- Email: [email protected]
Effect of blood high density lipoprotein cholesterol concentration on fecal steroid excretion in humans
Life Sciences, Vol. 32, pp. 2933-2937 Printed in the U.S.A.
Pergamon Press
EFFECT OF BLOOD HIGH DENSITY LIPOPROTEIN CHOLESTEROL CONCENTRATION ON FECAL STEROID EXCRETION IN HUMANS William T. Beher, Alan Gabbard,
Robert A. Norum and Sofia Stradnieks
Department of Medicine Gastroenterology Research Laboratory Henry Ford Hospital Detroit, Michigan 48202 (Received in final form April 15, 1983) Summary Daily excretion of fecal total bile acids and neutral steroids were compared in five controls and two patients with extremely low concentrations of plasma high density lipoprotein (3 to ii mg/dl) and severe atherosclerosis. There was no significant difference in steroid excretion rates in the groups. The predominant bile acid excreted in control feces was deoxycholic acid; lithocholic acid was predominant in the patients. The patients showed no signs of significant liver disease. Epidemiological studies have shown that an inverse relationship exists between the concentration of plasma high density lipoprotein (HDL) cholesterol and development of coronary heart disease (i). Whether this relationship is direct or causal is uncertain. In general, factors associated with increased incidence of coronary heart disease such as maleness, obesity, adult diabetes, hypertriglyceridemia and cigarette smoking, are associated with low plasma HDL cholesterol concentrations. On the other hand, those associated with high concentrations such as femaleness, nicotinic acid and estrogens are generally associated with lower incidence of coronary heart disease. Unfortunately, little is known of the mechanisms responsible for these associations. Several hypotheses have been advanced to explain how high density lipoproteins function in preventing coronary heart disease including (a) blood HDL mobilizes cholesterol from peripheral tissues (2), (b) plasma HDL or similar complexes block the uptake of plasma low density lipoprotein cholesterol by peripheral tissues (3,4), and (c) HDL scavanges cholesterol during normal metabolism of very low density lipoproteins to intermediate and low density lipoproteins in blood and delivers it to the liver for metabolism and excretion (4). Recently two patients with extremely low concentrations of blood HDL cholesterol ( = 6 mg/dl) have come to our attention {5). These subjects have precocious atherosclerosis, tendon xanthomas, and corneal clouding. With the exception of blood HDL concentrations they have normal blood lipid levels. To initiate studies of lipid metabolism in these subjects we have compared fecal neutral steroid and bile acid excretion in the patients and in age and weight-matched controls.
0024-3205/83 $3.00 + .00 Copyright (c) 1983 Pergamon Press Ltd.
2934
IIDL and Fecal Steroid Excretion
Vol.
32, No. 26,
1983
MATERIALS AND METHODS The purity of all chemicals, solvents and enzymes together with their availability were the same as previously reported (6-7). Total and individual fecal bile acids were determined by fluorometric methods described previously (6,7). Fecal neutral steroids were assayed according to Miettinen et al (8). Plasma total bile acids were determined fluorometrically using resazurin reagent according to Mashige et al (9). Plasma lipids were measured by lipid research clinic methods (i0). Experimental: Patients Case No. 1 was a 33-year-old Caucasian female. Examination showed welldefined yellow-orange plaques on her eyelids, neck, chest, arms and back. Cardiac catheterization and angiography revealed a dilated left ventricle and extensive obstructive lesions in all three coronary arteries. There was mild diffuse corneal clouding in both eyes. Serum glutamic oxaloacetic transaminase (SGOT) and glutamic pyruvic transaminase (SGPT) activities were normal. Plasma HDL cholesterol concentrations were extremely low (3-7 mg/dl). Plasma of this subject contained 0.0058, 18, i06, 2.5, 0.6, 0.0, 1.9 mg/dl of