bile acid metabolism in cirrhosis. viii. quantitative evaluation of bile acid synthesis from...

0270-9139/82/0201-0059$02.00/0 HEPATOLOGY Copyright 0 1982 by the American Association for the Study of Liver Diseases

Vol. 2, No. 1, p. 59, 1982 Printed in U.S.A.

Bile Acid Metabolism in Cirrhosis. VIII. Quantitative Evaluation of Bile Acid Synthesis

From [ 7~-3H]7a-Hydroxycholesterol and [ G-3H]26-Hydroxycholesterol

MARC GOLDMAN, Z. RENO VLAHCEVIC, CHARLES C. SCHWARTZ, JAN GUSTAFSSON, AND LEON SWELL

Division of Gastroenterology, Veterans Administration Medical Center, Medical College of Virginia, Richmond, Virginia 23249 and Departments of Pediatrics and Pharmaceutical

Biochemistry, University of Uppsala, Uppsala, Sweden

In order to evaluate more definitively the observed aberrations in the synthesis of cholic and chenodeoxycholic acids in patients with advanced cirrhosis, two bile acid biosynthesis pathways were examined by determining the efficiency of conversion of [3H]7a-hydroxycholesterol and r3H] 26-hydroxycholesterol to primary bile acids. Bile acid kinetics were determined by administration of ['4C]cholic and ['4C]chenodeoxycholic acids. Cholic acid synthesis in cirrhotic patients was markedly depressed (170 vs. 927 pmoles per day) while chenodeoxycholic acid synthesis was reduced to a much lesser degree (227 vs. 550 pmoles per day). The administration of [3H]7a-hydroxycholesterol allowed for an evaluation of the major pathway of bile acid synthesis via the 7a-hydroxylation of cholesterol. This compound was efficiently incorporated into primary bile acids by the two normal subjects (88 and 100%) and two cirrhotic patients (77 and 91%). However, the recovery of the label in cholic acid was slightly less in cirrhotic patients than in normal subjects. C3H]26- hydroxycholesterol was administered to ascertain the contribution of the 26-hydroxylation pathway to bile acid synthesis. All study subjects showed poor conversion (9 to 22%) of this intermediate into bile acids.

The results of this study suggest that a major block in the bile acid synthesis pathway in cirrhosis is at the level of 7a-hydroxylation of cholesterol (impairment of 7a-hydroxylase) and/or in the feedback triggering mechanism regulating bile acid synthesis. The data also suggest that the 26- hydroxylation pathway in normal subjects and patients with cirrhosis is a minor contributor to synthesis of the primary bile acids. Therefore, the relative sparing of chenodeoxycholic acid synthesis observed in cirrhotic patients is not due to preferential synthesis of this bile acid via the 26-hydroxylation pathway.

Patients with moderate to severe cirrhotic liver disease manifest a number of characteristic abnormalities of bile acid metabolism (1, 2). Cholic acid synthesis in these patients is markedly depressed and is 25 to 35% that of

Received April 3, 1981; accepted October 6, 1981. This work was supported in part by the Veterans Administration

and grants from the National Institutes of Health (AM,-14668 and AM- 23028) and the Swedish Medical Research Council (Project 03X-218).

The following systematic names are given to compounds referred to by trivial names: cholic acid, 3a,7a,12a-trihydroxy-5~-cholanoic acid; chenodeoxycholic acid, 3a,7a,12a-trihydroxy-5/3-cholanoic acid; 7a-hydroxycholesterol, 5-cholestene-3a,7a-diol; 26-hydroxycholesterol, 5- cholestene-3/3,26-diol.

Address reprint requests to: Z. R. Vlahcevic, M.D., Division of Gastroenterology (151), Veterans Administration Medical Center, 1201 Broad Rock Road, Richmond, Virginia 23249.

normal subjects; pool size and turnover rate of cholic acid are also significantly reduced. By contrast, chenodeoxycholic acid synthesis, pool size, and turnover rate are not altered to the same extent as those parameters of cholic acid metabolism; synthesis is reduced to a much lesser extent, and the pool size remains the same or is only slightly diminished. The reason for the relative sparing of chenodeoxycholic acid synthesis in these patients is not well understood. Secondary bile acid metabolism is also altered in cirrhotic patients as indicated by a virtual absence of deoxycholic acid from the bile (3).

