Neuronal gap junctions in the mouse main olfactory bulb: morphological analyses on transgenic mice

In the present study we analyzed the structural features of extraglomerular gap junction-forming processes in mouse olfactory bulb electron microscopically. This work complements a previous study in which we analyzed the structural features of neuronal gap junction-forming processes within the glomerulus itself. Furthermore we examined connexin 36 expressing cells in the mouse olfactory bulb by analyzing transgenic mice in which the connexin 36 coding sequence was replaced with histological reporters. In extraglomerular regions, the mitral/tufted cell somata, dendrites and axon hillocks made gap junctions and mixed synapses with interneuronal processes. These gap junctions and synapses were associated with various types of interneuronal processes, including a particular type of sheet-like or calyx-like process contacting the somata or large dendrites of mitral/tufted cells. In the olfactory bulbs of the transgenic mice, connexin 36 was expressed in mitral cells, tufted cells, presumed granule cells and periglomerular cells. Multiple immunofluorescent labelings further revealed that presumed interneurons expressing connexin 36 in the periglomerular region rarely expressed calbindin, calretinin or tyrosine hydroxylase and are likely to comprise a chemically uncharacterized class of neurons. Similarly, interneurons expressing connexin 36 in the granule cell layer were rarely positive for calretinin, which was expressed in numerous presumed granule cells in the mouse main olfactory bulb. In summary, these findings revealed that mitral/tufted cells make gap junctions with diverse types of neurons; in the glomeruli gap junction-forming interneuronal processes originated from some types of periglomerular cells but others from a hitherto uncharacterized neuron type(s), and in the extraglomerular region gap-junction forming processes originate mainly from a subset of cells within the granule cell layer.     

Immunocytochemical localization of GABA neurons and dopamine neurons in the rat main and accessory olfactory bulbs

Immunocytochemical localization of GABA neurons and dopamine neurons in the rat olfactory bulb was obtained with sheep antiserum to glutamate decarboxylase (GAD) and rabbit antiserum to tyrosine hydroxylase (TH). GAD-positive neurons include periglomerular cells, granule cells, superficial and deep short axon cells. TH-positive neurons represent periglomerular cells. Two-color immunocytochemistry shows that GABA and dopamine periglomerular cells are separate populations. The accessory olfactory bulb has rare dopamine cells and few superficial short axon cells. Radial gradients of GAD-immunostaining are evident in the main but not in the accessory olfactory bulb.     

Selective neuroinhibitory effects of taurine in slices of rat main olfactory bulb

Taurine is abundant in the main olfactory bulb, exceeding glutamate and GABA in concentration. In whole-cell patch-clamp recordings in rat olfactory bulb slices, taurine inhibited principal neurons, mitral and tufted cells. In these cells, taurine decreased the input resistance and caused a shift of the membrane potential toward the chloride equilibrium potential. The taurine actions were sustained under the blockade of transmitter release and were reversible and dose-dependent. At a concentration of 5 mM, typically used in this study, taurine showed 90% of its maximal effect. GABA(A) antagonists, bicuculline and picrotoxin, blocked the taurine actions, whereas the glycine receptor antagonist strychnine and GABA(B) antagonists, CGP 55845A and CGP 35348, were ineffective. These findings are consistent with taurine directly activating GABA(A) receptors and inducing chloride conductance. Taurine had no effect on periglomerular and granule interneurons. The subunit composition of GABA(A) receptors in these cells, differing from those in mitral and tufted cells, may account for taurine insensitivity of the interneurons. Taurine suppressed olfactory nerve-evoked monosynaptic responses of mitral and tufted cells while chloride conductance was blocked. This action was mimicked by the GABA(B) agonist baclofen and abolished by CGP 55845A; CGP 35348, which primarily blocks postsynaptic GABA(B) receptors, was ineffective. The taurine effect most likely was due to GABA(B) receptor-mediated inhibition of presynaptic glutamate release. Neither taurine nor baclofen affected responses of periglomerular cells. The lack of a baclofen effect implies that functional GABA(B) receptors are absent from olfactory nerve terminals that contact periglomerular cells. These results indicate that taurine decreases the excitability of mitral and tufted cells and their responses to olfactory nerve stimulation without influencing periglomerular and granule cells. Selective effects of taurine in the olfactory bulb may represent a physiologic mechanism that is involved in the inhibitory shaping of the activation pattern of principal neurons.     

