine dinucleotide phosphate, reduced form. Incubations were terminated by the addition of 5.5% zinc sulfate and centrifuged, and the supernatant was extracted with chloroform. The organic phase was evaporated to in a small volume of dryness under Nz, reconstituted chloroform, and applied to thin-layer chromatography plates (silica gel 60 F 254 type, 20 cm x 20 cm X 0.25 mm thickness and activated for 15 min at 100°C before use; E. Merck, Darmstadt, F.R.G.).
Materials and Methods Identification
Chemicals [4-r4C]Androstenedione (sp act 59 mCiimmo1; 98% purity by thin-layer chromatography) and [4-14C]testosterone (sp act 56.9 mCiimmo1; 97% purity by thin-layer chromatography) were obtained from Amersham Australia, Sydney, Australia. Unlabeled androstenedione, Sphydroxy- and 16a-hydroxyandrostenedione, testosterone, and 16a-hydroxytestosterone, as well as all biochemicals, were purchased from Sigma Chemical Co., St. Louis, MO. 7a-Hydroxyandrostenedione and 2a-hydroxytestosterone were obtained from Professor D. N. Kirk and the MRC Steroid Reference Collection, Queen Mary’s College, London, United Kingdom. 6P-Hydroxy- and 7a_hydroxytestosterone were purchased from Steraloids Inc., Wilton, N.H. 16P-Hydroxyandrostenedione was prepared enzymatitally by the action of 3P-hydroxysteroid dehydrogenase [Sigma) on 3P,16/3-dihydroxyandrost-5-ene-17-one (MRC Collection). Solvents and miscellaneous chemicals were from Ajax Chemicals, Sydney, Australia, and were at least analytical reagent grade. Animals Male Wistar rats (starting weight 150 g) were obtained from the animal house at the Westmead Hospital Institute of Clinical Pathology and Medical Research. Rats were given phenobarbital (1g/L) in their drinking water throughout the period of exposure to Ccl, (10 wk) (7). Controls were littermates that received phenobarbital as above. Experiments were performed 10 days after stopping phenobarbital and Ccl4 inhalation. Hepatic mixedfunction oxidase activity was stable at this time (7). Microsomal fractions were prepared as described elsewhere (9). The development of cirrhosis was assessed at the time of death from the macroscopic appearance of the liver. Liver function tests and histologic assessments were performed as described previously (7,8). Steroid
Hydroxylase
Activity
Microsomal androstenedione and testosterone hydroxylase activities were assayed essentially by the procedure of Gustafsson and Ingelman-Sundberg (13). Briefly, microsomal fractions from control or cirrhotic rat liver (0.75 mg protein/ml in 100 mM potassium phosphate buffer containing 1 mM ethylenediaminetetraacetic acid, pH 7.4) were incubated with 175 PM [‘4C]androstenedione or [‘4C]testosterone (4X105 dpm) for 10 min at 37°C. Reactions were initiated by the addition of 1 mM nicotin-
of Steroid
Metabolites
Androstenedione metabolites were resolved on thin-layer chromatography plates by development (twice) in the solvent system CHClJethyl acetate (1:2, vol/vol) as described by Waxman et al. (11).Testosterone metabolites were resolved similarly except that the first solvent system was dichloromethane/acetone (4:1, vol/vol) and the second system was chloroform/ethyl acetate/ethanol (4:1:0.7, vol/vollvol) (10). Zones corresponding to hydroxylated androstenedione and testosterone standards were visualized under ultraviolet light and scraped into vials for scintillation spectrometry (Aquasol scintillant, New England Nuclear Corp., Boston, Mass.). Other
Assays
Protein was determined by the method of Lowry et al. (14) with bovine serum albumin as standard, and P4s0 levels were estimated by the procedure of Omura and Sato (15). Analysis
of Data
In these experiments, mean values of data obtained from cirrhotic and control hepatic microsomes were compared using the unpaired Student’s t-test [two-tailed).
Results Microsomal in Control
P4s0 and Androgen Metabolism and Cirrhotic Rat Liver
Consistent with previous reports from this laboratory, levels of total microsomal P4sa were reduced in cirrhotic liver compared with controls (Table 1) (7,8). The metabolism of the androgens androstenedione and testosterone was studied in hepatic microsomal fractions from rats with CCll-induced cirrhosis, and in control animals. Formation of metabolites from both substrates was linear for at least 10 min (not shown). The capacity of cirrhotic rat liver microsomes to catalyze Clg-steroid metabolism was markedly lower than that of control fractions. In control microsomes 25.5% + 3.7% of the total substrate (androstenedione) was metabolized after 10 min, whereas only 12.0% +- 3.4% conversion occurred in cirrhotic microsomes (not shown). Very similar findings were noted with testosterone as substrate: 18.6% * 5.4% of the initial substrate was metabo-
STEROID HORMONE METABOLISM
July 1987
Table
1. Cytochrome
P4s0 and Androstenedione
Hydroxylation
in Hepatic
Microsomes
IN CIRRHOTIC RATS
From Cirrhotic
143
and Control
Rat9
Microsomes
Androstenedione hydroxylation (nmol product formedimin mg protein)
Cytochrome P450b (nmolimg protein)
Control Cirrhotic Percent of control
1.14 2 0.12
1.88 t 0.37
0.81t 0.08 71
1.19k 0.51 59 co.02
P”
W
7a
0.21t 0.06 0.27L 0.06 110 NS
16a
2.83+ 0.23 0.652 0.29 23
16P
0.43+ 0.14 0.30c 0.14 70 NS
NS, difference not significant. “Values are presented as mean + SD of six individual microsomal fractions. “Spectral determination cytochrome P4,0. cp values are for comparison of cirrhotic and control animal values.
lized in control microsomes after 10 min, and 9.1% ~fr3.4% in cirrhotic microsomal fractions under the same conditions (not shown]. It is therefore apparent that Clg-steroid metabolism is reduced to -50% of control levels in microsomes from cirrhotic rat liver. Non-P450 pathways were also examined in the present study and, from the data presented in Figure 1, it is clear that 17P-oxidoreductase activity (which converts androstenedione to testosterone) was significantly lower (p < 0.001) in cirrhotic rat liver. Thus, formation of testosterone was 0.45 nmol/ min . mg microsomal protein in control hepatic fractions and 40% of this rate (0.18 nmolimin . mg protein] in cirrhotic liver microsomes. Microsomal 5a-reductase activity was found to be negligible in both control and cirrhotic hepatic microsomes (data not shown]. Effect of Cirrhosis Hydroxylation
on Microsomal
Steroid
The point of major interest that emerged from this study is that androgen hydroxylation pathways
cmcontrol 0
Figure
1. 17pOxidoreductase
cirrhotic
activity in microsomes from control and cirrhotic rat liver. Significantly different from control. *p < 0.001.
