Insulin-like growth factors and their binding proteins in the venous effluents of ovary and adrenal gland in severely hyperandrogenic women.

Human Reproduction Insulin-like growth factors and their binding proteins in the venous effluents of ovary and adrenal gland in severely hyperandrogenic women Hannu Martikainen 0 Pasi Salmela Sinikka Nuojua-Huttunen 0 Jukka Per al a Sami Leinonen Mikael Knip Aimo Ruokonen 1 0 Department of Obstetrics and Gynecology 1 Clinical Chemistry, University Central Hospital of Oulu , Kajaanintie, 90220 Oulu , Finland 6To whom correspondence should be addressed - Insulin and insulin-like growth factors (IGF) are thought to play an important role in the pathogenesis of excessive androgen production. To explore this question further we measured the concentrations of IGF-I and -II and their binding proteins (IGFBP-1 and-3) in adrenal and ovarian vein samples of severely hyperandrogenic women (serum testosterone > 5 nmol/l) collected as part of their diagnostic work-up. The concentration of IGF-II was slightly but not significantly higher in the ovarian vein than in the adrenal and peripheral veins. The concentrations of IGF-I and IGFBP were identical in both the adrenal and ovarian veins and did not differ from those in the peripheral circulation. The concentration of IGFBP-1 was negatively correlated (rJ 0.60, P > 0.05) with insulin and IGFBP-3 showed a strong positive correlation with IGF-1 (r J 0.90, P > 0.01). These results indicate that neither the ovary nor the adrenal gland contributes significantly to the circulating pool of IGF or their binding proteins in severely hyperandrogenic subjects. Hyperinsulinaemia is associated with low circulating IGFBP-1 concentrations and IGFBP-3 seems to be an excellent indicator of the peripheral IGFI concentration. The concentrations of IGF-I suggested decreased somatotrophic activity in these obese, hyperinsulinaemic subjects. Key words: adrenal/hirsutism/IGF-system/insulin/ovary Hyperandrogenism and the polycystic ovary syndrome (PCOS) are often associated with hyperinsulinism. The precise pathophysiological mechanism(s) underlying these common endocrine disturbances have not been characterized in detail as yet but it is known that insulin is capable of modifying the androgen bioactivity by several mechanisms. Firstly, it may increase ovarian androgen synthesis, possibly synergistically with luteinizing hormone (LH) (Poretsky and Piper, 1994), or independently by activating insulin and insulin-like growth factor (IGF)-I receptors in theca cells (Bergh et al., 1993). Secondly, it can increase the biological activity of androgens by reducing sex hormone-binding globulin production in the liver (Plymate et al., 1988). IGFs in the circulation are bound to binding proteins (IGFBP), which are produced in the liver and also locally, like IGFs, in the ovary, and thereby function as autocrine and paracrine factors regulating folliculo- and steroidogenesis (El-Roeiy et al., 1994; Magoffin et al., 1995; Mason et al., 1996; Voutilainen et al., 1996). Insulin is known to decrease IGFBP-1 production and hence it may also modify these local regulatory systems in the ovary (Suikkari et al., 1988). Although the liver seems to be the main source of serum IGFBP-1, a previous study of ours showed increased serum concentrations of IGFBP-1 during ovarian stimulation for invitro fertilization (IVF) (Martikainen et al., 1991), suggesting an ovarian contribution to this binding protein in the circulation. In this study, we wanted to investigate further the role of the IGF-system in hyperandrogenic women by measuring the concentrations of IGF-I and -II and IGFBP-1 and -3 in the adrenal and ovarian veins. This study protocol allowed the evaluation of the adrenal and ovarian secretion of the factors into the circulation. Materials and methods A total of 13 consecutive women attending our unit with a serum testosterone concentration . 5 nmol/l (in at least one of the three to five measurements taken) participated in this study. The clinical and hormonal characteristics of the women have been previously reported in detail (Martikainen et al., 1996). The mean (6 SD) age was 30.4 (6 9.9) years and 12 patients were overweight with a body mass index (BMI) . 25. BMI indicated massive obesity in three cases (. 38). In all, 12 women (92%) were hirsute and had a Ferriman Gallwey score higher than 7. All the patients had cycle abnormalities, one of them having amenorrhoea and six oligomenorrhoea (46%). No sign of enzyme deficiencies or androgen secreting tumours was observed. The study protocol was approved by the Ethics Committee of the Medical Faculty of the University of Oulu, Oulu, Finland and informed consent was obtained from all the study subjects. Selective catheterization of the adrenal and ovarian veins The catheterization procedure, characterized in detail previously (Martikainen et al., 1996), was carried out between 0800 and 1100 h after an overnight fast (on the second day of hospitalization) during the cycle days 18 (if the patient menstruated). The percutaneous, transfemoral catheterization was carried out under local anaesthesia. The tip of a pre-curved 7.0 F catheter with a side hole 12 mm from the tip was guided into the left renal vein under fluoroscopic control. In order to clarify the anatomy, a renal European Society for Human Reproduction and Embryology 21.09610.2 16.5969.13 (14.24227.95) (10.46222.72) 147.2 640.3 148.5 633.2 (122.1 2173.4) (126.1 2169.3) 134.0 6139.8 166.7 6162.6 (34.0 2234.0) (68.4 2265.0) 0.6760.21 0.7360.15 (0.5220.82) (0.6120.84) 3.4960.81 3.3060.72 (3.0023.98) (2.8223.78) venography was performed. The tip of the catheter was introduced into the left ovarian vein and two samples of 10 ml each were slowly aspirated (at least 5 min/sample). This sampling time was optimized for reliable results in one patient prior to the study. Thereafter the catheter was changed to a Simmons II-type catheter, with one side hole 12 mm from the tip, which was introduced into the left adrenal vein and samples of 10 ml each were again slowly aspirated to obtain a pure effluent sample and avoid an admixing of peripheral blood. Subsequent to this aspiration, a small amount of non-ionic contrast medium (~0.5 ml for the suprarenal vein and 25 ml for the ovarian vein) was injected under fluoroscopy control to confirm the right position of the catheter. Prior to the completion of this procedure, two peripheral blood samples (each 10 ml) were aspirated from the lower part of the inferior vena cava. The venous sampling was successful in 10 out of the 13 cases. Two patients experienced minor discomfort during the procedure, but no major complications occurred. Measurement of insulin, insulin-like growth factors and binding proteins The concentrations of insulin, IGF and IGFBP were measured by commercial assays according the instructions provided by the manufacturers. IGFBP-1 concentrations were measured by immunoradiometric assay (Medix, Biochemica, Kauniainen, Finland), IGFBP3 by radioimmunoassay (Diagnostic Systems Laboratories, Webster, Tx, USA), I (...truncated)