Biosynthesis of cholesterol and cholesterol precursors in platelets of female rats treated with oral contraceptives

297

Biochimica et Biophysics @ Elsevier/North-Holland

Acta, 620 (1980) 297-307 Biomedical Press

BBA 57680

BIOSYNTHESIS IN PLATELETS CEPTIVES

M. CIAVATTI

OF CHOLESTEROL OF FEMALE RATS

a, G. MICHEL

AND CHOLESTEROL PRECURSORS TREATED WITH ORAL CONTRA-

b and S. RENAUD

a

a I.N.S.E.R.M., Unite 63, 22 Avenue du Doyen Lepine, 69500 Bron-Lyon, and b Laboratoire de Biochimie Microbienne, Universite Claude Bernard, Lyon I, 43 Boulevard du I1 No&m bre I91 8, 69622 Villeurbanne Cedex (France) (Received

January

25th,

Key words: Cholesterol (Rat platelet)

1980)

biosynthesis;

Cholesterol

precursor;

Oral contraceptive;

Summary Rat platelets were incubated with sodium [U-‘4C]acetate and labeled lipids were analyzed. The major part of the radioactivity was found in phospholipids and in acylglycerols. When the incubation was performed with platelets of female rats treated with contraceptives, the total incorporation of labeled acetate in lipids was 2-fold higher and 33.6% of the radioactivity was found in lanosterol plus 24-dihydrolanosterol. Moreover, there is labeling of cholesterol. All these compounds were analyzed and identified by gas-liquid chromatography/mass spectrometry. In the incubation with sodium [2-14C]mevalonate the labeling of lanosterol plus 24-dihydrolanosterol and of cholesterol was increased by 30% in platelets of treated rats. The effect of contraceptives on one step of cholesterol biosynthesis is discussed and a possible explanation of the correlation between oral contraceptives and thromboembolic accidents is suggested.

Introduction Oral contraceptives have been reported for some time to be associated with thromboembolic accidents [l-3], for instance platelet functions are modified with an increase in clotting activity [4] and in their ability to aggregate [4,5]. Platelet lipids have been implicated in the coagulation process [6] and in the

Abrreviations: lynestrenol, (lo)-trien-20-yne-3,17-diol.

19-nor-17a-pregn_4~n-20-yn-17-01:

