- Email: [email protected]
Central oxytocin potentiates excitatory responses of oxytocin neurones to stimulation of the dorsomedial hypothalamic nucleus in the suckled rat
Neuroscience Letters 317 (2002) 47–49 www.elsevier.com/locate/neulet
Central oxytocin potentiates excitatory responses of oxytocin neurones to stimulation of the dorsomedial hypothalamic nucleus in the suckled rat A.S. Cosgrave, J.B. Wakerley* Department of Anatomy, School of Medical Sciences, University of Bristol, University Walk, Bristol, BS8 1TD, UK Received 9 August 2001; received in revised form 16 October 2001; accepted 16 October 2001
Abstract Experiments were undertaken to investigate the effects of intracerebroventricular (i.c.v.) oxytocin (OT) on the response of supraoptic OT neurones to stimulation of the dorsomedial nucleus (DMH), in the suckled lactating rat. Under control conditions, the majority of OT neurones displayed either weak excitation to DMH stimulation, or no response. Following i.c.v. OT injection, all neurones showed a pronounced long-latency (70–115 ms) excitatory response, and the number of spikes evoked per stimulus pulse was significantly increased. This increased excitatory response was accompanied by facilitation of the milk-ejection reflex. Some OT neurones also displayed a short latency (8–13 ms) excitation to DMH stimulation, but this was unaffected by i.c.v. OT. In conclusion, the facilitation of bursting in OT neurones by i.c.v. OT is associated with potentiation of long-latency excitatory responses evoked by DMH stimulation. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Oxytocin neurones; Excitatory responses; Dorsomedial nucleus; Intracerebroventricular oxytocin; Potentiation; Milk-ejection reflex
Cosgrave et al. recently reported [2] that administration of intracerebroventricular (i.c.v) oxytocin (OT) to the suckled rat resulted in potentiation of the excitatory responses of OT neurones to stimulation of the mesencephalic ventral tegmentum, a region implicated in the milk-ejection reflex. This effect correlated with increased bursting in the OT neurones, and it was therefore proposed that the potentiation might provide a mechanism for the powerful facilitatory effect of central OT on the milk-ejection reflex [7]. However, whether such modulation can be observed at higher levels of the afferent pathway of the milk-ejection reflex has yet to be determined. The present study extends the work of Cosgrave et al. by investigating whether i.c.v. OT similarly modulates the responses of OT neurones to stimulation of the dorsomedial nucleus of the hypothalamus (DMH). The DMH is known to be an essential structure for activation of OT neurones by the suckling stimulus [8]. Experiments were performed on lactating Wistar rats (Bantin and Kingman, Hull, UK) mated within our own colony, and used at day 9–11 post-partum. After overnight * Corresponding author. Tel.: 144-117-928-7406; fax: 144-117929-1687. E-mail address: [email protected] (J.B. Wakerley).
separation from all but one of their pups, the rats were anaesthetised with urethane (1.2 g/kg i.p., supplemented with methohexitone sodium during surgery) and prepared for electrophysiological recording of antidromically-identified neurones and i.c.v. injection of OT into the third ventricle, as previously described [3]. A concentric bipolar stimulating electrode (outer diameter, 0.5 mm) was located within the DMH, using the following co-ordinates: 2.5 mm behind the bregma; 0.5 mm from the midline (ipsilateral to the recording site); and 7.5 mm deep. Following surgical preparation, the animals were left undisturbed for 1–2 h before commencement of electrophysiological recording. After locating a stable supraoptic (SO) neurone, stimulation was applied to the DMH at a frequency of one per 3 s, using biphasic pulses (2 ms; 0.2–0.8 mA). After application of at least 100 shocks (i.e. a minimum control period of 5 min), an i.c.v. injection of OT (2.2 ng; 1 mU) was given. The cell was then monitored for a further period spanning facilitation of the milk-ejection reflex, with maintained DMH stimulation. At the end of the experiment, the rats were killed by barbiturate overdose (50 mg sodium methohexitone i.v.), and the brains removed for histological analysis to confirm the stimulation sites.
