Inhibition of bile acid conjugation by cyclosporin A

Biochi~ic~a A~ta et Biophysica

ELSEVIER

Biochimica et Biophysica Acta 1272 (1995) 49-52

Inhibition of bile acid conjugation by cyclosporin A Donald A. Vessey a,b,*, Michael Kelley a Liver Study Unit, Department of Veterans' Affairs Medical Center, San Francisco, CA 94121, USA b Department of Medicine, Universi~ of California, San Francisco, CA 94143, USA Received 14 November 1994; accepted 6 March 1995

Abstract Each of the two steps involved in bile acid conjugation was tested in vitro for its sensitivity to inhibition by cyclosporin A (CsA). Bile acid-CoA: glycine/taurine N-acyltransferase, the enzyme which catalyzes the second step, was tested and found to be insensitive to inhibition by 20/xM CsA. Bile acid:CoA ligase, the enzyme which catalyzes the first step, was found to be inhibited by 25% at 10/xM CsA in the standard assay. The inhibition was competitive vs. bile acid and noncompetitive vs. ATP, and uncompetitive vs. CoA. CsA was also found to interfere with the divalent cation requirement of the enzyme at low concentrations of Mg 2+ the maximum inhibition was 70%. The maximum inhibition obtainable at physiologic Mg 2÷ concentration was 40%. The extent of inhibition was presumably limited by the insolubility of CsA. At concentrations of CsA reached in vivo during drug therapy, CsA can be expected to significantly inhibit bile acid conjugation. Keywords: Bile acid; Bile acid:CoA ligase; Bile secretion; Conjugation; Cyclosporin A

I. Introduction Cyclosporin A (CsA) is a fungus-derived immunosuppressive agent which is pivotal for successful organ transplantation and its' finding increased usage in the treatment of some autoimmune diseases and dermatological disorders [1,2]. The effectiveness of CsA improves with dose, but unfortunately CsA treatment can be associated with a number of side effects. One of these is hepatic dysfunction. CsA causes cholestasis in many patients [3], and has been found to consistently cause an accumulation of bilirubin and bile acids in blood [4]. Studies in rats in vivo and in human and rat hepatocytes in vitro have related this hepatic dysfunction to a CsA-induced decrease in the hepatic uptake of bile acids [5-8]. The focus of the research in our laboratory has been on the conjugation of bile acids with glycine or taurine, and we have previously found that the hepatic uptake and biliary excretion of bile acids is dependent upon efficient conjugation [9]. It thus seemed reasonable to hypothesize that CsA decreases the hepatic uptake of bile acids by inhibiting bile acid conjuga-

Abbreviations: CsA, cyclosporin A. * Corresponding author. Fax: + 1 (415) 750 6906. 0925-4439/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0 9 2 5 - 4 4 3 9 ( 9 5 ) 0 0 0 6 6 - 6

tion. To test this hypothesis we studied the effect of CsA on bile acid conjugation. Bile acid conjugation occurs in two steps, there is an initial formation of bile acid-CoA catalyzed by bile acid:CoA ligase (E.C. 6.2.1.17) followed by reaction of this intermediate with either glycine or taurine, catalyzed by bile acid-CoA:glycine/taurine Nacyltransferase (E.C. 2.3.1.65). In this work, each of these steps was examined in vitro for its sensitivity to inhibition by CsA.

2. Materials and methods [2,4-3H]Cholic acid was obtained from NEN Research Products, Wilmington, DE. Cyclosporin A was kindly provided by Sandoz Pharmaceuticals, Basel, Switzerland. Cholyl-CoA was prepared as described previously [10]. Coenzyme A (CoA), adenosine triphosphate (ATP), 5,5'dithiobis(2-nitrobenzoic) acid, ammonium sulfate, Hepes, Trizma Pre-set pH crystals, leupeptin, chymostatin, and pepstatin A were obtained from the Sigma Chemical, St. Louis, MO. Guinea pig liver microsomes were prepared by homogenizing freshly excised guinea pig liver in 4 vol. of cold 0.25 M sucrose, 5 mM EDTA and 5 mM EGTA, contain-

