Evidence that an altered prolactin release is consequent to abnormal ovarian activity in polycystic ovary syndrome

---~-------------------~--------~-

FERTILITY AND STERILITY Copyright

Vol. 64, No.6, December 1995

Printed on acid-free paper in U. S. A.

1995 American Society for Reproductive Medicine

Evidence that an altered prolactin release is consequent to abnormal ovarian activity in polycystic ovary syndrome

Anna Maria Paoletti, M.D.* Angelo Cagnacci, M.D. * Renza Soldani, M.D. * Marisa Orru, Ph.D. *

Silvia Ajossa, M.D. * Giovanna Pittorra, M.D. * Patrizio Mulas, M.D.t Gian Benedetto Melis, M.D. *:j:

Uniuersita degli Studi di Cagliari, Cagliari, Italy

Objective: To investigate whether endogenous dopaminergic activity is impaired in polycystic ovary syndrome (PCOS)-a£fected women and is normalized by medical ovariectomy. Patients: Women with PCOS untreated (n = 23) and treated for 3 months with GnRH analogue (GnRH-a) administration (n = 10) and normal cycling young women (n = 23) as controls. Interventions: Acute blockade of dopaminergic receptors by the IV administration of 5 mg of the dopaminergic receptor blocking agent sulpiride (sulpiride test) was performed 3 to 7 days after the initiation of spontaneous menses in cycling women or medroxyprogesterone acetateinduced menses in PCOS women. In PCOS women treated with GnRH-a administration (go serelin depot, 3.6 mg SC every 28 days), the sulpiride test was repeated 10 to 15 days after the third GnRH-a administration. Main Outcome Measure: Basal PRL levels and PRL increase induced by sulpiride. Results: Basal PRL levels and the PRL response to sulpiride were increased in women with PCOS. In women with PCOS medical ovariectomy induced by GnRH-a administration reversed to normal both basal and sulpiride-stimulated PRL levels. Conclusions: In women with PCOS the abnormal regulation of PRL and presumably of hypothalamic neurotransmitters controlling PRL secretion is not a primary alteration but it is Fertil Steril 1995;64:1094-8 likely dependent on abnormal ovarian functionality. Key Words: PCOS, sulpiride, dopamine, PRL

Polycystic ovary syndrome (peaS) is a complex reproductive disorder (1, 2) in which alterations of hypothalamic neurotransmitters are considered to increase the release of GnRH-LH and by inducing an altered stimulus on ovarian granulosa cells to lead to an ovarian androgen overproduction (3, 4). Studies on neurotransmitter alterations of women with peas mainly have focused on opioids and dopamine whose deficient inhibitory activity on GnRHLH may have a major role in inducing the endocrine

Received January 16, 1995; revised and accepted June 1, 1995. * Istituto di Ginecologia Ostetricia e Fisiopatologia della Riproduzione Umana. t Istituto di Clinica Dermosifilopatica. :j: Reprint requests: Gian Benedetto Melis, M.D., Istituto di Ginecologia Ostetricia e Fisiopatologia della Riproduzione Umana, Universita degli Studi di Cagliari, Via Ospedale 46, 09124 Cagliari, Italy (FAX: 39-70-668575). 1094

Paoletti et aI. Hyperprolactinemia in pcas

abnormalities of peas. Although data on opioid activity alteration are conflicting (3-6), data showing an abnormal activity of the endogenous dopaminergic system are more consistent (3, 7, 8). Peripheral indexes of central dopamine turnover are reduced in peas (9) and PRL, whose secretion is controlled strictly by an inhibitory dopaminergic effect, has been reported to be elevated in peas (3, 7, 8). Whether the supposed abnormal dopaminergic activity of women with peas represents a primary defect involved in the pathogenetic development of peas or it is an epiphenomenon of peas endocrine abnormalities still is unclear. Indeed, endogenous dopaminergic activity is influenced markedly by PRL that through a short-loop feed-back enhances hypothalamic dopamine turnover (10, 11), as well as by gonadal steroids. In particular, estrogens may enhance the dopaminergic activity by stimulating PRL release (12) and may inhibit the dopaminergic Fertility and Sterility

j

'1

Table 1 Clinical Data of 23 PCaS-affected Women (PCaS) and 23 Control Women (Controls) Menstrual cyelet

Controls

pcas

* Values are means

Age*

Body mass index*

y

kg/m2

23.7::+: 1.16 23.6 ::+: 1.72

22.77::+: 0.38 22.10 ::+: 0.29

:t::

SEM.

