Correlated morphological and neurochemical features identify different subsets of vasoactive intestinal polypeptide-immunoreactive interneurons in rat hippocampus

VIP enhances both pre- and postsynaptic GABAergic transmission to hippocampal interneurones leading to increased excitatory synaptic transmission to CA1 pyramidal cells

Vasoactive intestinal peptide (VIP) is present in the hippocampus in three subtypes of GABAergic interneurones, two of which innervate preferentially other interneurones, responsible for pyramidal cell inhibition. We investigated how pre- and postsynaptic modulation of GABAergic transmission (to both pyramidal cells and interneurones) by VIP could influence excitatory synaptic transmission in the CA1 area of the hippocampus. VIP (0.1-100 nM) increased [(3)H]GABA release from hippocampal synaptosomes (maximum effect at 1 nM VIP; 63.8 +/- 4.0%) but did not change [(3)H]glutamate release. VIP (0.3-30 nM) enhanced synaptic transmission in hippocampal slices (maximum effect at 1 nM VIP; field excitatory postsynaptic potentials (epsp) slope: 23.7 +/- 1.1%; population spike amplitude: 20.3 +/- 1.7%). The action on field epsp slope was fully dependent on GABAergic transmission since it was absent in the presence of picrotoxin (50 microM) plus CGP55845 (1 microM). VIP (1 nM) did not change paired-pulse facilitation but increased paired-pulse inhibition in CA1 pyramidal cells (16.0 +/- 0.9%), reinforcing the involvement of GABAergic transmission in the action of VIP. VIP (1 nM) increased muscimol-evoked inhibitory currents by 36.4 +/- 8.7% in eight out of ten CA1 interneurones in the stratum radiatum. This suggests that VIP promotes increased inhibition of interneurones that control pyramidal cells, leading to disinhibition of synaptic transmission to pyramidal cell dendrites. In conclusion, concerted pre- and postsynaptic actions of VIP lead to disinhibition of pyramidal cell dendrites causing an enhancement of synaptic transmission.

Multiple GABAA receptor subtypes regulate hippocampal ripple oscillations

High-frequency oscillations (140-200 Hz) were recorded in behaving rats from the CA1 area of the hippocampus. As generation of these synchronous patterns is assumed to depend on coordinated interneuronal inhibition, we studied the interference of benzodiazepines with the fine structure and occurrence of ripple oscillations. The nonselective GABAA receptor alpha-subunit agonist, diazepam, lowered the frequency of ripple oscillations and reduced their occurrence, amplitude and duration. Zolpidem, an alpha1-subunit selective benzodiazepine elevated ripple duration but acted similar to diazepam in other respects. The nonselective alpha-subunit benzodiazepine antagonist, flumazenil, reduced ripple numbers, amplitude and duration. Wavelet based analysis of the dynamics of intraripple frequency revealed a dramatic decay within a ripple. Only diazepam (1 mg/kg) accelerated this intraripple frequency accommodation. The effects were not due to increased behavioural activity and alertness as evident from vigilance state control. The results suggest a differential role of GABAA receptor subtype specific inhibitory mechanisms in the mediation and fine-tuning of the network synchronization during approximately 200 Hz hippocampal oscillations.

Immunocytochemically defined interneuron populations in the hippocampus of mouse strains used in transgenic technology

Transgenic mice are overtaking the role of model animals in neuroscience. They are used in developmental, anatomical, and physiological as well as experimental neurology. However, most results on the organization of the nervous system derive from the rat. The rat hippocampus and its neuronal elements have been thoroughly investigated, revealing remarkable functional and morphological diversity and specificity among hippocampal interneurons. Our aim was to examine the properties of distinct hippocampal interneuron populations, i.e., those immunoreactive for calcium-binding proteins (parvalbumin, calbindin, and calretinin), neuropeptides (cholecystokinin, neuropeptide Y, somatostatin, vasoactive intestinal polypeptide), and certain receptors (metabotropic glutamate receptor 1alpha, cannabinoid receptor type 1) in four strains of mice widely used in transgenic technology, and to compare their properties to those in the rat. Our data indicate that the distribution as well as the dendritic and axonal arborization of mouse interneurons immunoreactive for the different markers was identical in the examined mouse strains, and in most respects are similar to the features found in the rat. The postsynaptic targets of neurons terminating in the perisomatic (parvalbumin), proximal (calbindin), and distal (somatostatin) dendritic region, as well as on other interneurons (calretinin), also matched those found in the rat. However, a few significant differences could also be observed between the two species in addition to the already described immunoreactivity of mossy cells for calretinin: the absence of spiny calretinin-immunoreactive interneurons in the CA3 region, sparse contacts between calretinin-immunoreactive interneurons, and the axon staining for somatostatin and neuropil labeling for cholecystokinin. We can conclude that the morphofunctional classification of interneurons established in the rat is largely valid for mouse strains used in transgenic procedures.

The GABAergic septohippocampal pathway in control and reeler mice: target specificity and termination onto Reelin-expressing interneurons

The septohippocampal pathway contains two separate components: the cholinergic and the GABAergic. Whereas cholinergic fibers terminate on many hippocampal cell types, GABAergic septohippocampal fibers selectively contact the cell bodies of hippocampal interneurons. We examined whether the GABAergic septohippocampal system was altered in reeler mice. First, we found that both components of the septohippocampal pathway in mice present a distribution and target-cell specificity similar to that described in rats. We also show that GABAergic septohippocampal axons terminate on subpopulations of interneurons expressing reelin, which may implicate this extracellular matrix protein in the targeting of septohippocampal axons. We thus examined the septohippocampal pathway in reeler mice defective in Reelin. In contrast to wild-type animals, reeler mice displayed an ectopic location of both cholinergic and GABAergic fibers, which accumulate close to the hippocampal fissure. Despite their altered distribution, GABAergic septal axons maintain their target-cell selectivity innervating exclusively the perisomatic region of hippocampal interneurons. Thus, as in wild type, GABAergic septal fibers formed complex baskets around the cell body of GAD-positive hippocampal neurons in reeler mice. In addition, we found that reeler hippocampi have an altered distribution of hippocampal interneurons expressing PARV or CALB, many of which are located close to the hippocampal fissure. We thus conclude that although reeler mice have an altered distribution of hippocampal interneurons, GABAergic septohippocampal axons nevertheless terminate on their specific target interneurons. Thus, whereas target layer termination of septal fibers is severely impaired in reeler mice, our data indicate that the cell-specific targeting of GABAergic septohippocampal axons is governed by Reelin-independent signals.

Convergence of excitatory and inhibitory inputs onto CCK-containing basket cells in the CA1 area of the rat hippocampus

The number and distribution of excitatory and inhibitory inputs affect the integrative properties of neurons. These parameters have been studied recently for several hippocampal neuron populations. Besides parvalbumin- (PV) containing cells that include basket and axo-axonic cells, cholecystokinin (CCK)-containing interneurons also form a basket cell population with several properties distinct from PV cells. Here, at the light microscopic level, we reconstructed the entire dendritic tree of CCK-immunoreactive (IR) basket cells to describe their geometry, the total length and laminar distribution of their dendrites. This was followed by an electron microscopic analysis of serial ultrathin sections immunostained against gamma-aminobutyric acid, to estimate the density of excitatory and inhibitory synapses on their somata, axon initial segments and different subclasses of dendrites. The dendritic tree of CCK-IR basket cells has an average length of 6300 microm and penetrates all layers. At the electron microscopic level, CCK basket cells receive dendritic inputs with a density of 80-230 per 100 microm. The ratio of inhibitory inputs is relatively high (35%) and increases towards the soma (83%). The total numbers of excitatory and inhibitory synapses converging onto CCK-IR cells are approximately 8200. Comparison of the two, neurochemically distinct basket cells reveals that CCK-containing basket cells receive much less synaptic input than PV cells; however, the relative weight of inhibition is higher on CCK cells. Additional differences in their anatomical and physiological properties predict that CCK basket cells are under a more diverse, elaborate control than PV basket cells, and thus the function of the two populations must be different.

Opposing effects by pituitary adenylate cyclase-activating polypeptide and vasoactive intestinal peptide on hippocampal synaptic transmission

Pituitary adenylate cyclase-activating polypeptide (PACAP), vasoactive intestinal peptide (VIP), and their receptors have been localized within the hippocampus but their physiological function on synaptic transmission is still unclear. We investigated the effects of PACAP and VIP on evoked excitatory postsynaptic currents (EPSCs) recorded with patch clamp from CA1 pyramidal neurons in rat hippocampal slices. Bath application of PACAP reversibly reduced EPSC amplitude. This effect was partly prevented by intracellular addition of (R)-adenosine, cyclic 3',5'-hydrogenphosphorothioate (cAMPS-Rp), a cAMP antagonist inhibiting protein kinase A, but not by the calcium chelator 1,2-bis (2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA). Application of VIP induced a long-lasting increase of EPSC amplitude that was completely abolished when cAMPS-Rp was included in the intracellular solution. PACAP and VIP effects on EPSCs were mimicked by the cAMP agonist 8-bromoadenosine-3',5'-cyclic monophosphate (8-Br-cAMP). The differing abilities of PACAP and VIP to modulate transmission efficiency over long periods of time, through the cAMP/PKA pathway, suggest that these neuropeptides may exert opposing roles in synaptic plasticity.

