Nocturnal melatonin levels are unaltered by ovarian suppression in girls with central precocious puberty*t*

Transcription

1 FERTILITY AND STERILITY Copyright <> 1989 The American Fertility Society Vol. 52, No.6, December 1989 Printed on add-free paper in U.S.A. Nocturnal melatonin levels are unaltered by ovarian suppression in girls with central precocious puberty*t* Sarah L. Berga, M.D. 1f11 Kenneth Lee Jones, M.D.# Silvia Kaufmann, M.D.# Samuel S. C. Yen, M.D., D.Sc. ** University of California, San Diego School of Medicine (T-002J, and the General Clinical Research Center, La Jolla, California Girls with central precocious puberty were utilized as a model in which to study the melatonin secretory response to ovarian suppression. Eight girls with central precocious puberty documented by clinical and endocrine characteristics, including sleep-entrained augmentation of luteinizing hormone (LH) pulsatility, were investigated. Nocturnal (6:00 P.M. to 9:00 A.M.) plasma melatonin levels were measured hourly by a sensitive and specific radioimmunoassay before and after gonadotropin-ovarian downregulation with gonadotropin-releasing hormone (GnRH)-agonist. Although nocturnal melatonin elevations varied widely between girls, patterns within the same individual were remarkably reproducible and unaltered before and after treatment. Although estrogens have been shown to modulate melatonin synthesis and secretion, in this model, reduction of estrogen levels was not associated with alterations in plasma melatonin concentrations. Fertil Steril52:936, 1989 The role of pineal melatonin in the neuroendocrine events of reproduction, including puberty, remains unclear. In immature male rats, the appropriately timed administration of melatonin delayed Received February 7, 1989; revised and accepted August 3, * Supported by grants HD-21198, HD from the National Institute of Health, Bethesda, Maryland, and by grant RR from the General Clinical Research Center, La Jolla, California. t This work was conducted by the Clayton Foundation for Research. :j: Presented in part at the Sixty-Ninth Annual Meeting of the Endocrine Society, Indianapolis, Indiana, June 10 to 12, Department of Reproductive Medicine. 'If Recipient of the Bank of American-Giannini Foundation Fellowship, The American Fertility Society Ortho Distinguished Fellow in Reproduction, and Clinical Associate Physician of the National Institutes of Health (General Clinical Research Center). II Present address and request for reprints: Sarah L. Berga, M.D., The University of Pittsburgh School of Medicine, the onset of puberty and prevented the pubertal rise in gonadotropins as well as blocked naloxoneinduced luteinizing hormone (LH) pulsatility.l In ewes, the duration of the nocturnal melatonin rhythm underlies the photoperiod-induced onset of pubertr and gonadotropin -releasing hormone (GnRH)-LH pulsatility.3 These findings, together with studies showing hypothalamic melatonin receptor sites 4 and changes consistent with inhibition of GnRH release following hypothalamic melatonin implants,5 have led to the hypothesis that melatonin modulates GnRH release and that a decrease in melatonin secretion represents a component of the complex neuroendocrine regulation of Department of Obstetrics and Gynecology, Magee-Womens Hospital, Forbes Avenue and Halket Street, Pittsburgh, Pennsylvania # Department of Pediatrics. ** Clayton Foundation Investigator. 936 Berga et ai. Melatonin after ovarian suppression Fertility and Sterility

2 sexual maturation. An obligatory role for melatonin in pubertal events is refuted, however, by the finding in agonadal male monkeys that neonatal pinealectomy did not interrupt the onset of the prepubertal hiatus in gonadotropin secretion. 6 The role of melatonin in human pubertal events is also unclear. Studies in children have shown a decline in nocturnal melatonin levels from infancy to adolescence,7-9 but a causal link between this decline and the onset of puberty or the initiation of GnRH-LH pulsatility has not been demonstrated.1o-12 Further, melatonin levels have been shown to increase as pubertal stage advances10 and to be either decreased ll or unchanged12 in children with precocious puberty when compared with levels in age-matched prepubertal controls. Although the melatonin rhythm is entrainedprimarily by the environmental light:dark cycle,13,14 estrogens have been shown, in vitro, to alter pine a locyte {j-adrenoceptors,15 