- Email: [email protected]
Hypotriglyceridemic effects of levonorgestrel in rats
Atherosclerosis, 52 (1984) 329-338 EIsevier Scientific Publishers Ireland,
329 Ltd.
ATH 03518
Hypotriglyceridemic
Effects of Levonorgestrel Rats
Rama Khokha Departments
and Bernard
in
M. Wolfe
of Biochemistry and Medicine, The Unioersity of Western Ontario, London, Ont. (Canada) (Received 12 July, 1983) (Revised, received 17 February, 1984) (Accepted 23 March, 1984)
Summary The progestin, levonorgestrel administered orally to fed female rats significantly lowers both plasma total and very low density lipoprotein triglycerides. In contrast, plasma total cholesterol and low density lipoprotein cholesterol rose significantly. Suspensions of isolated hepatocytes were used to study the effects of levonorgestrel on triglyceride synthesis by examining the incorporation of labelled precursors ([9,10-3H]palmitate and [U-‘4C]glycerol) into triglycerides. Levonorgestrel (1O-4 M) significantly inhibited the incorporation of both precursors into hepatocyte triglycerides and also reduced their incorporation into the triglycerides (nearly all in d < 1.006) released into the medium. These results suggest that inhibition of hepatic triglyceride synthesis and release can account at least for part of the lowering of plasma VLDL which occurs during administration of levonorgestrel. Key words:
Hepatocytes - Lipoproteins - Levonorgestrel - Rats - Triglycerides
Introduction The progestin, dl-norgestrel [active isomer levonorgestrel] has been widely administered to women in the form of oral contraceptives [l]. Reports about the influence of dl-norgestrel-containing medications on serum lipids are conflicting. Although This research was supported by a grant from the Medical Research Research Studentship awarded to R.Khokha (1980-84) from the Medical appreciated.
0021-9150/84/$03.00
0 1984 Elsevier Scientific
Publishers
Ireland,
Ltd.
Council Research
of Canada Grant. A Council is also much
330
most investigators find that oral contraceptives containing dl-norgestrel lower serum triglycerides [2,3], this has not been observed by others [4- 71. This discrepancy likely relates to the type and dosage of estrogens used in the formulation, as estrogens tend to oppose the effect of progestins on plasma lipid and lipoprotein levels [8]. dl-Norgestrel alone, has been reported to lower triglycerides in premenopausal women [9]. The present study was undertaken to examine the effects of the active isomer levonorgestrel [lo] on lipid and lipoprotein levels in the rat and to elucidate its mechanism of action using isolated rat hepatocytes. Methods
and Materials
Materials [9,10-‘H]Palmitic acid (12 Ci/mmol) purchased from New England Nuclear obtained from Wyeth Ltd., Toronto, Ont. Millipore Corp., Mississauga, Ont. and supplied by Gibco Canada, Burlington, purchased from Sigma Chemical Co., St. Ralston-Purina Co., Minneapolis, MN.
and [U-14C]glycerol (12 pCi/pmol) were Corp., Boston, MA. Levonorgestrel was Collagenase (Type I) was obtained from Waymouth’s medium (MB 752/l) was Ont. Porcine albumin (fraction V) was Louis, MO. Rat chow was obtained from
Studies in vivo Female Sprague-Dawley rats (200-250 g), 7 in each group (control and experimental), were individually caged in an animal room illuminated from 8.00 to 20.00 h. Each rat in the experimental group received 4 pg levonorgestrel/day . kg body wtO.” (average dose 1.4 pg/day) for 15 days. The experimental diet was prepared by adding the levonorgestrel dissolved in absolute ethanol to one piece of the rat chow; ethanol was then evaporated. The levonorgestrel-containing chow was fed first (at 18.00 h) and the rat observed until it was completely consumed. The control diet was treated with an equal amount of ethanol which was likewise removed by evaporation. Thereafter, water and rat chow were available ad libitum to both levonorgestrel-treated and control rats. Rats received a fat-free diet for 12 h prior to killing to eliminate chylomicrons from plasma. They were venesected after receiving a sodium pentobarbital injection (13 mg in 0.2 ml) intraperitoneally and plasma samples were taken for determination of lipid content and for the separation of lipoproteins by the method of Have1 et al. [ll]. Plasma lipids were extracted in chloroform/methanol (2 : 1, v/v, [12]). Triglycerides, cholesterol and phospholipids were assayed by the methods previously described [13-161. Preparation of fat-free diet The fat-free diet prepared in the laboratory was a synthetic mixture of casein (18%), dextrose (73%), celluflour (5%), and salts (4%) as formulated by Bernhardt and Tomarelli [17]. The vitamin mixture was described by Greenfield et al. [18]. The composition of the salt mixture, water and fat-soluble vitamins was essentially the same as described earlier [19].
331
Studies in vitro using rat hepatocytes Female Sprague-Dawley rats (200-300 g) maintained on rat chow were used for preparation of hepatocytes by the method of Seglen [16,20]. Glucose (108, w/v) was added to the drinking water (to maintain a fed state) for 48 h prior to killing of the rats. Viability, determined by the trypan blue dye exclusion test, ranged from 60 to 90% (mean 71 + 3%) and the final intact hepatocyte yield ranged from 0.1 to 1.0 X 10’ cells per liver. Incubation of hepatocytes Each set of control and experimental preparation was carried out with cells freshly isolated from the same rat liver. In a typical experiment, incubations were carried out in a shaking water bath at 37 ‘C with hepatocytes (0.4-0.6 x 10” viable cells/ml) suspended in Waymouth’s medium in the presence or absence (control) of levonorgestrel (lop4 M). For dispersal in aqueous medium, levonorgestrel was dissolved in 0.1% dimethyl sulfoxide (DMSO); a similar amount of DMSO was present in the medium of the controls. 25 PCi of [9,10-3H]palmitate (complexed to porcine albumin, molar ratio approximately 1 : 1) and 25 PCi of [U-‘4C]glycerol were added to the medium immediately before the 0 sample. 50 PM palmitic acid complexed to albumin was also added to the medium (both experimental and control) in some experiments (see Fig. 4). Incubations were continued for 60 min. Samples of the suspended cells were removed at 0, 30, and 60 min and centrifuged at 3 500 rpm for 15 min at 5 ‘C to sediment the cells. The pellet (suspended in 0.5 ml ice-cold saline) and the supernatant were each extracted with chloroform/methanol (2 : 1) for determination of lipids. The chloroform layer of the chloroform/methanol extract of the lipids obtained from both the hepatocytes and the medium in which they were incubated was treated with silicic acid to remove phospholipids [13,20]. Labelled free fatty acids were removed by overalkalinization [16,21] from neutral lipids prior to determination of the 3H and 14C content of the neutral lipids (essentially triglycerides) of the hepatocytes and their supernatant. In several experiments, cells were sedimented from incubation media as described above and the supernatant was mixed with human serum (2: 1, v/v) prior to ultracentrifugation [ll] to determine the density fraction to which the secreted labelled lipoproteins corresponded. Statistics Differences between groups were evaluated according to Snedecor and Cochran [22] for both paired and unpaired samples. Variance is expressed as standard error of the mean. Results
Effects of levonorgestrel on plasma lipid and lipoprotein concentrations in vivo The lower mean plasma concentration of triglycerides (Table 1) was attributable to the significantly lower mean concentration of VLDL triglycerides (Table 2). The concentration of VLDL phospholipids fell in proportion to the decrease in VLDL
332
TABLE
1
EFFECTS
OF LEVONORGESTREL
ON THE CONCENTRATIONS
OF PLASMA
LIPIDS
IN VIVO
Plasma obtained from rats fed rat chow (control) or rat chow containing 4 pg levonorgestrel/day’kg body wt “j for 2 weeks was extracted and lipids were determined as described in Materials and Methods. Each value represents the mean + SE for 7 rats. Group
Control Levonorgestrel * Designates
Triglycerides
Cholesterol
Phospholipids
(mg/dl)
(mg/dl)
(mg/dl)
(Total)
(Total)
(Total)
34k2 25+3 a statistically
4652 54+4
*
significant
difference
62+4 67k6
*
from control
value at P < 0.025.