ethinyl estradiol, 19-nor-l’la-pregna-1,3,5-

298

aggregation of platelets [ 71. We have already reported an important increase in lipid synthesis in platelets of female rats treated with contraceptives [8]. This increase was observed with rats receiving ethinyl estradiol plus lynestrenol or ethinyl estradiol alone, while rats receiving lynestrenol alone gave results similar to those of the controls [8]. Then, ethinyl estradiol has been shown to be responsible for the increase of the incorporation of sodium li4C]acetate in platelet lipids. The purpose of the present work was to determine which lipid compounds were more actively synthesized in platelets when animals were treated with a contraceptive. Materials and Methods C~e~~cu~s. Cholesterol, lanosterol and 24-d~hydrolanosterol were from Sigma Co. (St. Louis, MO, U.S.A.); [U-‘4C]acetic acid, sodium salt (spec. act. 96.8 Ciimol) and [2-14C]mevalonic acid, sodium salt (spec. act. 36.5 Cilmol) were from the Commissariat a 1’Energie Atomique (Saclay, France); ethinyl estradiol (19-nor-l7~-pregna-1,3,5(10)-trien-20-yne-3,17-diol) was from Nutritional Biochemical Corporation (Cleveland, OH, U.S.A.); lynestrenol (19-nor17e-pregn-4en-20-yn-17-01) was a gift of Organon Co. (Oss, The Netherlands); ~,O-bis(trimethylsilyl)trifluoroacetamide was from Supelco (Bellefonte, PA, U.S.A.); trimethyl chlorosilane was from Hopkin and William Ltd. (Essex, U.K.). Preparation of platelets. This has been described previously [ES]. Female Sprague-Dawley rats (180-200 g) were used for the assays. Each experiment consisted of a control group (group I) receiving by intubation for 4 days 0.5 ml olive oil per 100 g body weight and another group receiving 0.5 ml olive oil with 10 I_tgethinyl estradiol plus 250 I.rglynestrenol (group II). The platelets were isolated from 4 ml of blood as previously described and were then resuspended in incomplete tyrode solution containing 0.2 g NaCl, 0.8 g NaHCO,, 1 g glucose, 0.07 g KH2P04, 1.2 g MgClz, 2.5 g gelatin, distilled water to 1000 ml, adjusted to pH 6.8. Incubations. 4 *10' platelets in 4 ml of incomplete tyrode solution were incubated at 37°C for 90 min (30, 60, 90, 120 min in the kinetic study) with 100 ~1 of an aqueous solution of 0.12 mM sodium [14C]acetate (20 ,uCi). Similarly, 4 * 10’ platelets from rats of groups I and II were incubated with 0.12 mM sodium [‘4C]mevalonate in 10 ~1 (10 #.Yi). Incubation of platelets from rats of group I with [ “C]acetate was performed in the presence of ethinyl estradiol, 1 pg/lO’ platelets in 5 ~1 ethanol, in comparison with a control containing only 5 ~1 ethanol. Lipid extraction and fractionation. After incubation, platelets were centrifuged and washed as previously described [ 81. Platelet lipids were extracted by the method of Folch et al. [9], the extract was evaporated under nitrogen and dissolved in 100 ~1 of CHCTJCH30H (2 : 1, v/v) for thin-layer chromatographic analysis. Glass plates (20 X 20 cm) coated with silica gel G (Merck) were used and thin-layer chromatography (TLC) was performed as described [ 81. The radioactive spots were localized by autoradiography (Kodirex film 2). The spots were identified by comparison with RF values of reference lipids, by specific coloration, and by scraping the gel from plates followed by elution,

299

300

hydrolysis and analysis of the products. In quantitative assays the spots were scraped into scintillation vials and the radioactivity was determined as described [ 81. Gas-liquid radiochromatography. Spots from thin-layer chromatography corresponding to phospholipids and neutral lipids were treated with BF3/ CH30H reagent (Merck) according to the technique of Morrison and Smith [lo]. Methyl fatty esters were purified on silica gel G with hexane/diethyl ether/acetic acid (80 : 20 : 1, v/v/v) and studied by gas-liquid radiochromatography on an Intersmat apparatus IGC 120 DFL combined with a Packard fraction collector No. 842 (column 2.5 m, 10% SP-2340 on 100-120 mesh Chromosorb W.A.W). Spots corresponding to cholesterol and other unsaponifiable compounds were silylated with a mixture of N,O-bis(trimethylsilyl)trifluoroacetamide/trimethylchlorosilane (80 : 20) and studied by gas-liquid radiochromatography (column 2.5 m, OV 1% on 100-120 mesh Chromosorb W.A.W.-D.M.C.S.), temperature program: 200-290” C, 4” K/min.

Mass spectrometry. Silylated samples were studied by gas-liquid chromatography-mass spectrometry in a VGMM 305 apparatus (ionisation potential 70 eV, temperature 250” C). Statistical analysis. Standard deviation was calculated by the usual method [ 111. Student’s paired t-test was used to determine the significance of data. Results

Incorporation

of radioactive

precursors

in platelet

lipids

The platelets of control and treated rats were incubated with sodium [ 14Clacetate for 90 min, then lipids were extracted from platelets and analyzed by TLC. The autoradiography of the plates showed seven radioactive spots (Fig. 1). Each spot was scraped and the radioactivity determined. The results are summarized in Table I. Compounds 1, 2,4,6, 7 were characterized respectively as phospholipids (no migration in thin-layer chromatography), monoacylglycerols (RF = 0.17), diaand chocylglycerols lesterol esters (R(; = 0.5), free fatty acids (R F = 0.65), triacylglycerols r = 0.9). Spot 3 was not found in control animals; it was observed only in rats treated with a contraceptive. Further studies permitted identification of the compound as cholesterol (see below). Spot 5 was very weak in control rats but strongly enhanced in treated rats. Two compounds were further characterized in spot 5, lanosterol and dihydrolanosterol (see below). The characterization of cholesterol and metabolic precursors of cholesterol in rats treated with contraceptive leads us to suggest that contraceptives induce a modification in cholesterol biosynthesis in platelets. Therefore, incorporation of sodium [ “C]mevalonate into platelet lipids was tested. The composition of platelet lipids is shown in Fig. 1. The total incorporation was 30% higher in lipids of treated rats than in those of controls (Table I). In control and treated animals the major part of mevalonate was incorporated in cholesterol precursor 5. The other radioactive spots have not been identified, but they probably consist of other precursors of cholesterol.

