This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kadva, A. Articles by Silman, R. Search for Related Content PubMed PubMed Citation Articles by Kadva, A. Articles by Silman, R.

Elevated Nocturnal Melatonin Is a Consequence of Gonadotropin-Releasing Hormone Deficiency in Women with Hypothalamic Amenorrhea

Department of Obstetrics and Gynaecology (A.K., O.D., W.L.D., R.S.), and Department of Endocrinology (J.M.), St. Bartholomew’s and the Royal London School of Medicine and Dentistry, Whitechapel, London E1 1BB, United Kingdom

Address all correspondence and requests for reprints to: Aban Kadva, Department of Obstetrics and Gynaecology, St. Bartholomew’s and the Royal London School of Medicine and Dentistry, Whitechapel, London E1 1BB, United Kingdom. E-mail: a.kadva{at}mds.qmw.ac.uk

	Abstract

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Elevated nocturnal melatonin is found in women with idiopathic hypogonadotropic hypogonadism (IHH), but it is not known whether this is implicated in the etiology of their GnRH deficiency. It is unlikely that nocturnal melatonin can be implicated in the etiology of the GnRH deficiency of Kallmann’s syndrome (KS), because this condition is caused by defective neuronal migration in embryonic life. We therefore measured nocturnal melatonin in women with IHH and KS to determine whether it was elevated in one or both conditions and thereby to determine whether it was implicated as cause or consequence of GnRH deficiency.

Four women with IHH, 3 women with KS, and 7 individually matched (age and body size) controls were recruited. Frequent day- and nighttime samples were taken for LH pulsatility studies. All patients showed absent or diminished LH pulsatility, compared with their respective controls. Samples were also taken over 24 h for melatonin and 6-sulphatoxymelatonin (the principle metabolite of melatonin and an independent marker of its secretion). Melatonin and 6-sulphatoxymelatonin levels were elevated in 6 of 7 patients (compared with their matched controls) and were significantly elevated in the KS group (compared with their controls).

The finding of elevated nocturnal melatonin (and its metabolite) in GnRH-deficient women with KS (as well as IHH) suggests that nocturnal melatonin is elevated as a consequence of GnRH deficiency, irrespective of its etiology.

	Introduction

Top Abstract Introduction Subjects and Methods Results Discussion References

 
IN LONG-DAY breeders, it is clear that increased nocturnal melatonin secretion in winter inhibits the GnRH pulse generator and thereby controls seasonal breeding. In humans, there is no evidence as to whether nocturnal melatonin can influence the GnRH pulse generator. However, it is known that, in women with idiopathic hypogonadotropic hypogonadism (IHH), where the etiology of the underlying GnRH deficiency is unknown, nocturnal melatonin secretion is elevated (1, 2, 3). It remains to be established whether it is this increased melatonin secretion in IHH patients that suppresses GnRH activity or whether it is the suppressed GnRH activity that increases melatonin secretion.

To explore this concept, we have studied a mixed group of GnRH-deficient female patients, some with IHH, others with Kallmann’s Syndrome (KS). If elevated nocturnal melatonin was playing a part in the etiology of IHH, we would expect it to be elevated in IHH but not in KS. The GnRH deficiency of KS is associated with a defect in neuronal migration (4), thereby making it unlikely that elevated nocturnal melatonin has any primary involvement in its etiology. However, if elevated nocturnal melatonin was a consequence of GnRH deficiency, irrespective of its etiology, we would expect it to be elevated in KS and in IHH patients.

We have studied nocturnal melatonin secretion in both groups of patients and have compared the results with matched controls. We have also studied the principal metabolite of melatonin, 6-sulphatoxymelatonin (SaMT), as an independent serum marker of melatonin secretion (5). Our aim was to determine whether nocturnal melatonin (and SaMT) levels were elevated in one or both groups of patients and thereby determine whether it was implicated as a cause or a consequence of GnRH deficiency.