triglycerides (Table 2). In contrast, the concentrations of plasma total and LDL cholesterol increased significantly (Tables 1 and 2). LDL phospholipids were also significantly elevated by levonorgestrel treatment (Table 2). However, no significant differences were found between control and treated rats in the concentrations of cholesterol and phospholipids in HDL or its subfractions HDL, and HDL, (data not shown). Time course of the incorporation of labelled precursor into hepatocyte triglycerides and triglycerides released into the medium The time course of the incorporation of both labelled palmitate and glycerol into hepatocyte triglycerides was curvilinear, up to 60 min, in the absence of added palmitate (Fig. l), whereas with the addition of 50 PM unlabelled palmitate, it was linear (Fig. 1). The release of labelled triglycerides into the medium was slow for the first 30 min, but increased 3-fold in the next 30 min. As expected, the observed pattern is compatible with a precursor-product relationship between hepatocyte
TABLE
2
EFFECT OF LEVONORGESTREL PROTEIN FRACTIONS IN VIVO
ON THE CONCENTRATIONS
OF LIPIDS
IN PLASMA
LIPO-
Plasma obtained from rats fed rat chow (control) or rat chow containing 4 pg levonorgestrel/day. kg body wt 075for 2 weeks was separated into standard lipoprotein density fractions by ultracentrifugation. The individual lipoprotein fractions were extracted and the lipids determined as described in Materials and Methods. Each value represents the mean + SE for 7 rats.
Control Levonorgestrel * Designates
Triglycerides
Cholesterol
Phospholipids
(mg/dl)
(mg/dl)
(mg/dl)
VLDL 25+2 21*1*
VLDL 4+1 4+1
a statistically
significant
difference
LDL 11+1 15+2*
HDL 31+2 35+3
from control
VLDL 7+1 5+2
value at P -C 0.05.
LDL 7+1 10+1*
HDL 48+3 52*4
333
0
15
30
60
0
15
30
60
MINUTES
Fig. 1. Effect of added palmitate 50 pM on the incorporation of [9,10-3H]palmitate into hepatocyte triglycerides. Isolated hepatocyte suspensions were incubated with labelled palmitate (A) and glycerol (B) with added palmitate (A) or without added aliquots of the hepatocytes, removed at 0, 15, 30 and 60 min were extracted and determined as described under Materials and Methods. Each value represents the experiments.
and [U-‘4C]glycerol 25 pCi of each of palmitate (0). The triglyceride counts mean for at least 2
triglycerides and the triglycerides released into the medium. Over the 60-min period of incubation with labelled precursors the release of labelled triglycerides into the medium accounted for only about 1% of labelled triglycerides synthesized in hepatocytes. Most of the labelled triglycerides released into the medium (89 + 5% of [ 3H] and 95 f 2% of [14C]), were in the density fraction d < 1.006 corresponding to VLDL, with only minor amounts in d > 1.006 (Table 3). Effect of levonorgestrel on triglyceride synthesis and release by isolated hepatocytes Levonorgestrel (10e4 M) significantly reduced the incorporation of both [9,103H]palmitate and [U-t4C]glycerol into hepatocyte triglycerides (P < 0.05) (Fig. 2). The inhibitory effect of levonorgestrel reached a plateau within 30 min of incubation. At that point in time, the per cent inhibition of the incorporation of labelled palmitate averaged 20 f 3% and that of glycerol 19 + 2%. Both values are significantly lower than control (P < 0.05) though similar to each other. Release of [ 3H]TABLE
3
PERCENT DISTRIBUTION OF LABELLED TIONS OF THE INCUBATION MEDIUM
TRIGLYCERIDES
AMONG
LIPOPROTEIN
FRAC-
Medium obtained from hepatocyte cell suspensions (at 0’ and 60’) incubated with 25 PCi of [9,103H]palmitate+ unlabelled palmitate (50 PM) and 25 gCi of [U-‘4C]glycerol was separated into lipoprotein fractions by ultracentrifugation. The individual lipoprotein fractions were extracted and triglycerides separated as described under Materials and Methods. The values represent the % increase of medium triglyceride counts in different density fractions over 60 min for 4 individual experiments. Density
13W f14Cl ND = not detected.
fractions
(n/ml)
1.006
1.006-1.019
1.019-1.063
1.063
g9+5 95+2
6+3 3k2
5+3
ND ND
2*1
E
a* ,’ ,
-P
-9’’
/’ I
/
, I
,
0
30
60
0
30
60
MINUTES
Fig. 2. Time course of incorporation of labelled precursors into hepatocyte triglycerides. The isolated rat hepatocyte suspensions were incubated with 25 PCi of [9,10-‘Hlpalmitate (A) and 25 PCi of [U“C]glycerol (B) in the presence (0) or absence (0) of 10e4 M levonorgestrel. The medium concentration of DMSO was 0.1%. The aliquots of the hepatocytes removed at 0, 30 and 60 min were extracted and triglyceride counts determined as described under Materials and Methods. Each value represents the mean for 5 experiments. Experimental values for both palmitate and glycerol were significantly lower than control at both 30 and 60 min (P < 0.05).
and [‘4C]triglycerides into the medium (Fig. 3) was also significantly reduced by 10e4 M levonorgestrel (P < 0.05). Maximum inhibition of the incorporations of both labelled palmitate and glycerol into triglycerides released into the medium was observed at 60 min (51 + 11% and 54 f 13%, respectively, P < 0.05). Effect of higher levels of DMSO on the hepatocyte response to levonorgestrel The incorporation of labelled precursors into hepatocyte triglycerides and triglycerides released into the medium was unaffected by 0.1% DMSO. Levonorgestrel (10m4 M), initially dissolved in 0.1% DMSO, partially precipitated upon addition to the medium. Progressively higher amounts of DMSO (O.lW, 0.25’%, 0.5% and 1.0%) were used to suspend the levonorgestrel (10e4 M) in the medium and the results
0
30
60
0
30
60
MINUTES
Fig. 3. Time course of release of [ 3H]- and [‘4C]triglycerides in the medium. The isolated hepatocyte suspensions were incubated with 25 PCi of [9,10-3H]palmitate (A) and 25 pCi of [U-‘4C]glycerol (B) in the presence (0) or absence (0) of 10e4 M levonorgestrel. The aliquots of the medium removed at 0, 30 and 60 mm were extracted and triglyceride counts determined as described under Materials and Methods. Each point represents the mean value for 5 different experiments. Experimental values for both palmitate and glycerol were significantly lower than control at both 30 and 60 min (P c 0.05).