No. with eumenorrhea

No. with oligomenorrhea

No. with amenorrhea

23 (100) 0(0)

0(0) 13 (56.52)

0(0) 10 (43.48)

t Values in parentheses are percentages.

activity by a direct action on hypothalamic dopaminergic neurons (13). The response ofPRL to thyrotropin-releasing hormone or specific dopamine receptor blocking agents has been considered an indirect way to test hypothalamic dopaminergic activity (14-16). In the present study the PRL response to the administration of the antidopaminergic agent sulpiride (17) was used as a tool to confirm abnormal dopaminergic activity in the hypothalamus of women with PCOS. Furthermore, its eventual dependence on an abnormal ovarian steroidogenetic activity also was evaluated by reducing ovarian steroidogenesis with the prolonged administration of a GnRH analogue (GnRH-a). MATERIALS AND METHODS

Twenty-three women with regular menstrual cycle and 23 women with clinical, endocrine, and ultrasound signs of PCOS gave their consent to participate in the study previously approved by the local Ethical Committee. None of the women referred to the study had been pregnant, and none of them wished to become pregnant at the time of the study. Mean age and body mass index were similar among the two groups, whereas menstrual cycle differed between the two groups (Table 1). In all enrolled women basal levels ofPRL evaluated in four samples collected at 20-minute intervals starting 2 hours after venipuncture were <20 ng/mL (conversion factor to SI unit, 1.0). The PRL response to the IV administration of 5 mg of the dopaminergic receptor blocking agent sulp-

iride (Dobren; Ravizza Farmaceutici SpA, Muggio, Milan, Italy) (sulpiride test) was evaluated during the early follicular phase (days 3 to 4) of the menstrual cycle in regular cycling and within 7 days from the first day of medroxyprogesterone acetate-induced menses in women with PCOS. Thereafter, all subjects were asked to give their consent to receive every 28 days for 6 cycles an SC implant of goserelin depot (3.6 mg Zoladex; Zeneca, Milan, Italy). Only 10 of them gave their consent. In these women the sulpiride test was repeated 10 to 15 days after the third goserelin depot implant. On each testing day, women were admitted at 7 :30 A.M. A polyethylene catheter was inserted immediately in an antecubital vein of the arm and was kept open by a slow infusion of saline solution. During tests, all women had bed rest and were not allowed to eat, sleep, or smoke during the sampling period. At times -30, 0, and 15, 30, 60, 90, and 120 minutes after the sulpiride administration, starting at 9:00 A.M., blood samples were withdrawn through the catheter and collected into heparinized tubes. Blood samples were centrifuged immediately and plasma samples were frozen at - 20°C until assayed. Prolactin concentrations were evaluated in all plasma samples, whereas LH, FSH, E 2 , androstenedione (A), total T, free T, and DHEAS were evaluated only in samples collected before sulpiride administration. Prolactin, LH, and FSH were measured by immunoradiometric assay methods (PRL: Ares Serono, Rome, Italy; LH and FSH: Sorin Biomedica, Saluggia Vercelli, Italy). Estradiol, total T, free T, A, and

Table 2 Specific Performance Characteristics ofLH, FSH, PRL, E 2, total T, free T, DHEAS, and A Assays*

Intra-assay coefficient of variationt Between-assay coefficient of variationt Sensitivity

LH

FSH

PRL

E2

Total T

Free T

DHEAS

A

miU/mL

miU/mL

ng/mL

pg/mL

ng/mL

pg/mL

ng/mL

ng/mL

2.8 3.4 0.25

3.8 3.3 0.15

1.4 3.7 0.3

4.0 3.9 8

9.5 11.63 0.1

3.2 3.7 0.15

6.0 8.5 0.00002

6.7 11.1 0.1

* Conversion factors to SI units: LH, FSH, and PRL, 1.0; E 2, 3.671; T, 3.467; DHEAS, 0.002714; A, 3.412. Vol. 64, No.6, December 1995

t Values are percentages.