	Subjects and Methods

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Patients

Seven adult female patients (four with IHH and three with KS) were recruited (Table 1). The patients were selected on the basis of having primary or secondary (minimum 6-months duration) amenorrhea, in the absence of: 1) recent stressful events; 2) psychiatric eating disorder; 3) weight loss; and 4) strenuous physical activity. None of the patients had withdrawal bleeding in response to progesterone.

Those patients who had been on hormone replacement therapy were discontinued from treatment at least 6 weeks before the study. Those patients who had previously been on ovulation induction therapy had not received any such therapy for at least 2 yr before the study.

The GnRH deficiency of the patients was assessed by intensive LH sampling; and when IHH was associated with anosmia, a diagnosis of KS was made.

Controls

Seven healthy women were recruited as individually matched controls (Table 1). Their age was within 5 yr, and their height and weight within 5%, of that of each patient. All controls had a normal menstrual cycle of 5–7/27–32 days; their ovulatory status was confirmed with a midcycle urinary rise of LH and raised midluteal serum progesterone levels (>30 nmol/L).

The controls were studied in the early follicular phase (days 2–5) of their cycle, and their GnRH activity was assessed by intensive LH sampling.

All subjects were admitted to the Assisted Conception and Reproductive Endocrinology Unit, Newham General Hospital. Approval for the study was obtained from the Local District Ethics Committee (NEC/93/009), and informed consent was obtained from all subjects.

Methods

Sampling Patients (nos. 1–7) and matched controls (nos. 1'-7')
Venous blood for serum LH (2 mL) and for melatonin and SaMT (10 mL) was obtained via an indwelling cannula throughout the 24-h study period. Before starting the intensive blood sampling schedule, blood was taken for baseline FSH and estradiol (E₂).

Blood samples were drawn for LH every 10 min, starting at 1000 h and continuing until 1600 h, and from 2200–0400 h. Two-hourly blood samples were also drawn for serum melatonin and serum SaMT till 2200 h, after which, hourly samples were collected till 0800 h the following morning. All subjects were ambulatory during the daytime (natural daylight) and were in bed (usually asleep) at night, when blinds were drawn and illumination was from night lights filtering from adjacent wards.

Assays Gonadotropins and E₂
Serum LH was assayed using a commercially available RIA kit (Serono Diagnostics, Serono, Italy). The interassay coefficient of variation (CV) was 12.5% at a concentration of 1.2 IU/L, and 4.3% at 41.7 IU/L; and the intraassay CV (IACV) was 10% at 1.1 IU/L, and 1.4% at 40.0 IU/L. The lower limit of assay detection was 0.5 IU/L. Samples from each subject were measured in the same assay.

The baseline FSH and E₂ samples were assayed using commercially available ACS-FSH, ACS-E2 chemiluminescence (Ciba Corning, Inc. Diagnostics Corp., Cedex, France) kits.

Melatonin
Samples for melatonin were assayed by RIA using I¹²⁵-labeled melatonin (6). The lower limit of assay detection was 16.8 pmol/L. The interassay CV was 14.8, 5.8, and 4.2%, and IACV was 9.9, 5.7, and 3.7%, at concentrations of 69, 215, and 1077 pmol/L, respectively.

SaMT
Samples for SaMT were assayed by RIA using I¹²⁵-labeled SaMT described in Peniston Bird et al. (7). The lower limit of assay detection was 33.6 pmol/L. The interassay CV was 18.8, 10.5, and 7.9%, and IACV was 2.6, 5.9, and 3.6%, at concentrations of 33, 215, and 646 pmol/L, respectively.

Sample preparation for melatonin and SaMT RIA
Serum samples were first extracted with methanol (AnalaR, BDH Chemicals, Dorset, Poole, UK). For melatonin 0.5 mL serum was added to 2.0 mL methanol, and for SaMT 0.25 mL serum was added to 1.0 mL methanol. The liquid phase was dried in a rotor-vac (Gyro-Vap, Howe, UK) under reduced pressure. The dried samples were reconstituted with 0.5 mL and 0.25 mL Tricine buffer, respectively.

