- Email: [email protected]
Effect of Organic Anions on Biliary Lipids in the Rat
GASTROENTEROLOGY 1982;83:1120-6
Effect of Organic Anions on Biliary Lipids in the Rat MICHAEL D. APSTEIN and SANDER
J.
ROBINS
Boston and West Roxbury Veterans Administration Medical Centers and Boston and Harvard University Schools of Medicine, Boston, Massachusetts
Bile salts enhance the biliary secretion of phospholipid and cholesterol. Other amphipilic molecules, organic anions, are secreted into bile as well. We studied the effects of bilirubin and iodipamide, two chemically dissimilar organic anions, on biliary lipid secretion in the rat. We infused bile salt pooldepleted rats with a stepwise infusion of taurocholate and a constant infusion of organic anion . Both organic anions markedly inhibited the biliary secretion of phospholipid and cholesterol without affecting bile salt secretion. This inhibition, at least with iodipamide , was dose-dependent and fully reversible. Using tritiated water as a precursor, we measured hepatic and biliary cholesterol synthesis in the presence or absence of an iodipamide infusion to see if decreased lipid synthesis couid explain decreased secretion. Despite the marked reduction in biliary cholesterol secretion, the specific activity of biliary cholesterol was unchanged during an iodipamide infusion. We suggest that organic anions interfere with the assembly of the biliary mixed micelle resulting in micelles that are deficient in phospholipid and cholesterol .
Micelle-forming bile salts can stimulate the biliary secretion of phospholipid and cholesterol (1). In addition, these bile salts can enhance the excretion of some (2,3) , but not all (4) , organic anions. Although organic anion excretion is noncompetitive Received September 15 , 1980. Accepted June 23 , 1982. Address requ ests for reprints to: Michael D. Apstein, M.D. , Division of Gastroenterology, West Roxbury Veterans Administration Medical Center, West Roxbury, Massachusetts 02132 . Dr. Apstein is the recipient of a Veterans Administration Career Development Award. This paper was presented in part at the meetings of the American Gastroenterological Association in New Orleans , La., May 19 79 , and a preliminary report was published in abstract form (GASTROENTEROLOGY 1979;76:1274). The authors thank Arthur Taylor for his help with the statistical analysis. © 19 82 by the American Gastroenterological Association 0016-5085 /82/1111 20-07$02.50
with biliary bile salt secretion and presumably along separate pathways (5), recent data suggest that at least one organic anion, bromsulphalein (BSP), inhibits biliary phospholipid and cholesterol secretion (6). We have studied the effects of two chemically dissimilar organic anions, bilirubin and iodipamide, on micellar lipid secretion in the rat. We found that both bilirubin and iodipamide depress biliary phospholipid and cholesterol secretion without affecting bile salt secretion. We then examined the effects of iodipamide on the synthesis of hepatic and biliary cholesterol and found that although biliary cholesterol secretion was inhibited its synthesis was unimpaired.
Methods and Materials Animals and Preparation Male, Sprague-Dawley rats (250-350 g) were maintained on an ad libitum Chow diet (Purina, St. Louis, Mo.). Animals were anesthetized with pentobarbital (45 mg/kg, intraperitoneally) and the bile duct and two iliac veins were cannulated. Animals were restrained in cages and allowed to recover from surgery for the next 18 h. Hydration was maintained by intravenous saline infusion (0 .9% at 0.9 ml/h) and body temperature maintained at 37°C with incandescent lamps (7).
Infusates The bilirubin solutions were made in the dark just before use by adding 170 /-Lmol of unconjugated bilirubin (Sigma Chemical Co., St. Louis , Mo .) to 50 ml of 0.45% NaHC0 3 /0.45% NaC!. Iodipamide (Cholegrafin, E. R. Squibb & Sons, Inc., Princeton, N.J.) was used as supplied by the manufacturer. Control infusions were either 0.9% NaCl or 0.45% NaHC0 3 /O.45% NaC!. Since results using either of the control solutions were not significantly different, they have been combined in all analyses. Sodium taurocholate (Calbiochem-Behring Corp., La Jolla, Calif., A grade, >99% pure) was infused as 10 mM in 0.9% NaC!. In early experiments we found that iodipamide interferes with enzymatic determination of bile salt. That is, when
November 1982
iodipamide in concentrations found in bile was added to salt standards and bile salt was measured by an enzymatic method (8). the optical density was increased by 50%100%. In fact, iodipamide alone has an optical density at 340 nm, the wavelength at which the enzymatic method of Talalay (8) is read. Therefore, in iodipamide and some control-infused animals, 0.75 /LCi (40 mCi/mmol) of [2414C]taurocholate (Mallinckrodt Inc., St. Louis, Mo., purified to >95% by thin-layer chromatography) was added to the bile salt infusates, and bile salt secretion was measured by recovery of radioactivity.
Experimental Design After 18 h of bile salt pool depletion when bile salt output is constant and at its nadir (9), rats were infused through one iliac vein with taurocholate in a stepwise fashion at rates of 108, 216, and 432 nmol/min. They were simultaneously infused through the other iliac vein with either a bilirubin solution that was kept in the dark (3.4 /Lmol/100 g, bolus followed by a constant infusion of 250 nmol/min· 100 g) (10). iodipamide (500 nmol/min . 100 g) (11), or a comparable volume of 0.9% NaCl (or 0.45% NaHC0 3/0.45% NaCl). Bile was collected on ice in tared tubes (and in the dark for the bilirubin-infused animals) for five 10-min periods before infusion of either anion or bile salt, and at each rate of bile salt infusion. The first two collections of each infusion period were allowed for equilibration. The next three collections were used to calculate biliary outputs since there was <10% variation between collections. Iodipamide-infused rats were further studied to determine the extent of reversibility of biliary lipid changes. In this group, after the stepwise taurocholate infusion, taurocholate was continued at 432 nmol/ min, iodipamide was replaced by saline, and bile was collected for an additional 26 h. To determine the dose-response effect of iodipamide on micellar lipid secretion, separate groups of 4 rats were infused with taurocholate at a constant rate of 432 nmol! min. After biliary lipid excretion was constant for 60 min, either a 0.9% saline or iodipamide infusion at 25, 100, 250, or 500 nmol/min . 100 g was begun and bile was collected for an additional 90 min. To measure hepatic and biliary cholesterol synthesis, * two additional groups of 4 animals each were bile salt pool depleted as outlined above. They were infused intravenously with taurocholate at a constant rate of 432 nmol/ min. After biliary lipid excretion had reached a steady state for 60 min, 30 mCi of tritiated water (New England Nuclear, Boston, Mass.) was given intravenously and bile was collected in 3D-min periods for 2.5 h. Then either an iodipamide solution (500 nmol/min . 100 g) or a 0.9% NaCl solution was begun and bile was collected for an additional 2.5 h. At the end of the 5 h, the animals were rapidly reanesthetized and the liver was flushed with cold saline through the portal vein and excised. * Biliary cholesterol synthesis refers to the amount of newly synthesized (labeled) cholesterol appearing in bile.
