Neuroendocrine self-control: dendritic release of vasopressin.

The past two decades have produced many new insights into the cellular mechanisms that regulate the firing patterns of hormone-secreting magnocellular neurons in the supraoptic and paraventricular nuclei of the hypothalamus. These firing patterns determine how much vasopressin or oxytocin is released into the systemic circulation. One such insight is the realization that these peptides are released not only from their axon terminals in the neurohypophysis, but also locally from their dendrites within the supraoptic or paraventricular nuclei (1, 2). These locally released peptides then act back on magnocellular neurons to influence their firing rate and thereby affect the secretion of hormone into the systemic circulation. In the case of vasopressin, the locally-released peptide inhibits the firing rate of vasopressin-secreting neurons, and hence their output of hormone into the bloodstream (2, 3). The article by Gillard et al. (4) in the current issue further advances our knowledge of the cellular mechanisms that control the local release of vasopressin within the rat supraoptic nucleus. Two aspects of this intranuclear release need to be kept in mind. First, dendritic release of vasopressin is uncoupled from the overall cellular electrical activity that determines action potential firing rate and thence the release of hormone into the circulation (2). Second, although it also responds to osmotic (and probably other) stimulation, dendritic vasopressin release occurs with a completely different timecourse from that of neurohypophysial release into the circulation: dendritic release lags hours behind systemic release (2, 5). Thus, questions arise concerning why there is a long delay for local release of vasopressin and what the physiological role is of this dendritic release of vasopressin. Although it has been known for more than 20 yr that there is local release of vasopressin and oxytocin within the supraoptic nucleus (6), it is only in recent years that the molecular and cellular mechanisms that mediate this intranuclear release have begun to be understood. As background to the work of Gillard et al., dendritically released vasopressin is known to act back, via V1a receptors, on magnocellular neuron excitability, inhibiting spiking activity and thus systemic release (2, 5). However, the vasopressin released by dendrites exerts a positive feedback on further dendritic vasopressin release. This mechanism is independent of spike activity, but dependent on intracellular calcium levels (7, 8). In contrast to the V1a receptor mechanism that mediates the inhibitory effect of vasopressin on its systemic release, Gillard et al. (4) show that the positive feedback action of vasopressin on its dendritic release is mediated mainly by V2 receptors on magnocellular neurons, consistent with evidence of V2 or V2-like receptor mediation of increased intracellular calcium by vasopressin action on magnocellular neurons (7). The current paper by Gillard et al. also addresses the role of other local paracrine mediators in the control of dendritic vasopressin release, under both basal and osmotically stimulated conditions, the osmotic challenge being administered systemically to live rats before their supraoptic nuclei were explanted for in vitro investigation. In particular, they studied the role of nitric oxide. Previously, neuronal nitric oxide synthase has been shown to be abundantly expressed in magnocellular neurons of the supraoptic and paraventricular nuclei (9, 10), and local application of nitric oxide inhibits neuronal excitability (via GABAergic mechanisms) and therefore hormone release into the circulation (11). However, Gillard et al. (4) produce evidence that both nitric oxide and glutamate promote the release of vasopressin in the supraoptic nucleus. They show that prior osmotic challenge increases the release of vasopressin, nitric oxide, glutamic acid, and aspartic acid from punches of the supraoptic nucleus. Working on the hypothesis that nitric oxide is likely to play a paracrine role in the dendritic release of vasopressin, they go on to show that osmotically stimulated local release of arginine vasopressin is blocked by inhibition of nitric oxide synthase or scavengers of nitric oxide. Intriguingly, the osmotically generated dendritic vasopressin release was also blocked by kynurenate, an inhibitor of ionotropic excitatory amino acid transmission, but in this case the release of nitric oxide was not blocked. Therefore, they propose that nitric oxide and glutamate are released within the supraoptic nucleus in response to osmotic stimulation, with the glutamate step being downstream to the paracellular action of nitric oxide in the stimulation of vasopressin release from dendrites (4). It is unclear, however, exactly how glutamatergic transmission fits into the scheme. What is the source of the glutamate that is released from the explanted supraoptic nucleus after in vivo osmotic challenge? It is unlikely to be the severed axon terminals of extranuclear input neurons. It could be from glia, but perhaps the most likely source is the dendrites of magnocellular neurons themselves (12). If so, dendritic release of both glutamate and vasopressin would have positive feedback actions to promote further dendritic vasopressin release. The ability of either kynurenate or V2 antagonists to block such release suggests a complex interaction that we do not yet understand. In terms of the physiological significance of dendritic vasopressin release, the authors speculate that these mechanisms promote efficient but not exhaustive secretion of vasopressin during osmotic challenge (4). Hard data on this interpretation are not yet available, but it is possible that such conservation mechanisms are at play during the declining Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community. 0013-7227/07/$15.00/0 Endocrinology 148(2):477–478 Printed in U.S.A. Copyright © 2007 by The Endocrine Society doi: 10.1210/en.2006-1531