335
0
0.1
0.25
0.5 %
DMSO
Fig. 4. The percent inhibition by levonorgestrel of incorporation of labelled precursors into hepatocyte triglycerides with increasing medium concentrations of dimethylsulfoxide (DSMO). The isolated hepatocyte suspensions were incubated with levonorgestrel (10s4 M) suspended with varying medium concentrations of DMSO (O.lW-1.0%) in the presence of 25 BCi of each [U-‘4C]glycerol (m) and [9,10-3H]palmitate (0) supplemented with 50 pM unlabelled palmitate. Control incubations were carried out with the same amounts of DMSO as in the experimental incubations, but in the absence of levonorgestrel. Hepatocyte triglycerides were extracted and their counts determined as described in Materials and Methods. Each value represents the mean f SE of 3 individual experiments.
were compared to controls (i.e. hepatocytes incubated without levonorgestrel, but the same amount of DMSO, Fig. 4). Mean values for the incorporation of palmitate into hepatocyte triglycerides in the presence of DMSO were not significantly different (P > 0.2) from their respective controls at any of the levels of DMSO that were tested as follows: 0.1% DMSO (16.8 + 1.9 vs. 15.4 + 2.4 x lo5 cpm/106 cells); 0.25% DMSO (19.8 f 1.6 vs. 17.9 + 0.8 X lo5 cpm/106 cells); 0.5% DMSO (20.2 rt 1.5 vs. 17.9 + 0.8 X lo5 cpm/106 cells); or 1.0% DMSO (18.2 f 1.8 vs. 16.2 + 0.7 x lo5 cpm/106 cells). Likewise, mean values for the incorporation of glycerol into hepatocyte triglycerides were not significantly (P > 0.2) different from their respective controls at any of the levels of DMSO tested as follows: 0.1% DMSO (14.8 k 1.5 vs. 14.3 + 1.8 X lo4 cpm/106 cells); 0.25% DMSO (16.7 & 0.3 vs. 16.1 k 0.6 x lo4 cpm/106 cells); 0.5% DMSO (15.7 _t 0.8 vs. 16.1 f 0.6 x lo4 cpm/106 cells); or 1.0% DMSO (16.7 k 10 vs. 14.7 + 0.5 X lo4 cpm/106 cells). However, in the presence of higher concentrations of DMSO the effect of levonorgestrel was enhanced. The incorporation of labelled palmitate into hepatocyte triglycerides in the presence of 10m4 M levonorgestrel was inhibited by only 20 + 3% when the concentration of DMSO was 0.1% compared to 26 &-0.6% with 0.25% DMSO, 31 & 1% with 0.5% DMSO and 39 k 0.8% with 1% DMSO. Likewise, the inhibition of the incorporation of labelled glycerol into hepatocyte triglycerides increased from 19 f 2% with 0.1% DMSO to 42 + 2% with 1.0% DMSO. Discussion The present studies demonstrate that a low dose of levonorgestrel, comparable to that used for human contraceptive purposes, significantly lowers serum total and VLDL triglycerides in fed female rats, while elevating total and LDL cholesterol levels. The parallel reductions of VLDL phospholipids and VLDL triglycerides
336
suggest that the lowering of VLDL is attributable to reduction of the number of VLDL particles rather than to the formation of abnormal VLDL particles. The measurements of serum lipids were made under fed-state conditions using a fat-free diet which excluded the effect of chylomicrons from the circulation which would have confounded the interpretation of the results. Although lipogenesis from carbohydrate is likely to be enhanced by a fat-free diet, the diet was the same for both norgestrel-treated and control rats. This allowed one to ascertain the effect of the levonorgestrel alone. Lower VLDL levels could result from either reduced hepatic triglyceride secretion or enhanced clearance of VLDL from the blood plasma. Our studies in vitro lend support to the former mechanism, but have not examined clearance mechanisms. Levonorgestrel rapidly inhibited the synthesis of triglycerides from palmitate and glycerol in isolated rat hepatocytes (Figs. 1 and 2). However, the extent of the inhibition was influenced by the amount of DMSO used to suspend the levonorgestrel (Fig. 3). Because incorporation of both substrates into triglycerides was inhibited to a similar extent by levonorgestrel, it is attractive to suggest that the inhibition may be at the level of either glycerol-3-phosphate acyltransferase, phosphatidic acid phosphatase and/or diacylglycerol acyltransferase, the three potentially rate-limiting enzymes shared by these substrates on their pathway to triglycerides. The inhibition by levonorgestrel of triglyceride release from hepatocytes (Figs. 3 and 4) could be accounted for solely by the reduction of hepatocyte triglyceride synthesis. However, because triglyceride release was inhibited by levonorgestrel to a greater extent (51-548) than hepatocyte triglyceride synthesis (19-20%) it is conceivable that levonorgestrel might have additional effects on the release of VLDL from hepatocytes. The present results suggest (i) that liver may be an important site of the hypolipidemic action of levonorgestrel and (ii) that inhibition of hepatic triglyceride synthesis and release can accaunt at least for part of the hypolipemic effect of levonorgestrel observed in vivo in fed rats. In contrast to the in vivo studies which were done over an extended time with a dose of levonorgestrel which is comparable to the conventional therapeutic doses given to women, the in vitro studies with isolated hepatocytes employed much higher pharmacologic doses which permitted much shorter studies. Although solubilization of the levonorgestrel posed a problem in the in vitro studies, we found that DMSO was useful in suspending the levonorgestrel. In the concentrations employed, DMSO alone inhibited neither hepatocyte triglyceride synthesis nor release. This contrasts with a previous report by Bell et al. [23] that a higher concentration of DMSO (5%) produces a 7-fold increase in the incorporation of [14C]acetate into triglycerides of surviving rat liver slices. The effects of DMSO levels lower than 5% have not previously been reported. Because the inhibitory effect of levonorgestrel occurred so rapidly, the inhibitory action in vitro of levonorgestrel on hepatocyte triglyceride synthesis and release may have occurred by a mechanism which is different from that generally known to account for steroid action. Steroids are generally thought to act by binding to receptors in the cytosol which translocate to the nucleus, bind to DNA and then mediate their action by inducing transcription of specific mRNA [24]. However,
337
progesterone is also known to alter CAMP levels by binding to membrane receptors [25,26], in addition to its known action via cytosolic receptors [27,28]. Levonorgestrel could act by altering CAMP levels by a mechanism similar to progesterone. Levonorgestrel, a 19-nortestosterone derivative, also has some androgenic activity [29]; however, cytosolic receptors for levonorgestrel have been shown to be competitively displaced by progesterone, but only weakly by testosterone [29]. It is not yet known whether there are membrane receptors for levonorgestrel. Additional explanations for the rapid effect of levonorgestrel could include (1) the phosphorylation-dephosphorylation mechanism regulation diacylglycerol acyltransferase proposed by Haagsman et al. [30] for the inhibitory action of glucagon on hepatic triglyceride synthesis, and/or (2) selective translation of mRNA independent of nucleus events [31]. Acknowledgements The authors are grateful for typing the manuscript.
