Metabolic interfaces between growth and reproduction. V. Pulsatile luteinizing hormone secretion is dependent on glucose availability.

To test the hypothesis that mechanisms controlling the secretion of LH are modulated by glucose availability, the acute effects of glucoprivation were studied. The model was the gonadectomized male lamb raised on a limited diet of artificial milk. The approach was to monitor LH secretion before and after the administration of a competitive antagonist of glucose metabolism, 2-deoxyglucose (2DG). We first determined whether LH secretion was influenced by glucose availability by administering 2DG at several doses. Peripheral administration of the glucose antagonist (240 and 480 mg/kg 2DG, single iv injection) transiently decreased LH pulse frequency, but not LH pulse amplitude. By contrast, LH secretion (frequency or amplitude) was not affected by lower doses (60 or 120 mg/kg) of the glucose antagonist. A second study was conducted to determine whether either the pituitary gland or the GnRH neurosecretory system per se is directly affected by short term glucoprivation. The competency of the pituitary was assessed by administering GnRH during the time when LH secretion is suppressed by pharmacological glucose blockade. Similarly, the function of the GnRH neurosecretory system was assessed by administering a GnRH secretagogue (N-methyl-D,L-aspartate) under the same glucoprivic conditions. In response to an optimized iv dose of 2DG, LH pulse frequency decreased. However, in lambs that received either GnRH or N-methyl-D,L-aspartate during the period of glucoprivation, LH pulse frequency was sustained at levels comparable to those before 2DG was given. To determine whether the effect of glucoprivation was central in origin, the glucose antagonist was administered into the lateral cerebral ventricle at 1/100th the doses used peripherally. Central administration of 2DG, independent of dose, transiently decreased LH pulse frequency, but not pulse amplitude. However, unlike the case with peripheral injection, plasma glucose values did not change after the administration of any dose of 2DG tested centrally. These findings indicate that glucose availability in the developing sheep influences LH secretion. Moreover, based upon analysis of LH pulse frequency, glucoprivation does not directly impair either the pituitary gland or the GnRH neurosecretory system. Collectively, these results suggest that glucose availability affects LH secretion by acting within the central nervous system at a detection site(s) peripheral to the GnRH neuron.