Earlier studies (4, 5) attributed the decrease in cholic acid synthesis in patients with liver disease to a deficiency of hepatic microsomal 12a-hydroxylase enzyme. This enzyme is specific for cholic acid synthesis and, therefore, has been the most likely candidate responsible

59

60 GOLDMAN ET AL. HEPATOLOGY

for preferential reduction of formation of this bile acid. However, several lines of evidence indicate that the defect in bile acid production in these patients may be more complex and encompass multiple blocks in both cholic and chenodeoxycholic acid synthesis. The possibility of bile acid synthesis in man, being both qualitatively and quantitatively different from what has been perceived earlier, has emerged from recent in vivo studies (6-8) as indicated in Figure 1. Results of these studies suggest that cholic and chenodeoxycholic acid synthesis can occur via multiple routes from cholesterol. Bile acid biosynthesis takes place from cholesterol via intermediate formation of 7a-hydroxycholesterol, 26-hydroxycholesterol, and possibly via an uncharacterized pathway to cholic acid involving neither an initial 7a- nor 26-hydroxylation of cholesterol. Bifurcation routes to both primary bile acids occur at the level of cholesterol, 7a-hydroxyd-cholesten-3-one, 5P-cholestane 3a,7a-diol, and 5P-cholestane 3a,7a,26-triol (Figure 1). In addition, side chain oxidation in the cholic acid pathway may involve mitochondria1 26-hydroxylation or microsomal25-hydroxylation (9, 10).

These recent observations provided an impetus for further studies on the nature of the metabolic defect in bile acid synthesis in cirrhosis. In an initial study (ll), it was shown that reduction of cholic acid synthesis in cirrhosis could not be accounted for by a defect in the side chain oxidation of cholesterol since administered labeled 5P-cholestane 3a,7a,12a-triol and 5P-cholestane 3a,7a, 12a,26-tetrol were efficiently converted to cholic acid. Bile acid synthesis was also shown to proceed efficiently from 7a-hydroxy-4-cholesten-3-one, but less cholic acid than chenodeoxycholic acid was formed from this compound as compared to normal controls. These findings suggested that l2a-hydroxylation of precursors

of cholic acid may be in part responsible for marked reduction of cholic acid synthesis in these patients. How- ever, in light of simultaneous (although less marked) reduction of chenodeoxycholic acid synthesis, the ques- tion arose whether cirrhotic patients may have an additional defect in the bile acid pathway such as deficiency of 7a-hydroxylase enzyme which is common to synthesis of both bile acids and/or unavailability of cholesterol substrate for bile acid synthesis. Less marked impairment of chenodeoxycholic acid synthesis in these patients also raised the possibility that this bile acid may be synthesized via the alternative 26-hydroxylation pathway (Fig- ure 1, Pathway 2). This pathway is of minor importance in normal subjects (7), but probably plays an important role in various cholestatic syndromes (12). Also, in uiuo (13) and in uitro (14) studies suggest that the major end product of this pathway is chenodeoxycholic acid.

The present study was designed to provide the information on the relative contributions of 7a-hydroxycholesterol and 26-hydroxycholesterol pathways to the bile acid economy in patients with cirrhosis and compare them to normal subjects. Also, an attempt was made to define further the site of blockage of bile acid synthesis in these patients. The experiments provided information in patients with intact enterohepatic circulation on the quantitative conversion of several bile acid intermediates at the initial stages of two distinct pathways (7a-hydroxycholesterol and 26-hydroxycholesterol) from cholesterol to bile acids.

MATERIALS AND METHODS LABELED AND REFERENCE COMPOUNDS

[24-’4C]cholic and [24-’*C]chenodeoxycholic acids were obtained from New England Nuclear Corp., Boston,

26- Hydroxycholesterol < ~~-HYDKWIASE ( c H ~ ~ E S T E R O L I - - ~ - - - ~ - - ~ - - , - - ~ ~ ~ ~ 1 qci* 0 I -hydroxy/ise

7a- hydroxycholesterol t L

\ \ Pa-hydroxyfose

I ~ G J ‘LCENODEOX Y CHOL IC ACID]

26 hydroxyfme 7a, 26-dthydroxy-4-t 7~-hydroxy-4-cholesten-3-one

* 7a, 12a-dihydroxy-4-cholesten-3-0ne

f2a -hydroxyflose I 1

5p-cholestane-3a. 7a,12a trio1

A\-.....‘-. cholesten-3-one

2,: ~ 7a-d10l

\ 4 \

26-hydroxy/me dt0’.

\ 26” f2a -hyd?Oxy/USe

5,hholestane 3a.7~,26-trlol 3 58 - c holes tone 3% 7u, I2a, 26-te t rol

I 4

3a.7a-dihydroxy-5~-cholestanoic acid .L J. .1

1 CHENODEOXYCHOLIC ACID - -

4 3a,7a, 12e trihydroxy-5~-cholestanoic acid

.L 5.

FIG. I. Pathways of cholesterol breakdown in man. Major (7a-hydroxycholesterol) pathway (1); minor (26-hydroxycholesteroI) pathway (2); “bypass” pathway to cholic acid (3); alternative pathway to chenodeoxycholic acid (4). The (--->) represents the pathway with an unknown sequence of intermediates (3 and 4); (-+) represents the pathway in which intermediates are known.

Vol. 2, No. 1, 1982 BILE ACID METABOLISM IN CIRRHOSIS 61

Mass. and checked for purity by thin-layer chromatography. If the labeled bile acids were found to be less than 97% pure, they were purified by thin-layer chromatography on Silica Gel G using the solvent system of ethyl acetate:isooctane:acetic acid ( 5 5 1 v/v/v).