GABAergic basket cells expressing cholecystokinin contain vesicular glutamate transporter type 3 (VGLUT3) in their synaptic terminals in hippocampus and isocortex of the rat

Vesicular glutamate transporter type 3 (VGLUT3) containing neuronal elements were characterized using antibodies to VGLUT3 and molecular cell markers. All VGLUT3-positive somata were immunoreactive for CCK, and very rarely, also for calbindin; none was positive for parvalbumin, calretinin, VIP or somatostatin. In the CA1 area, 26.8 +/- 0.7% of CCK-positive interneuron somata were VGLUT3-positive, a nonoverlapping 22.8 +/- 1.9% were calbindin-positive, 10.7 +/- 2.5% VIP-positive and the rest were only CCK-positive. The patterns of coexpression were similar in the CA3 area, the dentate gyrus and the isocortex. Immunoreactivity for VGLUT3 was undetectable in pyramidal and dentate granule cells. Boutons colabelled for VGLUT3, CCK and GAD were most abundant in the cellular layers of the hippocampus and in layers II-III of the isocortex. Large VGLUT3-labelled boutons at the border of strata radiatum and lacunosum-moleculare in the CA1 area were negative for GAD, but were labelled for vesicular monoamine transporter type 2, plasmalemmal serotonin transporter or serotonin. No colocalization was found in terminals between VGLUT3 and parvalbumin, vesicular acetylcholine transporter and group III (mGluR7a,b; mGluR8a,b) metabotropic glutamate receptors. In stratum radiatum and the isocortex, VGLUT3-positive but GAD-negative boutons heavily innervated the soma and proximal dendrites of some VGLUT3- or calbindin-positive interneurons. The results suggest that boutons coexpressing VGLUT3, CCK and GAD originate from CCK-positive basket cells, which are VIP-immunonegative. Other VGLUT3-positive boutons immunopositive for serotonergic markers but negative for GAD probably originate from the median raphe nucleus and innervate select interneurons. The presumed amino acid substrate of VGLUT3 may act on presynaptic kainate or group II metabotropic glutamate receptors.

Immunolocalization of metabotropic glutamate receptor 1? (mGluR1?) in distinct classes of interneuron in the CA1 region of the rat hippocampus

In the hippocampal CA1 region, metabotropic glutamate subtype 1 (mGluR1) receptors have been implicated in a variety of physiological responses to glutamate, which include modulation of synaptic transmission and plasticity, as well as neuronal excitability and synchronization. The mGluR1alpha isoform is characteristically expressed only by nonprincipal cells, and it is particularly enriched in somatostatin (SS)-containing interneurons in stratum oriensalveus. Anatomical and physiological data have indicated the presence of mGluR1alpha in several distinct classes of interneurons with their somata located also in strata pyramidale, radiatum, and lacunosum moleculare. Each different interneuron subtype, as defined by functionally relevant criteria, including input/ output characteristics and expression of selective molecular markers, subserves distinct functions in local hippocampal circuits. We have investigated which of the different CA1 interneuron classes express mGluR1alpha by immunofluorescent labeling, combining antibodies to mGluR1alpha, calcium-binding proteins, and neuropeptides, and by intracellular labeling in vitro. Several types of interneuron that are immunopositive for mGluR1alpha each targeted different domains of pyramidal cells and included (1) O-LM inter-neurons, found to coexpress both SS and parvalbumin (PV); (2) interneurons with target selectivity for other interneurons, expressing vasoactive intestinal polypeptide (VIP) and/or the calcium-binding protein calretinin; (3) procholecystokinin-immunopositive interneurons probably non-basket and dendrite-targeting; and (4) an as-yet unidentified SS-immunoreactive but PV-immunonegative interneuron class, possibly corresponding to oriens-bistratified cells. Estimation of the relative proportion of mGluR1alpha-positive interneurons showed 43%, 46%, and 30% co-labeling with SS, VIP, or PV, respectively. The identification of the specific subclasses of CA1 interneurons expressing mGluR1alpha provides the network basis for assessing the contribution of this receptor to the excitability of the hippocampus.

VIP-containing deep short-axon cells of the olfactory bulb innervate interneurons different from granule cells

This study investigates the targets of the population of vasoactive intestinal polypeptide (VIP)-containing deep short-axon cells of the rat olfactory bulb (OB), combining single- and double-immunocytochemical approaches under light and electron microscopy. It has been assumed that deep short-axon cells innervate granule cells in the mammalian OB, but their synaptic connectivity has not been demonstrated to date. Our results indicate that, instead of the accepted scheme of the bulbar circuitry, VIP-containing deep short-axon cells are gamma-aminobutyric acid (GABA)ergic interneurons specialized in the selective innervation of other GABAergic deep short-axon cells. Their axons contact with the perisomatic region and the dendritic portions of subsets of deep short-axon cells that contain VIP, calbindin D-28k and neuropeptide Y. Electron microscopy reveals axo-somatic and axo-dendritic symmetrical synapses from VIP-containing boutons. Taken altogether, our data show that the VIP-containing deep short-axon cells of the rat OB form an interneuronal network that modulates the function of other interneurons different from granule cells. They might be involved indirectly in the inhibition or disinhibition of principal cells or might participate in the generation of oscillatory activity and in the synchronization of populations of interneurons and, then, of principal cells. Present data demonstrate that modulation of the OB by local circuits is more complex than the simple inhibition from periglomerular cells and granule cells, and remark the importance of considering the contribution of other classes of GABAergic interneurons different from periglomerular cells and granule cells to the bulbar circuitry.

Differential modulation of CA1 and dentate gyrus interneurons during exploration of novel environments

Parallel recordings of hippocampal principal cells and interneurons were obtained as rats foraged in familiar and adjacent, novel environments. Firing rates of each cell type were assessed as a function of spatial location. Many CA1 interneurons exhibited large decreases in activity in the novel compared with the familiar environment. Dentate gyrus interneurons, however, were much more likely to exhibit large increases in firing in the novel environment. Neither effect was correlated with basic interneuron discharge properties such as degree of theta modulation, baseline firing rate or degree of spatially modulated discharge. Both CA1 and dentate gyrus interneuron rate changes extended into regions of the familiar environment bordering the novel environment. Principal cells in CA1 and dentate gyrus exhibited similar patterns of place specific activity each being indicative of incorporation of novel spatial information into the spatial representation of the familiar environment. The data indicate that inhibitory networks in the CA1 and dentate gyrus areas are modulated in a divergent fashion during the acquisition of novel spatial information and that interneuron activities can be used to detect those regions of an environment subject to redistribution of principal cell spatial activity patterns.

Interneuron Diversity series: Rhythm and mood in perisomatic inhibition

GABAergic interneurons innervating the perisomatic region of pyramidal cells control population discharge patterns, and thereby all cognitive operations in the cerebral cortex. A striking dichotomy in the function of this interneuron population seems to emerge from the synthesis of recent molecular, anatomical and electrophysiological data. Synaptically and electrically coupled networks of parvalbumin-containing basket cells operate as a non-plastic, precision clockwork for gamma and theta oscillations, and are indispensable for basic cortical processing. By contrast, a highly modifiable interneuron syncytium containing cholecystokinin carries information from subcortical pathways about the emotional, motivational and general physiological state of the animal, and appears to be involved in the fine-tuning of network cooperativity. Impairment of this inhibitory mechanism is likely to result in mood disorders such as anxiety.

Functional localization of cannabinoid receptors and endogenous cannabinoid production in distinct neuron populations of the hippocampus

The possible localization of cannabinoid (CB) receptors to glutamatergic and GABAergic synaptic terminals impinging upon GABAergic interneurons in the CA1 region of the rat hippocampus was examined using the electrophysiological measurement of neurotransmitter release in brain slices. Whereas activation of cannabinoid receptors via the application of the cannabinoid agonist WIN55,212-2 significantly and dose-dependently reduced evoked IPSCs recorded from interneurons possessing somata located in the stratum radiatum (S.R.) and stratum oriens (S.O.) lamellae, evoked glutamatergic EPSCs were unaffected in both neuronal populations. However, in agreement with previous reports, WIN55,212-2 significantly reduced EPSCs recorded from CA1 pyramidal neurons. Additional experiments confirmed that the effects of WIN55,212-2 on IPSCs were presynaptic and that they could be blocked by the CB1 receptor antagonist SR141716A. The involvement of endogenous cannabinoids in the presynaptic inhibition of GABA release was also examined in the interneurons and pyramidal cells using a depolarization-induced suppression of inhibition (DSI) paradigm. DSI was observed in CA1 pyramidal neurons under control conditions, and its incidence was greatly increased by the cholinergic agonist carbachol. However, DSI was not observed in the S.R. or S.O. interneuron populations, in either the presence or absence of carbachol. Whereas DSI was not present in these interneurons, the inhibitory inputs to these cells were modulated by the synthetic cannabinoid WIN55,212-2. These data support the hypothesis that cannabinoid receptors are located on inhibitory, but not excitatory, axon terminals impinging upon hippocampal interneurons, and that CA1 pyramidal neurons, and not interneurons, are capable of generating endogenous cannabinoids during prolonged states of depolarization.

