Effect of Complete Sulfation of Bile Acids on Bile Formation: Role of Conjugation and Number of Sulfate Groups IBRAHIM YOUSEF, DIANEMIGNAULT AND BEATRIZ TUCHWEBER Departmen.ts of Pharmacology and Nutrition, University of Montreal and Centre de Recherche Pediatrique, H6pital Sainte-Justine, Montreal, Canada H3C 3J7

The effect of complete sulfation of conjugated cholic, tors (1-41, but little information is available on their sulchenodeoxycholic and deoxycholic acids on bile for- fate esters, despite their apparent importance in some mation was investigated in rats. The sulfated bile acids cholestatic liver diseases (5-11).In systematic experiwere infused intravenously in stepwise increasing ments undertaken to characterize these compounds, we doses (1, 2, 3 and 4 pmol/min/100 gm body wt) in rats examined the effect of complete sulfation of unconjuafter 90 min of bile acid pool depletion. The effects of gated cholic acid (CA), chenodeoxycholic acid (CDCA), these bile acids on bile flow, bile salt, biliary phospholipid and cholesterol secretion rates were determined. deoxycholic acid (DOCA) and lithocholic acid (LCA) on In addition, their choleretic activity and their effect on bile formation (12). It was noted that the secretion of biliary lipid secretion were calculated. Appropriate these compounds was slower than that of nonsulfated controls infused with nonsulfated bile were also per- bile acids, demonstrating transport maximum characterformed. The sulfated bile acids increased bile flow with istics rather than secretory rate maximum (SRm) charincreasing the infusion doses, and the maximum bile acteristics of nonsulfated bile acids. Sulfation also inflow was significantly higher than nonsulfated bile creased bile salt-independent bile flow and the choleacids, Although cholestasis was developed during the retic activity of bile acids and significantlyprevented the infusion of nonsulfated bile acids, no cholestatic effect secretion of biliary phospholipid and cholesterol. This is was observed for sulfated bile acids. With the exception in accordance with the physico-chemical properties of of cholic acid, sulfation significantly increased the bile acid secretory rate maximum. The sulfates of cheno- these bile acids, as predicted by the studies of Armstrong deoxycholic and deoxycholic acids were further hy- and Carey (13). However, a recent investigation by Stedroxylated. The choleretic activities for all the sulfated vens et al. (14) showed that tauro 3 a-sulfate CDCA or 7 bile acids were significantly higher than the nonsul- a-sulfate CDCA infusion at 2.8 Fmo11100 gm body wtihr fated bile acids. All the sulfated bile acids significantly in hamsters resulted in a secretion rate and choleretic reduced the biliary lipid secretion, and a significant activity that were similar to those of taurochenodeoxycorrelation was found between the choleretic activity cholic acid (TCDCA) (17.0, 24.3 and 25.4 cl,l/pmol for a n d the phospholipid-dependent bile acid secretion. TCDCA, 3 a-sulfate TCDCA and 7 a-sulfate TCDCA, The data also showed that infusion of sulfated taurine- respectively). Furthermore, sulfated conjugated CDCA conjugated bile acids produced higher bile flow and significantly reduced biliary phospholipid secretion and bile acid secretion rate and was less effective when biliary lipid secretion rates were reduced compared had no significant effect on biliary cholesterol secretion. with glycine conjugates. It is concluded that sulfated The discrepancy between the results obtained by conjugated bile acids may have a role in protection Stevens et al. and our results may be attributed to difduring cholestasis either by stimulation of bile flow or ferent bile acid infusion and secretion rates. The objectiveof these experiments was to characterize by reduction of biliary lipid secretion, thus protecting cell membranes from the detergent properties of high the influence of various sulfated conjugated bile acids on concentrations of nonsulfated bile acids. (HEPATOLOGY bile formation in a manner similar to our previous study 1992;15:438-445.)

using completely sulfated conjugated bile acids. Although complete sulfation may not be directly relevant The physiological properties of common bile acids and t o an understanding of the role of these bile acids in their conjugates have been studied by many investiga- human cholestasis, it is necessary in comprehending the effect of conjugation and the position of the sulfate group in the bile acid molecule. Received March 5, 1991; accepted September 16, 1991. This work was supported by a grant from the Medical Research Council of Canada. Address reprint requests to: Dr. Ibrahim Yousef, Department of Pharmacology, University of Montreal, C.P. 6128, Succ. A, Montreal, Quebec, Canada H3C 357.

