L-5-HYDROXYTRYPTOPHAN DOES NOT STIMULATE LH-SECRETION DIRECTLY FROM THE PITUITARY IN PATIENTS WITH GONADOTROPIN-RELEASING-HORMONE DEFICIENCY

OBJECTIVE There is abundant histological and physiological evidence th at serotonin plays a role in the regulation of LH secretion in rats. S tudies in human subjects have been few, but their results include the finding that pulsatile administration of L-5-hydroxytryptophan (5-HTP, the immediate precursor of serotonin) amplifies LH secretion in women in the medium-late follicular phase, and that this effect is not due to 5-HTP directly inducing LH secretion by the pituitary, We have inve stigated whether 5-HTP amplifies LH secretion by enhancing the respons e of the pituitary to GnRH. PATIENTS Seven patients aged 20-40years wi th hypogonadotrophic hypogonadism (HH) of hypothalamic origin (3 men w ith Kallmann's syndrome, 2 women without anosmia and with GH deficienc y, and 2 women with anorexia nervosa). DESIGN To prime the pituitary, subcutaneous pulsatile GnRH was administered for 7 days at the rate of one 5-20 mu g pulse every 90 min. The day before the investigation, t his regimen was replaced by 1.5-3 mu g intravenous pulses at the same frequency, On the day of the investigation, 3 ml blood samples were ta ken every 10 min from 0850 to 19.00 hours. After the first two samples , the intravenous GnRH pulse frequency was increased to one per hour a nd was maintained at this level throughout the rest of the study. The first 4 h of the study acted as a control phase allowing determination of the pituitary response to GnRH, At 1300 h, 75mg of the aromatic-1- amino-acid decarboxylase inhibitor carbidopa was administered orally; carbidopa does not cross the blood-brain barrier, and prevents periphe ral conversion of 5-HTP to serotonin, At 1600 h, another 75 mg dose of carbidopa was administered, and administration of 8-20 mg pulses of 5 -HTP at a rate of one pulse per hour was begun. MEASUREMENTS LH was de termined in triplicate by an immunoradiometric assay (IRMA), and LH pu lses identified by means of a program developed in our laboratory. RES ULTS When pulsatile administration of GnRH was accompanied by administ ration of carbidopa and 5-HTP, LH pulse amplitude (2.32 +/- 0.71 IU/l) did not differ significantly from its value in either the GnRH + carb idopa phase (2.58 +/- 1.12 IU/l) or the unaccompanied GnRH phase (2.77 +/- 1.76 IU/l), CONCLUSIONS L-5-hydroxytryptophan-induced amplificati on of LH secretion in humans is not due to enhancement of the pituitar y response to GnRH, The effect of L-5-hydroxytryptophan must therefore be due to its action on the hypothalamus, where it may be hypothesize d that it increases GnRH release.