Brain Slice Technique Research Articles

Electrical resonances in central auditory neurons.

In the auditory pathway, signal processing depends on the filter functions of neurons. We used frequency analysis to investigate the contributions of intrinsic membrane properties to the input-output relationships in neurons. The whole-cell tight-seal recording technique in brain slices of chicks was used to study neurons at four levels of the auditory pathway. Neurons displayed resonant peaks in their voltage responses to injected sinusoidal currents that swept through a specified frequency range. Higher resonant frequencies tended to predominate at relatively lower stations in the auditory pathway (approximately 100 Hz in the nucleus magnocellularis, 24 Hz in the nucleus laminaris, 6 Hz in the nucleus ovoidalis). Field L neurons (cortex homologue) displayed low pass filter characteristics without resonance. We propose that the subthreshold membrane resonances amplify synaptic inputs at specific frequencies and contribute to the chick's ability to decode temporal sound parameters.

Transient and long-lasting actions of 5-HT on rat dentate gyrus neurones in vitro.

1. The actions of 5-hydroxytryptamine (5-HT) on rat dentate gyrus neurones were measured with conventional intracellular recording techniques in brain slices maintained in vitro at 32 degrees C. 2. Bath application of 5-HT (0.3-100 microM) hyperpolarized the membrane potential and reduced the input resistance; these effects persisted in tetrodotoxin (1 microM) and were abolished by MDL 73,005EF, a 5-HT1A receptor antagonist. 3. Local application of 5-HT via a pressure pipette also elicited a hyperpolarization and a reduction in resistance, and evoked a transient 'burst' of spontaneous inhibitory postsynaptic potentials (IPSPs) which were blocked by tetrodotoxin or bicuculline. 4. The 'burst' of IPSPs was subject to desensitization. It was completely abolished in the presence of the 5-HT3 receptor antagonist dolasetron. 5. In some cells, a longer lasting increase in spontaneous IPSP frequency was observed during application of 5-HT; this effect was blocked by the 5-HT2 receptor antagonist MDL 100,907. 6. 5-HT (30 microM) shortened the decay time constants of the glutamatergic and GABAergic evoked EPSPs and IPSPs without changing their amplitudes. 7. It is concluded that 5-HT hyperpolarizes granule cells via postsynaptic 5-HT1A receptors and increases spontaneous GABA release from inhibitory interneurones via the activation of 5-HT3 receptors and/or 5-HT2 receptors.

Combining patch-clamp and optical methods in brain slices

Combining patch-clamp and optical imaging techniques in brain slices offers several advantages for physiological studies of nerve cells. Numerous practical considerations weigh heavily in this design of an apparatus suitable for such combined measurements. These considerations include the thickness of the slices, the type of microscope to be used for imaging and the kind of optical signal to be measured. A system that combine optical and patch-clamp methods can be modified readily to permit studies of intracellular and extracellular signaling pathways via flash photolysis of caged compounds.

Differences in Ca 2+ permeability of AMPA-type glutamate receptor channels in neocortical neurons caused by differential GluR-B subunit expression

Fast excitatory synaptic transmission in the CNS is mediated by AMPA-type glutamate receptor (GIuR) channels. Heterologous expression suggested that the Ca 2+ permeability of these receptors critically depends on the sub-unit composition. Using patch-clamp techniques in brain slices, we found that the Ca 2+ permeability of native AMPA-type GluRs was markedly higher in nonpyramidal ( P Ca P K ≈ 0.63 ) than in pyramidal ( P CaG P K ≈ 0.05 ) neurons of rat neocortex. Analysis of mRNA in single cells indicated that the relative abundance of GIuR-B-specific mRNA was significantly lower in nonpyramidal ( GluR-B GluR-non-B ≈ 0.3 ) than in pyramidal ( sol GluR-B GluRnon-B ≈ 3 ) cells. This suggests that differences in relative abundance of GIuR-B-specific mRNA generate functional diversity of AMPA-type GIuRs in neurons with respect to Ca 2+ permeability.

Chronic cerebral hypoperfusion and impaired neuronal function in rats.