Paoletti et al. Hyperprolactinemia in

peas

1095

Table 3 Basal Values ofLH, FSH, PRL, E 2, Total T, Free T, DHEAS and A in Normal Women (Controls) and in PCDS-affected Women*

Controls (n = 23) PCDS (n = 23)

LH

FSH

PRL

E2

Total T

Free T

DHEAS

A

mIU/mL

mIU/mL

ng/mL

pg/mL

ng/mL

pg/mL

ng/mL

ng/mL

5.5::': 0.94

6.1 ::': 0.45

9.6::': 1.0

46.15::': 5.2

0.387 ::': 0.022

11.0::': 0.63§

5.0::': 0.45

15.2 ::': 1.0§

77.2::': 15.2t

0.633 ::': 0.054t

* Values are means::': SEM. Conversion factors to SI units: LH, FSH, and PRL, 1.0; E 2, 3.671; T, 3.467; DHEAS, 0.002714; A, 3.412.

DHEAS were measured by available RIA kits (E 2: Medical System, Genova, Italy; total T: ICN Biomedicals, Inc. Diagnostic Division, Costa Mesa, CA; free T: Diagnostic Product Corporation, Los Angeles, CA; A: Incstar Corporation, Stillwater, MN; DHEAS: Medical System). Intra-assay and interassay coefficients of variation, such as sensitivity and conversion factor for each assay are reported in Table 2. To avoid interassay variability all the samples of pcas women obtained prior and during the administration of the GnRH-a were run in the same assay. The PRL response to sulpiride was evaluated both as absolute values and as integrated area under the curve calculated by the method of triangulation and expressed as nglmL X minutes. Statistical analysis was performed by t-test for paired data or by oneway analysis of variance followed, when significant, by the Scheffe's post hoc test.

3.36::': 0.57 7.2::': 0.89*

2,000::': 150

1.5 ::': 0.089

2,300::': 200

3.4 ::': 0.25*

t p < 0.05. *P < 0.01. §P < 0.0001.

in pcas than in regular women as maximal PRL increase (205 ± 16 versus 113 ± 8.6 ng/mL; P < 0.0001), maximal percent PRL increase (17.9%

240

.*

200

--5, -en

+

§

..J

c

160

..J

W

RESULTS Prolactin and E2 basal levels were significantly higher in women with pcas (15.2 ± 1.0 nglmL; P < 0.0001 and 77.2 ± 15.2 pglmL; P < 0.05, respectively) than in control women (9.6 ± 1.0 ng/mL and 46.15 ± 5.2 pglmL, respectively) (Table 3). Moreover, in comparison to normal higher levels of LH, total T, free T, and A were observed in PCaS-affected women, whereas DHEAS and FSH values did not differ between two groups (Table 3). The administration of sulpiride induced a prompt and significant PRL increase in all women. In regular women, from basal levels of9.6 ± 1 ng/mL PRL increased to maximal values of 113 ± 8.6 ng/mL within 30 minutes, and then decreased progressively to values of 61.0 ± 5.4 ng/mL at the end of the observation (Fig. 1). In pcas women, PRL promptly increased from basal levels of 15.2 ± 1 ng/mL to maximal levels of 205 ± 16 ng/mL within 15 minutes (Fig. 1) and then decreased progressively, resulting in 109 ± 7.9 ngl mL at the end of the observation (Fig. 1). The statistical analysis between the two groups showed that PRL increase after sulpiride was significantly higher 1096

Paoletti et al. Hyperprolactinemia in

peas

> w 120 ..J
::i

en

c(

..J

D.

80

..J

a:

D.

