Differential Effects of Chenodeoxycholic and Ursodeoxycholic Acids on Interleukin 1, Interleukin 6 and Tumor Necrosis Factor-a Production by Monocytes YVON CALMUS,
JEROME GUECHOT,~ PHILIPPE PODEVIN, MARIE-THERESE JACQUELINE GIBOUDEAU~ AND RAOUL POUPON'
BONNEFIS,
lINSERM U181 and "Laboratoire de Biochimie, Hepita1 Saint-Antoine, 75014 Paris, France
Cell-mediated immunity and macrophage activity, tivity in vitro (8, 9). High blood concentrations of especially that of Kupffer cells, are impaired during endogenous bile acids could therefore play a role in the cholestasis. Some evidence exists that bile acids play a immune abnormalities observed during cholestasis. Urrole in these immune defects. The purpose of this study sodeoxycholic acid ( U D C A ) , which is able to partially was to evaluate the effects of individual bile acids on correct some of the immunological abnormalities obimmunity and to determine whether monocytes could served in PBC (10, 11) could counter the detrimental be a target. We assessed the effects of chenodeoxycholic acid, an endogenous bile acid, ursodeoxycholic effects of chenodeoxycholic acid ( C D C A ) on immunity. We explored whether immunosuppression by bile acid, which has been shown to partially correct the immunological abnormalities observed in primary acids could be the result of the inhibition of monocyte biliary cirrhosis, and their tauroconjugates on the function by assessing the effects of CDCA and UDCA on production of interleukin-1, interleukin-6 and tumor the production of three monokines, namely, necrosis factor-or. Chenodeoxycholic acid had a dose- interleukin-1 (ILl), interleukin-6 (IL6) and tumor nedependent inhibitory effect on interleukin-1 (inhibi- crosis factor-a (TNFa). IL1, IL6 and TNFa play a tory concentration 50% = 60 pmoUL), interleukin-6 critical role in the induction and the regulation of the (inhibitoryconcentration 50% = 80 pmoUL) and tumor immune response (12, 13) and are potent mediators of necrosis factor-a (inhibitory concentration 50% = 80 pmolb) production; inhibition was almost complete at inflammation (13). 250 pmoUL. In contrast, ursodeoxycholic acid had lesser or minimal inhibitoryeffects (inhibitoryconcenMATERIALS AND METHODS tration 50% = 100 pmol/L for interleukin-1and above CeZZ CuZture. Cells were obtained from six healthy blood 200 pmoUL for interleukin-6 and tumor necrosis factor-a). The inhibitory effects of taurochenodeoxy- donors after informed consent. Mononuclear cell suspensions cholic acid and tauroursodeoxycholicacid were similar were prepared from heparinized venous blood after the to those of chenodeoxycholic acid and ursodeowcholic removal of the platelet-rich plasma by centrifugation on acid, respectively. Ursodeoxycholic acid did not re- Lymphoprep (Nycomed, Oslo, Norway) for 12 min at 20" C. verse the chenodeoxycholicacid-induced inhibition of Cells were collected at the interface and washed twice. Cell interleukin-6or tumor necrosis fact0F-aproduction. In numbers and viability were assessed by trypan blue staining. conclusion,chenodeoxycholic acid exerts strong inhib- The percentage of monocytes was determined by nonspecific itory effects on monOcyte activity in uitro, whereas the esterase staining (14). The mononuclear cell suspension contained less than one platelet per mononuclear cell and less effects of ursodeoxycholic acid are minor. (HEPATOLOGY than 1%of polymorphonuclear leukocytes; more than 95% of 1992;16:719-723.) the cells excluded trypan blue. Monocytes were isolated from total mononuclear cells by Cell-mediated immunity (1-3) and macrophage ac- adherence. Cells were incubated for 90 min in RPMI 1640 tivity, especially that of Kupffer cells (4,5), are impaired supplemented with 10% FCS, 1% L-glutamine, 100 IU/ml during cholestasis, an observation that may be related to penicillin and 100 kg/ml streptomycin in 24-well culture the onset of severe systemic infections in this setting (6, plates. The adherent cells were washed five times; more than 7). Indeed, bile acids have immunosuppressive prop- 95%excluded trypan blue and more than 90%were monocytes. Monocytes were incubated (5 x lo5 cells/ml) in duplicate in erties on cell-mediated immunity and macrophage accomplete medmm for 18 hr in 24-well plates in the presence of lipopolysaccharide (LPS) from Escherichia coli 0 111/B4 or zymosan from Saccharomyces cerevisiae and in the presence or Received July 19,1991;accepted April 29, 1992. Address reprint requests to: Yvon Calmus, Service du Professeur Yves absence of bile acid. LPS and zymosan were both diluted in RPMI 1640 to yield stock concentrations of 1 mg/ml and Chapuis, H6pital &chin, 27 rue du Faubourg SaintJacques, 75014 Paris, 5 mg/ml, respectively.Aliquots were stored at - 20" C until use France. and then thawed and diluted in culture medium to yield final This work was supported by INSERM grant 90 CN 63. 31/1/39354 concentrations of 100ng/ml and 200 pg/ml, respectively.At the 7'19
CALMUS ET AL.