Studies in acute cerebral ischemia have shown that reductions in cerebral blood flow of up to 50% do not lead to infarction or alterations in neuronal electric activity. Little is known about the effects of chronic reductions in cerebral blood flow. The purpose of this study was to evaluate neuronal electrophysiological function in brain that had been subjected to a chronic reduction of cerebral blood flow of less than 50%. Based on existing knowledge of thresholds of cerebral ischemia, neuronal electrophysiological function should be unaffected by hypoperfusion of this magnitude. An arteriovenous fistula model in the rat was used to induce chronic cerebral hypoperfusion with reductions of cerebral blood flow of 25% to 50% as measured previously by 14C-labeled autoradiography. Using in vitro electrophysiological brain slice techniques, long-term potentiation in hippocampal CA1 neurons was examined extracellularly after 6 months of chronic noninfarctional cerebral hypoperfusion. Brains were also examined histologically at this time for evidence of cerebral infarction. There was no evidence of cerebral infarction. Long-term potentiation was produced in 9 of 12 control animals and only 2 of 8 hypoperfused animals. This difference was significant (P < .05) and demonstrated that long-term potentiation was impaired in animals with chronic hypoperfusion. Noninfarctional reductions in cerebral blood flow of up to 50% do impair neuronal function in chronic cerebral ischemia, a result quite distinct from that seen in acute ischemia. The threshold for neuronal dysfunction in chronic cerebral hypoperfusion is lower than that found in acute cerebral ischemia, suggesting that duration as well as severity of ischemic insult determines cellular viability. Chronic hypoperfusion may lead to a noninfarctional state with impaired neuronal function, a category of chronic cerebral ischemia not previously identified.

Angiotensin II responsiveness of rat paraventricular and subfornical organ neurons in vitro

The responsiveness of neurons in the hypothalamic paraventricular nucleus to angiotensin II was investigated using extracellular single unit recording techniques in rat brain slices. Bath application of angiotensin II at a concentration of 3 × 10 −7M for 2–5 min resulted in excitatory responses in 50.4% of 141 paraventricular cells tested. The mean increase in firing rate was 2.12 ± 0.20 (mean ± S.E.M.) spikes/s, which represents a mean increase in activity of 149.8 ± 16.5%. Angiotensin II-sensitive neurons usually dispkyed irregular, phasic, or very slow spontaneous activity, with the majority of these neurons located in the magnocellular region. Under physiological blockade of synaptic transmission with low Ca 2+/high Mg 2+ medium, neuronal responses to this peptide remained in 12 (92.3%) of 13 cells tested. Application of three successive doses of angiotensin II ranging from 3 × 10 −9−3 × 10 −7M showed that neuronal responses were dose-dependent with an estimated threshold of 10 −8M. In comparison with angiotensin III, angiotensin II not only stimulated more paraventricular cells, but usually induced larger excitatory responses. Angiotensin II subtype 1 receptor antagonist losartan completely blocked angiotensin II responsiveness in each of 14 paraventricular cells tested whereas PD 123319, an angiotensin II subtype 2 receptor antagonist, exhibited a partial inhibitory effect in about one half of another 13 cells. In addition, single unit in vitro subfornical organ recordings demonstrate that angiotensin II evokes greater excitatory responses than in the paraventricular nucleus and that these effects are abolished by losartan application. These results support the hypothesis that within both the paraventricular nucleus and subfornical organ angiotensin II is a bioactive peptide which modulates neuronal activity and thus may exert significant control over neuroendocrine and autonomic functions.

Platelet-activating factor inhibits excitatory transmission in rat neostriatum

Platelet-activating factor (PAF) and its receptor have been found in mammalian brain but its neuronal action is unclear. Using brain slice and whole-cell patch-clamp techniques, it was found that PAF dose-dependently inhibited excitatory postsynaptic current in neostriatal neurons but not the current induced by exogenously applied glutamate. These effects were also observed with the synthetic agonist 2-methylcarbamyl-PAF and were blocked by BN50739, a competitive antagonist at PAF receptors. It is concluded that PAF acts presynaptically to inhibit the release of glutamate at excitatory synapses in the neostriatum.

The organization of the suprachiasmatic circadian pacemaker of the rat and its regulation by neurotransmitters and modulators.

