Search

Alterations in hippocampal circuitry may underly age-related learning and memory impairment. We showed in a previous study that the GABAB-mediated slow inhibitory postsynaptic potential (IPSP) induced in CA1 pyramidal neurons by electrical stimulation of stratum radiatum, is depressed in the hippocampus of the aged rat. This could be due to alterations in GABAergic interneuron functions. We report in this study that the number of hippocampal calbindin-immunoreactive (CaBP-IR) GABAergic interneurons is decreased in the aged rat. The mean number of CaBP-IR interneurons per slice decreases by 50% in the aged rat. The most severe loss was observed in the stratum radiatum of CA1 (78%), with a less consistent loss of immunoreactivity in CA3 (35%). In contrast, the mean number of interneurons containing parvalbumin (PV), was not significantly decreased in the aged rat. Our results show a loss of CaBP immunoreactivity in a population of GABAergic interneurons, which might be related to an altered function of these interneurons and consequently of GABAergic synaptic transmission in the aged rat. In contrast, PV immunoreactivity in interneurons located close to the pyramidal layer does not decrease in the hippocampus of the aged rat.

The nature of the coupling mechanism of presynaptic calcium channels involved in the release of neurotransmitters in the mammalian central nervous system is unknown. Using intracellular recordings from CA1 neurons in the rat hippocampal slice preparation, we show that the N-type calcium channels antagonist omega-conotoxin GVIA (ω-CgTx) blocks partially the excitatory (EPSP) and totally the inhibitory (IPSP) synaptic transmission in CA1 hippocampal pyramidal neurons. In addition, the inhibitory effect of ω-CgTx on IPSPs is strongly depressed by intrahippocampal injection of PTX, while the effect on EPSP is not. The results suggest that the nature or the regulation of calcium channels might be different, depending on the location of these channels on excitatory or inhibitory terminals.

Regional brain incorporation of a radiolabeled unsaturated fatty acid, [1- 14 C]arachidonic acid ( 14 C-AA), was measured in awake rats following unilateral lesioning of the nucleus basalis magnocellularis (NBM). Right-sided lesions were produced in 3-month-old, male rats by stereotaxic injection of 10 μg ibotenic acid. Two weeks after lesioning, rats were subjected to one of two protocols: (1) 5 min intravenous infusion of 14 C-AA (170 μCi/kg); or (2) i.p.injection of arecoline (5 mg/kg), a cholinergic agonist, followed by 5 min intravenous infusion of 14 C-AA. All animals were killed 15 min postinfusion. Brains were frozen and sectioned for quantitative autoradiography or were stained for acetylcholinesterase (AChE). Animals with unilateral NBM lesions displayed reduced AChE staining in prefrontal, frontal and parietal cortices of the lesioned side, but there was no right-left difference in incorporation of 14 C-AA without cholinergic stimulation. Arecoline administration increased 14 C-AA incorporation into the prefrontal and frontal cortices ipsilateral to the NBM lesion as compared to the contralateral side and the increase was most prominent in deeper cortical layers such as layers IV and V. Right-left differences in incorporation were not apparent in parietal, temporal, or occipital cortices, where reduction of AChE activity was minimal or absent, nor in subcortical structures. The results suggest that the intravenous 14 C-AA technique combined with cholinergic stimulation can be used to detect compensatory regulation of phospholipid-coupled signal transduction caused by a deficit in cholinergic input into the cerebral cortex.

The anatomy of pathways containing the calcium binding protein calretinin was investigated in the forebrain of the rat, using a combination of immunohistochemical and retrograde tract tracing techniques. Numerous well identified pathways do contain calretinin, whereas others do not. Pathways arising from the substantia nigra/ventral tegmental area, the dorsal raphe, the lateral mammillary nucleus, the supramammillary nucleus, the triangular septal and septo-fimbrial nuclei, several thalamic nuclei, the parabrachial nucleus, the peripeduncular nucleus, the medial amygdala contain at least some calretinin. The proportion of projection neurons containing calretinin ranged from 2% (dorsal raphe to caudate) to about 75% (triangular septal nucleus to habenula, medial amygdala to the ventromedial hypothalamus). More than 50% of the nigro-striatal neurons contain calretinin immunoreactivity. In contrast, other pathways do not contain any calretinin immunoreactivity (for instance the pathways arising from cerebral cortex, locus coeruleus, cholinergic forebrain nuclei), although calretinin may be present in local neurons in these structures. The present study demonstrates that calretinin is not associated specifically with projection neurons or local neurons, identified transmitter systems or functionnally related pathways in the forebrain of the rat.