/o=Ym-"stimulated
HEPATOLOGY
1
T
T
0
TNFa
ILI
IL6
0
TNFa
FIG.1. Spontaneous and LPS-stimulated productions of IL1, IL6 and TNFa by monocytes, and spontaneous and zymosan-stimulated production of TNFa. Results are expressed as concentrations (picomoles per liter) of cytokines in the culture medium at the end of the incubation period.
end of the incubation period, supernatants were stored at - 80" C until use. Cytokine Assays. ILlp was measured using an RIA (Interleukin-lp [12511assay system, Amersham France, Les Ulis, France). The system uses a rabbit anti-ILlp antiserum, a high specific-activity [1251]ILlp (human recombinant) tracer, and human recombinant ILlp as standard. The sensitivity of the assay is 1.5 pmovl of supernatant; within assay and between assay variations were less than 4% and 9%, respectively. Preliminary tests showed that CDCA and UDCA (up to 250 FmoliL) do not interfere with the assay. IL6 was measured using an immunoradiometric assay (Interleukin-6 IRMA, Medgenix Diagnostics, Brussels, Belgium). The system uses several antibodies directed against IL6, a high-specific-activity [1251]IL6 (human recombinant) tracer and human recombinant IL6 as standard. The sensitivity of the assay is 0.5 pmol/L of supernatant; within-assay and between-assay variations were less than 7% and 8%, respectively. Preliminary tests showed that CDCA and UDCA (up to 250 Fmol/L) do not interfere with the assay. TNFa was measured using an immunoradiometric assay (TNF-a-IRMA, Medgenix Diagnostics, Brussels, Belgium). The system uses several monoclonal antibodies directed against distinct epitopes of human TNFa and recombinant TNFa as standard. The sensitivity of the assay is better than 0.3 pmoVI, within assay and between assay variations were less than 6% and 7%,respectively. Preliminary tests showed that CDCA and UDCA (up to 250 p r n o n ) do not interfere with the assay. The results for the three cytokine productions are expressed in most cases as the percentage of control values, calculated as follows: (concentration in presence of bile acid at the end of the incubation period - concentration in absence of bile acid) x 100. Bile Acid Measurement. Bile acid concentrations in the culture medium were measured by means of an enzymatic method (Enzabile, Nycomed, Oslo, Norway). Endotoxin Assay. Endotoxin levels in the culture medium were measured using a chromogenic limulus amoebocyte lysate test (Coatest endotoxin, Kabivitrum, Stockholm, Sweden). The detection limit is 5 pg/ml of culture medium.
100
200
D
Bile acid concentration (CIM)
FIG.2. Effects of CDCA (0) and UDCA ( 0 ) on LPS-stimulated IL1 production by monocytes. *p c 0.05 compared with value without bile acid. Dotted line corresponds to spontaneous IL1 production.