The long-term goal of our research is to understand how cells of the suprachiasmatic nucleus (SCN) are organized to form a 24-hr biological clock, and what roles specific neurotransmitters and modulators play in timekeeping and resetting processes. We have been addressing these questions by assessing the pattern of spontaneous neuronal activity, using extracellular and whole-cell patch recording techniques in long-lived SCN brain slices from rats. We have observed that a robust pacemaker persists in the ventrolateral region of microdissected SCN, and have begun to define the electrophysiological properties of neurons in this region. Furthermore, we are investigating changing sensitivities of the SCN to resetting by exogenous neurotransmitters, such as glutamate, serotonin, and neuropeptide Y, across the circadian cycle. Our findings emphasize the complexity of organization and control of mammalian circadian timing.

Physiological and morphological properties of accumbens core and shell neurons recorded in vitro.

The morphology and electrophysiological properties of neurons in the nucleus accumbens were studied using intracellular recording techniques in rat brain slices maintained in vitro. Neurons were subdivided according to their location in the shell or core region of the nucleus accumbens. Most of the cells in both regions had small to medium-sized (15.8 +/- 2.8 microns) somata with densely spinous dendrites, somewhat similar to the striatal medium spiny neuron. However, minor morphological differences between neurons from accumbens core and shell regions were found, such as fewer primary dendrites in shell neurons than in the core (3.8 +/- 0.8 vs. 4.4 +/- 1.0) and the spatial organization of their dendritic trees. In general, the passive membrane properties of neurons in each region were similar. However, shell neurons appeared to be less excitable in nature, as suggested by (1) a faster time constant, (2) the absence of TTX-insensitive events resembling low-threshold spikes, and (3) the lower probability of evoking spikes in shell neurons by stimulation of amygdaloid or cortical afferents in comparison to the responses of core neurons to cortical afferent stimulation. In most nucleus accumbens neurons the action potentials evoked by membrane depolarization were preceded by a slow Ca(2+)-dependent depolarization and showed firing-frequency adaptation. Following TTX administration, all-or-none spike-like events resembling high-threshold calcium spikes were observed in both regions. In summary, except for minor differences, most of the properties of core and shell neurons are similar, supporting their characterization as subdivisions of a single structure. Therefore, differences in the functional properties of these neuronal populations are likely to be due to their distinct connectivity patterns.

Differential responses of neocortical neurons to glucose and/or O2 deprivation in the human and rat.

1. Intracellular recordings were performed in human and rat neocortical neurons with in vitro brain slice techniques. Baseline cellular properties and the effect of O2 and glucose deprivation on these neurons were studied. 2. Intracellular labelings of electrophysiologically identified neurons showed that most neurons recorded from layers 4 and 5 of the neocortex in both rats and humans were pyramidal cells with a regular-spiking or a burst firing pattern. 3. A period of complete anoxia (4-5 min) induced little or no change in membrane potential (Vm) in rat and human neocortical neurons, contrasting with the major depolarization we have previously observed in rat brainstem neurons during a similar period of anoxia. Evident depolarization occurred only when the slices were exposed to a more prolonged period of anoxia (> 7 min in rats and > 10 min in humans). 4. Membrane input resistance (Rm) of neocortical neurons decreased in both species during anoxia. In human neocortical neurons, Rm decreased by a mean of 22% with a marked increase in rheobase and suppression in spontaneous excitatory postsynaptic potentials (EPSPs). Interestingly, the increase in rheobase in human cells occurred even at an early stage (post 2-3 min in anoxia), when Vm and Rm had not yet changed. 5. Perfusing slices with a glucose-free medium for 1-2 h produced a relatively modest change in Vm (mean congruent to 28 mV). However, combined deprivation of both glucose and O2 resulted in a major depolarization (mean congruent to 50 mV) within 5-10 min in both human and rat neocortical neurons.(ABSTRACT TRUNCATED AT 250 WORDS)

Anatomy and physiology of multipolar cells in the rat inferior collicular cortex using the in vitro brain slice technique