of total
were affected differently in the CC14-rat model of cirrhosis (Tables 1 and 2). Thus microsomal androthe principal stenedione S/?- and 16a-hydroxylation, routes of hydroxylation in control microsomes, were decreased in cirrhosis to 58% and 23% of their respective control activities (Table 1). In contrast, the rates of androstenedione 7a- and lG@hydroxylation in microsomes from cirrhotic rat liver were not significantly different from control. The data in Table 1 indicate absolute rates of product formation in both types of microsomes, but it is important to recognize that P 450in cirrhotic microsomes was only 71% of control levels (Table 1). As this enzyme system has a fundamental role in steroid hydroxylation, the change in P 450levels could be an important factor in the observed rates of hydroxylation. Consequently, rates of production of individual metabolites were corrected for differences in microsomal PdsO and the resultant turnover numbers are presented in Figure 2. Again, 16a-hydroxylation was significantly decreased (p < 0.001) to about 30% of control; 16/S and 6P-hydroxylation of androstenedione were not different in control and cirrhotic hepatic microsomes. However, in cirrhotic microsomes, turnover of androstenedione by 7a-hydroxylation was increased (Figure 2: 0.34 k 0.08 nmol product/min . nmol PasO in cirrhosis versus 0.21 k 0.07 in control liver; p < 0.02). Testosterone hydroxylation activities were also measured in control and cirrhotic rat hepatic microsomes (Table 2). The findings were consistent with those obtained using androstenedione as the steroid substrate. Thus, testosterone 2a-, 16~, and 17~ hydroxylations (the latter pathway produces androstenedione) were all catalyzed with much lower efficiency in microsomes from cirrhotic liver whether expressed as absolute activity (14%-38% of control; Table 2) or as turnover numbers (19%--53% of control; Figure 2). In contrast, the activity of the testosterone 6P-hydroxylation pathway was unchanged from control levels, and the 7a-hydroxylase pathway appeared more active in cirrhotic microsomes (p < 0.002; Table 2).
GASTROENTEROLOGY Vol. 93, No. 1
MURRAY ET AL.
144
3.0
1
androstenedione
hydroxylation
B/D ratio was 0.64:1.00 in control microsomes but, in cirrhosis, the ratio was quite different (1.41:1.00). From this finding it is clear that the preferred site of hydroxylation is the D ring in control microsomes and the B ring in cirrhotic liver fractions. Testosterone hydroxylation occurred in the A ring (2a position), B ring (Sp and 7a), and D ring (16a and IV_. . . . . ..c.,,,\ 111 T, L"IICI"15 ___L._l" +L_ -..a:,$ L_.A.._....l, l1c.I p"alu"us,. LUG ldl.llJ "I Ilyulunyld-
control uul cirrhotic
0
tions
occurring
in the A, B, and D rings was but in cirrhosis these values were dramatically altered to 0.06:1.00:0.34. Thus, D-ring hydroxylation was predominant in control hepatic microsomes, whereas the B ring was the preferred site of hydroxylation in microsomes from cirrhotic rats. A ring hydroxylation was also reduced in cirrhotic microsomes as a consequence of impaired testosterone 2a-hydroxylase activity. 0.46:1.00:1.35,
St3
@I
7a
160
testosterone
hydroxylation
.
Discussion
2a
60
70
16a
1701
Figure 2. Turnover of androstenedione and testosterone by different pathways of hydroxylation in microsomes from control and cirrhotic rat liver. Significantly different from control. *p < 0.001. **p < 0.02.
The most important finding of the present study is that selective impairment of specific pathways of &-steroid hydroxylation occurs in hepatic microsomes from rats with CC14-induced cirrhosis. There is now a large body of evidence that the positional hydroxylation of testosterone and androstenedione is catalyzed by individual forms of PaSO fan_I 7) (L”--su,.
IPrn~imlo l”Yl”UU
d,,rlinc frnm thio lcahnrotnmr lnrl ilLUUlU.3 AI”lll LA1113 lU”“lUL”lJ 1-u tn I”
The regioselectivity of androstenedione and testosterone hydroxylation in contr61 and cirrhotic rat hepatic microsomes was estimated from the turnover data in Figure 2. In control and cirrhotic hepatic microsomes the hydroxylation of androstenedione occurred in the B ring (at the 6p and YCY positions) and the D ring (at the 16~1 and 16p positions). The
the suggestion that the different effects of cirrhosis on individual drug oxidation pathways may be explained by selective effects on specific PdSOisozymes (8). We recently reported a selective reduction of one P450 subpopulation in partially purified P450 fractions from choline-deficient rat hepatic microsomes (9); a similar observation was made using partially purified PdSOfractions from CC14-cirrhotic rat liver microsomes (unpublished data). The present data provide stronger evidence of a selective effect of CC14-induced cirrhosis on individual forms of microsomal Pg5,, and extend earlier studies by providing information on the form (or forms) most affected. The selective decrease in testosterone turnover by 2a-, 16~, and 17a_hydroxylation, as well as the
Table
From
Changes in the Regioselectivity of Androgen Hydroxylation Produced by Hepatic Cirrhosis
2. Testosterone
Hydroxylation
in Hepatic
Microsomes
Cirrhotic
and Control
Rat?
Testosterone hydroxylation (nmol product formed/min mg protein) Microsomes
17ab
2a
6P
7a
160
Control Cirrhotic Percent of control PC
0.82 t 0.07 0.31 * 0.04 38
0.79 f 0.20 0.11 f 0.05 14
1.58 k 0.35 1.57 2 0.22 99 NS
0.16 L 0.02 (1.31f 0.12 190 co.02
1.47 k 0.34 0.32 f 0.07 22
NS, difference not significant. “Values are presented as mean 2 SD of six individual microsomal fractions. b17a-Oxidation of androstenedione formation from testosterone. ‘P values are for comparison of cirrhotic and control animal values.
indicates
rate
July 1987
STEROID HORMONE METABOLISM IN CIRRHOTIC RATS
St3
7a
16a androstenedione Figure
145
7a
2a testosterone
16a
3. Effect of hepatic cirrhosis on the disposition of C,,-steroids. Increased turnover by a particular pathway is indicated by solid lines, decreased turnover by broken lines, and unchanged turnover by dotted lines. The microsomal Ii’/%oxidoreductase is a “P-450 0 refers to those P,,, isozymes involved in the non-P,,, enzyme that converts androstenedione to testosterone. 17~oxidation of testosterone to androstenedione.
