The Journal of Neurophysiology Vol. 80 No. 1 July 1998, pp. 344-349
Copyright ©1998 by the American Physiological Society

Inhibitory Synapses Among Interneurons in the Glomerular Layer of Rat and Frog Olfactory Bulbs

Michelino Puopolo¹ and Ottorino Belluzzi²

¹ Dipartimento di Biologia, Sezione di Fisiologia Generale, Università di Ferrara, 44100 Ferrara; and ² Dipartimento di Biologia, Sezione di Fisiologia, Università di Modena, 41100 Modena, Italy

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Puopolo, Michelino and Ottorino Belluzzi. Inhibitory synapses among interneurons in the glomerular layer of rat and frog olfactory bulbs. J. Neurophysiol. 80: 344-349, 1998. The patch-clamp technique was applied to periglomerular (PG) cells from slices of frog and rat olfactory bulbs to characterize whole cell and single-channel properties of inhibitory synaptic currents. Spontaneous and electrically driven bicuculline-sensitive synaptic events were recorded under ionic conditions that excluded any possible interference of excitatory synapses. The peak amplitude distribution of spontaneous events could be fitted by several Gaussians having the same interpeak distance. Spontaneous currents reversed polarity at the chloride equilibrium potential and were suppressed by 2 mM Co²⁺; the decay phase could be fitted with a single exponential having a time constant of ~10 ms at 0 mV. Bicuculline-sensitive monosynaptic responses could be evoked in PG cells by electrical stimulations delivered at the distance of several glomeruli within the glomerular layer. Finally, in excised outside-out patches, single-channel analysis revealed the presence of typical -aminobutyric acid-A receptor channels, with a single-channel conductance of 28 pS in symmetrical chloride and mean open times of 3-4 ms. The simplest explanation of these data, effectively supported by pristine anatomic findings, is that there could be inhibitory contacts among interneurons in the glomerular layer.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Periglomerular (PG) cells are small, inhibitory interneurons, predominantly GABAergic, which likely play a key role in the processing of olfactory information. They are located in the glomerular layer of the olfactory bulb (OB) at the front end of the bulbar circuitry; they receive excitatory inputs from the olfactory nerve (ON) and from the apical dendrites of projection neurons [mitral and tufted (MT) cells] and send their terminals to the apical dendrite of MT cells, which are inhibited. Actually, even if this is the scheme adopted in most models of the OB, several experimental findings suggest that the situation could be more complex. In an electron microscopy (EM) study Pinching and Powell (1971b) provided evidence that, in the neuropil of the PG region, axons originating from PG cells also establish numerous connections with dendrites, somata, and the initial segment of both PG and short-axon cells, a different, more rare type of inhibitory interneuron that can be found in the glomerular region (Pinching and Powell 1971a). The observation has been confirmed in another EM study by White (1973), which observes these contacts within the glomerulus. Subsequently, histochemical evidence was provided for the presence of -aminobutyric acid-A (GABA_A) receptors in the PG cells (Bowery et al. 1987; Laurie et al. 1992); finally, the presence of GABAergic receptors (A-type) on PG cells was demonstrated by direct application of GABA (Bufler et al. 1992).

In previous works we observed spontaneous GABAergic synaptic events in PG cells, in both frogs (Bardoni et al. 1995) and rats (Puopolo and Belluzzi 1996). We expand these observations, providing the first evidence of the existence of functionally active GABAergic synapses on PG cells, and we suggest that this could be due to synaptic interconnections among interneurons.

	METHODS

Abstract Introduction Methods Results Discussion References

Rats

Rats (Wistar strain) from 8- to 20-days old were employed. The OB was sliced in the horizontal plane at a thickness of 150-170 mm. Slices were incubated for 30 min at 37°C and then were mounted in the observation bath (1 ml) and transferred onto a microscope stage with fixed table and DIC optics. Recordings were done at 30 ± 0.2°C (means ± SE); slices were continually perfused at the rate of 2 ml/min.