to E. Wolfe for helping
with the artwork
and S. Brock
References 1 Tyler, E.T., International 2
3
4
5
6
7
8 9 10 11 12 13
aspects of the use of norgestrel. In: The Second International Norgestrel Symposium, Excerpta Medica, Amsterdam, 1974, p. 1. Nielsen, F.H., Honore, E., Kristoffersen, K., Secher, N.J. and Pedersen, T., Changes in serum lipids during treatment with norgestrel oestradiol-valerate and cycloprogynon, Acta Obstet. Gynaecol. Stand., Suppl. 56 (1977) 367. Robertson, D.N., Alvarez, F., Sivin, I., Brache, V., Stern, J., Leon, P. and Faundes, A., Lipoprotein patterns in women in Santo Domingo using a levonorgestrel/estradiol contraceptive ring, Contraception, 24 (1981) 469. Foegh, M., Pederson, F.D., Gormsen, J., Knudsen, J.B. and Schou, G., Oral levonorgestrel-testosterone effects on spermatogenesis, hormone levels, coagulation factors and lipoproteins in normal men, Contraception, 24 (1980) 381. Riissner, S., Frankman, 0. and Marsk, L., Effects of various low dose contraceptive pills on serum lipoproteins. In: L.W. Hessel and H.H.J. Krans (Eds.), Lipoprotein Metabolism and Endocrine Regulation, Elsevier, Amsterdam 1979, pp. 91-98. Silfverstolpe, G., Gustafson, A., Samsioe, G. and Swanborg, A., Lipid metabolic studies in oophorectomized women - Effects of three different progestogens, Acta Obstet. Gynaecol. Stand.. Suppl. 88 (1979) 89. Pribicevic, V., Skrabalo, Z., Lipovac, V., Trenc, S., Zadjelovic, J. and Rajkovic, Z., Oral contraception, mechanical contraception and carbohydrate and lipid metabolism - A two year study. Int. J. Fertil., 24 (1979) 114. Beck, P., Contraceptive steroids - Modifications of carbohydrate and lipid metabolism, Metabolism, 22 (1973) 841. Spellacy, W.N., Buhi, N.C. and Birk, S.A., Norgestrel and carbohydrate-lipid metabolism - Glucose, insulin and triglyceride changes during six months time of use, Contraception, 9 (1974) 615. Robert, C.J., Singer, A.C. and Edgren, A.R., The biological activities of norgestrel and its enantiomers, Int. J. Fertil., 24 (1979) 39. Havel, R.J., Eder, H.A. and Bragdon, J.H., The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum, J. Clin. Invest., 35 (1956) 1345, Folch. J., Lees, M. and Sloane Stanley, G.H., A simple method for the isolation and purification of total lipids from animal tissues, J. Biol. Chem., 226 (1957) 497. Carlson, L.A., Determinations of serum triglycerides, J. Atheroscl. Res., 3 (1963) 334.
338
14 Sperry, W.M. and Webb, M.A.. A revision of the SchoenheimerrSperry method for cholesterol determination, J. Biol. Chem., 187 (1950) 97. 15 Morgan, T.E., Tinker, D.O. and Hanohan, D.J., Phospholipid metabolism in kidney, Part 1 (Isolation and identification of lipids of rabbit kidney), Arch. Biochem. Biophys., 103 (1963) 54. 16 Cheng, D.C.H. and Wolfe, B.M., Norenthindrone acetate inhibition of triglyceride synthesis and release by rat hepatocytes, Atherosclerosis, 46 (1983) 41. 17 Bernhard& F.W. and Tomarelli, R.M., A salt mixture supplying the National Research Council estimates of the mineral requirements of the rat, J. Nutr., 89 (1966) 495. 18 Greenfield, H., Briggs, G.M., Watson, R.H. and Yudkin, J., An improved diet for carbohydrate preference studies with rats - Some criticism of experimental diets. Proc. Nutr. Sot.. 28 (1969) 43A. 19 Hoehn, S.K. and Carroll, K.K., Effects of dietary carbohydrate on the incidence of mammary tumours induced in rats by 7,12-dimethylbenz (a) anthracene, Nutrition and Cancer, 1 (1979) 27. 20 Seglen, P.O., Preparation of isolated rat liver cells, Methods in Cell Biol.. 13 (1976) 29. 21 Borgstrom, G., Investigation of lipid separation methods - Separation of cholesterol esters. glycerides and free fatty acids, Acta Physiol. Stand., 25 (1952) 111. 22 Snedecor. G.W. and Cochran, W.G., Statistical Methods. 6th edition. Iowa State Umversity Press. Ames, IA, 1967, pp. 106 and 114. 23 Bell, F.P. and Hubert, E.V., Stimulation of hepatic squalene and triglyceride synthesis by dimethylsulfoxide, in vitro, Lipids, 17 (1982) 900. 24 Gorski, J. and Ganon, F., Current models of steroid hormone actton - ‘4 critique. Ann. Rev. Physiol.. 38 (1976) 425. 25 Schorbleret-Slatkine. S.. Schorderet. M. and Baulieu, E.E., Cyclic AMP-mediated control of meiosis - Effects of progesterone, cholera toxin, and membrane-active drugs in Xenopus lurvrs oocytes, Proc. Nat. Acad. Sci. (USA). 79 (1982) 850. 26 Mailer, J.L. and Sadler. SE., Regulation of steroid-induced cell diviston in amphibian oocytes by protein phosphorylation. In: O.M. Rosen and E.G. Krebs (Eds.), Conferences on Cell Proliferation. (Cold Spring Harbor Symp.), Cold Spring Harbor, Long Island, NY. Vol. 18. 1981. p. 1127. T., Progesterone receptor system of hen oviduct, J. 27 Murayama, A., Fukai, F. and Yamamoto. B&hem., 88 (1980) 1305. 28 Mulvihill, E.R. and Palmiter, R.D., Relationship of nuclear progesterone receptors to mduction of ovalbumin and conalbumin mRNA in chick oviduct, J. Biol. Chem., 255 (1980) 2085. 29 Uniyal, J.P., Buckshee, K., Bhargava. V.L., Hingorani, V. and Laumas. K.R., Binding of norgestrel to receptor proteins in the human endometrium and myometrium, J. Steroid Biochem., 8 (1977) 1183. 30 Haagsman, H.P., DeHaas, C.G.M., Geelen, M.J.H. and Van Golde. L.M.G.. Regulation of triacylglycerol synthesis in the liver - Modulation of diacylglycerol acyltransferase activity in vitro. J. Biol. Chem., 257 (1982) 10593. 31 Vydelingum, E.E., Insulin regulation of rat adipose tissue lipoprotein lipase synthesis. Diabetes. 31 (1982) 53A.