cyclic adenosine monophosphate (camp) levels/ 6 and the activity of hydroxyindole-o-methyltransferase (HIOMT),17 and to increase the density of melatonin binding sites in the hypothalamus.4 These data, in conjunction with the demonstration that estradiol administration suppressed daytime human melatonin secretion,18 raise the possibility that previous and current estrogen exposure may confound either the effect of melatonin in a given experimental paradigm or the interpretation of alterations in melatonin secretory patterns associated with reproductive dysfunction such as precocious puberty or amenorrhea.19,2o To discern the role of melatonin in human reproductive physiology, including puberty, the contribution of ovarian activity, especially estradiol secretion, to plasma melatonin levels must be elucidated. To address this issue, we determined nocturnal (6:00 P.M. to 9:00 A.M.) plasma melatonin concentrations in girls with central precocious puberty before and after downregulation of the gonadotropin-ovarian axis by a GnRH-agonist. Participants MATERIALS AND METHODS Eight girls ages 3.3 to 8.8 years with central idiopathic precocious puberty were enrolled in this study. Their clinical and biochemical characteristics are presented in Table 1. All girls had both advanced bone ages and Tanner stages. Lesions were excluded by computerized tomographic scans of the head, pelvic ultrasonography, serum beta human chorionic gonadotropin ({jhcg) levels, and thyroid and adrenal functions. As a confirmation of the central nature of the precocious puberty, nocturnal, sleep-associated, LH pulsatility was documented before the initiation of GnRH -agonist therapy. These studies were approved by the Human Subjects Committee of the University of California' San Diego. The eight girls were treated a mean of 8. 7 months (range 6 to 16.5 months) with daily subcutaneous injections of the GnRH-agonist, [(ImBzI)-D-His 6, Pr09-NEt]-GnRH (Histrelin, Ortho Pharmaceutical Corp., Raritan, NJ) at a daily dose of 6 to 8 p.,g/ kg. Clinically, all participants responded to treatment with either cessation or reversal of pubertal development by the time of the follow-up study as evidenced by a decrease in height velocity, stabilization or reversal of breast and pubic hair development, cessation of menses if present, and no further advancement of bone age. Study Protocol Participants were admitted to the Clinical Research Center between 3:00 P.M. to 4:00 P.M. After placement of an intravenous catheter, they were provided a mixed meal at 5:00 P.M. to 5:30 P.M. and then allowed to play from 5:30 P.M. to 6:00 P.M. Blood sampling began at 6:00 P.M. and continued at 15 minute intervals until 9:00 A.M. The girls were allowed snacks ad libitum. From 9:00 P.M. to 7:00 A.M., all subjects were allowed to sleep or rest in a windowless room which had a light intensity of <50 lux. The girls were otherwise in an ambient light environment of >500 lux. Activity, including sleep, was monitored by trained personnel. Hormone Measurements The concentrations of serum LH, follicle-stimulating hormone (FSH), cortisol, estradiol (E2), testosterone (T), and dehydroepiandrosterone-sulfate (DS) were determined by RIAs LH, FSH, cortisol, and E2 were measured in all samples from 6:00 P.M. to 9:00 A.M.; T and DS were determined from the 6:00 P.M. sample only. Plasma samples for melatonin determination were collected hourly in plastic tubes containing ethylenediaminetetraacetic acid (EDTA) and centrifuged. The plasma was frozen immediately and stored at -20 C until assay. Plasma melatonin was measured directly, without extraction, using pre- Vol. 52, No.6, December 1989 Berga et al. Melatonin after ovarian suppression 937

3 Table 1 Clinical and Biochemical Characteristics of Eight Girls with Central Precocious Puberty Before Treatment with GnRH-Agonist Patient Chronological Tanner stages no. age breast/pubic hair y II/II III/II II/II III/II II1/II III/I III/II IV/II Mean± SE 7.2 ± 0.7 Bone Menses age E2 DS y pmol/l JLmol/L Yes No No No No No No No ± ± ± 0.14 viously described methods. 