Oral contraceptives increase insulin-like growth factor binding protein-l concentration in women with polycystic ovarian disease*

Transcription

1 FERTILITY AND STERILITY Copyright e 1991 The American Fertility Society Printed on acid-free paper in U.S.A. Oral contraceptives increase insulin-like growth factor binding protein-l concentration in women with polycystic ovarian disease* Anne-Maria Suikkari, M.D.t Aila Tiitinen, M.D.t Ulf-Hakan Stenman, M.D.t Markku Seppala, M.D.t Timo Laatikainen, M.D.t:j: University of Helsinki, Helsinki, and University of Ouiu, Ouiu, Finland Insulin-like growth factor-i (IGF-I) stimulates ovarian androgen production. Insulin-like growth factor binding protein-1 (IGFBP-1) inhibits IGF actions in vitro. Objective: To investigate the effect of oral contraceptive (OC) pills, given for 3 months, on serum gonadotropin, androgen, IGF-I, and IGFBP-1 concentrations, and glucose tolerance in seven women with polycystic ovarian disease (PCOD) and in five healthy control subjects. Patients: Seven women with PCOD and five healthy control subjects. Interventions: An oral glucose tolerance test (OGTT) was performed before and after treatment with ~C. Results: After treatment with ~C, serum luteinizing hormone, androstenedione, and free testosterone levels decreased, and sex hormone-binding globulin concentration increased in the women with PCOD as well as in the control subjects. The cumulative response of serum insulin to OGTT was larger in the women with PCOD than in the control subjects both before and after treatment. Serum IGF-I concentration, which was unchanged during OGTT, decreased from basal level of 326 ± 70 ~g/l to 199 ± 28 ~g/l after treatment with OC in the women with PCOD, whereas no change was found in the control subjects (from 235 ± 11 ~g/l to 226 ± 11 ~g/l). Treatment with OC caused an increase of the mean basal IGFBP-1 concentration from 24 ± 7 ~g/l to 73 ± 14 ~g/l in the women with PCOD. This increase was constant during the OGTT. In the control subjects, treatment with OC did not result in any significant change in IGFBP-1 concentrations (from 44 ± 11 ~g/l to 61 ± 9 ~g/l). Conclusion: The combination of decreased total IGF-I concentration and increased IGFBP-1 concentration induced by OC may decrease ovarian androgen production in PCOD. Fertil Steril 55:895, 1991 Contraceptive pills are widely used in treatment of polycystic ovarian disease (PCOD) of women who do not wish to become pregnant. Combination oral Received August 28, 1990; revised and accepted January 16, * Supported by grants from the Finnish Cultural Foundation, Helsinki, Finland; the Paulo Foundation, Helsinki, Finland; the Sigrid Juselius Foundation, Helsinki, Finland; the Finnish Social Insurance Institution, Helsinki, Finland; and the Academy of Finland, Helsinki, Finland. t Department I of Obstetrics and Gynecology, University of Helsinki. :j: Present address: Department of Obstetrics and Gynecology, University of Oulu. Reprint requests: Timo Laatikainen, M.D., Department of Obstetrics and Gynecology, Oulu University Central Hospital, SF Oulu, Finland. contraceptives (OC) containing estrogen (E) and progestin suppress gonadotropin secretion resulting in normalization of excessive luteinizing hormone (LH) secretion. 1,2 Subsequently, the ovarian androgen secretion is reduced, often leading to relief of acne, hirsutism, and seborrhea. 1 Estrogen decreases the ovarian androgen secretion and increases the sex hormone-binding globulin (SHBG) synthesis in liver. Because of its antiandrogenic effect, OCs containing cyproterone acetate are effective in the treatment of androgenic symptoms in PCOD.3 Growth factors regulate ovarian androgen production. 4 Insulin-like growth factors (IGFs) and insulin stimulate androgen synthesis in ovarian stroma and in theca-interstitial cells. 5,6 Insulin-like growth factor binding proteins (IGFBP) modulate the bio- Suikkari et al. Effect of OC on serum IGFBP-l in pcon 895

