Electrophysiological evidence for the activation of supraoptic neurones during the release of oxytocin.

1. Antidromically identified supraoptic (SO) units were recorded in lactating rats anaesthetized with urethane (1.1 g/kg), and their activity was studied during milk ejection evoked by the suckling of the young. Fifty-eight SO units were recorded through 174 milk ejections. Each milk ejection was the result of a neurohypophysial release of an oxytocin pulse of 0.5-1.5 m-u.2. Fifty-five of the SO units displayed background activity and three were silent. The firing rates ranged from 0 to 15.4 spikes/sec, the distribution was exponential with 26% of the units firing at < 1 spike/sec. Sixteen (28%) of the SO units displayed a phasic pattern of activity characterized by periods of activity (6-132 sec) and silence (4-71 sec).3. Twenty-five of the non-phasic units displayed a large and stereotyped acceleration in spike activity some 10.5-17.4 sec before the rise in intramammary pressure at milk ejection. Units accelerated to rates between 9-66 spikes/sec, an increase of about thirtyfold (median) on background activity. The response was brief (0.9-4.7 sec) and was followed by a period of after-inhibition.4. It was concluded from studies of double recordings and observations of multi-unit activity that all the responsive units were synchronously activated. The mean latency of 13.3 sec between the onset of the neurosecretory response and milk ejection was similar to that observed following the experimental release of endogenous oxytocin by electrical stimulation of the neurohypophysis (50 pulses/sec for 2-4 sec).5. Four of the phasically active units were correlated with the oxytocin release for milk ejection. Three of these units displayed a burst of activity superimposed on the terminal portion of an active phase, some 10.2-14.7 sec before milk ejection. The fourth unit, recorded in conjunction with a responsive non-phasic unit, consistently switched from silence to activity coincident with the onset of the SO activation.6. The remaining SO units and a further ten units that were not antidromically activated by neurohypophysial stimulation displayed no change in activity during either the period of neurosecretory activation or the period of after-inhibition.7. This activation of the SO neurones, in the formulation of oxytocin release and milk ejection, is the same as that we have previously observed in recordings from the paraventricular (PV) region, and the proportion of neurones displaying the response is similar: 48% in the SO nuclei, 58% in the PV nuclei. We conclude, since the SO nuclei contain 80% of the neurosecretory cells that project to the neurohypophysis, that the SO nuclei are as important, if not more so, than the PV nuclei in the control of oxytocin release.