European Journal of Clinical Investigation (1990) 20, 143-149

Bile acid metabolism in familial dysbetalipoproteinaemia: studies in subjects with the apolipoprotein E-2/2phenotype B. ANGELIN, L. HOLMQUIST*, B. LEIJD & K. EINARSSON, Metabolism Unit, Department of Medicine, Huddinge University Hospital, and *King Gustaf V Research Institute, Karolinska Institute, Stockholm, Sweden Received 2 May 1989 and in revised form 26 September 1989

Abstract. Bile acid kinetics and biliary lipid composition were determined in seven subjects with primary dysbetalipoproteinaemia. They were all homozygous for the apolipoprotein E isoform E-2 and six of them were hyperlipidaemic (type I11 hyperlipoproteinaemia). With or without hyperlipidaemia,the apo E-2/2 phenotype was associated with increased bile acid formation (mean increase compared with 32 normolipidaemic controls, 43%; P < 0.025). The biliary lipid composition was not different from that seen in the controls. The results indicate that the uptake by the liver of apo E-containing remnant particles is of importance for the regulation of hepatic cholesterol metabolism in man. It is suggested that hepatic cholesterol synthesis is stimulated in dysbetalipoproteinaemia, and that this leads to a compensatory increase in bile acid synthesis. Keywords. Bile acids, biliary lipids, cholesterol, dysbetalipoproteinaemia, very low density lipoprotein. Introduction Familial dysbetalipoproteinaemia is an inherited disturbance of the metabolism of plasma lipoproteins [1,2]. In its most striking form, type I11 hyperlipoproteinaemia, it is associated with a considerably increased risk of premature vascular disease. The principal abnormal lipoproteins occurring in this disease are called 8-very low density lipoproteins (8VLDL), and they differ from normal very low density lipoproteins (VLDL) by their increased cholesterol to triglyceride ratio, by their P-electrophoretic mobility, and by their enrichment in apolipoprotein (apo) E. 8VLDL have been shown to include two distinct subclasses of lipoproteins: intestinal p-VLDL or chylomicron remnants, and hepatic P-VLDL or VLDL remnants [2]. The accumulation of these particles, which does not occur in normals, is the consequence of Correspondence: BO Angelin, MD. Department of Medicine, Huddinge University Hospital, S-141 86 Huddinge, Sweden.

the presence of a specific variant isoform of the E apoprotein referred to as apo E-2 [3,4]. This apolipoprotein isoform, which differs from the normal E-3 isoform by only a single amino acid substitution, has a markedly reduced affinity for binding to cell surface receptors recognizing the B and/or E apoproteins [5,6]. Recent genetic studies have indicated that apo E phenotypes are determined by two codominant alleles, and that three major isoforms (E-2, E-3, E-4) exist [2,7]. Normolipidaemic subjects with E-2 homozygosity occur with a frequency of x 1% in the general population; such individuals display reduced binding characteristics of their apo E and /3-VLDL, and thus have dysbetalipoproteinaemia but not hyperlipoproteinaemia [2]. Instead, their levels of low density lipoproteins (LDL) are generally lower than those found in the general population [8,9]. This has been hypothesized to be the consequence both of a defective conversion of VLDL to LDL, and of an hcreased removal of plasma LDL. Animal experiments indicate that the influx of remnant particles to the liver is Of importance for the regulation of hepatic cholesterol metabolism Thus, a diminished uptake of nmnant cholesterol leads to an accelerated synthesis rate ofcholesterol and to an increased expression of hepatic LDL receptors. It is presently not known whether remnant Particles have similar effects in man. Dysbetalipoproteinaemia Provides a suitable model for studying this Problem [8,91. Thus, we hypothesized that homozygosity for the aPo E-2 isofom would be associated with a defective down-regulation of the hepatic cholesterol synthesis and consequently an increased catabolism of cholesterol to bile acids. In the present work, we have studied the kinetics of the two primary bile acids, cholic acid and chenodeoxycholic acid, in patients with primary dysbetalipoproteinaemia. In addition, we have determined the influence of this condition on biliary lipid COmPoSitiOn. The results indicate that bile acid synthesis is moderately increased in dysbeblipoproteinaemias presumably reflectingan increased hepatic cholesterol synthesis.