301

6 5 43

i

Fig. 1. Autoradiogram of the labeled lipids from platelets after incubation with radioactive precursors. (A) platelets of control rats incubated with sodium [ 14CIacetate; (B) platelets of control rats incubated with sodium [14Clmevalonate: (C) platelets of rats treated with contraceptives. incubated with sodium [14C]acetate: (D) platelets of rata treated with contraceptives, incubated with [14Clmevalonate. 1. phospholipids; 2, monoacylglycerols: 3, cholesterol; 4, diacylglycerols: 5, lanosterol + dihydrolanosterol; 6. fatty acids; 7, triacylglycerols and cholesterol esters.

TABLE II DISTRIBUTION OF RADIOACTIVITY IN TOTAL WITH SODIUM [ 14C]ACETATE FOR 90 MIN

PLATELET

FATTY

ACIDS AFTER

INCUBATION

Results are given as percentage of total fatty acids. Fatty acid carbon number c12s16

Cl8

czo*22

c24

and

>c24

Platelet phospholipids

Control Treated

35.9 * 2.2 31.6 + 2.9

18.8 f 0.8 19.3 + 2.7

16.1 f 0.8 14.0 f 0.6

28.9 + 2.7 34.8 * 4.8

Platelet neutral lipids

Control Treated

34.9 * 2.9 29.2 -t 2.3

16.9 * 1.3 19.7 -f. 1.8

16.0 f 2.2 18.8 * 1.4

32.4 * 3.2 32.3 r 2.3

302

rncorpo~ation of sodium ~14~]a~etate in platelet fatty acids Platelet lipids were separated by TLC into two groups, phospholipids which have no mi~ation and mono-, di- and triacylglycerols which were pooled in one common group. From each group, fatty acids were obtained by alkaline hydrolysis, methylated and methyl esters analyzed by gas-liquid chromatography and determination of radioactivity. The results are summarized in Table II. No significant difference was observed in the distribution of the radioactivity in the fatty acids of control and treated animals. Kinetic study of i~~orpo~atio~ of s~di~rn [14C~ace~a~e The incorporation of sodium [ 14C]aceLate into total platelet lipids was measured after 30 and 120-min incubations. Results are given in Fig. 2. A biphasic curve was found in control animals in which incorporation is more rapid during the first hour of incubation. In contraceptive-treated rats incorporation of radioactive sodium acetate was linear during 2 h and total incorporation was 80% higher than in control. The incorporation was studied more precisely in each group of platelet lipids. Results are given in Figs. 3 and 4. In phospholipids the rates of incorporation do not show any significant difference during the first hour of incuba-

I

rats

0

control

with

contraceptives

rats

60 time Fig. 2. Kinetics of sodium tl%lacetate

90 120 of ~ncubetio~ (rn~n~ incorporation

into total lipids of rat platelets,

303

* rats

wtth contraceptives

PC

t

t me

of

incubation

@in)

Fig. 3. Kinetics of sodium [I4 CJacetate incorporation into complex lipids of rat platelets, PC, phosphatidylcholine; PS, phosphatidylserine; PI, phosphatidylinositol; PE, phosphatidylethanolamine; CER, ceramide.