31l1134453

MATERIALS AND METHODS Preparation of Sulfated Bile Acids. The sodium salts of taurocholic acid (TCA), TCDCA, taurodeoxycholic acid (TDOCA), glycocholic acid (GCA), glycochenodeoxycholic acid (GCDCA) and glycodeoxycholic acid (GDOCA), purchased

438

Vol. 15, No. 3, 1992

439

SULFATION AND BILE FORMATION

from Calbiochem (La Jolla, CA), were purified by extraction lighting (7 PM to AM darkness) with free access to food and with diethyl ether to remove unconjugated bile acids and by water before the experiment. They were anesthetized with repeated crystallization with methanol. No contaminants were pentobarbital (48 mg/gm body wt intraperitoneally), and a detectable by thin layer chromatography (TLC) using large polyethylene No. 50 catheter was inserted into the right samples and a solvent system of butanol, acetic acid and water internal jugular vein. The bile duct was cannulated with a ( l O : l : l , vol/vol/vol). The sulfate esters of bile acids were polyethylene No. 10 catheter. Body temperature, measured prepared by mixing the bile acids with a 3 to 4 molar excess of with a rectal thermometer, was maintained at 37” C sulfur trioxide trimethylamine (Aldrich, Milwaukee, WI), throughout the experiment using a rectal probe and thermodepending on the number of the hydroxyl group, in a mixture statically controlled infrared lamp. The length of the catheter of benzene, pyridine and acetic acid (10: 1: 1,vol/vol/vol).The was adjusted to 20 cm from the cannulation point. Bile output final sulfur trioxide concentration was 7% (wtivol). The was collected for 90 min after cannulation in successive 30-min mixture was refluxed for 3 to 4 hr, then cooled to room periods. The animals were then infused with 2 ml of 3.5% temperature; excess sulfur trioxide was removed by the albumin in 0.9% NaCl for 30 min, during which time bile was addition of an equal volume of 0.5 N NaOH in methanol. The collected in 10-min batches to assess basal bile flow (BF), bile sulfated bile acids were precipitated or crystallized by the salt and biliary lipid secretion before bile acid infusion. The bile addition of excess volume of cold diethyl ether, dried under acids were infused in four stepwise increasing doses. Each dose vacuum and dissolved again in NaOH-methanolto remove any was given in 2 ml of 3.5%albumin solution in 0.9% NaCl for further excess of sulfur trioxide. This process was repeated 30 min and bile was collected in 10-min batches. Preliminary three times. Finally, the crystallized bile acids were washed experiments were performed to establish the infusion dose of twice with excess volume of cold diethyl ether, dried under each bile acid that produces maximum bile flow and then vacuum, dissolved in distilled water and freeze-dried. The induces cholestasis. These doses were TCA = 1,1.5,2and 2.5; purity of the product was checked by TLC, using the GCA = 0.5, 1, 1.5 and 2; TCDCA and GCDCA 0.5, 1, 1.5 and above-mentioned solvent system. In all cases, one spot was 2; TDOCA and GDOCA = 0.2,0.4,0.6 and 1kmol/min/100 gm seen with a chromatographic mobility significantly less than body wt. The sulfated bile acids did not induce cholestasis at the parent compound. The bile acids did not react with the 3 the doses mentioned above, thus the following higher infusion a-hydroxy dehydrogenase. Purity was further checked by the doses were selected: 1, 2, 3 and 4 ~mol/min/100gm body wt. Calculations and Statistical Analysis. The data are pretitration of 20 kmol/ml in NaOH solution with 1 N HC1 and gave the equivalence of 96 for sulfated GCA, 101 for sulfated sented as means S.D. Differences between means were GCDCA and 96 for sulfated GDOCA. However, no equivalence evaluated by ANOVA and p < 0.05 was considered to be could be obtained for sulfated taurine-conjugated bile acids significant. The choleretic activity of bile acids was determined because they did not show precipitation at any pH. These from the slope of the regression line of the correlation between titration studies indicated that the purity of sulfated glyco-bile the bile salt secretion rate (BSSR) and bile flow, using values acids is greater than 96% (15). After solvolysis, the bile acids obtained until the SRm of the bile acids or their sulfate esters showed similar to that of the parent compound and reacted was reached. The effects of the bile acids or their sulfate esters with the 3 a-hydroxy dehydrogenase. Furthermore, solvolysis on phospholipid and cholesterol secretion were assessed from was complete because no sulfated bile acids could be detected the slope of the regression line of the correlation between the BSSR and phospholipid or cholesterol secretion rate until the by TLC. Bile Analysis. Bile volume was determined gravimetrically, SRm of the bile acids. The SRm was calculated by taking the assuming a density of 1gm/ml of bile. Bile acids were measured average of the highest three secretion rates of each bile acid enzymatically, using 3 a-hydroxysteroid dehydrogenase before (21-23). The data on choleretic activity were expressed as solvolysis (nonsulfatedbile acids) and after solvolysis (total bile microliters per micromole of bile acid and as nanomoles per acids) (16). Total lipids were extracted by the method of Bligh micromole of bile acid for phospholipid and cholesterol. and Dyer (17), and phospholipid was determined by the Bartlett technique after digestion in perchloric acid (18). RESULTS Cholesterol was measured by the cholesterol oxidase method (Boehringer-Mannheim, Indianapolis, IN). PhosphatidylSulfation of Bile Acids. The method described for the choline was determined by a choline oxidase method (Wako preparation of sulfated bile acids produces complete Pure Chemical Industries, Osaka, Japan) (19). Biliary bile sulfation of all available hydroxyl groups. The amount of acids were assessed by TLC and gas chromatography-mass sulfur trioxide varies between trihydroxy and dihydroxy spectrometry (GC-MS) after solvolysis, hydrolysis, methylation and acetylation, using a Hewlett-Packard 5890 Gas bile acids, depending on the number of hydroxyl groups Chromatograph (Hewlett-Packard, Waltham, MA) equipped available for sulfation. The time of the reaction is with a 12-meter fused silica capillary column cross-linked with important, because a preferential sulfation of the dif5% phenyl-methyl silicone having an internal diameter of 0.22 ferent hydroxyl groups depending on their position on mm (20).The gas chromatograph was connected to a Hewlett- the molecule appears to exist. The reaction is followed in Packard 5971A Mass Selective Detector (MSD), and bile acids time on the TLC system until only one spot is observed, were scanned between 50 and 650 ion mass. The spectra of the after which purity is checked as described in the bile acids were recorded by using MSD Chemstation software “Materials and Methods” section. supplied by Hewlett-Packard. The bile acids were identified CA Infisions. Figure 1 summarizes the effect of with a library of spectra obtained from known standards of bile infusion of TCA, GCA, S-TCA and S-GCA on BF, the acids that match spectra, specific ions and retention time (more details on the GC-MS method and identification of bile BSSR, biliary phospholipid secretion rate (BPLSR) and cholesterol secretion rates (BCHSR). BF (Fig. 1A) acids will be published elsewhere). Animals. Male Sprague-Dawley rats weighing 200 to 250 gm increased with escalating infusion doses of the bile acids, (Charles River. Saint-Constant, Quebec, Canada) were main- reaching a maximum (average of the highest three tained in an environment of constant temperature (22” C) and consecutive points) of 3.42 k 0.82, 3.71 2 0.30,

