Abnormal melatonin secretion in hypogonadal men

Clinical Endocrinology (1997) 47, 463–469

Abnormal melatonin secretion in hypogonadal men: the effect of testosterone treatment Rafael Luboshitzky*, Oded Wagner*, Shachar Lavi*, Paula Herer† and Peretz Lavie† *Endocrine Institute, Haemek Medical Center, Afula, †Sleep Research Centre, Haifa and Faculty of Medicine, Technion, Israel Institute of Technology, Haifa, Israel (Received 15 January 1997; returned for revision 25 February 1997; finally revised 7 April 1997; accepted 3 June 1997)

Summary OBJECTIVE We have recently demonstrated that GnRH deficient male patients have increased nocturnal melatonin secretion, whereas hypergonadotrophic hypogonadal males have decreased melatonin levels. We were interested in determining whether testosterone (T) treatment (when T levels were well matched with pubertal control values) has an effect on melatonin secretory profiles in these patients. DESIGN Prospective, controlled. SUBJECTS Six male patients with idiopathic hypogonadotrophic hypogonadism (IGD), six males with hypergonadotrophic hypogonadism due to Klinefelter’s syndrome (KS) and seven controls. Patients were examined before and during the administration of 250 mg testosterone enanthate/month for four months. MEASUREMENTS Serum samples for melatonin levels were obtained every 15 minutes from 1900 to 0700 h in a controlled light-dark environment. The results of FSH, LH, T and oestradiol (E2) (determined at hourly intervals) and melatonin profiles, were compared with the pre-treatment values in each group, and with values obtained in the control group. RESULTS All 12 patients had low pre-treatment T levels (1.4 6 0.7 in IGD and 2.0 6 0.4 in KS vs. 19.8 6 2.3 nmol/l in controls) and attained normal levels after four months of T treatment (19.5 6 7 in IGD and 22.7 6 3.8 nmol/l in KS). Serum LH, FSH and E2 levels (11 6 4 IU/l, 24 6 10 IU/l and 113 6 12 pmol/l, respectively) were still elevated in KS during T treatment as compared with values in controls (2 6 1 IU/l, 2 6 1 IU/l and 67 6 4 pmol/l, respectively). In IGD, serum LH (0.12 6 0.1 IU/l) and FSH (0.16 6 0.2 IU/l) levels during T treatment were suppressed. Pretreatment melatonin Correspondence: Dr Rafael LuboshitzkyEndocrine Institute, Haemek Medical Centre, Afula, 18101, Israel. Fax: 972-6-6525553. q 1997 Blackwell Science Ltd

levels in IGD were greater than those in age-matched pubertal controls while in KS, melatonin levels were lower than values in controls. Melatonin levels were equal in all 12 hypogonadal patients and controls when T levels were well matched. Mean (6SD) darktime melatonin levels decreased from 286 6 18 to 157 6 26 pmol/l in IGD and increased from 92 6 19 to 183 6 48 pmol/l in KS (vs 178 6 59 pmol/l in controls). The integrated melatonin values decreased in IGD (from 184 6 14 to 102 6 21 pmol/min. l × 103) and increased in KS (from 64 6 13 to 123 6 40, vs. 116 6 39 pmol/min. l × 103 in controls). No correlations were found between melatonin and LH, FSH or E2 levels. CONCLUSIONS These data indicate that male patients with GnRH deficiency have increased nocturnal melatonin secretion while in hypergonadotrophic hypogonadal males melatonin secretion is decreased. Testosterone treatment normalized melatonin concentrations in these patients. Taken together, the results suggest that GnRH, gonadotrophins and gonadal steroids modulate pineal melatonin in humans.

