Electrophysiological localization of distinct calcium potentials at selective somatodendritic sites in the substantia nigra

Biophysical Properties of Somatic and Axonal Voltage-Gated Sodium Channels in Midbrain Dopaminergic Neurons

Spiking activities of midbrain dopaminergic neurons are critical for key brain functions including motor control and affective behaviors. Voltage-gated Na⁺ channels determine neuronal excitability and action potential (AP) generation. Previous studies on dopaminergic neuron excitability mainly focused on Na⁺ channels at the somatodendritic compartments. Properties of axonal Na⁺ channels, however, remain largely unknown. Using patch-clamp recording from somatic nucleated patches and isolated axonal blebs from the axon initial segment (AIS) of dopaminergic neurons in mouse midbrain slices, we found that AIS channel density is approximately 4–9 fold higher than that at the soma. Similar voltage dependence of channel activation and inactivation was observed between somatic and axonal channels in both SNc and VTA cells, except that SNc somatic channels inactivate at more hyperpolarized membrane potentials (V_m). In both SNc and VTA, axonal channels take longer time to inactivate at a subthreshold depolarization V_m level, but are faster to recover from inactivation than somatic channels. Moreover, we found that immunosignals of Nav1.2 accumulate at the AIS of dopaminergic neurons. In contrast, Nav1.1 and Nav1.6 immunosignals are not detectible. Together, our results reveal a high density of Na⁺ channels at the AIS and their molecular identity. In general, somatic and axonal channels of both SNc and VTA dopaminergic neurons share similar biophysical properties. The relatively delayed inactivation onset and faster recovery from inactivation of axonal Na⁺ channels may ensure AP initiation at high frequencies and faithful signal conduction along the axon.

Striatal and extrastriatal dopamine in the basal ganglia: an overview of its anatomical organization in normal and Parkinsonian brains

Degeneration of the nigrostriatal dopaminergic system is the characteristic neuropathological feature of Parkinson's disease and therapy is primarily based on a dopamine replacement strategy. Dopamine has long been recognized to be a key neuromodulator of basal ganglia function, essential for normal motor activity. The recent years have witnessed significant advances in our knowledge of dopamine function in the basal ganglia. Although the striatum remains the main functional target of dopamine, it is now appreciated that there is dopaminergic innervation of the pallidum, subthalamic nucleus, and substantia nigra. A new dopaminergic- thalamic system has also been uncovered, setting the stage for a direct dopamine action on thalamocortical activity. The differential distribution of D1 and D2 receptors on neurons in the direct and indirect striato-pallidal pathways has been re-emphasized, and cholinergic interneurons are recognized as an intermediary mediator of dopamine-mediated communication between the two pathways. The importance and specificity of dopamine in regulating morphological changes in striatal projection neurons provides further evidence for the complex and multifarious mechanisms through which dopamine mediates its functional effects in the basal ganglia. In this review, the role of basal ganglia dopamine and its functional relevance in normal and pathological conditions will be discussed.

In vivo neurochemical monitoring and the study of behaviour

In vivo neurochemical monitoring techniques measure changes in the extracellular compartment of selected brain regions. These changes reflect the release of chemical messengers and intermediates of brain energy metabolism resulting from the activity of neuronal assemblies. The two principal techniques used in neurochemical monitoring are microdialysis and voltammetry. The presence of glutamate in the extracellular compartment and its pharmacological characteristics suggest that it is released from astrocytes and acts as neuromodulator rather than a neurotransmitter. The changes in extracellular noradrenaline and dopamine reflect their role in the control of behaviour. Changes in glucose and oxygen, the latter a measure of local cerebral blood flow, reflect synaptic processing in the underlying neuronal networks rather than a measure of efferent output from the brain region. In vivo neurochemical monitoring provides information about the intermediate processing that intervenes between the application of the stimulus and the resulting behaviour but does not reflect the final efferent output that leads to behaviour.