2 logical actions IGFs.7 Three human IGFBPs have been purified and cloned from various sources. Primarily produced by the liver,8 IGFBP-l is also synthesized in the secretory endometrium and in human granulosa cells. 9,10 In certain cell types, IGFBP-l inhibits the biological action of IGF-I in vitro, whereas in other cell types its action may be mitogenic.ll,12 In PCOD, serum IGF-I levels are elevated and those of IGFBP-l are decreased.13,14 These changes are proposed to enhance ovarian androgen production in PCOD. In nonpregnant individuals, the main site ofigf -I and IGFBP-l synthesis is believed to be the liver. As OCs are known to stimulate synthesis of certain proteins in the liver,15 it was of particular interest to study the effect of OCs on serum IGF-I, IGFBP-l, SHBG, insulin, and androgen levels in control subjects and PCOD patients. MATERIALS AND METHODS Seven women with PCOD, 16 to 31 years of age, and five control subjects, 19 to 28 years of age, were studied. The diagnosis of PCOD was based on clinical and biochemical criteria. All the subjects with PCOD had at least four of the following features: menstrual disturbances (n = 7), obesity (n = 2), hirsutism (n = 5), LH/follicle-stimulating hormone (FSH) ratio> 2 (n = 6), hyperandrogenemia (n = 6), or ultrasound findings of ovaries characteristic of PCOD (n = 6). None of the control subjects had more than one of these features. They were healthy women who needed contraception and had no contraindications for the use of ethinylestradiol and cyproterone acetate. The contraceptive pill used in this study was Diane nova (Leiras, Turku, Finland), which contains 35 J.Lg ethinylestradiol and 2 mg cyproterone acetate. The pill was selected because of the antiandrogenic nature of cyproterone acetate to suppress the local effects of hyperandrogenemia. An oral glucose tolerance test (OGTT) was performed after an overnight fast between 8:00 A.M. and 11:00 A.M. before and after 3 months' treatment. Blood samples were taken at 0, 30, 60, 120, and 180 minutes. Aliquots of serum were collected for the measurements of serum IGFBP-l, IGF-I, SHBG, insulin, and gonadotropin and androgen concentrations. Serum samples were stored at -20 C until analyzed. All subjects started taking the pill on the 1st day of the period. None ofthe subjects had taken any hormone medication for at least 3 months before the first OGTT. All subjects gave their informed consent before they volunteered to participate in the study. The protocol of the study was approved by the Ethical Committee of the hospital. Measurements of serum IGFBP-l concentrations were carried out by a radioimmunoassay (RIA) using purified IGFBP-l from human amniotic fluid and polyclonal rabbit anti-igfbp-l antiserum.16,17 Radioiodination was carried out by the lactoperoxidase method.18 In a final dilution of 1:50,000 to 1:150,000, the antiserum bound 25% to 30% of (1251)IGFBP-1. Unlabeled standards or serum samples (100 J.LL) were incubated with 100 J.LL radiolabeled IGFBP-l (about 15,000 cpm) and 100 J.LL antiserum at room temperature (RT) overnight. Antibody-bound and free radioactivity were separated by adding 500 J.LL Kaolin DASP (Farmos Diagnostica, Oulunsalo, Finland). After incubation for 15 minutes at RT, vortexing and incubating for additional 15 minutes, tubes were centrifuged at 1700 X g for 20 minutes at 4 C, and radioactivity of the pellets was measured in a gamma counter. Sensitivity of the assay was 2.3 J.Lg/L. This assay measures IGFBP-l both in bound and unbound forms, as evidenced by identical displacement curves for the binding protein preincubated with and without unlabeled IGF-1. The assay is unaffected by added IGF-I in vitro.19 The intraassay and interassay coefficients of variation were 6% to 10%. Insulin-like growth factor 1 concentrations were measured using RIA kits from ImmunoNuclear Corporation (IncStar Corp., Stillwater, MN). In this assay, IGF-I is separated from its binding proteins by extraction with octadodecylsilica. Serum insulin concentrations were measured by RIA kits (Phadeseph; Pharmacia, Uppsala, Sweden). Plasma glucose concentrations were measured by a glucose oxidase method. Serum SHBG concentrations