Effects of Feeding-Related Peptides on Neuronal Oscillation in the Ventromedial Hypothalamus.

The ventromedial hypothalamus (VMH) plays an important role in feeding behavior, obesity, and thermoregulation. The VMH contains glucose-sensing neurons, the firing of which depends on the level of extracellular glucose and which are involved in maintaining the blood glucose level via the sympathetic nervous system. The VMH also expresses various receptors of the peptides related to feeding. However, it is not well-understood whether the action of feeding-related peptides mediates the activity of glucose-sensing neurons in the VMH. In the present study, we examined the effects of feeding-related peptides on the burst-generating property of the VMH. Superfusion with insulin, pituitary adenylate cyclase-activating polypeptide, corticotropin-releasing factor, and orexin increased the frequency of the VMH oscillation. In contrast, superfusion with leptin, cholecystokinin, cocaine- and amphetamine-regulated transcript, galanin, ghrelin, and neuropeptide Y decreased the frequency of the oscillation. Our findings indicated that the frequency changes of VMH oscillation in response to the application of feeding-related peptides showed a tendency similar to changes of sympathetic nerve activity in response to the application of these substances to the brain.

Modulation of sympathetic preganglionic neuron activity via adrenergic receptors.

The sympathetic preganglionic neurons (SPNs) play a key role in the sympathetic nervous system. Previous reports have suggested that norepinephrine (NE) directly affects SPNs via both inhibitory hyperpolarization interactions mediated by α2 receptors and excitatory depolarization interactions mediated by α1 receptors. It remains poorly understood, however, whether the excitability of SPNs can be inhibited indirectly (presynaptically) as well as directly (postsynaptically). We intracellularly recorded 41 SPNs using the whole-cell patch-clamp technique in spinal cord slice preparations of neonatal rats. We examined the effects of NE or dexmedetomidine hydrochloride (Dxm) (α2-adrenergic receptor agonist) on SPNs by analyzing the excitatory postsynaptic potentials (EPSPs) and inhibitory postsynaptic potentials (IPSPs). EPSPs were dominant in 15 SPNs (EPSP-SPNs) and IPSPs were dominant in 7 SPNs (IPSP-SPNs) at baseline. We were unable to analyze the postsynaptic potentials in the other 19 SPNs, due to high frequency of action potential firings (firing-SPNs). At baseline, the membrane potentials and resistances of each type of SPN were similar. NE (1 µM) gradually depolarized the EPSP-SPNs and IPSP-SPNs (P < 0.001) and NE significantly increased the EPSP frequency of the EPSP-SPNs (P < 0.05). Dxm (10 nM) after application of NE decreased the EPSP frequency of the EPSP-SPNs (P < 0.001) and the EPSP voltage and IPSP voltage of the IPSP-SPNs (P < 0.05). In 5 of the 19 firing-SPNs, NE induced membrane hyperpolarization (P < 0.05) and completely inhibited firings. Dxm had no effect in these neurons. The SPNs received inhibitory modulation through α2-adrenergic receptors. Some SPNs can be directly inhibited via effects independent of the α2 receptors.

An in vitro experimental model for analysis of central control of sympathetic nerve activity.

Newborn rat brainstem-spinal cord preparations are useful for in vitro analysis of various brainstem functions including respiratory activity. When studying the central control of sympathetic nerve activity (SNA), it is important to record peripheral outputs of the SNA. We developed an in vitro preparation in which neuronal connections between the cardiovascular center in the medulla and SNA peripheral outputs are preserved. Zero- to 1-day-old rats were deeply anesthetized with isoflurane, and the brainstem and spinal cord were isolated with a partial right thoracic cage to record sympathetic nerve discharge from the right thoracic sympathetic nerve trunk (T9-T11). SNA in this preparation was strongly modulated by inspiratory activity. Single-shot electrical stimulation of the ipsilateral rostral ventrolateral medulla (RVLM) induced a transient increase of SNA. Bath application of angiotensin II induced an increase of SNA, and local ipsilateral microinjection of angiotensin II to the RVLM induced a transient increase of SNA. This preparation allows analysis of the central control of the SNA in vitro.