50

D.A. Vessey, M. Kelley/Biochimica et Biophysica Acta 1272 (1995) 49-52

ing 250 / z g / l each of the protease inhibitors leupeptin, chymostatin, and pepstatin A. Microsomes were isolated as described previously [11]. The final microsomal pellet was resuspended in cold 0.25 M sucrose to approx. 30 mg protein/ml and stored in aliquots at - 80 ° C. A partially purified preparation of bovine liver N-acyltransferase was prepared from liver obtained fresh from a local slaughterhouse. The liver was homogenized in 0.13 M KC1, 20 mM Hepes (pH 7.4), 1 mM EDTA, 1 mM EGTA, and 0.25 mg/1 of each of the protease inhibitors leupeptin, chymostatin, and pepstatin A at 5° C. The homogenate was centrifuged for 20 min at I0 K rpm, and the supernatant centrifuged an additional 45 rain at 38 K rpm. The resulting supernatant was filtered through cheesecloth, and a 40 to 55% ammonium sulfate fraction isolated. The final pellet was dissolved in a minimal volume of 25 mM potassium phosphate (pH 8.0) and dialyzed against 50 vol. of this buffer and stored at - 80 ° C. The reaction of [3H]cholate with coenzyme A as catalyzed by guinea pig liver microsomes was determined radiochemically as described previously [12]. Assays were conducted at 30 ° C and the standard reaction contained 100 mM Tris (pH 7.4), 1 mM MnC12, 10 /zM CoA, 7.0 /zM [3H]Cholate, 1 mM ATP, and 0 or 25 /xM CsA. The reaction of cholyl-CoA with glycine as catalyzed by bovine liver N-acyltransferase was determined spectrophotometricaily using the 5,5'-dithiobis(2-nitrobenzoic) acid reaction [13]. The standard reaction contained 0.1 mM 5,5'-dithiobis(2-nitrobenzoic) acid, 100 mM potassium phosphate (pH 8.0), 20 /~M choloyl-CoA, and 75 /zg of partially purified bovine liver N-acyitransferase protein in a 1 ml volume. The reaction was initiated by the addition of glycine and the absorbance monitored at 412 nm at 30 ° C.

'1, ~1

l

i

I

I

I

!.o

I

l

•

•

0.8

0.6

0.4

0.2

(],0

o.o

I

I

I

I

I

I

I

0.2

0.4

o.s

o.s

1.o

1.2

1.4

I/[c~,l~]

~u)

1.s

Fig. 1. Double reciprocal plot of the cholic acid dependence of the rate of the ligase reaction in the presence or absence of cyclosporin A. Ligase activity was determined at variable concentrations of [3H]cholate, in either the presence (O), or absence (C)) of 25/~M cyclosporin A (CsA).

inhibition. We first tested the inhibition as a function of the concentration of cholate (Fig. l). Inhibition by 25 /zM CsA was eliminated at extrapolated infinite concentration of cholate indicating that CsA is a competitive inhibitor of cholate binding. When the concentration of cholate was saturating, inhibition was still obtained if the concentration Mn 2+ or Mg 2+ was lowered (60% inhibition at 3 0 / x M Mg 2+). The double reciprocal plots of 1 / v vs. 1 / M g 2+ were non-linear, but extrapolated to a common intercept suggesting a competitive type inhibition. The ligase is known to have a complicated divalent cation requirement [14], and thus the Mg2+-dependence of CsA inhibition was not readily analyzable. When CsA was tested vs. CoA, the lines in the

3. Results Bile acid conjugation is catalyzed by two enzymes working sequentially. A CoA adduct of the bile acid is first formed by a microsomal membrane-bound bile acid:CoA ligase and then a soluble fraction bile acidCoA:glycine/taurine N-acyltransferase catalyzes the subsequent transfer of the acyl group to the amino group of glycine or taurine. We have examined each of these two enzymes in vitro for their sensitivity to inhibition by CsA. A partially purified preparation of bovine liver N-acyltransferase was tested for inhibition by 20 p~M CsA. The concentration of cholyl-CoA was varied from 2 to 2 0 / x M and the concentration of glycine from 0.5 to 50 mM. There was no inhibition detectable at any concentration of these substrates. We next tested the bile acid:CoA ligase for sensitivity to CsA using the microsomal cell fraction as the source of the enzyme. Using a standard assay, the addition of 10 /xM CsA to the reaction led to 25% inhibition of the ligase. We therefore proceeded to a kinetic analysis of the

0. R

|

|

|

|

|

|

|

|

|

¢

e

0.7 O.S 0.5

~,

0.4: 0.3 0.2 0.t

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

1/[coAl Fig. 2. Double reciprocal plot of the CoA dependence of the rate of the ligase reaction in the presence or absence of cyclosporin A. Ligase activity was determined at variable concentrations of CoA in either the presence (O), or absence (C)) of 25 /xM cyclosporin A (CsA).

