Serum squalene in postmenopausal women without and with coronary artery disease

Atherosclerosis 146 (1999) 61 – 64 www.elsevier.com/locate/atherosclerosis

Serum squalene in postmenopausal women without and with coronary artery disease Radhakrishnan A. Rajaratnam, Helena Gylling, Tatu A. Miettinen * Di6ision of Internal medicine, Department of Medicine, Uni6ersity of Helsinki, P.O. Box 340, FIN-00029 HYKS, Helsinki, Finland Received 30 June 1998; received in revised form 18 January 1999; accepted 8 March 1999

Abstract Squalene, found in earlier studies in human atherosclerotic plaques, was measured in the serum of postmenopausal women with coronary artery disease (CAD) (n=25) and randomly chosen age-matched healthy controls (n =30). The squalene concentrations of the whole population ranged from 37.5 to 115.5 mg/dl, and were higher in serum of the CAD than healthy women (91.4 92.6 versus 65.2 9 2.6 mg/dl, P =0.000), a finding observed also in relation to cholesterol (43.8 9 1.8 versus 32.9 9 1.1 102 × mmol/mol of cholesterol, P =0.000). The squalene concentration was also increased in chylomicrons, VLDL and d\1.006 g/ml lipoproteins, and the proportions to cholesterol in VLDL and d\ 1.006 g/ml lipoproteins. The respective squalene and cholesterol concentrations were related to each other in serum, VLDL and d\ 1.006 g/ml lipoproteins (r =0.52, 0.85 and 0.55, respectively), whereas the correlation with triglycerides was seen only in VLDL (r =0.84) over the whole population. Besides enhanced intestinal secretion, it remains to be shown whether higher serum squalene in postmenopausal coronary women is due to increased cholesterol synthesis or a defect in squalene conversion to lanosterol. © 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Postmenopausal women; Cholesterol; Coronary artery disease; Squalene

1. Introduction Squalene is a triterpenoid hydrocarbon intermediate of cholesterol biosynthesis. It is formed from farnesyl pyrophosphate and further undergoes cyclization to lanosterol by enzymes attached to microsomal particles, depending on reduced di- or triphosphopyridine nucleotides [1–3]. The rapid appearance of [14C]squalene in plasma after the intravenous administration of [14C]mevolanate indicates that squalene is immediately involved in cholesterol synthesis [4 – 6]. In addition, changes of cholesterol synthesis induced by hypolipidemic drugs are associated with corresponding changes of plasma squalene concentrations [7 – 9]. In humans, skin and adipose tissue contain high levels of squalene, whereas the liver and large intestine have moderate values [10]. Squalene is bound to sterolcarrier proteins in hepatocytes [11]. In fasting plasma, squalene is mainly transported either in triglyceride-rich lipoproteins [4,6] or in both low density lipoprotein * Corresponding author. Tel.: +358-9-47172220; fax: + 358-947174013.

(LDL) and high density lipoprotein (HDL) [12,13]. During the 24-h period, squalene varies considerably in human very low density lipoprotein (VLDL) and peaks at 04:00 [12]. Postprandial squalene clearance following a fat meal is retarded in type III hyperlipidemic apolipoprotein E 2/2 homozygotes [13,14] and in women with coronary artery disease (CAD) [15]. Squalene has been identified in arterial intima [10] and human atherosclerotic plaques [16]. Additionally, human arterial segments have the ability to produce squalene in vitro [17]. However, fasting serum squalene concentration and its distribution among chylomicrons, VLDL and d \ 1.006 g/ml lipoproteins have not been studied in CAD patients.