Analysis of data LH
The raw LH data were analyzed by two independent methods. The first identified pulses, using a moving threshold, which was calculated for each individual according to the IACV, so as to constrain the false-positive error rate to 1% (8). By applying the formula, (threshold x prepeak nadir value) + prepeak nadir, to the raw LH data, the beginning of a pulse was identified whenever the LH value exceeded the value calculated from the formula. The peak was the highest value noted in the pulse, the amplitude (IU/L) was the difference between the pulse peak and prepeak nadir value, and pulse frequency per hour was calculated from the number of pulses detected in 12 h. The second method used a computer program, Cluster analysis (9), with a 2- and 1-point moving test cluster for significant increases and decreases, respectively, and a t-statistic of 2.0 for both the upstroke and downstroke.

Melatonin and SaMT
The daytime value (melatonin or SaMT) was the mean of all serum concentrations between 1000 and 1800 h. The nocturnal onset was marked as the time at which the melatonin (or SaMT) concentration first exceeded a value greater than 2 SD of the daytime value. The nocturnal offset was noted as the time point before the melatonin (or SaMT) concentration fell below the onset value. The integrated nocturnal rise of melatonin (or SaMT) was calculated (pmol/L·h) as the area under the curve (AUC) between onset and offset times. The duration of nocturnal rise of melatonin (or SaMT) was the time (in hours) between the onset and the offset, and the peak nocturnal amplitude was the highest level assayed during this period.

Statistical analysis. The difference in mean LH values between patients and controls was assessed by the two-tailed Student’s t test.

The difference in mean LH pulse frequency and amplitude between patients and controls was assessed by nonparametric (Wilcoxon rank rum test) analysis.

The peak nocturnal amplitude, daytime value, AUC, and duration of melatonin (or SaMT) levels were expressed as mean ± SE. The difference between patients and controls was assessed nonparametrically by the Wilcoxon rank sum test.

The difference between serum melatonin (or SaMT), between patient groups and their respective control groups, was assessed by ANOVA. Significance was accorded at P < 0.05.

	Results

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Baseline gonadotropins and E₂

The mean (SE) LH concentration, calculated as the mean of all the intensively (n 72) sampled LH values, and the baseline serum FSH and E₂ values for all subjects are shown in Table 2. It can be seen that there was a highly significant difference in the mean LH values between patients and their respective controls.

The individual LH pulse patterns of all patients are shown in Fig. 1, and those of the controls are shown in Fig. 2. It can be seen that all patients show diminished pulsatility, compared with controls.

View larger version (17K): [in this window] [in a new window]   Figure 1. Daytime and nighttime pulsatile LH secretion in patients 1–7; () denotes a pulse detected by the moving threshold method and () by Cluster analysis.

View larger version (23K): [in this window] [in a new window]   Figure 2. Daytime and nighttime pulsatile LH secretion in controls 1'-7'; () denotes a pulse by the moving threshold method, and () by Cluster analysis.

Of the patients with IHH, patient 1 showed two negligible pulses at night, patient 2 showed pulses of low mean amplitude (0.7 IU/L), patient 3 (who was apulsatile during the day) showed two striking pulses (mean amplitude, 2.67 IU/L) at night, and patient 4 was essentially apulsatile. Of the patients with KS, patient 5 showed pulses of low mean amplitude (0.77 IU/L) during the day and night, and patients 6 and 7 were apulsatile.

All controls showed LH pulsatility, with frequency/hour ranging from 0.58–1.08 and amplitude ranging from 1.5–3.35 IU/L. The mean pulse amplitude was 2.01 (0.26) IU/L, and the mean pulse frequency was 0.76 (0.08) per hour.

Cluster analysis

Cluster analysis identified most (but not all) of the peaks identified by the moving threshold method. The mean pulse frequency/hour was 0.19 vs. 0.58, and the mean amplitude was 0.49 vs. 2.09 IU/L, in patients vs. controls, respectively. In patients, the pulse frequency and amplitude were significantly diminished, compared with controls (P < 0.05 and P < 0.01, respectively).