Platypnea-orthodeoxia syndrome in the right lateral decubitus position: a case report.

BACKGROUND: Platypnea-orthodeoxia syndrome is a rare syndrome characterized by dyspnea and hypoxia when the patient is sitting or standing. Here we report a case of platypnea-orthodeoxia syndrome caused by a right hemidiaphragmatic elevation with giant liver cyst that triggered a right-to-left shunt through the patent foramen ovale. This case report is the first presentation of a case secondary to hemidiaphragmatic elevation with giant liver cyst. In addition to this, a malposition of the pacemaker lead could be associated with platypnea-orthodeoxia syndrome in this case. CASE PRESENTATION: A 91-year-old Japanese woman presented to our hospital with hypoxia of unknown origin. Severe hypoxia and cyanosis were observed only in the right lateral decubitus position. A chest X-ray and computed tomography scan revealed right hemidiaphragmatic elevation, which was probably compressing the right atrium. A transesophageal echocardiogram showed a compressed right atrium and shunt blood flow in both directions: from the left to the right atrium and vice versa. The shunt flow was exacerbated by postural changes from the left to the right lateral decubitus. A transesophageal echocardiogram also confirmed compression of the right atrium due to giant liver cyst and a malposition of the pacemaker lead abnormally placed in the left atrium through patent foramen ovale. We concluded that the cause of hypoxia was platypnea-orthodeoxia syndrome with right-to-left interatrial shunt through patent foramen ovale. Surgical closure of patent foramen ovale was not performed due to the age of our patient, surgical difficulties, and failure to obtain informed consent. For these reasons she was discharged after receiving medical advice about her posture. CONCLUSIONS: Platypnea-orthodeoxia syndrome is rare and difficult to diagnose. The present case suggests that hypoxia due to postural changes should be considered a differential diagnosis of platypnea-orthodeoxia syndrome.

Interaction between novel oscillation within the ventromedial hypothalamus and the sympathetic nervous system.

The ventromedial hypothalamus (VMH) is known to play an important role in feeding behavior and the control of sympathetic nerve activity (SNA). We report the identification of novel neuron groups that showed oscillations on both sides of the VMH in hypothalamus slice preparations from juvenile rats of postnatal days 5-14. We detected spontaneous rhythmic burst activity with a frequency of around 0.06Hz typically in the dorsolateral region of the VMH (i.e., VMH oscillation) using optical recordings (voltage and calcium imaging), field potential recordings and intracellular membrane potential recordings. The oscillation was also confirmed after isolation of the VMH from other hypothalamic structures. The frequency of oscillation was increased by lowering the glucose concentration of the superfusate. To evaluate the relation between VMH oscillation and SNA, we simultaneously recorded VMH oscillation, SNA from the thoracic sympathetic nerve trunk and phrenic nerve discharge (Phr) in the decerebrate and arterially perfused in situ preparation from juvenile rats of postnatal days 5-11. Power spectral analysis in the arterially perfused in situ rat preparation revealed similar peak values to those of slice preparations within the low-frequency range between the VMH oscillation and sympathetic nerve trunk activity. In addition, we analyzed cross-correlations between the VMH, SNA and Phr. The results revealed that a predominant positive correlation of the VMH activity with the SNA existed with an average time lag of 2.4s, suggesting the presence of functional couplings between the VMH and SNA (and respiratory center) in the lower brainstem and spinal cord. We hypothesize that the VMH oscillation might be involved in low-frequency modulation of the SNA.

Role of dorsolateral periaqueductal grey in the coordinated regulation of cardiovascular and respiratory function.