Functional interactions between somatodendritic dopamine release, glutamate receptors and brain-derived neurotrophic factor expression in mesencephalic structures of the brain

Dopaminergic nigrostriatal neurons may be considered as bipolar functional entities since they are endowed with the ability to synthesize, store and release the transmitter dopamine (DA) at the somatodendritic level in the substantia nigra (SN). Such dendritic DA release seems to be distinct from the transmitter release occurring at the axon terminal and seems to rely preferentially on volume transmission to exert its physiological effects. An increased glutamatergic (Gluergic) transmission into the SN facilitates such dendritic DA release via activation of NMDA-receptors (NMDA-Rs) and to a lesser extent through group II metabotropic glutamate receptors (mGluRs). In addition, nigral mGluRs functionally interact with NMDA-Rs in the SN, further modulating the NMDA-R-mediated increase of DA release from dendrites in the SN. In turn, dendritically released DA may exert, via D1 receptors, a tonic inhibitory control upon nigral glutamate (Glu). Furthermore, released DA, via D2/D3 autoreceptors, produces an autoinhibitory effect upon DA cell firing and its own release process. An increased Gluergic transmission into the SN may also induce, via activation of NMDA-Rs, an augmented expression of different brain-derived neurotrophic factor (BDNF) gene transcripts in this brain area. Pharmacological evidence suggests that non-NMDA-Rs could also participate in the regulation of BDNF gene expression in the SN. Glu-mediated changes of nigral BDNF expression could regulate, in turn, the expression of important transmitter-related proteins in the SN, such as different NMDA-R subunits, mGluRs and DA-D3 receptors. In conclusion, Glu-DA-BDNF interactions in the SN may play an important role in modulating the flow of neuronal information in this brain structure under normal conditions, as well as during adaptive and plastic responses associated with various neurological and psychiatric disorders.

Glutamate-mediated [Ca2+]c dynamics in spontaneously firing dopamine neurons of the rat substantia nigra pars compacta

The mechanism by which glutamate regulates the cytosolic free Ca2+ concentration ([Ca2+]c) in spontaneously firing dopamine neurons is not clear. Thus we have investigated the glutamate-mediated [Ca2+]c dynamics in the acutely isolated dopamine neurons from the rat substantia nigra pars compacta by measuring [Ca2+]c and spontaneously occurring action potentials (SAPs). The freshly isolated dopamine neurons showed tetrodotoxin (TTX)-sensitive spontaneous firing of 2-3 Hz and the resting [Ca2+]c decreased with abolition of the SAPs. The level of [Ca2+]c was affected by the spontaneous firing rate. In the presence of the Na+ channel antagonist, TTX (0.5 microM), glutamate increased [Ca2+]c by activating different glutamate receptors depending on the glutamate concentration used. Addition of glutamate at low concentrations (<3 microM) raised [Ca2+]c mainly by activating metabotropic glutamate receptors (mGluR), whereas at high concentrations (>10 microM) it raised [Ca2+]c mainly by activating AMPA/kainate receptors. The contribution of NMDA receptors to the glutamate-mediated [Ca2+]c rises was largest at intermediate concentrations of glutamate. Activation of mGluR elicited a Ca2+ release from intracellular Ca2+ stores and continuous Ca2+ influx out of the cell. The spontaneous firing activities were highly enhanced by submicromolar levels of glutamate and abolished at levels above 10 microM. From these results, we conclude that at low glutamate concentrations the [Ca2+]c in the dopamine neurons is mainly governed by mGluR and the firing activities, whose rate is regulated at submicromolar glutamate concentrations, but at higher glutamate concentrations [Ca2+]c is dominantly affected by AMPA/kainate receptors.

K+ channel activation and low-threshold Ca2+ spike of rat cerebellar Purkinje cells in vitro

Using the whole-cell configuration of the patch-clamp recording method, we analyzed the role of K+ conductances in determining the characteristics of the dendritically-initiated low-threshold Ca+ spike (LTS) recorded at the somatic level of rat cerebellar Purkinje cells (PCs) in slice cultures. Blockade of tetra-ethyl-ammonium-(TEA)- and 4-aminopyridine-(4-AP)-sensitive K+ channels increased the amplitude of the LTS. This effect was prominent with 4-AP, which promotes the fast-decaying component of the LTS. Surprisingly, a shortening of the LTS was induced by the blockade of K+ channel activity instead of a broadening of spikes as generally observed. We propose that, when propagating to the soma, TEA- and 4-AP-sensitive K+ channel activity affects the electrical properties of dendrites such that the LTS is attenuated and slowed down.