The standard bicarbonate-buffered saline used had the following composition (in mM): 125 NaCl, 2.5 KCl, 26 NaHCO₃, 1.25 NaH₂PO₄, 2 CaCl₂, 1 MgCl₂, and 25 glucose; the solution was continually bubbled with 95% O₂-5% CO₂. The osmolarity was adjusted at 320 mosM with glucose. Two different intracellular solutions were used. The first (symmetrical chloride, E_Cl = 0 mV) had the following composition (in mM): 120 KCl, 10 NaCl, 2 MgCl₂, 0.5 CaCl₂, 5 ethylene glycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA), 10 N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES), 2 Mg-ATP, and 20 glucose. The second, used to study the synaptic events at 0 mV, was modified so that E_Cl was 70 mV and had the K⁺ ions substituted by Cs⁺; its composition was 145 Cs⁺ acetate, 4 MgCl₂, 0.5 CaCl₂, 5 EGTA, 10 HEPES, 2 Mg-ATP, and 20 glucose. In both, pH was set to 7.3 and osmolarity was adjusted to 310 mosM with glucose.

Frogs

Details of the method are given in a previous paper (Bardoni et al. 1995). The OB was extracted from adult frogs (Rana æsculenta, 40-60 g) and sliced (150-200 µm) in the coronal plane. The standard saline used had the following compositions (in mM): 125 NaCl, 2.5 KCl, 2 CaCl₂, 10 HEPES, and 5 glucose. The pH and osmolarity were adjusted to 7.8 (Howell et al. 1970; Reeves 1969) and 240 mosM, respectively. The pipettes were backfilled with solutions of the following composition (in mM): 1) symmetrical chloride (110 KCl, 0.5 CaCl₂, 5 K-EGTA, 10 K-HEPES, 2 Mg-ATP, and 8 glucose) and 2) E_Cl = 70 mV (107 K-gluconate, 7 KCl, 0.5 CaCl₂, 5 K-EGTA, 5 K-HEPES, 2 MgSO₄, and 8 glucose). pH was set to 7.8 with KOH and osmolarity was adjusted to 235 mosM with glucose.

Drugs, obtained from Sigma (St. Louis, MO) and from Tocris Neuramin (Bristol, UK), were dissolved in standard bath solution and perfused the entire preparation. Effects of drugs were observed within 1-2 min after changing bath solution.

ELECTROPHYSIOLOGY. The tight-seal, whole cell recording technique in thin slices was used (Edwards et al. 1989). The cells were not cleaned before patching. Seal resistances were always >3 G. The pipettes (borosilicate glass with internal filament, 1.5 OD, 0.86 ID; Hilgenberg), made with a Sutter P-97 puller, had tip resistances of 4-5 M (5-8 M in frog) when filled with a standard or Cs⁺ internal solutions and usually were not fire polished. Cells were accessed under direct visual control using a ×40 water immersion objective. Membrane currents were recorded with a patch amplifier (Axopatch 1D, Axon Instruments). The series resistance in these experiments was ~15 M (19 M in frog); series resistance compensation of 60-70% was routinely used. Leakage current was digitally subtracted from the recorded current tracings following the P4 protocol (Armstrong and Bezanilla 1974). Corrections for liquid-junction potential (Neher 1992) were applied a posteriori.

In the frog electrical stimulation was achieved with suction electrodes applied to the ON. The stimulus width was 1-1.5 ms and its amplitude ranged from 2 to 9 V. In the rat stimulation was obtained with Teflon-coated silver (Ag) wires (100-µm diam).

Stimulation and data acquisition were controlled by a PC-386 computer using a 12 bit A/D-D/A converter (TL-1-125, Axon Instruments). Before acquisition the signals were filtered at 3 kHz by a low-pass four-pole Bessel filter and digitized with sample times ranging from 50 µs to 1 ms. Data are given as means ± SE.

