Vasopressin and oxytocin action in the brain:Cellular neurophysiological studies

I.J.A. Urban, J.P.H. Burbach and D. De Wied (Eds.) Progress in Brain Research, Vol. 119 0 1998 Elsevier Science B.V. All rights reserved.

263

CHAPTER 3.2.1

Vasopressin and oxytocin action in the brain: Cellular neurophysiological studies M. Raggenbass", S. Alberi', M. Zaninetti, P. Pierson, J.J. Dreifuss Department of Physiology, University Medical Center, I , rue Michel Server, CH-1211 Geneva 4, Switzerland

During the last two decades it has become apparent that vasopressin (VP) and oxytocin (OT), in addition to playing a role as peptide hormones, also act as neurotransmitters. Morphological studies and electrophysiological recordings have shown a close anatomical correlation between the presence of these receptors and the neuronal responsiveness to VP or OT. These compounds have been found to affect membrane excitability in neurons located in the hippocampus, hypothalamus, lateral septum, brainstem, spinal cord and superior cervical ganglion. Sharp electrode intracellular and whole-cell recordings, done in brainstem motoneurons, have revealed that VP and OT can directly affect neuronal excitability by opening non-specific cationic channels. These neuropeptides can also influence synaptic transmis-

sion, by acting either postsynaptically or upon presynaptic target neurons or axon terminals. Whereas in some hypothalamic neurons OT appears to mobilize intracellular calcium, as revealed by calcium imaging techniques, in the brainstem the action of this neuropeptide is mediated by a second messenger which is distinct from the second messenger activated in peripheral target cells. Future studies should be aimed at elucidating the properties of the cationic channels responsible for the neuronal action of VP and OT, at identifying the brain-specific second messengers activated by these neuropeptides and at determining whether endogenous VP and OT can exert neuronal effects similar to those elicited by exogenous neuropeptides.

Introduction

OT may play a role in brain function, since in situ injection of VP and OT antagonists can interfere with behavior or physiological regulations; (4) specific binding sites, i.e., membrane receptors having high affinity for VP and OT are present in the central nervous system; (5) these receptors, or at least part of them, are localized on neurons, since application of exogenous VP and OT alters the rate of firing of single neurons present in regions where binding sites have been detected autoradiographically. The neuronal actions of VP and OT have been investigated initially following microinjection and microiontophoresis in situ and more recently in vitro. Studies done up to the early 1990s are reviewed in Argiolas and Gessa (1991); De Kloet et al. (1990); De Wied et al. (1993); Dreifuss and Raggenbass (1993); Raggenbass et al. (1995) and Richard et al. (1991). Here we summarize more recent work. We concentrate upon studies carried

Vasopressin (VP) and oxytocin (OT) are peptide hormones which act on a variety of target organs, including kidney, smooth muscle, liver and anterior pituitary. During the last two decades it has become apparent that these two peptides may in addition act as neurotransmitters. A number of arguments supports this conjecture: (1) VP and OT are not only synthesizedin hypothalamo-neurohyphy sial cells, but also in other hypothalamic and extrahypothalamic cell bodies whose axon projects to the limbic system, the brainstem and the spinal cord; (2) VP and OT can be shed from central axons as are classical neurotransmitters; (3) central VP and *Corresponding author. Tel.: 41-22-702-5386; fax: 4 1 22-702-5402; e-mail: mario.raggenbass@medecine. unige.ch. 'Present address: Novartis Pharma AG, CH-4002 Basel, Switzerland.

264

out at the cellular level, i.e., restrict our review to in vitro systems. VP and OT are treated together. Indeed, although acting via distinct receptors in distinct brain areas, these two neuropeptides appear to exert similar effects upon neuronal excitability.

Mapping neuropeptide sensitivity within the brain Light microscopic autoradiography has revealed the presence of V1,-type VP and OT receptors in selected regions of the central and autonomic nervous system (for reviews, see Barberis and Tribollet, 1996; Zingg, 1996). Electrophysiological recordings, done mostly in brain slices of the rat, have shown a close anatomical correlation between the presence of these receptors and the neuronal responsiveness to VP or OT (for a possible exception, however, see the following section). As expected from the distribution of receptors, these neuropeptides have been found to affect excitability in neurons located in the hippocampus, hypothalamus, lateral septum, brainstem, spinal cord and superior cervical ganglion. New results have been recently obtained in the hypothalamus, brainstem and spinal cord. Hypothalamus and limbic system

Studies in lactating rats suggest that OT may influence the milk-ejection reflex by acting in the bed nucleus of the stria terminalis (BST). Unit recordings, obtained from neurons located in the BST in hypothalamic slices from lactating rats, showed that about half of the neurons were excited by OT (Ingram et al., 1990). The proportion of responsive cells was unaltered during the peripartum period, but the sensitivity to the neuropeptide increased significantly during lactation (Ingram and Wakerley, 1993). In some OT-responsive neurons, orthodromic activation following hypothalamic electrical stimulation could be reversibly attenuated by an OT antagonist, suggesting the existence of an OTergic innervation of the BST (Ingram and Moos, 1992). Neurons in the dorsomedialdivision of the suprachiasmatic nucleus (SCN) synthesize VP,in addition, VP mRNA transcription displays intrinsic