tion, but during the second hour the incorporation is more significant in the treated rats, particularly in phosphatidylcholine. In neutral lipids the acylglycerol group shows an increase in incorporation during the last 30 min in the treated animals. Cholesterol and cholesterol esters, which are minor compounds in control animals, show significant increase in treated animals (3.5-fold). The greatest difference was observed with lanosterol and Z~-dihydrolanosterol (spot 5 in TLC) which show a linear incorporation during the entire experiment. After 2 h incubation, the level of the radioactive compound from spot 5 was 15-fold higher in contraceptive-treated rats than in controls. Iden t~f~~~~~o~ of compounds frum spok 3 artd 5 Both radioactive spots 3 and 5 were scraped from thin-layer plates. The com-

. rats

wi:h conlraceptives

Fig, 4. Kinetiesofsodium [14Cjacetate incorporation into neutral lipids of rat platelets. L, fanosterol; DL, dihydrolanosterol; MG, monoacylglycerols; DG, diacylglycerols; TG, triacylglycerols: CH, cholesterol: GE. cholesteryl esters.

pounds could not be methylated with diazomethane. Trimethylsilylation was performed under experiments conditions as described above and trimethylsibyl derivatives were analyzed by GC-MS in comparison with trimethylsilyl de~vatives of authentic cholesterol, ~anosterol and 2~-dihydrolanosterol. GLC of trimethylsilyf derivatives from spot 3 gave a major compound having the same retention time as trimethylsilyl cholesterol, The mass spectra of

305

trimethylsilyl derivatives of standard cholesterol and of the unknown compound are identical with a molecular ion at m/z = 458 and fr~entation peaks at m/z = 443 (M-15), 368 (M-90), 353 (M-90-15}, 329 (M-129), m/z = 255 elimination of the side chain and the trimethylsilyloxyl group, m/z = 247 rupture of the ring, B, m/z = 129 (TMS-0’ = CH-CH= CH2). Trimethylsilyl derivatives of the compounds from spot 5 were analyzed by GLC. Two compounds were found, and their retention times were respectively those of standard trimethylsilyl 24-dihydrolanosterol and trimethylsilyl lanosterol. The ratio of trimethylsilyl lanosterol to t~methylsilyl 24-dihydrolanosterol was about 5 : 1. The mass spectrum of the major compound was identical to the mass spectrum of standard trimethylsilyl lanosterol, molecular ion at m/z = 498 and fragments at m/z = 483 (M-15), 393 (M-90-15), 281,207. The mass spectrum of the minor compound was identical to the mass spectrum of standard trimethylsilyl 24-dihydrolanosterol: molecular ion at m/z = 500 and fragments at m/z = 485 (M-15), 395 (M-90-15), 281,227,207. Discussion A comparative study of sodium [ 14C]acetate incorporation into platelet lipids of control and contraceptive-treated rats had already shown a significant increase of lipid synthesis in the treated animals where estrogen was found to be responsible for this increase [ 81. The present work studies more precisely the lipid classes which were modified in platelets by contraceptives. Analysis of fatty acids shows no si~ificant difference in the incorporation of sodium [ 14C]acetate for 90 min in treated and untreated rats. When the incubation was extended by 30 min an increase of the incorporation of labeled sodium acetate was observed in acylglycerols and in some phospholipids, particularly in phosphatidylcholine. However, more drastic differences are found in the levels of cholesterol and of cholesterol precursors. In human platelets, Derksen and Cohen [12] have shown an accumulation of labeled cholesterol precursors by incubation with [ “C]mevalonic acid. When platelets were incubated with sodium [ “C]acetate, only a small part of labeled acetate could reach mevalonic acid and give rise to labeled lanosterol. Two steps of cholesterol biosynthesis seem to be inoperative in platelets: (i) the first sequence from acetate to mevalonate, and (ii) the final sequence from lanosterol to cholesterol. Our work confirms previous reports, since the platelets of control rats incubated with [14C]acetate give essentially labeled phospholipids and acylglycerols. The level of lanosterol plus dihydrol~osterol accounted for 7% of total labeled lipids and cholesterol for only 0.7%. When platelets of control rats were incubated with labeled mevalonate the level of lanosterol plus dihydrolanosterol reached 71.5% of total labeled lipids, whereas the level of cholesterol was 0.8%. We have previously found that treatment of rats with contraceptives enhances biosynthesis of platelet lipids [8]. in fact, incorporation of [ 14C]acetate into platelets from treated rats is more significant (Z-fold) in comparison with inco~oration into platelets of control rats. The distribution of radio-