D.A. Vessey, M. Kelley / Biochimica et Biophysica Acta 1272 (1995)49-52 |.6

i

i

i

i

i

i 0.01

i 0.02

i 0.0,1

J 0.04.

0.05

tion. This inhibition has a K l of 22 /zM. However, beyond 10 /zM CsA the extent of inhibition increases little, probably because one is approaching the limits of CsA solubility (approx. 10 to 20 /zM). Also, the concentration dependence of CsA inhibition was measured as a function of the Mg 2÷ concentration but in the presence of saturating cholate (10 /zM). A Dixon plot of the data was similar to that shown in Fig. 2. There was a linear high affinity inhibition from 0 to 1 0 / x M CSA which accounted for approx. 40% inhibition. This inhibition had a K~ of 11 ~M. Beyond 10 /zM CsA the extent of inhibition increased little, again probably due to the limited solubility of CsA.

1.6

1.4 1.2 1.0 "-

0.8 0.8

0.4 0.2 0.0 0.00

, J ....

1/[ATP] Fig. 3. Double reciprocal plot of the ATP dependence of the rate of the ligase reaction in the presence or absence of cyclosporin A. Ligase activity was determined at variable concentrations of ATP in either the presence (0), or absence (O) of 25 /xM cyclosporin A (CsA).

double reciprocal plot were parallel (Fig. 2). This suggests that CsA is an uncompetitive inhibitor vs. CoA. When CsA was tested for inhibition vs. ATP it gave a pattern of double reciprocal plots consistent with noncompetitive inhibition (Fig. 3). Veloso et al. [15] have estimated that the physiologic concentration of Mg 2÷ is approx. 1 mM. Therefore, to measure the dependence of the inhibition on the concentration of CsA, we employed 1 mM Mg 2+. We did this at several different non-saturating concentrations of cholate. A Dixon plot [16] of the data is shown in Fig. 4. The maximum inhibition obtainable at any concentration of cholate is approx. 40%. There is a high affinity inhibition from 0 to 10 /xM which accounts for approx. 30% inhibi-

O.g

!

I

!

!

i

0.8

Y

0.7 0.6

0.4 0.3 0.2 0.1 0.0 -30

-20

-10

51

I

I

!

I

I

I0

20

3o

4o

50

[Cyclosporin A] (uM) Fig. 4. Dixon plot of the inhibition of the ligase by cyclosporin A. Ligase activity was determined in the presence of the indicated concentrations of cyclosporin A (CsA), at each of 3 different concentrations of [3H]cholate, 0.8 /~M ('7), 2 /xM (O), and 10 p.M (O).

4. Discussion Bile acid conjugation occurs in two steps involving the formation of an intermediate CoA adduct of the bile acid, followed by transfer of the acyl group to the amino group of glycine or taurine. CsA failed to inhibit the final step in conjugation, i.e., the N-acyltransferase catalyzed reaction. However, this is less consequential because the first step, the bile acid:CoA ligase-catalyzed synthesis of an acyl-CoA is inhibited by CsA, and this is the rate determining step in bile acid conjugation [17]. Thus, the extent of the inhibition of the ligase by CsA should approximate the extent of the overall inhibition of bile acid conjugation. The data in Fig. 1 indicate that, at physiologic Mg 2+, CsA was a competitive inhibitor vs. cholate. This indicates that CsA inhibits by binding at the cholate binding site. The inhibition vs. CoA was uncompetitive. This is what one would expect since the ligase follows a ping-pong mechanism in which cholate and ATP bind first and then the product pyrophosphate is released prior to CoA binding [18]. For an inhibitor binding at the cholate binding site, there is an irreversible step (product release) between the binding of CsA and the binding of CoA, and this generates an uncompetitive inhibition pattern vs. CoA. CsA was also a competitive inhibitor versus Mg 2+, with a K l of 11 /~M. However, at physiologic concentrations of Mg ~+ (approx. 1 raM, see Ref. [15]), the effect of this inhibition was nil. Study of the concentration dependence of the inhibition by CsA indicated that complete inhibition was not obtained within the limits of solubility of CsA. Indeed, it would appear that it is the limited solubility of CsA that limits the extent of inhibition and accounts for the non-linearity of the plots in Fig. 2. The limit of solubility is approx. 10 /~M in buffer, but higher concentrations were obtained in the presence of microsomes. The maximum inhibition achieved was 40%; and 30% inhibition was achievable by 10 /zM CsA. This is a very low concentration and is in the range readily achieved during organ transplantation, particularly if an immediate post-operative period of intravenous infusion with CsA is employed. The