Patterns of expression of neuropeptides in GABAergic nonprincipal neurons in the mouse hippocampus: Quantitative analysis with optical disector

Neuropeptides are widely distributed in the central nervous system and are considered to play important roles in the regulation of neuronal activity. This study shows the patterns of expression of four neuropeptides [neuropeptide Y (NPY), somatostatin (SOM), cholecystokinin (CCK), and vasoactive intestinal polypeptide (VIP)] in gamma-aminobutyric acid (GABA)-ergic neurons of the mouse hippocampus, with particular reference to the areal and dorsoventral difference. First, we estimated the numerical densities (NDs) of GABAergic neurons containing these neuropeptides using the optical disector. The NDs of NPY- and SOM-positive GABAergic neurons were generally higher than those of CCK- and VIP-positive GABAergic neurons. In the whole area of the hippocampus, the ND of NPY-positive GABAergic neurons showed no significant dorsoventral difference (1.90 x 10(3)/mm(3) in the dorsal level, 2.09 x 10(3)/mm(3) in the ventral level), whereas the ND of SOM-positive GABAergic neurons was higher in the ventral level (1.44 x 10(3)/mm(3)) than in the dorsal level (0.80 x 10(3)/mm(3)). The ND of CCK-positive GABAergic neurons was also higher in the ventral level (0.57 x 10(3)/mm(3)) than in the dorsal level (0.33 x 10(3)/mm(3)). Similarly, the ND of VIP-positive GABAergic neurons was higher in the ventral level (0.61 x 10(3)/mm(3)) than in the dorsal level (0.43 x 10(3)/mm(3)). Next, we calculated the proportions of GABAergic neurons containing these neuropeptides among the total GABAergic neurons. In the whole area of the hippocampus, NPY-, SOM-, CCK-, and VIP-positive neurons accounted for about 31%, 17%, 7%, and 8% of GABAergic neurons, respectively. The present data establish a baseline for examining potential roles of GABAergic neurons in the hippocampal network activity in mice.

Mu opioid receptors are in discrete hippocampal interneuron subpopulations

In the rat hippocampal formation, application of mu opioid receptor (MOR) agonists disinhibits principal cells, promoting excitation-dependent processes such as epileptogenesis and long-term potentiation. However, the precise location of MORs in particular inhibitory circuits, has not been determined, and the roles of MORs in endogenous functioning are unclear. To address these issues, the distribution of MOR-like immunoreactivity (-li) was examined in several populations of inhibitory hippocampal neurons in the CA1 region using light and electron microscopy. We found that MOR-li was present in many parvalbumin-containing basket cells, but absent from cholecystokinin-labeled basket cells. MOR-li was also commonly in interneurons containing somatostatin-li or neuropeptide Y-li that resembled the "oriens-lacunosum-moleculare" (O-LM) interneurons innervating pyramidal cell distal dendrites. Finally, MOR-li was in some vasoactive intestinal peptide- or calretinin-containing profiles resembling interneurons that primarily innervate other interneurons. These findings indicate that MOR-containing neurons form a neurochemically and functionally heterogeneous subset of hippocampal GABAergic neurons. MORs are most frequently on interneurons that are specialized to inhibit pyramidal cells, and are on a limited number of interneurons that target other interneurons. Moreover, the distribution of MORs to different neuronal types in several laminae, some relatively far from endogenous opioids, suggests normal functional roles that are different from the actions seen with exogenous agonists such as morphine.

Synaptic vesicle protein synaptoporin is differently expressed by subpopulations of mouse hippocampal neurons

In the hippocampus, the synaptic vesicle protein synaptoporin (SPO) has been reported to be exclusively enriched in the granule cell axons, the mossy fibers. In this study, we show that in adult rats and mice SPO immunoreactivity (IR) is also detectable in strata oriens, radiatum, and lacunosum-moleculare of CA1-CA3, as well as perisomatically in the hippocampus proper and fascia dentata. In situ hybridization confirmed that SPO mRNA was present in granule cells and CA3 pyramidal cells but not in CA1 pyramidal cells. Importantly, cells scattered throughout the hippocampal layers resembling the distribution of interneurons were found to synthesize high amounts of SPO mRNA, too. Thus, these findings indicate that SPO expression in the hippocampus was underestimated until now. Moreover, double-labeling immunohistochemistry and confocal microscopy revealed selective colocalization of SPO and glutamate decarboxylase (GAD 65), a marker for gamma-aminobutyric acid (GABA)ergic terminals. To identify SPO expressing interneurons, in situ hybridization was combined with immunocytochemistry against parvalbumin (PV), calbindin (CB), calretinin (CR), cholecystokinin (CCK), and vasoactive intestinal polypeptide (VIP). We found that SPO transcripts were differentially expressed by various interneuron subpopulations in the hippocampus of C57Bl/6 mice (PV 44.2%, CB 46.3%, CR 19.3%, CCK 38.6%, VIP 59.9%). Immunoelectron microscopy for SPO labeled synaptic vesicle profiles in distinct symmetric and asymmetric synapses. In conclusion, our data demonstrate that hippocampal principal cells and interneurons display a variety of synaptic vesicles that are likely to contribute to the functional characteristics of their output synapses.

Complex effects of CNQX on CA1 interneurons of the developing rat hippocampus

We have investigated the effect of the alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor antagonist, 6-cyano-7-nitro-quinoxaline-2,3-dione (CNQX), on spontaneous GABA(A) receptor-mediated transmission in the hippocampal CA1 subfield. On average, simultaneous recordings from CA1 str. radiatum interneurons and pyramidal cells showed that CNQX application doubled the frequency of bicuculline sensitive spontaneous inhibitory postsynaptic currents (sIPSCs) without apparently changing their amplitude. However, despite the increase in sIPSC frequency, current-clamp recording showed that CNQX application was sufficient in most cases to depolarize interneurons to firing threshold. In contrast, CNQX application could not induce firing in pyramidal cells. In the presence of tetrado-toxin (TTX), CNQX increased interneuron membrane conductance, and depolarized interneurons from resting potentials. The axons of the studied interneurons ramify widely in the CA1 region and suggest that the cells of our sample are mostly involved with control of dendritic excitability. Our results indicate that CNQX-induced increase of sIPSC frequency is not limited to excitatory cells, but also impacts GABAergic interneurons. However, despite the increase of sIPSC frequency, CNQX-induced depolarization is sufficient to selectively generate firing in interneurons and thus modify the network properties mediated by GABA(A) receptors in the hippocampus.

GABA application to hippocampal CA3 or CA1 stratum lacunosum-moleculare excites an interneuron network

Whole cell voltage-clamp recording and focal application of the neurotransmitter gamma-aminobutyric acid (GABA) were used to investigate the ability of exogenous GABA applied to different locations within the guinea pig hippocampal slice to trigger a giant GABA-mediated postsynaptic current (GPSC) in pyramidal cells. A GPSC reflects the synchronous release of GABA from a group of interneurons. Recordings were done in the presence of 4-aminopyridine (4-AP) and blockers of ionotropic glutamatergic synaptic transmission. Spontaneous GPSCs occurred rhythmically in pyramidal cells under these conditions. Brief focal pressure application of GABA (500 microM; 30-200 ms) to CA3 stratum lacunosum-moleculare (SLM) or to the border between CA3 s. radiatum (SR) and SLM triggered an "all-or-none" GPSC in CA3 and CA1 pyramidal cells that looked like the spontaneous GPSCs. During the refractory period following a spontaneous GPSC, application of GABA could not trigger a GPSC. Both spontaneous GPSCs and GPSCs triggered by exogenous GABA were blocked by suppressing synaptic transmission with high Mg(2+)/low Ca(2+) bath solution. On the other hand, focal application of GABA to CA3 s. oriens (SO) or to proximal SR did not trigger a GPSC in the CA3 pyramidal cell; instead it produced a graded response. Focal application of GABA to regions other than CA3 was also tested. Focal application of GABA to CA1 SLM always triggered a GPSC in the CA3 pyramidal cell. Focal application of GABA within the outer two-thirds of the dentate molecular layer often elicited a GPSC in the CA3 pyramidal cell. In contrast, focal application of GABA to CA1 SO, to CA1 SR, or to the hilus elicited no current response in the CA3 pyramidal cell. These data indicate that the GPSC recorded in pyramidal cells that was triggered by focal GABA application resulted from the synchronous synaptic release of GABA from activated interneurons rather than from the binding of exogenous GABA to receptors on the pyramidal cell. Furthermore, the "all-or-none" nature of the response to SLM GABA applications of different durations indicates that the exogenous GABA was exciting (directly or indirectly) some members of a network of interneurons, which in turn recruited the rest of the network, rather than individually activating each interneuron that contributed to the GPSC. Interestingly, the effective sites of GABA application--CA3 SLM, CA1 SLM, and the outer two-thirds of the dentate molecular layer--are also the sites which receive direct innervation from the entorhinal cortex in an intact animal.