0304-3940/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 1) 02 40 2- 8
48
A.S. Cosgrave, J.B. Wakerley / Neuroscience Letters 317 (2002) 47–49
Results were obtained from seven OT neurones recorded for periods of 18–55 min of suckling, with continuous 0.33 Hz DMH stimulation. Background firing rates ranged from 0.4 to 5.7 spikes/s, and all of these neurones displayed facilitated high frequency bursting responses (mean interval, 4.9 ^ 0.5 min) following i.c.v OT. Examples of the effect of i.c.v OT on the responses of these OT neurones to DMH stimulation are shown in Figs. 1–3. Prior to i.c.v. OT, four out of seven neurones displayed a very weak, long-latency (105– 130 ms) excitation lasting 30–70 ms. In one case, the excitation was preceded by an inhibitory phase, and a further two neurones displayed only an inhibitory response. Following i.c.v. OT, all seven neurones displayed a strong excitation to DMH stimulation (latency, 100.4 ^ 5.8 ms; duration, 71.7 ^ 7.5 ms). The mean number of additional spikes evoked per stimulus shock (see [2] for explanation) for all seven tested neurones was 0.53 ^ 0.10, compared with 0.26 ^ 0.09 prior to OT (P , 0:001, paired t-test). Five of the neurones displayed a transient inhibition (latency, 20.6 ^ 3.1 ms; and duration, 77.0 ^ 5.9 ms) preceding their excitation. The enhanced excitatory response appeared after a delay of 3.8 ^ 1.0 min following i.c.v. OT, compared with a delay of 4.4 ^ 1.6 min to the first facilitated burst. As with ventral tegmentum stimulation [2], the response was strongest in the period leading up to each facilitated burst, and declined immediately afterwards (Fig. 2A).
Fig. 1. Example of the effect of i.c.v. OT on the response of an OT neurone to stimulation of the DMH in the suckled lactating rat. Part A, upper trace, shows a ratemeter recording of the activity of the neurone during continuous suckling and application of single-shock DMH stimulation (DMH stim) applied at 0.3 Hz. Note the occurrence of repetitive high frequency bursts (B) after OT administration. Part A, lower trace, shows a raster display of the response of the neurone to DMH stimulation. Note the weak excitatory response prior to i.c.v. OT administration, which became stronger following the OT injection. Part B shows peri-stimulus time-interval histograms for a 500 s control period (left) and a 500 s period after giving i.c.v. OT (right), to illustrate more clearly the increase in the response. Open arrows indicate one or more bins containing stimulus artefacts (these are truncated).
Fig. 2. Recording of an OT neurone to compare the response to single-shock (SS) versus double-shock (DS) stimulation of the DMH during administration of i.c.v. OT in the suckled lactating rat. This recording also illustrates how the response to DMH stimulation changed in intensity surrounding the high frequency bursts associated with milk ejection. Part A: segments of a raster display of the response of the neurone to SS DMH stimulation prior to OT (left, 0–300 s), the response to SS DMH stimulation after i.c.v. OT (centre, 500–1200 s), and the response to DS DMH stimulation after i.c.v. OT (right, 2500–3300 s). Vertical bars indicate the occurrence of each milk-ejection related burst (B). Part B: peri-stimulus time-interval histograms for a 300 s period of SS stimulation before i.c.v. OT (left), a 300 s period of SS stimulation after i.c.v. OT (centre), and a 300 s period of DS stimulation after i.c.v. OT (right). Time periods indicated on these histograms relate to the time course of the raster display in Part A. Open arrows indicate one or more bins containing stimulus artefacts (these are truncated). The single solid triangle indicates the onset of the response to SS stimulation; double solid triangles indicate the expected onset of each response to DS stimulation. Further explanation is provided in the text.