2. Methods

2.1. Subjects Twenty five 50–55 year-old postmenopausal CAD women were recruited from the University Central Hospital of Helsinki. Thirty age matched postmenopausal

0021-9150/99/$ - see front matter © 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 0 2 1 - 9 1 5 0 ( 9 9 ) 0 0 1 2 9 - X

62

R.A. Rajaratnam et al. / Atherosclerosis 146 (1999) 61–64

women were randomly chosen from the Helsinki population registry as controls. Postmenopause was defined as amenorrhoea and elevated follicle stimulating hormone (\ 30 U/I). Twelve of the CAD women had had a myocardial infarction earlier. Angioplasty was performed in 13 CAD women and coronary bypass surgery in nine women. The subjects were admitted to the study after at least a 6 month-period since infarction, angioplasty or bypass surgery. The controls were free of chest pain. One CAD and eight healthy women received peroral combined estrogen and progestin, two controls used transdermal estrogen alone and another two controls progestin alone. Hypolipidemic drug users were excluded. b-Blocking agents were used by 15 of the CAD patients and three of the control women, angiotensin converting enzyme inhibitors by one CAD patient and two of the control women, and two controls were on thiazide diuretics. The subjects were advised to consume their normal diet, and dietary fat and cholesterol were measured from 7 day-dietary recalls [18]. The cases and controls had a similar dietary intake of fat (74.29 4.5 versus 78.196.0 g/day), saturated fatty acids (15.5 90.8 versus 15.690.6 E%), monounsaturated fatty acids (13.4 90.6 versus 12.9 9 0.6 E%) mainly from rapeseed oil (virtually none from olive oil), polyunsaturated fatty acids (6.2690.39 versus 5.769 0.40 E%) and cholesterol (275.9920.7 versus 281.4 9 18.7 mg/day).

2.2. Lipoproteins and lipids Blood samples were drawn between 08:00 and 09:00 after a 12 h-fast. Chylomicrons, VLDL and d \1.006 g/ml lipoproteins were separated from plasma by ultracentrifugation in a fixed-angle Ti 50.4 rotor (Beckman Instruments®). For this purpose, 7.2 ml of plasma was overlayered with a salt solution of density 1.006 g/ml and centrifuged at 18 000 rpm for 30 min. The chylomicrons were isolated by aspirating the top 3.6 ml. The infranatant was then mixed with 1.006 g/ml salt solution and centrifuged at 35000 rpm for 18 h to separate VLDL and the bottom d \1.006 g/ml lipoproteins. Cholesterol and triglycerides were measured enzymatically by commercial kits (Boehringer Diagnostica, Germany). Squalene was measured in sera and the lipoproteins by gas liquid chromatography (GLC), as described previously [7,19]. Briefly, after adding 5-acholestane as an internal standard, the samples were saponified with 95% ethanol and KOH. Nonsaponified lipids were extracted by hexane, and squalene was quantitated by GLC on a 50 m long Hewlett Packard® Ultra I column, using commercial squalene to localize the squalene peak. Two baseline serum samples were analyzed for squalene and the average was used in the following analysis.

2.3. Data analysis Mean values were presented with S.E. Group differences were analyzed with Student’s t-test, and a Mann– Whitney rank-sum test was used when the data was not normally distributed. Association among continuous variables was tested with a Spearman rank correlation. Least squares linear regression equations were estimated and the difference between the regression lines was tested with the F-test. PB0.05 was considered significant.

3. Results The mean age and body mass index were similar without and with CAD, 52.99 0.4 versus 52.290.5 years and 27.191.0 versus 25.49 0.7 kg/m2, respectively, but cholesterol in the lipoprotein fraction d\ 1.006 g/ml was higher in the cases compared with the controls (Table 1). Squalene concentrations were normally distributed in serum and in d\ 1.006 g/ml lipoproteins, ranging from 37.5 to 115.5 and 27.6 to 104.6 mg/dl, respectively. The controls without and with hormone replacement had comparable serum squalene concentration (34.691.7 versus 31.49 1.3 mg/dl, P= NS). The percentage distribution of squalene, 5% in chylomicrons, 16% in VLDL and 79% in d\ 1.006 g/ml lipoproteins, was similar in the two groups and it was unaffected by hormone replacement. The decreasing order of the squalene to cholesterol ratio was chylomicrons, VLDL and d\ 1.006 g/ml lipoproteins, that of the squalene to triglyceride ratio was chylomicrons, d\ 1.006 g/ml lipoproteins and VLDL (Table 1). Squalene concentrations of all lipid fractions were clearly higher in the CAD than control women (Table 1), the respective ratios to cholesterol or triglycerides being also elevated in all except the chylomicron fraction of the CAD cases. The squalene and cholesterol concentrations were correlated with each other in serum (r=0.52, PB 0.001), VLDL (r=0.85, PB 0.001) and in d\1.006 g/ml lipoproteins (r= 0.55, PB 0.001), but not in the chylomicrons. In VLDL, the regression equations for cholesterol versus squalene differed (F= 3.84, PB0.05) in the two groups (Fig. 1). Thus, the squalene concentrations were 11.39 1.4 versus 7.479 0.47 mg/dl (PB 0.05) in the CAD versus control women for the lower half of VLDL cholesterol (50.35 mmol/l), the respective values being 18.592.2 versus 19.79 2.8 mg/dl (P= NS) in VLDL cholesterol \0.35 mmol/l. The fasting sterol mixture in VLDL was rich in squalene at the low VLDL cholesterol level (5 0.35 mmol/l) in the CAD versus control women (135.89 12.5 versus 98.39 4.7 mmol/mol of cholesterol, PB 0.01), the re-