The role of melatonin in the regulation of reproduction in seasonal mammals is well established (Tamarkin et al., 1985; Turek & Van Cauter, 1994). Although its role in humans is not fully elucidated, there is increasing evidence that melatonin plays a role in human reproduction (Turek & Van Cauter, 1994). Functional melatonin receptors were demonstrated in human granulosa cell membranes (Yie et al., 1995), prostate (Gilad et al., 1996), hypothalamic suprachiasmatic nuclei and pituitary pars tuberalis (Weaver et al., 1993), suggesting that melatonin may act on the reproductive system at three sites. Abnormal melatonin secretion was demonstrated in women with hypothalamic amenorrhoea (Berga et al., 1988; Brzezinski et al., 1988), anorexia nervosa (Tortosa et al., 1989), secondary amenorrhoea (Okatani & Sagara, 1994), in patients with precocious puberty (Waldhauser et al., 1991), and in hypogonadotrophic hypogonadism (Puig-Domingo et al., 1992; Ozata et al., 1996, Walker et al., 1996). Recently, we have demonstrated that GnRH deficient males have increased nocturnal melatonin secretion (Luboshitzky et al., 1995a) while hypergonadothrophic hypogonadal patients have decreased melatonin concentrations (Luboshitzky et al., 1996b). 463

464 R. Luboshitzky et al.

Since these two groups of hypogonadal male patients had low testosterone levels but different secretory profiles of melatonin, we assumed that melatonin and gonadotrophins are inversely related. On the other hand, the lack of correlation between LH and melatonin in GnRH deficient patients, and the inverse relations between melatonin and gonadal steroids in patients with precocious puberty (Waldhauser et al., 1991; Luboshitzky et al., 1995b), suggested that probably both gonadotrophins and gonadal steroids modulate melatonin secretion in humans. To examine this hypothesis we studied young males with hypogonadotrophic hypogonadism and hypergonadotrophic hypogonadism before and during treatment with testosterone. We determined nocturnal melatonin, LH, FSH, testosterone and E2 levels in these patients and in normal male subjects.

Subjects and methods

Participants Six males (aged 16–19 years; mean 17.3) with isolated gonadotrophin deficiency (IGD), six males (aged 16–20 years; mean 17.5) with hypergonadotrophic hypogonadism due to Klinefelter’s syndrome (KS) and seven normal young adult males (aged 17–20 years; mean 17.6) controls were enrolled in the study. The studies were approved by the local Human Subjects Committees. Participants and/or their parents gave their informed consent before the start of the study. Physical examination at the time of entering the study included assessment of pubertal development, testicular volume, height and weight (Tanner et al., 1983). All participants had an initial evaluation of serum FSH, LH, testosterone and E2 levels. All patients with KS had a 47XXY karyotype. All IGD patients had a blunted or absent LH response to LHRH (100 mg IV) and did not proceed spontaneously through normal puberty. All patients were examined for the evaluation of their nocturnal melatonin secretion in the untreated state and during testosterone treatment. Testosterone enanthate (Schering, Berlin, Germany, 250 mg) was given as a monthly IM injection. Each patient was re-examined ten days after the fourth testosterone injection. This testosterone dosage regimen was chosen as it was found to show normal testosterone levels (Finkelstein et al., 1991), although not necessarily would normalize gonadotrophin levels in patients with Klinefelter’s syndrome (Plymate, 1994).

attachment of the sleep recording electrodes, participants had a two-hour adaptation period from 1900 h to 2100 h (base-line ¼ light-time period) during which they remained in bed with lights on (approximately 150 LUX at eye levels). From 2100 h to 0700 h lights were off (dark-time period). Conventional sleep recordings were obtained from 2100 h to 0700 h to verify sleep quality. Blood samples (2 ml each) were drawn at 15 minute intervals from 1900 h to 0700 h. All studies in patients and controls were performed between December and May.