The midbrain periaqueductal grey (PAG) contains four longitudinal columns, referred to as the dorsomedial (dmPAG), dorsolateral (dlPAG), lateral (lPAG) and ventrolateral (vlPAG) subdivisions, which collectively have a pivotal role in integrating behavioural and physiological responses to external stressors as well as other functions. This review is focussed on the dlPAG, which is believed to be an important component of the central mechanisms that generate the defensive response to acute psychological stressors, such as the presence of a predator or other immediate threat. The anatomical connections of the dlPAG are highly specific and distinctly different from those of the other PAG subregions. The chemical properties of the dlPAG are also distinctly different from the other PAG subregions (e.g. there is a very high density of neurons that synthesize nitric oxide in the dlPAG but very few such neurons in the other PAG subregions). Recent functional studies have demonstrated that neurons in the dlPAG exert a powerful control over both sympathetic and respiratory activity, and that the pattern of the evoked respiratory changes is also distinctly different from those evoked from other PAG subregions. These studies also showed that the sympathetic and respiratory changes evoked from the dlPAG are highly correlated, suggesting the possibility that a common population of "command neurons" within this region may generate the sympathetic and respiratory changes that accompany defensive behavioural responses to acute psychological stressors. Finally, although the anatomical connections and functional properties of the dlPAG are distinctly different from the other PAG subregions, they have many similarities with adjacent parts of the superior colliculus, suggesting that the dlPAG and deep layers of the superior colliculus may be part of a common defence system in the midbrain.

Blockade of orexin receptors with Almorexant reduces cardiorespiratory responses evoked from the hypothalamus but not baro- or chemoreceptor reflex responses.

Orexin neurons form a restricted group in the dorsal hypothalamus. The group is centered on the perifornical area within the classic hypothalamic defense area, an area which when activated produces marked cardiovascular and respiratory effects. Central administration of orexin can produce cardiorespiratory effects, but the extent to which orexin contributes to such responses evoked from the perifornical hypothalamus is not clear. To determine this, we used the dual orexin receptor antagonist Almorexant to challenge the cardiorespiratory effects evoked by disinhibition of the perifornical hypothalamus. Bicuculline (10 and 20 pmol) was microinjected in the perifornical area before and after administration of Almorexant (15 mg/kg iv) or vehicle in urethane-anesthetized rats. Almorexant significantly reduced the pressor, tachycardic, renal sympathoexcitatory, and tachypneic responses to bicuculline (10 pmol, by 55%, 53%, 28%, 77%; 20 pmol, by 54%, 27%, 51%, 72%, respectively). Reductions of similar magnitude were observed with bicuculline microinjections centered on more caudal sites just peripheral to the orexin neuron group, which would likely have activated fewer orexin neurons. In contrast, Almorexant had no effect on the cardiorespiratory response of the chemoreflex (sodium cyanide injection) or the sympathetic component of the baroreflex. Thus orexin makes a major contribution to the cardiorespiratory response evoked from the perifornical area even though orexin neurons represent only a fraction of the output of this area. Orexin neurons may also mediate cardiorespiratory responses from non-orexin neurons in the caudal hypothalamus. However, under resting conditions, blockade of orexin receptors does not affect the chemo- and baroreflexes.

Synchronized activation of sympathetic vasomotor, cardiac, and respiratory outputs by neurons in the midbrain colliculi.