Unilateral nucleus basalis magnocellularis (NBM) ablation, which causes partial cholinergic denervation of the ipsilateral anterior neocortex, results in an acute but transient depression of regional cerebral metabolic rates for glucose (rCMRglc) in deafferented areas; rCMRglc normalizes within 2 weeks. To seek possible compensatory changes in cholinergic mechanisms following NBM ablation that could lead to rapid metabolic normalization, we studied rCMRglc responses to the receptor agonists nicotine and arecoline and the cholinesterase inhibitor physostigmine in rats at 2 weeks after unilateral NBM destruction. Physostigmine increased rCMRglc in 10 of 30 cortical areas contralateral to the NBM lesion. Compared to the unlesioned side, rCMRglc after physostigmine in the lesioned cortex was significantly lower in 2, significantly higher in 1 and not different (P0.05) in 27 areas. Neither arecoline nor nicotine treatment produced rCMRglc asymmetry in lesioned rats. These results demonstrate that responsivity to physostigmine is maintained in most regions of the rat neocortex after extrinsic cholinergic denervation by NBM ablation. This adaptive response appears not to result from cholinergic receptor upregulation and may reflect instead reorganization of cholinergic synapses.

The effects of muscarinic agonists (McN-A-343, pilocarpine, oxotremorine-M, carbachol) and antagonists (pirenzepine, gallamine) applied by iontophoresis were studied on several neuronal populations in the central nervous system of rats anesthetized with urethane. Septohippocampal neurons and neurons from hippocampus, subiculum and somatic sensory cortex were studied. Oxotremorine-M and carbachol had (almost exclusively) potent excitatory effects whereas pilocarpine had some and McN-A-343 had almost exclusively inhibitory effects on the 4 populations of neurons studied. Pirezepine blocked more easily the effects of pilocarpine and McN-A-343 than those of oxotremorine-M or carbachol. These results suggest (i) that many central neurons may bear different functional muscarinic receptors and (ii) that the various agonists studied might act through (at least partially) different mechanisms.

Medial septal-nucleus of the diagonal band of Broca area (MS-nDBB) neurons, identified by their antidromic response to the electrical stimulation of the fimbria and/or hippocampus, were studied in the rat under various conditions of anesthesia. These septo-hippocampal neurons (SHNs) were classified into 4 groups on the basis of: (i) their antidromic latency; and (ii) the presence or absence of a rhythmically bursting pattern of spontaneous discharge. The rhythmically bursting activity (43.5% of the SHNs) was highly dependent on the anesthetic conditions. The group of SHNs differed in their mean conduction velocity and rate of spontaneous activity. In contrast, irrespective of their classification in a particular group, the large majority of the SHNs could be excited by the iontophoretic application of cholinergic agonists. Beside the SHNs, two other populations of MS-nDBB neurons could be identified by electrical antidromic stimulation: neurons projecting to the amygdala (Am) and neurons projecting through the medial forebrain bundle (MFB). Half of the MS-nDBB neurons projecting to Am were also antidromically driven from the fimbria. The axonal branch projecting to Am had a slower conduction velocity than that projecting to hippocampus. In contrast MS-nDBB neurons projecting through the MFB were never antidromically driven from the fimbria, although they received orthodromic inputs. They had a slower conduction velocity than the other groups of MS-nDBB neurons.

The regional distribution of choline acetyltransferase activity, somatostatin levels and 125I-CGP 23996 (a somatostatin agonist analog) specific binding sites in 10 separate zones of the cerebral cortex was analyzed. The study was performed in normal rats as well as 15 days after unilateral excitotoxic lesion of the nucleus basalis. A significant correlation was found in the controls between the regional distribution of choline acetyltransferase activity and somatostatin concentrations, both most highly concentrated in the piriform and entorhinal cortex. In contrast, the regional density of 125I-CGP 23996 binding sites correlated neither with choline acetyltransferase activity nor with somatostatin levels. Unilateral lesions of the basal forebrain decreased choline acetyltransferase activity in the frontal and parietal cortex, while 125I-CGP 23996 binding decreased in frontal and occipital regions. No decrease in somatostatin content was observed. The results suggest that, in rats, cortical somatostatin receptors could be associated with cholinergic afferents from the nucleus basalis in the frontal cortex only.