reduced turnover of androstenedione by 16~ hydroxylation, suggests that levels of the malespecific P450,,, (16,17) [also termed P450,,T_A (18), and Pa5,,h (19)] are down-regulated in the CC& rat model of hepatic cirrhosis. This is a particularly interesting observation as it is now known that P450,6,is under hypothalamic-pituitary-gonadal control (15). Imprinting of the brain during the neonatal period permits the expression of P450,,, during adult life (16,17). Castration or hypophysectomy, or both, of male rats results in the disappearance of P450,,, apoprotein and its associated 16a-hydroxylase activity (16,18). This effect is now attributed to the abolition of a “male pattern” of growth hormone secretion by the pituitary (16,20). Thus, the selective decrease in the P450,80mediated pathway of steroid hydroxylation suggests that a similarity exists, in terms of microsomal oxidase activities, between hepatic cirrhosis produced by Ccl, and either hypophysectomy or castration of male rats. The function of the hypothalamic-pituitary-gonadal axis can be affected in a variety of ways in patients with chronic liver disease (21). It has also been reported that circulating levels of growth hormone (22) and somatomedins (23,24) are altered in individuals with hepatic cirrhosis. In view of the established role of growth hormone as a primary regulator
of the expression of specific forms of microsomal PJSOin rats, it now seems appropriate that the relationship between growth hormone and P450 isozyme composition should be studied in humans with cirrhosis. Other pathways of &,-steroid metabolism were less affected in hepatic cirrhosis. The 6@hydroxylation of androstenedione was reduced to 60% of control activity in cirrhotic microsomes; this decrease was comparable with the overall decrease in PJ5,, (Table 1). Changes in microsomal androstenedione 7a- and lG@hydroxylase activities were not significantly different from control, but testosterone 7a-hydroxylation was actually about twofold higher in cirrhotic microsomes. This apparent discrepancy between androstenedione and testosterone ?‘ahydroxylation activities in cirrhotic microsomes is probably due to the relative turnover capacity of form P450a(10) [also termed P450,,,~,(25)]. Wood et al. (10) have reported an approximate twofold greater turnover number for testosterone 7cr-hydroxylation than for androstenedione 7a-hydroxylation cataConsequently, it is likely that P,,,, is lyzed by PJSO,. able to compete more efficiently with other isozymes when testosterone is the substrate. These findings have implications for drug oxidation in cirrhotic liver. Male-specific PJSOs that are
146
GASTROENTEROLOGY Vol. 93, No. 1
MURRAY ET AL.
apparently identical with form P450,,, have high catalytic activity in the 4’- and 6-hydroxylation of Rand S-warfarin (E), and hexobarbital 3-hydroxylation (19). A number of additional xenobiotic oxidations are catalyzed efficiently by the male-specific P 4509 including aminopyrine N-demethylation (I 7, 25). Thus, it is not surprising that down-regulation of this enzyme results in impaired metabolic capacity in experimental liver disease. Our recent observations that ethylmorphine N-demethylase and benzo(a)pyrene hydroxylase activities are especially susceptible to the effects of CC14-induced cirrhosis (7) and long-term intake of a choline-deficient diet (9) are consistent with the proposed effects on P450,,,. Non-P,,,-mediated &,-steroid metabolism ip control and cirrhotic rat hepatic microsomes was also investigated. A finding of significance was reduced activity of microsomal 17p-oxidoreductase, an enzyme responsible for conversion of 17-keto steroids to 17/Lhydroxy steroids; e.g., androstenedione is converted to testosterone by this enzyme. This activity was decreased in cirrhotic liver to 40% of control (Figure 1). Interestingly, this enzyme also appears to be under hormonal control although its regulation has not been studied as intensively as that of the sex-differentiated forms of P450. For instance, Takeyama et al. (26) have reported that neonatal castration of mice, but not castration at 60 days of age, results in impaired 17fi-oxidoreductase activity in microsomes prepared at 120 days. One probable effect of diminished 17/3-oxidoreductase activity is that androstenedione would be relatively more available for P4,,-mediated hydroxylation as less of the substrate is converted to testosterone. As depicted in Figure 3, the interconversion between androstenedione and testosterone would be slower in cirrhotic microsomes. This is due not only to decreased l7poxidoreductase activity, but also to decreased P450mediated 17cr-hydroxylation of testosterone to androstenedione. In the present study we have demonstrated a selective reduction in the male-specific steroid 16~ hydroxylase in CC14-cirrhotic rat hepatic microsomes; other P,,,-mediated steroid hydroxylase pathways were not reduced. Thus it is possible to clearly distinguish the Ccl, rat model of hepatic cirrhosis from acute Ccl, hepatotoxicity. The present finding that androstenedione 16ghydroxylation (a pathway mediated by the phenobarbital PJ5&,) was unaffected in CC14-induced cirrhosis is in contrast to the evidence presented elsewhere that levels of P&ob are reduced after acute Ccl, intoxication of rats (27-29). We have observed that similar changes in steroid 16a-hydroxylase activity occur in both the CC14-inhalation and choline-deficient diet models of hepatic cirrhosis (30). Therefore, we be-
lieve that these selective changes in drug oxidation pathways are attributable to cirrhosis and not to the method of its production. Considered together, it is clear that chronic liver disease in experimental animals is associated with striking changes in several pathways of microsomal &-steroid metabolism. These substrates are excellent probes for the activity of several different P450s, and the present study established selective changes in the male-specific isozyme P4so,,,. Future studies will concentrate on the mechanism of this effect and the role of endocrine factors in hepatic cirrhosis in experimental animals.
References 1. Farrell GC, Cooksley WGE, Powell
LW. Drug metabolism in liver disease: activity of hepatic microsomal drug metabolising enzymes. Clin Pharmacol Ther 1979;26:483-92. 2. Klotz U, Fischer C, Muller-Seydlitz P, et al. Alterations in the disposition of differently cleared drugs in patients with cirrhosis. Clin Pharmacol Ther 1979;26:221-7. 3. Schenker S, Hoyumpa AM Jr, Wilkinson GR. The effect of parenchymal liver disease on the disposition and elimination of sedatives and analgesics. Med Clin North Am 1975;59: 887-96.
4. Marshall WJ, McLean AEM. The effect of cirrhosis of the liver on microsomal detoxications and cytochrome P-450. Br J Exp Path01 1969;50:578-83. 5. Murray M, Farrell GC. Different effects of carbon tetrachloride
toxicity and cirrhosis on substrate binding to rat hepatic microsomal cytochrome P-459. Biochem Pharmacol 1984; 33:687-9. 6. Villeneuve
10.
11.
12.
13.
14.