Neuronal gap junctions in the rat main olfactory bulb,with special reference to intraglomerular gap junctions

The structural features of neuronal gap junction-forming processes in the rat olfactory bulb were analyzed electron microscopically. Gap junctions were present in glomeruli and extraglomerular regions. In extraglomerular regions, mitral/tufted cell somata, dendrites and axon hillock-initial segments made gap junctions and mixed synapses with interneuronal processes, some of which were confirmed to be GABA positive. In glomeruli gap junctions were encountered mainly between mitral/tufted cell dendrites and diverse types of processes; a small population of them were conclusively identified as periglomerular cell dendrites. Gap junction-forming processes frequently received synapses from olfactory nerve terminals, suggesting that they could be type 1 periglomerular cells. However, the majority were GABA negative or only faintly positive and none were tyrosine hydroxylase positive, indicating that they were different from previously reported type 1 periglomerular cells. Furthermore serial sectioning analyses revealed that the majority of those processes forming gap junctions with mitral/tufted dendrites were smooth cylindrical and had few presynaptic sites, indicating that they were different from previously described periglomerular cells. These findings revealed that mitral/tufted cells make gap junctions with diverse types of neurons; and some of these gap junction-forming processes originated from some types of periglomerular cells but others from hitherto uncharacterized neuron type(s).     

Heterogeneity of nitric oxide synthase-containing neurons in the mouse main olfactory bulb

The distribution and structural features of nitric oxide [corrected] synthase (NOS) containing intrinsic neurons were studied in the mouse main olfactory bulb (MOB). NOS positive neurons were heterogeneous, including some subpopulations of periglomerular cells, granule cells, interneurons in the external plexiform layer, superficial and deep short-axon cells and stellate cells. NOS positive periglomerular cells were frequently calretinin immunoreactive and, although rarely, calbindin positive. Importantly, some middle and external tufted cells were also confirmed to be NOS positive, some of which were also cholecystokinin (CCK) positive. Retrograde tracer experiments showed that some NOS positive tufted cells, which were also CCK positive, constitute the intrabulbar association system and the projection system to the olfactory tubercle. In addition, another particular subpopulation of NOS positive neurons with no or little CCK immunoreactivity appeared to project to areas covering the dorsal endopiriform nucleus, claustrum and insular cortex. Furthermore, diverse types of neurons other than mitral/tufted cells were also suggested to be projection neurons of the MOB. The present study revealed the diversity of NOS positive neurons in the mouse MOB and further revealed that they were different from those reported previously in the rat MOB in structural and chemical properties.     

Intraglomerular inhibition: signaling mechanisms of an olfactory microcircuit

Microcircuits composed of principal neuron and interneuron dendrites have an important role in shaping the representation of sensory information in the olfactory bulb. Here we establish the physiological features governing synaptic signaling in dendrodendritic microcircuits of olfactory bulb glomeruli. We show that dendritic gamma-aminobutyric acid (GABA) release from periglomerular neurons mediates inhibition of principal tufted cells, retrograde inhibition of sensory input and lateral signaling onto neighboring periglomerular cells. We find that L-type dendritic Ca(2+) spikes in periglomerular cells underlie dendrodendritic transmission by depolarizing periglomerular dendrites and activating P/Q type channels that trigger GABA release. Ca(2+) spikes in periglomerular cells are evoked by powerful excitatory inputs from a single principal cell, and glutamate release from the dendrites of single principal neurons activates a large ensemble of periglomerular cells.     

Ultrastructural identification of substance P immunoreactive neurons in the main olfactory bulb of the hamster

The neurons containing substance P immunoreactivity in the main olfactory bulb of the hamster are located in the glomerular layer. Their cell bodies lie in the periglomerular region and contain spherical or ovoid nuclei which lack invaginations of the nuclear membrane and tend to be positioned eccentrically in the cell body. Dendrites of these neurons extend throughout the periglomerular region and project into the glomerular neuropil. Within the glomerular neuropil, processes with substance P immunoreactivity contain agranular, spherical synaptic vesicles. Primary olfactory axons, and processes of uncertain origin which contain pleomorphic synaptic vesicles, form synaptic contacts with substance P immunoreactive processes.These ultrastructural findings confirm that the substance P immunoreactive neurons are external tufted cells. Their likely physiological properties are considered in relation to the synaptic organization in the glomerular layer of the main olfactory bulb and to the other putative neurotransmitters or neuromodulators located in this layer.     