Hormone measurements Blood was centrifuged, immediately separated and stored at ¹208C, until assayed. Serum melatonin levels were measured following ethylether extraction. The RIA kits for the determination of melatonin were provided by Nichols Institute Diagnostics (Rotterdam, The Netherlands). The assay sensitivity was 11 pmol/l. Duplicate melatonin determinations were made from each sample. The intra-assay coefficients of variation (CV) of the assay were 5.5% for low concentrations (0–250 pmol/l) and 1% for high concentrations (above 250 pmol/l). The interassay CV was 7.8%. Serum FSH, LH, testosterone and E2 were determined at hourly intervals. Serum LH and FSH levels were determined by immunometric methods provided by Kodak Amerlite (Amersham, UK). The intra-assay CV’s for LH were 4.3% for low concentrations (2.2–3.3 IU/l), 2.3% for intermediate concentrations (9–13 IU/l) and 2.9% for high concentrations (27–41 IU/l). The inter-assay CV’s for LH were 2.6%, 1.3% and 1.1%, respectively. The intra-assay CV’s for FSH were 6.1%, 4.6% and 3.5% for low (2.0–4.0 IU/l), intermediate (5–15 IU/l) and high (25–45 IU/l) concentrations. The interassay CV’s for FSH were 3.9%, 2.9% and 2.0%, respectively. The sensitivities of the assays were 0.12 IU/l for LH and 0.5 IU/l for FSH. Serum testosterone and E2 levels were determined using RIA methods provided by Diagnostic Products Corp. (Los Angeles, CA). The intra-assay CV’s for testosterone were 9.4%, 2.4% and 2.9% for serum concentrations of 2.2–4.0, 16.2–25.0 and 29.4–62.0 nmol/l, respectively. The inter-assay CV’s for testosterone were 1.9%, 1.7% and 1.6%, respectively. The intra-assay CV’s for E2 were 6.4%, 2.7% and 2.3% for serum concentrations of 70–110, 150–320 and 400–700 pmol/l, respectively. The inter-assay CV’s for E2 were 2.8%, 1.5% and 1.2%, respectively. The sensitivity of the assays was 0.14 nmol/l for testosterone and 32 pmol/l for E2.

Study protocol Participants were admitted to the Technicon Sleep Research Center between 1600 h and 1700 h. All had sleeping accommodation with scalp electrodes in place. After inserting an IV catheter (kept patent with slow infusion of 0.9% NaCl) and

Analyses The onset of the nocturnal melatonin rise was defined as the time at which the first of three consecutive samples exceeded the mean concentrations of the base-line period by more than q 1997 Blackwell Science Ltd, Clinical Endocrinology, 47, 463–469

Melatonin levels in hypogonadal men 465

1 SD. The nocturnal melatonin peak was defined as the highest single concentration achieved at any time between the onset and the offset of the nocturnal rise. The time at which this occurred was defined as the time of the melatonin peak. The integrated values were determined as the area under the curve (AUC) from 1900 h to 0700 h.

Statistical analysis Paired two-tailed-t-tests as well as Wilcoxon signed rank tests were used to test whether there was any change in melatonin characteristics, LH, FSH, testosterone and E2 levels before and during treatment in each group of patients. The data in control subjects were compared to pretreatment and during testosterone treatment concentrations in IGD and KS separately by independent two-sided-t-tests and Wilcoxon 2 sample tests using unequal variance whenever necessary. Post-hoc Duncan’s Multiple Range test was performed on cell significant parameters.

Results All 12 patients had low pretreatment testosterone levels (1.4 6 0.7 nmol/l in IGD and 2.0 6 0.4 in KS) as compared with 19.8 6 2.3 nmol/l in controls (P < 0.006 for both groups). Serum testosterone levels during treatment were similar in IGD (19.5 6 3.7 nmol/l) and KS patients (22.7 6 3.8 nmol/l). Serum LH and FSH levels were suppressed in IGD during treatment (0.12 6 0.1 IU/l) and FSH (0.16 6 0.2 IU/l) levels. Pretreatment LH and FSH levels in KS were elevated (16 6 4 IU/l and 33 6 11 IU/l) and decreased to 11 6 4 and 24 6 10 during treatment (P < 0.05 for LH and P < 0.04 for FSH). Although lower than before treatment, gonadotrophin levels during testosterone treatment in KS were still significantly higher than in controls (P < 0.004 for LH and P < 0.01 for FSH). Pretreatment serum E2 levels (106 6 16 pmol/l) were elevated in KS patients (vs. 67 6 4 pmol/l in controls) and remained increased (113 6 12 pmol/l) during testosterone treatment. Pretreatment and during T therapy serum E2 levels were significantly higher than levels in controls (P < 0.01 for both groups). In IGD patients, during T therapy serum E2 levels (83 6 12 pmol/l) were not different from the values in untreated patients (70 6 4 pmol/l), but were significantly higher than the comparable values in controls (P < 0.03). Quality of sleep was assessed by actual hours for all participants and by sleep efficiency (ratio of actual hours of sleep to total hours of sleep) and was similar (65–69%) in all groups. However, both patients and controls had disturbed sleep, probably due to frequent blood sampling not performed from a separate room. Data from the nocturnal melatonin rhythms before and during testosterone q 1997 Blackwell Science Ltd, Clinical Endocrinology, 47, 463–469