rhythmicity. By acting within the SCN, VP may participate in the regulation of the circadian cycle. VP did indeed activate cells in the SCN in vitro. Interestingly, a V1, antagonist reduced the spontaneous basal activity in VP-sensitive neurons, a fact suggestive of an endogenous excitatory VPergic input (Mihai et al., 1994a,b). However, the basal activity of VP-responsive neurons showed a marked circadian activity in both heterozygous (VP-containing) and homozygous (VP-deficient) Brattleboro rats, indicating that VP was not required for the generation of this circadian pattern of activity (Ingram et al., 1996). In the ventromedial hypothalamus (VMH) of the rat, OT binding, as detected by autoradiography, is affected by modifications of the circulating gonadal steroid hormones. Castration drastically reduces this binding in animals of either sex, whereas gonadal steroid injection can restore it to normal levels. Accordingly, VMH neurons in hypothalamic slices from ovariectomized rats could be excited by OT, acting on OT receptors, and neuronal responsiveness increased following slice treatment with progesterone (Kow et al., 1991). However, the steroid dependence of OT receptors in the VMH appears to be species dependent, since in the guinea pig neither binding sites for OT nor neuronal responses to this neuropeptide were affected by gonadectomy (Inenaga et al., 1991; Tribollet et al., 1992). Central administration of OT can depress prolactin secretion and part of this effect may be mediated by a neuropeptide-induced activation of tuberoinfundibular dopaminergic neurons. Consistent with this conjecture, OT, at nanomolar concentrations, could increase the excitability of a majority of dorsomedial arcuate neurons, which may include tuberoinfundibular neurons (Yuan and Pan, 1996). In brain slices containing the subfornical organ (SFO), VP caused either an increase or a decrease in the discharge rate of angiotensinII-sensitive neurons. Whereas the excitatory effect was direct, the inhibitory effect was synaptically mediated and both were suppressed by a V1, receptor antagonist (Anthes et al., 1997). Since the SFO is a brain region devoid of a blood-brain barrier, circulating VP may influence the activity of SFO neurons. Functional V1, receptors have also been detected

265

in neurons of the central amygdaloid nucleus. In this brain region, which contains a high level of VP binding sites, VP could indeed exert a direct excitatory action 0.u et al., 1997). Electrophysiological recordings, performed in vivo and in vitro, indicate that VP and OT can alter the discharge rate of hypothalamic magnocellular neurons in the paraventricular and supraoptic nuclei (see, for example, Richard et al., 1991). In hypothalamic slices from virgin female rats, non-phasically firing neurons in these nuclei were inhibited by OT, whereas in slices from male or ovariectomized female animals, this neuropeptide had either no effect or caused excitation (Kuriyama et al., 1993). One should point out, however, that the presence of VP and OT receptors in the paraventricular and supraoptic nuclei is still an unresolved issue, different groups having reported conflicting results (Barberis and Tribollet, 1996). Thus, one cannot exclude that the electrophysiological effects cited above were exerted indirectly. In lactating rats, OT supraoptic neurons respond to suckling by generating synchronous bursts of action potentials. In an attempt to elucidate the mechanism of this intermittent neuronal activation, intracellular recordings were done in hypothalamic organotypic slice cultures. In identified oxytocinergic neurons, the bursting activity, when present, was similar to that described in situ, though the interburst interval was much shorter. OT, acting on OT receptors, reduced the interburst interval in spontaneouslybursting neurons and triggered bursting activity in some non-bursting neurons. Neither the spontaneous nor the OT-induced burst firing were due to intrinsic membrane properties of magnocellular neurons, but were dependent upon volleys of afferent excitatory postsynaptic potentials (EPSPs). This suggests that OT magnocellular neurons may be part of a hypothalamic rhythmic network, which is driven by glutamatergic synaptic transmission and which can be modulated by OT (Jourdain et al., 1998). In acutely dissociated neurons from the horizontal limb of the diagonal band of Broca, VP caused a slight increase in an outward current (Easaw et al., 1997). This effect, which was detectable only at strongly depolarized potentials, was dependent

upon extracellular calcium, was blocked by charybdotoxin and was mediated via V1, receptors. In some neurons, however, VP caused a reduction, rather than an increase, in this outward current. This latter effect was suppressed by a Vz receptor antagonist, an intriguing result in view of the fact that there is compelling evidence that brain VP receptors are of the V1,-type. Brainstem

In brainstem slices containing the dorsal vagal nucleus, a high proportion of neurons could be excited not only by OT, as found in our laboratory (see below), but also by VP, acting on distinct receptors (Mo et al., 1992). However, part of the recordings obtained by these authors may have been from adjacent solitary tract neurons, which are almost exclusively endowed with VP receptors (Raggenbass et al., 1989). In the dorsal vagal complex, the neuronal sensitivity to OT was reduced following estradiol pretreatment of the animals, and was slightly enhanced by progesterone, suggesting that OT responsiveness in this brain region may change during the estrus cycle (Tolchard and Ingram, 1993). In this same preparation, a V1, antagonist suppressed the response to VP but did not block completely the effect of vasotocin, providing suggestive evidence for a class of receptors distinct from either the OT or V1, subtypes (Ingram and Tolchard, 1994). Circulating VP is thought to modulate the sensitivity of the baroreceptor reflex by acting in the area postrema, which is located on the dorsal surface of the medulla and is outside the blood-brain barrier, whereas central VP may exert similar effects by modulating the activity of neurons in the nucleus of the solitary tract. Indeed, VP could affect the discharge rate of area postrema neurons in brainstem slices from the rat (Lowes et al., 1995) and the rabbit (Cai and Bishop, 1995). The neuropeptide action was mainly excitatory and, in the rat, it was at least in part direct and was mediated by VP receptors. In addition, by acting in area postrema (Cai et al., 1994), or upon vagal afferent neurons located in the nodose ganglion (Gao et