The [G-3H]cholest-5-ene-3fi,26-diol (26-hydroxycholesterol) was prepared from the methyl ester of generally tritium-labeled 3~-hydroxy-5-cholestenoic acid by reduction with lithium aluminum hydride. The 3P-hydroxy-5- cholestenoic acid was prepared by electrolysis in a man- ner analogous to the preparation of 3/?,7a-dihydroxy-5- cholestenoic acid as described earlier (15) and then ex- posed to tritium gas by the Wilzbach procedure. The material was purified by thin-layer chromatography using a solvent system of to1uene:ethyl acetate (1:l v/v) and consisted of equal parts of the R and S isomers. The specific activity was 400 pCi per pmole.

The [7fi-3H]7a-hydroxycholesterol was prepared by reduction of 7-ketocholesterol with [3H]sodium boro- hydride (16). The 7a- and 7p-isomers were separated by thin-layer chromatography on Silica Gel G (17). The plates were developed twice in diethyl ether. The specific activity of the labeled 7a-hydroxycholestero1 was 1 mCi per pmole.

PATIENTS Nine experiments were carried out on six patients.

Four were patients with advanced cirrhosis. One of the cirrhotics and both patients with no cirrhosis received both 3H intermediates on two separate occasions. Perti- nent information relevant to the patients is shown in Table 1. Informed consent was obtained from all patients prior to the study, The two patients without liver disease were hospitalized for minor gastrointestinal complaints. At the time of the experiments, these patients had no clinical or laboratory evidence of liver disease. The cirrhotic patients had advanced liver disease which was judged using criteria previously described by McCormick et al. (2). All medications were discontinued prior to the experiments. None of the cirrhotic patients received anti- biotics for a t least 2 months prior to the administration of the labeled intermediates.

TABLE 1. PERTINENT INFORMATION ON PATIENTS Dose" (pCi) Patient" Age 'H intermediate administered

No Cirrhosis

oc I 54 [G-3H]26-hydroxycholesterol 6.72 oc I1 54 [7P-:'H]7a-hydroxycholesterol 7.83 AM I 60 [7~-"H]7a-hydroxycholesterol 13.31 AM I1 60 [G-3H]26-hydroxycholesterol 8.05

Cirrhosis

GE 56 [7~-3H]7a-hydroxycholesterol 9.89 JD I 55 [G-3H]26-hydroxycholesterol 6.25

RE 47 ~G-~'H]26-hydroxycholesterol 5.78 RM 43 [G-3H]26-hydroxycholesterol 8.05

JD I1 55 [7~-3H]7a-hydroxycholestero1 5.73

All patients were males; patients OC, AM, and JD received the indicated compounds on two separate occasions. The lapsed time was 12, 75, and 107 days, respectively.

Each patient also received 5 pCi of [14C]cholic acid and 5 pCi of chenodeoxycholic acid.

TABLE 2. BILE ACID COMPOSITION OF PATIENTS 76 of total bile acids"

Patient Cholic Chenodeoxycholic Ueoxycholic

and othersh

No Cirrhosis

oc I 36.3 f 2.1 47.0 f 0.4 17.7 k 2.2 oc I1 41.2 f 1.2 36.4 f 0.5 22.4 k 0.9 AM I 53.3 f 1.0 37.3 f 0.9 9.4 f 1.5 AM I1 44.4 f 1.8 45.8 f 0.6 9.8 f 1.8

Cirrhosis

GE 39.2 f 0.9 59.6 f 1.1 1.2 f 0.2 JD I 25.0 f 0.3 73.3 f 0.3 1.7 f 0.2 JD I1 38.4 f 0.8 60.4 f 0.7 1.2 f 0.2 RE 44.8 f 1.9 53.0 f 1.9 2.2 f 0.4

1.2 f 0.3 RM 30.6 f 0.5

Represents the average of four to five determinations f S.E. on

Represents a small amount (2 to 3%) of other secondary bile acids.

68.2 f 0.8

duodenal bile obtained on 4 to 5 consecutive days.

Each patient received (in the morning) 5 pCi i.v. of ['4C]cholic and 5 pCi of ['4C]chenodeoxycholic acids plus 6 to 13 pCi of the 3H bile acid intermediate mixed in 20 ml of sterile human albumin. The albumin mixture was administered over a 2-min period, and the syringe was rinsed with 5 ml of albumin. The intermediates given to each patient are shown in Table 1. Patients OC, AM, and J D each received [3H]26-hydroxycholesterol and ["H]7a- hydroxycholesterol a t intervals between the compounds of 12, 75, 107 days, respectively. Sufficient time between the experiments had elapsed for the acid pools to turn over (Table 3) so that only small amounts of residual radioactivity were present in the bile from the first ex- periment prior to the administration of the second intermediate.