306

activity in the lipid classes is very different. The amount of lanosterol-dihydrolanosterol in treated rats accounts for 33.6% of total labeled lipids and cholesterol for 3.4%. In the case of mevalonate, the yield of the incorporation of mevalonate was 30% higher in platelets of treated rats than in control rats and the major part of the radioactivity was found in lanosterol plus dihydrolanosterol. No si~ific~t difference of the distribution of labeled lipids appears in treated and control rats; in both groups 1.8% of the radioactivity is found in cholesterol and 71-74% in lanosterol plus dihydrolanosterol. The lack of the in vitro biosynthesis of cholesterol precursors by incubation of platelets with sodium [ 14C]acetate could be due to an apparent vulnerability of the 3-hydroxy-3-methylglutaryl-CoA reductase which could be altered by in vitro manipulations [ 123. In fact, this hypothesis must be eliminated as lanosterol biosynthesis is observed in vitro with platelets of treated rats. It would thus appear that platelets have the capacity to convert acetate to lanosterol but the first steps from acetate to mevalonate may be highly depressed in normal platelets and only a slight part of the acetate could reach lanosterol. The platelets of contraceptive-treated rats seem to recover their capacity to synthesize lanosterol from acetate. We could suggest that contraceptives act on the repressor and subsequently permit transformation of acetate to mevalonate. The other sequence which seems to be missing in platelets is the conversion of lanosterol to cholesterol. Our results are less obvious on this point. A relatively high percentage of radioactivity was found in cholesterol when platelets of treated rats were incubated with labeled acetate. This result could lead us to suppose an enhancement by contraceptives of the final sequence of the biosynthesis. However, when platelets were incubated with [i4C]mevalonate the relative incorporations of labeled precursor in cholesterol were not different in platelets of control and treated rats. For this, no explanation can be given at the present time. The correlation between contraceptives and sterol biosynthesis has important consequences with regard to platelet function. Some preliminary assays have shown a hyperaggregability of platelets in. the presence of lanosterol and dihydrolanosterol. This observation could be related to the predisposition to thromboembolic accidents associated with treatment with oral contraceptives [l--3]. Further work is needed to throw more light on the role of contraceptives in the in vivo biosynthesis of cholesterol, and also on the mechanism of action of contraceptives on this biosynthesis. Acknowledgement This study was supported by a grant (Contrat libre) from the ‘Institute Nationale de la Sante et de la Recherche Medicale’, France. References 1 2 3 4 5

Markusb, R.E. and Seigel, D.G. (1969) Am. J. Publ. Health 59, 418434 Inman, W.H., Vessey, M.P., Westerholm, B. and Engelund. A. (1970) Brit. Med. J. 2, 203-209 Keown. D. (1969) &it. J. Surg. 56.486488 Lecompte, F. and Renaud, S. (1973) Thromb. Diath. Haemorrh. 30, 510-517 Poller. L.. Priest, C.M. and Thomson, J.M. (1969) Brit. Med. J. 4. 273-274

307 6 Marcus, A.J.. Ullman, H.L.. Safier, L.B. and Ballard, H.S. (1962) 7 Marcus, A.J. (1978) J. Lipid Res. 19,793-826

J. Clin. Invest. 41. 2199-2212

8 Ciavatti, M. and Renaud, S. (1979) Harmon. Metabol. Res. 7,441444 9 Folch, J.M., Lees, M., SloaneStanley, G.H., (1957) J. Biol. Chem. 226, 497-509 10 Morrison. W.R. and Smith, L.M. (1964) J. Lipid Res. 5. 600-608 11 Bum, J.H.. Finney. D.J. and Godwin, L.L. (1950) Biological Standardization, Oxford University. Oxford 12 Derksen, A. and Cohen, P. (1973) J. Biol. Chem. 248.7396-7403

2nd edn.. pp. 4243,