266

al., 1992), VP could indirectly influence the firing of solitary tract neurons. Spinal cord

Intracellular recordings obtained from sympathetic preganglionic neurons in transverse slices of spinal cord indicated that VP and a V1, agonist could cause cell depolarization, and that this effect was blocked by a V1, antagonist (Sermasi and Coote, 1994). However, some sympathetic preganglionic neurons could be shown to possess functional OT receptors, whose activation increased membrane excitability (Desaulles et al., 1995). Thus, neurons located in the intermediate-lateral nucleus of the thoraco-lumbar spinal cord appear to be heterogeneous in their sensitivity to neurohypophysial peptides.

Membrane mechanism of neuropeptide action The data reviewed so far suggest that VP and OT can affect the excitability of neurons located in various areas of the brain and spinal cord and may potentially subserve a variety of functions. However, with the exception of the studies done in hypothalamic organotypic slices and spinal cord, all these results were obtained using extracellular recordings and thus suffer from some limitations: the demonstration that the neuropeptideinduced effects are direct, rather than synaptically mediated, is diacult to obtain, the localization of the recorded neurons may be uncertain and no information can be deduced concerning the membrane mechanism of action of the neuropeptides. More recently, the membrane mechanism of action of VP and OT started to be studied. To this end, one needs to get stable sharp electrode intracellular recordings from identifiable neurons responsive to these neuropeptides. We met such conditions by using brainstem slices containing the facial nucleus, the hypoglossal nucleus and the dorsal motor nucleus of the vagus nerve. Facial nucleus

VP was found to increase the excitability of antidromically identified facial (VII) motoneurons

(Tribollet et al., 1991). The membrane mechanism responsible for this effect was characterized using the single-electrode voltage-clamp technique. The neuropeptide acted by generating a persistent inward current, which was voltage-gated, tetrodotoxin-insensitive and sodium-dependent. Blockade of transmembrane calcium currents or partial substitution of chloride ions by isethionate did not significantly alter the VP-induced current. It was not affected by a two-fold decrease in the transmembrane potassium gradient and was not modified by a variety of potassium channel blockers (Raggenbass et al., 1991). Current-voltage relationships, obtained from cesium-loaded motoneurons, showed that the VP current reversed in polarity at around 0 mV, suggesting that it was a nonspecific cationic current. Reducing the extracellularcalcium concentration caused a reversible increase in the amplitude of the VP current. Lowering the extracellular magnesium also increased this current, but less efficiently. These data indicate that divalent cations can modulate the VP response and suggest that in the normal physiological solution, which contains 2 mM calcium and 1 mM magnesium, the VP current is partially blocked (Alberi et al., 1993). By inducing a persistent voltage-dependent inward current, vasopressin can affect the inputoutput relationship of facial motoneurons. Indeed, by using whole-cell recordings in the current-clamp configuration, we have recently assessed that in the presence of the peptide the current input required to attain the firing threshold was decreased and the frequency-current relationship was shifted to the left. This suggests that descending vasopressinergic pathways of hypothalamic origin may modulate the motor output by enhancing brainstem motoneurons excitability. In transgenic mice overexpressing the protein Bc12, axotomy-induced neuronal death of neonatal facial motoneurons is prevented, as assessed by morphological criteria. However, the functional properties of these surviving motoneurons are unknown. To clarify this issue, we have carried out whole-cell patch clamp recordings in brainstem slices of mice containing the facial nucleus (Alberi et al., 1996). We found that axotomized motoneurons in transgenic animals had properties similar to

267

those of intact motoneurons. They fired repetitively following positive current injection and, under voltage clamp conditions, they responded to ionotropic glutamate receptor agonists (cf. Widmer et al., 1992) as well as to VP by generating sustained inward currents. However, cell input resistance was much higher in axotomized motoneurons, indicating that they were smaller in size - an observation which was consistent with the morphological data. Thus, lesioned surviving facial motoneurons in transgenic mice appear to be endowed with functional receptors to neurotransmittersheuromodulators. Hypoglossal nucleus

Autoradiography revealed the presence of VP receptors in the ventromedial and dorsal division of the hypoglossal (XII) nucleus and VP was found to generate a sustained inward current in a majority of hypoglossal neurons. Antidromic activation, following electrical stimulation of nerve XII axons, or morphological characterization of biocytin-labelled neurons, indicated that at least part of the VP-sensitive cells were motoneurons. When synaptic transmission was blocked by perfusing the slice with a low-calciumhigh-magnesium solution, the average peak amplitude of the VP-induced current decreased by 65%. Following tetrodotoxin treatment, this current decreased by a similar extent. In contrast, in a low-calciudnormalmagnesium concentration, i.e., in conditions of reduced synaptic transmission but of increased neuronal excitability, the VP current was not significantly altered. Thus, the action of VP was probably in part direct and in part presynaptic and the latter effect was dependent upon action potential propagation. Current-voltage relationships indicated that the inward current responsible for the postsynaptic effect reversed in polarity at around -15 mV, suggesting that it was carried by both sodium and potassium ions (Palouzier-Paulignan et al., 1994). Dorsal motor nucleus of the vagus nerve

