- Email: [email protected]
Ethynylestradiol-17β D-ring glucuronide conjugates are potent cholestatic agents in the rat
Life Sciences, Vol. 32, pp. 2989-2993 Printed in the U.S.A.
Pergamon Press
ETHYNYLESTRADIOL-176 D-RING GLUCURONIDE CONJUGATES ARE POTENT CHOLESTATIC AGENTS IN THE RAT Mary Vore, Harry Hadd* and William Slikker Jr.% Department of Pharmacology, University of Kentucky College of Medicine Lexington, Ky. 40536, *Department of Biochemistry, Northwest Center for Medical Education, Indiana University School of Medicine, Gary, In. 46408 and %Division of Teratogenesis Research, National Center for Toxicol-oical Research, Jefferson, Ar. 72079. (Received in final form April 15, 1983) Summary 17e-Ethynylestradiol-178(6-D-glucuronide) [EE2178(SG)], a metabolite of 17e-ethynylestradiol (EE2) identified in urine of women taking EE 2 in oral contraceptives, and its synthetic anomer, 17e-ethynylestradlol-176(~-D-glucuronide), [EE2178(~G)], were administered intravenously to female rats in order to determine their effects on bile flow. Both agents induced an immediate, profound and dose-dependent decrease in bile flow which returned to control levels within 1-8 hr. The logarithm of the dose vs the cholestatlc response curves for the two anomers were not parallel. EE2178(eG) was significantly more potent than EE2176(6G # such that the doses inhibiting bile flow by 50% were 1.25 and ii ~ o l / k g for the ~and 6-anomer respectively. 17e-Ethynylestradiol (EE2) is the major estrogenic component of oral contraceptives, drugs which are used by tens of millions of women world-wide. Oral contraceptives, and specifically, the estrogenic component, have been shown to decrease hepatic excretory function in nearly all women as indicated by retention of bromosulfophthalein (BSP), a dye excreted essentially exclusively in the bile (1-5). In some women who are particularly sensitive to estrogens, EE2 at doses of 0.16-1.0 mg, or mestranol, (the 3-methyl ether of EE2) at comparable doses, induces intrahepatic cholestasis typified by Jaundice and pruritis attributable to decreased biliary excretion of bilirubin and bile acids respectively (3-5). These effects are seen within a few days to weeks after administration of EE2 and revert to normal shortly after discontinuation of therapy. Treatment of rats chronically with EE 2 at doses of 0.5-1.0 mg/day for 5-9 days has also been shown to decrease bile flow, biliary excretion of BSP and synthesis of bile acids (6-9). In the present studies, the effect of acute administration of EE2176(6G), an endogenous metabollte of EE2, and its synthetic anomer, EE2178(~G) on bile flow in the rat was determined. Methods Chemicals: The e and 8 anomers of the 176-D-glucuronide conjugate of EE 2 were synthesized by the SnCI4 promoted reaction as described by Nadd et al (i0). The conjugates were resolved by crystallization and high pressure liquid chromatography. Positive identification was established by IR, UV, mass spectroscopy and IH- and 13C-nmr; the structure of the 6-anomer was confirmed by X-ray crystallographic analysis. The ~-anomer was refractory to hydrolysis by bovine liver 6-glucuronidase whereas EE2178(SG) was readily hydrolyzed. 0024-3205/83 $3.00 + .00 Copyright (c) 1983 Pergamon Press Ltd.
2990
EE2-glucuronide-induced
Cholestasis
Vol.
32, No. 26, 1983
Animals: Female Sprague-Dawley rats (200-230g) were anesthetized with urethane (i g/kg, ip) and the femoral vein and bile duct cannulated (PE-50 and PE-IO polyethylene tubing respectively). Body temperature was recorded via a rectal probe inserted 7 cm and maintained at 37°C with a heating pad connected to a temperature regulator. Bile was collected in tared vials and the volume determined gravimetrically assuming a density of i. Bile was collected for 30 min before administration of the glucuronides to determine initial bile flow. Glucuronides were administered in saline:propylene glycol:ethanol (10:4:1, v/v) (2 ml/kg). EE2 (33 ~mol/kg) was administered in propylene glycol (I ml/kg, iv). We have previously demonstrated that administration of vehicle alone has no effect on bile flow (ii). Data analysis. The log dose response curves were generated by linear regression. A test for parallelism showed no significant interaction (p>O.05) between the log dose response curve for estradiol-17B(B-D-glucuronide) (E217G) and EE217B(~G). A four-point parallel line bioassay was thus justified for determining the potency of EE217B(BG) relative to E217G. Four to six animals at each of two doses were used. The test for parallelism showed significant interaction between the log dose response curves for E217G and EE217~(~G).