Neuronal responses to the iontophoretic application of acetylcholine, carbachol and nicotine were studied in rats, 2 or 3 weeks following bilateral lesions of the nucleus basalis region, and compared to those obtained in normal animals. The percentage of cortical neurons excited by acetylcholine and their individual sensitivity were higher in lesioned animals. Furthermore, the laminar distribution of the responses to acetylcholine and the proportion of responses to carbachol and nicotine were also modified.

In both cats and rats, neurons with a distinctively narrow action potential were recognized as a small subset of all neurons isolated in the somatosensory cortex. These cells were characterized by generally having a spontaneous activity, some evidence of an afferent input, a sensitivity to glutamate but a relative resistance to depolarization block induced by glutamate and a marked insensitivity to acetylcholine. Two were filled with horseradish peroxidase (HRP) and recovered. Although others have suggested that such neurons are interneurons, following reconstruction it was apparent that the two cells filled with HRP were pyramidal cells. These observations suggest that there may be more than one class of cortical neurons with thin spikes.

Electrophysiological and pharmacological properties of neurons within solid basal forebrain transplants were studied in adult rats anesthetized with urethane. No specific topography of the neurons recorded was observed within the graft. The mean spontaneous activity of the grafted neurons (GNs) was relatively low (4.9 impulses/s) but not unlike that of other central neurons in situ. A large proportion of GNs fired with regular discharges, but other modes of discharge were also observed. A few rhythmically bursting GNs were recorded having a discharge pattern very much like that of the rhythmically bursting medial septal neurons. The responses of GNs to glutamate, gamma-aminobutyric acid, acetylcholine, serotonin and norepinephrine was fairly similar to those described in other central structures.

In many common experimental conditions the majority of the neurons isolated in cat and rat somatosensory cortex using extracellular recording techniques are unresponsive to somatic stimuli. This population of cells becomes amenable to experimental manipulation with iontophoretic administration of glutamate, other putative neurotransmitters, and their agonists and antagonists. During administration of glutamate or bicuculline methiodide as many as half of the unresponsive cells could be shown to receive somatic inputs that did not drive the cells without drug treatments. Several hypotheses are discussed concerning the possible origin and function of the unresponsive neurons and the conditions under which they might play an active role in cortical function such as during altered patterns of afferent input, enhanced release of acetylcholine, reduction of GABAergic inhibition and situations involving learning or directed attention. The unresponsive neurons may be a characteristic that differentiates somatosensory cortex from subcortical sensory pathways.

In vivo intracellular recordings were obtained from septohippocampal neurons identified by their antidromic response to electrical stimulation of the fimbria in rats anesthetized with pentobarbital or urethane. Beside the antidromic response, fimbria stimulation evoked a short-latency depolarizing potential (EPSP) followed by an hyperpolarizing potential which reversed polarity when recorded with KCl-filled electrodes. This IPSP is therefore likely to be chloride-dependent. It was followed by a long-lasting (80–250 ms) depolarizing potential often associated with a burst spikes. Septohippocampal neurons therfore receive an inhibitory, chloride-mediated, input which itself triggers a long-lasting excitatory event. These results are consistent with extracellular observations. Their significance in the septohippocampal circuitry is discussed.

The properties of the neuronal responses to different types of mechanical peripheral stimulation were studied during electrode penetrations in the first somatosensory cortex of anaesthetized rats with polyarthritis. Very few neurons were driven by light cutaneous stimulation (such as brushing) or by intense mechanical stimulation. Most of the neurons were driven by joint movement and/or moderate pressure on the skin. These neurons could be found in all cortical layers, the majority being located in layer V. These results contrast sharply with the properties and laminar distribution of the different functional categories of cortical neurons, as observed in normal animals.