were measured by immunofluorometric assay (IFMA) kits from LKB/Wallac, Turku, Finland. Serum LH and FSH concentrations were measured by IFMA using kits from LKB/Wallac.20 Serum testosterone (T) was measured after extraction with ethylether/ethylacetate (9:1 vol/vol) by RIA kits from Farmos Diagnostica, Oulunsalo, Finland. Serum androstenedione (A) was measured after extraction with ethylether/ethylacetate (9:1 vol/vol) by Coat-A Count RIA kits from Diagnostic Products Corp., Los Angeles, CA. Serum dehydroepiandrosterone sulfate (DHEAS) was measured by RIA kits from Diagnostic Products Corp., Los Angeles, CA. Comparisons between the groups were made using the unpaired Student's t-test after logarithmic transformation, and comparisons of values from the same 896 Suikkari et al. Effect of OC on serum IGFBP-l in PCOD Fertility and Sterility

3 individual were made using the paired Student's t test after logarithmic transformation. RESULTS The mean body mass index (BM!) was comparable between the control subjects and women with peod. Furthermore, there was no difference in BMI compared before and after treatment with oe (Table 1). Before treatment, the mean serum LH and free T concentrations were higher in the women with peod than in the control subjects. After treatment with oe, serum LH, A, and free T levels decreased in the women with peod as well as in the control subjects. Treatment with oe increased serum SHBG concentrations 4.6-fold in the women with peod and 3.6-fold in the control subjects. Mean serum SHBG concentration tended to be lower in the women with peod (P = 0.08). There was no change in mean serum SHBG concentration during OGTT in the peod patients and control subjects. The response of mean blood glucose levels to OGTT did not show any difference between the two groups studied (Fig. 1). Although the glucose concentrations tended to be higher in the women with peod compared with control subjects, the difference was not significant. There was no change in the glucose tolerance of any of the subjects after treatment with oe (Fig. 1). Basal serum insulin concentration was not higher in the women with peod compared with the control subjects. However, the cumulative response of serum insulin was larger in the women with peod than in the control subjects both before (75 ± 11 mull and 23 ± 3 mull, P < 0.001, respectively) and after treatment (69 ± 10 mull and 38 ± 8 mull, P < 0.05, respectively). Serum IGF-I concentration decreased after treatment with oe in the women with peod, whereas no change was found in the control subjects (Table 1). Insulin-like growth factor I concentration did not change during the OGTT. The mean serum IGFBP- 1 concentration decreased during the OGTT both in the peod patients and in the control subjects before and after the treatment with oe (Fig. 1). After treatment, a slightly delayed decrease was found in the peod patients, being significant at 180 minutes, as compared with 120 minutes before treatment and to 60 minutes in the control subjects both before and after treatment. Treatment with oe caused a threefold increase of the mean basal IGFBP-1 concentration in the women with peod. This increase was constant during the OGTT. In the control subjects, treatment with oe did not result in any significant change in IGFBP-1 concentrations. DISCUSSION This study confirmed that treatment with oral low-dose combination contraceptives normalizes the pre-existing elevated LH levels in the women with Table 1 Gonadotropins, Androgens, Insulin, Glucose, IGFBp 1, and IGF-I Values of the Women With PCOD and Control Subject Before and After Treatment With Contraceptive Pills Containing 35 /lg Ethinylestradiol and 2 mg Cyproterone Acetate" PCOD Control subjects Before treatment Parameter (n = 7) BMI (kg/m2) 24.6 ± 2.3 LH (U/L) 11.2 ± 1.2' FSH (U/L) 3.8 ± 0.4 LH/FSH 3.2 ± 0.7' T (nmol/l) 3.0 ± 0.3 Free T (nmol/l) ± e A (nmol/l) 13.8 ± 3.9 DHEAS (/lmol/l) 4.8 ± 0.7 SHBG (nmol/l) 27 ± 5 Glucose (mmol/l) 4.8 ± 0.5 Insulin (mu/l) 13.6 ± 3.5 IGFBP-1 (/lg/l) 24 ± 7 IGF-I (/lg/l) 326 ± 70 "Values are means ± SE. b Significant change after treatment, P < 0.01., Significant difference before treatment, P < After Before After treatment treatment treatment (n = 7) (n = 5) (n = 5) 23.5 ± ± ± ± 1.6 b 5.1 ± ± 0.7 b 2.5 ± ± ± ± ± ± ± 0.5d 2.2 ± ± ± 0.003f ± ± b 8.1 ± 1.6d 8.0 ± ± 0.3 d 5.3 ± ± ± ± 19f 41 ± ± 24f 4.6 ± ± ± ± ± ± ± 14d 44 ± ± ± 28d 235 ± ± 11 d Significant change after treatment, P < e Significant difference before treatment, P < f Significant change after treatment, P < Suikkari et al. Effect of DC on serum IGFBP-l in PCDD 897