treatment in all participants are shown in Table 1 and Figures 1 (a–c), and 2. The mean dark-time levels, peak nocturnal concentrations and the AUC values were significantly higher in untreated IGD patients compared with controls (Z ¼ 2.17, P < 0.02; Z ¼ 1.81, P < 0.04; Z ¼ 1.77; P < 0.04; respectively). During treatment, a pronounced decrease was observed in serum melatonin levels. The decrease in melatonin levels was evident with respect to the mean light-time and dark-time levels, peak nocturnal concentrations and the AUC values. The values in IGD during treatment were not different from the comparable values obtained in controls. The nocturnal melatonin onset time in untreated IGD patients occurred earlier than in controls (P < 0.03). During T treatment, values were still earlier, although not significantly different, than the comparable values in controls. Similarly, melatonin peak-time values during treatment occurred later than the comparable values in untreated IGD patients (P < 0.03). Prior to treatment, KS patients had statistically significant lower peak melatonin levels (Z ¼ ¹2.44, P < 0.008), dark-time levels Z ¼ ¹2.11, P < 0.02) and AUC levels Z ¼ ¹2.11, P < 0.02) in comparison to the control group. Treatment with testosterone increased melatonin mean light-time and dark-time levels as well as peak levels and AUC to values not different from those obtained in controls. Melatonin onset time and peak time values in untreated KS patients were not different from the comparable values in controls. These values did not change during T treatment. Discussion In the present study we demonstrated clear cut differences in nocturnal melatonin levels in GnRH deficient males (high melatonin concentrations versus controls) and in hypergonadotrophic hypogondal men (low melatonin versus controls). We demonstrated also that normalization of testosterone levels restored normal melatonin profiles in both patient groups. The IGD group by definition is GnRH deficient (i.e. no GnRH before and during testosterone treatment). As such, the suppression seen in FSH and LH levels takes place near the limits of assay sensitivity. Also, serum E2 levels were not different before and after testosterone treatment. Thus, in this group, the only significant change in reproductive hormones before and during testosterone treatment was the serum testosterone level. In the KS group, serum E2, FSH and LH levels were elevated prior to testosterone administration and remained so during treatment. Increased melatonin levels as observed in our IGD patients were also demonstrated in other studies (Puig-Domingo et al., 1992; Ozata et al., 1996; Walker et al., 1996). The effect of sex steroids on melatonin secretion was evaluated in several studies, both in male and female patients, but revealed conflicting results (Arendt et al., 1985;


Table 1 Nocturnal melatonin secretory patterns in hypogonadal men before and during testosterone treatment (mean 6 SD). Melatonin

Group Control (n ¼ 7) IGD (n ¼ 6) IGD þ T (n ¼ 6) P valuea KS (n ¼ 6) KS þ T (n ¼ 6) P valueb

Onset (clock h)

Peak-time (clock h)

Peak-level (pmol/l)

Light-time Dark-time AUC levels levels (pmol/min) Testosterone (nmol/l) (pmol/l) (pmol/l) l × 103

FSH (IU/l)