Coronal brain slices from 21-50-d-old hooded rats were used to characterize intracellular responses of cells in both the external and dorsal cortices of the inferior colliculus (IC). These cells could generate both sodium and calcium spikes. Depending on current amplitude, depolarizing current pulses could elicit either phasic or tonic firing patterns, with spike frequency adaptation. Spiking also occurred at the offset of a hyperpolarizing pulse. These patterns were due primarily to the activation of calcium conductances. Stimulation of the commissural pathway connecting the left and right IC produced a short-latency monosynaptic IPSP followed by an EPSP(s) and a late polysynaptic IPSP(s). Non-NMDA glutamate antagonists eliminated or reduced the amplitude of the EPSP and the late portion of the inhibition, while both IPSPs were blocked by GABAA antagonists. As described previously in guinea pig (Smith, 1986) and rat (Pierson et al., 1989), a large NMDA-mediated depolarizing event (paroxysmal depolarizing shift, or PDS) could be elicited by shocking the commissure of the IC in the presence of picrotoxin or bicuculline, NMDA, 4-aminopyridine, or in 0 Mg2+ Ringer's. The picrotoxin-induced PDS was significantly reduced or abolished in Ringer's containing aminophosphonovalerate. Cells displaying the responses described were labeled with neurobiotin. Those labeled are medium-sized multipolar cells. Their dendrites are usually spiny and can extend superficially up to the cortical surface. Their thin axons give rise to collaterals that branch profusely within the cortex. The main axons project laterally along the circumference of the IC or medially into the commissure separating the collicular hemispheres.

Development of GABA‐mediated, chloride‐dependent inhibition in CA1 pyramidal neurones of immature rat hippocampal slices.

1. gamma-Aminobutyric acid (GABA)-mediated, Cl(-)-dependent inhibitory postsynaptic potentials (IPSPs) and GABA currents in immature rat hippocampal CA1 neurones were studied using the whole-cell recording technique in brain slices. 2. IPSPs evoked by electrical stimulation were observed in postnatal 2- to 5- (PN2-5), 8- to 13-(PN8-13) and 15- to 20-(PN15-20)day-old CA1 neurones. In the presence of glutamate receptor blockers 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) and D-2-amino-5-phosphonovaleric acid (APV), the reversal potential for the IPSP (EIPSP) was near the resting membrane potential (RMP) in the PN2-5 neurones, but 13 and 25 mV more negative than the RMP in PN8-13 and PN15-20 neurones respectively. IPSPs and GABA currents were blocked by the GABAA-receptor antagonists bicuculline or picrotoxin. 3. The reversal potential for somatic GABA currents (EGABA) was examined in the presence of tetrodotoxin (TTX). There was a strong dependence of the EGABA upon the patch pipette [Cl-] ([Cl-]p). indicating that the GABA currents were mediated by a Cl- conductance. In PN2-5 neurones, EGABA agreed with the value predicted by the Goldman-Hodgkin-Katz equation at given concentrations of internal and external anions permeable through GABA-activated Cl- channels, whereas EGABA in older neurones was 8-18 mV more negative. 4. Examination of the relations between EGABA, holding potential, [Cl-]p and resting conductance indicated that the membrane of the PN2-5 neurones was readily permeable to Cl- which followed a passive Donnan equilibrium. Passive distribution of Cl- played a decreasing role in PN8-13 neurones and in PN15-20 neurones. 5. To assess the contribution of outward Cl- co-transport, bath applications of high K+ or furosemide were performed. High K+ and furosemide caused a reversible positive shift of EGABA in PN15-20 neurones. Raising the temperature moved EGABA to a more negative potential, with a Q10 of 5 mV. A similar change of EGABA in response to high K+, but not to furosemide, was found in PN8-13 neurones. 6. The present data indicate the existence of GABAA-mediated inhibitory synaptic connections in CA1 neurones at the earliest stages of postnatal life. During the first postnatal week, Cl- ions are passively distributed and the EIPSP and EGABA are near the RMP.(ABSTRACT TRUNCATED AT 400 WORDS)

Electrical properties of axons within Probst bundles of acallosal mice and callosi that have reformed upon glial-coated polymer implants

Lesioning of the developing cerebral midline and corpus callosum of mice results in the formation of bilateral masses of axons known as Probst's longitudinal bundles. These ectopic axons were examined using an in vitro brain slice technique, and they were found to be functional and to have electrical properties that were similar to those of the unlesioned corpus callosum at a comparable age. When a nitrocellulose bridge is properly oriented at the site of the callosal lesion in neonates, the implant will support the migration of glianates, in turn, promote the redirection of callosal axons across the midline. These axons were also analyzed by in vitro brain slice techniques, and they were found to be functional and to have electrical properties that were similar to the unlesioned corpus callosum with respect to conduction velocity and chronaxie but dissimilar with respect to charge threshold and rheobase current.