A. E F_..zl7/9(.8G)
B. E E217,8(,,G)
120-
I00
O ~D °
8O
~
a3 o
60
cO
4O
o~ 20
o
3'0
40
9'o ,~o ~o ,6o
3%
4o
9'o ,~'o ,~o
,~o
1i m e (minutes after dose) Fig. 1 Bile flow, expressed as a percentage of initial flow, after an iv dose of EE217B(BG) ( 0 , ll~mol/kg;, 0, 16.5~mol/kg;., 22~mol/kg; [] , 33Nmol/kg) or EE2178(~G) ( 0 , 0.85~mol/kg; 0, 1.25~mol/kg; , , 1.7~mol/kg; [7 , 2.5~mol/kg). The points represent the mean + S.E. of values obtained in four-six rats.
Vol. 32, No. 26, 1983
EE2-glucuronide-induced Cholestasis
2991
Results Administration of both EE217~(~G ) and EE217B(~G) induced an i ~ e d i a t e decrease in bile flow which was dose dependent (Fig. i). Inhibition of bile flow was readily and completely reversible following EE217B(BG) (Fig. IA); complete recovery after the higher doses of EE217B(~G) (Fig. IB) required seven to eight hours (data not shown). In contast, intravenous administration of the parent EE2 (33~mol/kg) significantly increased bile flow by 20-30% from 30-180 min after administration (data not shown). These data indicate that hydrolysis of the glucuronide conjugates to yield the parent EE 2 is not responsible for their cholestatic activity. When the maximal percent inhibition of bile flow was plotted against the logarithm of the dose, linear dose response curves were obtained (Fig. 2). The doses inhibiting bile flow by 50% were 1.25 and ll~mol/kg for EE217B(~G) and EE217~(BG) respectively. The log dose response curves for these two agents were not parallel. The log dose response curve for EE217B(BG) was parallel, however to those of four other steroid D-rlng glucuronide conjugates previously identified as cholestatlc (12). Calculation of the potency ratios and 95% confidence limits by means of a four point bioassay demonstrated that EE217B(~G ) is equlpotent to E217G and estriol-17B(B-D-glucuronide) (E317G) but significantly more potent than testosterone-17B(B-D-glucuronlde) (TG) or estriol-16~(~-D-glucuronide) (E316G). (Table 1).
100-
t, 80Z
0_1~6) I
//"
,g
N 2o-
/
I/
/
/
ISEE217,8(~G)
o,3 G
D 0 S E (/zmol/kg) FI 8. 2 The relationship between the logarithm of the dose of selected steroid D-ring glucuronldes and the maximal inhibition of bile flow. Data for E217G, E317G , TG and E316G are taken from Meyers et al (12). Discussion The identification of these two cholestatic agents, EE2!7B(aG) and EE217B(~G), confirms and extends our hypothesis that steroid D-ring glucuronide conJugates represent a new class of potent cholestatic agents. Efforts to synthesize the B-D-glucuronide at the phenolic 3 position of EE2 in sufficient quan-
2992
EE2-glucuronide-induced
Cholestasis
Vol. 32, No. 26, 1983
Table 1 Relative Cholestatic Potency and 95% Confidence Limits of Steroid D-Ring Glucuronides Compound
Relative potency 95% Confidence Limits
E217G
EE217B(BG)
1.0
0.75
-
0.51-1.09
E317G 0.68 0.58-0.80
Data for E317G, TG and E316G are taken from Meyers et al.