R.A. Rajaratnam et al. / Atherosclerosis 146 (1999) 61–64

63

spective ratios being similar in the higher VLDL cholesterol half. Triglycerides were associated with squalene, exclusively in VLDL (r = 0.84, P B0.001).

4. Discussion The present study revealed for the first time that serum squalene levels are higher in postmenopausal women with CAD than in those without CAD, and it may thus relate to the development of CAD. Squalene has been found, in fact, in human atherosclerotic plaques [16], suggesting its association with atherogenicity. Endothelial cells synthesize squalene and cholesterol [17], but deposition of squalene from the circulation to the atheroma could also take place. Lipoprotein particles were richer in squalene in the CAD patients compared with the controls, a finding which Table 1 Squalene, cholesterol and triglyceride distributiona Variable

Serum Squalene (mg/dl) Squalene/cholesterolb Squalene/ triglyceridesb Cholesterol (mmol/l) Triglycerides (mmol/l)

Cases (n= 25)

Controls (n= 30)

P-value

91.49 2.6 43.89 1.8 87.89 5.8

65.292.6 32.99 1.1 65.59 3.6

0.000 0.000 0.008

6.069 0.18 1.279 0.07

5.6290.15 1.229 0.08

NS NS

Chylomicrons Squalene (mg/dl) 4.119 0.32 Squalene/cholesterolb 304.09 48.9 Squalene/ 186.79 31.3 triglyceridesb Cholesterol (mmol/l) 0.04 90.00 Triglycerides 0.03 90.00 (mmol/l) VLDL Squalene (mg/dl) 14.8 91.4 Squalene/cholesterolb 120.6 910.1 Squalene/ 31.59 2.2 triglyceridesb Cholesterol (mmol/l) 0.359 0.03 Triglycerides 0.579 0.06 (mmol/l) d\1.006 g/ml lipoproteins Squalene (mg/dl) 71.39 3.0 Squalene/cholesterolb 52.59 14.9 Squalene/ 158.19 10.0 triglyceridesb Cholesterol (mmol/l) 5.199 0.18 Triglycerides 0.539 0.02 (mmol/l)

3.669 0.37 289.49 57.5 137.3922.9

0.030 NS NS

0.059 0.00 0.049 0.00

NS NS

11.191.3 97.89 3.7 25.69 1.2

0.022 0.042 0.046

0.3190.03 0.50 9 0.05

NS NS

54.59 2.5 32.89 1.4 119.395.3 4.6690.14 0.5390.02

0.000 0.021 0.003 0.022 NS

Values are mean 9 S.E. Squalene to cholesterol and triglyceride ratios are in 102×mmol/ mol of cholesterol and triglycerides, respectively. a

b

Fig. 1. Association of squalene with cholesterol in very low density lipoprotein (VLDL). Closed rings are cases and open rings controls. YCASES =22.9X+6.7 vs. YCONTROLS =36.7X−0.36, P B0.05.