ORGANIC ANIONS AND BILIARY LIPIDS
1121
Analytical Methods Bile volumes were determined gravimetrically. Phospholipid was measured by the method of Bartlett (12), and cholesterol by the method of Lie et al. (13). Iodipamide was measured as iodine by the method of Zak et al. (14). Bile salt output was measured by the enzymatic method of Talalay (8) in the bilirubin-infused animals, by bile radioactivity in the iodipamide-infused animals, and by both methods in the control-infused animals. Bile radioactivity for 14C was measured by scintillation counting using an internal standard to correct for quenching. Cholesterol synthesis was measured by the rate of tritiated water incorporation into digitonin-precipitable sterols in each half-hour bile collection and in the liver at the end of the experiment (15). Radioisotope incorporation was determined from samples that were first extracted for lipid by the method of Folch et al. (16). Cholesterol was precipitated as the digitonide after saponification. Precaution was taken to remove traces of precursor tritiated water, which could combine with the digitonin precipitate, by splitting the sterol-digitonin precipitate in pyridine and then extracting the free sterol with ether (15,17). Recovery of [7- 3Hlcholesterol (New England Nuclear) added to control bile was 98 ± 3%. 3H Radioactivity was measured by scintillation counting with internal standard to correct for quenching.
Statistical Analysis Values are presented as the mean ± SEM. Student's t-test for unpaired observations was used to compare groups and p values of <0.05 were considered to be statistically significant.
Results Figure 1 shows the relation of bile salt secretion to bile salt infusion. After biliary drainage, but before the bile salt infusion was started, baseline bile salt secretion was similar in all three groups (averaging 50 nmol/min). After the infusion of taurocholate, biliary bile salt secretion increased linearly to a similar extent in control-, bilirubin-, and iodipamide-infused groups. Whether bile salt was measured by the enzymatic method or by recovery of radioactivity, bile salt secretion equaled the sum of
Table 1. The Effect of Iodipamide on the Molar Ratio of Biliary Lipids Iodipamide infusion rate (nmol/min . 100 g)
o 25 100 250 500
Molar ratio of biliary bile saltl phospholipidlcholesterol 66:11:1 83:12:1 113:12:1 159:11:1 489:27:1
1122
APSTEIN AND ROBINS
GASTROENTEROLOGY Vo!' 83, No.5
-e
1
500
/~
c
••o ec
// ./ /'/ . /
400
..... Figure 1. The relation of biliary bile salt secretion to bile salt infusion. In the control-infused group (n = 4), biliary bile salt secretion was measured both enzymatically (ENZ, closed circle) and using [24- 14 C]taurocholate (14 C_ TC, open circle). There was no significant difference between groups in bile salt secretion at any given rate of infusion (x ± SEM).
I
...oew
a:: 300
U
T /;,~ ,6" //// CONTROL
200
w
...
/'V'
,'~/'i/
"/
...I
oe
.,'/ "
(/)
w
...I
/j~'
/~
...
w )(
/<,/;/./ ./
///
/:;/
Z
o
a::
//
/;/ /1 /
......
/
100
INFUSED (n=4) (14C-TC)
0
(ENZ)
•
BILIRUBIN INFUSED (n=6) IODIPAMIDE INFUSED (n=4)
0
A
III
0~1--------~-------4
o
100
________- L________~______~1
200
300
BILE SALT INFUSION RATE
baseline excretion plus the taurocholate infusion rate. After bile salt pool depletion. bile volume was similar in all groups and averaged 5.0 ILl/min' 100 g. In the saline-infused group in response to increasing bile salt infusion. bile volume increased by 13%. 23%. and 38%. respectively. In the bilirubin-infused group. the increase was similar: 23%. 28%. and 46%. respectively. In the iodipamide-infused group. the increase in bile volume was 143%. 160%. and 164%. respectively. Both organic anions markedly inhibited the biliary secretion of phospholipid and cholesterol (Figure 2). The first set of points in each panel shows that after bile salt pool depletion and before the start of either bile salt or anion infusion. the biliary secretion of cholesterol and phospholipid was the same in all groups. In controls. in response to increasing bile salt infusion. there was the expected and progressive rise of cholesterol and phospholipid secretion. However. in contrast. in both anion-infused groups. the secretion of cholesterol and phospholipid was significantly less (p < 0.01) than controls at all rates of bile salt infusion. To determine if this inhibition of secretion could be reversed. the experimental period was prolonged
400
500
(nmol •• /mln)
in 4 animals that had been previously infused with iodipamide and bile was collected for an additional 26 h. The results are shown in Figure 3. Before either bile salt or iodipamide infusion. the biliary secretion of phospholipid and cholesterol reflected the bile salt-depleted state of the animals. Within 20 min of the start of the iodipamide infusion. iodipamide secretion reached a maximum. rema.ined unchanged as the bile salt infusion rate was increased. and then slowly fell over the remainder of the study period. Phospholipid and cholesterol secretion. both of which were depressed by iodipamide. began to rise only when iodipamide secretion diminished. At about 12 h after iodipamide infusion was discontinued. they had reached values comparable to those in the control group. As shown in Figure 4. the inhibition of biliary phospholipid and cholesterol secretion by iodipamide is dose-dependent. With each increase in the iodipamide infusion rate there is a significant decrease in phospholipid and cholesterol secretion without any decrease in bile salt secretion. The resulting change in the molar ratio of biliary lipids is shown in Table 1. Depending on the infusion rate of iodipamide. the molar ratio of bile salt/phospholipid/cholesterol varies from 66: 11 : 1 to 489: 27: 1.
November 1982
ORGANIC ANIONS AND BILIARY LIPIDS
1123
Control Infused
8
n=8
6
...o
QI
c:
4
tJ)
QI
E
o
"'-
.J::.
QI
U
tJ)
Bilirubin Infused
2
o
E c:
lodipamide
n=4
z
o
120
~
w a: u
Control Infused
x!SEM
n=8
100
><
w
tJ)
> a:
'0
..J
.J::.
a.
«
80
Figure 2. The effect of bile salt infusion on biliary phospholipid and cholesterol secretion in bilirubin- (n = 4), iodipamide- (n = 6), and control-infused rats (n = 8). There was no difference (p > 0.05) in biliary lipid secretion between groups before bile salt infusion was started. Phospholipid and cholesterol secretion were markedly (p < 0.01) inhibited by either anion at each subsequent rate of bile salt infusion (iC ± SEM).