329 Ltd.
ATH 03518
Hypotriglyceridemic
Effects of Levonorgestrel Rats
Rama Khokha Departments
and Bernard
in
M. Wolfe
of Biochemistry and Medicine, The Unioersity of Western Ontario, London, Ont. (Canada) (Received 12 July, 1983) (Revised, received 17 February, 1984) (Accepted 23 March, 1984)
Summary The progestin, levonorgestrel administered orally to fed female rats significantly lowers both plasma total and very low density lipoprotein triglycerides. In contrast, plasma total cholesterol and low density lipoprotein cholesterol rose significantly. Suspensions of isolated hepatocytes were used to study the effects of levonorgestrel on triglyceride synthesis by examining the incorporation of labelled precursors ([9,10-3H]palmitate and [U-‘4C]glycerol) into triglycerides. Levonorgestrel (1O-4 M) significantly inhibited the incorporation of both precursors into hepatocyte triglycerides and also reduced their incorporation into the triglycerides (nearly all in d < 1.006) released into the medium. These results suggest that inhibition of hepatic triglyceride synthesis and release can account at least for part of the lowering of plasma VLDL which occurs during administration of levonorgestrel. Key words:
Hepatocytes - Lipoproteins - Levonorgestrel - Rats - Triglycerides
Introduction The progestin, dl-norgestrel [active isomer levonorgestrel] has been widely administered to women in the form of oral contraceptives [l]. Reports about the influence of dl-norgestrel-containing medications on serum lipids are conflicting. Although This research was supported by a grant from the Medical Research Research Studentship awarded to R.Khokha (1980-84) from the Medical appreciated.
0021-9150/84/$03.00
0 1984 Elsevier Scientific
Publishers
Ireland,
Ltd.
Council Research
of Canada Grant. A Council is also much
330
most investigators find that oral contraceptives containing dl-norgestrel lower serum triglycerides [2,3], this has not been observed by others [4- 71. This discrepancy likely relates to the type and dosage of estrogens used in the formulation, as estrogens tend to oppose the effect of progestins on plasma lipid and lipoprotein levels [8]. dl-Norgestrel alone, has been reported to lower triglycerides in premenopausal women [9]. The present study was undertaken to examine the effects of the active isomer levonorgestrel [lo] on lipid and lipoprotein levels in the rat and to elucidate its mechanism of action using isolated rat hepatocytes. Methods
and Materials
Materials [9,10-‘H]Palmitic acid (12 Ci/mmol) purchased from New England Nuclear obtained from Wyeth Ltd., Toronto, Ont. Millipore Corp., Mississauga, Ont. and supplied by Gibco Canada, Burlington, purchased from Sigma Chemical Co., St. Ralston-Purina Co., Minneapolis, MN.
and [U-14C]glycerol (12 pCi/pmol) were Corp., Boston, MA. Levonorgestrel was Collagenase (Type I) was obtained from Waymouth’s medium (MB 752/l) was Ont. Porcine albumin (fraction V) was Louis, MO. Rat chow was obtained from
Studies in vivo Female Sprague-Dawley rats (200-250 g), 7 in each group (control and experimental), were individually caged in an animal room illuminated from 8.00 to 20.00 h. Each rat in the experimental group received 4 pg levonorgestrel/day . kg body wtO.” (average dose 1.4 pg/day) for 15 days. The experimental diet was prepared by adding the levonorgestrel dissolved in absolute ethanol to one piece of the rat chow; ethanol was then evaporated. The levonorgestrel-containing chow was fed first (at 18.00 h) and the rat observed until it was completely consumed. The control diet was treated with an equal amount of ethanol which was likewise removed by evaporation. Thereafter, water and rat chow were available ad libitum to both levonorgestrel-treated and control rats. Rats received a fat-free diet for 12 h prior to killing to eliminate chylomicrons from plasma. They were venesected after receiving a sodium pentobarbital injection (13 mg in 0.2 ml) intraperitoneally and plasma samples were taken for determination of lipid content and for the separation of lipoproteins by the method of Have1 et al. [ll]. Plasma lipids were extracted in chloroform/methanol (2 : 1, v/v, [12]). Triglycerides, cholesterol and phospholipids were assayed by the methods previously described [13-161. Preparation of fat-free diet The fat-free diet prepared in the laboratory was a synthetic mixture of casein (18%), dextrose (73%), celluflour (5%), and salts (4%) as formulated by Bernhardt and Tomarelli [17]. The vitamin mixture was described by Greenfield et al. [18]. The composition of the salt mixture, water and fat-soluble vitamins was essentially the same as described earlier [19].
331
Studies in vitro using rat hepatocytes Female Sprague-Dawley rats (200-300 g) maintained on rat chow were used for preparation of hepatocytes by the method of Seglen [16,20]. Glucose (108, w/v) was added to the drinking water (to maintain a fed state) for 48 h prior to killing of the rats. Viability, determined by the trypan blue dye exclusion test, ranged from 60 to 90% (mean 71 + 3%) and the final intact hepatocyte yield ranged from 0.1 to 1.0 X 10’ cells per liver. Incubation of hepatocytes Each set of control and experimental preparation was carried out with cells freshly isolated from the same rat liver. In a typical experiment, incubations were carried out in a shaking water bath at 37 ‘C with hepatocytes (0.4-0.6 x 10” viable cells/ml) suspended in Waymouth’s medium in the presence or absence (control) of levonorgestrel (lop4 M). For dispersal in aqueous medium, levonorgestrel was dissolved in 0.1% dimethyl sulfoxide (DMSO); a similar amount of DMSO was present in the medium of the controls. 25 PCi of [9,10-3H]palmitate (complexed to porcine albumin, molar ratio approximately 1 : 1) and 25 PCi of [U-‘4C]glycerol were added to the medium immediately before the 0 sample. 50 PM palmitic acid complexed to albumin was also added to the medium (both experimental and control) in some experiments (see Fig. 4). Incubations were continued for 60 min. Samples of the suspended cells were removed at 0, 30, and 60 min and centrifuged at 3 500 rpm for 15 min at 5 ‘C to sediment the cells. The pellet (suspended in 0.5 ml ice-cold saline) and the supernatant were each extracted with chloroform/methanol (2 : 1) for determination of lipids. The chloroform layer of the chloroform/methanol extract of the lipids obtained from both the hepatocytes and the medium in which they were incubated was treated with silicic acid to remove phospholipids [13,20]. Labelled free fatty acids were removed by overalkalinization [16,21] from neutral lipids prior to determination of the 3H and 14C content of the neutral lipids (essentially triglycerides) of the hepatocytes and their supernatant. In several experiments, cells were sedimented from incubation media as described above and the supernatant was mixed with human serum (2: 1, v/v) prior to ultracentrifugation [ll] to determine the density fraction to which the secreted labelled lipoproteins corresponded. Statistics Differences between groups were evaluated according to Snedecor and Cochran [22] for both paired and unpaired samples. Variance is expressed as standard error of the mean. Results
Effects of levonorgestrel on plasma lipid and lipoprotein concentrations in vivo The lower mean plasma concentration of triglycerides (Table 1) was attributable to the significantly lower mean concentration of VLDL triglycerides (Table 2). The concentration of VLDL phospholipids fell in proportion to the decrease in VLDL
332
TABLE
1
EFFECTS
OF LEVONORGESTREL
ON THE CONCENTRATIONS
OF PLASMA
LIPIDS
IN VIVO
Plasma obtained from rats fed rat chow (control) or rat chow containing 4 pg levonorgestrel/day’kg body wt “j for 2 weeks was extracted and lipids were determined as described in Materials and Methods. Each value represents the mean + SE for 7 rats. Group
Control Levonorgestrel * Designates
Triglycerides
Cholesterol
Phospholipids
(mg/dl)
(mg/dl)
(mg/dl)
(Total)
(Total)
(Total)
34k2 25+3 a statistically
4652 54+4
*
significant
difference
62+4 67k6
*
from control
value at P < 0.025.