Physiological and morphological diversity of immunocytochemically defined parvalbumin- and cholecystokinin-positive interneurones in CA1 of the adult rat hippocampus

To investigate the electrophysiological properties, synaptic connections, and anatomy of individual parvalbumin-immunoreactive (PV-IR) and cholecystokinin-immunoreactive (CCK-IR) interneurones in CA1, dual intracellular recordings using biocytin-filled microelectrodes in slices of adult rat hippocampus were combined with fluorescence labelling of PV- and CCK-containing cells. Of 36 PV-IR cells, 29 were basket cells, with most of their axonal arbours in the stratum pyramidale (SP). Six were bistratified cells with axons ramifying throughout stratum oriens (SO) and stratum radiatum (SR). One was a putative axo-axonic cell with an axonal arbour confined to half of the SP and a narrow adjacent region of the SO. Of 27 CCK-IR neurones, 13 were basket cells, with most of their axonal arbours in the SP, and included basket cells with somata in the SP (6), SO (3), and SR (2) and at the border between the stratum lacunosum-moleculare (SLM) and the SR (2). In addition, several dendrite-targeting cell classes expressed CCK-IR: 4 of 9 bistratified cells with axons ramifying in the SO and SR; all five Schaffer-associated cells whose axons ramified extensively in the SR; both cells classified as quadrilaminar because their axons ramified in the SO, SP, SR, and SLM; one SO-SO cell whose dendritic and axonal arbours were contained within the SO; and one perforant path-associated cell with axonal and dendritic arbours within the distal SR and SLM. The majority (31 of 36) of PV-IR neurones recorded were fast-spiking, and most fast-spiking cells tested (25 of 29 basket, 1 axo-axonic, and 5 of 6 bistratified cells) were PV-IR. However, 1 of 6 regular-spiking basket, 1 of 4 regular-spiking bistratified, and 3 of 5 burst-firing basket cells were also PV-IR. In contrast, the majority (17 of 27) of the CCK-IR neurones recorded were regular-spiking, 3 were burst-firing, and 7 were fast-spiking. These data confirm that the majority of PV-IR and CCK-IR axon terminals innervate proximal portions of CA1 pyramidal cells. Most PV-IR cells are fast-spiking, whereas most CCK-IR cells are regular-spiking. In both neurochemical classes basket cells predominate, but both groups included subpopulations of dendrite-targeting cells. Despite these similarities, the two populations exhibited very different somatic distributions, and each contained cellular morphologies not represented in the other.

Input-dependent synaptic targeting of alpha(2)-subunit-containing GABA(A) receptors in synapses of hippocampal pyramidal cells of the rat

Pyramidal cells, expressing at least 14 subunits of the heteropentameric GABA(A) receptor, receive GABAergic input on their soma and proximal dendrites from basket cells, activating GABA(A) receptors and containing either parvalbumin or cholecystokinin and vasoactive intestinal polypeptide. The properties of GABA(A) receptors are determined by the subunit composition, and synaptic receptor content governs the effect of the presynaptic neuron. Using a quantitative electron microscopic immunogold technique, we tested whether the synapses formed by the two types of basket cell show a difference in the subunit composition of GABA(A) receptors. Terminals of one of the basket cells were identified by antibodies to parvalbumin. Synapses made by parvalbumin-negative terminals showed five times more immunoreactivity for the alpha(2) subunit than synapses made by parvalbumin-positive basket cells, whose synapses were frequently immunonegative. This difference is likely to be due to specific GABA(A) receptor alpha subunit composition, because neither synaptic size nor immunoreactivity for the beta(2/3) subunits, indicating total receptor content, was different in these two synapse populations. Synapses established by axo-axonic cells on axon initial segments showed an intermediate number of immunoparticles for the alpha(2) subunit compared to those made by basket cells but, due to their smaller size, the density of the alpha(2) subunit immunoreactivity was higher in synapses on the axon. Because the two basket cell types innervate the same domain of the pyramidal cell, the results indicate that pyramidal cells have mechanisms to target GABA(A) receptors, under presynaptic influence, preferentially to distinct synapses. The two basket cell types act via partially distinct GABA(A) receptor populations.

Interneurons unbound

Local-circuit, gamma-aminobutyric acid-releasing inhibitory interneurons of the hippocampus and cortex have traditionally been considered as the regulators of principal neuron activity--the yin to the excitatory yang. Recent evidence indicates that, in addition to that role, their network connectivity and the properties of their intrinsic voltage-gated currents are finely tuned to permit inhibitory interneurons to generate and control the rhythmic output of large populations of both principal cells and other populations of inhibitory interneurons. This review brings together recently described properties and emerging principles of interneuron function that indicate a much more complex role for these cells than just providers of inhibition.

Unusual target selectivity of perisomatic inhibitory cells in the hilar region of the rat hippocampus

Perisomatic inhibitory innervation of all neuron types profoundly affects their firing characteristics and vulnerability. In this study we examined the postsynaptic targets of perisomatic inhibitory cells in the hilar region of the dentate gyrus where the proportion of potential target cells (excitatory mossy cells and inhibitory interneurons) is approximately equal. Both cholecystokinin (CCK)- and parvalbumin-immunoreactive basket cells formed multiple contacts on the somata and proximal dendrites of mossy cells. Unexpectedly, however, perisomatic inhibitory terminals arriving from these cell types largely ignored hilar GABAergic cell populations. Eighty-ninety percent of various GABAergic neurons including other CCK-containing basket cells received no input from CCK-positive terminals. Parvalbumin-containing cells sometimes innervated each other but avoided 75% of other GABAergic cells. Overall, a single mossy cell received 40 times more CCK-immunoreactive terminals and 15 times more parvalbumin-positive terminals onto its soma than the cell body of an average hilar GABAergic cell. In contrast to the pronounced target selectivity in the hilar region, CCK- and parvalbumin-positive neurons innervated each other via collaterals in stratum granulosum and moleculare. Our observations indicate that the inhibitory control in the hilar region is qualitatively different from other cortical areas at both the network level and the level of single neurons. The paucity of perisomatic innervation of hilar interneurons should have profound consequences on their action potential generation and on their ensemble behavior. These findings may help explain the unique physiological patterns observed in the hilus and the selective vulnerability of the hilar cell population in various pathophysiological conditions.

Double immunofluorescence, peroxidase labelling and ultrastructural analysis of interneurones following prolonged electrophysiological recordings in vitro

Inhibitory hippocampal and neocortical interneurones comprise a physiologically, morphologically and neurochemically heterogenous cell population. To identify the roles each class of interneurone plays within a given circuit it is necessary to correlate the electrophysiological properties of individual cells with their neurochemistry and morphology at both the light and electron microscopic level. However, the optimal conditions required for any one part of the protocol typically compromise the results from another. We have developed a protocol which allows the neurochemical content, gross morphology and ultrastructure details of biocytin-filled neurones to be recovered following long, dual intracellular recordings in thick mature slices maintained in an interface recording chamber, helping define sub-populations which could not otherwise be determined. Dual immunofluorescence is performed by incubating the tissue in monoclonal and polyclonal antibodies simultaneously, prior to visualization of biocytin-labelling with precipitation of a peroxidase reaction product. By using a biotinylated anti-avidin D antibody (Vector Laboratories), the intensity of this precipitation can be enhanced further where necessary. It is envisaged that this protocol can not only help determine the neurochemical content of cells recorded in similar in vivo studies, but that the ability to amplify peroxidase labelling in poorly filled cells is also of interest.