The dorsal motor nucleus of the vagus nerve (X) in the rat contains OT binding sites (Dreifuss et al., 1988). In early experiments it was found that OT could increase the excitability of a large proportion

of neurons located in this nucleus (Charpak et al., 1984). Using morphological and electrophysiological criteria, some of the OT-responsive vagal neurons could be identified as being parasympathetic preganglionic motoneurons (Raggenbass et al., 1987; Tribollet et al., 1989; Dubois-Dauphin et al., 1992). The mechanism of action of OT was investigated in voltage-clamped vagal motoneurons. OT evoked a tetrodotoxin-insensitive, noninactivating inward current whose peak amplitude was concentration-related. The OT current-voltage curve contained a region of negative slope conductance. Partial replacement of extracellular sodium reversibly attenuated or suppressed the neuropeptide current, whereas substitution of extracellular chloride or blockade of calcium currents did not modify it. Neither a decrease in the transmembrane potassium gradient nor any of several potassium channel blockers affected the OT current. Lowering the extracellular calcium concentration caused a reversible enhancement of the response to OT (Raggenbass and Dreifuss, 1992). These results indicate that OT excites vagal motoneurons by inducing a sustained voltage-gated inward current which is sodium-dependent and is modulated by calcium.

Characterization of neuropeptide-activated second messengers Peripheral VP and OT receptors have been recently cloned in a variety of species. They belong to the G protein-coupled receptor family and possess seven hydrophobic transmembrane segments, connected by alternating extracellular and intracellular loops (for reviews, see Barberis and Tribollet, 1996; Zingg, 1996). In addition, the second messengers activated by VP and OT in peripheral target cells have been well characterized: while V1, and OT receptors are coupled to a phospholipase C-/3 (PLC-p), V2 receptors are linked to an adenylyl cyclase. By contrast, central receptors for VP and OT have not yet been cloned and the second messenger(s) involved in the central action of these peptides is (are) not yet known. Recently, we have begun to characterize this messenger by using whole-cell recordings in brainstem slices containing OT-sensitive vagal neurons.

268

When loaded with GTP-y-S,a non-hydrolyzable analogue of GTP, vagal neurons generated a persistent inward current in the absence of agonist; however, the OT effect was suppressed, suggesting that the neuropeptide-evoked current was mediated by G protein activation (Fig. 1). Loading vagal neurons with the calcium chelator, BAPTA, suppressed a calcium-dependent slowly decaying potassium aftercurrent, Zm, but did not affect the OT response, suggesting that the latter was not mediated by an agonist-induced increasein the intracellular free calciumconcentration.Protein kinase C (PKC) activation was probably not involved in the neuronal effect of OT, since the neuropeptideevoked current was not modified by loading neurons with PKC inhibitors (Raggenbasset al., 1995;Alberi et al., 1997).Thus, OTreceptorsin vagal neurons are probably not functionally coupled to a PLC-p. We are presently investigating whether at least part of the OT response may be mediated by a CAMPdependent intracellular pathway. Calcium imaging studies have recently revealed that in cultured rat supraoptic cells, VP and OT

A

induced a significant increase in the intracellular calcium concentration. While the OT effect was exclusivelydue to calcium released from thapsigargin-sensitive stores, the VP effect requkd calcium influx from the extracellular medium, mainly through L-, N- and T-type calcium channels (Lambert et al., 1994;Dayanithi et al., 1996; Sabatier et al., 1997).VP- and OT-induced calcium transients have also been observed in cells derived from the organum vasculosum of the lamina terminalis and the subfornical organ (Jurzak et al., 1995) and VP, as well as angiotensin II, could increase the intracellular free calcium concentration in cultured neurons from the area postrema (ConsolimColombo et al., 1996). In both systems, the neuropeptide effect persisted in the absence of extracellular calcium. Neuropeptide modulation of synaptic transmission

In addition to directly affect the membrane ionic permeability in selected neuronal populations, VP

C Oxytocin

1 5 0 PA 1 min

B Oxytocln

c I

NMDA

1 min

Control n=8

GTW n=7

Fig. 1. The inward current generated by OT in vagal neurons is mediated by G protein activation. The left panel shows wholecell voltage clamp recordings of membrane currents obtained in a control neuron (A) and in a neuron loaded with GTP-7-S (B). Compoundswere added to the perfusion solution for the 1.5-min period representedby the horizontal bar above each trace. The OT concentration was 0.2 pM in (A) and 1 pM in (B); the NMDA concentration was 50 pM. Note, in (B), that a persistent activation of G protein suppressed the OT but not the NMDA current. (C) Average inward current evoked by OT, at 1 pM,in control neurons and in GTP-y-S-loadedneurons. ***P< 0.001.

269

and OT can influence neurotransmission at some central synapses. We have recently uncovered an indirect effect of VP in the hypoglossal nvcleus, using whole-cell recordings. Under voltage clamp conditions, VP caused an increase in the frequency of spontaneous postsynaptic currents (PSCs) in a majority of the recorded neurons. These neuropeptide-sensitive PSCs were negative in polarity in neurons held at or near their resting membrane potential and reversed in polarity at around -50 mV, a value close to the chloride reversal potential (Fig. 2). In addition, in chloride-loaded neurons, these PSCs were amplified and were inward-going at all membrane potentials. The stimulatory effect of VP persisted in the presence of CNQX, MK-801 and bicuculline, suggesting that neither AMPN kainate nor NMDA nor GABAA receptors were involved. By contrast, it was suppressed by strychnine. We conjecture that, in addition to directly depolarize hypoglossal motoneurons (see above), VP may facilitate inhibitory synaptic transmission in the hypoglossal nucleus by acting upon the soma and/or axon terminals of putative glycinergic interneurons. In hypothalamic supraoptic neurons, OT reduced the amplitude of inhibitory postsynaptic currents Control 15 mV

-5mV -25 mV

-

I.

l l l L l

.’