apolipoproteins A-l, A-2, B, C-l, C-2, C-3, and E respectively (5). Case No. 2 was a 32-year-old Caucasian female (sister of the above). At age 30, a xanthoma was removed from around the tendons of her left foot. As in Case No. l, by age 31, there was mild diffuse corneal clouding in both eyes. Coronary and ventricular angiography at age 29 showed occlusion of the right coronary artery and severe stenosis of the circumflex coronary artery. The blood pressure, kidney, liver and thyroid function were normal. Plasma HDL cholesterol concentration was very low (4-ii mg/dl}. Plasma of this subject contained 0.0059, 20, Iii, 2.8, 0.2, 0.0, 2.6~ 0.5 mg/dl of apolipoproteins A-l, A-2, B, C-l, C-2, C-3, E, and F respectively (5). Controls These five subjects were age and weight-matched with Cases No. i and 2 and showed no signs of coronary heart disease or xanthomas. SGOT and SGPT activities, 2-hr postprandial bile acid concentrations, and blood lipoprotein levels were normal in each case. Controls were not sex-matched with the patients; however, this parameter has little or no effect on fecal steroid excretion (ii). Protocol Daily excretion of total fecal free and conjugated bile acids and total fecal neutral steroids were compared in two patients and five age-matched controls. Patients and controls consumed diets containing similar amounts of calories and cholesterol one week before and throughout the experiment. To enable estimation of caloric and cholesterol intake, the subjects, after detailed instruction by a nutrltionist, kept careful diaries of food intake. The cholesterol and caloric content of the patients' diets were adjusted to that of the controls by adding eggs and milk to the diet. The weights of patients and controls remained constant during the experimental interval. Quantitative seven-day feces collections were obtained from the two patients and three-day collections from the controls. Cr20_ markers were used to determine fecal flow. The data in Table I compares~several parameters in patients and control subjects.
Vol. 32, No. 26, 1983
HDL and Fecal Steroid Excretion
2935
TABLE I Profile of Subjects Parameter
Controls
Patients
Fasting plasma HDL cholesterol mgldl
50.0 ± 6.3*(20)**
5.5 ± 3.6(8)
Fasting plasma total cholesterol mg/dl
220 ± 60(i0)
176 t 42'(4)
Serum total bile acids (MMoles/liter) (2-hr postprandial)
4.3 + 2.1(i0)
3.2 t 1.8(4)
Dietary cholesterol intake (mMoles/day)
0.69 + 0.12
0.65 ± 0.09
Caloric intake (Kcal/day)
1979 ± 150.0
1886 ± 75
* Standard deviation. ** Number of samples. RESULTS The results of fecal bile acid and neutral steroid excretion rate studies are shown in Table II. There was no significant difference in either the excretion of total fecal bile acids or total fecal neutral steroids since the values for the two patients fell within the 95% confidence bands (mean ± two standard deviations) of the controls. Fig. I shows data on the excretion of individual fecal bile acids by the two groups. Feces from control subjects contained 48.3 ± 7.1% deoxycholic acid and 23.4 ± 8.0% lithocholic acid. Feces from patient No. 1 contained 35.8% deoxycholic acid, and 42.0% lithocholic acid. Feces from patient No. 2 contained 36.3% deoxycholic acid and 40.3% lithocholic acid. Clearly, subjects with low plasma HDL cholesterol excreted significantly more lithocholic acid than deoxycholic acid, while deoxycholic acid excretion dominated in controls. TABLE II Total Fecal Free and Conjugated Bile Acid and Total Neutral Steroid Excretion Subjects
Fecal bile acid excretion mg/day+
Fecal neutral steroid excretion mg/day@
Controls
242 ± 67*(5)**
506 ± 85(5)
Patients with low plasma HDL concentration
(155;239)++
480;560
* Standard deviation. ** Number of samples. + In terms of deoxycholic acid. In terms of cholesterol. ++ Individual values.
2936
lIDl, and
Fecal
Steroid
Excretion
Vol
32
No
26
1983
LOW PLASMA HDL PATIENTS w z w w 0
J
_J [L
c
!
o"
--E
z~ C O N T R O L S
!
-
D B R
A
2
4
6
8
10 12 14 l e 18 20 22 24 2e 28 30 32 34 36
SILICA GEL HR SECTION FIG.