could enhance squalene concentrations in atherosclerotic plaques by lipoprotein uptake. Most of the plasma squalene was carried by d\1.006 g/ml lipoproteins, and the strong correlation between squalene and cholesterol in those lipoproteins suggest that serum squalene is carried mainly by LDL in this population, in agreement with earlier studies [12,13]. LDL has been believed to be atherogenic for many decades. Therefore, it is obvious that squalene could accumulate in the arterial wall also by LDL infiltration. What is then the reason for the high squalene concentrations in serum of CAD women? Squalene concentrations in plasma lipoproteins are contributed by dietary intake, or by synthesis in intestinal or liver cells. Maximum dietary squalene could reach up to 200 mg for a 2000-calorie Mediterranean diet [10]. A considerable amount of squalene has been found in human bile and along with dietary olive oil, butter, cheese and egg yolk [10] this forms a small but significant amount of intestinal squalene. About 60% of intestinal squalene is absorbed and squalene consumption in high amounts increases plasma squalene concentrations [20–22]. Following a single dose of up to 500 mg of squalene in a fat meal, postprandial serum squalene concentrations increased within 6–9 h and returned to basal levels within 24 h [13], but the same amount of squalene consumption for 6 weeks increased only slightly its fasting plasma concentrations [22]. However, the dietary fat and monounsaturated fatty acid intake were similar in the two groups when virtually no olive oil was used. Rapeseed oil, which is very low in squalene [22], was the major source for monounsaturates. Dietary intake of squalene was calculated to be very low, B 20 mg/day, it appeared to be similar in both groups, and made hardly any difference to fasting squalene concentrations. The fasting squalene level in chylomicrons, slightly higher in the CAD than in the control women, suggests its intestinal origin, mostly synthesised by mucosal cells, while the level in VLDL and the VLDL lower half may point mainly to hepatic origin. Higher squalene and cholesterol concentrations in the

64

R.A. Rajaratnam et al. / Atherosclerosis 146 (1999) 61–64

VLDL lower half may be due to either enhanced conversion of VLDL to LDL or to decreased LDL clearance when the higher squalene to cholesterol ratio implies that hepatic sterol metabolism may have been altered in the coronary women. Compared to the afternoon levels, VLDL and LDL have several times higher squalene and methyl sterol values from midnight to the early morning hours in normal human life, indicating markedly enhanced de novo cholesterol synthesis up to about 04:00 during the daily diurnal rhythm [12]. There might be a delay in downregulation of this synthesis process in CAD women, resulting in the formation of squalene-rich VLDL particles in the low range of VLDL cholesterol and triglyceride concentrations (Fig. 1), and in higher fasting squalene in the circulating lipoproteins of the CAD women. Small estrogen doses appear to increase cholesterol synthesis in platlets [23], but the squalene values were not affected by the hormone replacement of the present controls. Elevated squalene concentrations in fasting sera do not always indicate high hepatic cholesterol synthesis [8,10,24], whereas fasting postsqualene precursors, which were not available for this study, are associated consistently with cholesterol synthesis [7,8,24]. Since squalene is at the top of the biosynthetic chain of cholesterol it could also accumulate in the circulation due to decreased turnover to lanosterol and gradually further to cholesterol. That postsqualene precursor sterols are not evenly increased is shown by findings that acute squalene feeding increases lanosterol and other methylated precursor sterols of cholesterol, while the values of the demethylated ones, occurring later in the synthesis chain, can remain unchanged [21]. In long term squalene treatment, only the demethylated sterols are increased [22]. In summary, raised squalene concentrations in the postmenopausal CAD women could partly be due to enhanced intestinal secretion or reduced turnover further to cholesterol or enhanced cholesterol synthesis. These possibilities should be studied further and related to the risk factors of atherosclerosis.

Acknowledgements This study was supported by grants from the Clinical Research Institute of the Helsinki University Central Hospital, the Research Foundation of Orion Corporation and the Ida Montin’s Foundation.