-.

-45 mV

Vasopressin 0.1 FM L

h

u -

c.l”

-65 rnV

-85 mV

(IPSCs) mediated by GABAA receptors (Brussaard et al., 1996); this effect was probably due to a neuropeptide-induced increase in intracellular calcium, which in turn depressed inhibitory synaptic transmission via a postsynaptic mechanism. These results suggest that magnocellular neurosecretory neurons may be endowed with functional OT receptors (but see the section on the hypothalamus and limbic system) and that OT may control the activity of these neurons by disinhibiting them. OT was also found to reduce the amplitude of excitatory postsynaptic currents (EPSCs) evoked in supraoptic neurons following electrical stimulation of afferent fibers. A similar reduction in excitatory input was observed following high frequency stimulation of afferents as well as by depolarizing single neurons by current injection. Since all these effects could be suppressed by an OT antagonist, it was suggested that dendritically released peptides can reduce excitatory synaptic input to magnocellular neurons by acting on presynaptic receptors (Kombian et al., 1997). A modulatory effect of OT upon glutamatergic synaptic transmission has been evidenced in cultured neonatal spinal cord dorsal horn neurons (lo et al., 1998). OT, or a selective OT agonist, caused a reversible increase in the frequency, but

P-V-

I‘

v

300 PA

05s

Fig. 2. VP enhances inhibitory postsynaptic currents (IPSCs) in hypoglossal motoneurons. The traces are current records, obtained in the whole-cell configuration,at holding potentials ranging from -85 to 15 mV. Currents were measured either in the absence (left panel) or in the presence of 0.1 pM VP (right panel). Patch pipettes contained a low-chloride solution. Note that the VP-enhanced IPSCs reversed in polarity at around -45 mV, i.e., close to the chloride reversal potential.

+

270

not the amplitude, of MA-receptor-mediated spontaneous, miniature and EPSCs. This peptide action was presynaptic and was dependent upon extracellular calcium influx. It might represent a neuronal mechanism by which descending hypothalamo-spinal OTergic pathways modulate sensory, and in particular nociceptive, input. VP (Van den Hooff and Urban, 1990) as well as the C-terminal glycopeptide of the VP precursor (Van den Hooff et al., 1990) could slightly increase the amplitude of glutamate-mediated EPSPs elicited by stimulation of the fimbria axons. A slight, long-lasting enhancement of excitatory synaptic transmission could also be induced by VP and VP 4-8 in the CA1 field in hippocampal slices (Chepkova et al., 1995; Rong et al., 1993). These data suggest that VP, as well as some related neuropeptides, could influence synaptic plasticity. However, the pharmacological basis of this action is unclear, since the facilitatory effect of VP in the lateral septum was not suppressed by a V1,receptor antagonist and it is not known whether the neuropeptide site of action was pre- or postsynaptic. Interestingly, VP was found to enhance synaptic transmission at the frog neuromuscular junction. The neuropeptide acted presynaptically, by producing a long-lasting increase in spontaneous and evoked transmitter release; no postsynaptic effect could be detected at the frog neuromuscular junction (Abdul-Ghani et al., 1990).

Conclusions and perspectives Extracellular recordings have been useful in mapping the VP and OT sensitivity in the central and peripheral nervous system and in showing that at least part of the binding sites having high affinity for these neuropeptides represent functional receptors located on neurons. The use of more advanced methods (intracellular and whole-cell recordings, calcium imaging) has allowed workers to begin to unravel the mode of action of these neuropeptides at a more mechanistic level. Some features emerge from these studies: (1) VP and OT can directly modulate neuronal excitability by opening nonspecific cationic channels, mainly permeable to sodium (facial, hypoglossal and vagal neurons; Delmas et al., 1997). (2) The neuro-

nal second messengers activated by VP-and OTreceptor binding may be different from the second messengers stimulated by agonist-receptor interaction in peripheral targqt cells (cf. vagal versus supraoptic neurons). (3) VP and OT can exert powerful indirect effects by acting either postsynaptically (supraoptic nucleus) or upon presynaptic target neurons (hypoglossal nucleus). In spite of some advances, however, a number of basic questions remain open. (1) The properties of the cationic channels responsible of the excitatory effects of VP and OT need to be further explored, possibly down to the single-channel level. To this end, cultured neurons, retaining their responsiveness to VP and OT, are needed. (2) The second messengers mediating the neuronal action of VP and OT need to be extensively characterized. Besides the whole-cell approach, microfluorimetric imaging of second messengers (calcium, CAMP, etc.), carried out in brain slices or in neuronal cultures, would be useful. In this context, it would also be important to determine how, besides exerting short term electrophysiological effects, VP and OT can have long term actions upon gene expression, thus influencing cell growth or morphology. (3) The demonstration that endogenous VP and OT can exert postsynaptic effects similar to those elicited by exogenously applied neuropeptides is still lacking. Dual recordings, carried out under visual control, in brain nuclei containing VP- and OT-synthesizing neurons as well as high affinity binding sites for these peptides may be helpful in this respect. VP-deficient Brattleboro rats are viable, though they behave abnormally (Bohus and De Wied, this volume). However, they are viable for studies. In addition, two recent studies present evidence that mice lacking a functional gene coding for OT are viable and fertile, and do not present any major reproductive behavioral or functional anomaly. The only functional defect was the inability of the OT-deficient females to nurse their offspring (Nishimori et al., 1996; Young et al., 1996). These data, however, do not rule out a role for VP and OT in the brain. The existence of a great number of neurotransmitter systems suggests a high degree of functional redundancy in central signaling pathways. Thus, in VP- or OT-deficient