Duodenal bile (5 ml) was collected on each patient via an Anderson tube after injection of Kinevac (E. R. Squibb and Sons, Princeton, N.J.) for 4 to 5 consecutive days.

METHODS Duodenal bile samples were extracted with 2:l chlo-

roform methanol (v/v) and washed with 0.2 volumes of water (18). The chloroform phase had less than 2% of either 3H or 14C activity. The water-methanol phase contained principally labeled bile acids plus small amounts (4%) of other radioactive products. Bile acid analysis was carried out as described earlier (19). Aliquots of the water-methanol phase were autoclaved for 3 hr with 10% KOH. The extract was brought to a pH of 1 to 2 with HC1 and the free bile acids extracted with diethyl ether. Bile acid methyl esters were prepared with diazo- methane and an aliquot of the methyl esters were subjected to gas-liquid chromatography to quantitate the individual bile acids; 5a-cholestane was used as an inter- nal standard. The stationary phase was 3% HIEFF 8BP (Applied Science Labs., State College, Pa.). Another aliquot of the methyl ester fraction was subjected to thin- layer chromatography on Silica Gel G and developed twice with ethyl acetate:isooctane:acetic acid ( 5 5 : 1 v/v/v). Methyl cholate and chenodeoxycholate were scraped from the plate and eluted with ethano1:chloro-


form:water:acetic acid (10032:2O:2 v/v/v/v). The mass of the bile acid methyl esters was determined by gas-liquid chromatography and radioactivity by liquid scintillation counting. Quench correction was applied by the external standard method (Mark 111, Tracor Analytic, Silver Spring, Md.). The identity of the isolated labeled cholic and chenodeoxycholic acids derived from the 3H intermediates was evaluated by migration on thin-layer plates, and homogeneity of mass and label as judged by no significant change in specific activity upon repeated crys- tallization of the methyl esters and also by their elutron pattern on silicic acid columns (20).

CALCULATIONS AND INTERPRETATIONS OF DATA Cholic and chenodeoxycholic acid synthesis and turn-

over rates were determined by the isotope dilution technique (19, 21) following the simultaneous administration of [ ''C]cholic and [ ''C]chenodeoxycholic acids. The va- lidity of the isotope dilution method for determining bile acid kinetics in man has been previously discussed (19, 22).

The fractional conversion of the administered [3H]7a- hydroxycholesterol and ["H]26-hydroxycholesterol to cholic and chenodeoxycholic acids was estimated from the area under specific activity curves, the amount of injected label, and the synthesis of the bile acids derived from the 14C isotope kinetic data (11).

The principal criteria for evaluating the quantitative significance of the 3H intermediates on the pathway to bile acids were the efficiency of their conversion to primary bile acids, the ratios of cholic to chenodeoxycholic acids by endogenous synthesis (determined by isotope dilution technique), and the incorporation of 3H label into these primary bile acids. Comparisons of the ratios by mass and radioactivity were indicative of whether the synthesized bile acids passed entirely through the administered intermediate. If the ratio by synthesis and 3H label were equal or similar, then it should be assumed that all of the synthesized cholic and chenodeoxycholic acids passed through the administered 3H intermediate. For example, if cholic and chenodeoxycholic acids were

synthesized solely via 7a-hydroxycholylation pathway, then after administration of labeled 7a-hydroxycholesterol, the ratios of synthesis (by isotope dilution) and 3H label in these primary bile acids should be identical. If the ratios of cholic to chenodeoxycholic acids (by synthesis and by the 3H label) were unequal then it must be assumed that synthesis of one of the primary bile acids in part bypassed the administered intermediate. This implies that the bile acid (either cholic or chenodeoxycholic) with a higher ratio of cholic to chenodeoxycholic acids by synthesis than label must have been diluted with unlabeled bile acid from an alternate precursor source proximal to the administered intermediate. An alternative explanation is that i.v. administered 7a-hydroxycholesterol might have been preferentially metab- olized into chenodeoxycholic acid. In the rat, however, the specific activities of cholic and chenodeoxycholic acids after i.v. administration of labeled 7a-hydroxycholesterol were identical (23).

In vitro and in vivo approaches to the study of bile acid pathways have problems in interpretation. Demon- stration of a reaction in vitro does not necessarily mean that it occurs in vivo. Conversely, the in vivo approach assumes that the i.v. administered bile acid intermediate upon entering the liver cell mixes homogenously with the appropriate endogenous precursor in the appropriate or- ganelle. This assumption cannot be validated directly, but it is the only approach available in man. It is important to emphasize that there is a good correlation between in vivo and in vitro studies in man and rat (24). Nevertheless, results of in vivo and in vitro studies of bile acid pathways have to be interpreted with caution.