( 0 . 5 cm e a c h )
I
Individual fecal bile acid excretion by controls and patients with low plasma HDL cholesterol concentration. Bile acids separated on activated (100°C for 1-hr) silica gel HR thin-layer c h r o m a t o g r a p h y plates (250 ~). Developing solvent isooctane:ethyl acetate:glacial acetic acid i0:10:2 (v/v). Q u a n t i f i c a t i o n of bile acids on 0.5 cm sections of gel by resazurin method. A) Conjugated bile acids; B) Cholic acid: C) Deoxycholic acid; D) 3 ~ - H y d r o x y - 1 2 - o x o - 5 B - c h o l a n o i c acid, and E) Lithocholic acid. DISCUSSION Interestingly, the rates of fecal bile acid and neutral steroid excretion were similar in controls and subjects with very low concentrations of HDL cholesterol. These o b s e r v a t i o n s do not appear to support observations (12-13) which suggest that the majority of excreted bile acids are derived from free cholesterol transported to the liver incorporated in HDL. However, since excreted bile acids are believed to originate from both cholesterol transported to the liver incorporated in lipoproteins, and newly synthesized liver cholesterol (4), it is possible that the relative c o n t r i b u t i o n s of the two sources of biliary steroids could change under different circumstances. In our patients, almost all of the steroids may be derived from the pool of newly synthesized liver cholesterol. This seems reasonable since bile acids are necessary for normal fat digestion and absorption, as well as representing an important pathway for cholesterol m e t a b o l i s m and excretion. Further studies of cholesterol and bile acid kinetics will be necessary to decide these points• The finding that lithocholic acid excretion dominates in subjects with low plasma HDL cholesterol, while deoxycholic acid excretion dominates in controls is interesting. Lithocholic acid arises from g a s t r o i n t e s t i n a l bacterial m e t a b o l i s m of c h e n o d e o x y c h o l i c acid, while deoxycholic acid is synthesized from cholic acid by the bacteria. Cholic and c h e n o d e o x y c h o l i c acids are the predominant bile acids found in human bile. In patients with low blood HDL
Vol. 32, No. 26, 1983
HDL and Fecal Steroid Excretion
2937
cholesterol, it follows that chenodeoxycholic acid must be the dominant acid in bile. Since there was little or no evidence of liver disease in our patients, which could affect the ratio of cholic to chenodeoxycholic acid in bile, it is possible that the activity of the 12~-hydroxylase which introduces the 12~-hydroxyl group during cholic acid synthesis, is, for some reason, reduced in the patients with low plasma HDL. It is also possible that alternate pathways of chenodeoxycholic acid synthesis are activated in these patients (14). REFERENCES 1. 2. 3. 4. 5.
6. 7. 8. 9. i0. Ii. 12. 13. 14.
H . A . TYROLER, Amer. Heart Assoc. Monograph No. 73, Circulation 62 Suppl. IV (1980). J . A . GLOMSET, Amer. J. Clin. Nutr. 23 i129-i136 (1970). N. E. MILLER, D. E. WEINSTEIN, T. E. CAREW, T. KOSCHINSKY and D. STEINBERG, J. Clln. Invest. 60 78-88 (1977). A. NICOLL, N. E. MILLER and B. LEWIS, Advances in Lipid Research Vol. 17 p 53-I06, Academic Press, Inc. New York (1980). R . A . NORUM, J. B. LAKIER, S. GOLDSTEIN, A. ANGEL, R. B. GOLDBERG, W. D. BLOCK, D. K. NOFFZE, P. J. DOLPHIN, J. EDELGLASS, D. O. BOGORAD and P. ALAUPOVIC, New Engl. J. Med. 306 1513-1519 (1982). W . T . BEHER, S. STRADNIEKS, G. J. LIN and J. SANFIELD, Steroids 38 281295 (1981). W . T . BEHER, S. STRADNIEKS and G. J. LIN, Steroids 39, 313-323 (1982). T . A . MIETTINEN, E. H. AHRENS and S. M. GRUNDY, J. Lipid Res 6 411-424 (1965). F. MASHIGE, K. IMAI and T. OSUGA, Clin. Chim. Acta 70 79-86 (1976). Lipid Research Clinics Program Manual of Laboratory Operations, Dept. of Health, Education and Welfare, Bethesda, MD, Publ. No. NIH-75-628 (1974). H . S . SODHI, B. J. KUDCHODKAR and D. T. MASON. Advances in Lipid Research, Vol. 17, p i07-153 Academic Press, Inc. New York (1980). C . C . SCHWARTZ, L. G. HALLORAN and Z. R. VLAHCEVIC, Science 200 62-64 (1978). C. C. SCHWARTZ, M. BERMAN, Z. R. VLAHCEVIC, L. G. HALLORAN, D. H. GREGORY and L. SWELL, J. Clin. Invest. 61 408-423 (1978). W . T . BEHER, Monographs on Atherosclerosis BILE ACIDS Vol 6, S. Karger! Basel (1976).