27 1

animals, alternative pathways may be called in action and substitute for the defective one. Acknowledgements This work was supported in part by the Swiss National Science Foundation (Grant 31.43436.95).

References Abdul-Ghani, M., Meiri, H. and Rahamimoff, R. (1990) Vasopressin produces long-lasting increase in transmitter release. Brain Res., 515: 355-357. Alberi, S., Dubois-Dauphin, M., Dreifuss, J.J. and Raggenbass, M. (1993) Modulation by divalent cations of the current generated by vasopressin in facial motoneurons. Brain Res., 624: 326-330. Alberi, S., Dreifuss, J.J. and Raggenbass, M. (1997) The oxytocin-induced inward current in vagal neurons of the rat is mediated by G protein activation but not by an increase in the intracellular calcium concentration. Eur. J. Neurosci., 9: 2605-2612. Alberi, S., Raggenbass, M., de Bilbao, F. and DuboisDauphin, M. (1996) Axotomized neonatal motoneurons overexpressing the Bc12 proto-oncogene retain functional electrophysiological properties. Proc. Natl. Acad. Sci. USA, 93: 3978-3983. Anthes, N., Schmid, H.A., Hashimoto, M., Riediger, T. and Simon, E. (1997) Heterogeneous actions of vasopressin on ANG 11-sensitive neurons in the subfornical organ of rats. Am. J. Physiol., 273: R2105-R2111. Argiolas, A. and Gessa, G.L. (1991) Central functions of oxytocin. Neurosci. Biobehav. Rev., 15: 217-231. Barberis, C. and Tribollet, E. (1996) Vasopressin and oxytocin receptors in the central nervous system. Crit. Rev. Neurobiol., 10: 119-154. Brussaard, A.B., Kits, K.S. and de Vlieger, T.A. (1996) Postsynaptic mechanism of depression of GABAergic synapses by oxytocin in the supraoptic nucleus of immature rat. J. Physiol. (Lond.), 497: 495-507. Cai, Y. and Bishop, V.S. (1995) Effects of arginine vasopressin and angiotensin II on area postrema neurons in rabbit brain slice preparation. Neurosci. Lett., 1 9 0 125128. Cai, Y., Hay, M. and Bishop, V.S. (1994) Stimulation of area postrema by vasopressin and angiotensin II modulates neuronal activity in the nucleus tractus solitarius. Brain Res., 647: 242-248. Charpak, S., Armstrong, W.E., Muhlethaler, M. and Drei-

fuss, J.J. (1984) Stimulatory action of oxytocin on neurones of the dorsal motor nucleus of the vagus nerve. Brain Res., 300: 83-89. Chepkova, A.N., French, P., De W i d , D., Ontskul, A.H., Ramakers, G.M.J., Skrebitski, V.G., Gispen, W.H. and Urban, I.J.A. (1995) Long-lasting enhancement of synaptic excitability of CAl/subiculum neurons of the rat ventral hippocampus by vasopressin and vasopressin(4-8). Brain Rex, 701: 255-266. Consolim-Colombo, F.M., Hay, M., Smith, T.C., ElizondoFoumier, M. and Bishop, V.S. (1996) Subcellular mechanisms of angiotensin 11 and arginine vasopressin activation of area postrema neurons. Am. J. Physiol., 271: R34R41. Dayanithi, G., Widmer, H. and Richard, P. (1996) Vasopressin-induced intracellular Ca2+ increase in isolated rat supraoptic cells. J. Physiol. (Lond.), 490: 713-727. De Kloet, E.R., Joels, M. and Urban, I.J.A. (1990) Central neurohypophyseal hormone receptors and receptor mediated cellular responses. In: D. De Wied (Ed.), Neuropeptides: Basics and Perspectives, Elsevier, Amsterdam, pp. 105-138. De Wied, D., Diamant, M. and Fodor, M. (1993) Central nervous system effects of the neurohypophyseal hormones and related peptides. Front. Neuroendocnnol., 14: 251-302. Delmas, P., Raggenbass, M. and Gola, M. (1997) Lowthreshold Na' currents: A new family of receptor-operated inward currents in mammalian nerve cells. Brain Res. Rev., 25: 246-254. Desaulles, E., Reiter, M.K. and Feltz, P. (1995) Electrophysiological evidence for oxytocin receptors on sympathetic preganglionic neurones: An in vitro study on the neonatal rat. Brain Rex, 699: 139-142. Dreifuss, J.J. and Raggenbass, M. (1993) Oxytocin-responsive cells in the mammalian nervous system. Regul. Peptides, 45: 109-1 14. Dreifuss, J.J., Raggenbass, M., Charpak, S., DuboisDauphin, M. and Tribollet, E. (1988) A role of central oxytocin in autonomic functions: Its action in the motor nucleus of the vagus nerve. Brain Res. Bull., 20: 765770. Dubois-Dauphin, M., Raggenbass, M., Widmer, H., Tribollet, E. and Dreifuss, J.J. (1992) Morphological and electrophysiological evidence for postsynaptic localization of functional oxytocin receptors in the rat dorsal motor nucleus of the vagus nerve. Brain Res., 575: 124-131. Easaw, J.C., Jassar, B.S. and Jhamandas, J.H. (1997) Vasopressin receptor subtypes differentially modulate calcium-activated potassium currents in the horizontal limb of the diagonal band of Broca. Neuroscience, 81: 5747.

