Enhances the Luteinizing. Hormone and. Follicle-Stimulating. Hormone Responses to. Gonadotropin-Releasing. Hormone in the Follicular, but. Not in the Luteal ...
Vol. 80, No. 4 Prmted m U.S.A.
OOZl-972w95/503.00/0 Journal of Clinical Endocrinology and Metabolism Copyright 0 1995 by The Endocrine Society
Melatonin Enhances the Luteinizing Hormone and Follicle-Stimulating Hormone Responses to Gonadotropin-Releasing Hormone in the Follicular, Not in the Luteal, Menstrual Phase ANGELO CAGNACCI, ANNA MARIA PAOLETTI, ELISABETTA MASCHIO, AND GIAN BENEDETTO Department
of Obstetrics
and Gynecology,
University
RENZA MELIS
of Cagliari,
ABSTRACT
T HAS BEEN widely reported that season-relatedphotoperiodic variations modify the nocturnal secretion of the pineal hormone melatonin and through this mechanism synchronize reproduction of seasonal breeders (l-4). By contrast, the influence of melatonin in reproduction of nonseasonalbreeders is unclear (5). Inhibitory influences of melatonin in human reproduction have been suggested by indirect evidence showing an association among defective ovulation or anovulation and an elevation in the circulating levels of melatonin (6-8). However, inconclusive results were obtained by studies evaluating the effect of exogenous melatonin on reproductive parameters and, in particular, on gonadotropin secretion (9-13). Despite evidence of a melatonin inhibitory effect on LH secretion in three postmenopausal women (10) and no effect in men (7), recent placebo-controlled experiments have shown that exogenous melatonin enhances the amplitude of spontaneous LH peaks and mean LH levels in women during the follicular (13), but not the luteal, phase of the menstrual cycle (14). Possible FSH modifications or variations of the pituitary response to GnRH were not investigated. The present experiments were performed to analyze the possibility that melatonin modifies the pituitary response of both LH and FSH to GnRH, and that this modification is modulated by the Received October 7, 1994. Revision received November 8, 1994. Accepted December 2, 1994. Address all correspondence and requests for reprints to: Angelo Cagnacci, M.D., Istituto di Ginecologia Ostetricia e Fisiopatologia della Riproduzione Umana, via Ospedale 46, 09124 Cagliari, Italy.
09124
Cagliari,
MARISA
ORRtJ,
Italy
administration of three submaximal doses of GnRH (1 pg at 0900 h, 5 Kg at 1100 h, and 10 pg at 1300 h). In the follicular phase, melatonin administration enhanced the LH and FSH responses to all three GnRH stimuli, whereas in the luteal phase, melatonin administration was ineffective. The present data indicate that an enhancing effect of melatonin on the LH and FSH responses to submaximal GnRH stimuli is evident in the follicular, but not the luteal, phase of the menstrual cycle and infer an endocrine window for the effect of melatonin on gonadotropin secretion. (J Clin Endocrinol Metab 80: 1095-1099, 1995)
Exogenous melatonin enhances LH pulse amplitude and mean LH levels in women during the follicular, but not the luteal, menstrual phase. In this study we investigated whether an increased pituitary response to GnRH is involved in the stimulatory effect of melatonin. Eight normal cycling women were studied on 2 consecutive days during the follicular stage (days 4-6), and eight were studied during the luteal phase (days 19-21) ofthe menstrual cycle. On 2 consecutive days, each woman received, randomly and in a double blind fashion, placebo or 3 mg melatonin (1 mg at 0800, 1000, and 1200 h), whereas the pituitary LH and FSH responses to GnRH were tested by the iv
I
SOLDANI,
but
different endocrine environments of the two menstrual phases.
Materials
and
Methods
The study, approved by the local ethical committee, was conducted in 16 regular cycling women, 25-35 yr of age, of whom 8 were studied in the follicular (days 4-6) and 8 in the luteal (days 18-21) phase of the menstrual cycle. Each woman was entrained to a normal light-dark cycle, was within 10% of her ideal body weight, and had been free from medications for at least 3 months. On 2 consecutive days, each woman received in a random and double blind fashion placebo or 3 mg melatonin in three divided doses (1 mg at 0800, 1000, and 1200 h). The time of administration and doses of melatonin were chosen on the basis of previous experiments, which showed that a similar melatonin administration keeps the circulating levels of the hormone in the pharmacological range for at least 8 h (13,151. On both days, submaximal GnRH stimuli were given at 0900 h (1 pg), 1100 h (5 fig), and 1300 h (10 pg). Blood samples (2.5 mL) were collected 30 and 0 before and 15,30,60,90, and 120 min after each GnRH stimulus. Plasma levels of melatonin, expressed as picomoles per L, were measured by a radioimmunological method (15) in all samples. The intraand interassay coefficients of variation (CVs) were 5% and 8%, respectively, and the sensitivity was 28 pmol/L (6.5 pg/mL). Plasma LH and FSH levels were measured in all samples by highly sensitive timeresolved immunofluorometric methods, using available kits (Delfia, Pharmacia-Wallac, Turku, Finland). The assay has a sensitivity of 0.05 IU/L for both LH and FSH, a mean intraassay CV of 5% for LH and 4% for FSH, and a mean interassay CV of 6% for LH and 5% for FSH. Estradiol (E,), progesterone (I’), testosterone (T), androstenedione (A), and PRL levels were measured in samples collected at 0900 h, using available radioimmunological kits. Circulating levels of LH, FSH, and melatonin were analyzed as absolute values and as integrated values [areas under the curve (AUC)], calculated by the method of triangulation, and expressed as international units per L or picomoles per L/min. Statistical analysis of the
1095
ET AL.