Pergamon Press
EFFECT OF BLOOD HIGH DENSITY LIPOPROTEIN CHOLESTEROL CONCENTRATION ON FECAL STEROID EXCRETION IN HUMANS William T. Beher, Alan Gabbard,
Robert A. Norum and Sofia Stradnieks
Department of Medicine Gastroenterology Research Laboratory Henry Ford Hospital Detroit, Michigan 48202 (Received in final form April 15, 1983) Summary Daily excretion of fecal total bile acids and neutral steroids were compared in five controls and two patients with extremely low concentrations of plasma high density lipoprotein (3 to ii mg/dl) and severe atherosclerosis. There was no significant difference in steroid excretion rates in the groups. The predominant bile acid excreted in control feces was deoxycholic acid; lithocholic acid was predominant in the patients. The patients showed no signs of significant liver disease. Epidemiological studies have shown that an inverse relationship exists between the concentration of plasma high density lipoprotein (HDL) cholesterol and development of coronary heart disease (i). Whether this relationship is direct or causal is uncertain. In general, factors associated with increased incidence of coronary heart disease such as maleness, obesity, adult diabetes, hypertriglyceridemia and cigarette smoking, are associated with low plasma HDL cholesterol concentrations. On the other hand, those associated with high concentrations such as femaleness, nicotinic acid and estrogens are generally associated with lower incidence of coronary heart disease. Unfortunately, little is known of the mechanisms responsible for these associations. Several hypotheses have been advanced to explain how high density lipoproteins function in preventing coronary heart disease including (a) blood HDL mobilizes cholesterol from peripheral tissues (2), (b) plasma HDL or similar complexes block the uptake of plasma low density lipoprotein cholesterol by peripheral tissues (3,4), and (c) HDL scavanges cholesterol during normal metabolism of very low density lipoproteins to intermediate and low density lipoproteins in blood and delivers it to the liver for metabolism and excretion (4). Recently two patients with extremely low concentrations of blood HDL cholesterol ( = 6 mg/dl) have come to our attention {5). These subjects have precocious atherosclerosis, tendon xanthomas, and corneal clouding. With the exception of blood HDL concentrations they have normal blood lipid levels. To initiate studies of lipid metabolism in these subjects we have compared fecal neutral steroid and bile acid excretion in the patients and in age and weight-matched controls.
0024-3205/83 $3.00 + .00 Copyright (c) 1983 Pergamon Press Ltd.
2934
IIDL and Fecal Steroid Excretion
Vol.
32, No. 26,
1983
MATERIALS AND METHODS The purity of all chemicals, solvents and enzymes together with their availability were the same as previously reported (6-7). Total and individual fecal bile acids were determined by fluorometric methods described previously (6,7). Fecal neutral steroids were assayed according to Miettinen et al (8). Plasma total bile acids were determined fluorometrically using resazurin reagent according to Mashige et al (9). Plasma lipids were measured by lipid research clinic methods (i0). Experimental: Patients Case No. 1 was a 33-year-old Caucasian female. Examination showed welldefined yellow-orange plaques on her eyelids, neck, chest, arms and back. Cardiac catheterization and angiography revealed a dilated left ventricle and extensive obstructive lesions in all three coronary arteries. There was mild diffuse corneal clouding in both eyes. Serum glutamic oxaloacetic transaminase (SGOT) and glutamic pyruvic transaminase (SGPT) activities were normal. Plasma HDL cholesterol concentrations were extremely low (3-7 mg/dl). Plasma of this subject contained 0.0058, 18, i06, 2.5, 0.6, 0.0, 1.9 mg/dl of apolipoproteins A-l, A-2, B, C-l, C-2, C-3, and E respectively (5). Case No. 2 was a 32-year-old Caucasian female (sister of the above). At age 30, a xanthoma was removed from around the tendons of her left foot. As in Case No. l, by age 31, there was mild diffuse corneal clouding in both eyes. Coronary and ventricular angiography at age 29 showed occlusion of the right coronary artery and severe stenosis of the circumflex coronary artery. The blood pressure, kidney, liver and thyroid function were normal. Plasma HDL cholesterol concentration was very low (4-ii mg/dl}. Plasma of this subject contained 0.0059, 20, Iii, 2.8, 0.2, 0.0, 2.6~ 0.5 mg/dl of apolipoproteins A-l, A-2, B, C-l, C-2, C-3, E, and F respectively (5). Controls These five subjects were age and weight-matched with Cases No. i and 2 and showed no signs of coronary heart disease or xanthomas. SGOT and SGPT activities, 2-hr postprandial bile acid concentrations, and blood lipoprotein levels were normal in each case. Controls were not sex-matched with the patients; however, this parameter has little or no effect on fecal steroid excretion (ii). Protocol Daily excretion of total fecal free and conjugated bile acids and total fecal neutral steroids were compared in two patients and five age-matched controls. Patients and controls consumed diets containing similar amounts of calories and cholesterol one week before and throughout the experiment. To enable estimation of caloric and cholesterol intake, the subjects, after detailed instruction by a nutrltionist, kept careful diaries of food intake. The cholesterol and caloric content of the patients' diets were adjusted to that of the controls by adding eggs and milk to the diet. The weights of patients and controls remained constant during the experimental interval. Quantitative seven-day feces collections were obtained from the two patients and three-day collections from the controls. Cr20_ markers were used to determine fecal flow. The data in Table I compares~several parameters in patients and control subjects.