In two neurones displaying inhibition preceding the longlatency excitation to DMH stimulation, tests were undertaken in which double shocks were applied to examine the interaction between the inhibitory and excitatory phases of the response (Fig. 2). The interval between the two stimulus pulses was arranged such that the inhibition evoked by the second pulse coincided with the excitation evoked by the first. In both cases, excitation to the first shock was strongly attenuated by the coincident inhibition resulting from the second shock, such that only a single excitatory phase was apparent (Fig. 2). Such interaction also resulted in truncation of the inhibitory period preceding this single excitation, but there was no loss in the amplitude of the excitatory response, arguing against the involvement of a rebound phenomenon in these responses. Two out of seven OT neurones also displayed a short latency (8–13 ms) excitatory response to stimulation of the DMH that, in each case, was present prior to i.c.v OT. There was no evidence that these short latency responses were potentiated by OT administration (Fig. 3). The absence of any effect on these short latency responses supports previous conclusions [2] that the potentiation of ascending excitatory inputs to OT neurones by i.c.v. OT is unlikely to involve a generalised change in post-synaptic excitability.
A.S. Cosgrave, J.B. Wakerley / Neuroscience Letters 317 (2002) 47–49
49
latency responses might also arise if DMH stimulation activated a short pathway ascending through the hypothalamus, with polysynaptic excitatory relays close to the SO nucleus. Indeed, participation of local polysynaptic circuitry has already been proposed to explain long-latency excitations observed in magnocellular neurones during peri-nuclear stimulation in vitro [1]. The site of action of i.c.v. OT in the current experiments remains to be established. OT receptors have been located in a number of structures involved in the afferent control of the milk-ejection reflex, including the bed nucleus, amygdala, and dorsal vagal complex [9]. However, recent electrophysiological studies have reported that OT can modulate afferent transmission [5] and burst generation [4] in OT neurones recorded from isolated hypothalamic slices or explants that are devoid of distant inputs. Hence, the possibility that the current observations involved an intra-hypothalamic action of OT, either within or close to the SO nuclei, also deserves further investigation. This work was supported by a grant from the Biotechnology and Biological Sciences Research Council. Fig. 3. Example of an OT neurone which displayed both a long and short latency excitatory response to stimulation of the DMH. Part A shows a raster display of the short latency response, indicated by the line of clustered spikes at a latency of 8–10 ms. Note how this short latency response was present throughout the recording, and was unaffected by i.c.v. OT administration. Vertical bars indicate the occurrence of milk-ejection related bursts (B). Part B shows peri-stimulus time-interval histograms of the short latency response, confirming the lack of potentiation by i.c.v. OT (time periods indicated on these histograms relate to the time course of the raster display in Part A). Open arrows indicate one or more bins containing stimulus artefacts (these are truncated). Part C shows peri-stimulus time-interval histograms of the long-latency response of this neurone, which was clearly potentiated by i.c.v. OT.
The finding that i.c.v. OT enhances long-latency excitatory responses of OT neurones to DMH stimulation is in line with previous results showing that OT potentiates ascending excitatory inputs to OT neurones during suckling [2]. As in these earlier studies, the potentiation observed following i.c.v. OT followed a similar time course to facilitation of the milk-ejection reflex, further supporting the view that facilitation of suckling-evoked bursting by central OT is associated with augmentation of excitatory inputs to the OT neurones. Interestingly, the latency of the potentiated excitations (.100 ms) are comparable with those previously observed with ventral tegmentum stimulation [2], and this might seem unexpected given the much closer proximity of the DMH to the SO nucleus. The long-latency (.100 ms) excitations observed in the present study might have resulted from activation of a long circuitous pathway from the DMH to the SO nucleus, perhaps involving extrahypothalamic structures involved in the milk-ejection reflex, such as the bed nucleus [6]. However, such long-
