Abstract
Several studies have demonstrated that the rat isolated mediobasohypothalamus can release gonadotropin-releasing hormone (GnRH) in a pulsatile manner in vitro. We have now used an in vitro perifusion system to characterize GnRH release from the adult male Sprague-Dawley rat isolated median eminence (ME) alone. After 1 h stabilization periods, eight MEs released GnRH in an episodic pattern during 4-hour experiments (samples collected every 4 min, RIA in triplicate) with mean (± SE) release rates of 2.2 ± 0.2 pg GnRH/4 min and peak amplitudes of 1.3 ± 0.2 pg/4 min. The intervals between significant peaks were variable, although 70% were in the range of 12-24 min, i.e. similar to the non-gonadal steroid-modulated frequency of pulsatile luteinizing hormone secretion in gonadectomized rats in vivo. During nine additional experiments, MEs were perifused with control medium (containing 2.5 mM calcium) through the first hour of sample collection, then either calcium-free medium containing 0.5 mM of the calcium chelator EGTA and 100 µM of the intracellular calcium antagonist TMB-8 or calcium-free medium containing just 100 µM TMB-8 for 2 h, followed by a final hour with control medium. Removal of calcium activity decreased the frequency of GnRH peaks from 3.2 ± 0.3 to 1.3 ± 0.2/h, which then increased to 2.3 ± 0.3/h with subsequent replacement of control medium. Removal of available calcium also decreased mean GnRH release by 51 ± 11 %, which returned to basal levels with subsequent calcium replacement. These results suggest that either (1) the most fundamental unit of the GnRH pulse-generating mechanism is expressed even at the level of isolated ME axons and secretory terminals, or (2) the terminals of transected GnRH-secreting neurons may be inherently unstable and that there is some mechanism for coordination of pulses of GnRH release located within the ME at the level of the nerve terminals or axons.