0
a. 60
!XI
tJ)
0 .J::. Q.
40
lodipamide
Infused
n=4
20 0
•
Bilirubin Infused
0
108
n=6
216
432
BILE SALT INFUSION RATE (n moles/min)
..J
Oc:
:fiE til 111_
6 4
...I02~ OE ~c:
zOO
~ ~.5 III ~~ a: ~.!!
o
Q.
0
><(I)E Oc
III
> a:
if
,~/________________________
120 80
40~
0 CJ 600
1118 ~ i~400 «
~~200
-G) cOO -Ec:
0 TAUROCHOLATE INFUSION 218
o
1
.32
2
n mol•• /mln
/ 3 TIME
6 HOURS
14
22
30
Figure 3. The relation of biliary secretion of iodipamide, phospholipid, and cholesterol over time (n = 4, iC ± SEM). Biliary phospholipid and cholesterol secretion was depressed during the period of iodipamide infusion and returned to control values after 12 h as the secretion of iodipamide decreased. The transient rise of phospholipid and cholesterol seen immediately after the start of the iodipamide infusion is unexplained, but may represent "washout" of existing lipid from the biliary system.
1124
GASTROENTEROLOGY Vo\. 83, No. 5
APSTEIN AND ROBINS
Figure 5 demonstrates the relation of iodipamide dose to biliary secretion of micellar lipid with time. In all groups the increase in cholesterol, phospholipid, and bile salt secretion in response to taurocholate infusion is the same. However within 45 min of the start of the iodipamide infusion, there is a dosedependent, sustained inhibition of biliary phospholipid and cholesterol while the biliary secretion of bile salt remains unimpaired. To determine if this inhibition of lipid secretion were due to decreased lipid synthesis, we measured the incorporation of tritiated water into hepatic and biliary cholesterol in the presence and absence of an iodipamide infusion. Figure 6 shows that the specific activity of biliary cholesterol increases after the intravenous administration of tritiated water. The specific activity of biliary cholesterol in the animals infused with iodipamide is the same as that in saline-infused controls. Similarly the amount of tritium incorporated into liver cholesterol is not significantly different between the two groups (7.5 ± 3.4 ILmol tritium incorporated/g dry wt· h in salineinfused animals vs 8.9 ± 1.1 in iodipamide-infused animals) .
Discussion We found that both bilirubin and iodipamide, two chemically dissimilar organic anions, markedly inhibit both biliary cholesterol and phospholipid secretion without affecting bile salt secretion. In addition, at least for iodipamide, this inhibition is reversible and dependent on the amount of iodipamide infused. Although the biliary secretion of organic anions and bile salt is noncompetitive (5), it is clear that both influence biliary phospholipid and cholesterol secretion. This phenomenon is not just limited to the rat; a depression of phospholipid and cholesterol secretion has been reported in patients receiving ioglycamide, an organic anion similar to iodipamide (18), and in dogs receiving BSP (6). There are several explanations for the observed depression of phospholipid and cholesterol secretion in the presence of organic anions. First we showed that these anions do not interfere with biliary lipid analysis by observing no change in either phospholipid or cholesterol concentrations after direct addition of anions to normal bile. Although possible, we do not believe that organic anions produce hepatotoxicity which results in depression of biliary cholesterol and phospholipid secretion. Even though iodipamide has been shown to be hepatotoxic in the rat, evidence of toxicity occurs only at doses much higher than we used (19). We are aware of no studies that demonstrate that
500
--E c
•
"'-
... ....I
cI:
400
II)
w 300
....I
III
G)
0
200 100
E
C
....I
z
0
.... W
0
II: W
...
II)
w
a:
....I
w
0
0
0
:r
UJ
> a:
80 II)
c(
C
..J
Do
m
;[
~
0
:r
60 40
Do
II)
0
:r
20
Do
0
0
25
100
250
500
IODIPAMIDE INFUSION RATE (nmol.s/minl 1 009)
Figure 4. The effects of iodipamide dose on biliary lipid secretion (n = 4, x ± SEM) . With increasing iodipamide dose, there is increasing inhibition of phospholipid and cholesterol secretion despite constant bile salt secretion.
unconjugated bilirubin at the rates of infusion we used is hepatotoxic. Furthermore, bile salt clearance by the liver is a good indicator of hepatic function (20). Figure 1 shows that the bile salt secretion rate equaled the sum of the infusion rate plus baseline secretion and was not different in the iodipamide-, bilirubin-, and control-infused groups. The inhibition of biliary phospholipid and cholesterol cannot be explained by depressed bile flow as bile volumes were similar in controls and bilirubininfused groups and were markedly increased in the iodipamide-infused group. The difference in bile volume in response to bilirubin and iodipamide infusions may reflect differences in the ability of organic anions to form self-aggregates or aggregates with the biliary mixed micelle. High levels of organic anions could depress cholesterol synthesis making it unavailable for biliary secretion. However, we showed that there was not a significant difference in the synthesis of either hepatic or biliary cholesterol in the presence or absence of an iodipamide infusion. Consequently, depressed cholesterol synthesis does not explain the inhibition of biliary cholesterol secretion. Perhaps lecithin synthesis was inhibited, resulting in de-
ORGANIC ANIONS AND BILIARY LIPIDS
November 1982
1125
TAUROCHOLATe INfUSION (432 ft",ol •• '.I.) 10DII'AliliDE INI'UIION
nmol.';min/100g 0 -0
100 ____ A 500 _ .-.
500
....
.E E
.,.,
'-
...C ~
II)
...
400 300
1&1
II
200
0
100
c
0
E
Figure 5. The relation of iodipamide dose to biliary lipid secretion over time (n = 4, x :t SEM). There is a prompt inhibition of phospholipid and cholesterol secretion, but no change in bile salt secretion, in response to an iodipamide infusion. This inhibition is dose-dependent and sustained over the time of infusion.
