Fasting suppresses pulsatile luteinizing hormone (LH) secretion and enhances orderliness of LH release in young but not older men

Pulsatile gonadotropin secretion and sex-steroid concentrations are suppressed reversibly in young fasted or malnourished human subjects. In this study, we investigated the impact of age on the dynamic neuroendocrine mechanisms underlying this stress response in healthy young (age, 28 ± 3 yr, n = 8) vs. older (age 67 ± 2 yr, n = 8) men with similar body mass indices (mean, 26 ± 0.6 vs. 26 ± 1.3 kg/m², respectively). Serum LH concentrations were measured by immunoradiometric assay (IRMA) in blood collected at 10-min intervals over 27 h on a control (fed) day and on the third day of a 3.5-day fast (water only) assigned in randomized order. After 24 h of basal sampling, GnRH (10 μg iv bolus) was administered to test gonadotrope responsiveness. Cortisol, dehydroepiandrosterone sulfate, androstenedione, testosterone, FSH, GH, and PRL were measured in 24-h pooled serum as positive and negative control hormones. Approximate entropy was used to quantitate the orderliness of LH release over 24 h, and a multiple-parameter deconvolution method was applied to quantify pulsatile LH secretion and LH half-life. In the fed state, older men exhibited elevated mean (24-h pooled) serum FSH and cortisol concentrations compared with young controls but equivalent serum LH concentrations and reduced serum GH, free testosterone, androstenedione, and dehydroepiandrosterone sulfate concentrations. Fed older men also manifested a lower frequency and amplitude of 24-h pulsatile LH secretion, and, by approximate entropy calculations, a more disorderly pattern of basal LH release than younger individuals. Three- and one-half days of fasting evoked 40% and 47% increases in mean (24-h) serum cortisol concentrations in young and older men, respectively (P < 0.01 vs. fed, but P = not significant for percentage rise in older vs. young men). Concurrently, fasting induced a 2.1- fold fall in the 24-h endogenous LH production rate in young men (fed 36 ± 9.7 vs. fasted 17 ± 2.0 IU/L of distribution volume/day, P < 0.01), but did not significantly affect the daily LH secretion rate in older men (fed 27 ± 4.5 vs. fasted 21 ± 3.4 IU/day). The reduced LH production rate in fasting young men was accounted for by a 1.7-fold decline in the mass of LH secreted per burst (fed 2.5 ± 0.45 vs. fasted 1.5 ± 0.16 IU/L, P < 0.05), whereas LH burst mass in older men remained unchanged (and low) during fasting. In addition, in young men, during the 3.5-day fast the number of computer- resolved LH secretory bursts per 24 h decreased (fed 15 ± 0.7 vs. fasted 11 ± 0.7, P < 0.01), and the interburst interval increased (fed 94 ± 4.2 vs. fasted 125 ± 8.7 min, P < 0.05). In contrast, in older men in the fed state, basal LH peak frequency and serum free testosterone concentrations were reduced compared with corresponding values in young men, and did not decline further with fasting. Whereas the orderliness of LH release patterns increased significantly during fasting in the young men, the approximate entropy measure failed to change significantly in unfed older subjects. By cosinor analysis, young men showed lower 24-h mesor (mean of nyctohemeral rhythm off serum LH concentrations than older volunteers during fasting. Moreover, young but not older men manifested preserved 24-h variations in LH interpulse intervals when fasting. Exogenously stimulated LH release (mean 3- h serum LH concentration or calculated mass of LH secreted) following a single iv injection of 10 μg GnRH was independent of age and fasting status. We conclude that the metabolic stressor of short-term fasting unmasks specific age-related neuroendocrine contrasts in the stress-responsive control of both the pulsatile and nyctohemeral regulation of the male hypothalamo-pituitary-gonadal-axis.