|
|
|
|
|
|
Copyright © 2001 by The Endocrine Society
The mechanisms leading to alterations in plasma melatonin (MT) levels with testosterone replacement in Klinefelter's syndrome (KS) remain elusive. We investigated early morning plasma MT levels, urinary 6-sulfatoxymelatonin (6-SM) levels, and urinary catecholamine levels before and 6 months after testosterone treatment in 31 patients with KS and 20 healthy males to demonstrate whether alterations in plasma MT levels in such patients are due to subtle changes in sympathoadrenal activity and/or alterations in the hepatic indolamine metabolism influenced by testosterone replacement.
The plasma MT level was measured by RIA. The sensitivity of the test was 10.7 pmol/L. The 6-SM level was measured by enzyme-linked immunosorbent assay. Urinary catecholamines were determined by high performance liquid chromatography. The pretreatment mean plasma MT level was insignificantly higher in the patient group than in the control group (72.57 ± 74.82 vs. 42.37 ± 29.02 pmol/L; z = -1.218; P = 0.223). The pretreatment urinary 6-SM and norepinephrine (NE) levels were significantly lower and, the epinephrine (E) and dopamine levels were insignificantly lower in the patient group than those in the control group [6-SM, 76.54 ± 31.92 vs. 125.49 ± 50.16 nmol/L (z = -3.727; P < 0.001); NE, 120.79 ± 58.33 vs. 178.84 ± 81.61 nmol/day (z = -2.585; P = 0.01); E, 31.27 ± 27.42 vs. 34.65 ± 28.33 nmol/day (z = -0.39; P = 0.692); dopamine, 1577.02 ± 863.02 vs. 1812.32 ± 677.59 nmol/day (z = -1.03, P = 0.308)]. After testosterone replacement, plasma MT levels were significantly decreased (72.57 ± 74.82 vs. 24.73 ± 23.61 pmol/L; z = -4.29; P < 0.001), and urinary 6-SM, NE, E, and dopamine levels were significantly increased [6-SM, 25.04 ± 10.44 vs. 40.05 ± 17.65 ng/mL (z = -4.78; P < 0.001); NE, 120.78 ± 58.33 vs. 154.08 ± 61.35 nmol/day (z = -4.27; P < 0.001); E, 31.27 ± 27.42 vs. 40.74 ± 30.04 nmol/day (z = -4.22; P < 0.001); dopamine, 1577.02 ± 863.02 vs. 2162.67 ± 823.15 (z = -6.127; P < 0.001)].
There was no relation between plasma MT levels, urinary 6-SM, and catecholamine levels and levels of gonadotropins or gonadal steroids either before or after treatment.
We demonstrate that in untreated KS, plasma MT levels tend to be higher than those in normal controls, whereas those of the melatonin metabolite 6-SM and those of NE in urine tend to be lower. After testosterone treatment, however, plasma MT levels fall significantly, whereas urinary levels of 6-SM and NE rise. Our data show that the effect of testosterone is mediated by enhanced metabolism of melatonin, not by any effect on net sympathetic outflow, and that the increase in plasma melatonin in untreated KS patients also results from an alteration in the rate of melatonin metabolism and not from increased net sympathetic activity. ( J Clin Endocrinol Metab 86: 738-743, 2001)
THE ROLE OF melatonin (MT) and the pineal gland in both physiological and pathological states in humans remains unclear. Seemingly, the effects of MT on reproductive function are also not clarified [1] [2] . MT synthesis is under strict adrenergic control. Increased MT synthesis in the pineal gland during darkness is induced by norepinephrine (NE) release from the postganglionic nerve endings via beta1 -receptors on the pinealocyte membrane [3] [4] . Stimulation of the beta-adrenergic receptor is further potentiated by alpha1 -receptors [5] . MT is primarily metabolized in liver via hydroxylation, and the major urinary metabolite of MT is 6-sulfatoxymelatonin (6-SM) [6] [7] [8] [9] .