Twenty-four-hour profiles of melatonin and SaMT

The individual 24-h profiles of melatonin and SaMT secretion in all patients are shown in Fig. 3. It can be seen that, in all cases, there was a clear diurnal rhythm. Though there was considerable interindividual variation, the secretion of melatonin and SaMT followed a closely similar pattern in each individual patient.

View larger version (27K): [in this window] [in a new window]   Figure 3. Twenty-four-hour profile of melatonin () and SaMT () in patients 1–7.

View larger version (28K): [in this window] [in a new window]   Figure 4. Twenty-four-hour profile of melatonin () and SaMT () in controls 1'-7'

From the above, it is clear that all patients (except patient 3) had higher nocturnal melatonin and SaMT secretion, compared with their respective matched controls.

The mean 24-h profile of melatonin in the all-patients group, or in patients grouped as IHH or grouped as KS, were compared with their respective controls (Fig. 5a). The mean nocturnal melatonin was elevated in the all-patients group and in both the IHH and KS patient groups, compared with their respective control groups. This difference was significant at five time points in the KS group, and at one time point in the IHH group and in the all-patients group. In Table 3, it can be seen that, though the mean peak nocturnal amplitude was higher in all categories (compared with their respective controls), the difference was only significant in the KS and all-patients groups. Similarly, the mean AUC was higher in all categories, compared with their respective controls, but the difference was only significant in the KS group.

View larger version (26K): [in this window] [in a new window]   Figure 5. a, Twenty-four-hour profile of melatonin (mean ± SE) in patients () and in controls () [(i), all patients vs. all controls; (ii), IHH patients vs. IHH controls; (iii), KS patients vs. KS controls]; b, 24-h profile of SaMT (mean ± SE) in patients () and in controls () [(i), all patients vs. all controls; (ii), IHH patients vs. IHH controls; (iii), KS patients vs. KS controls; *, significance at P < 0.05.

	Discussion

Top Abstract Introduction Subjects and Methods Results Discussion References

All pulses detected by Cluster analysis were detected by the moving threshold method. Cluster analysis detected fewer pulses because, in some cases, obvious peaks were not detected because of the edge effect (9). Whichever method was used, it was clear that there was significantly diminished amplitude and pulsatility in the patient group, compared with the control group.

Three out of four of our IHH patients showed elevated nocturnal melatonin and (SaMT) secretion, compared with their respective controls, corroborating the findings of others (1, 2, 3). One patient with IHH (patient 3) had lower nocturnal melatonin, compared with her control, demonstrating that, even with this condition, there can be substantial interindividual variation. Indeed, three of our seven controls had peak nocturnal melatonin levels of approximately 600 pmol/L, i.e. similar to some of the patients. As in all studies assessing abnormal melatonin secretory patterns, it is essential to individually match patients with controls, so as to compensate for the effect of interindividual variation.

When patients were regrouped differently, into those that were congenitally GnRH deficient (on the basis of absent LH pulsatility or primary amenorrhea; patients 1, 4, 6, and 7) and those that had acquired GnRH deficiency (on the basis of diminished LH pulsatility or secondary amenorrhea; patients 2, 3, and 5), it was still evident that the mean peak nocturnal melatonin amplitude and AUC were elevated in patients, compared with respective control groups (data not shown).

Thus, we have demonstrated that, in our three KS patients, nocturnal melatonin secretion was significantly elevated, compared with their respective controls. This has been previously described in one male with KS (14) but not in women. Our findings therefore offer strong evidence that the elevated nocturnal melatonin of GnRH deficiency is a consequence of the deficiency, irrespective of the etiology.