Intraglomerular dendritic link connected by gap junctions and chemical synapses in the mouse main olfactory bulb: electron microscopic serial section analyses

Glomeruli of the main olfactory bulb are considered to serve as functional units in processing the olfactory information. Thus the fine tuning of the output level from each glomerulus is important to the information processing in the olfactory system. The interactions among neuronal elements in glomeruli might be one of main mechanisms regulating this output level. In the mouse main olfactory bulb neuronal connections via chemical synapses and gap junction in glomeruli were analyzed by the serial electron microscopical reconstruction. Gap junctions were encountered between diverse types of dendritic processes, between mitral/tufted cell dendrites, between mitral/tufted cell dendrites and periglomerular cell dendrites and between mitral/tufted cell dendrites and dendrites of some interneurons different from periglomerular cells. Then these morphological observations indicate that we must consider both direct coupling between mitral/tufted cells via gap junctions and indirect coupling between mitral/tufted cells via intervening interneuronal processes. One of gap junction-forming processes presynaptic in asymmetrical synapses was traced back to the soma of its origin located in the glomerular layer, which was thus identified as an external tufted cell. However, interestingly, it showed apparently different ultrastructural features from other external tufted cells located at the border between the glomerular and external plexiform layers; the latter resemble so-called mitral/tufted cells located in the external plexiform and mitral cell layers. Then external tufted cells were assumed to be heterogeneous in their ultrastructural features. We occasionally encountered several dendrites connected by gap junctions, which furthermore made chemical synapses with each other and with other surrounding processes. Thus both chemical synapses and gap junctions interconnect complexly various processes in the glomerulus, where the local circuit among intermingled olfactory nerves, mitral/tufted cell dendrites and interneuron dendrites is far more complex than previously schematized.     

Immunohistochemical identification of two types of dopamine neuron in the rat olfactory bulb as seen by serial sectioning

Summary Several neurons around the glomeruli in the rat olfactory bulb contain the enzyme tyrosine hydroxylase as revealed by light and electron microscopic immunohistochemistry. Electron microscopic analysis of serial sections revealed that both superficial tufted cells and small periglomerular neurons were labelled. These results give further support for the view that dopamine neurons in the rat olfactory bulb, from a neuroanatomical point of view, do not represent a homogeneous cell population. Furthermore, taken together with previous results in the literature our findings indicate that, from a transmitter histochemical point of view, neither tufted cells nor periglomerular neurons represent a homogeneous cell population.     

An experimental study of the projection of the amygdala to the accessory olfactory bulb and its relationship to the concept of a dual olfactory system

Summary The present anatomical findings point to the existence of a separate subdivision of the olfactory system whose connections are quite different from the principal part. The main olfactory bulb has olfactory afferents from the receptors of the general olfactory mucosa, while the accessory bulb has afferents from receptors in the vomeronasal organ. The main bulb projects to the olfactory tubercle and pyriform cortex, while the accessory bulb projects to the amygdala. In turn these areas are further related with the medial forebrain bundle in the case of the pyriform cortex and olfactory tubercle, and with the medial preoptic area and medial hypothalamus in the case of the amygdala. The main and accessory olfactory bulbs are further distinguished by their centrifugal connections, the main bulb receiving fibres from the olfactory tubercle passing through the lateral olfactory tract, and the accessory olfactory bulb receiving fibres from the amygdala through the stria terminalis. The centrifugals to the accessory olfactory bulb resemble those to the main bulb in that both appear to terminate upon granule cells, although further projections to the external plexiform layer or to the periglomerular region have not been demonstrated for the accessory bulb. By virtue of its neural connections the accessory olfactory system is ideally placed to mediate the effects of olfactory stimuli on reproduction.     

GABAergic basal forebrain afferents innervate selectively GABAergic targets in the main olfactory bulb

In this work we have analyzed the targets of the GABAergic afferents to the main olfactory bulb originating in the basal forebrain of the rat. We combined anterograde tracing of 10 kD biotinylated dextran amine (BDA) injected in the region of the horizontal limb of the diagonal band of Broca that projects to the main olfactory bulb, with immunocytochemical detection of GABA under electron microscopy or vesicular GABA transporter (vGABAt) under confocal fluorescent microscopy. GABAergic afferents were identified as double labeled BDA-GABA boutons. Their targets were identified by their ultrastructure and GABA content. We found that GABAergic afferents from the basal forebrain were distributed all over the bulbar lamination, but were more abundant in the glomerular and inframitral layers (i.e. internal plexiform layer and granule cell layer). The fibers had thick varicosities with abundant mitochondria and large perforated synaptic specializations. They contacted exclusively GABAergic cells, corresponding to type 1 periglomerular cells in the glomerular layer, and to granule cells in inframitral layers. This innervation will synchronize the bulbar inhibition and consequently the response of the principal cells to the olfactory input. The effect of the activation of this pathway will produce a disinhibition of the bulbar principal cells. This facilitation might occur at two separate levels: first in the terminal tufts of mitral and tufted cells via inhibition of type 1 periglomerular cells; second at the level of the firing of the principal cells via inhibition of granule cells. The GABAergic projection from the basal forebrain ends selectively on interneurons, specifically on type 1 periglomerular cells and granule cells, and is likely to control the activity of the olfactory bulb via disinhibition of principal cells. Possible similarities of this pathway with the septo-hippocampal loop are discussed.     