TG 0.40 0.32-0.50
,E316G 0.29 0.24-0.35
(12).
tities for testing have not been successful (13); however the glucuronide conJugates of the phenolic 3-OH group of estradiol and estriol increase bile flow following intravenous administration (11,12). Thus, the localization of the glucuronic acid group to the D-ring of the steroid nucleus is essential for cholestatic activity. The mechanism by which these agents inhibit bile flow is not known; however inhibition of bile acid secretion follows a similar time course as inhibition of bile flow. The structural similarity between the bile acids and steroid D-ring glucuronides in the orientation of the carboxylic acid group of the bile acids or glucuronic acid to ring D of the steroid nucleus, suggests that these agents may act to inhibit the transport of bile acids from hepatocyte to bile and thus prevent the generation of bile acid-dependent bile flow. The greater potency of the ~-anomer is presumably due to its steric configuration which increases its intrinsic cholestatic activity or which decreases its inactivation (eg. hydrolysis by B-glucuronidase) or elimination (eg. biliary excretion). EE217B(~G ) has not been identified as an endogenous metabolite of EE2; indeed e-glucuronides are not known to be formed by mammalian UDP-glucuronyltransferases. EE2178(BG) however, has been identified in the urine of women taking oral contraceptives (14, 15) and in urine of rhesus monkeys administered EE2 (16). The present data suggest the strong possibility that formation of the cholestatic EE217B(BG) by hepatic or intestinal enzymes in vivo is responsible for the decreased hepatic excretory function associated with use of the oral contraceptives. The doses of EE217B(BG ) (11-33 ~mol/kg; 2.2-6.6 ~mol, total dose) which inhited bile flow acutely in the rat in the present studies are considerably higher on a per weight basis than doses of EE2 which induce cholestasis clinically (0.5-3 ~moi/50-60 kg). The relative sensitivity of rats vs humans to the hepatotoxic effects of this metabolite is not known however, nor are the consequences of its chronic exposure, as would occur in women taking oral contraceptives. References I. 2. 3. 4. 5. 6. 7. 8. 9.
G.J. KLEINER, L. KRESGH, I.M. ARIAS, N. Engl. J. Med. 273 420 (1965). M.N. MUELLER, A. KAPPAS, J. Clln. Invest. 43 1905 (1964). M.J. KREEK, M.H. SLEISENGER, G.H. JEFFRIES, Am. J. Med. 43 795 (1967). M.J. KREEK, E. WESER, M.H. SLEISENGER, G.H. JEFFRIES, New Engl. J. Med. 277 1391 (1967). F. KERN, W. ERFLING, F.R. SIMON, R. DAHL, A. MALLORY, T.E. STARZL, Gastroenterology 75 512 (1978). T.F. GALLAGHER, M.N. MUELLER, A. KAPPAS, Medicine 45 471 (1966). J.J. GUMUCIO, V.D. VALDIVIESO, Gastroenterology 6 1 3 3 9 (1971). M.J. KREEK, R.E. PETERSON, M.H. SLEISENGER, G.H. JEFFRIES, Proc. Soc. Exp. Biol. Med. 131 646 (1969). M. HARKAVY, N.B. JAVITT, Metabolic Effects of Gonadal Hormones and Contra-
Vol. 32, No. 26, 1983
EE2-glucuronide-lnduced Cholestasls
2993
eeptlve Steroids, (P. ii) H.A. Salhanick, D.M. Kipnes, R.L. Vande Wiele, Eds., Plenum Publishing Co., New York (1969). 10.H.E. HADD, W. SLIKKER, JR., D.W. MILLER, E.D. HELTON, W.L. DUAX, P.D. STRONG, D.C. SWENSON, J. STEROID BIOCHEM. 18 81 (1983). II.M. MEYERS, W. SLIKKER, G. PASCOE, M. VORE, J. Pharmacol. Exp. Ther. 214 87 (1980). 12. M. MEYERS, W. SLIKKER, M. VORE, J. Pharmacol. Exp. Ther. 218 63 (1981). 13.H.E. HADD, W. SLIKKER, E.D. HELTON, J. Steroid Biochem. 1 3 1 1 0 7 (1980). 14.M.C. WILLIAMS, J.W. GOLDZIEHER, Steroids 36 255 (1980). 15.SAHLBERG, B.-L., M. AXELSON, D.J. COLLINS,---J. SJOVALL, J. of Chromatography 217 453 (1981). 16.L.A. RAITANO, W. SLIKKER, D.E. HILL, H.E. HADD, T. CAIRNS, E.D. HELTON, Drug Metab. Dispos. 9 129 (1981).