[1] Boudaba, C., Schrader, L.A. and Tasker, J.G., Physiological evidence for local excitatory synaptic circuits in the rat hypothalamus, J. Neurophysiol., 77 (1997) 3396–3400. [2] Cosgrave, A.S., Richardson, C.M. and Wakerley, J.B., Permissive effect of centrally administered oxytocin on the excitatory response of oxytocin neurones to ventral tegmental stimulation in the suckled lactating rat, J. Neuroendocrinol., 12 (2000) 843–852. [3] Jiang, Q.B. and Wakerley, J.B., Analysis of bursting responses of oxytocin neurones in the rat in late pregnancy, lactation and after weaning, J. Physiol., 486 (1995) 237–248. [4] Jourdain, P., Israel, J.M., Dupouy, B., Oliet, S.H., Allard, M., Vitiello, S., Theodosis, D.T. and Poulain, D.A., Evidence for a hypothalamic oxytocin-sensitive pattern-generating network governing oxytocin neurons in vitro, J. Neurosci., 18 (1998) 6641–6649. [5] Kombian, S.B., Mouginot, D. and Pittman, Q.J., Dendritically released peptides act as retrograde modulators of afferent excitation in the supraoptic nucleus in vitro, Neuron, 19 (1997) 903–912. [6] Lambert, R.C., Moos, F.C., Ingram, C.D., Wakerley, J.B., Kremarik, P., Guerne, Y. and Richard, P., Electrical activity of neurons in the ventrolateral septum and bed nuclei of the stria terminalis in suckled rats: statistical analysis gives evidence for sensitivity to oxytocin and for relation to the milk-ejection reflex, Neuroscience, 54 (1993) 361–376. [7] Lambert, R.C., Moos, F.C. and Richard, P., Action of endogenous oxytocin within the paraventricular and supraoptic nuclei: a powerful link in the regulation of the bursting pattern of oxytocin neurones during the milk-ejection reflex in rats, Neuroscience, 57 (1993) 1027–1038. [8] Takano, S., Negoro, H., Honda, K. and Higuchi, T., Lesion and electrophysiological studies on the hypothalamic afferent pathway of the milk ejection reflex in the rat, Neuroscience, 50 (1992) 877–883. [9] Tribollet, E., Dubois-Dauphin, M., Dreifuss, J.J., Barberis, C. and Jard, S., Oxytocin receptors in the central nervous system. Distribution, development, and species differences, Ann. New York Acad. Sci., 652 (1992) 29–38.
Central oxytocin potentiates excitatory responses of oxytocin neurones to stimulation of the dorsomedial hypothalamic nucleus in the suckled rat A.S. Cosgrave, J.B. Wakerley* Department of Anatomy, School of Medical Sciences, University of Bristol, University Walk, Bristol, BS8 1TD, UK Received 9 August 2001; received in revised form 16 October 2001; accepted 16 October 2001
Abstract Experiments were undertaken to investigate the effects of intracerebroventricular (i.c.v.) oxytocin (OT) on the response of supraoptic OT neurones to stimulation of the dorsomedial nucleus (DMH), in the suckled lactating rat. Under control conditions, the majority of OT neurones displayed either weak excitation to DMH stimulation, or no response. Following i.c.v. OT injection, all neurones showed a pronounced long-latency (70–115 ms) excitatory response, and the number of spikes evoked per stimulus pulse was significantly increased. This increased excitatory response was accompanied by facilitation of the milk-ejection reflex. Some OT neurones also displayed a short latency (8–13 ms) excitation to DMH stimulation, but this was unaffected by i.c.v. OT. In conclusion, the facilitation of bursting in OT neurones by i.c.v. OT is associated with potentiation of long-latency excitatory responses evoked by DMH stimulation. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Oxytocin neurones; Excitatory responses; Dorsomedial nucleus; Intracerebroventricular oxytocin; Potentiation; Milk-ejection reflex
Cosgrave et al. recently reported [2] that administration of intracerebroventricular (i.c.v) oxytocin (OT) to the suckled rat resulted in potentiation of the excitatory responses of OT neurones to stimulation of the mesencephalic ventral tegmentum, a region implicated in the milk-ejection reflex. This effect correlated with increased bursting in the OT neurones, and it was therefore proposed that the potentiation might provide a mechanism for the powerful facilitatory effect of central OT on the milk-ejection reflex [7]. However, whether such modulation can be observed at higher levels of the afferent pathway of the milk-ejection reflex has yet to be determined. The present study extends the work of Cosgrave et al. by investigating whether i.c.v. OT similarly modulates the responses of OT neurones to stimulation of the dorsomedial nucleus of the hypothalamus (DMH). The DMH is known to be an essential structure for activation of OT neurones by the suckling stimulus [8]. Experiments were performed on lactating Wistar rats (Bantin and Kingman, Hull, UK) mated within our own colony, and used at day 9–11 post-partum. After overnight * Corresponding author. Tel.: 144-117-928-7406; fax: 144-117929-1687. E-mail address: [email protected] (J.B. Wakerley).
