Toxicity of Chenodeoxycholic Acid in the Rhesus Monkey

GASTROENTEROLOGY 69: 333-337, 1975 Copyright© 1975 by Th e Willia ms & Wilkins Co.

Vol.(i9. No.:! Printed ill U.S.A.

TOXICITY OF CHENODEOXYCHOLIC ACID IN THE RHESUS MONKEY H . DvRSZKA, P.H.D .

M.D .,

T.

CHEN, M.D.,

G.

SAI.EN, M . D ., AND

E . H.

MosnACII ,

Colle!fe of M edicine and Dent istry of N ew J ersey, New Jers ey Medical S chool, Newarh, N ew Jers ey, Public Hea lth Res ea rch Institut e of New Yorh, Netu Yorh, New Yorh , and Vetera ns Administration Hospital, East Oran!ie, New Jersey

Chenodeoxycholic acid is an important drug for the treatment of cholesterol cholelithiasis in man. Although no toxicity has been demonstrated in man, liver lesions develop in rhesus monkeys treated with chenodeoxycholic acid . To elucidate the mechanism of toxicity, chenodeoxycholic acid was fed daily to three groups of 6 animals each at the following dose: 10, 40, and 100 mg per kg; 2 separate animals were not treated and served as controls. After l month, the animals were killed. During the treatment period, most blood tests (e .g., blood count, blood urea nitrogen, albumin, SGOT, lactate dehydrogenase) remained within normal limits, but there was a significant dose-related increase in serum leucine aminopeptidase levels. The percentage of lithocholic acid, the 7 -dehydroxylated bacterial metabolite of chenodeoxycholic acid, rose from 1% in th e control animal to almost 14ry,, in the 100 mg per kg-treated group. Liver biopsies obtained before treatment and at necropsy showed no significant changes. Thus , exposure of the liver to increased amounts of lithocholic acid during chenodeoxycholic acid treatment might result in elevation of serum leucine aminopeptidase activity. Chenodeoxycholic acid (CDCA) has been shown to be effective in dissolving cholesterol gallstones in man. 1• 2 The intake of CDCA markedly alters the composition of bile and increases cholesterol solubility. 3 Studies in man have demonstrated no serious side effects or only mild transient serum transaminase elevations associated with the prolonged intake of CDCA. 1• 4 In contrast, investigations in some mammals have shown considerable functional and structural liver changes when CDCA was given (Huntingdon ReReceived January 31, 1975. Accepted March 31, 1975. Address requests for reprints to: G. Salen M.D . , Veterans Administration Hospital , East Orange, New Jersey 07019. This work was s upported hy a grant from the Intellectual Property Development Corporal ion, New Rochelle, New York 10804.

search Centre, unpublished results).r.-H The mechanism that leads to hepatotoxicity is not clear . However, biochemical and histological changes resemble those induced by feeding lithocholic acid (LCA). 9 • 10 Studies in rabbits suggest that the LCA formed from CDCA by intestinal bacteria can be absorbed and might be the toxic agent when CDCA is fed. 7 However, CDCA does not constitute a major component of rabbit bile. In contrast, the bile acid components of some subhuman primates are similar to that of man. 11 In this study, we therefore examined bile composition in relation to biochemical and histological findings in rhesus monkeys on various doses of CDCA.

Methods Animals. The studies were conducted in 20 rhesus monkeys (Macaca mulatta) weighing

334

DYRSZKA ET AL.