LH (IU/l)

Oestradiol BMI (pmol/l) (kg/m2)

22:06 6 0:20 03:56 6 142 293 6 89

61 6 31

178 6 59

116 6 39

19.8 6 2.3

261

261

67 6 4 22.6 6 0.6

21:17 6 0:40 02:23 6 1:57 430 6 45

95 6 27

286 6 18

184 6 14

1.4 6 0.7

1.4 6 0.8

1.0 6 0.7

70 6 4 23.1 6 0.8

21:38 6 0:26 04:23 6 1:55 241 6 58

61 6 14

157 6 26

102 6 21

19.5 6 3.7

0.16 6 0.2 0.12 6 0.1

83 6 12 23.5 6 1.1

n.s. 0.03 0.03 23:15 6 1:58 03:57 6 1:29 165 6 41

0.03 63 6 18

0.03 92 6 19

0.03 64 6 13

0.03 2.0 6 0.4

0.03 33 6 11

0.03 16 6 4

n.s. n.s. 106 6 16 22.7 6 0.8

21:31 6 0:18 03:27 6 1:19 272 6 95

77 6 35

183 6 48

123 6 40

22.7 6 3.8

24 6 10

11 6 4

113 6 12 23.2 6 0.9

n.s.

0.05

0.05

0.001

0.04

0.05

n.s.

n.s.

n.s.

n.s.

n.s.

AUC, Area under the curve; a Compared with untreated IGD patients; b Compared with untreated KS patients; T, testosterone treatment; IGD, Hypogonadotrophic hypogonadism; KS, Klinefelter’s syndrome; BMI, Body Mass Index; n.s., Not Significant.

Puig-Domingo et al., 1992; Delfs et al., 1994; Bartsch et al., 1995; Commentz & Willing, 1995). Recently Ozata et al. (1996) examined the effect of testosterone and GnRH treatment on melatonin levels in hypogonadal men. The authors failed to demonstrate any effect of testosterone or GnRH on melatonin profiles. However, in this study a single blood determination of melatonin level drawn at 0730 h was performed in untreated IGD patients with repeated single determination of melatonin level during treatment. Due to the large variability in the amplitude of the melatonin rhythm between subjects and the intra-individual variations in melatonin concentrations (Arendt, 1995), the interpretation of the results obtained in their study should be tentative, especially regarding the effect of testosterone or GnRH treatment on melatonin concentration. In the present study, patients were given a monthly injection of 250 mg testosterone enanthate. This treatment regimen results in non-physiological fluctuations in testosterone and E2 levels. Thus, it is likely that the effect of testosterone treatment on melatonin levels in IGD patients may be due to either testosterone or E2 as both are increased during treatment. The current results do not permit separation of the effects of testosterone from those of E2, given aromatization in vivo (Finkelstein et al., 1991; Bagatell et al., 1994). Recently, we have demonstrated that testosterone treatment given to male patients with constitutional delayed puberty or isolated hypogondotrophic hypogonadism normalized their melatonin levels. These data suggested that in GnRH deficiency, melatonin and sex steroids are inversely related (Luboshitzky

et al., 1996a). In contrast, the data in KS patients, in whom elevated gonadotrophin levels were associated with decreased melatonin secresion, suggest a different correlation between melatonin and the reproductive hormones. In other words, it is possible that gonadotrophins or GnRH rather than sex steroids may be suppressed melatonin concentrations in KS patients. Only when combining the data on melatonin profiles obtained in GnRH deficient males on one hand and KS patients on the other hand, before and during testosterone replacement therapy as presented in the current study, can we conclude that sex steroids, GnRH and gonadotrophin act together to modulate melatonin secretion. Similar to KS, in patients with precocious puberty, in whom gonadotrophins and gonadal steroids concentrations are increased, melatonin levels were suppressed (Waldhauser et al., 1991; Luboshitzky et al., 1995b). Animal studies decisively indicate that sex steroids affect pineal melatonin secretion. Decreased ability of the pineal gland to synthesize melatonin was shown in castrated male rats and the addition of testosterone restored their ability to synthesize melatonin comparable to controls (Daya & Potgeiter, 1985). The effects of testosterone on pineal melatonin secretion were examined in vitro in perifused pineal glands and revealed that at low concentrations, testosterone increases melatonin content while greater doses decreased pineal function (Cardinali et al., 1987; Yie & Brown, 1995; San Martin et al., 1996). Testosterone given to castrated male rats either increased or decreased pineal serotonin N-acetyltransferase (NAT) activity (Daya & Potgeiter, 1985; San Martin et al., 1996). Adding to the complexity of the possible modulatory q 1997 Blackwell Science Ltd, Clinical Endocrinology, 47, 463–469