Differential regulation of somatodendritic and nerve terminal dopamine release by serotonergic innervation of substantia nigra

Nigrostriatal dopaminergic neurons release dopamine from dendrites in substantia nigra and axon terminals in striatum. The cellular mechanisms for somatodendritic and axonal dopamine release are similar, but somatodendritic and nerve terminal dopamine release may not always occur in parallel. The current studies used in vivo microdialysis to simultaneously measure changes in dendritic and nerve terminal dopamine efflux in substantia nigra and ipsilateral striatum respectively, following intranigral application of various drugs by reverse dialysis through the nigral probe. The serotonin releasers (+/-)-fenfluramine (100 micro m) and (+)-fenfluramine (100 micro m) significantly increased dendritic dopamine efflux without affecting extracellular dopamine in striatum. The non-selective serotonin receptor agonist 1-(m-chlorophenyl)-piperazine (100 micro m) elicited a similar pattern of dopamine release in substantia nigra and striatum. NMDA (33 micro m) produced an increase in nigral dopamine of a similar magnitude to mCPP or either fenfluramine drug. However, NMDA also induced a concurrent increase in striatal dopamine. The D2 agonist quinpirole (100 micro m) had a parallel inhibitory effect on dopamine release from dendritic and terminal sites as well. Taken together, these data suggest that serotonergic afferents to substantia nigra may evoke dendritic dopamine release through a mechanism that is uncoupled from the impulse-dependent control of nerve terminal dopamine release.

Modulation of dendritic release of dopamine by metabotropic glutamate receptors in rat substantia nigra

A superfusion system was used to study the effects of metabotropic glutamate receptor (mGluR) ligands upon the release of [(3)H]dopamine ([(3)H]DA) previously taken up by rat substantia nigra (SN) slices. trans-(+/-)-1-Amino-(1S,3R)-cyclopentane dicarboxylic acid (trans-ACPD; 100 and 600 microM), a group I and II mGluR agonist, evoked the release of [(3)H]DA from nigral slices. This last effect was reduced significantly by (2S,3S,4S)-2-methyl-2-(carboxycyclopropyl)-glycine (MCCG; 300 microM), an antagonist of group II mGluR, or by the addition of tetrodotoxin (D-APV; 1 microM) to the superfusion medium. D-(-)-2-Amino-5-phosphono-valeric acid (100 microM), an N-methyl-D-aspartate receptor antagonist, or the presence of Mg(2+) (1.2mM) in the superfusion medium did not modify trans-ACPD-induced [(3)H]DA release. In addition, a group II mGluR agonist such as (2S,1'R,2'R,3'R)-2-(2',3'-dicarboxycyclopropyl)-glycine (DCG-IV; 100 microM) significantly induced the release of [(3)H]DA from nigral slices, whereas a group I mGluR agonist such as (RS)-3,5-dihydroxyphenylglycine (DHPG; 50 and 100 microM) did not modify the release of the [(3)H]-amine. Further experiments showed that the NMDA (100 microM)-evoked release of [(3)H]DA was decreased significantly by prior exposure of SN slices to trans-ACPD. Finally, partial denervation of the DA nigro-striatal pathway with 6-hydroxydopamine (6-OH-DA) increased trans-ACPD-induced release of [(3)H]DA, whereas it decreased trans-ACPD inhibitory effects on NMDA-evoked release of [(3)H]DA from nigral slices. The present results suggest that the dendritic release of DA in the SN is regulated by mGluR activation. Such nigral mGluR activation may produce opposite effects upon basal and NMDA-evoked release of DA in the SN. In addition, such mGluR-induced effects in the SN are modified in response to partial denervation of the DA nigro-striatal pathway.

Regulation of action potential size and excitability in substantia nigra compacta neurons: sensitivity to 4-aminopyridine