The medial septal area contains neurons which project to the hippocampal formation. A sizeable proportion of these septo-hippocampal neurons (SHNs) are cholinergic. About 40% of them also display a characteristic discharge pattern in rhythmic bursts. We hypothesized that SHNs with a rhythmically bursting activity (RBA) are the cholinergic ones. To test this hypothesis we studied the effects of acetylcholine synthesis blockade by hemicholinium-3 (HC-3) on the properties of the SHNs. HC-3 (16, 32 or 64 micrograms total dose) or saline were injected in the lateral ventricles of adult male Sprague-Dawley rats anesthetized with urethane. Extracellular recordings from SHNs in the medial septal area were obtained within hours after HC-3 injections (n = 24 animals). SHNs were identified by their antidromic response following electrical stimulation of the fimbria-fornix. The pharmacological properties of SHNs were studied in some animals using microiontophoretic applications from multibarreled electrodes filled with various neurotransmitters. The hippocampal rhythmic slow activity (RSA or theta) was abolished even after the lowest dose of HC-3 tested (16 micrograms). No significant change in SHNs conduction velocity or spontaneous activity was observed at any dose of HC-3. The percentage of SHNs with RBA was unchanged. In contrast the mean frequency of the RBA was decreased by HC-3 in a dose-dependent fashion. The mean frequency was lowest within the first 3 h after injection. Although the mean spontaneous activity was unchanged SHNs tended to have more spikes per burst. The effects of various neurotransmitters on SHNs were qualitatively unchanged after HC-3 injection. These results suggest that acetylcholine synthesis blockade by HC-3 leads not only to the disappearance of the hippocampal RSA in urethane-anesthetized animals, but also to a decrease in the frequency of the rhythmically bursting activity of the SHNs. Since the 4-Hz hippocampal theta is atropine-sensitive, the results provide indirect evidence that the SHNs with rhythmically bursting activity are the cholinergic SHNs.

In vivo differential pulse voltammetry was used for the detection of indoleamines during vertical electrode penetrations in rat first somatosensory cortex, for studying the laminar distribution of serotonin and/or its metabolites in that part of the cortex. The peak of current corresponding to 5-hydroxyindoles was maximum in the most superficial part of the cortex and diminished gradually in the deeper layers. These results suggest that the cortical serotonergic innervation is predominant in the superficial layers.

The excitability of the axonal terminals of medial septal neurons projecting to the dentate gyrus has been studied in the anesthetized rat under various experimental conditions: spontaneous or drug-induced variations in neuronal soma discharge rate, conditioning stimulation of afferent pathways (perforant path, commissural pathway, fimbria-formix). It has been observed that terminals excitability is inversely correlated to the level of neuronal ongoing activity. These effects were observed on virtually all septal neurons projecting to the dentate gyrus. Since about one half of the septohippocampal neurons likely to be cholinergic, it follows that such a phenomenon is not transmitter specific.

Neurons located in the medial septum-nucleus of the diagonal band (vertical limb) area and antidromically activated by electrical stimulation of the fimbria were recorded in urethane anesthetized rats. Forty-five percent of these septo-hippocampal neurons (SHNs) discharged rhythmically in short bursts (mean burst frequency, 4 Hz). They were antidromically driven at short latencies from the fimbria. SHNs driven at long latencies (above 5 ms) were never bursting neurons. Fimbria stimulation also had a powerful inhibitory effect on the spontaneous activity of SHNs. The vast majority of the septo-hippocampal neurons were excited by the iontophoretic application of acetylcholine or cholinergic agonists, carbachol being the most effective. The acetylcholine-induced excitations were readily abolished by atropine. In contrast hexamethonium and mecamylamine were less effective. The rhythmic bursting activity could not be consistently altered by the iontophoretic application of cholinergic agonists or antagonists or of divalent cations. SHNs were also sensitive to various other putative neurotransmitters (substance P, GABA) known to play a role in the medial septal area. Bursting neurons and ACh-sensitive neurons were less frequent among unidentified medial septal neurons. The regulation of the septo-hippocampal cholinergic pathway is therefore not likely to be due to a direct feedback inhibition by locally released acetylcholine. However a strong inhibitory feedback could be exerted by the hippocampo-septal pathway impinging directly or indirectly on septo-hippocampal neurons.