(a) 400 350 300 250 200 150 100 50 0 1900

2100

2300

0100 Lights off

0300

0500

0700

2100

2300

0100 Lights off

0300

0500

0700

(b) 400

Melatonin (pmol/l)

350 300 250 200 150 100 50 0 1900

(c) 350 300 250 200 150 100 50 0 1900

2100

2300

0100 0300 Time (24h clock)

q 1997 Blackwell Science Ltd, Clinical Endocrinology, 47, 463–469

0500

0700

Fig. 1 Nocturnal serum melatonin concentrations before (solid line) and during (broken line) testosterone treatment (mean 6 SD). a. in hypogonadotrophic hypogonadal males (n ¼ 6), b. hypergonadotrophic hypogonadal males (n ¼ 6). c. nocturnal serum melatonin levels in 12 hypogonadal males during testosterone treatment (dotted line, Hypogonadotrophic hypogonadism, n ¼ 6; broken line, Klinefelter’s syndrome, n ¼ 6) and in seven control subjects (solid line). The shaded area represents 1 SD above and below the mean in controls.

Melatonin (AUC) (pmol/ 3 min. l × 10 )


200 150 100 50 0 IGD

Ks

Control

Fig. 2 Integrated melatonin values (AUC) before (B) and during (A) testosterone treatment in male patients with hypogonadism g controls. 1 GD, Hypogonadotrophic hypogonadism, KS, Hypergonadotrophic hypogonadism.

hormones of melatonin secretion are the studies demonstrating the presence of GnRH in the monkey pineal gland (Pe´vet, 1983) and the capability of GnRH to influence protein synthesis in the rat pinealocytes (Halder-Misra & Pe´vet, 1983). These studies indicated that GnRH directly or indirectly may also participate in the modulation of melatonin. These data and the results in the present study suggest that it is most likely that the modulation of melatonin secretion is mediated by the combined action of GnRH, gonadotrophins and gonadal steroids. Supporting this hypothesis is our novel observation that LH, FSH, androgen and oestrogen receptors are present in human pinealocytes (Luboshitzky et al., 1997). It is possible that GnRH receptors are also present in human pineal glands. In conclusion, male patients with GnRH deficiency have higher melatonin levels than normal controls, whereas in hypergonadotrophic hypogonadism melatonin secretion is suppressed. Testosterone treatment resulted in normalization of melatonin levels in both groups. Our data suggest that GnRH, gonadotrophins and gonadal steroids modulate melatonin secretion in humans. Acknowledgements We are indebted to Mrs Frances Nachmani for her excellent secretarial assistance, to Mr Isaam Thuma for technical assistance and to Ms Yael Meshulam for graphical assistance. Supported by grants from the B. Rappaport Research Fund, The Maurice and Arlene King Fund (USA), and the Israel Ministry of Health and Technion grants (No. 181-466 and 181645). References Arendt, J., Labib, M.H., Bojkowski, C., Hanson, S. & Marks, V. (1985) Rapid decrease in melatonin production during successful treatment of delayed puberty. Lancet, I, 1326.