RESULTS

Bile acid composition data are presented in Table 2. The bile of the four cirrhotic patients was found to be virtually devoid of deoxycholic acid. In the patients without cirrhosis, the percentages of cholic and chenodeoxycholic acids were on the average close to unity. The ratio of cholic to chenodeoxycholic acid in the cirrhotic pa-

TABLE 3. KINETICS AND SYNTHESIS OF PRIMARY BILE ACIDS Cholic Chenodeoxycholic Correlation coefficient

Cholic Chenodeoxycholic* Synthesis K Synthesis K Patient"

(pmoles/day) (days-') (pmoles/day) (days-')

No Cirrhosis

oc I 808 0.552 534 0.292 0.99 0.99 oc I1 998 0.348 599 0.270 0.99 0.99

AM I1 884 0.454 529 0.336 0.98 0.96 AM I 1,019 0.356 538 0.334 0.99 0.98

Cirrhosis

GE 216 0.113 286 0.121 0.98 0.98 JD I 157 0.165 294 0.120 0.99 0.99 J D I1 250 0.285 245 0.182 0.99 0.99 RE 205 0.106 243 0.129 0.99 0.98 IZM 122 0.106 317 0.165 0.99 0.99

Bile acid isotope dilution kinetics were carried out on patients OC, AM, and JD on two separate occasions; time between the first and second

Represents the least-squares regression fit between time in days vs. log-specific activity of cholic and chenodeoxycholic acids; p values were determinations was 12, 75, and 107 days, respectively.

all significant 0.01 to 0.05.


tients was <1 which is a reflection of the higher percent- age of chenodeoxycholic acid in these patients.

Several representative ['HI- and ['4C]cholic and chenodeoxycholic acid specific activity curves for patients with and without liver disease are shown in Figure 2. The decay of the bile acids was log-linear with time (in all experiments) which is indicative of first order kinetics.

Table 3 shows the computer fit slopes (K) of the regression line and the calculated daily synthesis rates for both primary bile acids derived from the I4C isotope dilution data. In patients with cirrhosis, the fractional turnover rate of cholic acid was considerably slower than in patients without cirrhosis. The cholic acid pool in patients with no cirrhosis turned over in 2 to 3 days and approximately 6 to 9 days in the cirrhotic patients. Che- nodeoxycholic acid turnover was also suppressed in cirrhotic patients by about 40 to 50%. Cholic acid synthesis in the two subjects without cirrhosis averaged 927 pmoles or 261 mg per day which is within the range of normal subjects (17). By contrast, all four patients with cirrhosis had markedly depressed cholic acid synthesis rates; the average was 190 pmoles or 76 mg per day. Thus, the cirrhotic patients synthesized about '/s as much cholic acid as the patients without cirrhosis. Chenodeoxycholic acid synthesis was also reduced in patients with liver disease; the average in the patients without and with cirrhosis was 500 pmoles or 220 mg per day and 227 p o l e s or 111 mg per day, respectively. The decrease in chenodeoxycholic acid synthesis exhibited by the cirrhotic patients was about 50%, but this reduction was considerably less than was noted for cholic acid.

10.000

5000

2000 w J 0 I 3 1 1000 E n

500

200

1 i i b 4 DAYS

Bile acid kinetic data obtained from the ['H]cholic and [3H]chenodeoxycholic acid specific activity decay curves are shown in Table 4. The observed fractional turnover rate constants for cholic and chenodeoxycholic acids were similar to those observed with administered ["C] cholic and ['4C]chenodeoxycholic (Table 3).

The efficiency of conversion of the two 'H intermediates to primary bile acids is shown in Table 5. The ["HI 26-hydroxycholesterol compound was poorly converted to primary bile acids in both groups of patients; the range of conversion was 9 to 22%. Comparison of the ratios of cholic to chenodeoxycholic acids by synthesis and 'H label indicates that this intermediate favored chenodeoxycholic acid in both subjects without cirrhosis and in 2 of the 3 patients with liver disease. The ['H]7a-hydroxycholesterol was efficiently converted to bile acids in both groups of patients; the conversion ranged from 77 to 100%. In the two patients without cirrhosis, about equal amounts of labeled cholic and chenodeoxycholic acids were formed from this precursor. Comparison of the synthesis and the 'H label ratios indicate that ['HI 7a-hydroxycholesterol favored chenodeoxycholic over cholic acid by about 1.8/1. Similar comparisons in the two cirrhotic patients showed that chenodeoxycholic acid was also preferred over cholic acid by about 1.5/1. It is noteworthy that even though cholic acid synthesis was considerably more depressed in the cirrhotic patients than in chenodeoxycholic acid synthesis (Table 3), the incorporation of [3H]7a-hydroxycholesterol into cholic acid was reduced to only a small extent.

The inequality of the ratios of cholic to chenodeoxy-

1 "

I I I I 1 1 2 3 4 5

DAYS FIG. 2. Representative log plot of specific activity of cholic acid (0, 0) and chenodeoxycholic acid (A, A) of two typical patients without

cirrhosis (A) and two patients with cirrhosis (B); ['H]cholic acid (0); ['4C]cholic acid (0); [3H]chenodeoxycholic acid (A); and ['4C]chenodeoxycholic acid (A).