Z
0
t-
W
~
(J
W
e
...0 lao
80 60
~
:z: lao 40 en 0 :z: 20
-'
0
U)
>
tC
lao
ID
8
...0
e
en
~
4
0
2
III: 1&1
... 1&1
:z: U
0
\
105
150
MINUTES
creased biliary lecithin secretion which in turn depressed cholesterol secretion (21). We believe this chain of events is unlikely because newly synthesized lecithin accounts for only 3% of biliary lecithin secretion in the rat (22) . It is possible that organic anions compete with phospholipid and cholesterol uptake by the liver, decreasing the availability of these lipids for biliary secretion. This seems highly unlikely since in studies of biliary lipid secretion using the isolated perfused rat liver, lipid excretion was not dependent on the lipid content of the perfusion medium (22). We believe that there is a physical interaction between organic anions and biliary lipids that inhibits secretion of phospholipid and cholesterol. In vitro studies demonstrate that this kind of interaction may occur. Binding of both conjugated and unconjugated bilirubin with various phospholipids
has been described (23). Conjugated bilirubin was bound more strongly than unconjugated bilirubin and maximal binding was attained with lecithin, the major phospholipid in bile. Other investigators found that bilirubin binds to red blood cells and causes leakage of hemoglobin and glucose 6-phosphate dehydrogenase (24). The binding can be reversed by exposing bilirubin-red cell complexes to lipid micelles and can be prevented by exposing the bilirubin to the phospholipid fraction of red cell membrane before incubating with intact red cells. If, as suggested by Small (25), the translocation of bile salt leaches lecithin and cholesterol from the canalicular membrane, then a high concentration of organic anions in the membrane could compete with lecithin and cholesterol for incorporation into the biliary mixed micelle. If, as others have suggested (26), the macromolecular complex of mixed micelle
1126
GASTROENTEROLOGY Vol. 83, No. 5
APSTEIN AND ROBINS
•
• Controls o lodipamide
/
/
/
~
/
/
/
/
/0
;I
;I
;I
/
. /
/
/
/
/
.
/
0
/
/
.
0---
/'
/'
/
/
_ -cr
/'
/ //0
/
/
/
\y =2931 + 3453 LnX R2 =0.97
o
2
3
4
5
HOURS Figure 6. The effect of an iodipamide infusion of bile cholesterol specific activity. Tritiated water was administered at time O. In the iodipamide-infused group (n = 4), the iodipamide infusion was begun 2.5 h later. The salineinfused controls (n = 4) are shown by the closed circles. The regression line is calculated from the control data points only. The dashed lines represent 95% confidence bands. Bile cholesterol specific activity of the iodipamide-infused animals falls within these confidence bands. There is no difference in bile cholesterol specific activity after the start of iodipamide infusion when compared to controls.
is assembled in the hepatocyte before excretion, the high concentration of organic anions in the cytosol could displace lecithin and cholesterol from the micelle after it was assembled. The exact mechanism by which organic anions inhibit micellar lipid excretion is unknown. It is possible that they interfere with the intrahepatic assembly of mixed micelles, or perhaps disrupt previously assembled ones, resulting in biliary mixed micelles which are deficient in cholesterol and phospholipid.
References 1. Hardinson WGM, Apter JT. Micellar theory of biliary cholesterol excretion. Am J Physiol 1972;222:61-7. 2. Berk RN, Goldberger LE, Loeb PM. The role of bile salts in the hepatic excretion of iopanoic acid. Invest RadioI1974;9:7-15. 3. Goresky CA, Haddad HH, Kluger WS, et al. The enhancement of maximal bilirubin excretion with taurocholate-induced increments in bile flow. Can J PharmacoI1974;52:389-403. 4. Loeb PM, Barnhart JL, Berk RN. Iotroxamide-a new intravenous cholangiographic agent. Comparison with iodipamide and the effect of bile salts. Radiology 1977;125:323-9.
5. Wheeler HO. Secretion of bile. In: Schiff L, ed. Diseases of the liver. Philadelphia: Lippincott, 1975:87-110 . 6. Shaffer EA, Preshaw RM. Effects of sulfobromophthalein excretion on biliary lipid secretion in humans and dogs. Am J Physiol 1981;240:G85-9. 7. Roberts RJ, Klassen CD, Plaa GL. Maximum biliary excretion of bilirubin and BSP during anesthesia-induced alterations of rectal temperature. Proc Soc Exp BioI Med 1967;125:313-6. 8. Talalay P. Enzymatic analysis of steroid hormones. Methods Biochem Anal 1960;8:119-43. 9. Myant NB, Eder HA. The effect of biliary drainage upon the synthesis of cholesterol in the liver. J Lipid Res 1961;2:363-8. 10. Klassen CD, Roberts RJ, Plaa GL. Maximum biliary excretion of bilirubin and BSP during various rates of infusion in rats of different weights and strains. Toxicol Appl Pharmacol 1969; 15:143-51. 11. Sperber I, Sperber G. Hepatic excretion of radiocontrast agents. In: Knoefel PK, ed. International encyclopedia of pharmacology and therapy. New York: Pergamon Press, 1971:165-235. 12. Bartlett GR. Phosphorus assay in column chromatography. J BioI Chern 1959;234:466-8. 13. Lie RF, Schmitz JM, Pierre KJ, et al. Cholesterol oxidase-based determination, by continuous flow analysis, of total and free cholesterol in serum. Clin Chern 1976;22:1627-30. 14. Zak B, Boyle AJ. A simple method for the determination of organic bound iodine. J Am Pharmacol Assoc Sci Ed 1952; 41:260-2. 15. Anderson JM, Dietschy JM. Absolute rates of cholesterol synthesis in extrahepatic tissues measured with 3H-Iabeled water and 14C-Iabeled substrates. J Lipid Res 1979;20:740-52. 16. Folch JL, Lees M, Sloane-Stanley GH. A single method for the isolation and purification of total lipids from animal tissues. J BioI Chern 1957;266:497-509. 17. Sperry WM. Quantitative isolation of sterols. J Lipid Res 1963;4:221-5. 18. Bell GO, Doran J, Fayadh M, et al. Effects of ioglycamide (biligram) on bile flow and biliary lipid secretion in man. Gut 1978;19:300-7. 19. Burk RF, Barnhart JL. Iodipamide hepatotoxicity in the rat. Gastroenterology 1979;76:1363-7. 20. LaRusso NF, Hoffman NE, Hoffman AF, et al. Validity and sensitivity of an intravenous bile acid tolerance test in patients with liver disease. N Engl J Med 1975;292:1209-14. 21. Robins SJ, Armstrong MJ. Biliary lecithin secretion. II. Effects of dietary choline and biliary lecithin synthesis. Gastroenterology 1976;70:397-402. 22. Robins SJ, Brunengraber H. Origin of biliary cholesterol and lecithin in the rat: contribution of new synthesis and performed hepatic stores. J Lipid Res 1982;23:604-8. 23. Talafant E. Bile pigment-phospholipid interaction. Biochim Biophys Acta 1971;231:394-8. 24. Kapoor CL, Murti CRMU, Bajpai PC. Toxic effect of bilirubin on human red blood cell and its reversal by RBC lipids. Indian J Med Res 1972;60:918-28. 25. Small OM. The formation of gallstones. Adv Intern Med 1970;16:243-64. 26. Swell L, Bell CC, Entenmann C. Bile acids and lipid metabolism. III. Influence of bile acids on phospholipids in liver and bile of the isolated perfused dog liver. Biochim Biophys Acta 1968;164:278-84.