E f f m t of Bile Acids on Cell Viability. Fresh mononuclear cells ( > 95%trypan blue negative) were pelleted and incubated with 200 FCi N+51Cr0, (20 ~1 from a 10 mCi/ml stock solution) for 90 min at 37" C in 5% CO,. The cells were then washed three times in RPMI and incubated in complete medium for 60 min at 37" C in 5%CO,. After resuspension in complete medium, the cells were incubated at 1 x lo6 cellslml in 96-well microplates at 37" C in 5% CO, for 18 hr, with various concentrations of CDCA or UDCA. At the end of the incubation period, the plates were spun at 200 g for 10 min, and 50 pl of the supernatant was removed for counting in a y-counter. The percent specific lysis was calculated as: (counts per minute in experiment - counts per minute spontaneous lysislcounts per minute maximum release - counts per minute spontaneous release) x 100. Sources. Cell-culture (12 x 75 mm) polypropylene tubes, 96-well U-bottomed microtiter and 24-well plates were purchased from Falcon (Grenoble,France). CDCA, UDCA (sodium salts)and their tauroconjugates (TCDCA and TUDCA) with a purity of 99% were obtained from Calbiochem Co. (San Diego, CA). LPS prepared by Westphal phenolic extraction from Escherichia coli 0111:B4, zymosan from Succharomyces cereuisiae, penicillin, streptomycin and L-glutamine were obtained from Sigma Chemical Co. (St. Louis, MO). RPMI 1640 and FCS were obtained from Gibco (Cergy-Pontoise, France), and Lymphoprep was from Nycomed (Oslo, Norway). Statistical Analysis. The data were expressed as mean S.E.M. of six separate experiments. Statistical analysis was performed using nonparametric analysis of variance (Kruskal-Wallis).
*
RESULTS Effmt of CDCA and UDCA on ILl Production by Monocytes. Figure 1 shows that IL1 production (pico-
moles per liter culture medium) was enhanced by endotoxin stimulation (LPS, 100 ng/ml). Figure 2 shows the effect of CDCA and UDCA on IL1 production by LPS-stimulated monocytes. CDCA had a dosedependent inhibitory effect on IL1 secretion; 50% inhibition was obtained at a CDCA concentration of 60
Vol. 16, No. 3, 1992
72 1
DIFFERENTIAL, EFFECTS OF CDCA AND UDCA ON MONOKINES
1
0
Bile acid concentrationCM)
FIG.3. Effects of CDCA (a),UDCA ( 0 ) and CDCA plus UDCA (25 KmoVL) (A) on LPS-stimulated IL6 production by monocytes. *p < 0.05 compared with value without bile acid. Dotted line corresponds to spontaneous IL6 production.
Fmol/L in the culture medium, and almost complete inhibition was obtained at 250 p,moVL. Sigdicant inhibition relative to control values was observed at concentrations from 50 Fmol/L to 250 FmoVL. In contrast, 50% inhibition by UDCA was only reached at a concentration of 100 pmoVL. CDCA and UDCA had sigmficantly different degrees of inhibition at concentrations ranging from 50 to 150 Fmol/L. Effect of CDCA and UDCA on IL6 Production by Morrocytes. IL6 production (picomoles per liter culture
medium) was enhanced by endotoxin stimulation (Fig. 1). CDCA had a dose-dependent inhibitory effect (Fig. 31, with 50% inhibition at 80 FmoVL and almost complete inhibition at 250 pmoVL. Significant. inhibition relative to control values was observed at concentrations from 25 FmoVL to 250 FmoVL. In contrast, 50% inhibition by UDCA was only reached at concentrations above 200 Fmol/L. Differences in the inhibitory effects of CDCA and UDCA were significant for concentrations ranging from 50 to 250 kmol/L. UDCA (25 FmoVL) had no significant effect on the inhibition induced by CDCA (Fig. 3). Efect of CDCA and UDCA on TNFa Production by Morrocytes. TNFa production (picomoles per liter culture medium) was enhanced by endotoxin stimulation (Fig. 1). CDCA had a dose-dependent inhibitory effect on TNFa secretion by LPS-stimulated monocytes (Fig. 4);50% inhibition was obtained at 80 pmoVL, and almost complete inhibition was obtained at 150 FmoVL. Sigmficant inhibition relative to control values was observed at concentrations from 50 FmolL to 250 Fmol/L. In contrast, 50%inhibition by UDCA was only reached at concentrations above 200 FrnoVL. Differences in the inhibitory effects of CDCA and UDCA were significant at concentrations ranging from 100 to 250 FmoVL. UDCA (25 to 100 FmoVL) did not reverse the inhibition induced by CDCA (100 Fmol/L) (Table 1).
FIG.4. Effects of CDCA ( 0 )and UDCA ( 0 )on LPS-stimulated TNFa production by monocytes. *p < 0.05 compared with value without bile acid. Dotted line corresponds to spontaneous TNFa production.