Vol. 32, No. 26, 1983
HDL and Fecal Steroid Excretion
2935
TABLE I Profile of Subjects Parameter
Controls
Patients
Fasting plasma HDL cholesterol mgldl
50.0 ± 6.3*(20)**
5.5 ± 3.6(8)
Fasting plasma total cholesterol mg/dl
220 ± 60(i0)
176 t 42'(4)
Serum total bile acids (MMoles/liter) (2-hr postprandial)
4.3 + 2.1(i0)
3.2 t 1.8(4)
Dietary cholesterol intake (mMoles/day)
0.69 + 0.12
0.65 ± 0.09
Caloric intake (Kcal/day)
1979 ± 150.0
1886 ± 75
* Standard deviation. ** Number of samples. RESULTS The results of fecal bile acid and neutral steroid excretion rate studies are shown in Table II. There was no significant difference in either the excretion of total fecal bile acids or total fecal neutral steroids since the values for the two patients fell within the 95% confidence bands (mean ± two standard deviations) of the controls. Fig. I shows data on the excretion of individual fecal bile acids by the two groups. Feces from control subjects contained 48.3 ± 7.1% deoxycholic acid and 23.4 ± 8.0% lithocholic acid. Feces from patient No. 1 contained 35.8% deoxycholic acid, and 42.0% lithocholic acid. Feces from patient No. 2 contained 36.3% deoxycholic acid and 40.3% lithocholic acid. Clearly, subjects with low plasma HDL cholesterol excreted significantly more lithocholic acid than deoxycholic acid, while deoxycholic acid excretion dominated in controls. TABLE II Total Fecal Free and Conjugated Bile Acid and Total Neutral Steroid Excretion Subjects
Fecal bile acid excretion mg/day+
Fecal neutral steroid excretion mg/day@
Controls
242 ± 67*(5)**
506 ± 85(5)
Patients with low plasma HDL concentration
(155;239)++
480;560
* Standard deviation. ** Number of samples. + In terms of deoxycholic acid. In terms of cholesterol. ++ Individual values.
2936
lIDl, and
Fecal
Steroid
Excretion
Vol
32
No
26
1983
LOW PLASMA HDL PATIENTS w z w w 0
J
_J [L
c
!
o"
--E
z~ C O N T R O L S
!
-
D B R
A
2
4
6
8
10 12 14 l e 18 20 22 24 2e 28 30 32 34 36
SILICA GEL HR SECTION FIG.