J-P, Wood AJJ, Shand DG, et al. Impaired drug metabolism in experimental cirrhosis in the rat. Biochem Pharmacol 1978;27:2577-81. Farrell GC, Zaluzny L. Microsomal protein synthesis and induction of cytochrome P-450 in cirrhotic rat liver. Aust J Exp Biol Med Sci 1984;62:291-301. Farrell GC, Zaluzny L. Hepatic heme metabolism and cytochrome P,,, in cirrhotic rat liver. Gastroenterology 1985; 89:172-g. Murray M, Zaluzny L, Farrell GC. Drug metabolism in cirrhosis: selective changes in cytochrome P-459 isozymes in the choline-deficient rat model. Biochem Pharmacol 1986; 35:iai7-24. Wood AWW, Ryan DE, Thomas PE, et al. Regio- and stereoselective metabolism of two C19 steroids by five highly purified and reconstituted rat hepatic cytochrome P-450 isozymes. J Biol Chem 1983;258:8839-47. Waxman DJ, Ko A, Walsh C. Regioselectivity and stereoselectivity of androgen hydroxylations catalyzed by cytochrome P-450 isozymes purified from phenobarbital-induced rat liver. J Biol Chem 1983;258:11937-47, Cheng K-C, Schenkman JB. Testosterone metabolism by cytochrome P-450 isozymes RLM 3 and RLM 5 and by microsomes. J Biol Chem 1983;258:11738-44. Gustafsson J-A, Ingelman-Sundberg M. Regulation and substrate specificity of a steroid sulfate-specific hydroxylase system in female rat liver microsomes. J Biol Chem 1975; 250:3451-a. Lowry OH, Rosebrough NJ, Farr AL, et al. Protein measure-
STEKOID HORMONE METABOLISM
July 1987
ment with the Folin phenol reagent. J Biol Chem 1951; 193265-75. 15. Omura T, Sato R. The carbon monoxide binding pigment of liver microsomes. I. Evidence for its hemoprotein nature. J Biol Chem 1964;239:2370-8. 16. Morgan ET, MacGeoch C, Gustafsson J-A. Hormonal and developmental regulation and expression of the hepatic microsomal 16Lu-hydroxylase cytochrome P-450 apoprotein in the rat. J Biol Chem 1985;260:11895-8. 17. Morgan ET, MacGeoch C, Gustafsson J-A. Sexual differentiation of cytochrome P-450 in rat liver: evidence for a constitutive isozyme as the male-specific 16a-hydroxylase. Mol Pharmacol
18. Waxman
1985;27:471-9. DJ, Dannan
hepatic
GA, Guengerich
cytochrome
monal
imprinting,
isozymes.
P-450:
and xenobiotic
Biochemistry
FP. Regulation
age-dependent inducibility
purified
of adult
20. Mode ation
rats. J Biol Chem
A, Gustafsson
between
P-450
plasma
of hepatic
J-A,
level
from hepatic
of three microsomes
1984;259:1239-50.
Jansson
of growth
steroid
metabolism
21. Van Thiel DH, Gavaler
JS, Schade
J-O,
hormone
et al.
Association
and sex differenti-
in the rat. Endocrinology
1982;111:1692-7, hypothalamic
pituitary
gonadal
RS. Liver axis.
Semin
disease Liver
and the Dis 1985;
5:35-45.
22. Cameron
DP, Burger
human
growth
failure,
and
HG, Catt KJ, et al. Metabolic
hormone
in the
in patients
isolated
perfused
with
clearance
hepatic
pig liver.
and
of
renal
Metabolism
1972;21:895-904.
23. Wu A, Grand
DB, Hambley
J, et al. Reduced
serum
somato-
in patients
with chronic
liver disease.
147
Clin Sci
24. Schimpff
RM, Lebrec D, Donnadieu M. Serum somatomedin activity measured as sulphation factor in peripheral, hepatic, and renal veins of patients with alcoholic cirrhosis. Acta Endocrinol 1978;88:729-36. FP, Dannan GA, Wright ST, et al. Purification and 25. Guengerich characterization of liver microsomal cytochromes P-450:electrophoretic. spectral, catalytic and immunochemical properties, and inducibility of eight isozymes isolated from rats treated with phenobarbital or p-naphthoflavone. Biochemistry 1982;21:6019-30. 26. Takeyama M, Nagareda T, Matsumoto K. 17/j-Oxidoreductase activity in kidneys of male and female mice of various ages. J Steroid Biochem 1985;22:617-21.
27. Head
1985;24:4409-17.
cytochromes
male
hor-
of sex-specific
19. Ryan DE, Iida S, Wood AWW, et al. Characterization highly
of rat
expression,
medin activity 1974;47:359-66.
IN CIRRHOTIC RATS
B, Moody DE, Woo CH, et al. Alterations of specific forms of cytochrome P-450 in rat liver during acute carbon tetrachloride intoxication. Toxic01 Appl Pharmacol 1981; 61:286-95. 28. Noguchi T, Fong K-L, Lai EK, et al. Selective early loss of polypeptides in liver microsomes of C&-treated rats: relationship to cytochrome P-450 content. Biochem Pharmacol 1982;31:609-14. 29. Noguchi T, Fong K-L, Lai EK, et al. Specificity of a phenobarbital-induced cytochrome P-450 for metabolism of carbon tetrachloride to the trichloromethyl radical. Biochem Pharmacol 1982;31:615-624, of 30. Murray M, Zaluzny L, Dannan GA, et al. Altered regulation cytochrome P-450 enzymes in choline-deficient cirrhotic male rat liver: impaired regulation and activity of the malespecific androst-4-ene-3,17-dione 16a-hydroxylase, cytochrome P-450 UT-A, in hepatic cirrhosis. Mol Pharmacol [in press).
Impaired Androgen 16wHydroxylation Hepatic Microsomes From Carbon Tetrachloride-Cirrhotic Male Rats MICHAEL
MURRAY,
LOUISE
ZALUZNY,
1987:93:141-7
in
and
GEOFFREY C. FARRELL Department
of Medicine,
University
of Sydney,
Hepatic cirrhosis produced by repeated inhalation of carbon tetrachloride is associated with reduced levels of microsomal cytochrome P450. In this study the C,,-steroids androstenedione and testosterone were used as specific probes of the functional activity of several forms of cytochrome PdSOin microsomal fractions from control and cirrhotic rat liver. The 16~ principal finding, that androstenedione hydroxylation and testosterone Za-, 16~, and 17~ hydroxylation were reduced to 14%--38% of control activity, strongly suggests that levels of the male sexually differentiated cytochrome PdsO (P450,,,] are decreased in hepatic cirrhosis. The activity of other cytochrome P,,,-mediated &-steroid hydroxylases, with the exception of androstenedione Go-hydroxylase, appeared essentially unaltered in microsomes from cirrhotic rats. Cirrhosis induced by carbon tetrachloride was also associated with greatly decreased activity of the microsomal cytochrome PdSOindependent 17P-oxidoreductase, an enzyme that catalyzes the conversion of androstenedione to testosterone. Consequently, and in view of the impaired activity of cytochrome P4,,,-mediated testosterone 17a-hydroxylation, the capacity of cirrhotic microsomes to catalyze the interconversion of androstenedione and testosterone was much lower than that of control microsomes. The present data confirm and extend earlier observations that selective impairment of drug oxidation pathways occurs Received August 5, 1986. Accepted November 10, 1986. Address requests for reprints to: Michael Murray, Ph.D., Department of Medicine, Westmead Hospital, Westmead NSW 2145, Australia. This work was supported by a grant from the Australian National Health and Medical Research Council. M. Murray is a National Health and Medical Research Council Research Fellow. The authors thank Diane West for secretarial assistance in the preparation of the manuscript and the Department of Anatomical Pathology at Westmead Hospital for histologic studies. 0 1987 by the American Gastroenterological Association 0016-5085/87/$3.50
Westmead
Hospital,
Westmead,
Australia
in hepatic cirrhosis. These changes are unrelated to the acute toxicity produced by carbon tetrachioride exposure. The available evidence supports the assertion that specific forms of cytochrome Pd5,, are subject to altered regulation in cirrhosis. Cytochrome PJSO (P4& is a family of isozymes, situated in the hepatic endoplasmic reticulum, that is active in the oxidation of a variety of endogenous and exogenous compounds, including drugs and steroid hormones. Oxidative metabolism of many drugs is impaired in patients with liver disease (l-3). In vitro studies of microsomal P450-mediated mixed-function oxidase activity (1)and in vivo studies of hepatic drug clearance (2) have led to the suggestion that selective changes in individual forms of PJSOmay occur in chronic liver disease. Studies of experimental cirrhosis produced in rats by the repeated inhalation of carbon tetrachloride (CC14) (4) have demonstrated that hepatic microsomes from cirrhotic rats have a decreased capacity to bind (5) and to catalyze the oxidative metabolism (6,7)of drug substrates. In this model hepatic heme turnover was normal (8), and the capacity for microsomal enzyme induction was retained (7). From these findings it was proposed that impaired drug oxidation may be a consequence of altered regulation of certain forms of PJ5,, (8). Recent studies, undertaken in the choline-deficient rat model of hepatic cirrhosis, suggested that the suppression of P 450 levels in chronic liver disease is selective for some forms of cytochrome (9). A large body of evidence now exists that individual PdsOs catalyze the position-specific hydroxylation of C19-steroids, including androst-4-ene-3,17dione (androstenedione) and 17P-hydroxyandrost4-ene-3-one (testosterone) (10-12). The present study used these steroid substrates as probes to Abbreviation
used
in this paper:
P,,,,
cytochrome
P,,,.