The superior and inferior colliculi are believed to generate immediate and highly coordinated defensive behavioral responses to threatening visual and auditory stimuli. Activation of neurons in the superior and inferior colliculi have been shown to evoke increases in cardiovascular and respiratory activity, which may be components of more generalized stereotyped behavioral responses. In this study, we examined the possibility that there are "command neurons" within the colliculi that can simultaneously drive sympathetic and respiratory outputs. In anesthetized rats, microinjections of bicuculline (a GABA(A) receptor antagonist) into sites within a circumscribed region in the deep layers of the superior colliculus and in the central and external nuclei of the inferior colliculus evoked a response characterized by intense and highly synchronized bursts of renal sympathetic nerve activity (RSNA) and phrenic nerve activity (PNA). Each burst of RSNA had a duration of ∼300-400 ms and occurred slightly later (peak to peak latency of 41 ± 8 ms) than the corresponding burst of PNA. The bursts of RSNA and PNA were also accompanied by transient increases in arterial pressure and, in most cases, heart rate. Synchronized bursts of RSNA and PNA were also evoked after neuromuscular blockade, artificial ventilation, and vagotomy and so were not dependent on afferent feedback from the lungs. We propose that the synchronized sympathetic-respiratory responses are driven by a common population of neurons, which may normally be activated by an acute threatening stimulus.

Baro-excited neurons in the caudal ventrolateral medulla (CVLM) recorded using the whole-cell patch-clamp technique.

Caudal ventrolateral medulla (CVLM) neurons have important roles in the regulation of sympathetic nerve activity and blood pressure through their tonic inhibition of rostral ventrolateral medulla neurons. As few reports have demonstrated CVLM neuronal activity using the whole-cell patch-clamp technique, we attempted to find neurons in the CVLM that are depolarized by the stimulation of baroreceptors. To record the membrane potentials of the neurons in the CVLM, we developed a modified brainstem-spinal cord preparation that enabled us to change the pressure exerted on the aortic arch and carotid sinuses. We were able to identify neurons in the CVLM in which they were depolarized and the action potential (AP) frequency was increased upon baroreceptor stimulation. We referred to these neurons as baro-excited CVLM neurons. When these preparations were superfused with an angiotensin-II (Ang-II) solution, the frequency of the APs increased in 10 of the 14 baro-excited CVLM neurons. Superfusion with a low-Ca(2+), high-Mg(2+) solution abolished the APs in all seven baro-excited CVLM neurons, suggesting that the baro-excited CVLM neurons did not fire spontaneously. When the preparation was superfused with a low-Ca(2+) solution, 6 of the 7 baro-excited CVLM neurons did not respond to Ang-II superfusion. We for the first time found the baro-excited CVLM neurons, which depolarized pressure dependently but may not fire spontaneously. As Ang-II did not change the activity of the CVLM neurons during superfusion with a low-Ca(2+), high-Mg(2+) solution, the presynaptic neurons may be mandatory for the Ang-II-induced activation of postsynaptic baro-excited CVLM neurons.

Importance of rostral ventrolateral medulla neurons in determining efferent sympathetic nerve activity and blood pressure.

Accentuated sympathetic nerve activity (SNA) is a risk factor for cardiovascular events. In this review, we investigate our working hypothesis that potentiated activity of neurons in the rostral ventrolateral medulla (RVLM) is the primary cause of experimental and essential hypertension. Over the past decade, we have examined how RVLM neurons regulate peripheral SNA, how the sympathetic and renin-angiotensin systems are correlated and how the sympathetic system can be suppressed to prevent cardiovascular events in patients. Based on results of whole-cell patch-clamp studies, we report that angiotensin II (Ang II) potentiated the activity of RVLM neurons, a sympathetic nervous center, whereas Ang II receptor blocker (ARB) reduced RVLM activities. Our optical imaging demonstrated that a longitudinal rostrocaudal column, including the RVLM and the caudal end of ventrolateral medulla, acts as a sympathetic center. By organizing and analyzing these data, we hope to develop therapies for reducing SNA in our patients. Recently, 2-year depressor effects were obtained by a single procedure of renal nerve ablation in patients with essential hypertension. The ablation injured not only the efferent renal sympathetic nerves but also the afferent renal nerves and led to reduced activities of the hypothalamus, RVLM neurons and efferent systemic sympathetic nerves. These clinical results stress the importance of the RVLM neurons in blood pressure regulation. We expect renal nerve ablation to be an effective treatment for congestive heart failure and chronic kidney disease, such as diabetic nephropathy.