Protein kinase A mediates the modulation of the slow Ca(2+)-dependent K(+) current, I(sAHP), by the neuropeptides CRF, VIP, and CGRP in hippocampal pyramidal neurons

We have studied modulation of the slow Ca(2+)-activated K(+) current (I(sAHP)) in CA1 hippocampal pyramidal neurons by three peptide transmitters: corticotropin releasing factor (CRF, also called corticotropin releasing hormone, CRH), vasoactive intestinal peptide (VIP), and calcitonin gene-related peptide (CGRP). These peptides are known to be expressed in interneurons. Using whole cell voltage clamp in hippocampal slices from young rats, in the presence of tetrodotoxin (TTX, 0.5 microM) and tetraethylammonium (TEA, 5 mM), I(sAHP) was measured after a brief depolarizing voltage step eliciting inward Ca(2+) current. Each of the peptides CRF (100-250 nM), VIP (400 nM), and CGRP (1 microM) significantly reduced the amplitude of I(sAHP). Thus the I(sAHP) amplitude was reduced to 22% by 100 nM CRF, to 17% by 250 nM CRF, to 22% by 400 nM VIP, and to 40% by 1 microM CGRP. We found no consistent concomitant changes in the Ca(2+) current or in the time course of I(sAHP) for any of the three peptides, suggesting that the suppression of I(sAHP) was not secondary to a general suppression of Ca(2+) channel activity. Because each of these peptides is known to activate the cyclic AMP (cAMP) cascade in various cell types, and I(sAHP) is known to be suppressed by cAMP via the cAMP-dependent protein kinase (PKA), we tested whether the effects on I(sAHP) by CRF, VIP, and CGRP are mediated by PKA. Intracellular application of the PKA-inhibitor Rp-cAMPS significantly reduced the suppression of I(sAHP) by CRF, VIP, and CGRP. Thus with 1 mM Rp-cAMPS in the recording pipette, the average suppression of I(sAHP) was reduced from 78 to 26% for 100 nM CRF, from 83 to 32% for 250 nM CRF, from 78 to 30% for 400 nM VIP, and from 60 to 7% for 1 microM CGRP. We conclude that CRF, VIP, and CGRP suppress the slow Ca(2+)-activated K(+) current, I(sAHP), in CA1 hippocampal pyramidal neurons by activating the cAMP-dependent protein kinase, PKA. Together with the monoamine transmitters norepinephrine, serotonin, histamine, and dopamine, these peptide transmitters all converge on the cAMP cascade modulating I(sAHP).

Nucleus reuniens thalami innervates gamma aminobutyric acid positive cells in hippocampal field CA1 of the rat

The aim of the present study was to investigate whether the nucleus reuniens thalami (RE) innervates inhibitory cells in hippocampal field CA1. Therefore, we examined the RE-CA1 input at the ultrastructural level. RE axons were anterogradely labeled with biotin-conjugated dextran amine (BDA), in combination with pre-embedding gamma aminobutyric acid (GABA)-immunolabeling of neurons in CA1. We observed that part of the BDA-labeled axons formed asymmetrical (i.e. excitatory) synapses on GABA-positive dendrites. Based on these data, which are in line with our previously published electrophysiological observations (Dolleman-Van der Weel, M.J., Lopes da Silva, F.H. and Witter, M.P., Nucleus reuniens thalami modulates activity in hippocampal field CA1 through excitatory and inhibitory mechanisms. J. Neurosci., 17 (1997) 5640-5650), we propose that RE-CA1 input partially influences hippocampal transmission through activation of local inhibitory interneurons.

Alterations of perisomatic GABA synapses on hippocampal CA1 inhibitory interneurons and pyramidal cells in the kainate model of epilepsy

In the kainate model of epilepsy, electrophysiological and anatomical modifications occur in inhibitory circuits of the CA1 region of the rat hippocampus. Using postembedding GABA immunocytochemistry and electron microscopy, we characterized perisomatic GABA and non-GABA synaptic contacts in CA pyramidal cells, and GABAergic interneurons of stratum oriens/alveus and stratum lacunosum-moleculare, and examined if changes occurred at these synapses at two weeks post-kainate treatment. We found that, in control rats, the number and total length of perisomatic GABA synapses were significantly smaller (approximately 40-50%) in lacunosum-moleculare interneurons than in oriens/alveus interneurons and pyramidal cells. Additionally, the number and total length of perisomatic non-GABA synapses were different among all cell types, with these parameters increasing significantly in the following order: pyramidal cells<lacunosum-moleculare interneurons<oriens/alveus interneurons. Following kainate treatment, we found that the number and total length of GABA synapses were significantly increased in lacunosum-moleculare interneurons (by 76% and 100%, respectively), but were unchanged in pyramidal cells and oriens/alveus interneurons. In addition, the mean length of individual GABA synapses was significantly increased (by 17%) in pyramidal cells after kainate treatment. In contrast, no changes were observed at non-GABA synapses in any cell type examined after kainate treatment. These results indicate that, in control animals, the ultrastructural correlates of perisomatic GABA inhibition are less pronounced in lacunosum-moleculare than oriens/alveus interneurons or pyramidal cells, whereas those of perisomatic excitation are more prominent in oriens/alveus than lacunosum-moleculare interneurons, and much less present in pyramidal cells. In addition, our results with kainate-treated animals suggest that cell-specific changes in perisomatic inhibition may occur in CA1 inhibitory interneurons in the chronically hyperexcitable hippocampus. The ultrastructural correlates of perisomatic inhibition were increased in lacunosum-moleculare interneurons, which may thus suggest some disinhibition of pyramidal cells. However, the ultrastructural correlates of perisomatic inhibition were increased in pyramidal cells, implying some enhancement of perisomatic inhibition of principal cells in the hyperexcitable hippocampus.

Changes in excitatory and inhibitory circuits of the rat hippocampus 12-14 months after complete forebrain ischemia

Changes in interneuron distribution and excitatory connectivity have been investigated in animals which had survived 12-14 months after complete forebrain ischemia, induced by four-vessel occlusion. Anterograde tracing with Phaseolus vulgaris leucoagglutinin revealed massive Schaffer collateral input even to those regions of the CA1 subfield where hardly any surviving pyramidal cells were found. Boutons of these Schaffer collaterals formed conventional synaptic contacts on dendritic spines and shafts, many of which likely belong to interneurons. Mossy fibres survived the ischemic challenge, however, large mossy terminals showed altered morphology, namely, the number of filopodiae on these terminals decreased significantly. The entorhinal input to the hippocampus did not show any morphological alterations. The distribution of interneurons was investigated by neurochemical markers known to label functionally distinct GABAergic cell populations. In the hilus, spiny interneurons showed a profound decrease in number. This phenomenon was not as obvious in CA3, but the spiny metabotropic glutamate receptor 1alpha-positive non-pyramidal cells, some of which contain calretinin or substance P receptor, disappeared from stratum lucidum of this area. In the CA1 region, somatostatin immunoreactivity disappeared from stratum oriens/lacunosum-moleculare-associated cells, while in metabotropic glutamate receptor 1alpha-stained sections these cells seemed unaffected in number. Other interneurons did not show an obvious decrease in number. In stratum radiatum of the CA1 subfield, some interneuron types had altered morphology: the substance P receptor-positive dendrites lost their characteristic radial orientation, and the metabotropic glutamate receptor 1alpha-expressing cells became extremely spiny. The loss of inhibitory interneurons at the first two stages of the trisynaptic loop coupled with a well-preserved excitatory connectivity among the subfields suggests that hyperexcitability in the surviving dentate gyrus and CA3 may persist even a year after the ischemic impact. The dorsal CA1 region is lost; nevertheless hyperactivity, if it occurs, may have a route to leave the hippocampus via the longitudinally extensive axon collaterals of CA3 pyramidal cells, which may activate the subiculum and entorhinal cortex with a relay in the surviving ventral hippocampal CA1 region.

Presynaptically located CB1 cannabinoid receptors regulate GABA release from axon terminals of specific hippocampal interneurons

To understand the functional significance and mechanisms of action in the CNS of endogenous and exogenous cannabinoids, it is crucial to identify the neural elements that serve as the structural substrate of these actions. We used a recently developed antibody against the CB1 cannabinoid receptor to study this question in hippocampal networks. Interneurons with features typical of basket cells showed a selective, intense staining for CB1 in all hippocampal subfields and layers. Most of them (85.6%) contained cholecystokinin (CCK), which corresponded to 96.9% of all CCK-positive interneurons, whereas only 4.6% of the parvalbumin (PV)-containing basket cells expressed CB1. Accordingly, electron microscopy revealed that CB1-immunoreactive axon terminals of CCK-containing basket cells surrounded the somata and proximal dendrites of pyramidal neurons, whereas PV-positive basket cell terminals in similar locations were negative for CB1. The synthetic cannabinoid agonist WIN 55,212-2 (0.01-3 microM) reduced dose-dependently the electrical field stimulation-induced [3H]GABA release from superfused hippocampal slices, with an EC50 value of 0. 041 microM. Inhibition of GABA release by WIN 55,212-2 was not mediated by inhibition of glutamatergic transmission because the WIN 55,212-2 effect was not reduced by the glutamate blockers AP5 and CNQX. In contrast, the CB1 cannabinoid receptor antagonist SR 141716A (1 microM) prevented this effect, whereas by itself it did not change the outflow of [3H]GABA. These results suggest that cannabinoid-mediated modulation of hippocampal interneuron networks operate largely via presynaptic receptors on CCK-immunoreactive basket cell terminals. Reduction of GABA release from these terminals is the likely mechanism by which both endogenous and exogenous CB1 ligands interfere with hippocampal network oscillations and associated cognitive functions.