5-Hydroxytryptamine 2 and 5-hydroxytryptamine 1A receptors mediate opposing responses on membrane excitability in rat association cortex

The effects of serotonin on pyramidal cells of layer V of the medial prefrontal cortex were examined using intracellular recording techniques in rat brain slices in vitro. Bath administration of serotonin (0.3–100 μM) produced two distinct responses which could be differentiated physiologically and pharmacologically. The first of these responses was a membrane hyperpolarization. This effect of serotonin was associated with a decrease in input resistance and was independent of the transmembrane chloride gradient, suggesting that it was mediated by an increase in potassium conductance. The ability of serotonin to induce a hyperpolarization was mimicked by (±)-8-hydroxy-dipropylaminotetralin hydrobromide and was blocked by BMY 7378 and spiperone but not by ketanserin, indicating that it was mediated by the activation of receptors of the 5-hydroxytryptamine 1A subtype. The second response to serotonin involved a membrane depolarization, the replacement of the afterhyperpolarization that follows a burst of spikes in these cells by a slow depolarizing afterpotential, and a decrease in spike frequency accommodation. These effects were mimicked by 4-bromo-2,5-dimethoxyphenyl-isopropylamine and antagonized by ketanserin and by low concentrations of spiperone, indicating that they were mediated by the activation of 5-hydroxytryptamine 2 receptors. Interestingly, qualitatively identical responses could be elicited in these cells by activation of muscarinic and α 1-adrenergic receptors suggesting that 5-hydroxytryptamine 2, muscarinic and α 1-adrenergic receptors converge onto a common set of membrane mechanisms to increase cellular excitability. Although 5-hydroxytryptamine 1A and 5-hydroxytryptamine 2 receptors mediated opposing effects on membrane excitability, most pyramidal neurons appeared to express both receptor subtypes on their membrane surface. The coactivation of both receptor subtypes resulted in a selective enhancement of responsiveness to strong excitatory stimuli with little effect on weaker stimuli. The paradoxical presence of two serotonin receptors mediating opposite effects on membrane excitability in the same cell provides a flexible mechanism by which serotonin might regulate how pyramidal neurons encode incoming excitatory stimuli onto firing activity.

The lateral spread of ictal discharges in neocortical brain slices

The in vitro brain slice technique was used to examine the lateral propagation of spontaneous electrographic ictal episodes across adjacent areas of guinea pig neocortex. Epileptiform activity was induced by perfusing slices with Mg-free artificial CSF. Simultaneous field potential recordings of ictal episodes were obtained from 4 micropipettes placed 1–3 mm apart across coronal slices in middle-cortical layers. Two types of lateral spread were characterized. Ictal episodes often developed focally and then spread as a slowly moving wavefront traveling at <0.3 mm/sec into adjacent, uninvolved cortex. By contrast, other episodes began nearly synchronously at all cortical sites. The individual afterdischarges that composed each ictal episode propagated rapidly across the cortex at >30 mm/sec and were triggered by multiple pacemakers. Ictal episodes always terminated abruptly across the entire slice. The NMDA-receptor antagonist, 2-amino-phosphono-valerate, applied focally between recording sites, blocked rapid propagation across treated areas and resulted in the emergence of spatially separate, independent pacemakers. Pacemaker failure is the proposed mechanism for simultaneous and generalized termination of ictal episodes in this in vitro model of epileptogenesis.

Depolarization and stimulation of neurons in nucleus tractus solitarii by carbon dioxide does not require chemical synaptic input

The effects of elevated CO 2(i.e. hypercapnia) on neurons in the nucleus tractus solitarii were studied using extracellular ( n = 82) and intracellular ( n = 33) recording techniques in transverse brain slices prepared from rat. Synaptic connections from putative chemosensitive neurons in the ventrolateral medulla were removed by bisecting each transverse slice and discarding the ventral half. In addition, the response to hypercapnia in 20 neurons was studied during high magnesium low calcium synaptic blockade. Sixty-five per cent of the neurons ( n = 75) tested were either insensitive or inhibited by hypercapnia. However, 35% ( n = 40) were depolarized and/or increased their firing rate during hypercapnia. Nine out of 10 CO 2 excited neurons retained their chemosensitivity to CO 2 in the presence of high magnesium low calcium synaptic blockade medium. Our findings demonstrate that many neurons in the nucleus tractus solitarii were depolarized and/or increased their firing rate during hypercapnia. These neurons were not driven synaptically by putative chemosensitive neurons of the ventrolateral medulla since this region was removed from the slice. Furthermore, because chemosensitivity persisted in most neurons tested during synaptic blockade, we conclude that some neurons in the nucleus tractus solitarii are inherently CO 2-chemosensitive. Although the function of dorsal medullary chemosensitive neurons cannot be determined in vitro, their location and their inherent chemosensitivity suggest a role in cardiorespiratory central chemoreception.