Pergamon Press
ETHYNYLESTRADIOL-176 D-RING GLUCURONIDE CONJUGATES ARE POTENT CHOLESTATIC AGENTS IN THE RAT Mary Vore, Harry Hadd* and William Slikker Jr.% Department of Pharmacology, University of Kentucky College of Medicine Lexington, Ky. 40536, *Department of Biochemistry, Northwest Center for Medical Education, Indiana University School of Medicine, Gary, In. 46408 and %Division of Teratogenesis Research, National Center for Toxicol-oical Research, Jefferson, Ar. 72079. (Received in final form April 15, 1983) Summary 17e-Ethynylestradiol-178(6-D-glucuronide) [EE2178(SG)], a metabolite of 17e-ethynylestradiol (EE2) identified in urine of women taking EE 2 in oral contraceptives, and its synthetic anomer, 17e-ethynylestradlol-176(~-D-glucuronide), [EE2178(~G)], were administered intravenously to female rats in order to determine their effects on bile flow. Both agents induced an immediate, profound and dose-dependent decrease in bile flow which returned to control levels within 1-8 hr. The logarithm of the dose vs the cholestatlc response curves for the two anomers were not parallel. EE2178(eG) was significantly more potent than EE2176(6G # such that the doses inhibiting bile flow by 50% were 1.25 and ii ~ o l / k g for the ~and 6-anomer respectively. 17e-Ethynylestradiol (EE2) is the major estrogenic component of oral contraceptives, drugs which are used by tens of millions of women world-wide. Oral contraceptives, and specifically, the estrogenic component, have been shown to decrease hepatic excretory function in nearly all women as indicated by retention of bromosulfophthalein (BSP), a dye excreted essentially exclusively in the bile (1-5). In some women who are particularly sensitive to estrogens, EE2 at doses of 0.16-1.0 mg, or mestranol, (the 3-methyl ether of EE2) at comparable doses, induces intrahepatic cholestasis typified by Jaundice and pruritis attributable to decreased biliary excretion of bilirubin and bile acids respectively (3-5). These effects are seen within a few days to weeks after administration of EE2 and revert to normal shortly after discontinuation of therapy. Treatment of rats chronically with EE 2 at doses of 0.5-1.0 mg/day for 5-9 days has also been shown to decrease bile flow, biliary excretion of BSP and synthesis of bile acids (6-9). In the present studies, the effect of acute administration of EE2176(6G), an endogenous metabollte of EE2, and its synthetic anomer, EE2178(~G) on bile flow in the rat was determined. Methods Chemicals: The e and 8 anomers of the 176-D-glucuronide conjugate of EE 2 were synthesized by the SnCI4 promoted reaction as described by Nadd et al (i0). The conjugates were resolved by crystallization and high pressure liquid chromatography. Positive identification was established by IR, UV, mass spectroscopy and IH- and 13C-nmr; the structure of the 6-anomer was confirmed by X-ray crystallographic analysis. The ~-anomer was refractory to hydrolysis by bovine liver 6-glucuronidase whereas EE2178(SG) was readily hydrolyzed. 0024-3205/83 $3.00 + .00 Copyright (c) 1983 Pergamon Press Ltd.
2990
EE2-glucuronide-induced
Cholestasis
Vol.
32, No. 26, 1983
Animals: Female Sprague-Dawley rats (200-230g) were anesthetized with urethane (i g/kg, ip) and the femoral vein and bile duct cannulated (PE-50 and PE-IO polyethylene tubing respectively). Body temperature was recorded via a rectal probe inserted 7 cm and maintained at 37°C with a heating pad connected to a temperature regulator. Bile was collected in tared vials and the volume determined gravimetrically assuming a density of i. Bile was collected for 30 min before administration of the glucuronides to determine initial bile flow. Glucuronides were administered in saline:propylene glycol:ethanol (10:4:1, v/v) (2 ml/kg). EE2 (33 ~mol/kg) was administered in propylene glycol (I ml/kg, iv). We have previously demonstrated that administration of vehicle alone has no effect on bile flow (ii). Data analysis. The log dose response curves were generated by linear regression. A test for parallelism showed no significant interaction (p>O.05) between the log dose response curve for estradiol-17B(B-D-glucuronide) (E217G) and EE217B(~G). A four-point parallel line bioassay was thus justified for determining the potency of EE217B(BG) relative to E217G. Four to six animals at each of two doses were used. The test for parallelism showed significant interaction between the log dose response curves for E217G and EE217~(~G).