Autoradiographic analysis of [3H]dopamine and [3H]dopa uptake in the turtle olfactory bulb

Uptake and retention of exogenous tritiated dopamine and L-dopa was observed within turtle olfactory bulb slices. In the more superficial layers, periglomerular and superficial tufted cells, as well as their processes, and intraglomerular dendrites were recognized as labeled. Within the deeper part of the bulb, some labeled cells between the tanycytes, as well as nerve fibers and terminals within the granule cell layer, are reported. The results confirm the presence of specific intrinsic dopaminergic cells within the reptilian olfactory bulb.     

Vasoactive intestinal polypeptide-immunoreactive neurons in the main olfactory bulb of the hedgehog (Erinaceus europaeus)

Vasoactive intestinal polypeptide (VIP) immunoreactivity was localized by the indirect antibody enzyme method (PAP technique) in the main olfactory bulb of the hedgehog. Most VIP-immunoreactive cells were located in the glomerular layer and throughout the external plexiform layer. Fewer cells were observed in the granule cell layer. At the morphological level they exhibit the characteristics of periglomerular, external tufted, superficial short axon, horizontal and Van Gehuchten cells. It should be mentioned that another specific neuronal type was found in the inner third of the external plexiform layer, which is not described in other animals. These results revealed that a high number of intrinsic neuronal types of the olfactory bulb of the hedgehog display a strong VIP immunoreactivity.     

Distribution of enkephalin-like immunoreactivity in the rat main olfactory bulb

Enkephalin-like immunoreactivity was localized within the main olfactory bulb of the rat using immunohistochemical techniques. These studies utilized well characterized antisera directed to either leu⁵- or met⁵-enkephalin. Specificity was established by absorption of the antisera with either 10 μM synthetic leu⁵- or met⁵-enkephalin.Specific enkephalin-like immunoreactivity was observed within several different cell populations including (1) periglomerular cells, (2) granule cells and their processes within the external plexiform layer and (3) occasional short-axon (horizontal) cells within the granule and external plaxiform layers. The granule cell layer contained the greatest number of immunoreactive cells. Only a limited number of immunoreactive cells were found in both the periglomerular and granule cell layers, suggesting the enkephalin-containing neurons represent a sub-population within each layer.The absence of immunoreactive processes in the periventribular white matter, as well as the morphologies of immunoreactive bulbar neurons, indicates that enkephalin is found exclusively within intrinsic olfactory bulb neurons.     

Origin and migration of interneurons of the olfactory bulb in the musk shrew,Suncus murinus

The olfactory bulb of the musk shrew, Suncus murinus, is characterized by the presence of various interneurons. Our previous report (Kakuta et al., 2001) demonstrated that positive immunoreactions for calretinin were observed in periglomerular and perinidal cells in the glomerular layer, small ovoid neurons in the external plexiform layer, and granule cells in the granule cell layer of the olfactory bulb in the musk shrew aged 1 to 5 weeks, in addition to calretinin-immunoreactive bipolar cells distributed in the anterior subependymal layer and in each layer of the olfactory bulb. To examine the origin and migration of interneurons of the olfactory bulb, we labeled generated cells by injecting 28-day-old musk shrews with 5-bromo-2'-deoxyuridine (BrdU), and detected the labeled progeny cells that survived after several intervals. BrdU-labeled cells originated in the subependymal layer around the anterior horn of the lateral ventricle, and rostrally migrated in the subependymal layer from the anterior wall of the lateral ventricle into the center of the olfactory bulb, where they radially migrated into the granule cell layer, external plexiform layer, and glomerular layer. It took 2 days to migrate rostrally in the subependymal layer from the anterior lateral ventricle to the center of the olfactory bulb, and 2 to 6 days to migrate radially from the bulbar subependymal layer into the three layers mentioned. The rate of rostralward migration of the labeled cells was estimated to be 38 microm/h, while that of radial migration, 7 to 25 microm/h. The present BrdU-labeling study, together with our previous immunohistochemical study (Kakuta et al., 2001), indicates that anterior subependymal cells differentiate into granule cells in the granule cell layer, into Van Gehuchten cells in the external plexiform layer, and into periglomerular and perinidal cells in the glomerular layer of the olfactory bulb in the musk shrew.     