Arendt, J. (1995) The pineal in human physiology and pathology. In: Melatonin and the mammalian pineal gland. pp 201–246, Chapman & Hall, London. Bagatell, C.J., Dahl, K.D. & Bremner, W.J. (1994) The direct pituitary effect of testosterone to inhibit gonadotrophin secretion in men is partially mediated by aromatization to estradiol. Journal of Andrology, 15, 15–21. Bartsch, C., Seeger, H., Muck, A.O. & Lippert, T.H. (1995) The effect of estradiol on the production of melatonin in postmenopausal women. International Journal of Clinical Pharmacological Therapeutics, 33, 401–403. Berga, S.L., Mortola, J.F. & Yen, S.S.C. (1988) Amplification of nocturnal melatonin secretion in women with functional hypothalamic amenorrhea. Journal of Clinical Endocrinology and Metabolism, 66, 242–244. Brzezinski, A., Lynch, J.H.J., Seibel, M.M., Deng, M.H., Dader, T.M. & Wurtman, R.J. (1988) The circadian rhythm of plasma melatonin during the normal menstrual cycle and in amenorrheic women. Journal of Clinical Endocrinology and Metabolism, 66, 891–895. Cardinali, D., Vacas, M.I., Keller-Sarmiento, M.I., Etchegoyen, G.S., Pereyra, E.N. & Chulugan, H.F. (1987) Neuroendocrine integrative mechanisms in mammalian pineal gland: Effects of steroids and adenohypophysial hormones on melatonin synthesis in vivo. Journal of Steroid Biochemistry, 27, 565–571. Commentz, J. & Willig, R. (1995) No effect of ethinyl-estradiol treatment on melatonin secretion in healthy pubertal girls. Experimental Clinical Endocrinology, 103, 52–57. Daya, S. & Potgeiter, C. (1985) The effect of castration, testosterone and estradiol on 14C-Serotonin metabolism by organ cultures of male rat pineal gland. Experientia, 41, 275–276. Delfs, T.M., Baars, S., Fock, C., Schumacher, M., Olcese, J. & Zimmermann, R.C. (1994) Sex steroids do not alter melatonin secretion in humans. Human Reproduction, 9, 49–54. Finkelstein, J.S., Whitcomb, R.W., O’Dea, L.L., Langcopes, C., Shoenfeld, D.A. & Crowley, W.F. Jr. (1991) Sex steroid control of gonadotrophin secretion in the human male. I Effects of testosterone administration in normal and gonadotropin releasing hormone deficient men. Journal of Clinical Endocrinology and Metabolism, 73, 609–620. Gilad, E., Laudon, M., Matzkins, H., Pick, E., Sofer, M., Braf, Z. & Zisapel, N. (1996) Functional melatonin receptors in human prostate epithelial cells. Endocrinology, 137, 1412–1417. Halder-Misra, C. & Pe´vet, P. (1983) The influence of luteinizinghormone-releasing hormone (LHRH) on the process of protein and/ or peptide secretion characterized by the formation of granular vesicles in mammalian pinealocytes. Cell Tissue Research, 232, 529–538. Luboshitzky, R., Lavi, S., Thuma, I. & Lavie, P. (1995a) Increased nocturnal melatonin secretion in male patients with hypogonadotropic hypogonadism and delayed puberty. Journal of Clinical Endocrinology and Metabolism, 80, 2144–2148. Luboshitzky, R., Tiosano, D.I., Ben-Harush, M.I., Thuma, I., Ayach, A., Lavie, P. & Etzioni, A. (1995b) Pseudo-precocious puberty in a male patient and the melatonin-testosterone relationship. Journal of Pediatric Endocrinology and Metabolism, 8, 295–299. Luboshitzky, R., Lavi, S., Thuma, I. & Lavie, P. (1996a) Testosterone treatment alters melatonin concentrations in male patients with gonadotropin-releasing hormone deficiency. Journal of Clinical Endocrinology and Metabolism, 81, 770–774. Luboshitzky, R., Wagner, O., Lavi, S., Herer, P. & Lavie, P. (1996b) q 1997 Blackwell Science Ltd, Clinical Endocrinology, 47, 463–469