A. E F_..zl7/9(.8G)
B. E E217,8(,,G)
120-
I00
O ~D °
8O
~
a3 o
60
cO
4O
o~ 20
o
3'0
40
9'o ,~o ~o ,6o
3%
4o
9'o ,~'o ,~o
,~o
1i m e (minutes after dose) Fig. 1 Bile flow, expressed as a percentage of initial flow, after an iv dose of EE217B(BG) ( 0 , ll~mol/kg;, 0, 16.5~mol/kg;., 22~mol/kg; [] , 33Nmol/kg) or EE2178(~G) ( 0 , 0.85~mol/kg; 0, 1.25~mol/kg; , , 1.7~mol/kg; [7 , 2.5~mol/kg). The points represent the mean + S.E. of values obtained in four-six rats.
Vol. 32, No. 26, 1983
EE2-glucuronide-induced Cholestasis
2991
Results Administration of both EE217~(~G ) and EE217B(~G) induced an i ~ e d i a t e decrease in bile flow which was dose dependent (Fig. i). Inhibition of bile flow was readily and completely reversible following EE217B(BG) (Fig. IA); complete recovery after the higher doses of EE217B(~G) (Fig. IB) required seven to eight hours (data not shown). In contast, intravenous administration of the parent EE2 (33~mol/kg) significantly increased bile flow by 20-30% from 30-180 min after administration (data not shown). These data indicate that hydrolysis of the glucuronide conjugates to yield the parent EE 2 is not responsible for their cholestatic activity. When the maximal percent inhibition of bile flow was plotted against the logarithm of the dose, linear dose response curves were obtained (Fig. 2). The doses inhibiting bile flow by 50% were 1.25 and ll~mol/kg for EE217B(~G) and EE217~(BG) respectively. The log dose response curves for these two agents were not parallel. The log dose response curve for EE217B(BG) was parallel, however to those of four other steroid D-rlng glucuronide conjugates previously identified as cholestatlc (12). Calculation of the potency ratios and 95% confidence limits by means of a four point bioassay demonstrated that EE217B(~G ) is equlpotent to E217G and estriol-17B(B-D-glucuronide) (E317G) but significantly more potent than testosterone-17B(B-D-glucuronlde) (TG) or estriol-16~(~-D-glucuronide) (E316G). (Table 1).
100-
t, 80Z
0_1~6) I
//"
,g
N 2o-
/
I/
/
/
ISEE217,8(~G)
o,3 G
D 0 S E (/zmol/kg) FI 8. 2 The relationship between the logarithm of the dose of selected steroid D-ring glucuronldes and the maximal inhibition of bile flow. Data for E217G, E317G , TG and E316G are taken from Meyers et al (12). Discussion The identification of these two cholestatic agents, EE2!7B(aG) and EE217B(~G), confirms and extends our hypothesis that steroid D-ring glucuronide conJugates represent a new class of potent cholestatic agents. Efforts to synthesize the B-D-glucuronide at the phenolic 3 position of EE2 in sufficient quan-
2992
EE2-glucuronide-induced
Cholestasis
Vol. 32, No. 26, 1983
Table 1 Relative Cholestatic Potency and 95% Confidence Limits of Steroid D-Ring Glucuronides Compound
Relative potency 95% Confidence Limits
E217G
EE217B(BG)
1.0
0.75
-
0.51-1.09
E317G 0.68 0.58-0.80
Data for E317G, TG and E316G are taken from Meyers et al.