separation from all but one of their pups, the rats were anaesthetised with urethane (1.2 g/kg i.p., supplemented with methohexitone sodium during surgery) and prepared for electrophysiological recording of antidromically-identified neurones and i.c.v. injection of OT into the third ventricle, as previously described [3]. A concentric bipolar stimulating electrode (outer diameter, 0.5 mm) was located within the DMH, using the following co-ordinates: 2.5 mm behind the bregma; 0.5 mm from the midline (ipsilateral to the recording site); and 7.5 mm deep. Following surgical preparation, the animals were left undisturbed for 1–2 h before commencement of electrophysiological recording. After locating a stable supraoptic (SO) neurone, stimulation was applied to the DMH at a frequency of one per 3 s, using biphasic pulses (2 ms; 0.2–0.8 mA). After application of at least 100 shocks (i.e. a minimum control period of 5 min), an i.c.v. injection of OT (2.2 ng; 1 mU) was given. The cell was then monitored for a further period spanning facilitation of the milk-ejection reflex, with maintained DMH stimulation. At the end of the experiment, the rats were killed by barbiturate overdose (50 mg sodium methohexitone i.v.), and the brains removed for histological analysis to confirm the stimulation sites.
0304-3940/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 1) 02 40 2- 8
48
A.S. Cosgrave, J.B. Wakerley / Neuroscience Letters 317 (2002) 47–49
Results were obtained from seven OT neurones recorded for periods of 18–55 min of suckling, with continuous 0.33 Hz DMH stimulation. Background firing rates ranged from 0.4 to 5.7 spikes/s, and all of these neurones displayed facilitated high frequency bursting responses (mean interval, 4.9 ^ 0.5 min) following i.c.v OT. Examples of the effect of i.c.v OT on the responses of these OT neurones to DMH stimulation are shown in Figs. 1–3. Prior to i.c.v. OT, four out of seven neurones displayed a very weak, long-latency (105– 130 ms) excitation lasting 30–70 ms. In one case, the excitation was preceded by an inhibitory phase, and a further two neurones displayed only an inhibitory response. Following i.c.v. OT, all seven neurones displayed a strong excitation to DMH stimulation (latency, 100.4 ^ 5.8 ms; duration, 71.7 ^ 7.5 ms). The mean number of additional spikes evoked per stimulus shock (see [2] for explanation) for all seven tested neurones was 0.53 ^ 0.10, compared with 0.26 ^ 0.09 prior to OT (P , 0:001, paired t-test). Five of the neurones displayed a transient inhibition (latency, 20.6 ^ 3.1 ms; and duration, 77.0 ^ 5.9 ms) preceding their excitation. The enhanced excitatory response appeared after a delay of 3.8 ^ 1.0 min following i.c.v. OT, compared with a delay of 4.4 ^ 1.6 min to the first facilitated burst. As with ventral tegmentum stimulation [2], the response was strongest in the period leading up to each facilitated burst, and declined immediately afterwards (Fig. 2A).