Increased MT levels in male hypogonadism have raised the question of whether an association exists between MT and reproductive state. Previous studies have reported elevated nocturnal MT levels in some patients with hypothalamic amenorrhea, anorexia nervosa, and delayed puberty, suggesting the regulatory action of MT on reproductive function [10] [11] [12] [13] . Moreover, Puig-Domingo et al. [14] reported MT-related hypogonadotropic hypogonadism in a patient with pineal calcification and hypermelatoninemia. We have previously demonstrated that patients with idiopathic hypogonadotropic hypogonadism have increased early morning MT levels and are not influenced by short-term gonadotropin treatment [15] . However, plasma MT levels in primary hypogonadism are controversial, and more studies are needed to clarify this issue [16] [17] [18] [19] [20] . Luboshitzky et al. [17] [18] found low MT levels in primary hypogonadism, whereas Tortosa et al. [19] and recently Rajmil et al. [20] demonstrated increased plasma MT levels in primary hypogonadism. Moreover, we previously found a slight, but not significant, increase in primary hypogonadism [15] . The mechanisms
In the present study we aimed to clarify whether alterations in plasma MT levels in Klinefelter's syndrome (KS) were due to differences in sympathoadrenal activity and/or alterations in the hepatic indolamine metabolism influenced by T replacement treatment.
Thirty-one untreated patients with KS (mean age, 20.9 ± 1.2 yr) and 20 age-matched healthy men (mean age, 21.3 ± 1.4 yr) were enrolled in the study. The diagnosis of KS was based on eunuchoid appearance, decreased serum T concentration below normal range for adults, increased gonadotropin levels, and abnormal karyotype (47,XXY). All patients and control subjects gave informed consent, and the study was approved by the local ethical committee of Gulhane School of Medicine.
Patients with KS were treated with an im injection of Sustanon 250 (Organon, Oss, The Netherlands) every 2 weeks that contained 30 mg T propionate, 60 mg T phenylpropionate, 60 mg T isocaproate, and 100 mg T decanoate for 6 months. Follow-up evaluations were performed 6 months later. Plasma levels of FSH, LH, total T (TT), free T (FT), TSH, GH, PRL, estradiol (E2 ), dehydroepiandrosterone sulfate (DHEAS), cortisol, and MT were measured 7 days after the last injection of Sustanon. Similarly, a 24-h urine sample was collected for epinephrine (E), NE, dopamine, and 6-SM measurements.
Blood samples for MT measurement were drawn in the early morning (at 0730 h) in all patients and control subjects before and 6 months after treatment. We evaluated MT levels in blood samples drawn at the same time (0730 h) before and after therapy to prevent the effect of intra individual variation on MT levels, as previous studies have demonstrated that MT secretory patterns are specific for each individual [21] , and that intraindividual variation of MT tends to be relatively small [22] [23] . Moreover, it is known that 6-SM excretion is highly reproducible in an individual subject.
Figure 1. Plasma melatonin levels before and after treatment in patients with Klinefelter's
syndrome.
Figure 2. Urinary 6-sulphatoxymelatonin levels before and after treatment in patients
with Klinefelter's sydrome.
Plasma MT was measured by RIA using a commercial kit supplied by Immuno Biological Laboratories (Direct 125 I RIA Kit, Hamburg, Germany). The assay sensitivity was 10.7 pmol/L. Duplicate MT determinations were made from each sample. The intraassay coefficient of variation of the assay at 39 pmol/L was 1% (n = 4).
Duplicate 24-h urine samples were collected from each individual for determination of 6-SM and catecholamine levels. Urine was collected in HCl-preadded bottles for catecholamine measurements. Blood samples for MT measurements were drawn simultaneously with 24-h urine collection for 6-SM measurement. Ten milliliters of each urine sample were stored at -70 C. 6-SM was measured by enzyme-linked immunosorbent assay with reagents from Immuno Biological Laboratories (melatonin sulfate and enzyme-linked immunosorbent assay kit). The assay sensitivity was 3.057 nmol/L. The intraassay coefficient of variation at 21.4 nmol/L was 5.4% (n = 10), and that at 137.6 nmol/L was 3.4% (n = 10). Urinary catecholamines were determined by high performance liquid chromatography (Millipore Corp., Milford, MA) using the same colon.
Serum FSH, LH, TT, E2 , PRL, TSH, cortisol (reagents from Chiron Corp., Halstead, UK), FT (Diagnostics Systems Laboratories, Inc., Webster, TX), and DHEAS and GH (Diagnostic Products, Los Angeles, CA) were measured by chemiluminescence techniques at the Nuclear Medicine Laboratory. The normal levels of these hormones as follows: FSH, less than 15 IU/L; LH, less than 15 IU/L; PRL, 2.1-17.7 mug/L; TT, 10.76-36.94 nmol/L; FT, 0.36-2.46 pmol/L; cortisol, 137.95- 689.75 nmol/dL; DHEAS, 0.1-1.20 mumol/L; GH, less than 8 mug/L; TSH, less than 6.5 mIU/L; and E2 , less than 220.26 pmol/L.