Activation of 5-hydroxytryptamine-1A receptors suppresses cardiovascular responses evoked from the paraventricular nucleus.

Activation of central 5-hydroxytryptamine-1A (5-HT(1A)) receptors powerfully inhibits stress-evoked cardiovascular responses mediated by the dorsomedial hypothalamus (DMH), as well as responses evoked by direct activation of neurons within the DMH. The hypothalamic paraventricular nucleus (PVN) also has a crucial role in cardiovascular regulation and is believed to regulate heart rate and renal sympathetic activity via pathways that are independent of the DMH. In this study, we determined whether cardiovascular responses evoked from the PVN are also modulated by activation of central 5-HT(1A) receptors. In anesthetized rats, the increases in heart rate and renal sympathetic nerve activity evoked by bicuculline injection into the PVN were greatly reduced (by 54% and 61%, respectively) by intravenous administration of (±)-8-hydroxy-2-(di-n-propylamino)tetralin (8-OH-DPAT), an agonist of 5-HT(1A) receptors, but were then completely restored by subsequent administration of WAY-100635, a selective antagonist of 5-HT(1A) receptors. Microinjection of 8-OH-DPAT directly into the PVN did not significantly affect the responses to bicuculline injection into the PVN, nor did systemic administration of WAY-100635 alone. In control experiments, a large renal sympathoexcitatory response was evoked from both the PVN and DMH but not from the intermediate region in between; thus the evoked responses from the PVN were not due to activation of neurons in the DMH. The results indicate that activation of central 5-HT(1A) receptors located outside the PVN powerfully inhibits the tachycardia and renal sympathoexcitation evoked by stimulation of neurons in the PVN.

Topographical specificity of regulation of respiratory and renal sympathetic activity by the midbrain dorsolateral periaqueductal gray.

The midbrain periaqueductal gray (PAG) mediates the physiological responses to a wide range of stressors. It consists of four longitudinal columns that have different anatomical connections and functional properties. Previous anatomical and behavioral studies have led to the hypothesis that the dorsolateral PAG, but not the adjacent lateral and dorsomedial subregions, is a key center that integrates the behavioral response to acute psychological threatening stimuli. In this study, we tested whether, consistent with this hypothesis, activation of neurons in the dorsolateral PAG evokes a pattern of cardiovascular and respiratory responses that is distinct from that evoked from surrounding regions. Arterial pressure, heart rate, renal sympathetic nerve activity (RSNA), and phrenic nerve activity (PNA) were recorded simultaneously in urethane-anesthetized rats. Microinjections of very small amounts of d,l-homocysteic acid (750 pmol in 15 nl) were made in sites throughout the dorsomedial, dorsolateral, and lateral PAG subregions. Increases in RSNA of similar magnitude accompanied by small to moderate increases in arterial pressure and heart rate were evoked from all three PAG subregions. In contrast, large increases in both PNA burst rate (respiratory rate) and overall respiratory activity were evoked only from a highly circumscribed region that corresponded closely to the dorsolateral PAG subregion at an intermediate to caudal level. Within this region, the evoked increases in RSNA and respiratory activity were highly correlated (r = 0.914, P < 0.001), suggesting the possibility that a common population of "command neurons" within the dorsolateral PAG may generate both sympathetic and respiratory responses from this region.

Electrophysiological responses of sympathetic preganglionic neurons to ANG II and aldosterone.