40

o

-30

6 3"0 do dO MINUTES

1'20 MINUTES

Figure 1 Prolactin response to sulpiride stimulation test in normally menstruating women (n = 23,0) (left); in PCDS-affected women (n = 23, .) (left); in PCDS-affected women before (n = 10; .) (right) and during goserelin depot treatment (n = 10, D) (right). The arrows indicate the IV administration of 5 mg of sulpiride. The values are expressed as means::': SEM. Conversion factor to SI unit, 1.0. *p < 0.0001 versus normally menstruating women; §P < 0.01 versus GnRH-a-treated PCDS women.

Fertility and Sterility

------~

t

§

24000

crease of PRL after 30 minutes from sulpiride (198 : :': : 18 versus 126 : :': : 24 ng/mL; P < 0.01) (Fig. 1), as maximal percent PRL increase (17.6% : :': : 3.1 % versus 10.5% : :': : 1.7%; P < 0.05) and as area under the curve (21,811 : :': : 1,815 versus 14,037 : :': : 1,960 ng/mL X 120 minutes; P < 0.05) (Fig. 2) .

z

2w

tis 20000

a:::::1 oC

WE

Ul o ..IN

a:: ... a. >< 11...1

16000

DISCUSSION

°e wCj,

~.s 12000 ~w

00

w-

::t:!!:

1-0.

8000

a::..1

w~

OUI

za:: ~w

.di:

4000

~cC cC

0

Figure 2 Area under the curves of the PRL response to the IV administration of 5 mg of sulpiride in normal women (D) (left), in PeOS-affected women (.) (left), and in 10 peOS-affected women before (.) (right) and during GnRH-a treatment (~) (right). The values are expressed as the mean ofPRL secretion X 120 minutes ± SEM. §P < 0.01 versus normal menstruating women; tP < 0.05 versus GnRH-a-treated peos women.

: :': : 1.1% versus 13.2% : :': : 1.3%; P < 0.05) and as area under the curve (21,694 : :': : 1,699 versus 11,818 : :': : 873 ng/mL X 120 minutes; P < 0.01) (Fig. 2). In the 10 PCOS-affected women who received GnRH-a treatment, a significant decrease of the values of LH, FSH, PRL, E 2 , total T, free T, and A was observed after 3 months of GnRH-a treatment, and the resulting values did not differ from those of the control women with regular menstrual cycle (Table 4). The administration of sulpiride induced within 30 minutes a prompt and significant increase ofPRL secretion, from basal values of 8.5 : :': : 1.5 ng/mL to maximal values of 126 : :': : 24 ng/mL (Fig. 1). Thereafter, PRL decreased to reach values of 66 : :': : 13.8 ng/ mL at the end of observation (Fig. 1). The area under the PRL curve was 14,037 : :': : 1,960 ng/mL X 120 minutes (Fig. 2). In women with PCOS the PRL response to sulpiride observed during GnRH-a administration was significantly lower than prior GnRHa treatment when evaluated both as maximal in-

Present data agree with previous studies demonstrating an increased PRL secretion in PCOS-affected women (3, 7, 8). Both basal and sulpiridestimulated PRL release were increased in PCOS, suggesting an excessive stimulus or a deficiency of the mechanisms controlling PRL secretion. Indeed, some authors believe that in PCOS-affected women hyperprolactinemia derives from an altered dopaminergic activity (18-20), but whether this defect is primary or secondary to an abnormal gonadal steroid environment never has been investigated. Because the amount of PRL release evoked by dopaminergic receptor blocking agent is dependent on the activity of central dopamine activity (16, 17), the enhanced PRL response to the administration of sulpiride in PCOS may be associated with an altered dopaminergic activity. Several observations have indicated that an increased estrogen stimulus may enhance PRL secretion by a direct stimulation oflactotroph cells (12) and by a direct inhibition on central dopaminergic neurons (13). Although we cannot exclude that desensitization of GnRH receptors after GnRH-a treatment might reverse the hyperprolactinemic effect of endogenous GnRH at pituitary level (21), the hypothesis that impaired dopaminergic system in PC OS can be secondary to an abnormal ovarian steroid secretion (22) is suggested strongly by present data showing that medical ovariectomy dependent on desensitization of GnRH-a at pituitary or possibly at the ovary (23) normalizes both basal and sulpiride-stimulated PRL secretion. In our opinion, the present findings clarifying that hyperprolactinemia of PC OS is consequent to an altered ovarian steroidogenesis furnish novel insights on the role that the abnormal ovarian steroidogenesis of PCOS exerts in inducing a secondary neurotransmitter un-