TABLE1. Effect of CDCA (100 pmol/L) plus UDCA on LPS-stimulated TNFa production by monocytes Bile acid in the medium
None CDCA alone CDCA + UDCA (25 p,mol/L) CDCA + UDCA (50 p,moVL) CDCA + UDCA (75 p,moVL) CDCA + UDCA (100 +mol/L)
TNFa production (pmovL) 315 ? 63 126 ? 31" 89 ? 1 1 " ~ ~ 51 2 2",b 56 ? 8",' 46 ? 5".'
"p < 0.05 vs. controls. 'NS vs. CDCA alone.
The effect of CDCA was also assayed on the zymosanstimulated production of TNFa. TNFa production (picomoles per liter culture medium) was enhanced by zymosan stimulation (200 Fg/ml) (Fig. 1). CDCA had a dose-dependent inhibitory effect (Fig. 5); 50%inhibition was obtained at 70 pmol/L, and almost complete inhibition was obtained at 100 p,mol/L. Differential Efects of CDCA, UDCA and their Tauromqjugates on TiVFaProdudion. In another series of
experiments, the effects of TCDCA and TUDCA (200 p,moVL) on TNFa production were compared with those of CDCA and UDCA (200 FmolL). TCDCA had less of an inhibitory effect than CDCA (Fig. 6), although the difference was not significant. TUDCA and UDCA had no significant inhibitory effect on TNFa production at 200 FrnoVL. E m t of CDCA and UDCA on Cell Viability. Figure 7 shows 51Crrelease from mononuclear cells incubated for 18 hr with various concentrations of CDCA or UDCA. UDCA had no cytotoxic effect at concentrations up to 250 pmoVL. Cytolysis, assessed by 51Cr release, was significant at 100 FmoVL CDCA (5.3%)but reached only 10.8% at a concentration of 250 pmol/L.
722
CALMUS ET AL.
HEPATOLOGY
J
-- -----43 Y
0
I
'
50
0
I
'
100
I
'
150
I
200
I
250 Bile acid concentration (pM)
CDCA concentration(CIM)
FIG. 5. Effects of CDCA on zymosan-stimulated TNFa production by monocytes. *p < 0.05 compared with value without bile acid. Dotted line corresponds to spontaneous TNFa production.
T
T
Control
CDCA
UDCA
TCDCA
TUDCA
Bile acid (200 pM)
FIG.6. Effects of CDCA, UDCA, TCDCA and TUDCA (200 KmoVL) on LPS-stimulated TNFa production by monocytes expressed as concentrations (picomoles per liter) in the culture medium a t the end of the incubation period. *p < 0.05 compared with value without bile acid.
Endotoxin Assays. No endotoxin was detected in the culture medium. In the presence of LPS, the endotoxin concentration in the culture medium was not modified by 250 pmoVL CDCA or 250 pmolL UDCA (data not shown). DISCUSSION
Cell-mediated immunity (1-3)and macrophage functions, especially those of Kupffer cells (4, 51, are impaired during cholestasis. The mechanisms of these immune defects are not fully understood, although reduced cytokine production has been shown to occur (15, 16). Our results show that CDCA induces a dose-dependent inhibition of IL1, IL6 and TNFa production by monocytes, suggesting that high concentrations of endogenous bile acids could play a role in the
FIG.7. Effects of CDCA and UDCA on cell viability. Cell viability was estimated by 51Cr release after 18 hr of incubation with bile acids. *p < 0.05 compared with value without bile acid.