Fig. 1. Example of the effect of i.c.v. OT on the response of an OT neurone to stimulation of the DMH in the suckled lactating rat. Part A, upper trace, shows a ratemeter recording of the activity of the neurone during continuous suckling and application of single-shock DMH stimulation (DMH stim) applied at 0.3 Hz. Note the occurrence of repetitive high frequency bursts (B) after OT administration. Part A, lower trace, shows a raster display of the response of the neurone to DMH stimulation. Note the weak excitatory response prior to i.c.v. OT administration, which became stronger following the OT injection. Part B shows peri-stimulus time-interval histograms for a 500 s control period (left) and a 500 s period after giving i.c.v. OT (right), to illustrate more clearly the increase in the response. Open arrows indicate one or more bins containing stimulus artefacts (these are truncated).
Fig. 2. Recording of an OT neurone to compare the response to single-shock (SS) versus double-shock (DS) stimulation of the DMH during administration of i.c.v. OT in the suckled lactating rat. This recording also illustrates how the response to DMH stimulation changed in intensity surrounding the high frequency bursts associated with milk ejection. Part A: segments of a raster display of the response of the neurone to SS DMH stimulation prior to OT (left, 0–300 s), the response to SS DMH stimulation after i.c.v. OT (centre, 500–1200 s), and the response to DS DMH stimulation after i.c.v. OT (right, 2500–3300 s). Vertical bars indicate the occurrence of each milk-ejection related burst (B). Part B: peri-stimulus time-interval histograms for a 300 s period of SS stimulation before i.c.v. OT (left), a 300 s period of SS stimulation after i.c.v. OT (centre), and a 300 s period of DS stimulation after i.c.v. OT (right). Time periods indicated on these histograms relate to the time course of the raster display in Part A. Open arrows indicate one or more bins containing stimulus artefacts (these are truncated). The single solid triangle indicates the onset of the response to SS stimulation; double solid triangles indicate the expected onset of each response to DS stimulation. Further explanation is provided in the text.
In two neurones displaying inhibition preceding the longlatency excitation to DMH stimulation, tests were undertaken in which double shocks were applied to examine the interaction between the inhibitory and excitatory phases of the response (Fig. 2). The interval between the two stimulus pulses was arranged such that the inhibition evoked by the second pulse coincided with the excitation evoked by the first. In both cases, excitation to the first shock was strongly attenuated by the coincident inhibition resulting from the second shock, such that only a single excitatory phase was apparent (Fig. 2). Such interaction also resulted in truncation of the inhibitory period preceding this single excitation, but there was no loss in the amplitude of the excitatory response, arguing against the involvement of a rebound phenomenon in these responses. Two out of seven OT neurones also displayed a short latency (8–13 ms) excitatory response to stimulation of the DMH that, in each case, was present prior to i.c.v OT. There was no evidence that these short latency responses were potentiated by OT administration (Fig. 3). The absence of any effect on these short latency responses supports previous conclusions [2] that the potentiation of ascending excitatory inputs to OT neurones by i.c.v. OT is unlikely to involve a generalised change in post-synaptic excitability.
A.S. Cosgrave, J.B. Wakerley / Neuroscience Letters 317 (2002) 47–49
49
latency responses might also arise if DMH stimulation activated a short pathway ascending through the hypothalamus, with polysynaptic excitatory relays close to the SO nucleus. Indeed, participation of local polysynaptic circuitry has already been proposed to explain long-latency excitations observed in magnocellular neurones during peri-nuclear stimulation in vitro [1]. The site of action of i.c.v. OT in the current experiments remains to be established. OT receptors have been located in a number of structures involved in the afferent control of the milk-ejection reflex, including the bed nucleus, amygdala, and dorsal vagal complex [9]. However, recent electrophysiological studies have reported that OT can modulate afferent transmission [5] and burst generation [4] in OT neurones recorded from isolated hypothalamic slices or explants that are devoid of distant inputs. Hence, the possibility that the current observations involved an intra-hypothalamic action of OT, either within or close to the SO nuclei, also deserves further investigation. This work was supported by a grant from the Biotechnology and Biological Sciences Research Council. Fig. 3. Example of an OT neurone which displayed both a long and short latency excitatory response to stimulation of the DMH. Part A shows a raster display of the short latency response, indicated by the line of clustered spikes at a latency of 8–10 ms. Note how this short latency response was present throughout the recording, and was unaffected by i.c.v. OT administration. Vertical bars indicate the occurrence of milk-ejection related bursts (B). Part B shows peri-stimulus time-interval histograms of the short latency response, confirming the lack of potentiation by i.c.v. OT (time periods indicated on these histograms relate to the time course of the raster display in Part A). Open arrows indicate one or more bins containing stimulus artefacts (these are truncated). Part C shows peri-stimulus time-interval histograms of the long-latency response of this neurone, which was clearly potentiated by i.c.v. OT.