CAGNACCI TABLE
1. Mean (*SE) cycle during
menstrual
plasma placebo
levels of E,, P, PRL, T, and A of women (PL) or melatonin (MEL) administration
during
JCE C M . 1995 I’0180
the follicular
phase
(n = 8) and luteal
Follicular
102.0 1.5 2.0 2.3 8.17
(pmolh)
P (nmoVL) T (nmol/L) A (nmol/L) PRL (/&J a P < 0.01 US. corresponding ’ P < 0.05 vs. corresponding results was Mann-Whitney
ML
-t -c ? 2 ?
follicular follicular
21.1 0.4 0.01 0.1 1.6
phase phase
performed by the Wilcoxon LJ test for paired or unpaired
(n = 8) of the
Luteal
PL E,
phase
103.1 1.7 1.8 2.2 8.1
2 ? 2 ? ?
PL
19.2 0.1 0.02 0.2 0.1
400.3 55.1 2.4 3.1 11.9
2 * + i t
ML
55.5” 15.4” 0.01 0.1 1.2b
397.2 46.0 2.2 2.5 12.4
k I! 2 k 2
43.0” 11.6” 0.03 0.1 1.0’
values. values.
signed rank test data, respectively.
or
phase of the menstrual cycle (221,600 2 58,500 VS.233,900 5 65,600 AUC). In the follicular phase, melatonin administration was associated with increased LH and FSH responsesto GnRH. Peak LH and FSH levels after each GnRH stimulus were significantly higher during melatonin than placebo administration (Fig. 1). The sameresults were obtained when data were expressed as areas under the LH or FSH peaks after each GnRH stimulus or as the overall gonadotropin AUCs (Table 2). Similarly, net area increases above baseline were higher during the administration of melatonin than placebo for both LH (2348.0 ? 376.4 IIS. 1569.0 + 174.4 AUC; P < 0.05) and FSH (780.2 ? 110.7 VS.561 ? 60.0 AUC;
the
Results
Compared to the follicular phase, the luteal menstrual phasewas characterized by increasedlevels of E,, I’, and PRL (Table l), an increased GnRH-stimulated LH AUC (P < O.Ol), and a decreasedGnRH-stimulated FSH AUC (P < 0.05;Table 2). However, when the gonadotropin responseto GnRH was evaluated as the net area increase above baseline area, both the LH (3818.9 ? 677.1I)S.1569.0+- 174.4AUC; P < 0.01) and the FSH (673.9 ? 59.5 DS. 561.0 + 60.0 AUC; P < 0.05) responsesto GnRH were significantly higher in the luteal phase. In both the follicular and luteal menstrual phases, plasma concentrations of E,, I’, T, A, and PRL during placebo treatment were similar to those during melatonin administration (Table 1). After the administration of placebo, circulating melatonin ranged below 25 pmol/L in both the follicular and luteal menstrual phases. By contrast, within 60 min after its administration, melatonin reached pharmacological levels during both the follicular (7,150 2 1,360 pmol/L) and luteal (6,820 * 1,600 pmol/L) menstrual phases, ranged around these values for all 6 h of the test, and at the end of the study was still in the pharmacological range, with values of 4,830 * 1,131 pmol/L in the follicular phase and 5,290 ? 985 pmol/L in the luteal phase. After melatonin administration, melatonin AUCs observed in the follicular phase were similar to those observed in the luteal
P < 0.02). In the luteal phase, melatonin administration was not associated with an increased gonadotropin responseto GnRH (Fig. 2). Peak gonadotropin levels after each GnRH stimulus were similar during melatonin and placebo administrations. The same results were obtained when data were expressed as areas under the LH and FSH peaks after each GnRH stimulus or as overall gonadotropin AUCs (Table 2). Similarly, net area increasesabove baseline were similar during the administrations of placebo and melatonin for both LH (3818.9? 677.1VS.4081.8 ? 707.3AUC) and FSH (673.9? 59.5 vs. 708.9 t 101.0 AUC). Discussion
This is the first study showing a stimulatory effect of melatonin on the gonadotropin response to physiological or
TABLE 2. Mean (?SE) integrated LH and FSH responses to each individual GnRH stimulus (1 pg = peak 1; 5 pg = peak 2; 10 pg = peak 3) and overall response to the three stimuli (total) of eight women studied in the follicular phase and eight women studied in the luteal phase of the menstrual cycle during the administration of placebo (pl) or melatonin (mel; 1 mg at 0800, 1000, and 1200 h) Peak
Follicular LH pl LH me1 FSH pl FSH me1 Luteal LH pl LH me1 FSH FSH aP bP ’ P dP
< < <