The intermediolateral cell column (IML) of the spinal cord is an important area where sympathetic impulses propagate to peripheral sympathetic organs. ANG II and aldosterone are important components of the renin-angiotensin-aldosterone system (RAAS), which activate the sympathetic nervous system. Each is partly synthesized in the brain and plays a paracrine role in the regulation of blood pressure independently of RAAS in the periphery. Our purpose in the present study was to clarify the contributions of sympathetic preganglionic neurons in the IML (IML neurons) and the effects of ANG II and aldosterone on the sympathetic nervous system. To examine responses to ANG II and aldosterone, we intracellularly recorded 104 IML neurons using a whole cell patch-clamp technique in spinal cord slice preparations. IML neurons were classified into two types: silent and firing. Both neuron types were significantly depolarized by ANG II, and candesartan inhibited this depolarization. After pretreatment with TTX, firing neurons (but not silent neurons) were significantly depolarized by ANG II. Aldosterone significantly increased the number of excitatory postsynaptic potentials (EPSPs) in both neuron types, but this response disappeared after pretreatment with TTX. ANG II and aldosterone had no synergistic effects on the IML neurons. The silent neurons had large cell soma, and many more dendrites than the firing neurons. These results suggest that ANG II acts presynaptically and postsynaptically in IML neurons, while aldosterone acts mainly presynaptically. Thus, the physiological effects of these substances are likely to be transmitted via specific membrane receptors of IML and/or presynaptic neurons.

Relation of blood pressure quantitative trait locus on rat chromosome 1 to hyperactivity of rostral ventrolateral medulla.

Genetic factors that induce essential hypertension have been examined using genome-wide linkage analyses. A quantitative trait locus (QTL) region that is closely linked to hypertension has been found on chromosome 1 in stroke-prone spontaneously hypertensive rats (SHRSPs). We used 2 congenic rats in which the blood pressure QTL on rat chromosome 1 was introgressed from SHRSP/Izm to Wistar-Kyoto (WKY)/Izm (WKYpch1.0) and from WKY/Izm to SHRSP/Izm (SHRSPwch1.0) rats by repeated backcrossing. Previous studies reported that the intermediate phenotype of this QTL for hypertension is characterized by the hyperactivity of the sympathetic nervous system in response to physiological and psychological stress. We performed intracellular patch-clamp recordings of rostral ventrolateral medulla (RVLM) neurons from WKY, WKYpch1.0, SHRSPwch1.0, and SHRSPs and compared the basal electrophysiological activities of RVLM neurons and the responses of these neurons to angiotensin II. The basal membrane potential of RVLM neurons from WKYpch1.0 was significantly "shallower" than that of the neurons from WKY. The depolarization of RVLM neurons from WKYpch1.0 in response to angiotensin II was significantly larger than that in neurons from WKY rats, whereas the depolarization of RVLM neurons from SHRSPwch1.0 was significantly smaller than that in neurons from SHRSPs. The response to angiotensin II of RVLM neurons from WKYpch1.0 and SHRSPs was sustained even after the blockade of all of the synaptic transmissions using tetrodotoxin. The QTL on rat chromosome 1 was primarily related to the postsynaptic response of RVLM bulbospinal neurons to brain angiotensin II, whereas both the QTL and other genomic regions influenced the basal activity of RVLM neurons.

Candesartan and insulin reduce renal sympathetic nerve activity in hypertensive type 1 diabetic rats.

The nonlinearity of cardiovascular regulation is higher in normal physiology, whereas several diseases are characterized by a reduction in this nonlinearity. Reduced nonlinearity of heart rate regulation is a robust risk factor for high mortality in patients with myocardial infarction. We investigated the changes in linear and nonlinear correlations of cardiovascular regulation after administering drugs in hypertensive diabetic rats. Type 1 diabetes was induced in rats by intraperitoneally injecting spontaneously hypertensive rats with streptozotocin. The animals were then divided into 4 groups and each group was given vehicle, candesartan, amlodipine, or insulin for 2 weeks. Blood pressure, heart rate, renal sympathetic nerve activity, and renal blood flow were simultaneously recorded in the conscious state, and the linear and nonlinear correlations were compared by using coherence and the mutual information method. Candesartan and amlodipine decreased blood pressure to a similar extent, but renal sympathetic nerve activity was significantly lower in the candesartan group than in the vehicle group. The renal sympathetic nerve activity in the insulin group was also lower than in the vehicle group. There were no significant differences in linear correlation among the 4 groups. In contrast, the nonlinear correlations between renal sympathetic nerve activity and blood pressure in the candesartan group and the insulin group were significantly higher than in the vehicle group. Candesartan and insulin decreased renal sympathetic nerve activity and increased the nonlinearity. These results suggest that reducing the activity of renin-angiotensin system and insulin that lowers blood glucose level may improve autonomic nervous system dysfunction and neurohumoral regulation of the cardiovascular system in diabetic hypertensive rats.