triglycerides (Table 2). In contrast, the concentrations of plasma total and LDL cholesterol increased significantly (Tables 1 and 2). LDL phospholipids were also significantly elevated by levonorgestrel treatment (Table 2). However, no significant differences were found between control and treated rats in the concentrations of cholesterol and phospholipids in HDL or its subfractions HDL, and HDL, (data not shown). Time course of the incorporation of labelled precursor into hepatocyte triglycerides and triglycerides released into the medium The time course of the incorporation of both labelled palmitate and glycerol into hepatocyte triglycerides was curvilinear, up to 60 min, in the absence of added palmitate (Fig. l), whereas with the addition of 50 PM unlabelled palmitate, it was linear (Fig. 1). The release of labelled triglycerides into the medium was slow for the first 30 min, but increased 3-fold in the next 30 min. As expected, the observed pattern is compatible with a precursor-product relationship between hepatocyte
TABLE
2
EFFECT OF LEVONORGESTREL PROTEIN FRACTIONS IN VIVO
ON THE CONCENTRATIONS
OF LIPIDS
IN PLASMA
LIPO-
Plasma obtained from rats fed rat chow (control) or rat chow containing 4 pg levonorgestrel/day. kg body wt 075for 2 weeks was separated into standard lipoprotein density fractions by ultracentrifugation. The individual lipoprotein fractions were extracted and the lipids determined as described in Materials and Methods. Each value represents the mean + SE for 7 rats.
Control Levonorgestrel * Designates
Triglycerides
Cholesterol
Phospholipids
(mg/dl)
(mg/dl)
(mg/dl)
VLDL 25+2 21*1*
VLDL 4+1 4+1
a statistically
significant
difference
LDL 11+1 15+2*
HDL 31+2 35+3
from control
VLDL 7+1 5+2
value at P -C 0.05.
LDL 7+1 10+1*
HDL 48+3 52*4
333
0
15
30
60
0
15
30
60
MINUTES
Fig. 1. Effect of added palmitate 50 pM on the incorporation of [9,10-3H]palmitate into hepatocyte triglycerides. Isolated hepatocyte suspensions were incubated with labelled palmitate (A) and glycerol (B) with added palmitate (A) or without added aliquots of the hepatocytes, removed at 0, 15, 30 and 60 min were extracted and determined as described under Materials and Methods. Each value represents the experiments.
and [U-‘4C]glycerol 25 pCi of each of palmitate (0). The triglyceride counts mean for at least 2
triglycerides and the triglycerides released into the medium. Over the 60-min period of incubation with labelled precursors the release of labelled triglycerides into the medium accounted for only about 1% of labelled triglycerides synthesized in hepatocytes. Most of the labelled triglycerides released into the medium (89 + 5% of [ 3H] and 95 f 2% of [14C]), were in the density fraction d < 1.006 corresponding to VLDL, with only minor amounts in d > 1.006 (Table 3). Effect of levonorgestrel on triglyceride synthesis and release by isolated hepatocytes Levonorgestrel (10e4 M) significantly reduced the incorporation of both [9,103H]palmitate and [U-t4C]glycerol into hepatocyte triglycerides (P < 0.05) (Fig. 2). The inhibitory effect of levonorgestrel reached a plateau within 30 min of incubation. At that point in time, the per cent inhibition of the incorporation of labelled palmitate averaged 20 f 3% and that of glycerol 19 + 2%. Both values are significantly lower than control (P < 0.05) though similar to each other. Release of [ 3H]TABLE
3
PERCENT DISTRIBUTION OF LABELLED TIONS OF THE INCUBATION MEDIUM
TRIGLYCERIDES
AMONG
LIPOPROTEIN
FRAC-
Medium obtained from hepatocyte cell suspensions (at 0’ and 60’) incubated with 25 PCi of [9,103H]palmitate+ unlabelled palmitate (50 PM) and 25 gCi of [U-‘4C]glycerol was separated into lipoprotein fractions by ultracentrifugation. The individual lipoprotein fractions were extracted and triglycerides separated as described under Materials and Methods. The values represent the % increase of medium triglyceride counts in different density fractions over 60 min for 4 individual experiments. Density
13W f14Cl ND = not detected.
fractions
(n/ml)
1.006
1.006-1.019
1.019-1.063
1.063
g9+5 95+2
6+3 3k2
5+3
ND ND
2*1
E
a* ,’ ,
-P
-9’’
/’ I
/
, I
,
0
30
60
0
30
60
MINUTES
Fig. 2. Time course of incorporation of labelled precursors into hepatocyte triglycerides. The isolated rat hepatocyte suspensions were incubated with 25 PCi of [9,10-‘Hlpalmitate (A) and 25 PCi of [U“C]glycerol (B) in the presence (0) or absence (0) of 10e4 M levonorgestrel. The medium concentration of DMSO was 0.1%. The aliquots of the hepatocytes removed at 0, 30 and 60 min were extracted and triglyceride counts determined as described under Materials and Methods. Each value represents the mean for 5 experiments. Experimental values for both palmitate and glycerol were significantly lower than control at both 30 and 60 min (P < 0.05).
and [‘4C]triglycerides into the medium (Fig. 3) was also significantly reduced by 10e4 M levonorgestrel (P < 0.05). Maximum inhibition of the incorporations of both labelled palmitate and glycerol into triglycerides released into the medium was observed at 60 min (51 + 11% and 54 f 13%, respectively, P < 0.05). Effect of higher levels of DMSO on the hepatocyte response to levonorgestrel The incorporation of labelled precursors into hepatocyte triglycerides and triglycerides released into the medium was unaffected by 0.1% DMSO. Levonorgestrel (10m4 M), initially dissolved in 0.1% DMSO, partially precipitated upon addition to the medium. Progressively higher amounts of DMSO (O.lW, 0.25’%, 0.5% and 1.0%) were used to suspend the levonorgestrel (10e4 M) in the medium and the results
0
30
60
0
30
60
MINUTES
Fig. 3. Time course of release of [ 3H]- and [‘4C]triglycerides in the medium. The isolated hepatocyte suspensions were incubated with 25 PCi of [9,10-3H]palmitate (A) and 25 pCi of [U-‘4C]glycerol (B) in the presence (0) or absence (0) of 10e4 M levonorgestrel. The aliquots of the medium removed at 0, 30 and 60 mm were extracted and triglyceride counts determined as described under Materials and Methods. Each point represents the mean value for 5 different experiments. Experimental values for both palmitate and glycerol were significantly lower than control at both 30 and 60 min (P c 0.05).