TABLE 4. KINETICS OF PRIMARY BILE ACIDS DERIVED FROM ADMINISTERED 'H INTERMEDIATES" ~~~ ~

Cholic Chenodeoxycholic 'H intermediate

administered yo Intercept K Correlation yo Intercept K Correlation (DPM/Wmole) (days-') coefficient (DPM/mnole) (days-') coefficient

Patient

No Cirrhosis oc I [G-3H]26-hydroxycholesterol 938 0.534 0.99 860 0.252 0.96

AM I [7/3-3H]7a-hydroxycholesterol 5,213 0.343 0.98 8,280 0.315 0.96 AM 11 [ G-3H]26-hydroxycholesterol 593 0.411 0.97 1,233 0.326 0.97

oc I1 [ 7/3-3H]7a-hydroxycholesterol 2,481 0.342 0.99 3,377 0.251 0.99

Cirrhosis GE [ 7/3-3H]7a-hydroxycholesterol 3,800 0.132 0.99 4,522 0.122 0.99

JD I1 [ 7~-3H]7a-hydroxycholesterol 5,199 0.300 0.98 4,222 0.142 0.99

RM [G-3H]26-hydroxycholesterol 549 0.121 0.97 663 0.185 0.99

JD I [G-3H]26-hydroxycholesterol 882 0.171 0.99 789 0.134 0.95

RE [G-~'H]26-hydroxycholesterol 158 0.156 0.99 539 0.144 0.96

~

Represents the values derived from a least-squares regression fit of time in days vs. log-specific activity of cholic and chenodeoxycholic acids; p values were all significant at the level of 0.01 to 0.05. Number of specific activity values per patient was 4 to 5.

TABLE 5. CONVERSION OF 3H INTERMEDIATES TO PRIMARY BILE ACIDS % of administered dose recovered as: Cholic/Chenodeoxycholic "H intermediate

administered Cholic Chenodeoxycholic Total Synthesis :IH ,'H/Synthesis Patient

oc I oc I1 AM I AM I1

GE JD I JD I1 RE RM

[G-3H]26-hydroxycholesterol [ 7/3-3HH]7a-hydroxycholesterol [ 7/3-3H]7a-hydroxycholesterol [G-3H]26-hydroxycholesterol

[7~-3H]7a-hydroxycholesterol [G-3H]26-hydroxycholesterol [ 7~-3H]7a-hydroxycholesterol [G-3H]26-hydroxycholesterol rG-3H126-hydroxycholesterol

No Cirrhosis 9.5 12.2 41.6 46.3 52.4 47.9 7.1 11.2

Cirrhosis

28.3 48.3 5.8 12.4 34.0 57.1 1.6 7.7 3.1 6.4

~~

21.7 87.9 100.32 18.3

76.6 18.2 91.1 9.3 9.5

~~

1.51 1.67 1.89 1.67

0.76 0.53 1.02 0.84 0.38

~

0.78 0.90 1.09 0.63

0.59 0.47 0.60 0.21 0.48

-~

0.50 0.54 0.58 0.38

0.78 0.89 0.59 0.25 1.26

cholic acids by synthesis and 'H label probably reflects the presence of additional alternate pathways to bile acids. In patients with and without cirrhosis, a considerable amount of cholic acid was synthesized which did not appear to pass through 7a-hydroxycholesterol since in each of the four patients, the ratio of cholic to chenodeoxycholic acids by synthesis was greater than the ratio of those bile acids by 3H label.

DISCUSSION Patients with cirrhosis and an intact enterohepatic

circuit synthesize 50 to 100 mg per day of cholic acid or only 20 to 25% of that observed in normal subjects. Chenodeoxycholic acid synthesis is also depressed in these patients but only by about 50% as compared to normal subjects. The data of the present study suggest that a deficiency of 12a-hydroxylase cannot entirely account for the markedly depressed cholic acid synthesis in patients with liver disease for the following reasons: (a) the reduction in chenodeoxycholic acid synthesis in these patients implies that other portions of the bile acid pathway common to the synthesis of both bile acids must also be defective, and (b) the overall conversion of [3H] 7a-hydroxycholesterol to cholic and chenodeoxycholic acids in two patients with severe cirrhosis was similar to that observed in normal subjects. The incorporation of

H label into cholic acid appeared to be somewhat depressed probably as a result of a defect in 12a-hydroxylation step. If l2a-hydroxylase activity was the major contributing factor responsible for the depression of cholic acid synthesis, there should have been only a very small amount of labeled cholic acid formed from 7a- hydroxycholesterol to account for the marked differences between the synthesis of the primary bile acids.