GASTROENTEROLOGYVol. 93, No. 1
142 MURRAY ET AL.
pathways in microsomal fracassess hydroxylation tions from Ccl,-cirrhotic and control rat liver. From these studies it is apparent that selective changes in individual forms of P4sa occur in the Ccl, rat model of chronic liver disease and that these changes are attributable to the pathophysiologic state of cirrhosis itself.
amide adenine dinucleotide phosphate, reduced form. Incubations were terminated by the addition of 5.5% zinc sulfate and centrifuged, and the supernatant was extracted with chloroform. The organic phase was evaporated to in a small volume of dryness under Nz, reconstituted chloroform, and applied to thin-layer chromatography plates (silica gel 60 F 254 type, 20 cm x 20 cm X 0.25 mm thickness and activated for 15 min at 100°C before use; E. Merck, Darmstadt, F.R.G.).
Materials and Methods Identification
Chemicals [4-r4C]Androstenedione (sp act 59 mCiimmo1; 98% purity by thin-layer chromatography) and [4-14C]testosterone (sp act 56.9 mCiimmo1; 97% purity by thin-layer chromatography) were obtained from Amersham Australia, Sydney, Australia. Unlabeled androstenedione, Sphydroxy- and 16a-hydroxyandrostenedione, testosterone, and 16a-hydroxytestosterone, as well as all biochemicals, were purchased from Sigma Chemical Co., St. Louis, MO. 7a-Hydroxyandrostenedione and 2a-hydroxytestosterone were obtained from Professor D. N. Kirk and the MRC Steroid Reference Collection, Queen Mary’s College, London, United Kingdom. 6P-Hydroxy- and 7a_hydroxytestosterone were purchased from Steraloids Inc., Wilton, N.H. 16P-Hydroxyandrostenedione was prepared enzymatitally by the action of 3P-hydroxysteroid dehydrogenase [Sigma) on 3P,16/3-dihydroxyandrost-5-ene-17-one (MRC Collection). Solvents and miscellaneous chemicals were from Ajax Chemicals, Sydney, Australia, and were at least analytical reagent grade. Animals Male Wistar rats (starting weight 150 g) were obtained from the animal house at the Westmead Hospital Institute of Clinical Pathology and Medical Research. Rats were given phenobarbital (1g/L) in their drinking water throughout the period of exposure to Ccl, (10 wk) (7). Controls were littermates that received phenobarbital as above. Experiments were performed 10 days after stopping phenobarbital and Ccl4 inhalation. Hepatic mixedfunction oxidase activity was stable at this time (7). Microsomal fractions were prepared as described elsewhere (9). The development of cirrhosis was assessed at the time of death from the macroscopic appearance of the liver. Liver function tests and histologic assessments were performed as described previously (7,8). Steroid
Hydroxylase
Activity
Microsomal androstenedione and testosterone hydroxylase activities were assayed essentially by the procedure of Gustafsson and Ingelman-Sundberg (13). Briefly, microsomal fractions from control or cirrhotic rat liver (0.75 mg protein/ml in 100 mM potassium phosphate buffer containing 1 mM ethylenediaminetetraacetic acid, pH 7.4) were incubated with 175 PM [‘4C]androstenedione or [‘4C]testosterone (4X105 dpm) for 10 min at 37°C. Reactions were initiated by the addition of 1 mM nicotin-
of Steroid
Metabolites
Androstenedione metabolites were resolved on thin-layer chromatography plates by development (twice) in the solvent system CHClJethyl acetate (1:2, vol/vol) as described by Waxman et al. (11).Testosterone metabolites were resolved similarly except that the first solvent system was dichloromethane/acetone (4:1, vol/vol) and the second system was chloroform/ethyl acetate/ethanol (4:1:0.7, vol/vollvol) (10). Zones corresponding to hydroxylated androstenedione and testosterone standards were visualized under ultraviolet light and scraped into vials for scintillation spectrometry (Aquasol scintillant, New England Nuclear Corp., Boston, Mass.). Other
Assays
Protein was determined by the method of Lowry et al. (14) with bovine serum albumin as standard, and P4s0 levels were estimated by the procedure of Omura and Sato (15). Analysis
of Data
In these experiments, mean values of data obtained from cirrhotic and control hepatic microsomes were compared using the unpaired Student’s t-test [two-tailed).