between 2.0 and 3.0 kg. The animals were housed at the New York University Primate Center at Sterling Forest (Laboratory tor Experimental Medicine and Surgery in Primates). Prior to the study, chest X-rays and parasitological studies were obtained and any diseased animals were eliminated from the study. All monkeys were fed a standard diet (Purina monkey chow) throughout the experimental period. There were four study groups. Group I comprised 2 animals and served as a control group. Each of the other three groups comprised 6 animals with 3 males and 3 females per group. These groups were treated with CDCA (I.P.D. Corp., New Rochelle, N.Y .; 99% pure CDCA and 1% cholic acid; no LCA was detected by gas-liquid chromatography) at the following dose levels: group II, 10 mg per kg per day ; group III, 40 mg per kg per day; group IV, 100 mg per kg per day. CDCA was administered once a day in the morning by force feeding gelatin capsules. Individual doses were approx imated with combinations of 100-, 50-, and 10-mg capsules. The control animals received placebo capsules. All test animals were killed after 33days of CDCA administration with the exception of 1 animal from each treated group . These animals were killed 9 days after discontinuation of CDCA. Laboratory inuesti{!ation. Laboratory studies obtained prior to and on termination of the treatment period were as follows: total blood count, SGOT, total and direct serum bilirubin, serum total protein and albumin, serum cholesterol, blood urea nitrogen , creatinine, electrolytes , and blood glucose. Serum leucine aminopeptidase (LAP) was measured 19 days after the initiation of CDCA treatment (Roche Clinical Laboratories, Nutley, N.J.). Determination of biliary lipids . Bile was obtained by needle aspiration from the gallbladder of the animals after death. Biliary bile acid concentration and composition were determined as described by Salen et at. 1 2 The phosphorus content of a Folch extract of bile was determined by the method of Chen et at ." Phospholipid concentration was calculated by multiplying the phosphorus content by 25 . PatholoRical studies. Liver tissue obtained prior to CDCA administration by percutaneous biopsy was examined histologically and com pared to liver tissue obtained after death. The treatment groups were not known to the examiner . Gross examination of the following organs revealed no abnormalities: brain, heart, lungs, stomach, small intestine, colon, gallbladder, liver, kidneys, gonads, and eyes.

Vol. 69, No. 2

Results Fatalities No deaths occurred. Clinical Signs Bowel fun ction Group I. No change in bowel function. Group II. No diarrhea occurred throughout study period. Group III. One animal had diarrhea for 4 days during the study period. Group IV. Three animals developed diarrhea for periods from 3 to 5 days, adding up to 15 days of liquid feces for a total of 196 study days in this group. Despite changes in bowel function, the animals continued to receive the prescribed dose of CDCA. Food intak e. There was no difference in food intake before and during CDCA treatment in groups II and III. One animal in group IV showed diminished food intake during the 2nd week of CDCA administration. Weight . There were only slight changes in weight during the study period. All treated groups showed changes comparable to those of the control animals. Laboratory Investigations All laboratory studies with the exception of serum leucine aminopeptidase remained unchanged after CDCA administration in comparison to control and pretreatment values. Serum LAP was elevated in all three treatment groups as compared to the control values (the control values for LAP were obtained from the 2 control animals and 4 healthy, untreated monkeys of the same age kept under identical conditions; however, these animals are otherwise not included in this study), and the increase was proportional to the dose of CDCA administered. The mean values and statistical analysis a.re presented in figure 1. Biliary Bile Acid Composition Administration of CDCA caused a marked decrease in the proportion of cholic acid and deoxycholic acid in the bile (table 1) . CDCA became the major component of

August 1975

CHENODEOXYCHOLIC ACID TOXICITY

the biliary bile acids. The proportion of LCA increased by a factor of 7 to 10 and exceeded the amount of cholic acid and deoxycholic acid in the 40- and 100-mg per kg groups. We did not search for sulfated bile acids and, therefore, the total LCA content might have been even higher. There were no significant changes in biliary phospholipid and cholesterol concentrations and their proportions relative to bile acids did not change in the treated ani15

II'!.