Immunocytochemical localization of the GABAB2 receptor subunit in the glomeruli of the mouse main olfactory bulb

The olfactory input to the brain is carried out by olfactory nerve axons that terminate in the olfactory bulb glomeruli and make synapses onto dendrites of glutamatergic projection neurons, mitral and tufted cells, and GABAergic interneurons, periglomerular cells. The dendrites are reciprocally connected through asymmetric synapses of mitral/tufted cells with periglomerular cells and symmetric synapses of the opposite direction. Transmission at the first synapse in the olfactory pathway is regulated presynaptically, and this regulation is mediated, in part, by metabotropic GABAB receptors that, when activated, inhibit transmitter release from the olfactory nerve. Functional GABAB receptors are heterodimers composed of the GABAB1 and GABAB2 subunits. Studies using double immunofluorescence have shown colocalization of both subunits in the glomerular neuropil, and ultrastructural studies have localized GABAB1 to extrasynaptic, synaptic, and perisynaptic sites on the plasma membrane of olfactory nerve terminals. We studied the subcellular localization of GABAB2 in the mouse olfactory glomeruli using a subunit-specific antibody and preembedding immunogold labeling. Immunoreactivity for GABAB2 was associated with symmetric dendrodendritic synapses of periglomerular cells with mitral/tufted cells and was localized to the extrasynaptic plasma membrane of presynaptic dendrites, and extrasynaptic, synaptic, and perisynaptic sites on the plasma membrane of postsynaptic dendrites. The results suggest that postsynaptic, and perhaps presynaptic, GABAB receptors may be expressed at GABAergic synapses between dendrites of periglomerular interneurons and projection neurons. Immunolabeling was observed at junctions of the olfactory nerve with mitral/tufted cell dendrites, providing ultrastructural evidence for the expression of the GABAB2 subunit at the primary olfactory synapse.     

NADPH-diaphorase and cytosolic urea cycle enzymes in the rat accessory olfactory bulb

The nitric oxide cycle consists of nitric oxide synthase, argininosuccinate synthetase and argininosuccinate lyase to form nitric oxide. We have examined the colocalization of nitric oxide synthase and the cytosolic urea cycle enzymes (argininosuccinate synthetase, argininosuccinate lyase and arginase) in the accessory olfactory bulb of the rat by using a double labeling procedure combining reduced-nicotinamide-adenine-dinucleotide-phosphate-diaphorase (NADPH-d) reaction with fluorescent immunocytochemistry. Each glomerulus showed a different NADPH-d activity, and those with the strongest NADPH-d activities were assembled in the caudomedial part of the accessory olfactory bulb. Argininosuccinate synthetase-like immunoreactive glomeruli were distributed in the caudomedial part of the accessory olfactory bulb, and most of them were also strongly NADPH-d positive. The mitral or tufted cells were argininosuccinate synthetase-, argininosuccinate lyase- and arginase-like immunoreactive, but were not NADPH-d positive. The granule cells were NADPH-d positive or argininosuccinate lyase-like immunoreactive, but were not argininosuccinate synthetase- or arginase-like immunoreactive. Some granule cells were both NADPH-d positive and argininosuccinate lyase-like immunoreactive. The results indicate the heterogeneity of glomeruli of the accessory olfactory bulb with respect to the distribution of these enzymes. The granule cells have nitric oxide synthase and argininosuccinate lyase, and thus may efficiently produce nitric oxide.     