Slow, pacemaker-like firing is due to intrinsic membrane properties in substantia nigra compacta (SNc) neurons in vitro. How these properties interact with afferent synaptic inputs is not fully understood. In this study, intracellular recordings from SNc neurons in brain slices showed that spontaneous action potentials (APs) were attenuated when generated from lower than normal threshold. Such APs were blocked by 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) and could be related to non-N-methyl-D-aspartate (NMDA) receptor-mediated spontaneous excitatory postsynaptic potentials (EPSPs). The AP attenuation was reproduced by stimulus-evoked EPSPs and by current injections to the soma. APs evoked from holding potentials between -40 and -60 mV were reduced in width by Cd(2+) (0. 2 mM). Tetraethylammonium chloride (TEA, 10 mM) or 4-aminopyridine (4-AP, 5 mM) increased the AP width. However, at more negative holding potentials, Cd(2+) and TEA were inefficacious, whereas 4-AP enlarged the AP, partly via induction of a Cd(2+)-sensitive component. A monophasic afterhyperpolarization (AHP), following attenuated APs, was little affected by either Cd(2+) or TEA, but inhibited by 4-AP, which induced an additional, slow component, sensitive to Cd(2+) or apamin (100 nM). The AP delay showed a discontinuous relation to the amplitude or slope of the injected current (delay shift), which was sensitive to low doses of 4-AP (0. 05 mM). The initial time window before the delay shift was longer than the rise time of EPSPs. It is suggested that a 4-AP-sensitive current prevents or postpones discharge during slow depolarization's, but allows direct excitation by fast EPSPs. Fast excitation leads to AP attenuation, primarily due to strong activation of 4-AP-sensitive current. This seems to cause inhibition of the Ca(2+) current during the AP and reduction of Ca(2+)-dependent K(+) currents. Together, these properties are likely to influence the excitability and the local, somatodendritic effects of the AP, in a manner that discriminates between firing induced by the intrinsic pacemaker mechanism and fast synaptic potentials.

Fast Na+ spike generation in dendrites of guinea-pig substantia nigra pars compacta neurons

Electric fields were applied to study the regenerative properties of substantia nigra pars compacta neurons in guinea-pig brain slices. Two types of spikes, of high or low amplitude, were generated in both the soma-hyperpolarizing and the soma-depolarizing directions of the field. The different sensitivity of the spikes to somatic polarization suggested that the high-amplitude spikes were generated near the cell body, whereas the low-amplitude spikes were generated at a distance from the soma. Application of tetrodotoxin or intracellular injection of QX 314 abolished both types of spike. The spikes were not inhibited in the presence of glutamate receptor antagonists or during Ca2+ channel blockade. Blockers of gap junctional conductance (sodium propionate, octanol and halothane) did not affect the field-induced spikes. The spike generation was highly sensitive to changes in membrane conductance induced by current injection in the soma or by external field application. The ability of a conditioning field stimulation to affect the spike generation in different neuronal compartments suggested that a transient outward current was generated in the dendrites. The field-induced spikes were facilitated by synaptic stimulation and, in some neurons, low-amplitude spikes were generated by synaptic potentials in the absence of field application. These results suggest that channels responsible for Na+ spike generation reside in the dendrites, and are influenced by spatially distributed voltage-dependent K+ currents and by synaptic input.

Modeling N-methyl-D-aspartate-induced bursting in dopamine neurons

Burst firing of dopaminergic neurons of the substantia nigra pars compacta can be induced in vitro by the glutamate agonist N-methyl-D-aspartate. It has been suggested that the interburst hyperpolarization is due to Na+ extrusion by a ouabain-sensitive pump [Johnson et al. (1992) Science 258, 665-667]. We formulate and explore a theoretical model, with a minimal number of currents, for this novel mechanism of burst generation. This minimal model is further developed into a more elaborate model based on observations of additional currents and hypotheses about their spatial distribution in dopaminergic neurons [Hounsgaard (1992) Neuroscience 50, 513-518; Llinás et al. (1984) Brain Res. 294, 127-132]. Using the minimal model, we confirm that interaction between the regenerative, inward N-methyl-D-aspartate-mediated current and the outward Na(+)-pump current is sufficient to generate the slow oscillation (approximately 0.5 Hz) underlying the burst. The negative-slope region of the N-methyl-D-aspartate channel's current-voltage relation is indispensable for this slow rhythm generation. The time-scale of Na(+)-handling determines the burst's slow frequency. Moreover, we show that, given the constraints of sodium handling, such bursting is best explained mechanistically by using at least two spatial, cable-like compartments: a soma where action potentials are produced and a dendritic compartment where the slow rhythm is generated. Our result is consistent with recent experimental evidence that burst generation originates in distal dendrites [Seutin et al. (1994) Neuroscience 58, 201-206]. Responses of the model to a number of electrophysiological and pharmacological stimuli are consistent with known responses observed under similar conditions. These include the persistence of the slow rhythm when the tetrodotoxin-sensitive Na+ channel is blocked and when the soma is voltage-clamped at -60 mV. Using our more elaborate model, we account for details of the observed frequency adaptation in N-methyl-D-aspartate-induced bursting, the origin of multiple spiking and bursting mechanisms, and the interaction between two different bursting mechanisms. Besides reproducing several well established firing patterns, this model also suggests that new firing modes, not yet recorded, might also occur in dopaminergic neurons. This model provides mechanistic insights and explanations into the origin of a variety of experimentally observed membrane potential firing patterns in dopaminergic neurons, including N-methyl-D-aspartate-induced bursting and its dendritic origin. Such a model, capable of reproducing a number of realistic behaviors of dopaminergic neurons, could be useful in further studies of the basal ganglia-thalamocortical motor circuit. It may also shed light on bursting that involves N-methyl-D-aspartate channel activity in other neuron types.