0LAP

~

LCA %of

10

BILE

150

,--

II LCA

-

100

~

ACIDS

LAP

;.,

!:;

5

0

al Control

mu"

r

50

/~ Gr.n IOmqlkq

~

**

Gr.m 40mqlkq

**

.,

0

Gr. Ill IOOmq/kq

**

FIG . 1. Rel a tionship between serum leucine aminopeptidase (LAP) and percentage of lithocholic acid (LCA) in bile. *, milliunits per ml of serum; **,dose of chenodeoxycholic acid in milligrams per kg of body weight per day . The mean values ± sn for LAP and percentage of biliary LCA were as follows: Animal group

No. of animals

-----Control group

2

II III IV

5 5 5

LAP

18 ± 3 (Mean LAP obtained from6 animals) 42 ± 26 101 ± 52 145 ±55

Range

( 13- 22)

(17- 82) (44-195) (66-220)

LCA

1.24

8.4 11.8 13 .:!

± ± ±

1.8 :l.5 2.1

The differences of LAP activ ities between controls and group II, and between group II and group III are statistically significant (P < 0 .05 by Student's t-test) . There is no statistically significant difference between groups III and IV (P> 0.05). Percentage of biliary LCA differs significantly between control animals and treated animals (P < 0.005) . No statistically significant difference for LCA was found between groups II and III and groups III and IV (P > 0.05). There is positive correlation between serum LAP activi ties and percentage of biliary LCA at the 95% confidence level by linear regression.

335

mals. The three animals killed 9 days after CDCA was discontinued showed marked improvement in bile acid composition: the proportions of LCA and CDCA decreased and were replaced by endogenous cholic acid. Li,;ht Microscopy of Liver Needle biopsies prior to the study were compared to wedge specimens obtained after death. Mild inflammatory changes compatible with nonspecific hepatitis were found in some animals before and after dosing. These changes were unrelated to the administration of the test material . In group II, 1 animal was found to have mild nonspecific focal hepatitis prior to the study with normal histology on termination. Another animal had similar changes at the end of the test period. In group III, 2 animals had mild nonspecific f<>cal changes prior to dosing and unremarkable histological findings after death. In group IV, initial nonspecific mild inf1ammatory changes in 1 animal could no longer be appreciated after dosing. Another animal in this group showed mild focal changes at death. Correlation of Histology and Laboratory Findings with Biliary Composition After 33 days of CDCA treatment, no specific histological abnormalities were observed. However, in all treated groups the LAP levels were elevated proportionally to the dose given. Similarly, the amount of LCA in the bile increased in proportion to the dose of CDCA administered. The relation between serum LAP and biliary LCA is shown in figure 1. For statistical analysis, see legend to figure 1.

Discussion The feeding of CDCA for 1 month was not associated with any significant histological liver abnormalities in our animals. The absence of significant hepatic lesions is in contrast to previous studies and is most likely explained by the shorter duration of CDCA administration in our animals (Huntingdon Research Centre, un-

336

DYRSZKA ET AL. TABLE

Vol . 69, No. 2

1. Mean biliary bile acid composition as p er cent of total bile acids"

Group' 1: controls (2) II: 10 mg/ kg" (5) III: 40 m g/kgc (5) IV: 100 mg/kg'. (5) Treatment discontinued (3)

CA

COCA

LCA

DCA

51.7 (45 .9-57.5)" 12.8 (4.5- 17 .8) 2.6 (0.4- 4.4) 3.8 (1.0-7 .6) 42.4 (37 .6- 50.4)

19.9 (19.0- 20.9) 64 .6 (69 .5- 74.4) 82 .5 (58.1 -87.0) 78 .4 (74 .0- 81.0) 17.7 (14 .7- 23.9)

1.2 (0.8-1.6) 8.4 (6 .2-10.3) 11.9 (8.6- 16.9) 13.3 (1 1.!3-16.9) 3.5 (3.2-4 .0)

27. 1 (19 .9-34.3) 14.1 (10 .9- 19.7) 3.0 (1.1 -4 .6) 4.2 (1.7-6.5) 36.6 (31.6- 42.6)

a CA, c holic acid; CDCA, chenodeoxycholic acid; LCA, lithocholic acid; DCA, deoxyc holic acid. ' Figure in parentheses is number of animals in each group. ,. Dose of CDCA in milligrams per kg of body weight per day. "Figures in parentheses are ranges for th e various groups .