	RESULTS

Abstract Introduction Methods Results Discussion References

Recordings were obtained from 50 neurons in in vitro rat OB slice preparations. Neurons in the glomerular layer were identified as PG cells on the basis of the location around the glomerular neuropile, value of membrane capacity (6.9 ± 2 pF, n = 50), and single action potential in response to the injection of depolarizing current pulses. We rule out any possibility of misidentifying PG cells and external tufted cells for the following two reasons: 1) the two populations had quite distinct membrane capacity (7.2 ± 0.2 pF, n = 109 for PG cells; 20.3 ± 0.66 pF, n = 119 for external tufted cells) and 2) they responded in entirely different ways to direct injection of depolarizing currents; PG cells responded with a single action potential (Fig. 1C), and external tufted cells responded with trains of spikes (Fig. 1D). The identification was confirmed by Lucifer yellow injection in ~10 cells in this study (Fig. 1, A and B) and in ~20 in a previous study (Puopolo and Belluzzi 1996).

View larger version (35K): [in this window] [in a new window]   FIG. 1. Morphology and basic properties of a rat periglomerular cell and of an external tufted cell. A: Camera lucida reconstruction of Lucifer yellow-filled periglomerular cells. B: external tufted cell. Calibration bars for A and B = 25 mm. C: voltage responses of a typical PG cell (7.1 pF) to injected currents (from 65 to 45 pA in steps of 20 pA) from a resting membrane potential of 80 mV. D: voltage responses of an external tufted cell (17 pF) to injected currents from a membrane potential of 80 mV. Represented are the responses to five current pulses from 100 to +10 pA in steps of 30 pA and to a pulse of 140 pA, the first giving rise to action potentials. C and D are represented in the same scale.

For a more extensive description of the fundamental electrophysiological properties, we refer to previous papers (Bardoni et al. 1995; Puopolo and Belluzzi 1996). Basically, in rat PG cells the resting membrane potential was 52 ± 7 mV, input resistance was 750 ± 63 M, and membrane charging could be fitted by a single exponential function, suggesting that PG cells usually behave as a single, electrotonically compact compartment (Johnston 1981; Rall 1977).

In the rat both spontaneous kynurenate-sensitive and bicuculline-sensitive postsynaptic events were observed in 20% of the PG cells under current- or voltage-clamp conditions, whereas in the frog only GABAergic events were observed. For brevity, we shall refer to GABAergic synaptic events as inhibitory postsynaptic currents (IPSCs; see DISCUSSION). In spite of the variety of neurotransmitters described in this region (Shipley and Ennis 1996; Trombley and Shepherd 1993), we have never observed spontaneous events that could not be blocked by either bicuculline or kynurenate.

Spontaneous excitatory postsynaptic currents (EPSCs) were blocked by 1 mM kynurenate (Fig. 2A), were insensitive to bicuculline, reversed ~0 mV, and had a reversal potential independent of E_Cl. An analysis of the amplitudes of EPSCs is shown in Fig. 2B for a 5-min recording. The amplitude distribution histogram could be fitted by three or four equally spaced Gaussian curves with the mean centered at multiples of 18 pA at 70 mV. The EPSCs occurred with frequencies ranging from 2 to 4 Hz and decayed according to a single exponential with a time constant of 5 ms. These events, provisionally ascribed to spontaneous activity from the ON terminals (Bardoni et al. 1996; Berkowicz et al. 1994), were not analyzed further in the current study.

View larger version (23K): [in this window] [in a new window]   FIG. 2. Excitatory postsynaptic currents (EPSCs) in rat. A: voltage-clamp recordings of spontaneous activity in rat PG cell in the presence of 10 µM bicuculline. Recordings in symmetrical chloride, holding potential 70 mV. B: histogram of the amplitude distribution of the EPSCs in control conditions (5 min). The histogram was fitted with 3 Gaussian curves with peaks at multiples of 18 pA. Bin width 5 pA, 560 events.