Amygdala efferents form inhibitory-type synapses with a subpopulation of catecholaminergic neurons in the rat Nucleus tractus solitarius

The central nucleus of the amygdala (CNA) integrates visceral responses to stress partially through efferent projections to portions of the medial nuclei of the solitary tracts (mNTS) containing catecholaminergic neurons. To determine anatomical sites for CNA modulation of these neurons, immunoperoxidase detection of anterogradely transported Phaseolus vulgaris-leucoagglutinin (PHA-L) or biotinylated dextran amine (BDA) was combined with immunogold-silver labeling of the catecholamine-synthesizing enzyme, tyrosine hydroxylase, in adult rat mNTS. From 350 anterogradely labeled terminals identified within the intermediate mNTS, 30% formed symmetric, inhibitory-type synapses and the remainder lacked recognized junctions as seen within a single plane of section. Of the terminals forming symmetric synapses, 16% were presynaptic to tyrosine hydroxylase immunoreactive dendrites and the remainder to unlabeled dendrites. The level of tyrosine hydroxylase immunoreactivity as assessed by density of gold-silver particles was significantly lower in dendrites receiving synaptic input from CNA efferents as compared with dendrites of the same sizes (2.0 microns 2 in mean area) which received synapses from unlabeled terminals or lacked recognizable synaptic inputs. When separately examined without regard to afferent input, the medium- and larger-sized dendrites having mean cross-sectional areas of 1-3 microns 2 also contained significantly less tyrosine hydroxylase immunoreactivity than small (< 1 micron 2) dendrites. These results suggest that CNA efferents to the mNTS inhibit non-catecholamine-containing neurons and a subpopulation of catecholaminergic neurons distinguished by their low levels of tyrosine hydroxylase. The findings also indicate that small, presumably more distal, dendrites in the intermediate mNTS may synthesize and/or release catecholamines.

Cholinergic stimulation of rostral and caudal substantia nigra pars compacta produces opposite effects on circling behavior and striatal dopamine release measured by brain microdialysis

Turning in circles is among the behaviors elicited by unilateral cholinergic stimulation of the substantia nigra. Recent studies have shown that microinjection of cholinergic agonists into the substantia nigra pars compacta increases dopamine release and turnover in the striatum of anesthetized rats [Hernández-López et al. (1992) Brain. Res. 598, 114-120; Blaha and Winn (1993) J. Neurosci, 13, 1035-1044]. In this study, the relationship between circling behavior and striatal dopamine release following cholinergic stimulation of the substantia nigra pars compacta neurons was assessed by brain microdialysis in awake rats. The results indicate that cholinergic stimulation of the substantia nigra pars compacta with the mixed nicotinic-muscarinic cholinergic agonist carbachol modulates striatal dopamine release, and this effect is accompanied by circling behavior and stereotypies. Microinjection of carbachol (109 nmol) in the caudal portions of the substantia nigra pars compacta induced contralateral circling associated with an increase of dopamine release in neostriatum. On the contrary, ipsilateral circling and reduction of striatal dopamine release was elicited when the same dose of the drug was applied in the rostral portions of the substantia nigra pars compacta. The above findings are in accordance with recent electrophysiological studies suggesting the existence of sub-populations of nigrostriatal dopaminergic neurons, and indicate that the substantia nigra pars compacta is functionally compartmentalized. We conclude that the cholinergic input to the substantia nigra pars compacta could modulate the motor behavior through regulating the firing rate of nigrostriatal dopaminergic neurons and dopamine release in the neostriatum.