The question arises as to how GnRH deficiency can influence the pineal to elevate nocturnal melatonin secretion. It has been suggested by Okatani and Sagara (3) that, in the GnRH deficient state, it is the reduced E₂ levels that result in increased nocturnal melatonin secretion as a consequence of reduced negative feedback. To support their hypothesis, they showed that the elevated nocturnal melatonin levels in GnRH-deficient patients were lowered after E₂-replacement therapy was commenced. Furthermore, they showed that nocturnal melatonin levels were elevated in patients with endometriosis, who became E₂ deficient after receiving GnRH analogue therapy.

Whether the suppressant effect of high (exogenous) E₂ levels on melatonin is only effective in a milieu of preexisting E₂ deficiency is a consideration. Okatani and Sagara (3) could not demonstrate any fall in nocturnal melatonin during the early follicular phase in normal controls who received the same regime of E₂ therapy that was administered to their GnRH-deficient group. Also, when melatonin levels were studied in normal women during different phases (i.e. with varying levels of the normal range of E₂) of the normal menstrual cycle, there was no significant difference in nocturnal melatonin secretion (2, 15). Thus, the effect of E₂ on pineal receptors may be one that takes time for full expression and hence may not be seen with short-term administration of E₂ or the dynamic changes that occur during the menstrual cycle. However, Delfs et al. (16) did not show any reduction in nocturnal melatonin in women during persistently high E₂ states, e.g. during early pregnancy. Also, supraphysiological levels of E₂ in patients undergoing ovarian hyperstimulation for induction of ovulation did not decrease melatonin secretion.

In men, however, the story seems different. Some, but not all, patients with Klinefelter’s syndrome (who are characteristically known to have elevated E₂ levels) showed lowered nocturnal melatonin levels (17). In this study, the patients with low testosterone and elevated E₂ showed lowered nocturnal melatonin levels, whereas patients with normal testosterone and elevated E₂ did not, suggesting that there is more than a simple negative feedback effect of elevated E₂ on melatonin secretion.

In conclusion, we have demonstrated that GnRH deficiency in women with IHH and KS is associated with elevated nocturnal melatonin levels. These findings support the concept that endogenously deficient E₂ levels (prevalent in GnRH deficiency) elevate nocturnal melatonin secretion.

	Acknowledgments

 
We wish to express our gratitude to Dr. Karen Price for assistance and support with laboratory facilities within the Department of Obstetrics and Gynaecology, St. Bartholomew’s and the Royal London School of Medicine and Dentistry. We also wish to thank the Chemical Pathology Laboratory at Newham General Hospital for performing the FSH and E₂ assays, and Dr. Michael Johnson for supplying the Cluster Analysis program.

	Footnotes

 
¹ Clinical Research Fellow.

² Reader in Obstetrics and Gynaecology.

³ Reader in Medicine.

⁴ Research Fellow.

⁵ Senior Lecturer in Reproductive Physiology.

Received February 18, 1998.

Revised June 18, 1998.

Accepted June 24, 1998.

	References

Top Abstract Introduction Subjects and Methods Results Discussion References

 