In ~30% of the PG cells studied in the rat (50% in frog), spontaneous GABAergic postsynaptic currents were observed under voltage-clamp conditions. Spontaneous IPSCs were first studied in the frog (Fig. 3). In some cases the basic properties of these events were analyzed at the holding potential of 70 mV, with solutions having an E_Cl of 0 mV (Fig. 3A). In most of the experiments, however, we adopted the condition shown in Fig. 3C (i.e., E_Cl = 70 mV, holding potential 0 mV) corresponding to the EPSC's equilibrium potential, even if in the frog excitatory spontaneous events were never observed. Spontaneous IPSCs could be reversibly blocked by the GABA_A antagonist bicuculline (10 mM; Fig. 3, A and C), were insensitive to kynurenate 1 mM and reversed at the chloride equilibrium potential. An example of amplitude distribution is shown in Fig. 3B for a 5-min recording. The amplitude distribution histogram could be fitted by three equally spaced Gaussian curves with the mean centered at multiples of ~50 pA, and the IPSCs decayed following a single exponential with time constants of 38 ± 2 ms.

View larger version (35K): [in this window] [in a new window]   FIG. 3. GABAergic spontaneous activity in frog. A: voltage-clamp recordings of spontaneous activity in frog PG cell in the indicated experimental conditions. Recordings in symmetrical chloride, holding potential 70 mV. B: histogram of the amplitude distribution of the IPSCs in control conditions (5 min). The histogram was fitted with 3 Gaussian curves (interpeak distance 48 pA). Bin width 10 pA, 342 events. C: voltage-clamp recordings of spontaneous inhibitory postsynaptic currents (IPSCs) in frog PG cell. Recordings in ECl 70 mV, holding potential 0 mV. D: fit with a single exponential function of the decay of one of the IPSCs shown in C; t = 38.5 ms.

The second of the two protocols used in the frog (E_Cl = 70 mV, V_hold 0 mV) was chosen to study IPSCs in the rat because the holding potential set at the equilibrium potential of the glutamatergic synapses excluded any possible interference caused by spontaneous activity of excitatory terminals.

In the rat at 30°C, spontaneous IPSCs occurred at a frequency ranging from 1 to 4 Hz and their decay phase could be fitted by a single exponential with time constant of 11 ± 2 ms (Fig. 4C). The amplitude distribution histogram could be fitted by two, occasionally three, equally spaced Gaussian curves with the mean centered at multiples of 17 pA at 0 mV (Fig. 4B). The spontaneous activity was Ca²⁺ dependent and could be reversibly blocked by Co²⁺ 2 mM or by elimination of Ca²⁺ ions in the extracellular solution. The spontaneous IPSCs were not sustained by autapses because their frequency was not increased by depolarization (not shown).

View larger version (27K): [in this window] [in a new window]   FIG. 4. GABAergic spontaneous activity in rat. A: voltage-clamp recordings of spontaneous activity in rat PG cell in the indicated experimental conditions. Recordings in ECl 70 mV, holding potential 0 mV. B: histogram of the amplitude distribution of the IPSCs in control conditions (5 min). The histogram was fitted with two Gaussian curves (interpeak distance 17 pA). Bin width 4 pA, 465 events. C: fit with a single exponential function of the decay of 1 of the IPSCs shown in A; t = 10.8 ms.

To characterize GABA-activated currents in greater detail, outside-out patches from PG cells were obtained (n = 6). Figure 5A shows an example of single-channel opening activated by the addition of 10 µM GABA to the solution. GABA-activated single-channel currents revealed amplitudes of 2.9 pA at ±100 mV with the use of symmetrical chloride. The conductance obtained from the current-voltage relationship was 28 ± 1 pS (5 patches, Fig. 5B), and examination of the gating properties revealed open times of 3-4 ms (3 patches; Fig. 5C).