Effect of Organic Anions on Biliary Lipids in the Rat MICHAEL D. APSTEIN and SANDER
J.
ROBINS
Boston and West Roxbury Veterans Administration Medical Centers and Boston and Harvard University Schools of Medicine, Boston, Massachusetts
Bile salts enhance the biliary secretion of phospholipid and cholesterol. Other amphipilic molecules, organic anions, are secreted into bile as well. We studied the effects of bilirubin and iodipamide, two chemically dissimilar organic anions, on biliary lipid secretion in the rat. We infused bile salt pooldepleted rats with a stepwise infusion of taurocholate and a constant infusion of organic anion . Both organic anions markedly inhibited the biliary secretion of phospholipid and cholesterol without affecting bile salt secretion. This inhibition, at least with iodipamide , was dose-dependent and fully reversible. Using tritiated water as a precursor, we measured hepatic and biliary cholesterol synthesis in the presence or absence of an iodipamide infusion to see if decreased lipid synthesis couid explain decreased secretion. Despite the marked reduction in biliary cholesterol secretion, the specific activity of biliary cholesterol was unchanged during an iodipamide infusion. We suggest that organic anions interfere with the assembly of the biliary mixed micelle resulting in micelles that are deficient in phospholipid and cholesterol .
Micelle-forming bile salts can stimulate the biliary secretion of phospholipid and cholesterol (1). In addition, these bile salts can enhance the excretion of some (2,3) , but not all (4) , organic anions. Although organic anion excretion is noncompetitive Received September 15 , 1980. Accepted June 23 , 1982. Address requ ests for reprints to: Michael D. Apstein, M.D. , Division of Gastroenterology, West Roxbury Veterans Administration Medical Center, West Roxbury, Massachusetts 02132 . Dr. Apstein is the recipient of a Veterans Administration Career Development Award. This paper was presented in part at the meetings of the American Gastroenterological Association in New Orleans , La., May 19 79 , and a preliminary report was published in abstract form (GASTROENTEROLOGY 1979;76:1274). The authors thank Arthur Taylor for his help with the statistical analysis. © 19 82 by the American Gastroenterological Association 0016-5085 /82/1111 20-07$02.50
with biliary bile salt secretion and presumably along separate pathways (5), recent data suggest that at least one organic anion, bromsulphalein (BSP), inhibits biliary phospholipid and cholesterol secretion (6). We have studied the effects of two chemically dissimilar organic anions, bilirubin and iodipamide, on micellar lipid secretion in the rat. We found that both bilirubin and iodipamide depress biliary phospholipid and cholesterol secretion without affecting bile salt secretion. We then examined the effects of iodipamide on the synthesis of hepatic and biliary cholesterol and found that although biliary cholesterol secretion was inhibited its synthesis was unimpaired.
Methods and Materials Animals and Preparation Male, Sprague-Dawley rats (250-350 g) were maintained on an ad libitum Chow diet (Purina, St. Louis, Mo.). Animals were anesthetized with pentobarbital (45 mg/kg, intraperitoneally) and the bile duct and two iliac veins were cannulated. Animals were restrained in cages and allowed to recover from surgery for the next 18 h. Hydration was maintained by intravenous saline infusion (0 .9% at 0.9 ml/h) and body temperature maintained at 37°C with incandescent lamps (7).
Infusates The bilirubin solutions were made in the dark just before use by adding 170 /-Lmol of unconjugated bilirubin (Sigma Chemical Co., St. Louis , Mo .) to 50 ml of 0.45% NaHC0 3 /0.45% NaC!. Iodipamide (Cholegrafin, E. R. Squibb & Sons, Inc., Princeton, N.J.) was used as supplied by the manufacturer. Control infusions were either 0.9% NaCl or 0.45% NaHC0 3 /O.45% NaC!. Since results using either of the control solutions were not significantly different, they have been combined in all analyses. Sodium taurocholate (Calbiochem-Behring Corp., La Jolla, Calif., A grade, >99% pure) was infused as 10 mM in 0.9% NaC!. In early experiments we found that iodipamide interferes with enzymatic determination of bile salt. That is, when
November 1982
iodipamide in concentrations found in bile was added to salt standards and bile salt was measured by an enzymatic method (8). the optical density was increased by 50%100%. In fact, iodipamide alone has an optical density at 340 nm, the wavelength at which the enzymatic method of Talalay (8) is read. Therefore, in iodipamide and some control-infused animals, 0.75 /LCi (40 mCi/mmol) of [2414C]taurocholate (Mallinckrodt Inc., St. Louis, Mo., purified to >95% by thin-layer chromatography) was added to the bile salt infusates, and bile salt secretion was measured by recovery of radioactivity.
Experimental Design After 18 h of bile salt pool depletion when bile salt output is constant and at its nadir (9), rats were infused through one iliac vein with taurocholate in a stepwise fashion at rates of 108, 216, and 432 nmol/min. They were simultaneously infused through the other iliac vein with either a bilirubin solution that was kept in the dark (3.4 /Lmol/100 g, bolus followed by a constant infusion of 250 nmol/min· 100 g) (10). iodipamide (500 nmol/min . 100 g) (11), or a comparable volume of 0.9% NaCl (or 0.45% NaHC0 3/0.45% NaCl). Bile was collected on ice in tared tubes (and in the dark for the bilirubin-infused animals) for five 10-min periods before infusion of either anion or bile salt, and at each rate of bile salt infusion. The first two collections of each infusion period were allowed for equilibration. The next three collections were used to calculate biliary outputs since there was <10% variation between collections. Iodipamide-infused rats were further studied to determine the extent of reversibility of biliary lipid changes. In this group, after the stepwise taurocholate infusion, taurocholate was continued at 432 nmol/ min, iodipamide was replaced by saline, and bile was collected for an additional 26 h. To determine the dose-response effect of iodipamide on micellar lipid secretion, separate groups of 4 rats were infused with taurocholate at a constant rate of 432 nmol! min. After biliary lipid excretion was constant for 60 min, either a 0.9% saline or iodipamide infusion at 25, 100, 250, or 500 nmol/min . 100 g was begun and bile was collected for an additional 90 min. To measure hepatic and biliary cholesterol synthesis, * two additional groups of 4 animals each were bile salt pool depleted as outlined above. They were infused intravenously with taurocholate at a constant rate of 432 nmol/ min. After biliary lipid excretion had reached a steady state for 60 min, 30 mCi of tritiated water (New England Nuclear, Boston, Mass.) was given intravenously and bile was collected in 3D-min periods for 2.5 h. Then either an iodipamide solution (500 nmol/min . 100 g) or a 0.9% NaCl solution was begun and bile was collected for an additional 2.5 h. At the end of the 5 h, the animals were rapidly reanesthetized and the liver was flushed with cold saline through the portal vein and excised. * Biliary cholesterol synthesis refers to the amount of newly synthesized (labeled) cholesterol appearing in bile.