19 The assay sensitivity was 43 pmoljl (10 pg/ml). Duplicate melatonin determinations were made from each sample and plasma samples from a given individual before and after GnRH -agonist treatment were run in the same assay. The intra-assay and interassay coefficients of variation of this assay were 7.2% and 15.5%, respectively. Analyses The onset of the nocturnal melatonin rise was defined as the time at which the plasma melatonin level first exceeded 64.5 pmoljl (15 pg/ml). The offset was the first value < 64.5 pmoljl after the nocturnal rise. The duration of melatonin rise was the time between onset and offset. The amplitude was the peak nocturnal value. The integrated melatonin level was measured as the area under the curve (AVC) from 6:00 P.M. to 9:00 A.M. The pulsatility of LH was assessed by the computerized algorithm "Cluster Analysis".25 A Cluster width of two for both nadirs and peaks and a t statistic of 3.34 for both upstroke and downstroke was used. Two-tailed paired t-tests were employed to determine if there were significant differences in the reported variables before and after treatment. A P value of <0.05 was considered significant. RESULTS As shown in Table 2, GnRH -agonist treatment induced dramatic declines in LH pulse number/15 hours (P < 0.001), LH pulse amplitude (P < 0.005), and mean FSH levels (P < 0.005). While LH, T, cortisol, and DS levels were unchanged, there was a sharp decline in mean levels of E2 (P < 0.01), so that after treatment the nocturnal mean E2 level of each girl was at the lower end of the range in our laboratory for agonadal women (18.4 to pmoljl). In addition, the difference between bone age and chronological age declined after GnRH treatment. The mean ± standard error (SE) of this difference was 3.0 ± 0.4 before and 2.2 ± 0.4 after treatment, P < The melatonin profiles of each individual before and after treatment are shown in Figure 1. While each girl had a clear nocturnal elevation of melatonin levels, there was marked variability between individuals. Despite the uniform reduction in E2 levels after GnRH-agonist treatment, however, there were no consistent changes in the nocturnal melatonin profiles. Before and after values (Fig. 2) were comparable for amplitude (398 ± 106 versus 376 ± 52 pmoljl), duration (9.8 ± 0.9 versus 10.1 ± 0.8 hours), and AVC (167 ± 35 versus 162 ± 28 pmol-min/l X 10 3 ). Linear correlation and regression analysis revealed a negative correlation between age as a independent variable and the pretreatment nocturnal melatonin duration (r = -0.73, P = 0.04), but there were no significant linear correlations between age and nocturnal melatonin amplitude or AVC nor between E 2 1evels and any of the melatonin parameters. Although studies were scheduled without regard to season, inspection of the individual graphs of the nocturnal melatonin profiles revealed no striking evidence of seasonal variation in duration, amplitude, or AVC, but a formal analysis of acrophase was precluded by the absence of 24-hour sampling. DISCUSSION Before GnRH-agonist treatment, melatonin levels in girls with central precocious puberty showed 938 Berga et al. Melatonin after ovarian suppression Fertility and Sterility

4 Table 2 Biochemical Characteristics of Eight Girls with Central Precocious Puberty Before and After Treatment with GnRH-Agonist a LHpulse LH LHpulse amplitude FSH lull no./15 h lull lull Before 14.0± ± ± ±0.7 After 10.9± ± 0.5 d 2.9 ±0.6 c 4.3 ± O.4 c a Values are means ± SE. b P < E2 T Cortisol DS pmoill nrnolll nrnolll umoill ± ± ± ± ± 5.0 b 0.33 ± ± ±0.24 c P< d P<O.OO1. a modest (P < 0.04) age-related decline. This was primarily related to the duration, but not amplitude, of nocturnal melatonin secretion. When compared with women in the early follicular phase (mean age ± SE of25 ± 2 yrs),19 the mean nocturnal melatonin levels of the eight girls with central precocious puberty (mean age of 7.2 ± 0.7 yrs) were 4 hours longer in duration and 73 pmol/l greater in amplitude. Further, like those of adult women,20 melatonin secretory profiles in girls with central precocious puberty exhibited wide variation between individuals whereas patterns within individuals remained stable. Thus the dimension and heterogeneity of the melatonin secretory patterns in girls with central precocious puberty are consistent with previous reports.7-9,20 The present investigation demonstrates that despite clear evidence of pituitary-ovarian suppression by GnRH -agonist as documented by clinical responses and the declines in the levels of E2, FSH, and LH pulsatility, nocturnal plasma melatonin levels, as assessed by amplitude, duration and integrated concentrations, were unaltered in girls with central precocious puberty. This lack of change in nocturnal melatonin secretion after gonadal suppression is surprising because substantial