All results are given as the mean ± SD. According to the distribution of the data, either paired t test or Wilcoxon signed ranks test was used to compare related samples data and either unpaired t test or Mann-Whitney U test was used to compare independent sample data. Correlations between various parameters were determined by Pearson correlation analysis. A calculated P < 0.05 is considered statistically significant.
All variables in the patient and the control groups are given in Table 1 . No significant differences regarding age, PRL, or TSH were detected between patient and control groups (z = -0.875, P = 0.382; z = -1.245, P = 0.213; and t = 0.866, P = 0.391, respectively). Plasma FSH, LH, and DHEAS levels in patients with KS were significantly higher than those in control subjects (z = -3.727, P < 0.001; z = -5.982, P < 0.001; and t = -2.211, P = 0.032, respectively). Plasma T and FT levels, cortisol levels, and right and left testicular volumes were significantly lower than those in the normal men (z = -8.833, P < 0.001; z = -7.413, P < 0.001; z = -2.383, P = 0.017; z = -70.448, P < 0.001; and z = -60.644, P < 0.001, respectively). Plasma GH levels were significantly higher, and MT levels were insignificantly higher (z = 2.662, P = 0.01; and z = -1.218, P = 0.223, respectively), whereas urinary 6-SM levels were significantly lower in the patient group than in the control group (z = -3.727; P < 0.001). Urinary NE levels were significantly lower, and E and dopamine levels were insignificantly lower in the patient group than in the control group (z = -2.585, P = 0.010; z = -0.396, P = 0.692; and t = -1.030, P = 0.308, respectively).
As shown in
Table 2
, plasma levels of GH, PRL, TSH, DHEAS,
and cortisol and right and left testicular volumes did not change significantly after
6 months of T treatment (t = -0.135, P = 0.894;
z = -1.695, P = 0.090; z = -0.425, P
= 0.671; t = 0.281, P = 0.781; t = -1.969,
P = 0.058; z = -1.368, P
= 0.171; and t = 1.945, P
= 0.061, respectively). Plasma FSH and LH levels were significantly decreased ( t
= 6.368, P < 0.001; and z = -4.860, P
< 0.001, respectively), whereas plasma T and FT were restored to normal levels
( t = -14.196, P
< 0.001; and t = -12.066, P
< 0.001, respectively). Interestingly, plasma MT levels were significantly decreased
(72.57 ± 74.82 vs. 24.73 ± 23.61 pmol/L;
z = -4.286; P < 0.001) (
Fig. 1
), whereas urinary 6-SM (
Fig. 2
), NE, E, and
There was no correlation among plasma MT levels, urinary 6-SM, catecholamine levels, and plasma levels of FSH, LH, T, FT, PRL, TSH, GH, DHEAS, E2 , and cortisol (data not shown).
Our study demonstrated that plasma MT levels are slightly, but not significantly, higher in untreated patients with KS than in controls, and T treatment for 6 months results in a significant decrease in plasma MT levels. Recent studies have demonstrated that plasma MT levels were high in males with hypogonadotropic hypogonadism, and gonadotropin administration leads to a significant decrease in plasma MT levels in such patients [14] [15] . Rajmil et al. [20] found higher plasma MT levels in male patients with primary hypogonadism and demonstrated that T replacement for 3 months normalized plasma MT levels. Moreover, we previously found a slight, but not significant, increase in primary hypogonadism [15] . Thus, more studies are needed to clarify this issue in primary hypogonadism [16] . It may be argued that different genetic trait of KS may account for the variable MT production or regulation.