Table 4 Basal Values ofLH, FSH, PRL, E 2 , Total T, Free T, DHEAS and A in peOS-affected Women* Before and During Goserelin Depot Treatmentt LH

PC OS before GnRH-a PC OS during GnRH-a

FSH

PRL

Total T

Free T

A

DHEAS

mlU/mL

mlU/mL

ng/mL

pg/mL

ng/mL

pg/mL

ng/mL

ng/mL

12.2 :t 0.94 0.957:t 0.113§

5.9:t 0.55 1.8 :t 0.3§

14.5 :t 1.6 9.5 :t 1.5§

74.8 :t 9.5 36.8 :t 7.6:1:

0.710:t 0.120 0.354 :t 0.042:1:

7.95 :t 0.99 4.3:t 0.89:1:

2,300:t 160 2,100 :t 130

3.7 :t 0.092 1,036 :t 0.08lt

* n = 10.

t Values are means :t SEM. Conversion factors to SI units: LH, FSH,

:l:P < 0.05. §P < 0.0001.

and PRL, 1.0; E z, 3.671; T, 3.467; DHEAS, 0.002714; A, 3.412.

Vol. 64, No.6, December 1995

Paoletti et al. Hyperprolactinemia in peDS

1097

balance. Recent studies from our laboratory have not supported the hypothesis of an opioid system alteration in PC OS-women (6), whereas present data confirm the abnormal activity of the endogenous dopaminergic system and its dependence on ovarian abnormalities. As suggested by several studies (24, 25), disorders in ovarian steroidogenesis probably are crucial factors in the etiopathogenesis of PCOS, but the consequent alterations of neurotransmitter function can sustain the vicious circle of chronic anovulation by inducing an abnormal gonadotropin regulation and stimulus to ovarian steroidogenesis.

13.

Acknowledgments. We acknowledge Ms. Franca Fadda and Ms. Antonella Piselli for their collaboration in the typing and revision of the manuscript, respectively.

16.

12.

14.

15.

REFERENCES 1. Insler V, Lunenfeld B. Polycystic ovarian disease: a challenge and controversy. Gynecol Endocrinol 1990;5:51-69. 2. Yen SSC. Chronic anovulation caused by peripheral endocrine disorders. In: Yen SSC, Jaffe RB, editors. Reproductive endocrinology. Philadelphia: Saunders, 1991:576-630. 3. Barnes RB, Lobo RA. Central opioid activity in polycystic ovary syndrome with and without dopaminergic modulation. J Clin Endocrinol Metab 1985;61:779-82. 4. Cumming DC, Reid RL, Quigley ME, Rebar RW, Yen SSC. Influences on LH secretion in polycystic ovary syndrome. Clin Endocrinol (Oxf) 1984;20:643-8. 5. Armeanu MC, Berkhout GMJ, Schoemaker J. Pulsatile luteinizing hormone secretion in hypothalamic amenorrhea, anorexia nervosa and polycystic ovarian disease during naltrexone treatment. Fertil Steril 1992;57:762-70. 6. Cagnacci A, Soldani R, Paoletti AM, Falqui A, Melis GB. Prolonged opioid blockade with naltrexone and luteinizing hormone modifications in women with polycystic ovarian syndrome. Fertil Steril 1994; 62:269-72. 7. Luciano AA, Chapler FK, Sherman BM. Hyperprolactinemia in polycystic ovary syndrome. Fertil Steril 1984;41:719-25. 8. Carmina E, Rosato F, Maggiore M, Gagliano AM, Indovina D, Janni A. Prolactin secretion in polycystic ovary syndrome (PCO): correlation with the steroid pattern. Acta Endocrinol (Copenh) 1984; 105:99-104. 9. Paradisi R, Grossi G, Venturoli S, Capelli M, Porcu E, Fabbri R, et al. Evidence for an hypothalamic alteration of catecholamine metabolism in polycystic ovary syndrome. Clin Endocrinol (Oxf) 1988;29:317-26. 10. Barton AC, Lahti RA, Piercy MF, Moore KE. Autoradiographic identification of prolactin binding sites in rat median eminence. Neuroendocrinology 1989;49:649-53. 11. Gudelsky GA, Porter JC. Release of dopamine from tubero-

1098

Paoletti et al. Hyperprolactinemia in peDS

17.