*

440

YOUSEF, MIGNAULT AND TUCHWEBER

HEPATOLOGY

BILE SALT SECRETION RATE 400

albumin

0.5

1.0

2.0

1.5

2.5

3.0

4.0

urnole/ min/ 100 g body weight

CHOLESTEROL SECRETION RATE

3.500 0-OS-TCA

- OSo0

0 .-.TCA A - ...... - 0AS-GCA GCA

o.0 .0 .?

~- -O.~ - 2.500 .?

4.

t

A.

0.5

1.0

..--1.5

2.0

2.5

3.0

4

n

prnole/ min/ 100 g body weight

FIG.1.Effect of the infusion of stepwise increasing doses of sulfated and nonsulfated TCA and GCA on bile flow (A), bile salt secretion rate (B), phospholipid secretion rate (C) and cholesterol secretion rate (D). The infusion rates for sulfated taurocholic acid (S-TCA)and sulfated glycocholic acid (S-GCA) were 1, 2, 3 and 4 ~mol/min/100gm body wt and for TCA were 1, 1.5, 2 and 2.5 Fmol/min/100 gm body wt and for gm body wt. Each dose was infused for 30 min, and bile was collected every 10 min. Basal values during GCA were 0.5,1,1.5 and 2 ~mol/min/100 albumin infusion before the bile salt infusion are also shown. The data represent the mean values for six to eight experiments for each bile acid and the standard deviations were omitted for clarity.

6.07 +- 1.69 and 5.00 -t 0.33 pJmin/gm ofliver for TCA, GCA, S-TCA and S-GCA, respectively. With the nonsulfated bile acids, BF declined at the end of infusion of 2.5 pmol of TCA/min/100gm body w t and 2.0 pmol of GCA. However, BF was not significantlyreduced until the end of infusion of 4 bmol of S-TCA or S-GCA. Figure 1B summarizes the BSSR from data obtained by using enzymatic techniques for the measurement of total bile acid secretion. As with BF, the BSSR rose with increasing infusion doses, reaching SRm and then declining. The SRm (the highest three consecutive points) was 254 2 59, 257 39, 245 f 110 and 107 +- 8 nmol/min/gm of liver for TCA, GCA, S-TCA and S-GCA, respectively. No significant differences were found between the SRm of TCA and S-TCA,but the SRm of S-GCA was significantly lower than that of GCA and S-TCA. The decline in the BSSR after the SRm was reached was not significant. Bile acid analysis by GC showed that after the first sample was taken (10 min after the start of infusion), all the bile acids secreted after the TCA and GCA infusions were CA, but 10 to 15 nmol/min/gm of liver of nonsulfated bile acids was continuously secreted after the S-TCA and S-GCA