immunodeficiency observed during cholestasis and that monocytes could be an important target of bile acids. We chose to study CDCA because it is one of the two major primary endogenous bile acids and its concentration is strongly increased during cholestatic disorders (17). In this work, unconjugated forms of the bile acids were used at concentrations of 0 to 250 pmoVL. Blood cells are exposed essentially to conjugated forms of bile acids, which have less marked toxic effects than unconjugated forms (18,19);however, the inhibitory effects of TCDCA and TUDCA in this work were similar to those of their unconjugated counterparts. During chronic cholestatic diseases, tissue concentrations of bile acids of 100 to 250 pmol/L are commonly observed, although blood concentrations are lower (18,20). Our results can reasonably be extrapolated to the in viuo situation because the contact time between the immune cells and bile acids was much shorter in our in uitro conditions (18 hr) than during cholestasis. Indeed, the concentration time product, which reflects the effects of a substance more closely than concentration alone (21), is of the same order of magnitude in the two situations. In our model, the cytotoxicity of CDCA was low, reaching only 11%at 250 pmoVL after 24 hr of incubation. Previous studies have shown that the viability of Kupffer or blood cells is not significantly reduced at taurocholate concentrations below 1 mmoVL (18, 22). A direct cytotoxic effect thus cannot account for the inhibitory effect of CDCA in our conditions. Endotoxin concentration, measured at the end of the incubation period, was not significantly modified by the presence of 250 pmoVL CDCA or UDCA, a direct effect of bile acids on endotoxin can thus be excluded. Although bile acids could reduce the uptake of LPS by monocytes, it has been reported that LPS uptake by Kupffer cells is only significantly reduced in the presence of 1 mmol/L taurocholate (22). Inhibition of LPS uptake is therefore unlikely to explain the immune effect of bile acids in our model. To exclude a potential interaction between LPS and bile acid in our
Vol. 16, No. 3, 1992
DIFFERENTIAL EFFECTS OF CDCA AND UDCA ON MONOKINES
conditions, the effect of CDCA was also assayed on TNFa production stimulated by zymosan. The inhibitory effect was of the same order of magnitude as that on LPS-stimulated production. In the presence of high concentrations of CDCA, LPS-induced cytokine production was lower than spontaneous control production (Figs. 2 to 4), suggesting that CDCA may have an inhibitory effect on intracellular pathways of cytokine production. It has been shown that activation of monocytes (23), particularly by LPS (24), is dependent on transduction pathways involving both an increase in the intracellular calcium concentration and the activation of protein kinase C. Because bile acids can modulate protein b a s e C activity (25) and enhance intracellular calcium concentrations (261, it is tempting to speculate that their effects on monocyte activity are mediated by a direct effect on these transducing signals. The effects of CDCA on intracellular calcium concentrations and the protein kinase C and protein kinase A pathways are currently under study. Immunosuppressive drugs such as cyclosporine and corticosteroids have inhibitory effects on monocytes (27-29) and exert negative effects on cytokine production (28). Bile acids could exert their immune effects by a mechanism similar to that of immunosuppressants. UDCA is a bile acid for which the physicochemical and biological effects are different from those of endogenous bile acids (19).The inhibitory effect of UDCA on IL1, IL6 and TNFa production was significantly less marked than that of CDCA. In our model, the addition of 25 FmoVL UDCA had no significant effect on the CDCA (25 to 250 FmoVL)-induced inhibition of IL6 production, and UDCA (25 to 100 p,mol/L) did not reverse the inhibition induced by CDCA (100 pmoVL) on TNFa production. This suggests that the beneficial effects of UDCA in humans cannot be ascribed to a competition between endogenous bile acids and UDCA at the target cell level but rather to a reduction in the endogenous bile acid concentration induced by long-term administration of UDCA (30). In conclusion, CDCA exhibits strong inhibitory properties in uitro on monocyte activity assessed in terms of IL1, IL6 and TNFa production. This effect could play a role in the immune abnormalities observed during cholestasis. UDCA has less immunosuppressive properties. REFERENCES 1. Roughneen PT, Drath DB, Kulkarni AD, Rowlands BJ. Impaired nonspecific cellular immunity in experimental cholestasis. Ann Surg 1987;206:578-582. 2. h s t r o n g CP, Dixon JM, Taylor TV, Davies GC. Surgical experience of deeply jaundiced patients with bile duct obstruction. Br J Surg 1984;71:234-238. 3. Cainzos M, Potel J, Puente JL. Anergy in jaundiced patients. Br J Surg 1988;75:147-149. 4. Drivas G, James 0, Wardle N. Study of the reticuloendothelial phagocytic capacity in patients with cholestasis. BMJ 1976:l: 1568-1569. 5. Halpem BN, Biozzi G, Nicol T, Bilberg DLJ. Effect of experimental biliary obstruction on the phagocytic activity of the reticuloendothelial system. Nature 1957;180:503-504. 6. Feduccia TD, Scott-Conner CEH, Grogan JB. Profound suppression of lymphocyte function in early biliary obstruction. Am J Med Sci 1988;296:39-44.
723
7. Pain JA, Cahill CJ, Bailey ME. Perioperative complications in obstructive jaundice: therapeutic considerations. Br J Surg 1985: 72:942-945.