272 Gao, X., Phillips, P.A., Widdop, R.E., Trinder, D., Jarrott, B. and Johnston, C.I. (1992) Presence of functional vasopressin V1 receptors in rat vagal afferent neurones. Neurosci. Lett., 145: 79-82. Inenaga, K., Kannan, H., Yamashita, H., Tribollet, E., Raggenbass, M. and Dreifuss, J.J. (1991) Oxytocin excites neurons located in the ventromedial nucleus of the guinea-pig hypothalamus. J. Neuroendocrinol., 3: 569-573. Ingram, C.D. and Moos, F. (1992) Oxytocin-containing pathway to the bed nuclei of the stria terminalis of the lactating rat brain: Immunocytochemical and in vitro electrophysiological evidence. Neuroscience, 47: 43% 452. Ingram, C.D. and Tolchard, S. (1994) [ArgSIvasotocin excites neurones in the dorsal vagal complex in vitro: Evidence for an action through novel class(es) of CNS receptors. J. Neuroendocrinol., 6 415-422. Ingram, C.D. and Wakerley, J.B. (1993) Post-partum increase in oxytocin-induced excitation of neurones in the bed nuclei of the stria terminalis in vitro. Brain Res., 602: 325-330. Ingram, C.D., Cutler, K.L. and Wakerley, J.B. (1990) Oxytocin excites neurones in the bed nucleus of the stria terminalis of the lactating rat in vitro. Brain Res., 527: 167-170. Ingram, C.D., Snowball, R.K. and Mihai, R. (1996) Circadian rhythm of neuronal activity in suprachiasmatic nucleus slices from the vasopressin-deficientBrattleboro rat. Neuroscience, 75: 635-641. Jo, Y.-H., Stoekel, M.-E., Freund-Mercier, M.-J. and Schlichter, R. (1998) Oxytocin modulates glutamatergic synaptic transmission between cultured neonatal spinal cord dorsal horn neurons. J. Neurosci., 18: 2377-2386. Jourdain, P., Israel, J.-M., Dupouy, B., Oliet, S.H.R., Allard, M., Vitiello, S., Theodosis, D.T. and Poulain, D.A. (1998) Evidence for a hypothalamic oxytocin-sensitive pattern-generating network governing oxytocin neurons in vitro. J. Neurosci., 18: 6641-6649. Jurzak, M., Miiller, A.R. and Gerstberger, R. (1995) Characterization of vasopressin receptors in cultured cells derived from the region of rat brain circumventricular organs. Neuroscience, 65: 1145-1 159. Kombian, S.B., Mouginot, D. and Pittman, Q.J. (1997) Dendritically released peptides act as retrograde modulators of afferent excitation in the supraoptic nucleus in vitro. Neuron, 1 9 903-912. Kow, L-M., Johnson, A.E., Ogawa, S. and Pfaff, D.W. (1991) Electrophysiological actions of oxytocin on hypothalamic neurons in vitro: Neuropharmacological characterization and effects of ovarian steroids. Neuroendocrinology, 54: 526-535.

Kuriyama, K., Nakashima, T., Kawarabayashi, T. and Kiyoham, T. (1993) Oxytocin inhibits nonphasically firing supraoptic and paraventricular neurons in the virgin female rat. Brain Res. Bull., 31: 681-687. Lambert, R.C., Dayanithi, G., Moos, F.C. and Richard, P. (1994) A rise in the intracellular Ca2+concentration of isolated rat supraoptic cells in response to oxytocin. J. Physiol. (Lomi.),478: 275-288. Lowes, V.L., Sun, K., Li, Z. and Ferguson, A.V. (1995) Vasopressin actions on area postrema neurons in vitro. Am. J. Physiol., 269: R463-R468. Lu, Y.-F., Moriwaki, A., Tomizawa, K., Onuma, H., Cai, X.-H.and Matsui, H. (1997) Effects of vasopressin and involvement of receptor subtypes in the rat central amygdaloid nucleus in vitro. Brain Res., 768: 266-272. Mihai, R., Coculescu, M., Wakerley, J.B. and Ingram,C.D. (1994) The effects of [Arg8]vasopressin and [Arg8]vasotocin on the firing rate of suprachiasmatic neurons in vim. Neuroscience, 62: 783-792. Mihai, R., Juss, T.S. and Ingram, C.D. (1994) Suppression of suprachiasmaticnucleus neurone activity with a vasopressin receptor antagonist: Possible role for endogenous vasopressin in circadian activity cycles in vitro. Neurosci. LRtt., 179: 95-99. Mo, ZL., Katafuchi, T., Muratani, H. and Hori, T. (1992) Effects of vasopressin and angiotensin 11on neurones in the rat dorsal motor nucleus of the vagus, in vitro. J. Physiol. (Lond.),458: 561-577. Nishimori, K., Young, L.J., Guo, Q., Wang, Z., Insel, T.R. and Matzuk, M.M. (1996) Oxytocin is required for nursing but is not essential for parturition or reproductive behavior. Proc. Natl. Acad. Sci. USA, 93: 11699-11704. Palouzier-Paulignan,B., Dubois-Dauphin, M., Tribollet, E., Dreifuss, J.J. and Raggenbass,M. (1994) Action of vasopressin on hypoglossal motoneurones of the rat: Presynaptic and postsynaptic effects. Brain Res., 650 117126. Raggenbass, M. and Dreifuss, J.J. (1992) Mechanism of action of oxytocin in rat vagal neurones: Induction of a sustained sodium-dependent current. J. Physiol. (Lond.), 457: 131-142. Raggenbass, M., Dubois-Dauphin, M., Charpak, S. and Dreifuss, J.J. (1987) Neurons in the dorsal motor nucleus of the vagus nerve are excited by oxytocin in the rat but not in the guinea pig. Proc. Natl. Acad. Sci. USA, 84: 3926-3930. Raggenbass, M., Tribollet, E., Dubois-Dauphin, M. and Dreifuss, J.J. (1989) Vasopressin receptors of the vasopressor (Vl) type in the nucleus of the solitary tract of the rat mediate direct neuronal excitation. J. Neurosci., 9: 3929-3936. Raggenbass, M.,Goumaz, M., Sermasi, E., Tribollet, E. and