Neurotensin excitation of rat ventral tegmental neurones

1. Whole-cell patch-clamp recordings were made from ventral tegmental area neurones in rat midbrain slices in vitro. In principal cells, which are presumed to contain dopamine, neurotensin (< or = 1 microM) caused an inward current at -60 mV in thirty of forty-seven neurones and had no effect on the remainder. In secondary neurones, neurotensin caused an inward current in twelve of thirty-three cells. 2. The inward current evoked by neurotensin reached a maximum amplitude of about 80 pA, and declined over several minutes when the application was discontinued. The current was most commonly accompanied by a decrease in membrane conductance and reversed polarity at a strongly hyperpolarized potential; this reversal potential was less negative in a higher extracellular potassium concentration. Neurotensin also caused an inward current even in potassium-free internal and external solutions; this current was accompanied by a conductance increase, reversed close to 0 mV and was inhibited by reduction of the extracellular sodium concentration (from 150 to 20 mM). 3. The inward current was associated with a large increase in noise; this persisted in calcium-free solutions but was inhibited by low sodium concentration. The increase in noise was more prominent at hyperpolarized potentials. The amplitude of the unitary current underlying the increase in noise was estimated from the ratio of the variance to the mean as about 1.5 pA at -100 mV. 4. When the recording was made with an electrode containing guanosine 5'-thio-triphosphate, the steady inward current evoked by neurotensin did not reverse when the application was discontinued. When the recording electrode contained pertussis toxin, the action of neurotensin was not different although outward currents evoked by dopamine and baclofen declined with time. 5. It is concluded that neurotensin excites ventral tegmental area neurones by activating a pertussis toxin-insensitive guanosine nucleotide-binding protein. This leads to a reduction in membrane potassium conductance and an increase in membrane sodium conductance, the relative contribution of which varies from cell to cell.

Topographic heterogeneity of substantia nigra neurons: diversity in intrinsic membrane properties and synaptic inputs

The passive and active membrane properties of substantia nigra neurons were recorded in vitro at various locations throughout its anterior-posterior extent and their responses to extracellular electrical stimulation within the pars reticulata were analysed. One class of nigral pars compacta cell showed the well-established electrophysiological characteristics of mesencephalic dopaminergic neurons, i.e. spontaneous discharge in a very rhythmic, pacemaker fashion without bursting activity and with broad action potentials. However, these neurons could be subdivided further according to differences in electrophysiological properties which correlated with their position within the substantia nigra. Thus, neurons recorded from the anterior part of the substantia nigra, at the level of the mammilary bodies displayed a significantly higher firing rate and shorter action potential than those located in posterior slices at the level of the accessory optic tract. The location of the cell was also a critical factor in its response to stimulation of the pars reticulata: in anterior slices only 45.5% of the cells responded with inhibitory postsynaptic potentials to stimulation, while in posterior slices inhibitory postsynaptic potentials occurred in 85.7% of the neurons (n = 44). In addition, anteriorly located neurons were more sensitive to direct electrical stimulation than posteriorly located cells and they also exhibited excitatory postsynaptic potentials (33%) on pars reticulata stimulation. However, the actual properties of inhibitory postsynaptic potentials were essentially the same in these neurons irrespective of whether they were located either in the anterior or posterior part of the nigra: reversal potentials of inhibitory postsynaptic potentials were found at two distinct potentials indicating involvement of both GABAA and GABAB receptors. This deduction is also supported by additional pharmacological findings: application of the GABAA antagonist, bicuculline methiodide and/or GABAB antagonist, 2-hydroxysaclofen blocked both the inhibitory postsynaptic potentials and the cessation of spontaneous firing activity of the cells to stimulation of the pars reticulata. The other type of pars compacta neuron recorded discharges phasically and was located exclusively in the anterior pole of the substantia nigra. These cells showed a wide range of spontaneous firing activity, a non-rhythmic, irregular pattern of firing, a shorter action potential width and the presence of a low-threshold calcium conductance. These "phasic" neurons also differed greatly from other compacta neurons in their response to pars reticulata stimulation: spontaneous activity of these cells was not inhibited nor did they show inhibitory postsynaptic potentials. Instead, the majority was preferentially activated by direct stimulation of the dendrites, although excitatory postsynaptic potentials could also be evoked.(ABSTRACT TRUNCATED AT 400 WORDS)