evidence supports a direct influence of estrogen on pineal gland function: estrogens have been shown to stimulate HIOMT activity and enhance melatonin synthesis in female rat pineals,17 cause accumulations of camp in the female guinea pig pineal complex,16 and alter the receptor density and binding affinity of pineal J9-adrenoceptors in ovariectomized ewes. IS The failure to discern an effect of endogenous gonadal steroids on nocturnal melatonin secretion in this pubertal model suggests that either species differences exist or that experimentally imposed in vitro manipulations cause acute changes which are not sustained. Alternatively, the possibility exists that the effects of changing estrogen milieu on pi /2...!!!!!!!!!!!!!!!!!! 2/ J age 3.3 :::;.30 o E / /5... age 6.3 = /B...!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 6/7... _ age / ""!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 7/7... <= age o... /. \ e.. 12/ /29 --:- '4' 4/ ::!!!!!!!!!!!!!!!!!!!!!!!!!!!! 7/ / /24... c..j age B.l age B.3 age B.B... 7/ age B.B :::; i 2 21'0 21'0 21'0 21'... -'--"-"--"-- ---":..J tloo Clock Hours Figure 1 Overnight melatonin profiles determined at hourly intervals starting at 6:00 P.M. in eight girls with central precocious puberty before (closed circles) and after (open triangles) treatment with GnRH-agonist. The dates of each study are indicated in the upper left of each plot and in all instances the interval between studies was greater than 6 months. The closed bars above each plot indicate sleep duration in the initial study and the open bars above each plot indicate sleep duration in the follow-up study. Vol. 52, No.6, December 1989 Berga et ai. Melatonin after ovarian suppression 939

5 :J :::: II x a 150 E B oj "- c c.g E I j D.. L "0 j 50 :: 100 E 0 Q. E < 0 0 u 0 8 :J < Figure 2 Mean (±SE) values before (B) and after (A) GnRHagonist treatment of eight girls with central precocious puberty for each of the parameters employed to define the nocturnal melatonin profile: amplitude, duration, and AUC. alone do not adequately reflect central neurophysiology. Acknowledgments. We are grateful for the assistance of Ms. Del Alsobrook, B.A., Pam Malcom, B.S.N., Roseanne Brady, B.S., Shannon Petze, B.A., Anne Warkentin, B.S., Mary Moscinsky, B.A., Jeff Wong, B.S., and the Clinical Research Center staff of the University of California, San Diego Medical Center and for the statistical advice of Allen Lein, Ph.D. and Paul Shragg, M.A. of the University of California, San Diego School of Medicine. neal secretion are not seen until decrements of estrogen greater than those seen in this model are attained. This latter possibility is unlikely, however, because each girl showed clear clinical signs of estrogenization before treatment which were either arrested or reversed by GnRH -agonist treatment. In particular, the diminished difference between bone age and chronological age after therapy represents an in vivo index of sustained hypoestrogenism. While the possibility that centrally produced estrogens could influence melatonin synthesis and secretion cannot be excluded, if alterations in the serum levels of estrogens cause sustained alterations in nocturnal melatonin secretion in humans, such changes should have been seen in this model. Interpretation of the role of melatonin in human physiology or pathophysiology needs to account for the finding that estrogens, by increasing the density of melatonin binding sites, may augment melatonin action on hypothalamic neurosecretion. 4 Therefore, estrogen may influence hypothalamic responses, such as GnRH-LH pulsatility, to melatonin without altering plasma melatonin levels. This interpretation is supported by the observation that in ovariectomized ewes the stimulation of Gn RH-LH pulse frequency by melatonin infusions given to mimic the progression from a long to a short photoperiod was enhanced by the presence of physiological concentrations of estradiol without a coincident augmentation of pituitary responsivity to GnRH. 3 The finding that nocturnal melatonin profiles remained stable in girls with central precocious puberty despite marked reductions in estrogen levels suggests that differences in estrogen exposure do not confound comparisons of plasma melatonin levels between groups such as children with central precocious puberty and age-matched prepubertal control children, but the possibility remains that peripheral melatonin concentrations REFERENCES 