335
0
0.1
0.25
0.5 %
DMSO
Fig. 4. The percent inhibition by levonorgestrel of incorporation of labelled precursors into hepatocyte triglycerides with increasing medium concentrations of dimethylsulfoxide (DSMO). The isolated hepatocyte suspensions were incubated with levonorgestrel (10s4 M) suspended with varying medium concentrations of DMSO (O.lW-1.0%) in the presence of 25 BCi of each [U-‘4C]glycerol (m) and [9,10-3H]palmitate (0) supplemented with 50 pM unlabelled palmitate. Control incubations were carried out with the same amounts of DMSO as in the experimental incubations, but in the absence of levonorgestrel. Hepatocyte triglycerides were extracted and their counts determined as described in Materials and Methods. Each value represents the mean f SE of 3 individual experiments.
were compared to controls (i.e. hepatocytes incubated without levonorgestrel, but the same amount of DMSO, Fig. 4). Mean values for the incorporation of palmitate into hepatocyte triglycerides in the presence of DMSO were not significantly different (P > 0.2) from their respective controls at any of the levels of DMSO that were tested as follows: 0.1% DMSO (16.8 + 1.9 vs. 15.4 + 2.4 x lo5 cpm/106 cells); 0.25% DMSO (19.8 f 1.6 vs. 17.9 + 0.8 X lo5 cpm/106 cells); 0.5% DMSO (20.2 rt 1.5 vs. 17.9 + 0.8 X lo5 cpm/106 cells); or 1.0% DMSO (18.2 f 1.8 vs. 16.2 + 0.7 x lo5 cpm/106 cells). Likewise, mean values for the incorporation of glycerol into hepatocyte triglycerides were not significantly (P > 0.2) different from their respective controls at any of the levels of DMSO tested as follows: 0.1% DMSO (14.8 k 1.5 vs. 14.3 + 1.8 X lo4 cpm/106 cells); 0.25% DMSO (16.7 & 0.3 vs. 16.1 k 0.6 x lo4 cpm/106 cells); 0.5% DMSO (15.7 _t 0.8 vs. 16.1 f 0.6 x lo4 cpm/106 cells); or 1.0% DMSO (16.7 k 10 vs. 14.7 + 0.5 X lo4 cpm/106 cells). However, in the presence of higher concentrations of DMSO the effect of levonorgestrel was enhanced. The incorporation of labelled palmitate into hepatocyte triglycerides in the presence of 10m4 M levonorgestrel was inhibited by only 20 + 3% when the concentration of DMSO was 0.1% compared to 26 &-0.6% with 0.25% DMSO, 31 & 1% with 0.5% DMSO and 39 k 0.8% with 1% DMSO. Likewise, the inhibition of the incorporation of labelled glycerol into hepatocyte triglycerides increased from 19 f 2% with 0.1% DMSO to 42 + 2% with 1.0% DMSO. Discussion The present studies demonstrate that a low dose of levonorgestrel, comparable to that used for human contraceptive purposes, significantly lowers serum total and VLDL triglycerides in fed female rats, while elevating total and LDL cholesterol levels. The parallel reductions of VLDL phospholipids and VLDL triglycerides
336
suggest that the lowering of VLDL is attributable to reduction of the number of VLDL particles rather than to the formation of abnormal VLDL particles. The measurements of serum lipids were made under fed-state conditions using a fat-free diet which excluded the effect of chylomicrons from the circulation which would have confounded the interpretation of the results. Although lipogenesis from carbohydrate is likely to be enhanced by a fat-free diet, the diet was the same for both norgestrel-treated and control rats. This allowed one to ascertain the effect of the levonorgestrel alone. Lower VLDL levels could result from either reduced hepatic triglyceride secretion or enhanced clearance of VLDL from the blood plasma. Our studies in vitro lend support to the former mechanism, but have not examined clearance mechanisms. Levonorgestrel rapidly inhibited the synthesis of triglycerides from palmitate and glycerol in isolated rat hepatocytes (Figs. 1 and 2). However, the extent of the inhibition was influenced by the amount of DMSO used to suspend the levonorgestrel (Fig. 3). Because incorporation of both substrates into triglycerides was inhibited to a similar extent by levonorgestrel, it is attractive to suggest that the inhibition may be at the level of either glycerol-3-phosphate acyltransferase, phosphatidic acid phosphatase and/or diacylglycerol acyltransferase, the three potentially rate-limiting enzymes shared by these substrates on their pathway to triglycerides. The inhibition by levonorgestrel of triglyceride release from hepatocytes (Figs. 3 and 4) could be accounted for solely by the reduction of hepatocyte triglyceride synthesis. However, because triglyceride release was inhibited by levonorgestrel to a greater extent (51-548) than hepatocyte triglyceride synthesis (19-20%) it is conceivable that levonorgestrel might have additional effects on the release of VLDL from hepatocytes. The present results suggest (i) that liver may be an important site of the hypolipidemic action of levonorgestrel and (ii) that inhibition of hepatic triglyceride synthesis and release can accaunt at least for part of the hypolipemic effect of levonorgestrel observed in vivo in fed rats. In contrast to the in vivo studies which were done over an extended time with a dose of levonorgestrel which is comparable to the conventional therapeutic doses given to women, the in vitro studies with isolated hepatocytes employed much higher pharmacologic doses which permitted much shorter studies. Although solubilization of the levonorgestrel posed a problem in the in vitro studies, we found that DMSO was useful in suspending the levonorgestrel. In the concentrations employed, DMSO alone inhibited neither hepatocyte triglyceride synthesis nor release. This contrasts with a previous report by Bell et al. [23] that a higher concentration of DMSO (5%) produces a 7-fold increase in the incorporation of [14C]acetate into triglycerides of surviving rat liver slices. The effects of DMSO levels lower than 5% have not previously been reported. Because the inhibitory effect of levonorgestrel occurred so rapidly, the inhibitory action in vitro of levonorgestrel on hepatocyte triglyceride synthesis and release may have occurred by a mechanism which is different from that generally known to account for steroid action. Steroids are generally thought to act by binding to receptors in the cytosol which translocate to the nucleus, bind to DNA and then mediate their action by inducing transcription of specific mRNA [24]. However,
337
progesterone is also known to alter CAMP levels by binding to membrane receptors [25,26], in addition to its known action via cytosolic receptors [27,28]. Levonorgestrel could act by altering CAMP levels by a mechanism similar to progesterone. Levonorgestrel, a 19-nortestosterone derivative, also has some androgenic activity [29]; however, cytosolic receptors for levonorgestrel have been shown to be competitively displaced by progesterone, but only weakly by testosterone [29]. It is not yet known whether there are membrane receptors for levonorgestrel. Additional explanations for the rapid effect of levonorgestrel could include (1) the phosphorylation-dephosphorylation mechanism regulation diacylglycerol acyltransferase proposed by Haagsman et al. [30] for the inhibitory action of glucagon on hepatic triglyceride synthesis, and/or (2) selective translation of mRNA independent of nucleus events [31]. Acknowledgements The authors are grateful for typing the manuscript.