Expression of neurotrophins in hippocampal interneurons immunoreactive for the neuropeptides somatostatin, neuropeptide-Y, vasoactive intestinal polypeptide and cholecystokinin

Using a double detection method, which combines in situ hybridization for the detection of neurotrophin messenger RNA with immunocytochemistry against the neuropeptides somatostatin, neuropeptide Y, vasoactive intestinal polypeptide and cholecystokinin, we have analysed the expression of the neurotrophins, nerve growth factor, brain-derived neurotrophic factor and neurotrophin-3, in distinct populations of neuropeptide-immunoreactive hippocampal interneurons. Nerve growth factor messenger RNA expression was found in subsets of the four subpopulations of neuropeptide-immunoreactive interneurons. The highest degree of co-localization was observed in the neuropeptide-Y-positive cells (up to 70%) and in somatostatin-immunoreactive cells (48%). Only small subsets of cholecystokinin- and vasoactive intestinal polypeptide-positive neurons (21% and 10%, respectively) displayed nerve growth factor hybridization signals. In contrast, expression of neurotrophin-3 messenger RNA was exclusively observed in 26% of neuropeptide-Y-immunoreactive cells. Brain-derived neurotrophic factor hybridization signals were never detected in the neuropeptide-positive hippocampal interneurons. Morphological analysis of neuropeptide-immunoreactive interneurons that express or lack nerve growth factor messenger RNA revealed that most perisomatic inhibitory neurons, such as large vasoactive intestinal polypeptide/ cholecystokinin-immunoreactive cells, showed positive nerve growth factor hybridization signals. In addition, some somatostatin/neuropeptide-Y-immunoreactive interneurons, which are responsible for dendritic inhibition of principal hippocampal neurons, expressed nerve growth factor messenger RNA. In contrast, interneurons specialized to innervate other GABAergic cells, such as small vasoactive intestinal polypeptide-positive cells, lacked nerve growth factor expression. All these data indicate that expression of neurotrophins is differentially regulated in functionally distinct classes of hippocampal interneurons immunoreactive for neuropeptides. We also analysed whether neuropeptide-immunoreactive interneurons expressing neurotrophins were targets of the GABAergic septohippocampal pathway. We used a triple detection method, combining anterograde tracing of this connection, with in situ hybridization for the detection of neurotrophin mRNA, and immunocytochemistry against neuropeptides. Our data showed that the four populations of hippocampal interneurons studied (somatostatin, neuropeptide-Y, vasoactive intestinal polypeptide and cholescystokinin) received GABAergic afferents from the septum. However, no preference for neuropeptide-immunoreactive cells expressing neurotrophins was observed, compared to neuropeptide-positive neurons lacking neurotrophin expression.

IPSPs elicited in CA1 pyramidal cells by putative basket cells in slices of adult rat hippocampus

CA1 basket cells are identifiable by an axonal arbour largely confined to, and spanning, the entire depth of stratum pyramidale where they innervate pyramidal somata and proximal dendrites. Basket cells display a range of electrophysiological properties and the inhibitory postsynaptic potentials (IPSPs) they elicit in pyramidal cells vary widely in duration. To determine whether these parameters are correlated, we used paired intracellular recordings, with biocytin filling, in pyramidal cells of adult hippocampal slices, and studied gamma-aminobutyric acid (GABAA) IPSPs (n = 43) elicited by putative basket cells (n = 35) with axons largely confined to stratum pyramidale in simultaneously recorded pyramidal cells. Fast-spiking interneurons elicited relatively brief IPSPs, while IPSPs elicited by burst-firing cells were amongst the slowest. Regular spiking interneurons elicited fast and slow GABAA IPSPs, but any one interneuron elicited IPSPs with remarkably similar durations in two to four pyramidal targets. However, with different types of target for a single putative basket cell, IPSPs elicited in postsynaptic interneurons were briefer than in pyramidal cells. Vertical oriens cells with somata in stratum oriens and a narrow, sparse axonal arbour in stratum pyramidale in transverse hippocampal slices, elicited IPSPs whose rise times and half widths clustered around intermediate values. Durations of IPSPs in pyramidal cells thus correlate, to a degree, with the physiological properties of presynaptic basket cells. The seven-fold range of durations observed (10-70 ms half widths) may underlie contributions made by different basket cells to hippocampal rhythms of different frequencies.

Structural basis of the cholinergic and serotonergic modulation of GABAergic neurons in the hippocampus

Ascending subcortical pathways effectively modulate hippocampal information processing. Two components, the cholinergic and serotonergic pathways have been demonstrated to play an important role in the generation of behaviour-dependent hippocampal EEG patterns. Several findings suggest that the above projections influence the activity of hippocampal interneurons. Here we review the available data from physiological, pharmacological and receptor localization experiments, drawing attention to the crucial role of interneurons in the transfer and amplification of subcortical effects on cortical information processing. We hypothesize that, by exerting diverse actions on different subsets of interneurons, the cholinergic and serotonergic systems might change the balance of somatic and dendritic inhibition, and consequently change the integrative properties of hippocampal principal cells.

Medial septal and median raphe innervation of vasoactive intestinal polypeptide-containing interneurons in the hippocampus

Vasoactive intestinal polypeptide-immunoreactive interneurons are known to form three anatomically and neurochemically well-characterized neuron populations in the hippocampus. Two of these establish synaptic contacts selectively with other GABAergic cells (interneuron-selective cells), whereas the third type innervates pyramidal cell bodies and proximal dendrites like a conventional basket cell. Our aim was to examine which of the vasoactive intestinal polypeptide-containing interneuron populations are among the targets of GABAergic septohippocampal and serotonergic raphe-hippocampal pathways. Anterograde tracing with Phaseolus vulgaris leucoagglutinin combined with double immunocytochemistry for vasoactive intestinal polypeptide was used at the light and electron microscopic levels. Our results show that both interneuron-selective cells and vasoactive intestinal polypeptide-containing basket cells receive synaptic input from the medial septum and median raphe nucleus. The GABAergic component of the septohippocampal pathway establishes multiple contacts on both cell types. In the case of the raphe-hippocampal projection, single or double contacts were more frequent on vasoactive intestinal polypeptide-positive interneuron selective cells (76%), whereas multiple contacts predominated on basket cells (83%). The extrinsic GABAergic innervation of interneuron-selective cells in the hippocampus indicates a complex interaction among GABAergic systems, which might ensure the timing and rhythmic synchronization of inhibitory processes in the hippocampus. On the other hand, our results suggest that the serotonergic effect on perisomatic inhibition is exerted via vasoactive intestinal polypeptide-containing basket cells that are functionally distinct from their parvalbumin-positive relatives, which appear to escape control of serotonergic as well as local interneuron-selective cells.

Nicotinic receptor activation excites distinct subtypes of interneurons in the rat hippocampus

We examined the function of nicotinic acetylcholine receptors (nAChRs) in interneurons of area CA1 of the rat hippocampus. CA1 interneurons could be classified into three categories based on nicotinic responses. The first class was depolarized by alpha7 nAChRs, found in all layers of CA1 and as a group, had axonal projections to all neuropil layers of CA1. The second class had both fast alpha7 and slow non-alpha7 nAChR depolarizing responses, was localized primarily to the stratum oriens, and had axonal projections to the stratum lacunosum-moleculare. The third group had no nicotinic response. This group was found in or near the stratum pyramidale and had axonal projections almost exclusively within and around this layer. Low concentrations (500 nM) of nicotine desensitized fast and slow nAChR responses. These findings demonstrate that there are distinct subsets of interneurons with regard to nicotinic receptor expression and with predictable morphological properties that suggest potential cellular actions for nicotinic receptor activation in normal CNS function and during nicotine abuse.

Neurochemistry and pharmacology of the major hippocampal transmitter systems: synaptic and nonsynaptic interactions

Hippocampus plays a crucial role in important brain functions (e.g. memory, learning) thus in the past two decades this brain region became a major objective of neuroscience research. During this period large number of anatomical, neurochemical and electrophysiological data have been accumulated. While excellent reviews have been published on the anatomy and electrophysiology of hippocampal formation, the neurochemistry of this area has not been thoroughly surveyed. Therefore the aim of this review is to summarize the neurochemical and pharmacological data on the release of the major neurotransmitters found in the hippocampal region: glutamate (GLU), gamma-amino butyric acid (GABA), acetylcholine (ACh), noradrenaline (NA) and serotonin (5-HT). In addition, this review analyzes the synaptic and nonsynaptic interactions between hippocampal neuronal elements and overviews how auto- and heteroreceptors are involved in the presynaptic modulation of transmitter release. The presented data clearly show that transmitters released from axon terminals without synaptic contact play an important role in the fine tuning of communication between neurons within a neuronal circuit.