View larger version (30K): [in this window] [in a new window]   FIG. 5. Single-channel properties of -aminobutyric acid (GABA) receptor channels. A: strips of single-channel recordings at 0 mV and filtered at 1 kHz; outside-out patches. Single channel activity was started by applying 10 µM GABA. Holding potential +100 mV, ECl 70 mV. B: amplitude of single-channel currents as a function of membrane potential. Each point represents the mean of the Gaussian describing the current amplitude. C: normalized open time distribution of GABA-activated, single-channel currents. Bin size 1 ms, 257 openings.

In addition to the spontaneous events, electrically driven IPSCs could also be observed in four cases in response to electrical shocks applied in the glomerular layer with a couple of Teflon-coated Ag wires. The tracings shown in Fig. 6 represent pure GABAergic responses evoked in PG cells by stimuli delivered through an electrode placed in the middle of the glomerular layer. A test shock followed a conditioning shock (0.5-ms duration by a 2-s interval) to test for the presence of paired-pulse depression (see next paragraph and DISCUSSION). The recordings in the example shown were taken from a PG cell located rostrally to the point of stimulation at a distance of ~0.5 mm, corresponding to about three diameters of an average glomerulus in rat.

View larger version (8K): [in this window] [in a new window]   FIG. 6. Evoked synaptic activity recorded in rat PG cells under voltage-clamp conditions following stimulation in the glomerular layer. Paired pulses, 0.5-ms duration and separated by an interval of 2 s, were delivered in each experimental condition. Recordings in symmetrical chloride, Vhold 70 mV.

These responses were completely and reversibly blocked by bicuculline (10 µM) and were insensitive to kynurenate. The delay between the stimulus artifact and the rising phase of the synaptic response was 3-4 ms, suggesting a monosynaptic response. The evidence that the presynaptic element involved was not an ON terminal is both pharmacological (sensitivity to bicuculline) and electrophysiological; no sign of paired-pulse depression could be observed with paired pulses, a peculiar feature of the ON stimulation (Gusel'nikova et al. 1973; Freeman 1974; personal observations).

	DISCUSSION

Abstract Introduction Methods Results Discussion References

This study provides the first physiological evidence that both spontaneous and electrically driven GABAergic synaptic input can be observed in PG cells.

We are confident that the cells studied in this work were actually PG cells and did not belong to one of the other neuronal types described in the glomerular layer, namely, external tufted cells, short-axon (SA) cells, or astrocytes. In fact, even if external tufted cells may sometimes be mistaken for PG cells at DIC microscopy, the error of identification is immediately revealed by the measure of the membrane capacity and by the excitability profile, as explained above and shown in Fig. 1. Some of the classification criteria adopted (location around the glomeruli, membrane capacity, and single action potential) cannot entirely rule out the possibility that some of the cells studied in the rat were SA cells. However, it should be observed that SA cells were rare enough to escape the careful investigation of Golgi (1874), Ramon y Cajal (1890), and Blanes (1898) in the same preparation used for most of these experiments. Furthermore, in the frog where, as we have shown, GABAergic synapses are present on PG cells, SA cells are not present. Finally, SAs are slightly larger than PG cells (Getchell and Shepherd 1975; Pinching and Powell 1971a) and could belong to a relatively well-identifiable small group of cells averaging 12-13 pF that are not included in this study.

From our experiment it cannot be decided with certainty if these GABAergic terminals originate from other interneurons in the glomerular layer or from centrifugal fibers, but it should be observed that, although there is a good histological evidence sustaining the first hypothesis, there are no findings supporting the second. In fact, ultrastructural studies from serial slices have clearly established that axons originating from PG cells impinge onto the proximal dendrites, soma, and initial segment of other PG and SA cells (Pinching and Powell 1971b; White 1973) and that these synaptic contacts had the typical hallmarks of inhibitory synapses (flattened vesicles and symmetrical thickening). On the other hand, GABA was found to be involved in centrifugal innervation of the OB both in the frog (Kratskin et al. 1992) and in mammals (Zaborszky et al. 1986). However, although there is good anatomic and functional evidence that GABAergic centrifugal fibers terminate in the granule cell layer (Kratskin et al. 1992; Kunze et al. 1992), the centrifugal projections to the glomerular layer appear to be mostly cholinergic (Le Jeune and Jourdan 1993; Macrides and Davis 1983; Pinching and Powell 1972) and serotoninergic (McLean and Shipley 1987); to our knowledge there is no evidence to date of GABAergic centrifugal fibers impinging onto interneurons in the glomerular layer of rat or frog (for a recent review see also Duchamp-Viret and Duchamp 1997).