1. Aubert ML, Rivest RW, Lang U, Winiger BP, Sizonenko PC: Delayed sexual maturation induced by daily melatonin administration eliminates the LH response to naloxone despite normal responsiveness to GnRH in juvenile male rats. Neuroendocrinology 48:72, Yellon SM, Foster DL: Melatonin rhythms time photoperiod-induced puberty in the female lamb. Endocrinology 119:44, Bittman EL, Kaynard AH, Olster DH, Robinson JE, Yellon SM, Karsch F J: Pineal melatonin mediates photoperiodic control of pulsatile luteinizing hormone secretion in the ewe. Neuroendocrinology 40:409, Zisapel N, Shaharabani M, Laudon M: Regulation of mel a tonin's activity in the female rat brain by estradiol: effects on neurotransmitter release and on iodomelatonin binding sites. Neuroendocrinology 46:207, Glass JD, Knotts LK: A brain site for the antigonadal action of melatonin in the white footed mouse (Peromyscus leucopus): involvement of the immunoreactive GnRH neuronal system. Neuroendocrinology 46:48, Plant TM, Zorub DS: Pinealectomy in agonadal infantile male Rhesus monkeys (Macaca mulatta) does not interrupt initiation of the prepubertal hiatus in gonadotropin secretion. Endocrinology 118:227, Attanasio A, Borrelli P, Gupta D: Circadian rhythms in serum melatonin from infancy to adolescence. J Clin Endocrinol Metab 61:388, Waldhauser F, Frisch H, Waldhauser M, Weiszenbacher G, Zeithuber U, Wurtman RJ: Fall in nocturnal serum melatonin during prepuberty and pubescence. Lancet 1:362, Waldhauser F, Weiszenbacher G, Tatzer E, Gisinger B, Waldhauser M, Schemper M, Frisch H: Alterations in nocturnal serum melatonin levels in humans with growth and aging. J Clin Endocrinol Metab 66:648, Penny R: Episodic secretion of melatonin in pre- and postpubertal girls and boys. J Clin Endocrinol Metab 60:751, Attanasio A, Borrelli P, Marini R, Cambiaso P, Cappa M, Gupta D: Serum melatonin in children with early and delayed puberty. Neuroendocrinol Lett 5:387, Ehrenkranz JRL, Tamarkin L, Comite F, Johnsonbaugh RE, Bybee DR, Loriaux DL, Cutler GB, Jr: Daily rhythm of plasma melatonin in normal and precocious puberty. J Clin Endocrinol Metab 55:307, Lynch HJ, Jimerson DC, Ozaki Y, Post RM, Bunney WE, Wurtman RJ: Entrainment of rhythmic melatonin secre- 940 Berga et ai. Melatonin after ovarian suppression Fertility and Sterility

6 tion in man to a 12-hour phase shift in the light/dark cycle. Life Sci 23:1557, Lewy AJ, Wehr TA, Goodwin FK, Newsome DA, Markey SP: Light suppresses melatonin secretion in humans. Science 210:1267, Foldes A, Maxwell CA, Scaramuzzi RJ, Donnelly JB, Hoskinson RM, Rintoul AJ: Seasonal changes in the sensitivity of ovine pineal,b-adrenoceptors to steroids. Neuroendocrinology 41:494, Cardinali DP, Vacas MI, Solveyra CG, Sarmiento MIK, Vollrath L: In vitro effects of estradiol, testosterone, and progesterone on 5-methoxyindole content, cyclic adenosine 3',5'-monophosphate synthesis, and norepinephrine release in different parts of the female guinea pig pineal complex. J Pineal Res 3:351, Cardinali DP, Nagle CA, Rosner JM: Effects of estradiol on melatonin and protein synthesis in the rat pineal organ. Horm Res 5:304, Penny R, Goebelsmann U: Effect of estradiol on plasma melatonin levels. J Endocrinol Invest 7:55, Berga SL, Mortola JF, Yen SSC: Amplification of noctur- nal melatonin secretion in women with functional hypothalamic amenorrhea. J Clin Endocrinol Metab 66:242, Brzezinski A, Lynch HJ, Seibel MM, Deng MH, Nader TM, Wurtman RJ: The circadian rhythm of plasma melatonin during the normal menstrual cycle and in amenorrheic women. J Clin Endocrinol Metab 66:891, Yen SSC, Llerena 0, Little G, Pearson OH: Disappearance rates of endogenous luteinizing hormone and chorionic gonadotropin in man. J Clin Endocrinol Metab 28:1763, Yen SSC, Llerena LA, Pearson LH, Littell AS: Disappearance rates of endogenous follicle stimulating hormone in serum following surgical hypophysectomy in man. J Clin Endocrinol Metab 30:325, Judd HL, Parker DC, Rakoff JS, Hopper BR, Yen SSC: Elucidation of mechanisms of the nocturnal rise of testosterone in men. J Clin Endocrinol Metab 38:134, Anderson DC, Hopper BR, Lasley BL, Yen SSC: A simple method for the assay of eight steroids in small volumes of plasma. Steroids 28:179, Veldhuis JD, Johnson ML: Cluster analysis: a simple, versatile, and robust algorithm for endocrine pulse detection. Am J Physiol 250:E486, 1986 Vol. 52, No.6, December 1989 Berga et al. Melatonin after ovarian suppression 941