Postsynaptic targets of somatostatin-immunoreactive interneurons in the rat hippocampus

Two characteristic interneuron types in the hippocampus, the so-called hilar perforant path-associated cells in the dentate gyrus and stratum oriens/lacunosum-moleculare neurons in the CA3 and CA1 regions, were suggested to be involved in feedback circuits. In the present study, interneurons identical to these cell populations were visualized by somatostatin-immunostaining, then reconstructed, and processed for double-immunostaining and electron microscopy to establish their postsynaptic target selectivity. A combination of somatostatin-immunostaining with immunostaining for GABA or other interneuron markers revealed a quasi-random termination pattern. The vast majority of postsynaptic targets were GABA-negative dendritic shafts and spines of principal cells (76%), whereas other target elements contained GABA (8%). All of the examined neurochemically defined interneuron types (parvalbumin-, calretinin-, vasoactive intestinal polypeptide-, cholecystokinin-, substance P receptor-immunoreactive neurons) received innervation from somatostatin-positive boutons. Recent anatomical and electrophysiological data showed that the main excitatory inputs of somatostatin-positive interneurons originate from local principal cells. The present data revealed a massive GABAergic innervation of distal dendrites of local principal cells by these feedback driven neurons, which are proposed to control the efficacy and plasticity of entorhinal synaptic input as a function of local principal cell activity and synchrony.

Neurochemical features and synaptic connections of large physiologically-identified GABAergic cells in the rat frontal cortex

Physiological and morphological properties of large non-pyramidal cells immunoreactive for cholecystokinin, parvalbumin or somatostatin were investigated in vitro in the frontal cortex of 18-22-day-old rats. These three peptides were expressed in separate populations including large cells. Cholecystokinin cells and parvalbumin cells made boutons apposed to other cell bodies, but differed in their firing patterns in response to depolarizing current pulses. Parvalbumin cells belonged to fast-spiking cells. Parvalbumin fast-spiking cells also included chandelier cells. In contrast, cholecystokinin cells were found to be regular-spiking non-pyramidal cells or burst-spiking non-pyramidal cells with bursting activity from hyperpolarized potentials (two or more spikes on slow depolarizing humps). Large somatostatin cells belonged to the regular-spiking non-pyramidal category and featured wide or ascending axonal arbors (wide arbor cells and Martinotti cells) which did not seem to be apposed to the somata so frequently as large cholecystokinin and parvalbumin cells. For electron microscopic observations, another population of eight immunohistochemically-uncharacterized non-pyramidal cells were selected: (i) five fast spiking cells including one chandelier cell which are supposed to contain parvalbumin, and (ii) three large regular-spiking non-pyramidal cells with terminals apposed to somata, which are not considered to include somatostatin cells, but some of which may belong to cholecystokinin cells. The fast-spiking cells other than a chandelier cell and the large regular-spiking non-pyramidal cells made GABA-positive synapses on somata (4% and 12% of the synapses in two small to medium fast-spiking cells, 22% and 35% of the synapses in two large fast-spiking cells, and 10%, 18% and 37% of the synapses in three large regular-spiking non-pyramidal cells). A few terminals of the fast-spiking and regular-spiking non-pyramidal cells innervated GABAergic cells. About 30% of the fast-spiking cell terminals innervated spines, but few of the regular-spiking non-pyramidal cell terminals did. A fast-spiking chandelier cell made GABA-positive synapses on GABA-negative axon initial segments. These results suggest that large GABAergic cells are heterogeneous in neuroactive substances, firing patterns and synaptic connections, and that cortical cells receive heterogeneous GABAergic somatic inputs.

GABAergic axon terminals at perisomatic and dendritic inhibitory sites show different immunoreactivities against two GAD isoforms, GAD67 and GAD65, in the mouse hippocampus: a digitized quantitative analysis

Glutamic acid decarboxylase (GAD), the gamma-aminobutyric acid (GABA)-synthetic enzyme, consists of two isoforms, GAD67 and GAD65. Although distributions of the two GAD isoforms at the somatic level are known to be heterogeneous among different subpopulations of GABAergic neurons, those at the synaptic level have not been investigated. In order to analyze quantitatively the two GAD-isoform immunoreactivities in axon terminals, we combined confocal laser scanning microscopy with digitized image analysis to measure the gray levels of immunofluorescent signals for the two GAD isoforms in a large number of individual boutons in each hippocampal and dentate layer of the mouse. Synaptic boutons exhibited lamina-specific immunoreactivities against the GAD isoforms. Boutons in the principal cell layers (stratum pyramidale of the hippocampus proper and the granule cell layer of the dentate gyrus) showed more intense immunoreactivity against GAD67 than those in the dendritic layers (strata lacunosum-moleculare, radiatum, and oriens of the hippocampus proper and the molecular layer of the dentate gyrus). By contrast, boutons in the dendritic layers showed more intense immunoreactivity against GAD65 than those in the principal cell layers. Such differential distributions could be correlated to the GAD-isoform immunoreactivities in the axon terminals originating from parvalbumin-containing neurons, a particular subpopulation of hippocampal GABAergic neurons mainly innervating the perisomatic domain of principal neurons. In addition to previously reported physiological and pharmacological differences between the GABAergic synapses on perisomatic domain and those on distal dendrites, the present results suggest a functional differentiation of GABAergic synapses between these two inhibitory sites.

Light and electron microscopic study of neuronal nitric oxide synthase-immunoreactive neurons in the rat subiculum

Neurons in the rat subiculum that are capable of producing nitric oxide were studied by using an antibody to the neuronal isoform of nitric oxide synthase (nNOS). In the light microscope, the staining pattern with the nNOS antibody closely resembled that seen following histochemical processing with nicotinamide adenine dinucleotide phosphate diaphorase. Immunostained neurons were found in all layers, and, in addition, large dendrites in the apical dendrite layer were also immunopositive. Although a few immunolabelled cells had the typical morphology of interneurons, most were found to have the characteristics of pyramidal neurons. In the subiculum, these immunoreactive pyramidal neurons were concentrated mainly in the most superficial cell layers and closest to the CA1 region, but pyramidal neurons in the CA1 layer of the hippocampus were consistently immunonegative. Immunopositive profiles in the subiculum were studied in the electron microscope and compared with unlabelled structures. Ultrastructural criteria suggest that both pyramidal and nonpyramidal subicular neurons are immunopositive for nNOS. Large, spiny dendrites and smaller, varicose dendrites were found to be immunoreactive for nNOS. Vesicle-containing profiles were probably presynaptic axons, and immunopositive boutons were seen to make symmetrical and asymmetrical synaptic contacts.

Neurocalcin-immunoreactive cells in the rat hippocampus are GABAergic interneurons

Neurocalcin (NC) is a recently described calcium-binding protein isolated and characterized from bovine brain. NC belongs to the neural calcium-sensor proteins defined by the photoreceptor cell-specific protein recoverin that have been proposed to be involved in the regulation of calcium-dependent phosphorylation in signal transduction pathways. We analyzed the distribution and morphology of the NC-immunoreactive (IR) neurons in the rat dorsal hippocampus and the coexistence of NC with GABA and different neurochemical markers which label perisomatic inhibitory cells [parvalbumin (PV) and cholecystokinin (CCK)], mid-proximal dendritic inhibitory cells [calbindin D28k (CB)], distal dendritic inhibitory cells [somatostatin (SOM) and neuropeptide Y (NPY)], and interneurons specialized to innervate other interneurons [calretinin (CR) and vasoactive intestinal polypeptide (VIP)]. NC-IR cells were present in all layers of the dentate gyrus and hippocampal fields. In the dentate gyrus, NC-IR cells were concentrated in the granule cell layer, especially in the hilar border, whereas in the CA fields they were most frequently found in the stratum radiatum. NC-IR cells were morphologically heterogeneous and exhibited distinctive features of non-principal cells. In the dentate gyrus, pyramidal-like, multipolar and fusiform (horizontal and vertical) cells were found. In the CA3 region most NC-IR cells were multipolar, but vertical and horizontal fusiform cells also appeared. In the CA1 region, where NC-IR cells showed most frequently vertically arranged dendrites, multipolar, bitufted and fusiform (vertical and horizontal) cells could be distinguished. All the NC-IR cells were found to be GABA-IR in all hippocampal layers and regions, and they represented about 19% of the GABA-positive cells. NC/CB, NC/CR and NC/VIP double-labeled cells were found in all hippocampal regions, and represented 29%, 24% and 18% of the NC-IR cells, respectively. NC and CCK did not coexist in the dentate gyrus; however, 9% of the NC-IR cells in the CA fields also contained CCK. No coexistence of NC with PV, SOM or NPY was found in any hippocampal region. We conclude that NC is exclusively expressed by interneurons in the rat hippocampus. NC-IR cells are a morphologically and neurochemically heterogeneous subset of GABAergic non-principal cells, which, on the basis of the known termination pattern of the colocalizing markers, are also functionally heterogeneous and are mainly involved in feed-forward dendritic inhibition in the commissural-associational and Schaffer collateral termination zones (CB containing cells), in innervation of other interneurons (CR- and VIP-containing cells), and in perisomatic inhibition (CCK-containing cells). NC is never present in perisomatic inhibitory PV-containing cells, or in feed-back distal dendritic inhibitory SOM/NPY-containing cells.