The effects of drugs administered iontophoretically were studied on 302 neurons isolated from partially deafferented hindlimb granular cortex of the rat and were compared to a previously studied sample from normal granular cortex. The proportions of cells affected by glutamate and acetylcholine (ACh) were not markedly different after partial deafferentation, but the cells were more readily depolarized by glutamate and their responses to a fixed dose of glutamate were larger. Fewer cells were excited strongly by ACh (up to the point of depolarization block) and the amplitudes of responses to test pulses of ACh were reduced in infragranular layers after partial deafferentation. Fifteen cells (5.9%) were inhibited by ACh administration, whereas in normal cortex this value was less than 1%. Fewer receptive fields were uncovered by glutamate after partial deafferentation, but more receptive fields were enlarged by this substance after nerve transection. These data were interpreted to mean that the cells from deafferented cortex had fewer excitatory inputs and as well, were apparently under a comparatively weaker degree of inhibitory control after deafferentation. The administration of ACh uncovered fewer somatic inputs than in normal animals. Responses were enhanced by ACh less frequently, and repeated treatments with ACh often led to a reduction in the effectiveness of the afferent stimulus. Increases in neuronal thresholds for somatic stimuli also were observed. The laminar distribution of the effects of ACh was similar to the distribution observed in normal animals. The responses to iontophoretically administered ACh and its agonists appeared to be mediated through both nicotinic and muscarinic receptor processes after partial deafferentation. Often the time course of the effects of ACh was abnormal, being characterized by oscillations between silence and very high rates of discharge with a period of 6-12 s. These observations are consistent with the hypothesis that after a major deafferentation (i) there is a reduction of the inhibitory controls normally present in the somatosensory cortex, and (ii) neuronal responses to ACh are modified in partially deafferented cortex. The magnitude of the responses of cells to ACh in the supragranular layers are larger after deafferentation whereas the magnitude of responses in the infragranular layers are reduced by this procedure. These changes may be related to changes induced by deafferentation in the distribution of receptors and/or in their pharmacological properties.

The properties of two populations of acetylcholine-sensitive neurons, defined by their different laminar distribution (group A, layers V and upper VIa; group B, layer VIb and underlying white matter) were studied in the SmI cortex of urethane-anesthetized rats. The spontaneous activity of group B neurons was significantly lower than in group A. The proportion of neurons excited by iontophoretic acetylcholine, the duration and the pharmacological properties of the excitations were the same in both groups. The dose-response curves obtained for sets of neurons of both groups, recorded with the same electrode, had comparable slopes. Carbachol had a greater potency than acetylcholine in both groups. In contrast, the latency of the response to acetylcholine was significantly shorter in group B. The amount of current required to increase the spontaneous discharge rate of a given neuron by 200% was lower in group B than in group A. These data suggest that the relative change in firing frequency induced by a given cholinergic input is likely to be larger in group B than in group A. Finally, neuronal excitations could still be obtained in both groups when acetylcholine was applied 300 μm, but not 600 μm, from the cell body, suggesting that the spatial distribution of acetylcholine receptors is similar in groups A and B.

Neurons located in the medial septum-nucleus of the diagonal band of Broca (vertical limb) and antidromically activated by electrical stimulation of the fimbria were recorded in urethane anesthetized rats. Forty-three percent of these septohippocampal neurons (SHNs) were excited by the iontophoretic application of thyrotropin-releasing hormone (TRH). Rhythmically bursting SHNs were more often excited (63%) by TRH than the non-bursting SHNs (35%). The majority of the TRH-sensitive SHNs could also be excited by cholinergic agonists. TRH-induced excitations were not abolished by the simultaneous application of atropine. Potentiation by TRH of acetylcholine, carbachol or glutamate-induced excitations of SHNs were rarely observed. Cyclo (His-Pro) and (3-Me-His2)-TRH were observed to have similar, although less dramatic, effects. These results demonstrate that the SHNs, which are the neurons of origin of the septohippocampal pathway, are readily excited by TRH.

Differential pulse voltammetry was used for the detection of 5-hydroxyindoles in the cerebral of rats anaesthetized with urethane. The stimulation of the lateral hypothalamus or of the dorsal raphe nucleus induced a 10–40% increase in the amplitude of the signal. The signal recorded from p-chlorophenylalanine (pCPA)-pretreated animals was much smaller than in normal animals and could be increased by 5-HTP administration. The stimulation of the serotonergic pathways was ineffective in the pCPA-pretreated animals.