273 Dreifuss, J.J. (1991) Vasopressin generates a persistent voltage-dependent sodium current in a mammalian motoneuron. J. Neurosci., 11: 1609-1616. Raggenbass, M., Alberi, S. and Dreifuss, J.J. (1995) Neuronal effects of vasopressin and oxytocin: A progress report. In: T. Saito, K. Kurokawa and S. Yoshida (Eds.), Neurohypophysis: Recent Progress of Vasopressin and Oxytocin Research, Elsevier, Amsterdam, pp. 331-338. Richard, P., Moos, F. and Freund-Mercier, M.-J. (1991) Central effects of oxytocin. Physiol. Rev., 71: 331-370. Rong, X.-W., Chen, X.-F. and Du, Y.-C. (1993) Potentiation of synaptic transmission by neuropeptide AVP4-8 (ZNC(C)PR) in rat hippocampal slices. NeuroReport, 4: 1135-1 138. Sabatier, N., Richard, P. and Dayanithi, G. (1997) L-, N- and T-, but neither P- nor Q-type Ca2+channels control vasopressin-induced Ca2+influx in magnocellular vasopressin neurones isolated from the rat supraoptic nucleus. J. Physiol. (Lond.), 503.2: 253-268. Sermasi, E. and Coote, J.H. (1994) Oxytocin acts at V1 receptors to excite sympathetic preganglionic neurones in neonate rat spinal cord in vitro. Brain Res., 647: 323332. Tolchard, S . and Ingram, C.D. (1993) Electrophysiological actions of oxytocin in the dorsal vagal complex of the female rat in vitro: Changing responsiveness during the oestrous cycle and after steroid treatment. Brain Res., 609: 21-28. Tribollet, E., Charpak, S., Schmidt, A., Dubois-Dauphin, M. and Dreifuss, J.J. (1989) Appearance and transient expression of oxytocin receptors in fetal, infant, and peripubertal rat brain studied by autoradiography and electrophysiology.J. Neurosci., 9: 1764-1773.

Tribollet, E., Goumaz, M., Raggenbass, M., DuboisDauphin, M. and Dreifuss, J.J. (1991) Early appearance and transient expression of vasopressin receptors in the brain of rat fetus and infant. An autoradiographical and electrophysiological study. Dev. Brain Res., 58: 13-24. Tribollet, E., Li, Z., Inenaga, K., Yamashita, H., Raggenbass, M., Dubois-Dauphin, M. and Dreifuss, J.J. (1992) Functional neuronal binding sites for oxytocin in the ventromedial hypothalamus of the guinea pig after gonadectomy. Brain Res., 588: 346-350. Van den Hooff, P. and Urban, I.J.A. (1990) Vasopressin facilitates excitatory transmission in slices of the rat dorso-lateral septum. Synapse, 5: 201-206. Van den Hooff, P., Seger, M.A., Burbach, J.P.H. and Urban, I.J.A. (1990) The C-terminal glycopeptide of propressophysin potentiates excitatory transmission in the rat lateral septum. Neuroscience, 37: 647-653. Widmer, H., Dreifuss, J.J. and Raggenbass, M. (1992) NMethyl-Baspartate and vasopressin activate distinct voltage-dependent inward currents in facial motoneurones. Brain Res., 593: 215-220. Young, W.S., Shepard, E., Amico, J., Hennighausen, L., Wagner, K.-U., LaMarca, M.E., McKinney, C. and Ginns, E.I. (1996) Deficiency in mouse oxytocin prevents milk ejection, but not fertility or parturition. J. Neuroendocrinol., 8: 847-853. Yuan, Z.F. and Pan, J.-T. (1996) Stimulatory effect of central oxytocin on tuberoinfundibular dopaminergic neuron activity and inhibition on prolactin secretion: Neurochemical and electrophysiological studies. Endocrinology, 137: 4120-4125. Zingg, H.H. (1996) Vasopressin and oxytocin receptors. BalliLre 's Clin. Endocrinol. Metab., 10: 75-96.