Enkephalin-containing interneurons are specialized to innervate other interneurons in the hippocampal CA1 region of the rat and guinea-pig

Enkephalins are known to have a profound effect on hippocampal inhibition, but the possible endogenous source of these neuropeptides, and their relationship to inhibitory interneurons is still to be identified. In the present study we analysed the morphological characteristics of met-enkephalin-immunoreactive cells in the CA1 region of the rat and guinea-pig hippocampus, their coexistence with other neuronal markers and their target selectivity at the light and electron microscopic levels. Several interneurons in all subfields of the hippocampus were found to be immunoreactive for met-enkephalin. In the guinea-pig, fibres arising from immunoreactive interneurons were seen to form a plexus in the stratum oriens/alveus border zone, and basket-like arrays of boutons on both enkephalin-immunoreactive and immunonegative cell bodies in all strata. Immunoreactive boutons always established symmetric synaptic contacts on somata and dendritic shafts. Enkephalin-immunoreactive cells co-localized GABA, vasoactive intestinal polypeptide and calretinin. Postembedding immunogold staining for GABA showed that all the analysed enkephalin-immunoreactive boutons contacted GABAergic postsynaptic structures. In double-immunostained sections, enkephalin-positive axons were seen to innervate calbindin D28k-, somatostatin-, calretinin- and vasoactive intestinal polypeptideimmunoreactive cells with multiple contacts. Based on these characteristics, enkephalin-containing cells in the hippocampus are classified as interneurons specialized to innervate other interneurons, and represent a subset of vasoactive intestinal polypeptide- and calretinin-containing cells. The striking match of ligand and receptor distribution in the case of enkephalin-mediated interneuronal communication suggests that this neuropeptide may play an important role in the synchronization and timing of inhibition involved in rhythmic network activities of the hippocampus.

Distinct interneuron types express m2 muscarinic receptor immunoreactivity on their dendrites or axon terminals in the hippocampus

In previous studies m2 muscarinic acetylcholine receptor-immunoreactive interneurons and various types of m2-positive axon terminals have been described in the hippocampal formation. The aim of the present study was to identify the types of interneurons expressing m2 receptor and to examine whether the somadendritic and axonal m2 immunostaining labels the same or distinct cell populations. In the CA1 subfield, neurons immunoreactive for m2 have horizontal dendrites, they are located at the stratum oriens/alveus border and have an axon that project to the dendritic region of pyramidal cells. In the CA3 subfield and the hilus, m2-positive neurons are multipolar and are scattered in all layers except stratum lacunosum-moleculare. In stratum pyramidale of the CA1 and CA3 regions, striking axon terminal staining for m2 was observed, surrounding the somata and axon initial segments of pyramidal cells in a basket-like manner. The co-localization of m2 with neurochemical markers and GABA was studied using the "mirror" technique and fluorescent double-immunostaining at the light microscopic level and with double-labelling using colloidal gold-conjugated antisera and immunoperoxidase reaction (diaminobenzidine) at the electron microscopic level. GABA was shown to be present in the somata of most m2-immunoreactive interneurons, as well as in the majority of m2-positive terminals in all layers. The calcium-binding protein parvalbumin was absent from practically all m2-immunoreactive cell bodies and dendrites. In contrast, many of the terminals synapsing on pyramidal cell somata and axon initial segments co-localized parvalbumin and m2, suggesting a differential distribution of m2 receptor immunoreactivity on the axonal and somadendritic membrane of parvalbumin-containing basket and axo-axonic cells. The co-existence of m2 receptors with the calcium-binding protein calbindin and the neuropeptides cholecystokinin and vasoactive intestinal polypeptide was rare throughout the hippocampal formation. Only calretinin and somatostatin showed an appreciable degree of co-localization with m2 (20% and 15%, respectively). Using retrograde tracing, some of the m2-positive cells in stratum oriens were shown to project to the medial septum, accouting for 38% of all projection neurons. The present results demonstrate that there is a differential distribution of m2 receptor immunoreactivity on the axonal vs the somadendritic membranes of distinct interneuron types and suggest that acetylcholine via m2 receptors may reduce GABA release presynaptically from the terminals of perisomatic inhibitory cells, while it may act to increase the activity of another class of interneuron, which innervates the dendritic region of pyramidal cells.

Synaptic communication among hippocampal interneurons: properties of spontaneous IPSCs in morphologically identified cells

The properties of spontaneous IPSCs (sIPSCs) recorded with whole-cell patch-clamp techniques were investigated in various anatomically identified hippocampal CA1 interneurons and were compared with those recorded in pyramidal cells. Neurons labeled with biocytin or neurobiotin were classified on the basis of their dendritic and axonal arborizations, leading to the identification of previously unknown interneuron types projecting to the dendritic region of pyramidal cells. In most interneurons, the average sIPSCs decayed slower than did those observed in pyramidal cells. The properties of sIPSCs were homogeneous within a given morphologically identified neuron type. Many interneurons had comparable somatic size, location, and dendritic arbor but displayed extremely different axonal projections paralleled by distinct sIPSC properties. Thus, physiological comparisons are only meaningful after the complete morphological identification of the recorded cells. The decay of sIPSCs matched for amplitudes and rise times could vary over 10-fold in a given interneuron, consistent with electrotonic filtering and possibly with different GABAA receptor subunit assemblies present at distinct synapses. Our findings demonstrate an extensive connectivity among hippocampal interneurons through GABAA synapses of various properties that may underlie complex network oscillations at different frequencies.

The expression pattern of somatostatin and calretinin by postnatal hippocampal interneurons is regulated by activity-dependent and -independent determinants

Hippocampal interneurons form distinct populations identified on the basis of their projection pattern and neurochemical characteristics, which includes the expression of specific neuropeptides and/or calcium-binding proteins. The neurochemical maturation of hippocampal interneurons is largely a postnatal event, and factors which govern this maturation are presently unknown. Using slice cultures, we have investigated the role of neuronal activity in regulating the expression of somatostatin and calretinin during the postnatal maturation of hippocampal interneurons. Blocking inhibitory activity with bicuculline, or excitatory activity with 6,7-dinitroquinoxaline-2,3-dione, for 14 days in slice cultures from seven-day-old rat increased and decreased, respectively, the number of somatostatin-immunoreactive neurons. Withdrawal of the blocking agents resulted in a reversal of the effects on somatostatin immunoreactivity. In addition, bicuculline slightly increased the number of calretinin-positive neurons, while 6,7-dinitroquinoxaline-2,3-dione exerted no effect. However, bicuculline and 6,7-dinitroquinoxaline-2,3-dione markedly increased and decreased, respectively, the number of calretinin-labelled axons. Despite activity-linked modifications of immunoreactivity levels, no change in the organotypic location of somatostatin-labelled neurons was observed, whatever the treatment. Double labelling studies demonstrated that somatostatin and calretinin were expressed by different neurons, even when the number of labelled cells was highly increased. These results show that the levels of expression of somatostatin and calretinin in maturing hippocampal interneurons are tuned to the endogenous balance of excitatory and inhibitory activity. In contrast, the neurochemical specificity of each subtype of interneurons does not depend upon variations in neuronal activity.

Synaptic effects of identified interneurons innervating both interneurons and pyramidal cells in the rat hippocampus

GABAergic interneurons sculpt the activity of principal cells and are themselves governed by GABAergic inputs. To determine directly some of the sources and mechanisms of this GABAergic innervation, we have used dual intracellular recordings with biocytin-filled microelectrodes and investigated synaptic interactions between pairs of interneurons in area CA1 of the adult rat hippocampus. Of four synaptically-coupled interneuron-to-interneuron cell pairs, three presynaptic cells were identified as basket cells, preferentially innervating somata and proximal dendrites of pyramidal cells, but one differing from the other two in the laminar distribution of its dendritic and axonal fields. The fourth presynaptic interneuron was located at the border between strata lacunosum moleculare and radiatum, with axon ramifying within stratum radiatum. Action potentials evoked in all four presynaptic interneurons were found to elicit fast hyperpolarizing inhibitory postsynaptic potentials (mean amplitude 0.35 +/- 0.10 mV at a membrane potential of -59 +/- 2.8 mV) in other simultaneously recorded interneurons (n=4). In addition, three of the presynaptic interneurons were also shown to produce similar postsynaptic responses in subsequently recorded pyramidal cells (n=4). Electron microscopic evaluation revealed one of the presynaptic basket cells to form 12 synaptic junctions with the perisomatic domain (seven somatic synapses and five synapses onto proximal dendritic shafts) of the postsynaptic interneuron in addition to innervating the same compartments of randomly-selected local pyramidal cells (50% somatic and 50% proximal dendritic synapses, n=12). In addition, light microscopic analysis also indicated autaptic self-innervation in basket (12 of 12) and bistratified cells (six of six). Electron microscopic investigation of one basket cell confirmed six autaptic junctions made by five of its boutons. Together, these data demonstrate that several distinct types of interneuron have divergent output to both principal cells and local interneurons of the same (basket cells) or different type. The fast synaptic effects, probably mediated by GABA in both postsynaptic interneurons and principal cells are similar. These additional sources of GABA identified here in the input to GABAergic cells could contribute to the differential temporal patterning of distinct GABAergic synaptic networks.