8. Gianni L, Di Padova F, Zuin M, Podda M. Bile acid-induced inhibition of the proliferative response to phytohemagglutinin and pokeweed mitogen: an in uitro study. Gastroenterology 1980;78: 231-235. 9. Keane RM, Gadacz TR, Munster AM, Birmingham W, Winchurch RA. Impairment of human lymphocyte functions by bile salts. Surgery 1984;95:439-443. 10. Calmus Y, Gane P, Rouger P, Poupon R. Hepatic expression of class I and class I1 major histocompatibility complex molecules in primary biliary cirrhosis: effect of ursodeoxycholic acid. HEPATOLOGY 1990;11:12-15. 11. Poupon RE, Balkau B, Eschwege E, Poupon R, the UDCA-PBC Study Group. A multicenter double blind trial of ursodiol for the treatment of primary blliary cirrhosis. N Engl J Med 1991;324: 1548-1554. 12. Mizel S. The interleukins. FASEB J 1989;3:2379-2388. 13. O’Garra A. Interleukins and the immune system. Lancet 1989;l: 943-947. 14. Tucker SB, Pierre RV, Jordan RE. Rapid identification of monocytes in a mixed mononuclear cell preparation. J Immunol Methods 1977;14:267-269. 15. Haga Y, Sakamoto K, Egami H, Yokoyama Y, Arai M, Mori K, Akag~M. Changes in production of interleukin-1and interleukin-2 associated with obstructive jaundice and biliary drainage in patients with gastrointestinal cancer. Surgery 1989;106:842-848. 16. Kakumu S, Yoshioka K, Tsutsumi Y, Wakita T, Arao M. Production of tumor necrosis factor and interleukin-1 by peripheral blood mononuclear cells from patients with primary biliary cirrhosis. Clin Immunol Immunopathol 1990;56:54-65. 17. Stiehl A. Disturbances of bile acid metabolism in cholestasis. In: Paumgartner G , ed. Clinics in gastroenterology: bile acids. London: WB Saunders, 1977:45-89. 18. Coleman R, Iqbal S, Godfrey PP, Billington D. Composition of several mammalian biles and their membrane-damaging properties. Biochem J 1979;178:201-208. 19. Hofmann AF, Roda A. Physicochemical properties of bile acids and their relationship to biological properties: an overview of the problem. J Lipid Res 1984;25:1477-1489. 20. Tujm a n N, Nair PP. Tissue-bound bile acids. In: Setchell KDR, Kritchevsky D, Nair PP, eds. The bile acids: chemistry, physiology and metabolism. New York: Plenum Press, 1988:373-378. 21. Rowland M, Tozer TN. Therapeutic response and toxicity. In: Clinical pharmacokinetics: concepts and applications. Philadelphia: Lea and Febiger, 1980:155-172. 22. Van Bossuyt H, Desmaretz C, Gaeta GB, Wisse E. The role of bile acids in the development of endotoxemia during obstructive jaundice in the rat. J Hepatol 1990;10:274-279. 23. Klein JEi, Schepers TM, Dean WL, Sonnenfeld G, McLeish KR. Role of intracellular calcium concentration and protein kinase C activation in IFN-y stimulation of U937 cells. J Immunol 1990;144:4305-4311. 24. Kovacs EJ, Brock B, Varesio L, Young HA. IL-2 induction of IL-1 beta mRNA expression in monocytes: regulation by agents that block second messenger pathways. J Immunol 1989;143:35323537. 25. Fitzer CJ,O’Brian CA, Guillem JG, Weinstein IB. The regulation of protein b a s e C by chenodeoxycholate, deoxycholate and several structurally related bile acids. Carcinogenesis 1987;8: 21 7-220. 26. Anwer MS, Engelking LR, Nolan K, Sullivan D, Zimniak P, Lester R. Hepatotoxic bile acids increase cytosolic C a + + activity of isolated rat hepatocytes. HEPATOLOGY 1988;8:887-891. 27. Baybutt HN, Holsboer F. Inhibition of macrophage differentiation and function by cortisol. Endocrinology 1990;127:476-480. 28. Narumi S, Hamilton TA. Dexamethasone selectively regulates LPS-inducible gene expression in murine peritoneal macrophages. Immunopharmacology 1990;19:93-101. 29. Benson A, Ziegler HK. Macrophage as targets for inhibition by cyclosporine. Transplantation 1989;47:696-703. 30. Chretien Y, Poupon R, Gherardt MF, Chazouilleres 0, Labbe D, Myara A, Trivin F. Bile acid glycine and taurine conjugates in serum of patients with primary biliary cirrhosis: effect of ursodeoxycholic treatment. Gut 1986;30:1110-1115.