Several hypotheses can be advanced to attribute a physiological meaning to the existence of GABAergic synapses among interneurons in the glomerular layer. Assuming that these synapses are inhibitory, the most obvious hypothesis is that this arrangement is designed to trim the PG cell excitation during prolonged depolarization. Furthermore, spontaneous GABAergic input would also extend the potential range over which the PG cell activity can be modulated. Finally, in the rat hippocampus and neocortex, a mechanism for fast oscillation is sustained by a network of inhibitory interneurons connected by GABAergic synapses and driven by activation of glutamatergic synapses (Whittington et al. 1995). Interestingly, olfactory information has also been proposed to be encoded by oscillating neural assemblies (Laurent and Davidowitz 1994; Wehr and Laurent 1996).

However, it should be observed that GABAergic synapses are not necessarily inhibitory. In fact it has been shown that GABA in the central nervous system can be an excitatory neurotransmitter during the early postnatal period, attaining the inhibitory transmitter features only 8 days after birth (Cherubini et al. 1991). In this context, more directly related to the preparation that we were using, is the observation that reversed chloride gradients have actually been measured by using cytochemical methods in a subpopulation of PG cells (Siklós et al. 1995), which would render the opening of chloride channels excitatory rather than inhibitory in these cells. Unfortunately, it is difficult with conventional electrophysiological methods to see the true sign of GABAergic synapses in PG cells because their dimension makes these cells too fragile to use the classical intracellular microelectodes, and the whole cell patch-clamp method does not allow maintaining the original intracellular chloride concentration for a time sufficient for measurement.

The observation of GABAergic synapses in the PG cells of the frog could be of some significance because during evolution these interneurons appear for the first time in amphibia, where they are rare, becoming increasingly numerous in reptilia and mammalians, where they completely surround the glomeruli. Whatever the role of PG cells will prove to be, it is of some interest that in amphibia the GABAergic connections are already present, as if they were essential for PG cell function rather than constituting a refinement of their capability.

As for the role of these connections, and more in general for the role of PG cells, only speculations are possible at the moment. Our guess is that, in general terms, PG cells endow the bulbar circuitry of terrestrial animals with some additional skill, and it is tempting to think that this new capacity could be related to the distinct requirements imposed by the terrestrial odors, which are hydrophobic and highly volatile, in contrast to aquatic odors, which are hydrophilic and more persistent.

However, we wonder if trying to define the role of PG cells in a holistic way, assuming that PG cells are a homogeneous population, does not risk to be an ill-posed issue. Granule cells, the other population of interneurons, are comparatively much more homogeneous, and their role in lateral inhibition has been clearly defined (Yokoi et al. 1995). The situation is very different for PG cells; they have distinct histochemical properties (Kosaka et al. 1997), different excitability profiles because of distinct composition in voltage-dependent channels (Puopolo and Belluzzi 1998), and, probably, different intracellular concentration of chloride ions (Siklós et al. 1995), with the obvious consequence that this could have on GABAergic signals. We conclude that an essential prerequisite for any reasonable guess about the role of PG cells is a much better understanding of their different types, properties, and synaptic connections, for which a considerable effort must still be spent.

	ACKNOWLEDGEMENTS

  We are grateful to P. Duchamp-Viret, A. Duchamp, and I. Kratskin for useful comments on this manuscript.

  This laboratory work was carried out with a personal grant to Prof. Virgilio Perri.