published results). 8 A 6- month study on rhesus monkeys was carried out at the Huntingdon Research Center in England and revealed ductular hyperplasia and mononuclear cell infiltration ("triaditis") after administration of comparable doses of CDCA (Huntingdon Research Centre, unpublished results). In our experience, histological examination of control and pretrial biopsies is important because they show that cellular infiltration of portal triads is not an uncommon finding in untreated rhesus monkeys and is not necessarily related to CDCA treatment. Ductular proliferation probably requires more prolonged exposure to CDCA than that in our study. Similarly, the absence of elevated SGOT levels is at variance with other investigations; however, it is consistent with our normal histological findings. In contrast, serum levels of LAP were elevated in all but 1 animal treated with CDCA. However, the possibility of transient SGOT elevation prior to our determination and, similarly, decrease of LAP activity following our measurement cannot be ruled out. Serum LAP has the same significance as alkaline phosphatase in liver disease. 14 However, it is a more specific liver enzyme . In some instances it has been found to be elevated in humans with various hepatobiliary diseases in the absence of raised alkaline phosphatase and transaminase activities. 15 Since it is not present in bone, 16 it was used in our study in place of

alkaline phosphatase which was elevated in our rhesus monkeys with active bone growth. Serum LAP apparently became elevated despite the absence of histological liver abnormalities . Thus, if histological abnormalities occur in prolonged administration of CDCA, increased serum LAP levels may indicate early hepatic toxicity of CDCA. As observed in other species and man, 3 • 7 the administration of CDCA led to marked changes in biliary bile acid composition. The proportion of CDCA and LCA increased and was associated with diminished cholic acid and deoxycholic acid . These changes were rapidly reversed in the animals killed 9 days after cessation of CDCA feeding. Lithocholic acid has been shown by Cooper to cause severe liver damage in rhesus monkeys. 9 Hunt showed that in primates treated with LCA, liver lesions develop that closely resemble the bile duct proliferation seen in monkeys treated with CDCA for long periods .'° Fischer et a\. 7 have shown that liver damage is produced in rabbits fed CDCA, and histological lesions correlated with the proportion of biliary LCA. Although no histological changes were seen in our treated animals, elevated serum LAP levels parallel the proportion of biliary LCA. The increase in serum LAP was not directly related to biliary CDCA concentrations, since the proportion of CDCA was higher in the bile of group III than group IV animals, whereas

August 1975

CHENODEOXYCHOLIC ACID TOXICITY

LAP was greatest in group IV animals with the highest proportion of biliary LCA. Thus, it appears that LAP levels were related to the relative concentrations of LAC in the bile. We suggest, therefore, that the hepatotoxic effect of CDCA in rhesus monkeys can possibly result from the continuous exposure of the liver to LCA. LCA is produced by 7a-dehydroxylation of CDCA through the action of intestinal bacteria. The absorption of LCA and its incorporation into the enterohepatic circulation lead to its appearance in the bile. In some species, LCA can be metabolized by the liver to other bile acids, such as hyodeoxycholic acid and CDCA. In addition to little absorption ofLCA, 17 these hydroxylation mechanisms appear to be active in the rat, 18 • 19 which is relatively resistant to the toxic effect of LCA, 20 and shows relatively little LCA in the bile when fed CDCA. In man, only small amounts of LCA are detected in the bile. 12 However, it is important to emphasize that substantial amounts may be sulfated by the liver and excreted in this form. 21 Our methods would not detect sulfated LCA. Thus, even larger amounts of LCA may be present in monkey bile than shown in table 1. Another difference in the metabolism of CDCA between man and the rhesus monkey is that large amounts of ursodeoxycholic acid, the 7{Jhydroxy epimer of CDCA, are found in humans treated with CDCA. 12 The ursadeoxycholic acid may represent a detoxification mechanism to control the concentration of CDCA in the human liver since it is formed from CDCA. Thus, the hepatic toxicity that develops with the administration of CDCA to the rhesus monkey may result from an inability to metabolize CDCA or its bacterial metabolite LCA.