Based on the information obtained in this report, it seems plausible that a major block in bile acid synthesis in patients with liver disease resides at one or more of the following levels in the bile acid pathway: (a) the 7a- hydroxylation of cholesterol; (b) an alternate pathway from cholesterol; (c) the unavailability of cholesterol substrate for bile acid synthesis, and (d) a defect in the feedback regulation of bile acid synthesis. In view of the fact that severe cirrhotics can increase their bile acid synthesis considerably when administered cholestyramine (25) or in response to complete biliary diversion (26), the lack of cholesterol substrate is probably not the chief causative factor responsible for the impairment of bile acid synthesis. Alternate pathways to primary bile acids may be of considerable significance in man, but the major portion (60 to 70%) of the bile acids appears to be synthesized via 7a-hydroxycholesterol. Several recent re- ports (6, 7) have presented evidence which suggests that

3

man may synthesize considerable amounts of cholic acid via a pathway from cholesterol which does not involve an initial 7a-hydroxylation. The results of the present study suggest that this alternative pathway is also present in normal subjects with an intact enterohepatic circuit and in patients with advanced cirrhosis.

The existence of a bile acid pathway initiated by the 26-hydroxylation of cholesterol has been demonstrated in man (13), and it has been suggested that this pathway may assume an important quantitative role in certain types of cholestatic liver disease (12). Specifically, 3p- hydroxy-5-cholenoic acid, a known intermediate of the 26-hydroxylation pathway, has been found in large amounts in the urine of children with biliary atresia (27). If the 26-hydroxylation pathway was also more active relative to the 7a-hydroxylation pathway in patients with cirrhosis, this could account for the observation that chenodeoxycholic acid synthesis is impaired to a much lesser extent than cholic acid synthesis. When 26-hydroxycholesterol was tested for its ability to form bile acids in bile fistula patients, it was found to be poorly converted to bile acids; it favored chenodeoxycholic over cholic acid (7). Similar findings were made in the present study in patients with and without liver disease and in an intact enterohepatic circuit. However, the observed value for the efficiency of conversion of the administered generally labeled 26-hydroxycholesterol is probably subject to a small error since an unknown amount of radioactivity was lost during cleavage of the three carbons from the side chain. It should also be pointed out that a mixture of R and S isomers of 26-hydroxycholesterol were administered. Selective utilization of one of the isomers is a possibility, since a recent report (28) indicates that 26- hydroxylation of 5P-cholestane 3a,7a-di01 leads to the formation of predominantly 25-R,5P-cholestane 3a,7a,26- triol. However, recently we administered labeled R and S 3a,7a,12a-trihydroxy-5~-cholestanoic acid to bile fistula patients and found equal incorporation of these isomers into cholic acid (29).

Circumstantial evidence suggests that failure to invoke the feedback mechanism regulating 7a-hydroxylase activity in the microsomes could be the prime factor responsible for the depressed bile synthesis in cirrhosis. Recent observations (26) in a severe cirrhotic patient with a bile fistula indicate that the synthetic feedback mechanism could only be invoked to produce a large increase in cholic acid synthesis when the enterohepatic circuit was totally disrupted.

In conclusion, it appears that the impairment of bile acid synthesis in patients with cirrhosis is probably the result of defects in both the feedback mechanism triggering synthesis and in the enzymatic steps in the pathways. The relative sparing of chenodeoxycholic acid synthesis in these patients is not due to compensatory alternative 26-hydroxylation pathway which preferentially synthes- izes chenodeoxycholic acid. In this respect, patients with parenchymal liver cell disease may differ from those with cholestasis. It is unlikely that the reduction in effective parenchymal liver cell mass is a sole factor responsible for diminished bile acid synthesis in cirrhosis; if this were the case. svnthesis of both bile acids should have been


I "

effected proportionally. J Dig Dis 1978; 23:1115-1120.

REFERENCES 1. Vlahcevic ZR, Juttijudata P, Bell CC, Jr, et al. Bile acid metabolism

in patients with cirrhosis. 11. Cholic and chenodeoxycholic acid metabolism. Gastroenterology 1972; 62:1174-1181.

2. McCormick WC, 111, Bell CC, Jr., Swell L, et al. Cholic acid synthesis as an index of the severity of liver disease in man. Gut

3. Yoshida T, McCormick WC, 111, Swell L, et al. Bile acid metabolism in cirrhosis. IV. Characterization of the abnormality in deoxycholic acid metabolism. Gastroenterology 1975; 68335-341.

4. Carey JB, Jr. The serum trihydroxy-dihydroxy bile acid ratio in liver and biliary tract disease. J Clin Invest 1958 371494-1503.

5. Vlahcevic ZR, Prugh MF, Gregory DH, et al. Disturbances of bile acid metabolism in parenchymal liver cell disease. Clinic in Gastro- enterology 1977; 625-43.

6. Vlahcevic ZR, Schwartz CC, Gustafsson J, et al. Biosynthesis of bile acids in man. Multiple pathways to cholic acid and chenodeoxycholic acid. J Biol Chem 1980; 2552925-2933.

7. Swell L, Gustafsson J, Schwartz CC, et al. An in vivo evaluation of the quantitative significance of several potential pathways to cholic and chenodeoxycholic acids from cholesterol in man. J Lipid Res

1973; 14~895-902.