18.

19.

20.

21.

22.

23.

24.

25.

infundibolar neurons into pituitary stalk blood after prolactin and haloperidol administration. Endocrinology 1980; 106: 526-29. Strangers SA, Brunner RM, Bethea CL. Estrogen and progestin receptor immunocytochemistry in lactotrophes versus gonadotropes of monkey pituitary cell cultures. Endocrinology 1989; 124:1462-70. Rasmussen DD. The interaction between mediobasohypothalamic dopaminergic and endorphinergic neuronal systems as a key regulation of reproduction: an hypothesis. J Endocrinol Invest 1991; 14:323-52. Pourmand M, Rodriguez-Arnan MD, Weightman DR, Hall R, Cook DB, Scanlon MF. Domperidone: a novel agent for the investigation of anterior pituitary function and control in man. Clin Endocrinol (Dxf) 1980; 12:211-5. Niboe Anderson A, Tabor A. PRL, TSH, GH and LH responses to metoclopramide and breast-feeding in normal and hyperprolactinemic women. Acta Endocrinol (Copenh) 1982; 100:177 -83. Caviezel F, De Pasqua A, Pozza G. Studies on prolactin secretion by sulpiride in man. In: Spano PF, Trabucchi M, Corsini GU, Gessa GL, editors. Sulpiride and other benzamides. Milan: Italian Brain Research Foundation Press, 1979:215-20. Tissani AH, Stefanini E, Gessa GL. Stimulation of dopamine synthesis and activation of tyrosine hydroxylase by (- )sulpiride, sultopride and tiapride. In: Spano PF, Trabucchi M, Corsini GU, Gessa GL, editors. Sulpiride and other benzamides. Milan: Italian Brain Research, 1979:3-9. Alger M, Vasquez-Matute L, Mason M, Canales ES, Zarate A. Polycystic ovarian disease associated with hyperprolactinemia and defective metoclopramide response. Fertil Steril 1980; 34:70-1. Velardo A, Pantaleoni M, Zironi C, Zizzo G, Marrame P. Evidence of altered dopaminergic modulation of prolactin and thyrotropin secretion in patients with polycystic ovary syndrome. Horm Res 1991;35:4-7. Prelevic GM, Wurzburger MI, Peric LJA. Metoclopramide effects on serum prolactin, LH and FSH in patients with polycystic ovary syndrome. J Endocrinol 1988; 11:255-9. Mais V, Melis GB, Paoletti AM, Strigini F, Antinori D, Fioretti P. Prolactin-releasing action of a low-dose of exogenous gonadotropin-releasing hormone throughout the human menstrual cycle. Neuroendocrinology 1986;44:326-9. Barnes RB, Mileikowsky GN, Cha KY, Spencer CA, Lobo RA. Effects of dopamine and metoclopramide in polycystic ovary syndrome. J Clin Endocrinol Metab 1986;63:506-9. Minaretzis D, Jakubowski M, Mortola JF, Pavlou SN. Gonadotropin-releasing hormone receptor gene expression in human ovary and granulosa-lutein cells. J Clin Endocrinol Metab 1995;80:430-4. Pache TD, de Jong FH, Hop WC, Fauser BCJM. Association between ovarian changes assessed by transvaginal sonography and clinical and endocrine signs of the polycystic ovary syndrome. Fertil Steril 1993;59:544-9. Erickson GF, Magoffin DA, Garzo VG, Cheung AP, Chang RJ. Granulosa cells of polycystic ovaries: are they normal or abnormal? Hum Reprod 1992;7:293-9.

Fertility and Sterility