ORGANIC ANIONS AND BILIARY LIPIDS
1121
Analytical Methods Bile volumes were determined gravimetrically. Phospholipid was measured by the method of Bartlett (12), and cholesterol by the method of Lie et al. (13). Iodipamide was measured as iodine by the method of Zak et al. (14). Bile salt output was measured by the enzymatic method of Talalay (8) in the bilirubin-infused animals, by bile radioactivity in the iodipamide-infused animals, and by both methods in the control-infused animals. Bile radioactivity for 14C was measured by scintillation counting using an internal standard to correct for quenching. Cholesterol synthesis was measured by the rate of tritiated water incorporation into digitonin-precipitable sterols in each half-hour bile collection and in the liver at the end of the experiment (15). Radioisotope incorporation was determined from samples that were first extracted for lipid by the method of Folch et al. (16). Cholesterol was precipitated as the digitonide after saponification. Precaution was taken to remove traces of precursor tritiated water, which could combine with the digitonin precipitate, by splitting the sterol-digitonin precipitate in pyridine and then extracting the free sterol with ether (15,17). Recovery of [7- 3Hlcholesterol (New England Nuclear) added to control bile was 98 ± 3%. 3H Radioactivity was measured by scintillation counting with internal standard to correct for quenching.
Statistical Analysis Values are presented as the mean ± SEM. Student's t-test for unpaired observations was used to compare groups and p values of <0.05 were considered to be statistically significant.
Results Figure 1 shows the relation of bile salt secretion to bile salt infusion. After biliary drainage, but before the bile salt infusion was started, baseline bile salt secretion was similar in all three groups (averaging 50 nmol/min). After the infusion of taurocholate, biliary bile salt secretion increased linearly to a similar extent in control-, bilirubin-, and iodipamide-infused groups. Whether bile salt was measured by the enzymatic method or by recovery of radioactivity, bile salt secretion equaled the sum of
Table 1. The Effect of Iodipamide on the Molar Ratio of Biliary Lipids Iodipamide infusion rate (nmol/min . 100 g)
o 25 100 250 500
Molar ratio of biliary bile saltl phospholipidlcholesterol 66:11:1 83:12:1 113:12:1 159:11:1 489:27:1
1122
APSTEIN AND ROBINS
GASTROENTEROLOGY Vo!' 83, No.5
-e
1
500
/~
c
••o ec
// ./ /'/ . /
400
..... Figure 1. The relation of biliary bile salt secretion to bile salt infusion. In the control-infused group (n = 4), biliary bile salt secretion was measured both enzymatically (ENZ, closed circle) and using [24- 14 C]taurocholate (14 C_ TC, open circle). There was no significant difference between groups in bile salt secretion at any given rate of infusion (x ± SEM).
I
...oew
a:: 300
U
T /;,~ ,6" //// CONTROL
200
w
...
/'V'
,'~/'i/
"/
...I
oe
.,'/ "
(/)
w
...I
/j~'
/~
...
w )(
/<,/;/./ ./
///
/:;/
Z
o
a::
//
/;/ /1 /
......
/
100
INFUSED (n=4) (14C-TC)
0
(ENZ)
•
BILIRUBIN INFUSED (n=6) IODIPAMIDE INFUSED (n=4)
0
A
III
0~1--------~-------4
o
100
________- L________~______~1
200
300
BILE SALT INFUSION RATE
baseline excretion plus the taurocholate infusion rate. After bile salt pool depletion. bile volume was similar in all groups and averaged 5.0 ILl/min' 100 g. In the saline-infused group in response to increasing bile salt infusion. bile volume increased by 13%. 23%. and 38%. respectively. In the bilirubin-infused group. the increase was similar: 23%. 28%. and 46%. respectively. In the iodipamide-infused group. the increase in bile volume was 143%. 160%. and 164%. respectively. Both organic anions markedly inhibited the biliary secretion of phospholipid and cholesterol (Figure 2). The first set of points in each panel shows that after bile salt pool depletion and before the start of either bile salt or anion infusion. the biliary secretion of cholesterol and phospholipid was the same in all groups. In controls. in response to increasing bile salt infusion. there was the expected and progressive rise of cholesterol and phospholipid secretion. However. in contrast. in both anion-infused groups. the secretion of cholesterol and phospholipid was significantly less (p < 0.01) than controls at all rates of bile salt infusion. To determine if this inhibition of secretion could be reversed. the experimental period was prolonged
400
500
(nmol •• /mln)
in 4 animals that had been previously infused with iodipamide and bile was collected for an additional 26 h. The results are shown in Figure 3. Before either bile salt or iodipamide infusion. the biliary secretion of phospholipid and cholesterol reflected the bile salt-depleted state of the animals. Within 20 min of the start of the iodipamide infusion. iodipamide secretion reached a maximum. rema.ined unchanged as the bile salt infusion rate was increased. and then slowly fell over the remainder of the study period. Phospholipid and cholesterol secretion. both of which were depressed by iodipamide. began to rise only when iodipamide secretion diminished. At about 12 h after iodipamide infusion was discontinued. they had reached values comparable to those in the control group. As shown in Figure 4. the inhibition of biliary phospholipid and cholesterol secretion by iodipamide is dose-dependent. With each increase in the iodipamide infusion rate there is a significant decrease in phospholipid and cholesterol secretion without any decrease in bile salt secretion. The resulting change in the molar ratio of biliary lipids is shown in Table 1. Depending on the infusion rate of iodipamide. the molar ratio of bile salt/phospholipid/cholesterol varies from 66: 11 : 1 to 489: 27: 1.
November 1982
ORGANIC ANIONS AND BILIARY LIPIDS
1123
Control Infused
8
n=8
6
...o
QI
c:
4
tJ)
QI
E
o
"'-
.J::.
QI
U
tJ)
Bilirubin Infused
2
o
E c:
lodipamide
n=4
z
o
120
~
w a: u
Control Infused
x!SEM
n=8
100
><
w
tJ)
> a:
'0
..J
.J::.
a.
«
80
Figure 2. The effect of bile salt infusion on biliary phospholipid and cholesterol secretion in bilirubin- (n = 4), iodipamide- (n = 6), and control-infused rats (n = 8). There was no difference (p > 0.05) in biliary lipid secretion between groups before bile salt infusion was started. Phospholipid and cholesterol secretion were markedly (p < 0.01) inhibited by either anion at each subsequent rate of bile salt infusion (iC ± SEM).