TG 0.40 0.32-0.50
,E316G 0.29 0.24-0.35
(12).
tities for testing have not been successful (13); however the glucuronide conJugates of the phenolic 3-OH group of estradiol and estriol increase bile flow following intravenous administration (11,12). Thus, the localization of the glucuronic acid group to the D-ring of the steroid nucleus is essential for cholestatic activity. The mechanism by which these agents inhibit bile flow is not known; however inhibition of bile acid secretion follows a similar time course as inhibition of bile flow. The structural similarity between the bile acids and steroid D-ring glucuronides in the orientation of the carboxylic acid group of the bile acids or glucuronic acid to ring D of the steroid nucleus, suggests that these agents may act to inhibit the transport of bile acids from hepatocyte to bile and thus prevent the generation of bile acid-dependent bile flow. The greater potency of the ~-anomer is presumably due to its steric configuration which increases its intrinsic cholestatic activity or which decreases its inactivation (eg. hydrolysis by B-glucuronidase) or elimination (eg. biliary excretion). EE217B(~G ) has not been identified as an endogenous metabolite of EE2; indeed e-glucuronides are not known to be formed by mammalian UDP-glucuronyltransferases. EE2178(BG) however, has been identified in the urine of women taking oral contraceptives (14, 15) and in urine of rhesus monkeys administered EE2 (16). The present data suggest the strong possibility that formation of the cholestatic EE217B(BG) by hepatic or intestinal enzymes in vivo is responsible for the decreased hepatic excretory function associated with use of the oral contraceptives. The doses of EE217B(BG ) (11-33 ~mol/kg; 2.2-6.6 ~mol, total dose) which inhited bile flow acutely in the rat in the present studies are considerably higher on a per weight basis than doses of EE2 which induce cholestasis clinically (0.5-3 ~moi/50-60 kg). The relative sensitivity of rats vs humans to the hepatotoxic effects of this metabolite is not known however, nor are the consequences of its chronic exposure, as would occur in women taking oral contraceptives. References I. 2. 3. 4. 5. 6. 7. 8. 9.
G.J. KLEINER, L. KRESGH, I.M. ARIAS, N. Engl. J. Med. 273 420 (1965). M.N. MUELLER, A. KAPPAS, J. Clln. Invest. 43 1905 (1964). M.J. KREEK, M.H. SLEISENGER, G.H. JEFFRIES, Am. J. Med. 43 795 (1967). M.J. KREEK, E. WESER, M.H. SLEISENGER, G.H. JEFFRIES, New Engl. J. Med. 277 1391 (1967). F. KERN, W. ERFLING, F.R. SIMON, R. DAHL, A. MALLORY, T.E. STARZL, Gastroenterology 75 512 (1978). T.F. GALLAGHER, M.N. MUELLER, A. KAPPAS, Medicine 45 471 (1966). J.J. GUMUCIO, V.D. VALDIVIESO, Gastroenterology 6 1 3 3 9 (1971). M.J. KREEK, R.E. PETERSON, M.H. SLEISENGER, G.H. JEFFRIES, Proc. Soc. Exp. Biol. Med. 131 646 (1969). M. HARKAVY, N.B. JAVITT, Metabolic Effects of Gonadal Hormones and Contra-
Vol. 32, No. 26, 1983
EE2-glucuronide-lnduced Cholestasls
2993
eeptlve Steroids, (P. ii) H.A. Salhanick, D.M. Kipnes, R.L. Vande Wiele, Eds., Plenum Publishing Co., New York (1969). 10.H.E. HADD, W. SLIKKER, JR., D.W. MILLER, E.D. HELTON, W.L. DUAX, P.D. STRONG, D.C. SWENSON, J. STEROID BIOCHEM. 18 81 (1983). II.M. MEYERS, W. SLIKKER, G. PASCOE, M. VORE, J. Pharmacol. Exp. Ther. 214 87 (1980). 12. M. MEYERS, W. SLIKKER, M. VORE, J. Pharmacol. Exp. Ther. 218 63 (1981). 13.H.E. HADD, W. SLIKKER, E.D. HELTON, J. Steroid Biochem. 1 3 1 1 0 7 (1980). 14.M.C. WILLIAMS, J.W. GOLDZIEHER, Steroids 36 255 (1980). 15.SAHLBERG, B.-L., M. AXELSON, D.J. COLLINS,---J. SJOVALL, J. of Chromatography 217 453 (1981). 16.L.A. RAITANO, W. SLIKKER, D.E. HILL, H.E. HADD, T. CAIRNS, E.D. HELTON, Drug Metab. Dispos. 9 129 (1981).