1980; 21:455-466. 8. Swell L. Schwartz CC. Gustafsson J. et al. A quantitative evaluation

of the conversion of '25-hydroxycholesterof to bile acids in man. Biochim Biophys Acta 1981; 663163-168.

9. Salen G, Shefer S, Setoguchi T, et al. Bile alcohol metabolism in man. Conversion of 5,&cholestane 3a,7a,12~,-25-tetrol to cholic acid. J Clin Invest 1975; 56226-231.

10. Bjorkhem I, Gustafsson J, Johansson G, et al. Biosynthesis of bile acids in man. Hydroxylation of the Cz,-steroid side chain. J Clin Invest 1975; 55:478-486.

11. Patteson TE, 111, Vlahcevic ZR, Schwartz CC, et al. Bile acid metabolism in cirrhosis. VI. Sites of blockage in the bile acid pathway to primary bile acids. Gastroenterology 1980; 79620-628.

12. Goldman MA, Schwartz CC, Swell L, et al. Bile acid metabolism in health and disease. In: Fopper H, Schaffner F, eds. Progress in liver diseases, Vol. VI. New York Grune and Stratton, 1979225-241.

13. Anderson KE, Kok E, Javitt NB. Bile acid synthesis in man: metabolism of 7a-hydroxycholesterol I4C and 26-hydroxycholesterol. J Clin Invest 1972; 51:112-117.

14. Mitropoulous KA, Myant NB. The formation of lithocholic acid, chenodeoxycholic acid and alpha- and beta-muricholic acids from cholesterol incubated with rat liver metochondria. Biochem J 1967;

15. Danielsson H. On the metabolism of 7a-hydroxycholesterol in mouse liver monogenates. Ark Kemi 1961; 17:363-372.

16. Molinow MR, McLaughlin P, Popworth L, et al. A new method for determining liver microsomal cholesterol 7a-hydroxylase. Steroids

17. Dayal B, Batta AK, Shefer S , et al. Preparation of 24(R) and 246) 5P-cholestane 3a,7a,24-triols and 25(R) and 25(S) cholestane 3a,7a,26-triols by a hydroboration procedure. J Lipid Res 1978; 1 9

18. Folch J, Lees M, Sloan-Stanley GH. A simple method for the isolation and purification of total lipids from animal tissues. J Biol Chem 1957; 226:497-509.

19. Vlahcevic ZR, Miller JR, Farrar JT, et al. Kinetics and pool size of primary bile acids in man. Gastroenterology 1971; 61:85-90.

20. Schwartz CC, Cohen BI, Vlahcevic ZR, et al. Quantitative aspects of the conversion of 5P-cholestane intermediates to bile acids in man. J Biol Chem 1976; 251:6308-6314.

21. Lindstedt S. Turnover of cholic acid in man. Bile Acids and Steroids Acta Physiol Scand 1957; 401-9.

22. Vlahcevic ZR, Gregory DH, Swell L. Validation of isotope dilution technique for determination of bile acid synthesis in man. Gastro- enterology 1974; 67:1299-1300.

23. Shefer S, Hauser S, Bekersky I, et al. Biochemical site of regulation of bile acid biosynthesis in the rat. J Lipid Res 1970; 11:404-411.

103:472-479.

1975; 25663-672.

191-196.

24. Danielsson H, Sjovall J. Bile acid metabolism. Ann Rev Biochem

25. Angelin B, Einarsson K, Hellstrom K. Effect of cholestyramine on bile acid kinetics in patients with portal cirrhosis of the liver. Evidence of a selective defect in the formation of cholic acid. Am

1975; 44:233-253.

GOLDMAN ET AL. HEPATOLOGY 66

26. Vlahcevic ZR, Goldman M, Schwartz CC, et al. Bile acid metabo- cholic acid in man. Stereospecific side-chain hydroxylation of 5p- lism in cirrhosis. VII. Evidence for defective feedback control of cholestane 3a,7a-diol. J Clin Invest 1978; 62539-545. bile acid synthesis. Hepatology 1981; 1:146-150. 29. Swell L, Gustafsson J, Danielsson H, et al. Biosynthesis of bile

27. Makino I, Sjovall J, Norman A, et al. Excretion of 3P-hydroxy-5- acids in man. An in uiuo evaluation of the conversion of R and S cholenoic and 3-hydroxy-5a-cholanoic acids in the urine of infants 3a,7a,l2a-trihydroxy-5/3-cholestanoic and 3a,7a,12a-24r-tetrahy- with biliary atresia. FEBS Lett 1971; 15:161. droxy-5p-cholestanoic acids to cholic acid. J Biol Chem 1981; 256:

28. Shefer S, Cheng FW, Batta AK, et al. Biosynthesis of chenodeoxy- 912-916.

bile acid metabolism in cirrhosis. viii. quantitative evaluation of bile acid synthesis from...

Documents