Decreased nocturnal melatonin secretion in patients with Klinefelters Syndrome. Clinical Endocrinology, 45, 749–754. Luboshitzky, R., Dharan, M., Goldman, D., Hiss, Y., Herer, P. & Lavie, P. (1997) Immunohistochemical localization of gonadotropin and gonadal steroid receptors in human pineal glands. Journal of Clinical Endocrinology and Metabolism, 82, 977–981. Okatani, Y. & Sagara, Y. (1994) Amplification of nocturnal melatonin secretion in women with functional secondary amenorrhea: relation to endogenous oestrogen concentration. Clinical Endocrinology, 41, 763–770. Ozata, M., Bulur, M., Bingal, N., Beyhan, Z., Corakci, A., Bolu, F. & Gundogan, M.A. (1996) Daytime plasma melatonin levels in male hypogonadism. Journal of Clinical Endocrinology and Metabolism, 81, 1877–1881. Pe´vet, P. (1983) The different classes of proteic and peptidic substances present in the pineal gland. In: The pineal gland and its endocrine role. (Ed. Axelrod, J., Fraschini F. & Velo, G.P.) pp 113–149, Plenum Publ., New York. Plymate, S. (1994) Hypogonadism. Endocrinology Metabolism Clinics of North America, 23, 609–620. Puig-Domingo, M., Webb, S.M., Serrano, J., Peinado, M.A., Corcoy, R., Ruscalleda, J., Reiter, R.J. & De Leiva, A. (1992) Melatoninrelated hypogonadotropic hypogonadism. New England Journal of Medicine, 327, 1356–1359. San Martin, M., Bogdan, A. & Touitou, Y. (1996) Day-night differences in the effects of gonadal hormones on melatonin release from perifused rat pineals. Evidence of a circadian control. Steroids, 61, 27–32. Tamarkin, I., Baird, C.I. & Almeida, O.E.X. (1985) Melatonin: A

q 1997 Blackwell Science Ltd, Clinical Endocrinology, 47, 463–469

coordinating signal for mammalian reproduction. Science, 227, 714– 720. Tanner, J.M., Whitehouse, R.H., Cameron, N., Marschal, W.A., Healy, M.J.R. & Goldstein, H. (1983) Assessment of skeletal maturity and prediction of adult height. pp 1–106, Academic Press, London. Tortosa, F., Puig-Domingo, M., Peinado, M.A., Oriola, J., Webb, S.M. & De Leiva, A. (1989) Enhanced circadian rhythm of melatonin in anorexia nervosa. Acta Endocrinologica, 120, 574–578. Turek, F.W. & Van Cauter, E. (1994) Rhythms in reproduction. In: The Physiology of Reproduction (Ed. Knobil E. & Neil D.) pp 487–540, Raven Press, New York. Waldhauser, F., Boepple, P.A., Schemper, M., Mansfield, M.J. & Crowley, W.F. Jr (1991) Serum melatonin in central precocious puberty is lower than in age-matched prepubertal children. Journal of Clinical Endocrinology and Metabolism, 73, 793–796. Walker, A.B., English, J., Arendt, J. & MacFarlane, I.A. (1996) Hypogonadotropic hypogonadism and primary amenorrhea associated with increased melatonin secretion from a cystic pineal lesion. Clinical Endocrinology, 45, 353–356. Weaver, D.R., Stehle, J.H., Stopa, E.G. & Reppert, S.N. (1993) Melatonin receptors in human hypothalamus and pituitary: Implications for circadian and reproductive responses to melatonin. Journal of Clinical Endocrinology and Metabolism, 76, 295–301. Yie, S.M., Niles, L.P. & Younglai, E.V. (1995a) Melatonin receptors on human granulosa cell membranes. Journal of Clinical Endocrinology and Metabolism, 80, 1747–1749. Yie, S.M. & Brown, G.M. (1995) Effects of sex hormones on the pineal response to isoproterenal and on pineal beta-adrenergic receptors. Neuroendocrinology, 62, 93–100.