*

infusions. These values were subtracted from the results of the total bile acids to obtain the actual SRm for S-TCA and S-GCA. Figure 1,C, shows the BPLSR during infusions. With nonsulfated TCA and GCA, the BPLSR increased with escalating infusion doses, reaching a SRm of 11.2 & 4.8 and 16.0 +- 1.8 nmol/min/gm of liver, after which it declined until the end of the infusion. With S-TCA infusion, the BPLSR decreased during infusion but large differences existed between rats. With S-GCA, however, an immediate significant reduction of the BPLSR (10 min after the start of the infusion), which continued during the entire experiment, was noted. Similar results were obtained for the BCHSR (Fig. 1D). With TCA and GCA, the BCHSR reached a maximum of 1.95 k 0.96 and 2.72 k 0.38 nmol/min/gm of liver, after which it declined. With sulfated CA, the BCHSR was reduced after 10 min of the start of infusion and continued to be below control (nonsulfates infusion) values until the end of infusion. CDC Infusions. At the dose levels used, TCDCA and GCDCA were cholestatic; BF was, therefore, reduced within 10 min after the start of infusion and continued

441

SULFATION AND BILE FORMATION

Vol. 15, No. 3, 1992

BILE SALT SECRETION RATE

CHENODEOXYCHOLIC ACID BILE FLOW

L

0

.Ln

.-E 1 '.

albumin

0.5

1.0

1.5

2.0

3.0

4.0

urnole/ min/ 100 g body weight

CHOLESTEROL SECRETION RATE

PHOSPHOLIPID SECRETION RATE

12 I

3.500

O-OS-TDOC A - - AS-GDOC 8-8 TOOC o ----..oGOOC

Z 2.500-

. . . . . . . . . -. . . . . . . . . . . . . . . 1

0.000 albumin albumin

0.5

1.0

1.5

2.0

3.0

4.0

0.2

0.4

:

0.6

1.0

2.0

3.0

4.0

urnole/ min/ 100 g body weight

urnole/ rnin/ 100 g body weight

FIG.2. Effect of infusion of stepwise increasing doses of sulfated and nonsulfated TCDCA and GCDCA on bile flow (A), bile salt secretion The infusion rates for sulfated taurochenodeoxycholic acid (S-TCDCA) rate (B), phospholipid secretion rate (C) and cholesterol secretion rate (D). and sulfated glycochenodeoxycholicacid (S-GCDCA)were 1 , 2 , 3 and 4 p,mol/min/100 g m body wt; for taurochenodeoxycholic acid (TCDCAI and glycochenodeoxycholic acid (GCDCA) were 0.5, 1.0,1.5 and 2.0 pmol/min/lOO gm body wt. Each dose was infused for 30 min, and bile was collected every 10 min. Basal values during albumin infusion before bile salt infusion are also shown. The data represent the mean values for six to eight experiments for each bile acid, and standard deviations were omitted for clarity.

to decline during the stepwise increasing doses of infusion until complete cholestasis occurred at the infusion of 2.0 pmol/min/100 gm body wt. S-TCDCA and S-GCDCAinfusion, on the other hand, increased BF up to a maximum of 5.17 -+ 1.48 and 5.05 k 0.28 pl/min/gm of liver, which declined thereafter until the infusion of 4 pmol/min/100 gm body wt (Fig. 2A). Similarly, the BSSR was highest in the first 30 min of infusion of 0.5 pmol of TCDCA and GCDCN100 gm body wt, after which it declined to a full stop on infusion of 1.5 pmol/min/100 gm body wt. The SRm was 107 & 47 for TCDCA and 44 2 23 nmol for GCDCNminigm of liver. With S-TCDCA and S-GCDCA, the BSSR rose with increasing infusion doses, reaching a maximum of 188 t 12 and 182 k 13 nmol/min/gm of liver, respectively, with no significant difference between them. However, the SRm for sulfated CDCA was significantly higher than that of its parent compounds. After the SRm, the BSSR declined to a significant level at the end of infusion (4 pmol/min/100 gm body wt, Fig. 2B). Bile acid analysis by GC showed that with TCDCA and GCDCA the bile acids secreted after 10 min of infusion were mainly CDCA. With S-TCDCA and

S-GCDCA, 10 to 15 nmol/min/gm of liver was unsulfated and subtracted from the total BSSR to obtain the true S-TCDCA and S-GCDCA secretion rates. In both cases, 3a, 7a disulfate of a-muricholic acid, identified by the MSD, accounted for up to 90% of the bile acids secreted, indicating that S-TCDCA and S-GCDCA were 6 p-hydroxylated to a similar degree. The BPLSR was increased during TCDCA infusion to a maximum of 9.4 f 2.7 from a basal value of 6.8 t 0.9 nmol/min/gm of liver. On the other hand, GCDCA infusion reduced the BPLSR to 4.5 f 2.1 (value obtained at the GCDCA SRm) compared with a basal value of 6.1 +- 1.4 nmol/min/gm of liver. With both bile acids, the BPLSR continued to decline thereafter until the end of the experiment. S-TCDCA and S-GCDCA infusions immediately reduced the BPLSR (Fig. 2 0 . The BCHSR was similar to the BPLSR, increasing during TCDCA infusion to a maximum of 1.93 k 0.76 compared with a basal value of 1.53 f 0.5 nmol/min/gm of liver. During GCDCA infusion, it was reduced to 1.28 -+ 0.66 compared with a basal value of 1.46 k 0.26 nmol/min/gm of liver. The BCHSR declined thereafter, reaching undetectable amounts with 1 pmol infusion.