5.

6.

7.

8.

9.

10.

ll.

12.

13.

14.

15.

16. 17 .

REFERENCES 1. Thistle JL, Hofmann AF: Efficacy and specificity of chenodeoxycholic acid therapy for dissolving gallstones. N Eng! J Med 289:655-659, 1973 2. Bell GD, Whitney B , Dowling RH: Gallstone dissolution in man using chenodeoxycholic acid. Lancet 2:1213-1216, 1972 3. Thistle JL, Schoen field JL: Induced alterations in composition of bile of persons having cholelithiasis. Gastroenterology 61:488-496, 1971 4. Bell GD, Mok HYI, Thwc M, et a!: Liver st ruc-

18. 19.

20. 21.

337

ture and function in cholelithiasis: effect of chenodeoxycholic acid. Gut lfi : 16fl- 172, 1974 Holsti P: Bile acids as a cause of liver injury: cirrhogenic effect of chenodeoxycholic acid in rabbits. Acta Pathol Microbiol Scall(! fi4:479. 1962 Bergmann F, Vader!inden W: Liver morphology and gallstone format ion in hamsters and mice treated with chenodeoxycholic ac id . Acta Pat hoi Microbial Scand 81:21:1-221, 197:1 Fischer CD, Cooper NS. Rothsc hild MA. et a!: Effect of dietary chenodeoxycholic acid and lithocholic ac id in the rabbit. Am ,J Dig Dis 19:877-886, 1974 Webster HK, Lancaster MC, Wease DF: The effect of primary bil e acid feeding on cholesterol metabolism and hepa tic function and morphology in the rhesus monkey (abstr). Gastroenterology 6fi:fi76, 1973 Cooper RA, Garcia FA, Trey C: The effect of lithocholic acid on red cell membranes in vivo . ,J Lab Clin Med 79:7-18, 1972 Hunt RD: Proliferation of bile ductu!es (the ductu!ar cell reaction) induced by lithocholic acid (abstr). Fed Proc 24:431, 196fi Has!ewood GAD: The biologica l significance of chemical differenc es in bile acids. Hiol Rev 39:537-!;74, 1964 Salen G, Tint GS, Eliav B, et al: In creased formation of ursodeoxycholic ac id in patients treated with chenodeoxycholic acid. J Clin Invest 5:J:G12- G24, 1974 Chen PS ,Jr, Toribara TY, Warner H: Microdetermination of phosphorus. Anal Chern 28:175G- 17fi8, 19fi6 Hutterer F: Recent progress in clinical enzymology fi>r the diagnosis of liver diseases . In The Liver and Its Diseases. Edited by F Schaffner, S Sherlock, CM Leevy. New York, Intercontinental Medical Book Corp, 1974, p 76- 84 Mericas G, Anagnostou E, Hadziyannis S, et a!: The di agnostic value of se rum leucine amino peptidase. J Clin Pathol 17:52-55, 1964 Schwartz M: Enzymes in Cancer. Clin Chern 19:10-22, 1973 Danielsson H, Johansson G: Effects of long-term feeding of chenodeoxycholic acid on biosynthesis and metabolism of bile acids in the rat. Gastroenterology 67:126-134, 1974 Einarsson K: On the formation of hyodeoxycholic acid in the rat. J Bioi Chern 241:534-539, 1966 Thomas PJ, Hsia SL, Matschiner JT, eta!: Bile acids . XIX. Metabolism of lithocholic acid-24-"C in the rat. J Bioi Chern 239:102-105, 1964 Palmer RH: Bile acids, liver injury and liver disease. Arch Intern Med 130:606-617, 1972 Cowen AE, Korman MG, Hofmann AF: Metabolis m of lithocholic acid in man (abstrl. Gastroenterology 67 :78fi, 197 4