Electrophysiology of interneurons in the glomerular layer of the rat olfactory bulb

In the mammalian olfactory bulb, glomeruli are surrounded by a heterogeneous population of interneurons called juxtaglomerular neurons. As they receive direct input from olfactory receptor neurons and connect with mitral cells, they are involved in the initial stages of olfactory information processing, but little is known about their detailed physiological properties. Using whole cell patch-clamp techniques, we recorded from juxtaglomerular neurons in rat olfactory bulb slices. Based on their response to depolarizing pulses, juxtaglomerular neurons could be divided into two physiological classes: bursting and standard firing. When depolarized, the standard firing neurons exhibited a range of responses: accommodating, nonaccommodating, irregular firing, and delayed to firing patterns of action potentials. Although the firing pattern was not rigorously predictive of a particular neuronal morphology, most short axon cells fired accommodating trains of action potentials, while most delayed to firing cells were external tufted cells. In contrast to the standard firing neurons, bursting neurons produced a calcium-channel-dependent low-threshold spike when depolarized either by current injection or by spontaneous or evoked postsynaptic potentials. Bursting neurons also could oscillate spontaneously. Most bursting cells were either periglomerular cells or external tufted cells. Based on their mode of firing and placement in the bulb circuit, these bursting cells are well situated to drive synchronous oscillations in the olfactory bulb.     

Vasoactive intestinal polypeptide-containing elements in the olfactory bulb of the hedgehog (Erinaceus europaeus)

The distribution of vasoactive intestinal polypeptide (VIP)-immunopositive elements was analyzed in the olfactory bulb (OB) of the Western European hedgehog (Erinaceus europaeus) under light and electron microscopy. The immunoreactivity appeared in an abundant population of periglomerular cells of the glomerular layer, in interneurons of the external plexiform layer, and in a restricted group of deep short-axon cells of the internal plexiform layer, the granule cell layer and the white matter. In the glomerular layer, VIP-containing periglomerular cells constituted a population of non-GABAergic neurons and did not receive synapses from olfactory axons. In the EPL, VIP-immunoreactivity appeared in a morphologically heterogeneous population of GABAergic interneurons, most of them identified as satellite cells and Van Gehuchten cells. These interneurons exerted an abundant and selective innervation of the somata, primary and secondary dendrites of the principal mitral and tufted cells, but did not contact granule cells. Perisomatic innervation of the principal cells followed two different patterns. The first included 'normal' basket-like arrangements of VIP-containing varicosities surrounding the somata of mitral and tufted cells. In the second, a set of satellite cells gave rise to short dendritic shafts that embraced the somata of principal cells in an 'exuberant' basket-like arrangement. These two morphological patterns of perisomatic innervation of principal cells were correlated with a neurochemical specificity of the target. In this sense, the 'exuberant' basket-like structures were always found surrounding a subpopulation of principal cells that did not contain the calcium-binding protein parvalbumin (PV). By contrast, they were never found surrounding the subpopulation of PV-containing principal cells, which only showed 'normal' basket-like structures. This study provides new data on the connectivity and neurochemical features of the hedgehog olfactory bulb and suggests that the olfactory circuits in this species are more complex than those described in other mammals.     

Postnatal changes in expression of vesicular glutamate transporters in the main olfactory bulb of the rat

Olfactory information is initially processed through intricate synaptic interactions between glutamatergic projection neurons and GABAergic interneurons in the olfactory bulb. Although bulbar neurons and networks have been reported to develop even postnatally, much is yet unknown about the glutamatergic neuron development. To address this issue, we studied the postnatal ontogeny of vesicular glutamate transporters (VGLUT1 and VGLUT2) in the main olfactory bulb of rats, using in situ hybridization, immunohistochemistry, and their combination. In situ hybridization data showed that VGLUT1 mRNA is intensely expressed in differentiating mitral cells and smaller cells of the mitral cell layer (MCL) on postnatal day 1 (P1), and also at lower levels in small- and medium-sized cells, presumably tufted cell populations, of the external plexiform layer (EPL) from P5 onward. VGLUT2 mRNA was expressed in many MCL cell populations on P1, also in small- and medium-sized cells of the EPL at almost the same level as MCL cells between P5 and P7, and became apparently less intense in the MCL than in the EPL from P10 onward. The expression, unlike VGLUT1 mRNA, was also found in small-sized cells of the interglomerular region. In partial agreement with these data, immunohistochemical analyses demonstrated that subsets of mitral and EPL cells are stained for VGLUT1 or VGLUT2, with the former cells coexpressing both subtypes until P5. Moreover, a combined fluorescence in situ hybridization–immunohistochemical dual labeling of the P10 bulb revealed that neither VGLUT1 nor VGLUT2 mRNA is expressed in GABAergic or dopaminergic periglomerular cells, implying their expression in other periglomerular cell subclasses, external tufted cells and/or short-axon cells. Thus, the present study suggests that early in the postnatal development distinct glutamatergic bulbar neurons of rats express spatiotemporally either or both of the two VGLUT subtypes as a specific vesicular transport system, specifically contributing to glutamate-mediated neurobiological events.