The mechanisms leading to alterations in plasma MT levels with T replacement also remain elusive. One possible explanation is that T leads to an increase in hepatic indolamine metabolism. It is known that T induces liver enzyme activity [20] . An increased metabolic rate for MT would apparently lead to higher levels of urinary 6-SM. On the other hand, no changes would have occurred in the case of decreased MT production. Interestingly, in this present study
It is interesting to note that Walker et al. [24] demonstrated that atenolol, but not E2 , treatment in a patient with hypogonadotropic hypogonadism reduced MT production. It was previously shown that NE increases MT biosynthesis via beta1 -receptors on the pinealocyte membrane [3] [4] . Furthermore, alpha 1 -receptors located on these cells enhance beta-receptor stimulation [5] . Yie and Brown [25] showed that sex hormones regulate pineal MT production by modifying the beta-adrenergic mechanism. As androgens have been reported to modulate sympathoadrenal activity in the rat, some investigators have also focused on plasma catecholamine levels in hypogonadal patients. However, no study to date addressed the question of whether alteration of sympathetic nervous system activity could modulate MT levels before and after treatment in male hypogonadism. In support of our findings, Del Rio et al. showed that plasma NE, but not E, levels are significantly low in hypogonadal patients, and T replacement leads to restoration of plasma NE levels [26] . Moreover, Coletta et al. [27] demonstrated that sympathoadrenal activity is reduced in KS and that T treatment is able to restore normal activity of the sympathetic nervous system. However, we could not find any correlation between MT and urinary catecholamine levels either before or after treatment. Thus, our data demonstrate that the effect of T on plasma MT levels is not mediated by any effect on net sympathetic outflow. In our study we also observed that urinary NE, but not E and dopamine, levels in untreated KS patients were lower than those in normal men. Thus, our findings provide evidence that the tendency for plasma MT to increase observed in untreated KS patients also does not result from increased net sympathetic activity. Using urinary NE and E as indexes of relevant sympathoadrenal activity relies on the presumption that the firing of the sympathetic nerves to the pineal gland, such as from superior cervical ganglion, behave like sympathetic nerves elsewhere in the body. However, there is evidence that this is often not the case. For example, sympathetic outflow from the superior cervical ganglion is affected by light, but not posture, whereas that from the celiac axis shows opposite effects [28] . In other words, distinct pathways in the central nervous system modulate the changes in sympathoadrenal outflow that occur in response to drugs or particular changes in the physiological state [28] . However, it is difficult to assess only superior cervical ganglion activity before and after T therapy. Sensitive tools to assess the activity of the sympathetic nervous system in humans include measurements of catecholamine levels, norepinephrine spillover techniques, and microneurography. For instance, microneurography determines muscle sympathetic outflow. Hence, as described previously, we employed urinary NE, E, and dopamine levels to evaluate the sympathoadrenal status before and after androgen replacement [26] [27] .
The exact role of the pineal gland in human reproductive function is not well understood. A direct modulatory effect of the gonadal hormones on pineal MT synthesis is well established in animal studies [29] [30] . In the case of humans, abnormal MT release associated with disorders of the reproductive system can only be argued in the presence of compelling evidence suggesting a relationship between MT and the hypothalamo-pituitary-gonadal axis [14] . The demonstration of MT receptors in the brain [31] [32] , hypothalamic suprachiasmatic nuclei [33] , human granulosa cell membranes [34] , and prostate [35] together with the enhancing effect of MT on the gonadotropin response to submaximal GnRH stimulation in the follicular, but not the luteal, phase of the menstrual cycle in normal women [36] suggest the existence of a putative interaction between GnRH and MT
Recent studies also demonstrated that the Mel 1a receptor is expressed in the hypophyseal pars tuberalis and the suprachiasmatic nucleus, which are the presumed sites of the reproductive and circadian actions of MT. The Mel 1b MT receptor is expressed mainly in the retina and, to a lesser extent, in the brain [45] [46] . Taken together, these findings suggest that MT has multiple sites of action in reproductive function. Molecular and cellular studies of the MT signaling system, its regulation, and effects on downstream functional events in the future may provide new insights into the relationship between gonadal steroids and MT secretion in humans.
In summary, we demonstrate that in untreated KS, plasma MT levels tend to be higher than those in normal controls, whereas those of the MT metabolite 6-SM and of NE in urine tend to be lower. After T treatment, however, plasma MT levels fall significantly, whereas urinary levels of 6-SM and NE rise. Our data show that the effect of T is mediated by enhanced metabolism of MT, not by any effect on net sympathetic outflow, and that the increase in plasma MT in untreated KS is a consequence of alterations in the rate of MT metabolism and not of increased net sympathetic activity.
We are grateful to Ismail Hakki Ulus, M.D., Department of Pharmacology, Uludag University Medical School (Bursa, Turkey), for his critical review of the manuscript.