The mammalian circadian clock in the suprachiasmatic nuclei is reset in vitro by cAMP

A circadian clock located in the suprachiasmatic nuclei (SCN) of the hypothalamus controls the daily behavioral and physiological rhythms of mammals. While the mammalian circadian system has been the focus of research for many years, very little work has been directed at understanding its underlying biochemical mechanisms. In these experiments we used the hypothalamic brain slice technique to investigate these mechanisms, focusing specifically on the intrinsic resetting properties of the circadian clock of the rat. We monitored a primary expression of the clock or pacemaker, the circadian rhythm of electrical activity of SCN neurons. This rhythm continues the oscillatory pattern seen in vivo for up to 60 hr in vitro, with an activity peak near midday that shows very little variation among SCN from different rats. The stability of the rhythm in vitro enabled us to use the time of peak activity to monitor the phase of the underlying pacemaker. Bath application of membrane-soluble cAMP analogs in 1 hr pulses induced robust advances in the phase of the rhythm that remained stable for 2 cycles. This effect depended on the phase of the pacemaker at the time of treatment: the peak was maximally advanced (4-6 hr) by treatments during the middle of the subjective day (projected from the donor's cycle); treatments during most of the subjective night and early subjective day induced no phase changes. Half-maximal phase resetting was induced at 1 x 10(-10) M concentrations of active cAMP analog.(ABSTRACT TRUNCATED AT 250 WORDS)

Effect of amino acid derivatives of ?-carboline on evoked activity of hippocampal neurons

The action of methylamide-beta-carboline-3-carboxylate (1), glycinamide-beta-carboline-3-carboxylate methyl ester (2), leucinamide-beta-carboline-3-carboxylate methyl ester (3) on evoked activity of hippocampal neurons was investigated using brain slice technique. Both 1 and 2 application (5 mkM, 15 min) reduced paired-pulse inhibition (PPI). Single-pulse and frequency stimulation revealed population spike (PS) increase or generation of additional PS after 1 or 2 application, the former drug being more active at all stimulation patterns. 3 had no definitive activity in the test-system used. The data obtained suggest similarity of pharmacological profiles of 1 and 2, and anxiogenic activity of 2 in vivo.

Direct excitatory action of vasopressin in the lateral septum of the rat brain

The electrophysiological action of arginine vasopressin on neurones in the lateral septum of the rat brain was studied using extracellular recordings and the in vitro brain slice technique. Of 177 neurones tested in the presence of vasopressin at 1–1000 nM, 77 (about 44%) responded by a reversible increase in firing rate, 12 (about 7%) were inhibited and the remaining were not affected. The lowest peptide concentration effective in exciting septal neurones ranged between 1 and 50 nM, and the magnitude of the excitatory effect was concentration dependent. At high vasopressin concentrations, the peptide-induced excitation was often followed by a transient pause in firing; this was probably due to action potential inactivation, brought about by the vasopressin-induced neuronal membrane depolarization. The excitatory effect of vasopressin was postsynaptic, since it was not abolished following synaptic blockade in a low calcium-high magnesium perifusion solution. A comparison of the effects of vasopressin and oxytocin suggested that most of the septal vasopressin-sensitive neurones are endowed with vasopressin receptors, whereas a minority of them bear oxytocin receptors.

Neuron responses to substance P and enkephalin in rat dorso-lateral septum in vitro

Using an in vitro brain slice technique the responses of spontaneously active neurons in the rat dorso-lateral septum to 10 nM substance P (SP) and enkephalin were determined. Fewer neurons responded to SP (41%) than to enkephalin (55%). The SP responses were 13 excitations, 14 inhibitions, the enkephalin responses were 13 excitations, 14 inhibitions and 11 responded to both, 6 of these were inhibited by both. Immunocytochemical techniques have shown there is a discrete localisation of SP and enkephalin axons and terminals in the rat septum. SP responsive neurons were associated with the SP terminal-rich region ( p = 0.01) but no association was found for enkephalin responses in the enkephalin terminal-rich region ( p = 0.7).