442

YOUSEF, MIGNAULT AND TUCHWEBER DEOXYCHOLIC ACID

HEPATOLOGY

BILE SALT SECRETION RATE

BILE FLOW

0-OS-TDOC A - - AS-GDOC

350.-

.-.TDOC o -0GDOC

300 T

>-

200 250/ 1504

I

0'

0

E

o'bumin

0.4

0.2

0.6

1.0

3.0

2.0

albumin

4.0

0.2

0.4

0.6

1.0

2.0

PHOSPHOLIPID SECRETION RATE 0-0

C

0

A-

-.m o

0

0

-

4.0

CHOLESTEROL SECRETION RATE

25 20 --

3.0

urnale/ min/ 100 g body weight

urnole/ min/ 100 g body weight

O-OS-TDOC

S-TDOC

AS-GDOC TDOC oGDOC

o GDOC

2 2500 cp

. . 0.0004albumin

0.2

0.4

0.6

1.0

2.0

3.0

4.0

albumin

urnole/ min/ 100 g body weight

; !

0.2

0.4

0.6

1.0

: 2.0

; ;

: : ; : :-I 3.0 4.0

urnole/ min/ 100 g body weight

FIG.3. Effect of the infusion of stepwise increasing doses of sulfated and nonsulfated TDOCA and GDOCA acids on bile flow (A),bile salt secretion rate (B),phospholipid secretion rate (C) and cholesterol secretion rate (D). The infusion rates for sulfated taurodeoxycholic acid (S-TDOCA)and sulfated glycodeoxycholicacids (S-GDOCA) were 1,2, 3 and 4 ~mol/min/100gm body wt; for taurodeoxycholic acid (TDOCA) and glycodeoxycholicacid (GDOCA) were 0.2,0.4, 0.6 and 1+mol/min/100 gm body wt. Each dose was infused for 30 min, and bile was collected every 10 min. Basal values during albumin infusion before bile salt infusion are also shown. The data represent the mean values for six to eight experiments for each bile acid, and standard deviations were omitted for clarity.

The BCHSR was immediately reduced to undetectable amounts on infusion of 1 pmol of S-TCDCA and to 0.19 t 0.32 on infusion of 2 kmol of S-GCDCA (Fig. 2D). DOC Infusion. As with CDCA infusions, TDOCA and GDOCA rapidly decreased BF after the start of infusion of 0.2 pmol/min/100 gm body wt, and the decline continued with stepwise increasing infusion doses. On the other hand, S-TDOCA and S-GDOCA enhanced BF, reaching a maximum of 5.03 t 1.29 and 4.60 t 0.98 pyminigm liver, respectively. BF did not change significantly thereafter. No significant difference in BF was found after S-TDOCA and S-GDOCA infusions (Fig. 3A). The BSSR rose with increasing infusion doses to reach a SRm of 76.7 k 30.9 and 72.0 -t 23.3 nmol/min/gm liver with TDOCA and GDOCA, respectively, after which it declined to a complete stop with 1 kmol infusions. The BSSR was elevated with escalating infusion doses of S-TDOCA and S-GDOCA, reaching a maximum of 254.0 -+ 138.3 and 115.0 -C 31.9 nmol/min/gm liver and continuing at this level until the end of the infusion of 4 pmol (Fig. 3b). Bile acid analysis

by GC showed that with TDOCA and GDOCA, DOCA was the only bile acid secreted after 10 min of infusion. Similarly, with S-TDOCA, DOCA was secreted unchanged, but with S-GDOCA, CA (identified by the MSD) accounted for 61% of the bile acid secreted. The BPLSR was increased during TDOCA and GDOCA infusion, reaching a maximum of 15.7 k 8.0 and 20.1 k 8.2 nmol/min/gm liver, after which it declined to undetectable amounts on infusion of 0.6 p,mol. On the other hand, S-TDOCA and S-GDOCA immediately reduced the BPLSR during the entire experiment (Fig. 3 0 . Similarly, the BCHSR was increased during TDOCA and GDOCA infusion, reaching a maximum of 2.73 0.95 and 2.89 k 0.93 nmol/min/gm of liver, respectively, and declining thereafter. The BCHSR fell continuously after 10 min of S-TDOCA and S-GDOCA infusion (Fig. 3D). Choleretic Activity. The choleretic activity of the bile acids was calculated from the slope of the regression line between the BSSR and BF (Fig. 4). In all experiments, a significant correlation was found between the BSSR and BF. The choleretic activity of TCA and GCA was 6.3 and 8.0 pl/pmol, respectively, with no significant dif-