4 -0- PCOOmean.. PCOlmean... KOmean 2+--'--~--~~--~~ 150 ~ 125.s 100.!: ~ 75.!: E 50 2 ~ ::J ~ 80 ci.. 60 '" " E 2 (J) " 20 0 O+--,---.--~-.----~ Time (min) Figure 1 Responses of concentrations of glucose in plasma and insulin and IGFBP-1 in serum to OGTT in PCOD patients (peo) and in healthy control subjects (K) before (0) and after (1) treatment with OC pills. induced thecal cell activity. Both these changes are likely to counteract the insulin-induced enhancement of LH action and thus ameliorate the disturbed intraovarian regulatory mechanisms in peod. Treatment with oe did not have any effect on insulin or glucose response to OGTT. The decrease of IGFBP-l concentrations during OGTT was similar to those reported previously.22,23 Despite the elevated basal IGFBP-l levels after treatment with oe, serum IGFBP-l concentration decreased during the OGTT in women with peod, although the response was somewhat delayed. In the control subjects, the decrease of IGFBP-l concentration during OGTT was similar both before and after treatment with oe. This shows that oe does not modulate the acute effect of insulin on serum IGFBP-l concentration. The larger cumulative insulin response in the women with peod compared with control subjects suggests that there is a decrease of insulin resistance in women with peod in spite of their near normal basal insulin concentrations, normal glucose tolerance, and BMI. It has been shown by several authors that insulin resistance is not improved by suppression of hyperandrogenism in the peod patients.24,25 In the present study, insulin response to OGTT remained unchanged during treatment with oe in the peod patients. peod. This decreases the stimulation of ovarian thecal and stromal cells by LH. The increase of serum SHBG concentration, both in the women with peod and the control subjects, resulted in a decrease of serum free T concentration. The decreased LH secretion and increased circulating SHBG concentration during treatment with oe were previously thought to be the main factors reducing hyperandrogenemia in women with peod.1 The present findings suggest that insulin, IGF-I, and IGFBP-l may also playa part. Insulin and IGF-I are proposed to enhance the stimulatory effects of LH on ovarian androgen production.21 Treatment with the oe increased the serum IGFBP-l concentration and decreased the IGF-I concentration in the peod patients. A plausible explanation for increased serum IGFBP-l concentration is E-induced stimulation of IGFBP- 1 synthesis in the liver. A possible effect of oe on IGFBP-l synthesis in the granulosa cells lo cannot be excluded. The concomitant decrease of IGF-I concentration is difficult to explain. By increasing the IGFBP-l concentration and decreasing the total IGF-I concentration, oes could reduce the IGF-I- REFERENCES 1. Goldzieher JW: Polycystic ovarian disease. Fertil Steril 35: 371, Talbert LM, Sloan C: The effect oflow-dose oral contraceptive on serum testosterone levels in polycystic ovarian disease. Obstet Gynecol 53:694, Frohlich M, Vader HL, Walma ST, De Rooy H: The influence of long-term treatment with cyproterone acetate-ethinylestradiol combination on androgen levels in blood of hirsute women. J Steroid Biochem 12:499, Adashi EY, Resnick CE, Hernandex ER, Svoboda ME, Van Wyk JJ: Potential relevance of insulin-like growth factor I to ovarian physiology: from basic science to clinical application. Semin Reprod Endocrinol 7:94, Barbieri RL, Makris A, Randall RW, Daniels G, Kistner RW, Ryan KJ: Insulin stimulates androgen