1. Berga SL, Mortola JF, Yen SSC. 1988 Amplification of nocturnal melatonin secretion in women with functional amenorrhoea. J Clin Endocrinol Metab. 66:242–244.[Abstract]
2. Brzezinsky A, Lynch HJ, Seibel MM, Deng MH, Wurtman RJ. 1988 The circadian rhythm of plasma melatonin during the normal menstrual cycle and in amenorrhoeic women. J Clin Endocrinol Metab. 66:891–895.[Abstract]
3. Okatani Y, Sagara Y. 1994 Amplification of nocturnal melatonin secretion in women with functional secondary amenorrhoea: relation to endogenous oestrogen concentration. Clin Endocrinol (Oxf). 41:763–770.[Medline]
4. Schwanzel-Fukuda M, Bick D, Pfaff DW. 1989 Luteinizing hormone-releasing hormone (LHRH) expressing cells do not migrate normally in inherited hypogonadal (Kallmann) syndrome. Mol Brain Res. 6:311–326.[Medline]
5. Bojowski CJ, Arendt J, Shih MC, Markey SP. 1987 Melatonin secretion in human urine assessed by measuring its metabolite, 6-sulphatoxymelatonin. Clin Chem. 33:1343–1348.[Abstract/Free Full Text]
6. Di WL, Kadva A, Djahanbakhch O, Silman R. 1998 The RIA of bound and free melatonin in plasma. Clin Chem 44:304–310.
7. Peniston-Bird JF, Di WL, Street CA, Edwards R, Little JA, Silman R. 1996 An enzyme-immunoassay for 6-sulphatoxymelatonin in human urine. J Pineal Res. 20:51–56.[Medline]
8. Veldhuis JD, Rogol AD, Johnson ML. 1985 Minimizing false-positive errors in hormonal pulse detection. Am J Physiol. 248:(Endocrinology Metabolism 11) E475–E481.
9. Veldhuis JD, Johnson M. 1986 Cluster analysis: a simple and robust algorithm for endocrine pulse detection. Am J Physiol. 250:(Endocrinology Metabolism 13) E486–E493.
10. Yen SSC. 1993 Female hypogonadotropic hypogonadism: hypothalamic amenorrhoea syndrome. Endocrinol Metab Clin North Am. 22:29–58.[Medline]
11. Crowley WF, Jameson JL. 1992 Clinical counterpoint: gonadotropin-releasing hormone deficiency: perspectives from clinical investigation. Endocr Rev. 13:4:635–640.
12. Backstrom CT, McNeilly AS, Leask RM, Baird DT. 1982 Pulsatile secretion of LH, FSH, prolactin, oestradiol and progesterone during the human menstrual cycle. Clin Endocrinol (Oxf). 17:29–42.[Medline]
13. Raeme N, Sauder SE, Kelch RP, Marshall JC. 1984 Pulsatile gonadotrophin secretion during the human menstrual cycle: evidence for altered frequency of gonadotrophin-releasing hormone secretion. J Clin Endocrinol Metab. 59:328–337.[Abstract]
14. Luboshitzky R, Lavi S, Thuma S, Lavi P. 1995 Increased nocturnal melatonin secretion in male patients with hypogonadotrophic hypogonadism and delayed puberty. J Clin Endocrinol Metab. 80:2144–2148.[Abstract]
15. Berga SL, Yen SS. 1990 Circadian pattern of plasma melatonin concentrations during four phases of the human menstrual cycle. Neuroendocrinology. 51:606–612.[Medline]
16. Delfs TM, Baars S, Fock C, Schumacher M, Olcese J, Zimmerman RC. 1994 Sex steroids do not alter melatonin secretion in the human. Hum Reprod. 9:40–54.
17. Luboshitzky R, Wagner O, Lavi S, Herer P, Lavi P. 1996 Decreased nocturnal melatonin secretion in patients with Klinefelter’s syndrome. Clin Endocrinol (Oxf). 45:749–754.[CrossRef][Medline]

This article has been cited by other articles:

				S. Andrico, A. Gambera, C. Specchia, C. Pellegrini, L. Falsetti, and E. Sartori Leptin in functional hypothalamic amenorrhoea Hum. Reprod., August 1, 2002; 17(8): 2043 - 2048. [Abstract] [Full Text] [PDF]

				L. Falsetti, A. Gambera, L. Barbetti, and C. Specchia Long-Term Follow-Up of Functional Hypothalamic Amenorrhea and Prognostic Factors J. Clin. Endocrinol. Metab., February 1, 2002; 87(2): 500 - 505. [Abstract] [Full Text] [PDF]

				R. Luboshitzky, M. Levi, Z. Shen-Orr, Z. Blumenfeld, P. Herer, and P. Lavie Long-term melatonin administration does not alter pituitary-gonadal hormone secretion in normal men Hum. Reprod., January 1, 2000; 15(1): 60 - 65. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kadva, A. Articles by Silman, R. Search for Related Content PubMed PubMed Citation Articles by Kadva, A. Articles by Silman, R.