*

443

SULFATION AND BILE FORMATION

Vol. 15, No. 3, 1992

ference between them. Sulfation increased the choleretic activity of TCA and GCA to 17.8 and 36.6 p,Vbmol, respectively. The choleretic activity of S-GCA was significantly higher than that of S-TCA. CDCA and DOCA had no choleretic activity and, in fact, they reduced BF and may, therefore, be considered t o have a cholestatic effect. This cholestatic action of DOC was more marked than that of CDCA with no significant difference between taurine or glycine conjugates. Sulfation of CDCA prevented the cholestatic activity and produced choleretic activity of 19.8 and 19.1 p,l/pmol for S-TCDCA and S-GCDCA, respectively. Similar results were obtained for S-TDOCA and S-GDOCA because both had choleretic activity of 12.1 and 28.6 p,l/p,mol, respectively, with the S-GDOCA values being significantly higher than those of S-TDOCA. Bile Acid-dependent Secretion of Phospholipid and Cholesterol. Figure 5 illustrates bile acid-dependent phospholipid secretion at the dose levels used in this study, calculating the slope of the regression line between the BSSR and the BPLSR using data obtained before cholestasis (reduction in BSSR). It can be seen that TCA had the lowest and GDOCA the highest effect on phospholipids. All sulfated bile acids reduced phospholipid secretion, with S-TCA and T-DOCA being the least effective. Similar results were obtained for bile acid-dependent cholesterol secretion (Fig. 6). DISCUSSION

Little information is available on the physicochemical and physiological properties of sulfate esters of bile acids despite their importance in cholestatic liver diseases (5-11). Recently we investigated the effects of complete sulfation of common free bile acids on bile formation in rats (12). It was shown that the secretion of sulfated bile acids was slower and less than that of nonsulfated bile acids, demonstrating transport maximum kinetics rather than SRm characteristics of nonsulfated bile acids. Choleretic activity was higher and ranged from 15 to 30 Fl/Frnol for DOCA, CDCA and CA. These bile acids did not produce cholestasis and reduced the secretion of biliary lipids. Therefore it was suggested that sulfation may prevent the cholestatic effect of nonsulfated bile acids because of high choleretic activity and the protection of cellular membranes against the detergent properties of high bile salt concentrations that result in solubilization of phospholipids in cellular membranes (22). Because the physicochemical properties of bile acids are affected by the type of conjugation and because almost all bile salts in bile are conjugated with glycine and taurine in humans (24), we investigated the effects of sulfated, conjugated common bile acids on bile formation in rats. Although the physiological significance of this study in human cholestasis may not be apparent because monosulfated, conjugated bile salts predominate (lo), it is nevertheless important in defining the role that conjugation may play in the physiological action of these bile acids. These data are in general agreement with those

T C A

S T

G C A

A

S G C A

T C D C

S T C D C

G C D C

S C D C

T D O C

S T D 0 C

G D O C

S G D

C

Bile Acids FIG.4.Choleretic activity of sulfated and nonsulfated bile salts. The values were obtained from the slope of the regression line of the correlation between bile flow and bile salt secretion rate during the infusion of the bile salt and until the SRm of bile salt. Values are expressed as microliters ofbile per micromole ofbile salt, mean 2 S.D. nmole / umole bile acid

300 I

-100"

"

T C A

S T A

a C A

"

S Q C

T C O C

'

1

'

S T C

O C D C

S Q C

C

C

"

T D O C

S T D C

' G D O C

"

S G D

C

Bile Acids

FIG.5 . Bile acid-dependent secretion of phospholipid for sulfated and nonsulfated bile salts. The values were calculated from the slope of the regression line of the correlation between bile salt and phospholipid secretion rates during the infusion of the bile salt and until the SRm of the bile salt. Values are expressed as nanomoles of phospholipid per micromole of bile salt, mean 5 S.D.

obtained with sulfated free bile acids with regard to their choleretic activity and their effect on biliary lipid secretion. All sulfated bile acids had high choleretic activity and reduced biliary phospholipid and cholesterol secretion. Nonsulfated bile acids had lower choleretic activity, and CDCA and DOCA were even cholestatic at the dose levels used in this study. These bile acids increased the secretion of biliary lipids until their SRm, after which it declined. The cholestatic effect of bile acids may be related to the exhaustion of the biliary phospholipid pool available for bile salt secretion, after which the bile acids solubilize cellular membrane phospholipids (22). It has been suggested that the reduction of biliary phospholipid secretion during the infusion of sulfated bile acids may be a mechanism by which they protect cellular membranes. In this study, a significant correlation was found between the changes in bile aciddependent phospholipid secretion and choleretic activity (r = 0.7977, Fig. 7). It is apparent that sulfation increased the bile salt SRm for conjugated CDCA and DOCA but had no effect on the SRm of TCA and reduced the SRm of GCA. This is in contrast to data obtained