accumulation in incubations of ovarian stroma obtained from women with hyperandrogenism. J Clin Endocrinol Metab 62:904, Cara JF, Rosenfield RL: Insulin-like growth factor I and insulin potentiate luteinizing hormone-induced androgen synthesis by rat ovarian thecal-interstitial cells. Endocrinology 123:733, Baxter RC, Martin JL: Binding proteins for the insulin-like growth factors: structure, regulation and function. Prog Growth Factor Res 1:49, Julkunen M, Koistinen R, Aalto-Setala K, Seppala M, Janne OA, Kontula K: Primary structure of human insulin-like 898 Suikkari et al. Effect of oe on serum IGFBP-l in peod Fertility and Sterility

5 growth factor-binding protein/placental protein 12 and tissuespecific expression of its mrna. FEBS Lett 236:295, Rutanen E-M, Koistinen R, Wahlstrom T, Bohn H, Ranta T, Seppala M: Synthesis of placental protein 12 by human decidua. Endocrinology 116:1304, Suikkari A-M, Jalkanen J, Koistinen R, Biitzow R, Ritvos 0, Ranta T, Seppala M: Human granulosa cells synthesize low molecular weight insulin-like growth factor-binding protein. Endocrinology 124:1088, Ritvos 0, Ranta T, Jalkanen J, Suikkari A-M, Voutilainen R, Bohn H, Rutanen E-M: Insulin-like growth factor (IGF) binding protein from human decidua inhibits the binding and biological action of IGF-I in cultured choriocarcinoma cells. Endocrinology 122:2150, Elgin RG, Busby WH, Jr, Clemmons DR: An insulin-like growth factor (lgf) binding protein enhances the biologic response to IGF-1. Proc Nat! Acad Sci VSA 84:3254, Vrdl W: Polycystic ovarian disease: endocrinological parameters with specific reference to growth hormone and somatomedin-c. Arch Gynecol Obstet 243:13, Suikkari A-M, Ruutiainen K, Erkkola R, Seppala M: Low levels of low molecular weight insulin-like growth factorbinding protein in patients with polycystic ovarian disease. Hum Reprod 4:136, Von Schoulz B, Stigbrand T: Pregnancy zone protein: chemistry, biology and clinical studies. In Pregnancy Proteins: Biology, Chemistry and Clinical Application, Edited by JG Grudzinskas, B Teisner, M Seppala. New York, Academic Press, 1982, p Koistinen R, Huhtala M-L, Stenman V-H, Seppala M: Purification of placental protein PP12 from human amniotic fluid and its comparison with PP12 from placenta by immunological, physicochemical and somatomedin-binding properties. Clin Chim Acta 164:293, Koistinen R, Stenman V-H, Alfthan H, Seppala M: Timeresolved immunofluorometric assay of 34-kDa somatomedinbinding protein. Clin Chem 33:1126, Seppiila M, Ronnberg L, Karonen S-L, Kauppila A: Micronized oral progesterone increases the circulating level of endometrial secretory PP14/i3-lactoglobulin homologue. Hum Reprod 2:453, Suikkari A-M, Rutanen E-M, Seppala M: Circulating levels of immunoreactive insulin-like growth factor-binding protein in non-pregnant women. Hum Reprod 12:297, Stenman V-H, Alfthan H, Koskimies A, Seppiila M, Pettersson K, LOvgren T: Monitoring the LH surge by uitrarapid and highly sensitive immunofluorometric assay. Ann NY Acad Sci 442:544, Hernandez ER, Resnick CE, Svoboda ME, Van Wyk JJ, Payne DW, Adashi EY: Somatomedin-C/insulin-like growth factor I as an enhancer of androgen biosynthesis by cultured rat ovarian cells. Endocrinology 122:1603, Suikkari A-M, Koivisto VA, Koistinen R, Seppala M, Yki Jarvinen H: Dose-response characteristics for suppression of low molecular weight plasma insulin-like growth factorbinding protein by insulin. J Clin Endocrinol Metab 68:135, Pekonen F, Laatikainen T, Buyalos R, Rutanen E-M: Decreased 34K insulin-like growth factor binding protein in polycystic ovarian disease. Fertil Steril51:972, Geffner ME, Kaplan SA, Bersch N, Golde DW, Landaw EM, Chang RJ: Persistence of insulin resistance in polycystic ovarian disease after inhibition of ovarian steroid secretion. Fertil Steril45:327, Dunaif A, Green G, Futterweit W, Dobrjansky A: Suppression of hyperandrogenism does not improve peripheral or hepatic insulin resistance in the polycystic ovary syndrome. J Clin Endocrinol Metab 70:699, 1990 Suikkari et al. Effect of OC on serum IGFBP-l in PCOD 899