143

144

B. ANGELIN et al. Table 1.Clinical data on patients with familial dysbetalipoproteinaemia

Patient number

Sex*

Body Plasma Plasma VLDL Age weight Relativet cholesterol triglycerides cholesterol/ Previous history, (year) (kg) (%) mmol I-' mmol I-' triglyceride present symptoms

I VG 2KA

F F

68 62

62 75

93 109

19.0 13.9

15.0

3MBB

F

53

63

113

11.6

5.7

I .30

4HF 5MB 6BEM 7BM

M M M M

53 36 32 42

85 90 85 78

99

8.7 15.8 14.9 5.8

2.9 7.1 9.0 I .8

1.43 1.70 1.42 I 40

Controlst F(14) 4 9 f 3 M(18) 3 9 f 3

61k2 76f2

107

108 108

94k2 99*3

1.54

7.5

5.9f0.3 5.7f0.3

Hypertension Claudication, xanthoma Angina pectoris, xanthoma Cholecystectomy Xanthoma Xanthoma Normal

1.2iO.1 1.1iO.1

F, female; M, male. weight (kg) x 100%. height (cm) - lOq 2 Meanf SEM. Numbers of subjects within parentheses.

t Calculated as

Patients and methods Patients

The study included seven patients (four males and three females) with dysbetalipoproteinaemia (Table I). Their ages ranged from 32 to 68 (mean age, 49 years), and they were not obese. None of the patients showed clinical or laboratory evidence of intestinal, hepatic, or renal disease, hyper- or hypothyroidism, or overconsumption of ethanol. Evidence of ischaemic heart disease was present in two subjects. No drugs or diets known to influence lipid metabolism had been given prior to the study. Cholecystograms and/or examinations with ultrasound had not shown gallstones or cholecystopathy in any of the individuals studied; patient no. 4 had been cholecystectomized many years previously, however. Dysbetalipoproteinaemia was defined by the presence of /3-VLDL upon agarose gel electrophoresis, a cholesterol to triglyceride ratio in VLDL of more than 0.75, and an apo E-2/2 phenotype [1,2]. Six of the patients had hyperlipidaemia and thus hyperlipoproteinaemia type 111; four of those displayed palmar and eruptive xanthomas. Subjects nos 6 and 7 were brothers, both of them being homozygous for the apo E-2 phenotype, but only one of them (no. 6) had hyperlipidaemia. This subject was studied on two occasions: in the basal state, and after 1 year of therapy with nicotinic acid. A total of 32 normolipidaemic, non-obese controls without gallbladder disease (18 males and 14 females) were studied (Table I). The results on some of these subjects have been reported previously [1 1,121.

Experimental procedure

The patients were hospitalized in a metabolic ward during the study and were given a standardized diet of natural type for 4-7 days before and during the period of investigation [13]. Constant body weight was maintained, and the daily intake of cholesterol was about 0.5 mmol(200 mg) in each subject. The controls, who were given the same diet, were followed closely as outpatients at the metabolic ward. Bile acid kinetics were determined as described previously [l l-131. The sodium salts of ['4C]-cholic acid (4 pCi) and ['4C]-chenodeoxycholicacid (4 pCi) were dissolved in water and given orally in the morning before breakfast. Four samples of fasting duodenal bile were collected at intervals of 2-4 days. Cholecystokinin was administered intravenously, and about 5 ml of concentrated bile were obtained through a thin polyvinyl tube. The specific radioactivities of cholic acid and chenodeoxycholic acid were determined in each sample. Biliary lipid composition was analysed in 2-4 of the samples. Venous blood samples were drawn repeatedly during the study and analysed for cholesterol, triglycerides and lipoprotein pattern. Informed consent was obtained from all subjects and research was carried out according to the Declaration of Helsinki. The ethical aspects of the study were approved by the Ethical Committee of Karolinska Institute, Stockholm, Sweden. Materials