A sample of 302 neurons in rat hindlimb granular cortex was studied between 2 and 3 weeks after transection and ligation of the sciatic nerve. These neurons were compared to a control sample obtained from normal rats under similar experimental conditions. After sciatic nerve transection the proportion of neurons driven by somatic stimuli applied to the hindlimb (29.1%) was not significantly different from the proportion observed in the control sample (25.8%). The proportion of neurons with cutaneous receptive fields was also the same before and after nerve transection although the proportion of neurons responding to higher threshold taps was reduced. Spontaneously active neurons were encountered more frequently after sciatic nerve transection than in the control (45% vs 32%) and their mean discharge frequency was higher (8.6 vs 6.4 imp/s). Twice as many (10.2% vs 5.2%) spontaneously active neurons whose discharge was modulated by afferent stimuli were found after sciatic nerve transection. This and several other lines of evidence suggest that the cortical neurons were released from inhibition by the sciatic nerve transection. The number of spontaneously active neurons and the mean discharge rate were increased in each cortical lamina suggesting that the increased excitability seen after deafferentation occurred in all cortical layers. Neurons throughout the portion of the hindlimb representation studied could be driven from the remaining innervated region of the hindlimb, including one and sometimes two digits, part of the palm and most of the dorsum of the foot. More neurons were driven by receptive fields on the ankle than was the case for normal rats. Most neurons with cutaneous receptive fields were distributed in the same layers as those in the control group and had normal shapes and appearances. However, there were few of small size. Most were of moderate dimensions well within the normal range. Seven examples were found with unusually large proportions of their field extending from the foot onto the ankle and covering most of the posterior quadrant of the animal. In normal cortex the largest receptive fields were found in the middle layers. This distribution was not as clear after sciatic nerve section and a mixture of large and medium-sized receptive fields occurred at each depth. Further, some neurons with cutaneous receptive fields were found above 300 μm and below 1000 μm, depths where receptive fields werre uncommon in the normal sample.

Septo-hippocampal neurons (SHNs) were recorded from the medial septum-diagonal band area of rats anaesthetized with either urethane or fluothane. They were identified by their antidromic response to the electrical stimulation of the fimbria. Their responses to peripheral somatic noxious and non-noxious stimulation were studied. Non-noxious natural stimulations were relatively ineffective. In contrast, 68% of the SHNs were driven by noxious stimulation. The SHNs could be driven either by mechanical or thermal stimulation. Intraperitoneal injection of bradykinin excited about half of the SHNs. Some neurons were able to encode stimulus intensity (strength of the mechanical stimulation and/or temperature of the thermal stimulation). The receptive fields of the SHNs were large, usually involving half of the body or the whole body surface. These results suggest that SHNs, which are at the origin of the cholinergic septo-hippocampal pathway, might be involved in cerebral mechanisms related to nociception.

Neurons located in the ventromedial globus pallidus (nucleus basalis) and substantia innominata, that were antidromically driven by electrical stimulation of the frontoparietal cortex, were recorded in the urethane anesthetized rat. The basalocortical neurons (BCNs) were antidromically driven with latencies of 1.1–13.5 ms, giving conduction velocities of 0.6–6.8 m/s. Many BCNs had regular patterns of spontaneous discharge (mean spontaneous activity: 20 impulses/s). Most BCNs were not responsive to non-noxious peripheral somatic stimulation. BCNs were readily excited by the iontophoretic application of glutamate and strongly inhibited by GABA. Eighty-five percent of the BCNs could be excited by acetylcholine. They could also be excited by cholinergic agonists. Muscarinic agonists excited a higher proportion of BCNs than nicotinic agonists. Excitatory responses to acetylcholine, carbachol and muscarinic agonists were abolished by atropine.

The septo-hippocampal pathway is one of the best characterized cholinergic pathways of the mammalian central nervous system. It is very likely that this pathway is involved in the pathophysiology of dementia of the Alzheimer's type and perhaps of normal aging also. The properties of the septo-hippocampal neurons identified by their antidromic response to the electrical stimulation of the fimbria-fornix were altered in aged (27 months) rats as compared to young (2–3 months) controls: their pattern of spontaneous activity (including the frequency of occurrence of a rhythmically bursting activity) and their axonal conduction velocity were altered. The intensity of these changes was dependent on the presence of pituitary tumor which developed spontaneously in about1/3 of the animals. These results provide direct evidence of the involvement of septo-hippocampal neurons in the aging process.