Results Microsomal in Control
P4s0 and Androgen Metabolism and Cirrhotic Rat Liver
Consistent with previous reports from this laboratory, levels of total microsomal P4sa were reduced in cirrhotic liver compared with controls (Table 1) (7,8). The metabolism of the androgens androstenedione and testosterone was studied in hepatic microsomal fractions from rats with CCll-induced cirrhosis, and in control animals. Formation of metabolites from both substrates was linear for at least 10 min (not shown). The capacity of cirrhotic rat liver microsomes to catalyze Clg-steroid metabolism was markedly lower than that of control fractions. In control microsomes 25.5% + 3.7% of the total substrate (androstenedione) was metabolized after 10 min, whereas only 12.0% +- 3.4% conversion occurred in cirrhotic microsomes (not shown). Very similar findings were noted with testosterone as substrate: 18.6% * 5.4% of the initial substrate was metabo-
STEROID HORMONE METABOLISM
July 1987
Table
1. Cytochrome
P4s0 and Androstenedione
Hydroxylation
in Hepatic
Microsomes
IN CIRRHOTIC RATS
From Cirrhotic
143
and Control
Rat9
Microsomes
Androstenedione hydroxylation (nmol product formedimin mg protein)
Cytochrome P450b (nmolimg protein)
Control Cirrhotic Percent of control
1.14 2 0.12
1.88 t 0.37
0.81t 0.08 71
1.19k 0.51 59 co.02
P”
W
7a
0.21t 0.06 0.27L 0.06 110 NS
16a
2.83+ 0.23 0.652 0.29 23
16P
0.43+ 0.14 0.30c 0.14 70 NS
NS, difference not significant. “Values are presented as mean + SD of six individual microsomal fractions. “Spectral determination cytochrome P4,0. cp values are for comparison of cirrhotic and control animal values.
lized in control microsomes after 10 min, and 9.1% ~fr3.4% in cirrhotic microsomal fractions under the same conditions (not shown]. It is therefore apparent that Clg-steroid metabolism is reduced to -50% of control levels in microsomes from cirrhotic rat liver. Non-P450 pathways were also examined in the present study and, from the data presented in Figure 1, it is clear that 17P-oxidoreductase activity (which converts androstenedione to testosterone) was significantly lower (p < 0.001) in cirrhotic rat liver. Thus, formation of testosterone was 0.45 nmol/ min . mg microsomal protein in control hepatic fractions and 40% of this rate (0.18 nmolimin . mg protein] in cirrhotic liver microsomes. Microsomal 5a-reductase activity was found to be negligible in both control and cirrhotic hepatic microsomes (data not shown]. Effect of Cirrhosis Hydroxylation
on Microsomal
Steroid
The point of major interest that emerged from this study is that androgen hydroxylation pathways
cmcontrol 0
Figure
1. 17pOxidoreductase
cirrhotic
activity in microsomes from control and cirrhotic rat liver. Significantly different from control. *p < 0.001.
of total
were affected differently in the CC14-rat model of cirrhosis (Tables 1 and 2). Thus microsomal androthe principal stenedione S/?- and 16a-hydroxylation, routes of hydroxylation in control microsomes, were decreased in cirrhosis to 58% and 23% of their respective control activities (Table 1). In contrast, the rates of androstenedione 7a- and lG@hydroxylation in microsomes from cirrhotic rat liver were not significantly different from control. The data in Table 1 indicate absolute rates of product formation in both types of microsomes, but it is important to recognize that P 450in cirrhotic microsomes was only 71% of control levels (Table 1). As this enzyme system has a fundamental role in steroid hydroxylation, the change in P 450levels could be an important factor in the observed rates of hydroxylation. Consequently, rates of production of individual metabolites were corrected for differences in microsomal PdsO and the resultant turnover numbers are presented in Figure 2. Again, 16a-hydroxylation was significantly decreased (p < 0.001) to about 30% of control; 16/S and 6P-hydroxylation of androstenedione were not different in control and cirrhotic hepatic microsomes. However, in cirrhotic microsomes, turnover of androstenedione by 7a-hydroxylation was increased (Figure 2: 0.34 k 0.08 nmol product/min . nmol PasO in cirrhosis versus 0.21 k 0.07 in control liver; p < 0.02). Testosterone hydroxylation activities were also measured in control and cirrhotic rat hepatic microsomes (Table 2). The findings were consistent with those obtained using androstenedione as the steroid substrate. Thus, testosterone 2a-, 16~, and 17~ hydroxylations (the latter pathway produces androstenedione) were all catalyzed with much lower efficiency in microsomes from cirrhotic liver whether expressed as absolute activity (14%-38% of control; Table 2) or as turnover numbers (19%--53% of control; Figure 2). In contrast, the activity of the testosterone 6P-hydroxylation pathway was unchanged from control levels, and the 7a-hydroxylase pathway appeared more active in cirrhotic microsomes (p < 0.002; Table 2).
GASTROENTEROLOGY Vol. 93, No. 1
MURRAY ET AL.
144
3.0
1
androstenedione
hydroxylation
B/D ratio was 0.64:1.00 in control microsomes but, in cirrhosis, the ratio was quite different (1.41:1.00). From this finding it is clear that the preferred site of hydroxylation is the D ring in control microsomes and the B ring in cirrhotic liver fractions. Testosterone hydroxylation occurred in the A ring (2a position), B ring (Sp and 7a), and D ring (16a and IV_. . . . . ..c.,,,\ 111 T, L"IICI"15 ___L._l" +L_ -..a:,$ L_.A.._....l, l1c.I p"alu"us,. LUG ldl.llJ "I Ilyulunyld-
control uul cirrhotic
0
tions
occurring
in the A, B, and D rings was but in cirrhosis these values were dramatically altered to 0.06:1.00:0.34. Thus, D-ring hydroxylation was predominant in control hepatic microsomes, whereas the B ring was the preferred site of hydroxylation in microsomes from cirrhotic rats. A ring hydroxylation was also reduced in cirrhotic microsomes as a consequence of impaired testosterone 2a-hydroxylase activity. 0.46:1.00:1.35,
St3
@I
7a
160
testosterone
hydroxylation
.
Discussion
2a
60
70
16a
1701
Figure 2. Turnover of androstenedione and testosterone by different pathways of hydroxylation in microsomes from control and cirrhotic rat liver. Significantly different from control. *p < 0.001. **p < 0.02.
The most important finding of the present study is that selective impairment of specific pathways of &-steroid hydroxylation occurs in hepatic microsomes from rats with CC14-induced cirrhosis. There is now a large body of evidence that the positional hydroxylation of testosterone and androstenedione is catalyzed by individual forms of PaSO fan_I 7) (L”--su,.
IPrn~imlo l”Yl”UU
d,,rlinc frnm thio lcahnrotnmr lnrl ilLUUlU.3 AI”lll LA1113 lU”“lUL”lJ 1-u tn I”
The regioselectivity of androstenedione and testosterone hydroxylation in contr61 and cirrhotic rat hepatic microsomes was estimated from the turnover data in Figure 2. In control and cirrhotic hepatic microsomes the hydroxylation of androstenedione occurred in the B ring (at the 6p and YCY positions) and the D ring (at the 16~1 and 16p positions). The
the suggestion that the different effects of cirrhosis on individual drug oxidation pathways may be explained by selective effects on specific PdSOisozymes (8). We recently reported a selective reduction of one P450 subpopulation in partially purified P450 fractions from choline-deficient rat hepatic microsomes (9); a similar observation was made using partially purified PdSOfractions from CC14-cirrhotic rat liver microsomes (unpublished data). The present data provide stronger evidence of a selective effect of CC14-induced cirrhosis on individual forms of microsomal Pg5,, and extend earlier studies by providing information on the form (or forms) most affected. The selective decrease in testosterone turnover by 2a-, 16~, and 17a_hydroxylation, as well as the
Table
From
Changes in the Regioselectivity of Androgen Hydroxylation Produced by Hepatic Cirrhosis
2. Testosterone
Hydroxylation
in Hepatic
Microsomes
Cirrhotic
and Control
Rat?