Monosynaptic excitatory connection from the rostral ventrolateral medulla to sympathetic preganglionic neurons revealed by simultaneous recordings.

To directly investigate whether a monosynaptic connection exists between neurons in the rostral ventrolateral medulla (RVLM) and sympathetic preganglionic neurons (SPNs), we used simultaneous extracellular recordings of RVLM neurons and whole-cell patch-clamp recordings of SPNs at the Th2 level and analyzed them by spike-triggered averaging. We averaged 200 sweeps of membrane potentials in SPN triggered by the spikes in the RVLM neuron. No clear postsynaptic potentials were detected in the averaged wave of SPNs before angiotensin II (Ang II) superfusion, whereas during superfusion with Ang II (6 micromol/L) on the medulla oblongata side alone excitatory postsynaptic potentials (EPSPs) were clearly found in the SPN of 3 out of 10 pairs at 40 +/- 1 ms after the averaged triggering spike in the RVLM neuron. We consider them to be monosynaptic EPSPs, because 1) the averaged EPSPs exhibited a sharp rise time, 2) the onset latency of the averaged EPSPs in the SPNs after the trigger spike in the RVLM was the same as the latency of the antidromic action potentials in the RVLM neurons in response to electrical stimulation of the SPNs, and 3) the amplitude of the averaged EPSPs was over 2 mV. In summary, combining simultaneous recording and spike-triggered averaging allowed us to demonstrate a monosynaptic excitatory connection between a single RVLM neuron and a single SPN in the thoracic spinal cord. Such connections provide the basis for the maintenance of sympathetic tone and the integrative reflex that relays through the RVLM. The results explain the mechanism by which Ang II in the RVLM area increases peripheral sympathetic activity and blood pressure.

Novel axonal projection from the caudal end of the ventrolateral medulla to the intermediolateral cell column.

We used an optical imaging technique to investigate whether axons of neurons in the caudal end of the ventrolateral medulla (CeVLM), as well as axons of neurons in the rostral ventrolateral medulla (RVLM), project to neurons in the intermediolateral cell column (IML) of the spinal cord. Brain stem-spinal cord preparations from neonatal normotensive Wistar-Kyoto and spontaneously hypertensive rats were stained with a voltage-sensitive dye, and responses to electrical stimulation of the IML at the Th2 level were detected as changes in fluorescence intensity with an optical imaging apparatus (MiCAM-01). The results were as follows: 1) depolarizing responses to IML stimulation during low-Ca high-Mg superfusion were detected on the ventral surface of the medulla at the level of the CeVLM, as well as at the level of the RVLM, 2) depolarizing responses were also detected on cross sections at the level of the CeVLM, and they had a latency of 24.0 +/- 5.5 (SD) ms, 3) antidromic action potentials in response to IML stimulation were demonstrated in the CeVLM neurons where optical images were detected, and 4) glutamate application to the CeVLM increased the frequency of excitatory postsynaptic potentials (EPSPs) and induced depolarization of the IML neurons. The optical imaging findings suggested a novel axonal and functional projection from neurons in the CeVLM to the IML. The increase in EPSPs of the IML neurons in response to glutamate application suggests that the CeVLM participates in the regulation of sympathetic nerve activity and blood pressure and may correspond to the caudal pressor area.