References [1] Langdon RG, Bloch K. The utilization of squalene in the

.

biosynthesis of cholesterol. J Biol Chem 1953;200:135 –43. [2] Tchen TT, Bloch K. On the conversion of squalene to lanosterol in vitro. J Biol Chem 1957;226:921 – 30. [3] Goodman DS, Popjak G. Studies on the biosynthesis of cholesterol: XII. Synthesis of allyl pyrophosphates from mevalonate and their conversion into squalene with liver enzymes. J Lipid Res 1960;1:286 – 300. [4] Goodman DS. Squalene in human and rat blood plasma. J Clin Invest 1964;43:1480 – 5. [5] Liu GCK, Ahrens EH Jr., Schreibman PH, Samuel P, McNamara DJ, Crouse JR. Measurement of cholesterol synthesis in men by isotope kinetics of squalene. Proc Nat Acad Sci USA 1975;72:4612 – 6. [6] Saudek CD, Frier BM, Liu GCK. Plasma squalene: lipoprotein distribution and kinetic analysis. J Lipid Res 1978;19:827–35. [7] Miettinen TA. Serum squalene and methyl sterols as indicators of cholesterol synthesis in vivo. Life Sci 1969;8:713 – 21. [8] Miettinen TA. Detection of changes in human cholesterol metabolism. Ann Clin Res 1970;2:300 – 20. [9] Nestel PJ, Kudchodkar B. Plasma squalene as an index of cholesterol synthesis. Clin Sci Mol Med 1975;49:621 – 4. [10] Liu GCK, Ahrens EH Jr., Schreibman PH, Crouse JR. Measurement of squalene in human tissues and plasma: validation and application. J Lipid Res 1976;17:38 – 45. [11] Scallen TJ, Schuster MW, Dhar AK. Evidence for a noncatalytic carrier protein in cholesterol biosynthesis. J Biol Chem 1971;246:224 – 30. [12] Miettinen TA. Diurnal variation of cholesterol precursors squalene and methyl sterols in human plasma lipoproteins. J Lipid Res 1982;23:466 – 73. [13] Gylling H, Miettinen TA. Postabsorptive metabolism of dietary squalene. Atherosclerosis 1994;106:169 – 78. [14] Gylling H, Relas H, Miettinen TA. Postprandial vitamin A and squalene clearances and cholesterol synthesis off and on lovastatin treatment in type III hyperlipoproteinemia. Atherosclerosis 1995;115:17 – 26. [15] Rajaratnam R, Gylling H, Miettinen TA. Postprandial lipoproteins, apolipoprotein B-48 and squalene metabolism in coronary postmenopausal women. Atherosclerosis 1997;134:340–1. [16] Brooks CJW, Harland WA, Steel G. Squalene, 26-hydroxycholesterol and 7-ketocholesterol in human atheromatous plaques. Biochim Biophys Acta 1966;125:620 – 2. [17] Clair RWS, Lofland HB Jr., Prichard RW, Clarkson TB. Synthesis of squalene and sterols by isolated segments of human and pigeon arteries. Exp Mol Pathol 1968;8:201 – 15. [18] Knuts LR, Rastas M, Haapala P. Micro-Nutrica, Version 1.0. Kansanela¨kelaitos (National Pensions Institute), Helsinki, 1997. [19] Miettinen TA. Cholesterol metabolism during ketoconazole treatment in man. J Lipid Res 1988;29:43 – 51. [20] Tilvis RS, Miettinen TA. Dietary squalene increases tissue sterols and fecal bile acids in the rat. Lipids 1983;18:32 – 6. [21] Strandberg TE, Tilvis RS, Miettinen TA. Metabolic variables of cholesterol during squalene feeding in humans: comparison with cholestyramine treatment. J Lipid Res 1990;31:1637 – 43. [22] Miettinen TA, Vanhanen H. Serum concentration and metabolism of cholesterol during rapeseed oil and squalene feeding. Am J Clin Nutr 1994;59:356 – 63. [23] Ciavatti M, Davenas E, Blache D, Monnier MA, Renaud S. Biosynthesis of platelet lipids in relation to aggregation in women using oral contraceptives. Contraception 1982;25:629– 38. [24] Miettinen TA. Cholesterol precursors and their diurnal rhythm in lipoproteins of patients with jejuno-ileal by-pass and ileal dysfunction. Metabolism 1985;34:425 – 30.