Testosterone hydroxylation (nmol product formed/min mg protein) Microsomes
17ab
2a
6P
7a
160
Control Cirrhotic Percent of control PC
0.82 t 0.07 0.31 * 0.04 38
0.79 f 0.20 0.11 f 0.05 14
1.58 k 0.35 1.57 2 0.22 99 NS
0.16 L 0.02 (1.31f 0.12 190 co.02
1.47 k 0.34 0.32 f 0.07 22
NS, difference not significant. “Values are presented as mean 2 SD of six individual microsomal fractions. b17a-Oxidation of androstenedione formation from testosterone. ‘P values are for comparison of cirrhotic and control animal values.
indicates
rate
July 1987
STEROID HORMONE METABOLISM IN CIRRHOTIC RATS
St3
7a
16a androstenedione Figure
145
7a
2a testosterone
16a
3. Effect of hepatic cirrhosis on the disposition of C,,-steroids. Increased turnover by a particular pathway is indicated by solid lines, decreased turnover by broken lines, and unchanged turnover by dotted lines. The microsomal Ii’/%oxidoreductase is a “P-450 0 refers to those P,,, isozymes involved in the non-P,,, enzyme that converts androstenedione to testosterone. 17~oxidation of testosterone to androstenedione.
reduced turnover of androstenedione by 16~ hydroxylation, suggests that levels of the malespecific P450,,, (16,17) [also termed P450,,T_A (18), and Pa5,,h (19)] are down-regulated in the CC& rat model of hepatic cirrhosis. This is a particularly interesting observation as it is now known that P450,6,is under hypothalamic-pituitary-gonadal control (15). Imprinting of the brain during the neonatal period permits the expression of P450,,, during adult life (16,17). Castration or hypophysectomy, or both, of male rats results in the disappearance of P450,,, apoprotein and its associated 16a-hydroxylase activity (16,18). This effect is now attributed to the abolition of a “male pattern” of growth hormone secretion by the pituitary (16,20). Thus, the selective decrease in the P450,80mediated pathway of steroid hydroxylation suggests that a similarity exists, in terms of microsomal oxidase activities, between hepatic cirrhosis produced by Ccl, and either hypophysectomy or castration of male rats. The function of the hypothalamic-pituitary-gonadal axis can be affected in a variety of ways in patients with chronic liver disease (21). It has also been reported that circulating levels of growth hormone (22) and somatomedins (23,24) are altered in individuals with hepatic cirrhosis. In view of the established role of growth hormone as a primary regulator
of the expression of specific forms of microsomal PJSOin rats, it now seems appropriate that the relationship between growth hormone and P450 isozyme composition should be studied in humans with cirrhosis. Other pathways of &,-steroid metabolism were less affected in hepatic cirrhosis. The 6@hydroxylation of androstenedione was reduced to 60% of control activity in cirrhotic microsomes; this decrease was comparable with the overall decrease in PJ5,, (Table 1). Changes in microsomal androstenedione 7a- and lG@hydroxylase activities were not significantly different from control, but testosterone 7a-hydroxylation was actually about twofold higher in cirrhotic microsomes. This apparent discrepancy between androstenedione and testosterone ?‘ahydroxylation activities in cirrhotic microsomes is probably due to the relative turnover capacity of form P450a(10) [also termed P450,,,~,(25)]. Wood et al. (10) have reported an approximate twofold greater turnover number for testosterone 7cr-hydroxylation than for androstenedione 7a-hydroxylation cataConsequently, it is likely that P,,,, is lyzed by PJSO,. able to compete more efficiently with other isozymes when testosterone is the substrate. These findings have implications for drug oxidation in cirrhotic liver. Male-specific PJSOs that are
146
GASTROENTEROLOGY Vol. 93, No. 1
MURRAY ET AL.
apparently identical with form P450,,, have high catalytic activity in the 4’- and 6-hydroxylation of Rand S-warfarin (E), and hexobarbital 3-hydroxylation (19). A number of additional xenobiotic oxidations are catalyzed efficiently by the male-specific P 4509 including aminopyrine N-demethylation (I 7, 25). Thus, it is not surprising that down-regulation of this enzyme results in impaired metabolic capacity in experimental liver disease. Our recent observations that ethylmorphine N-demethylase and benzo(a)pyrene hydroxylase activities are especially susceptible to the effects of CC14-induced cirrhosis (7) and long-term intake of a choline-deficient diet (9) are consistent with the proposed effects on P450,,,. Non-P,,,-mediated &,-steroid metabolism ip control and cirrhotic rat hepatic microsomes was also investigated. A finding of significance was reduced activity of microsomal 17p-oxidoreductase, an enzyme responsible for conversion of 17-keto steroids to 17/Lhydroxy steroids; e.g., androstenedione is converted to testosterone by this enzyme. This activity was decreased in cirrhotic liver to 40% of control (Figure 1). Interestingly, this enzyme also appears to be under hormonal control although its regulation has not been studied as intensively as that of the sex-differentiated forms of P450. For instance, Takeyama et al. (26) have reported that neonatal castration of mice, but not castration at 60 days of age, results in impaired 17fi-oxidoreductase activity in microsomes prepared at 120 days. One probable effect of diminished 17/3-oxidoreductase activity is that androstenedione would be relatively more available for P4,,-mediated hydroxylation as less of the substrate is converted to testosterone. As depicted in Figure 3, the interconversion between androstenedione and testosterone would be slower in cirrhotic microsomes. This is due not only to decreased l7poxidoreductase activity, but also to decreased P450mediated 17cr-hydroxylation of testosterone to androstenedione. In the present study we have demonstrated a selective reduction in the male-specific steroid 16~ hydroxylase in CC14-cirrhotic rat hepatic microsomes; other P,,,-mediated steroid hydroxylase pathways were not reduced. Thus it is possible to clearly distinguish the Ccl, rat model of hepatic cirrhosis from acute Ccl, hepatotoxicity. The present finding that androstenedione 16ghydroxylation (a pathway mediated by the phenobarbital PJ5&,) was unaffected in CC14-induced cirrhosis is in contrast to the evidence presented elsewhere that levels of P&ob are reduced after acute Ccl, intoxication of rats (27-29). We have observed that similar changes in steroid 16a-hydroxylase activity occur in both the CC14-inhalation and choline-deficient diet models of hepatic cirrhosis (30). Therefore, we be-
lieve that these selective changes in drug oxidation pathways are attributable to cirrhosis and not to the method of its production. Considered together, it is clear that chronic liver disease in experimental animals is associated with striking changes in several pathways of microsomal &-steroid metabolism. These substrates are excellent probes for the activity of several different P450s, and the present study established selective changes in the male-specific isozyme P4so,,,. Future studies will concentrate on the mechanism of this effect and the role of endocrine factors in hepatic cirrhosis in experimental animals.