to E. Wolfe for helping
with the artwork
and S. Brock
References 1 Tyler, E.T., International 2
3
4
5
6
7
8 9 10 11 12 13
aspects of the use of norgestrel. In: The Second International Norgestrel Symposium, Excerpta Medica, Amsterdam, 1974, p. 1. Nielsen, F.H., Honore, E., Kristoffersen, K., Secher, N.J. and Pedersen, T., Changes in serum lipids during treatment with norgestrel oestradiol-valerate and cycloprogynon, Acta Obstet. Gynaecol. Stand., Suppl. 56 (1977) 367. Robertson, D.N., Alvarez, F., Sivin, I., Brache, V., Stern, J., Leon, P. and Faundes, A., Lipoprotein patterns in women in Santo Domingo using a levonorgestrel/estradiol contraceptive ring, Contraception, 24 (1981) 469. Foegh, M., Pederson, F.D., Gormsen, J., Knudsen, J.B. and Schou, G., Oral levonorgestrel-testosterone effects on spermatogenesis, hormone levels, coagulation factors and lipoproteins in normal men, Contraception, 24 (1980) 381. Riissner, S., Frankman, 0. and Marsk, L., Effects of various low dose contraceptive pills on serum lipoproteins. In: L.W. Hessel and H.H.J. Krans (Eds.), Lipoprotein Metabolism and Endocrine Regulation, Elsevier, Amsterdam 1979, pp. 91-98. Silfverstolpe, G., Gustafson, A., Samsioe, G. and Swanborg, A., Lipid metabolic studies in oophorectomized women - Effects of three different progestogens, Acta Obstet. Gynaecol. Stand.. Suppl. 88 (1979) 89. Pribicevic, V., Skrabalo, Z., Lipovac, V., Trenc, S., Zadjelovic, J. and Rajkovic, Z., Oral contraception, mechanical contraception and carbohydrate and lipid metabolism - A two year study. Int. J. Fertil., 24 (1979) 114. Beck, P., Contraceptive steroids - Modifications of carbohydrate and lipid metabolism, Metabolism, 22 (1973) 841. Spellacy, W.N., Buhi, N.C. and Birk, S.A., Norgestrel and carbohydrate-lipid metabolism - Glucose, insulin and triglyceride changes during six months time of use, Contraception, 9 (1974) 615. Robert, C.J., Singer, A.C. and Edgren, A.R., The biological activities of norgestrel and its enantiomers, Int. J. Fertil., 24 (1979) 39. Havel, R.J., Eder, H.A. and Bragdon, J.H., The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum, J. Clin. Invest., 35 (1956) 1345, Folch. J., Lees, M. and Sloane Stanley, G.H., A simple method for the isolation and purification of total lipids from animal tissues, J. Biol. Chem., 226 (1957) 497. Carlson, L.A., Determinations of serum triglycerides, J. Atheroscl. Res., 3 (1963) 334.
338
14 Sperry, W.M. and Webb, M.A.. A revision of the SchoenheimerrSperry method for cholesterol determination, J. Biol. Chem., 187 (1950) 97. 15 Morgan, T.E., Tinker, D.O. and Hanohan, D.J., Phospholipid metabolism in kidney, Part 1 (Isolation and identification of lipids of rabbit kidney), Arch. Biochem. Biophys., 103 (1963) 54. 16 Cheng, D.C.H. and Wolfe, B.M., Norenthindrone acetate inhibition of triglyceride synthesis and release by rat hepatocytes, Atherosclerosis, 46 (1983) 41. 17 Bernhard& F.W. and Tomarelli, R.M., A salt mixture supplying the National Research Council estimates of the mineral requirements of the rat, J. Nutr., 89 (1966) 495. 18 Greenfield, H., Briggs, G.M., Watson, R.H. and Yudkin, J., An improved diet for carbohydrate preference studies with rats - Some criticism of experimental diets. Proc. Nutr. Sot.. 28 (1969) 43A. 19 Hoehn, S.K. and Carroll, K.K., Effects of dietary carbohydrate on the incidence of mammary tumours induced in rats by 7,12-dimethylbenz (a) anthracene, Nutrition and Cancer, 1 (1979) 27. 20 Seglen, P.O., Preparation of isolated rat liver cells, Methods in Cell Biol.. 13 (1976) 29. 21 Borgstrom, G., Investigation of lipid separation methods - Separation of cholesterol esters. glycerides and free fatty acids, Acta Physiol. Stand., 25 (1952) 111. 22 Snedecor. G.W. and Cochran, W.G., Statistical Methods. 6th edition. Iowa State Umversity Press. Ames, IA, 1967, pp. 106 and 114. 23 Bell, F.P. and Hubert, E.V., Stimulation of hepatic squalene and triglyceride synthesis by dimethylsulfoxide, in vitro, Lipids, 17 (1982) 900. 24 Gorski, J. and Ganon, F., Current models of steroid hormone actton - ‘4 critique. Ann. Rev. Physiol.. 38 (1976) 425. 25 Schorbleret-Slatkine. S.. Schorderet. M. and Baulieu, E.E., Cyclic AMP-mediated control of meiosis - Effects of progesterone, cholera toxin, and membrane-active drugs in Xenopus lurvrs oocytes, Proc. Nat. Acad. Sci. (USA). 79 (1982) 850. 26 Mailer, J.L. and Sadler. SE., Regulation of steroid-induced cell diviston in amphibian oocytes by protein phosphorylation. In: O.M. Rosen and E.G. Krebs (Eds.), Conferences on Cell Proliferation. (Cold Spring Harbor Symp.), Cold Spring Harbor, Long Island, NY. Vol. 18. 1981. p. 1127. T., Progesterone receptor system of hen oviduct, J. 27 Murayama, A., Fukai, F. and Yamamoto. B&hem., 88 (1980) 1305. 28 Mulvihill, E.R. and Palmiter, R.D., Relationship of nuclear progesterone receptors to mduction of ovalbumin and conalbumin mRNA in chick oviduct, J. Biol. Chem., 255 (1980) 2085. 29 Uniyal, J.P., Buckshee, K., Bhargava. V.L., Hingorani, V. and Laumas. K.R., Binding of norgestrel to receptor proteins in the human endometrium and myometrium, J. Steroid Biochem., 8 (1977) 1183. 30 Haagsman, H.P., DeHaas, C.G.M., Geelen, M.J.H. and Van Golde. L.M.G.. Regulation of triacylglycerol synthesis in the liver - Modulation of diacylglycerol acyltransferase activity in vitro. J. Biol. Chem., 257 (1982) 10593. 31 Vydelingum, E.E., Insulin regulation of rat adipose tissue lipoprotein lipase synthesis. Diabetes. 31 (1982) 53A.