( 0 . 5 cm e a c h )
I
Individual fecal bile acid excretion by controls and patients with low plasma HDL cholesterol concentration. Bile acids separated on activated (100°C for 1-hr) silica gel HR thin-layer c h r o m a t o g r a p h y plates (250 ~). Developing solvent isooctane:ethyl acetate:glacial acetic acid i0:10:2 (v/v). Q u a n t i f i c a t i o n of bile acids on 0.5 cm sections of gel by resazurin method. A) Conjugated bile acids; B) Cholic acid: C) Deoxycholic acid; D) 3 ~ - H y d r o x y - 1 2 - o x o - 5 B - c h o l a n o i c acid, and E) Lithocholic acid. DISCUSSION Interestingly, the rates of fecal bile acid and neutral steroid excretion were similar in controls and subjects with very low concentrations of HDL cholesterol. These o b s e r v a t i o n s do not appear to support observations (12-13) which suggest that the majority of excreted bile acids are derived from free cholesterol transported to the liver incorporated in HDL. However, since excreted bile acids are believed to originate from both cholesterol transported to the liver incorporated in lipoproteins, and newly synthesized liver cholesterol (4), it is possible that the relative c o n t r i b u t i o n s of the two sources of biliary steroids could change under different circumstances. In our patients, almost all of the steroids may be derived from the pool of newly synthesized liver cholesterol. This seems reasonable since bile acids are necessary for normal fat digestion and absorption, as well as representing an important pathway for cholesterol m e t a b o l i s m and excretion. Further studies of cholesterol and bile acid kinetics will be necessary to decide these points• The finding that lithocholic acid excretion dominates in subjects with low plasma HDL cholesterol, while deoxycholic acid excretion dominates in controls is interesting. Lithocholic acid arises from g a s t r o i n t e s t i n a l bacterial m e t a b o l i s m of c h e n o d e o x y c h o l i c acid, while deoxycholic acid is synthesized from cholic acid by the bacteria. Cholic and c h e n o d e o x y c h o l i c acids are the predominant bile acids found in human bile. In patients with low blood HDL
Vol. 32, No. 26, 1983
HDL and Fecal Steroid Excretion
2937
cholesterol, it follows that chenodeoxycholic acid must be the dominant acid in bile. Since there was little or no evidence of liver disease in our patients, which could affect the ratio of cholic to chenodeoxycholic acid in bile, it is possible that the activity of the 12~-hydroxylase which introduces the 12~-hydroxyl group during cholic acid synthesis, is, for some reason, reduced in the patients with low plasma HDL. It is also possible that alternate pathways of chenodeoxycholic acid synthesis are activated in these patients (14). REFERENCES 1. 2. 3. 4. 5.
6. 7. 8. 9. i0. Ii. 12. 13. 14.
H . A . TYROLER, Amer. Heart Assoc. Monograph No. 73, Circulation 62 Suppl. IV (1980). J . A . GLOMSET, Amer. J. Clin. Nutr. 23 i129-i136 (1970). N. E. MILLER, D. E. WEINSTEIN, T. E. CAREW, T. KOSCHINSKY and D. STEINBERG, J. Clln. Invest. 60 78-88 (1977). A. NICOLL, N. E. MILLER and B. LEWIS, Advances in Lipid Research Vol. 17 p 53-I06, Academic Press, Inc. New York (1980). R . A . NORUM, J. B. LAKIER, S. GOLDSTEIN, A. ANGEL, R. B. GOLDBERG, W. D. BLOCK, D. K. NOFFZE, P. J. DOLPHIN, J. EDELGLASS, D. O. BOGORAD and P. ALAUPOVIC, New Engl. J. Med. 306 1513-1519 (1982). W . T . BEHER, S. STRADNIEKS, G. J. LIN and J. SANFIELD, Steroids 38 281295 (1981). W . T . BEHER, S. STRADNIEKS and G. J. LIN, Steroids 39, 313-323 (1982). T . A . MIETTINEN, E. H. AHRENS and S. M. GRUNDY, J. Lipid Res 6 411-424 (1965). F. MASHIGE, K. IMAI and T. OSUGA, Clin. Chim. Acta 70 79-86 (1976). Lipid Research Clinics Program Manual of Laboratory Operations, Dept. of Health, Education and Welfare, Bethesda, MD, Publ. No. NIH-75-628 (1974). H . S . SODHI, B. J. KUDCHODKAR and D. T. MASON. Advances in Lipid Research, Vol. 17, p i07-153 Academic Press, Inc. New York (1980). C . C . SCHWARTZ, L. G. HALLORAN and Z. R. VLAHCEVIC, Science 200 62-64 (1978). C. C. SCHWARTZ, M. BERMAN, Z. R. VLAHCEVIC, L. G. HALLORAN, D. H. GREGORY and L. SWELL, J. Clin. Invest. 61 408-423 (1978). W . T . BEHER, Monographs on Atherosclerosis BILE ACIDS Vol 6, S. Karger! Basel (1976).