References 1. Farrell GC, Cooksley WGE, Powell
LW. Drug metabolism in liver disease: activity of hepatic microsomal drug metabolising enzymes. Clin Pharmacol Ther 1979;26:483-92. 2. Klotz U, Fischer C, Muller-Seydlitz P, et al. Alterations in the disposition of differently cleared drugs in patients with cirrhosis. Clin Pharmacol Ther 1979;26:221-7. 3. Schenker S, Hoyumpa AM Jr, Wilkinson GR. The effect of parenchymal liver disease on the disposition and elimination of sedatives and analgesics. Med Clin North Am 1975;59: 887-96.
4. Marshall WJ, McLean AEM. The effect of cirrhosis of the liver on microsomal detoxications and cytochrome P-450. Br J Exp Path01 1969;50:578-83. 5. Murray M, Farrell GC. Different effects of carbon tetrachloride
toxicity and cirrhosis on substrate binding to rat hepatic microsomal cytochrome P-459. Biochem Pharmacol 1984; 33:687-9. 6. Villeneuve
10.
11.
12.
13.
14.
J-P, Wood AJJ, Shand DG, et al. Impaired drug metabolism in experimental cirrhosis in the rat. Biochem Pharmacol 1978;27:2577-81. Farrell GC, Zaluzny L. Microsomal protein synthesis and induction of cytochrome P-450 in cirrhotic rat liver. Aust J Exp Biol Med Sci 1984;62:291-301. Farrell GC, Zaluzny L. Hepatic heme metabolism and cytochrome P,,, in cirrhotic rat liver. Gastroenterology 1985; 89:172-g. Murray M, Zaluzny L, Farrell GC. Drug metabolism in cirrhosis: selective changes in cytochrome P-459 isozymes in the choline-deficient rat model. Biochem Pharmacol 1986; 35:iai7-24. Wood AWW, Ryan DE, Thomas PE, et al. Regio- and stereoselective metabolism of two C19 steroids by five highly purified and reconstituted rat hepatic cytochrome P-450 isozymes. J Biol Chem 1983;258:8839-47. Waxman DJ, Ko A, Walsh C. Regioselectivity and stereoselectivity of androgen hydroxylations catalyzed by cytochrome P-450 isozymes purified from phenobarbital-induced rat liver. J Biol Chem 1983;258:11937-47, Cheng K-C, Schenkman JB. Testosterone metabolism by cytochrome P-450 isozymes RLM 3 and RLM 5 and by microsomes. J Biol Chem 1983;258:11738-44. Gustafsson J-A, Ingelman-Sundberg M. Regulation and substrate specificity of a steroid sulfate-specific hydroxylase system in female rat liver microsomes. J Biol Chem 1975; 250:3451-a. Lowry OH, Rosebrough NJ, Farr AL, et al. Protein measure-
STEKOID HORMONE METABOLISM
July 1987
ment with the Folin phenol reagent. J Biol Chem 1951; 193265-75. 15. Omura T, Sato R. The carbon monoxide binding pigment of liver microsomes. I. Evidence for its hemoprotein nature. J Biol Chem 1964;239:2370-8. 16. Morgan ET, MacGeoch C, Gustafsson J-A. Hormonal and developmental regulation and expression of the hepatic microsomal 16Lu-hydroxylase cytochrome P-450 apoprotein in the rat. J Biol Chem 1985;260:11895-8. 17. Morgan ET, MacGeoch C, Gustafsson J-A. Sexual differentiation of cytochrome P-450 in rat liver: evidence for a constitutive isozyme as the male-specific 16a-hydroxylase. Mol Pharmacol
18. Waxman
1985;27:471-9. DJ, Dannan
hepatic
GA, Guengerich
cytochrome
monal
imprinting,
isozymes.
P-450:
and xenobiotic
Biochemistry
FP. Regulation
age-dependent inducibility
purified
of adult
20. Mode ation
rats. J Biol Chem
A, Gustafsson
between
P-450
plasma
of hepatic
J-A,
level
from hepatic
of three microsomes
1984;259:1239-50.
Jansson
of growth
steroid
metabolism
21. Van Thiel DH, Gavaler
JS, Schade
J-O,
hormone
et al.
Association
and sex differenti-
in the rat. Endocrinology
1982;111:1692-7, hypothalamic
pituitary
gonadal
RS. Liver axis.
Semin
disease Liver
and the Dis 1985;
5:35-45.
22. Cameron
DP, Burger
human
growth
failure,
and
HG, Catt KJ, et al. Metabolic
hormone
in the
in patients
isolated
perfused
with
clearance
hepatic
pig liver.
and
of
renal
Metabolism
1972;21:895-904.
23. Wu A, Grand
DB, Hambley
J, et al. Reduced
serum
somato-
in patients
with chronic
liver disease.
147
Clin Sci
24. Schimpff
RM, Lebrec D, Donnadieu M. Serum somatomedin activity measured as sulphation factor in peripheral, hepatic, and renal veins of patients with alcoholic cirrhosis. Acta Endocrinol 1978;88:729-36. FP, Dannan GA, Wright ST, et al. Purification and 25. Guengerich characterization of liver microsomal cytochromes P-450:electrophoretic. spectral, catalytic and immunochemical properties, and inducibility of eight isozymes isolated from rats treated with phenobarbital or p-naphthoflavone. Biochemistry 1982;21:6019-30. 26. Takeyama M, Nagareda T, Matsumoto K. 17/j-Oxidoreductase activity in kidneys of male and female mice of various ages. J Steroid Biochem 1985;22:617-21.
27. Head
1985;24:4409-17.
cytochromes
male
hor-
of sex-specific
19. Ryan DE, Iida S, Wood AWW, et al. Characterization highly
of rat
expression,
medin activity 1974;47:359-66.
IN CIRRHOTIC RATS
B, Moody DE, Woo CH, et al. Alterations of specific forms of cytochrome P-450 in rat liver during acute carbon tetrachloride intoxication. Toxic01 Appl Pharmacol 1981; 61:286-95. 28. Noguchi T, Fong K-L, Lai EK, et al. Selective early loss of polypeptides in liver microsomes of C&-treated rats: relationship to cytochrome P-450 content. Biochem Pharmacol 1982;31:609-14. 29. Noguchi T, Fong K-L, Lai EK, et al. Specificity of a phenobarbital-induced cytochrome P-450 for metabolism of carbon tetrachloride to the trichloromethyl radical. Biochem Pharmacol 1982;31:615-624, of 30. Murray M, Zaluzny L, Dannan GA, et al. Altered regulation cytochrome P-450 enzymes in choline-deficient cirrhotic male rat liver: impaired regulation and activity of the malespecific androst-4-ene-3,17-dione 16a-hydroxylase, cytochrome P-450 UT-A, in hepatic cirrhosis. Mol Pharmacol [in press).