[24-'4Cl-cholic acid (1 38 pCi mg-I) and [24-I4C]chenodeoxycholic acid (1 38 pCi mg- I ) were obtained

BILE ACID METABOLISM IN DYSBETALIPOPROTEINAEMIA from New England Nuclear Corp. (Boston, MA, USA). The radiochemical purity, about 99%, was checked by autoradiography of thin-layer chromatograms. Cholecystokinin was obtained from AB Kabi Diagnostics (Stockholm, Sweden). A purified 3 a-hydroxysteroid dehydrogenase preparation, supplied as a kit (Sterognost-3a), was purchased from Nyegaard & Co. A/S (Oslo, Norway). Methods

Cholesterol and triglycerides were determined with a Technicon Autoanalyzer (Technicon Instruments Corp., Tarrytown, NY, USA). Plasma lipoprotein separation was performed by electrophoresis on agarose gel [I41 and by ultracentrifugation [ 151. Isoelectric focusing of apo E in VLDL

Blood was collected into heparin-containing vacuum tubes by venipuncture in the morning after fasting for 12 h, and plasma was immediately separated at 4°C. Solutions containing 270 mmol I-' EDTA and 50 mmol I-' merthiolate were then added to a final concentration in plasma of 1-3 and 0.25 mmol I - I , respectively. VLDL were isolated by density gradient ultracentrifugation [161. Apolipoproteins were extracted from VLDL (0.5 mg protein ml-I) by treatment with aqueous isopropanol [ 171. Phospholipids were removed from the isopropanol extract by shaking with 2 vol. of ethyl ether. The remaining apolipoprotein solution was dialysed against 8 mol I-' urea, and analysed by polyacrylamide gel isoelectric focusing as previously described [ 181.

145

acidification, the free bile acids were extracted with ethyl ether, methylated and purified by thin-layer chromatography. One aliquot was then quantitated with gas-liquid chromatography after preparation of the trimethylsilyl derivatives, using an 1.5% SE-30 column. Another aliquot was analysed for radioactivity by liquid scintillation and the specific activities of cholic and chenodeoxycholic acids were determined. The pool size and turnover of the bile acids were calculated according to the principles described by Lindstedt [19]. Details of the procedure have been published previously [ 13,201. Determination of biliary lipid composition

The total bile acid concentration was determined enzymatically with a 3 a-hydroxysteroid dehydrogenase method in a portion of each bile sample [21]. Another portion was extracted immediately after collection with 20 vols of chloroform/methanol (2: 1, v/v) and the chloroform phase was analysed with respect to cholesterol [22] and phospholipids [23]. Lipid composition of bile was expressed as molar per cent cholesterol, bile acids and phospholipids. The cholesterol saturation of bile was calculated according to Carey & Small [24], assuming a biliary lipid concentration of 10 g dl-' [25]. Statistical analysis

Data are presented individually or as mean+ SEM. Significancesof differences between groups were tested with Student's t-test, and correlations are expressed as correlation coefficients, r [26]. Results

Measurements of bile acid kinetics

The duodenal bile samples were hydrolysed with 1 mol I - ' KOH in closed tubes for 12 h at ll0"C. After

Cholic acid

Patient

I 2 3

Pool size mmol 0.49 1.86 I .48 2.81 2.52

Chenodeoxycholicacid

Fractional Synthesis turnover mmolday-' day-'

0.39 0.59 0.39 4 1.27 5 1.26 6 1.18 I .09 7 2.41 0.93 Mean f SEM 1.82 f0.31 0.85f0.15 Controls 1.96i0.18 0.66f0.06

In the six patients with type 111hyperlipoproteinaemia, the pool sizes of cholic acid (1.72f0.35 mmol) and chenodeoxycholic acid (1 -61 +@22 mmol) were within

0.80 0-32 0.26 0.45 0.50 0.93 0.39

032k0.09 0.39f0.03

Significantly different from controls, p

2 phenotype.

Bile acid kinetics and biliary lipid composition were determined in seven subjects with primary dysbetalipoproteinaemia. They were all homozygous for ...
640KB Sizes 0 Downloads 0 Views