0
a. 60
!XI
tJ)
0 .J::. Q.
40
lodipamide
Infused
n=4
20 0
•
Bilirubin Infused
0
108
n=6
216
432
BILE SALT INFUSION RATE (n moles/min)
..J
Oc:
:fiE til 111_
6 4
...I02~ OE ~c:
zOO
~ ~.5 III ~~ a: ~.!!
o
Q.
0
><(I)E Oc
III
> a:
if
,~/________________________
120 80
40~
0 CJ 600
1118 ~ i~400 «
~~200
-G) cOO -Ec:
0 TAUROCHOLATE INFUSION 218
o
1
.32
2
n mol•• /mln
/ 3 TIME
6 HOURS
14
22
30
Figure 3. The relation of biliary secretion of iodipamide, phospholipid, and cholesterol over time (n = 4, iC ± SEM). Biliary phospholipid and cholesterol secretion was depressed during the period of iodipamide infusion and returned to control values after 12 h as the secretion of iodipamide decreased. The transient rise of phospholipid and cholesterol seen immediately after the start of the iodipamide infusion is unexplained, but may represent "washout" of existing lipid from the biliary system.
1124
GASTROENTEROLOGY Vo\. 83, No. 5
APSTEIN AND ROBINS
Figure 5 demonstrates the relation of iodipamide dose to biliary secretion of micellar lipid with time. In all groups the increase in cholesterol, phospholipid, and bile salt secretion in response to taurocholate infusion is the same. However within 45 min of the start of the iodipamide infusion, there is a dosedependent, sustained inhibition of biliary phospholipid and cholesterol while the biliary secretion of bile salt remains unimpaired. To determine if this inhibition of lipid secretion were due to decreased lipid synthesis, we measured the incorporation of tritiated water into hepatic and biliary cholesterol in the presence and absence of an iodipamide infusion. Figure 6 shows that the specific activity of biliary cholesterol increases after the intravenous administration of tritiated water. The specific activity of biliary cholesterol in the animals infused with iodipamide is the same as that in saline-infused controls. Similarly the amount of tritium incorporated into liver cholesterol is not significantly different between the two groups (7.5 ± 3.4 ILmol tritium incorporated/g dry wt· h in salineinfused animals vs 8.9 ± 1.1 in iodipamide-infused animals) .
Discussion We found that both bilirubin and iodipamide, two chemically dissimilar organic anions, markedly inhibit both biliary cholesterol and phospholipid secretion without affecting bile salt secretion. In addition, at least for iodipamide, this inhibition is reversible and dependent on the amount of iodipamide infused. Although the biliary secretion of organic anions and bile salt is noncompetitive (5), it is clear that both influence biliary phospholipid and cholesterol secretion. This phenomenon is not just limited to the rat; a depression of phospholipid and cholesterol secretion has been reported in patients receiving ioglycamide, an organic anion similar to iodipamide (18), and in dogs receiving BSP (6). There are several explanations for the observed depression of phospholipid and cholesterol secretion in the presence of organic anions. First we showed that these anions do not interfere with biliary lipid analysis by observing no change in either phospholipid or cholesterol concentrations after direct addition of anions to normal bile. Although possible, we do not believe that organic anions produce hepatotoxicity which results in depression of biliary cholesterol and phospholipid secretion. Even though iodipamide has been shown to be hepatotoxic in the rat, evidence of toxicity occurs only at doses much higher than we used (19). We are aware of no studies that demonstrate that
500
--E c
•
"'-
... ....I
cI:
400
II)
w 300
....I
III
G)
0
200 100
E
C
....I
z
0
.... W
0
II: W
...
II)
w
a:
....I
w
0
0
0
:r
UJ
> a:
80 II)
c(
C
..J
Do
m
;[
~
0
:r
60 40
Do
II)
0
:r
20
Do
0
0
25
100
250
500
IODIPAMIDE INFUSION RATE (nmol.s/minl 1 009)
Figure 4. The effects of iodipamide dose on biliary lipid secretion (n = 4, x ± SEM) . With increasing iodipamide dose, there is increasing inhibition of phospholipid and cholesterol secretion despite constant bile salt secretion.
unconjugated bilirubin at the rates of infusion we used is hepatotoxic. Furthermore, bile salt clearance by the liver is a good indicator of hepatic function (20). Figure 1 shows that the bile salt secretion rate equaled the sum of the infusion rate plus baseline secretion and was not different in the iodipamide-, bilirubin-, and control-infused groups. The inhibition of biliary phospholipid and cholesterol cannot be explained by depressed bile flow as bile volumes were similar in controls and bilirubininfused groups and were markedly increased in the iodipamide-infused group. The difference in bile volume in response to bilirubin and iodipamide infusions may reflect differences in the ability of organic anions to form self-aggregates or aggregates with the biliary mixed micelle. High levels of organic anions could depress cholesterol synthesis making it unavailable for biliary secretion. However, we showed that there was not a significant difference in the synthesis of either hepatic or biliary cholesterol in the presence or absence of an iodipamide infusion. Consequently, depressed cholesterol synthesis does not explain the inhibition of biliary cholesterol secretion. Perhaps lecithin synthesis was inhibited, resulting in de-
ORGANIC ANIONS AND BILIARY LIPIDS
November 1982
1125
TAUROCHOLATe INfUSION (432 ft",ol •• '.I.) 10DII'AliliDE INI'UIION
nmol.';min/100g 0 -0
100 ____ A 500 _ .-.
500
....
.E E
.,.,
'-
...C ~
II)
...
400 300
1&1
II
200
0
100
c
0
E
Figure 5. The relation of iodipamide dose to biliary lipid secretion over time (n = 4, x :t SEM). There is a prompt inhibition of phospholipid and cholesterol secretion, but no change in bile salt secretion, in response to an iodipamide infusion. This inhibition is dose-dependent and sustained over the time of infusion.