444

HEPATOLOGY

YOUSEF, MIGNAULT AND TUCHWEBER nmole / urnole _______ bile acid ~

40

~

30 I

20

-10r

-*O'L

L

-

1 -

L

I 1-

; s E s ; s g s ; s : s A

T C

A

A

G C A

D C

T C

D C

G C

C

C

O C

T D

O C

c

G D C

Bile Acids

FIG.6. Bile acid-dependent secretion of cholesterol of sulfated and nonsulfated bile salts. The values were calculated from the slope of the regression line of the correlation between bile salt and cholesterol secretion rates during the infusion of the bile salt and until the SRm of the bile salt. Values are expressed as nanomoles of cholesterol per micromole of bile salt, mean t S.D. ul / urnole bile acid eo

r

I

401

30

3 o L-100 --

that S-TCA and S-TDOCA, which are not metabolized, had a similar SRm. These data suggest that all sulfated, conjugated bile acids may have a similar SRm, and when variation occurs, they may be attributed to biotransformation. This theory cannot explain differences in the SRm of S-TCA and S-GCA, especially because no metabolites were detected in these perfusions. The decrease in phospholipid and cholesterol secretion produced by sulfated bile acids is in accordance with their physicochemical properties because sulfation would affect hydrophobic/hydrophilic balance and diminished phospholipid secretion as predicated from the previous {n vi& stbdies of Armstrong and Carey (13). However, these data may not concur with the conclusion of Stevens et al. (14). These authors showed that phospholipid secretion was reduced after monosulfated TCDCA infusion, but they classified this decrease as slight (0.23 vs. 0.13 kmol/min/gm liver for TCDCA and STCDCA, respectively). It is possible that the dose used in these studies was too low to have a larger effect on phospholipid secretion because the BSSR was 1 to 3 kmol/min/kg compared with values of up to 15 p,mol/min/kg obtained in our investigation. Furthermore, it is possible that the secretion of de nouo synthesized bile acids in their experiments (caused by overnight depletion of bile acid pool) was the cause of this discrepancy. Different transport systems have been demonstrated for sulfated and nonsulfated bile acids (25.26). and thus the infusion of small concentrations of S-TCDCA would not affect the secretion of de nouo synthesized bile acids that stimulate lipid output in bile. In this investigation the infusion of a high dose of sulfated bile acid probably masked any effect of the nonsulfated bile acid secreted on lipid output. This suggests that the ratio of sulfatedlnonsulfated bile acids would regulate biliary lipid secretion. The data presented in this study extend our knowledge on the biological effect of sulfated bile acids and shed some light on their possible role in cholestasis. These bile acids exhibited high choleretic activity, were not cholestatic at doses from 1 to 4 kmol/min/100 gm body wt and reduced biliary lipid secretion. Thus sulfation may influence cholestasis by stimulating bile flow and preserving cellular structure in the presence of high concentrations of detergent cholestatic agents such as bile acids. The data also show that the type of conjugation may influence the biological function of sulfated bile acids, with taurine-conjugated bile acids causing increased BF and having a higher SRm but being the least effective in reducing biliary lipid secretion.

-40 -20 -10 -6" o s" m zo " 40 nmole phospholipid / prnole bile acid "

' I

100

FIG.7. Effect of changes in biliary secretion of phospholipid on the choleretic activity of the bile salt. The graph represents the correlation between the increase or the decrease in biliary phospholipid secretion (nanomoles per micromole of bile salt) and the choleretic activity of the bile salt (microliter per micromole of bile salt), mean S.D.

*

with free sulfated bile acids, where the SRm was always lower than with nonsulfated compounds. The SRm of nonsulfated bile acids is apparently inversely related to their detergent properties (20, 23). These data are in agreement with this suggestion with the exception of GCDCA, which showed the lowest SRm value perhaps because of its cellular toxicity. This toxicity could also explain the unusual pattern of biliary lipid secretion during the infusion of this bile acid as compared with TCDCA (Fig. 2 C and D). The SRm of sulfated, conjugated bile acids does not conform with this trend since all the compounds used would be expected to have similar detergent properties because all the hydroxyl groups were sulfated. Therefore it is possible that variations in the SRm between different conjugates may be due to their metabolism. Both S-TCDCA and S-GCDCA were metabolized by 90% to 3a, 7a-disulfate, 6p-hydroxy 5p-cholanoic acid (3a, 7a disulfate of a-muricholic acid) and thus both had a similar SRm. S-GDOCA was metabolized by 60%to 3a, 12a-disulfate 7a-hydroxy 5p-cholanoic acid (3a, 7a disulfate of CA) and had a SRm similar to that of S-GCA. It is interesting

REFERENCES 1. Small DM. The physical chemistry of cholanic acid. In: Nair PP, Kritchevsky D, eds. The bile acids. Vol. 1. New York: Plenum

Press, 1971249-256. 2. Carey MC. Physical-chemical properties of bile acids and their salts. In: Danielsson H, Sjovall S, eds. Sterols and bile acids. Amsterdam: Elsevier, 1985:345-403. 3. Hofmann AF',Roda A. Physico-chemical properties of bile acids