52

D.A. Vessey, M. Kelley/Biochimica et Biophysica Acta 1272 (1995) 49-52

upper limit of therapeutic blood level is approx. 1 to 2 /zM [19] and hepatic levels are higher [20]. Further, CsA is primarily metabolized in the liver and eliminated in the bile [21]. Thus, in the case of transplantation of the liver where postoperatively the level of bile secretion is low [22] and the occurrence of biliary obstruction common [23], the cyclosporin level will be even higher in this period. Assuming that the ligase from human liver has the same sensitivity to CsA as the guinea pig enzyme, then it would be predicted that significant inhibition of bile acid conjugation occurs in vivo during CsA therapy. Such inhibition would contribute to the decreased hepatic uptake of bile acids noted previously [5-8]. Since bilirubin uptake is also decreased by CsA [4], it is suggested that other factors may also contribute. In conclusion, CsA has been found to be an effective inhibitor of bile acid:CoA ligase activity in vitro due to binding at the bile acid binding site. The low inhibition constant suggests that cyclosporin levels common to drug therapy can be expected to lead to inhibition of bile acid conjugation in vivo. Acknowledgements This work was supported by a grant from the National Institutes of Health (DK19212). References [1] Beveridge, T. (1992) Transplant. Proc. 24, 64-66. [2] Fathman, C.G. and Myers, B.C. (1992) N. Engl. J. Med. 325, 1693-1695 (1992).

[3] Schade, R.R., Guglielmi, A., Van Thiel, D.H., Thomson, M.G., Wartu, V., Griffith, B., Sanghvi, A., Bahnson, H. and Hardesty, R. (1983) Transplant. Proc~ 15, 2757-2760. [4] Mason, J. (1989) Pharmacol. Rev. 42, 423-434. [5] Azer, S.A. and Stacey, N.H. (1988) Biochem. Pharmacol. 42, 510512. [6] Ziegler, K. and Frimmer, M. (1986) Biochim. Biophys. Acta 855, 136-142. [7] Stacey, N.H. and Kotecka, B. (1988) Gastroenterology 95, 780-786. [8] Azer, S. and Stacey, N.H. (1993) Biochem. Pharmacol. 46, 813-819. [9] Vessey, D.A., Whitney, J. and Gollan, J.L. (1983) Biochem. J. 214, 923-927. [10] Vessey, D.A., Erdman, A., Kelley, M. and St. Claire, T. (1991) Anal. Biochem. •95, 97-100. [11] Vessey, D.A., Benfatto, A.M. and Kempner, E.S. (1987) J. Biol. Chem. 262, 5360-5365. [12] Vessey, D.A. and Zakim, D. (1977) Biochem. J. 163, 357-362. [13] Nandi, D.L., Lucas, S.V. and Webster, L.T., Jr. (1979) J. Biol. Chem. 254, 7230-7237. [14] Kelley, M. and Vessey, D.A. (1994) Biochim. Biophys. Acta 1209, 51-55. [15] Veloso, D., Guynn, R.W., Oskarsson, M. and Veech, R.L. (1973) J. Biol. Chem. 248, 4811-4819. [16] Dixon, M. (1953) Biochem. J. 55, 170-171. [17] Vessey, D.A. (1978) Biochem. J. 174, 621-626. [18] Kelley, M. and Vessey, D.A. (1995) Biochem. J., in press. [19] Oka, T., Ohmori, Y., Aikawa, I., Ioka, J., Kadotani, Y., Nomura, H., Suzuki, S. and Hashimoto, I. (1983) Transplant. Proc. 15 (Suppl. 1,2), 2501-2506. [20] Lensmeyer, G.L., Siebe, D.A., Carlson, I.H. and Subramanian, R. (1991) J. Anal. Toxicol. 15, 110-115. [21] Goodman, L.S., Gilman, A. and Gilman, A.G. (1990) Pharmacological Basis of Therapeutics, Pergamon Press, Elmsford NY, p. 1269. [22] Javitt, N.B., Shiu, M.H. and Fortner, J.G. (1971) Gastroenterol. 60, 405 -408. [23] Carithers, R.L., Jr., Fairman, R.P., Mendez-Picon, G., Posner, M.P. and Mills, A.S. (1988) in Transplantation of the Liver (Maddrey, W.C., ed.), pp. 111-141, Elsevier Science, NY.