Z
0
t-
W
~
(J
W
e
...0 lao
80 60
~
:z: lao 40 en 0 :z: 20
-'
0
U)
>
tC
lao
ID
8
...0
e
en
~
4
0
2
III: 1&1
... 1&1
:z: U
0
\
105
150
MINUTES
creased biliary lecithin secretion which in turn depressed cholesterol secretion (21). We believe this chain of events is unlikely because newly synthesized lecithin accounts for only 3% of biliary lecithin secretion in the rat (22) . It is possible that organic anions compete with phospholipid and cholesterol uptake by the liver, decreasing the availability of these lipids for biliary secretion. This seems highly unlikely since in studies of biliary lipid secretion using the isolated perfused rat liver, lipid excretion was not dependent on the lipid content of the perfusion medium (22). We believe that there is a physical interaction between organic anions and biliary lipids that inhibits secretion of phospholipid and cholesterol. In vitro studies demonstrate that this kind of interaction may occur. Binding of both conjugated and unconjugated bilirubin with various phospholipids
has been described (23). Conjugated bilirubin was bound more strongly than unconjugated bilirubin and maximal binding was attained with lecithin, the major phospholipid in bile. Other investigators found that bilirubin binds to red blood cells and causes leakage of hemoglobin and glucose 6-phosphate dehydrogenase (24). The binding can be reversed by exposing bilirubin-red cell complexes to lipid micelles and can be prevented by exposing the bilirubin to the phospholipid fraction of red cell membrane before incubating with intact red cells. If, as suggested by Small (25), the translocation of bile salt leaches lecithin and cholesterol from the canalicular membrane, then a high concentration of organic anions in the membrane could compete with lecithin and cholesterol for incorporation into the biliary mixed micelle. If, as others have suggested (26), the macromolecular complex of mixed micelle
1126
GASTROENTEROLOGY Vol. 83, No. 5
APSTEIN AND ROBINS
•
• Controls o lodipamide
/
/
/
~
/
/
/
/
/0
;I
;I
;I
/
. /
/
/
/
/
.
/
0
/
/
.
0---
/'
/'
/
/
_ -cr
/'
/ //0
/
/
/
\y =2931 + 3453 LnX R2 =0.97
o
2
3
4
5
HOURS Figure 6. The effect of an iodipamide infusion of bile cholesterol specific activity. Tritiated water was administered at time O. In the iodipamide-infused group (n = 4), the iodipamide infusion was begun 2.5 h later. The salineinfused controls (n = 4) are shown by the closed circles. The regression line is calculated from the control data points only. The dashed lines represent 95% confidence bands. Bile cholesterol specific activity of the iodipamide-infused animals falls within these confidence bands. There is no difference in bile cholesterol specific activity after the start of iodipamide infusion when compared to controls.
is assembled in the hepatocyte before excretion, the high concentration of organic anions in the cytosol could displace lecithin and cholesterol from the micelle after it was assembled. The exact mechanism by which organic anions inhibit micellar lipid excretion is unknown. It is possible that they interfere with the intrahepatic assembly of mixed micelles, or perhaps disrupt previously assembled ones, resulting in biliary mixed micelles which are deficient in cholesterol and phospholipid.
References 1. Hardinson WGM, Apter JT. Micellar theory of biliary cholesterol excretion. Am J Physiol 1972;222:61-7. 2. Berk RN, Goldberger LE, Loeb PM. The role of bile salts in the hepatic excretion of iopanoic acid. Invest RadioI1974;9:7-15. 3. Goresky CA, Haddad HH, Kluger WS, et al. The enhancement of maximal bilirubin excretion with taurocholate-induced increments in bile flow. Can J PharmacoI1974;52:389-403. 4. Loeb PM, Barnhart JL, Berk RN. Iotroxamide-a new intravenous cholangiographic agent. Comparison with iodipamide and the effect of bile salts. Radiology 1977;125:323-9.
5. Wheeler HO. Secretion of bile. In: Schiff L, ed. Diseases of the liver. Philadelphia: Lippincott, 1975:87-110 . 6. Shaffer EA, Preshaw RM. Effects of sulfobromophthalein excretion on biliary lipid secretion in humans and dogs. Am J Physiol 1981;240:G85-9. 7. Roberts RJ, Klassen CD, Plaa GL. Maximum biliary excretion of bilirubin and BSP during anesthesia-induced alterations of rectal temperature. Proc Soc Exp BioI Med 1967;125:313-6. 8. Talalay P. Enzymatic analysis of steroid hormones. Methods Biochem Anal 1960;8:119-43. 9. Myant NB, Eder HA. The effect of biliary drainage upon the synthesis of cholesterol in the liver. J Lipid Res 1961;2:363-8. 10. Klassen CD, Roberts RJ, Plaa GL. Maximum biliary excretion of bilirubin and BSP during various rates of infusion in rats of different weights and strains. Toxicol Appl Pharmacol 1969; 15:143-51. 11. Sperber I, Sperber G. Hepatic excretion of radiocontrast agents. In: Knoefel PK, ed. International encyclopedia of pharmacology and therapy. New York: Pergamon Press, 1971:165-235. 12. Bartlett GR. Phosphorus assay in column chromatography. J BioI Chern 1959;234:466-8. 13. Lie RF, Schmitz JM, Pierre KJ, et al. Cholesterol oxidase-based determination, by continuous flow analysis, of total and free cholesterol in serum. Clin Chern 1976;22:1627-30. 14. Zak B, Boyle AJ. A simple method for the determination of organic bound iodine. J Am Pharmacol Assoc Sci Ed 1952; 41:260-2. 15. Anderson JM, Dietschy JM. Absolute rates of cholesterol synthesis in extrahepatic tissues measured with 3H-Iabeled water and 14C-Iabeled substrates. J Lipid Res 1979;20:740-52. 16. Folch JL, Lees M, Sloane-Stanley GH. A single method for the isolation and purification of total lipids from animal tissues. J BioI Chern 1957;266:497-509. 17. Sperry WM. Quantitative isolation of sterols. J Lipid Res 1963;4:221-5. 18. Bell GO, Doran J, Fayadh M, et al. Effects of ioglycamide (biligram) on bile flow and biliary lipid secretion in man. Gut 1978;19:300-7. 19. Burk RF, Barnhart JL. Iodipamide hepatotoxicity in the rat. Gastroenterology 1979;76:1363-7. 20. LaRusso NF, Hoffman NE, Hoffman AF, et al. Validity and sensitivity of an intravenous bile acid tolerance test in patients with liver disease. N Engl J Med 1975;292:1209-14. 21. Robins SJ, Armstrong MJ. Biliary lecithin secretion. II. Effects of dietary choline and biliary lecithin synthesis. Gastroenterology 1976;70:397-402. 22. Robins SJ, Brunengraber H. Origin of biliary cholesterol and lecithin in the rat: contribution of new synthesis and performed hepatic stores. J Lipid Res 1982;23:604-8. 23. Talafant E. Bile pigment-phospholipid interaction. Biochim Biophys Acta 1971;231:394-8. 24. Kapoor CL, Murti CRMU, Bajpai PC. Toxic effect of bilirubin on human red blood cell and its reversal by RBC lipids. Indian J Med Res 1972;60:918-28. 25. Small OM. The formation of gallstones. Adv Intern Med 1970;16:243-64. 26. Swell L, Bell CC, Entenmann C. Bile acids and lipid metabolism. III. Influence of bile acids on phospholipids in liver and bile of the isolated perfused dog liver. Biochim Biophys Acta 1968;164:278-84.