Vol. 15, No. 3, 1992

SULFATION AND BILE FORMATION

and their relationship to physiological properties: an overview of the problem. J Lipid Res 1984;25:1477-1489. 4. Hofmann AF. Bile acids. In: Arias IM, Jakoby WB, Popper H, Schacter D, Shafritz DA, eds. The liver; biology and pathophysiology. 2nd ed. New York Raven Press, 1988:533-572. 5. Stiehl A. Bile salt sulfates in cholestasis. Eur J Clin Invest 1974;4:59-63. 6. Makino I, Hashimoto H, Shinozaki K, Yoshino K, Nakagawa S. Sulfated and nonsulfated bile acids in urine, serum and bile of patients with hepatobiliary diseases. Gastroenterology 1975;68: 545-553. 7. Summerfield JA, Gullen J, Barnes S, Billing BH. Evidence for control of urinary excretion of bile acids and bile acid sulfate in the cholestatic syndromes. Clin Sci Mol Med 1977;52:51-65. 8. Alme B, Bremmelgaard A, Sjovall S, Thomassen. Analysis of metabolic profiles of bile acid in urine using a lipophilic anion exchanger and computerized gas liquid chromatography-mass spectrometry. J Lipid Res 1977;18:339-362. 9. Stiehl A, Ast E, Gzyan P, Frohling W, Raedsch R, Kommerell B. Pool size, synthesis and turnover of sulfated and nonsulfated cholic acid and chenodeoxycholic acid in patients with cirrhosis of the liver. Gastroenterology 1978;74:572-577. 10. Bartholomew TC, Summerfield JA, Billing BH, Lawson AM, Setchell KDR. Bile acid profiles of human serum and skin interstitial fluid and their relationship to pruritus studied by gas chromatography-mass spectrometry. Clin Sci 1982;63:65-73. 11. Niessen KH, Teufel HM, Brugmann G. Sulfated bile acids in duodenal juice of healthy infants and children compared with sulfated bile acids in paediatric patients with various gastroenterological diseases. Gut 1984;25:26-31. 12. Yousef IM, Barnwell SG, Tuchweber B, Weber A, Roy CC. Effect of complete sulfation of bile acids on bile formation in rats. HEPATOLOGY 1987;7:535-542. 13. Armstrong MJ, Carey MC. The hydrophobic-hydrophilic balance of bile salts: inverse correlation between reverse-phase high performance liquid chromatographic mobilities and micellar cholesterol solubilizing capacities. J Lipid Res 1982;23:70-80.

445

14. Stevens RD, Lack L, Collins RH, Meyers WC Jr, Killenberg PG. Effect of monosulfate esters of taurochenodeoxycholate on bile flow and biliary lipids in hamsters. J Lipid Res 1989;30:673-679. 15. Igimi H, Carey MC. pH solubility relations of chenodeoxycholic and ursodeoxycholic acids: physical-chemical basis for dissimilar solution and membrane phenomena. J Lipid Res 1980;21:72-90. 16. Talalay P. Enzymatic analysis of steroid hormones. Methods Biochem Anal 1960;8:119-143. 17. Bligh EG, Dyer WJ. A rapid method for total lipid extraction and purification. Can J Biochem Physiol 1959;37:911-917. 18. Bartlett GR. Phosphorous assay in column chromatography. J Biol Chem 1959;234:466-468. 19. Gurantz D, Laker MF, Hofmann AF. Enzymatic measurement of choline containing phospholipids in bile. J Lipid Res 1981;22: 373-376. 20. Yousef IM, Mignault D, Weber AW, Tuchweber B. Influence of dehydrocholic acid on secretion of biliary lipids in rats. Digestion 1990;45:40-51. 21. Hardison WGM, Hotoff DE, Miyai K, Weiner RG. Nature of bile acid maximum secretion rate in the rat. Am J Physiol 1981;241: G337-G343. 22. Yousef IM, Barnwell SG, Gratton F, Tuchweber B, Weber A. Liver cell membrane solubilization may control the maximum secretory rate of cholic acid in rat. Am J Physiol 1987;252:G84-G91. 23. Barnwell SG, Yousef IM, Tuchweber B. Biliary lipid secretion in rats during infusion of increasing doses of unconjugated bile acids. Biochim Biophys Acta 1987;992:221-233. 24. Haslewood GAD. The biological importance of bile salts. Amsterdam: Elsevier-North Holland Publishing Co., 1978:28-32. 25. Kuipers F, Enserink M, Havinga R, van der Steen Ad BM, Hardonk MJ, Fevery J, Vonk RJ. Separate transport systems for biliary secretion of sulfated and unsulfated bile acids in the rat. J Clin Invest 1988;81:1593-1599. 26. Yousef IM, Tuchweber B, Mignault D, Weber A. Effect of co-infusion of cholic acid and sulfated cholic acid on bile formation in rats. Am J Physiol 1989;256:G62-G66.

Effect of complete sulfation of bile acids on bile formation: role of conjugation and number of sulfate groups.

The effect of complete sulfation of conjugated cholic, chenodeoxycholic and deoxycholic acids on bile formation was investigated in rats. The sulfated...
876KB Sizes 0 Downloads 0 Views