The finding that i.c.v. OT enhances long-latency excitatory responses of OT neurones to DMH stimulation is in line with previous results showing that OT potentiates ascending excitatory inputs to OT neurones during suckling [2]. As in these earlier studies, the potentiation observed following i.c.v. OT followed a similar time course to facilitation of the milk-ejection reflex, further supporting the view that facilitation of suckling-evoked bursting by central OT is associated with augmentation of excitatory inputs to the OT neurones. Interestingly, the latency of the potentiated excitations (.100 ms) are comparable with those previously observed with ventral tegmentum stimulation [2], and this might seem unexpected given the much closer proximity of the DMH to the SO nucleus. The long-latency (.100 ms) excitations observed in the present study might have resulted from activation of a long circuitous pathway from the DMH to the SO nucleus, perhaps involving extrahypothalamic structures involved in the milk-ejection reflex, such as the bed nucleus [6]. However, such long-
[1] Boudaba, C., Schrader, L.A. and Tasker, J.G., Physiological evidence for local excitatory synaptic circuits in the rat hypothalamus, J. Neurophysiol., 77 (1997) 3396–3400. [2] Cosgrave, A.S., Richardson, C.M. and Wakerley, J.B., Permissive effect of centrally administered oxytocin on the excitatory response of oxytocin neurones to ventral tegmental stimulation in the suckled lactating rat, J. Neuroendocrinol., 12 (2000) 843–852. [3] Jiang, Q.B. and Wakerley, J.B., Analysis of bursting responses of oxytocin neurones in the rat in late pregnancy, lactation and after weaning, J. Physiol., 486 (1995) 237–248. [4] Jourdain, P., Israel, J.M., Dupouy, B., Oliet, S.H., Allard, M., Vitiello, S., Theodosis, D.T. and Poulain, D.A., Evidence for a hypothalamic oxytocin-sensitive pattern-generating network governing oxytocin neurons in vitro, J. Neurosci., 18 (1998) 6641–6649. [5] Kombian, S.B., Mouginot, D. and Pittman, Q.J., Dendritically released peptides act as retrograde modulators of afferent excitation in the supraoptic nucleus in vitro, Neuron, 19 (1997) 903–912. [6] Lambert, R.C., Moos, F.C., Ingram, C.D., Wakerley, J.B., Kremarik, P., Guerne, Y. and Richard, P., Electrical activity of neurons in the ventrolateral septum and bed nuclei of the stria terminalis in suckled rats: statistical analysis gives evidence for sensitivity to oxytocin and for relation to the milk-ejection reflex, Neuroscience, 54 (1993) 361–376. [7] Lambert, R.C., Moos, F.C. and Richard, P., Action of endogenous oxytocin within the paraventricular and supraoptic nuclei: a powerful link in the regulation of the bursting pattern of oxytocin neurones during the milk-ejection reflex in rats, Neuroscience, 57 (1993) 1027–1038. [8] Takano, S., Negoro, H., Honda, K. and Higuchi, T., Lesion and electrophysiological studies on the hypothalamic afferent pathway of the milk ejection reflex in the rat, Neuroscience, 50 (1992) 877–883. [9] Tribollet, E., Dubois-Dauphin, M., Dreifuss, J.J., Barberis, C. and Jard, S., Oxytocin receptors in the central nervous system. Distribution, development, and species differences, Ann. New York Acad. Sci., 652 (1992) 29–38.