	FOOTNOTES

  Address for reprint requests: O. Belluzzi, Dipartimento di Scienze Biomediche e Terapie Avanzate-Sezione di Fisiologia Umana, Via Fossato di Mortara 17/19, 44100 Ferrara, Italy.

  Received 8 January 1998; accepted in final form 25 March 1998.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• ARMSTRONG, C. M., BEZANILLA, F. Charge movement associated with the opening and closing of the activation gates of Na channels. J. Gen. Physiol. 63: 533-552, 1974.[Abstract/Free Full Text]
• BARDONI, R., MAGHERINI, P. C., BELLUZZI, O. Sodium current in periglomerular cells of frog olfactory bulb in vitro. Brain Res. 703: 19-25, 1995.[Medline]
• BARDONI, R., MAGHERINI, P. C., BELLUZZI, O. Excitatory synapses in the glomerular triad of frog olfactory bulb in vitro. Neuroreport 7: 1851-1855, 1996.[Medline]
• BERKOWICZ, D. A., TROMBLEY, P. Q., SHEPHERD, G. M. Evidence for glutamate as the olfactory receptor cell neurotransmitter. J. Neurophysiol. 71: 2557-2561, 1994.[Abstract/Free Full Text]
• BLANES, T. Sobre algunos puntos dudosos de la estructura del bulbo olfatorio. Rev. Trimestral Micrográfica 3: 99-127, 1898.
• BOWERY, N. G., HUDSON, A. L., PRICE, G. W. GABA_A and GABA_B receptor site distribution in the rat central nervous system. Neuroscience 20: 365-383, 1987.[Medline]
• BUFLER, J., ZUFALL, F., FRANKE, C., HATT, H. Patch-clamp recordings of spiking and nonspiking interneurons from rabbit olfactory bulb slices: GABA- and other transmitter receptors. J. Comp. Physiol. [A] 170: 153-159, 1992.[Medline]
• CHERUBINI, E., GAIARSA, J. L., BEN-ARI, Y. GABA: an excitatory transmitter in early postnatal life. Trends Neurosci. 14: 515-519, 1991.[Medline]
• DUCHAMP-VIRET, P., DUCHAMP, A. Odor processing in the frog olfactory system. Prog. Neurobiol. 53: 561-602, 1997.[Medline]
• EDWARDS, F. A., KONNERTH, A., SAKMANN, B., TAKAHASHI, T. A thin slice preparation for patch clamp recordings from neurones of the mammalian central nervous system. Pflügers Arch. 414: 600-612, 1989.[Medline]
• FREEMAN, W. J. Attenuation of transmission through glomeruli of olfactory bulb on paired shock stimulation. Brain Res. 65: 77-90, 1974.[Medline]
• GETCHELL, T. V., SHEPHERD, G. M. Short-axon cells in the olfactory bulb: dendrodendritic synaptic interactions. J. Physiol. (Lond.) 251: 523-548, 1975.[Abstract/Free Full Text]
• GOLGI, C. Sulla fina struttura dei bulbi olfattorii. Riv. Sper. Freniatr. Med. Leg. Alienazioni Ment. 1: 405-425, 1874.
• GUSEL'NIKOVA, K. G., TSITOLOVSKY, L. E., ENGOVATOV, V. V. Early inhibition in the frog olfactory bulb. Zh. Evol. Biokhim. Fiziol. 9: 593-598, 1973.
• HOWELL, B. J., BAUMGARDNER, F. W., BONDI, K., RAHN, H. Acid-base balance in cold-blooded vertebrates as a function of body temperature. Am. J. Physiol. 218: 600-606, 1970.[Free Full Text]
• JOHNSTON, D. Passive cable properties of hippocampal CA3 pyramidal neurons. Cell. Mol. Neurobiol. 1: 41-55, 1981.[Medline]
• KOSAKA, K., TOIDA, K., MARGOLIS, F. L., KOSAKA, T. Chemically defined neuron groups and their subpopulations in the glomerular layer of the rat main olfactory bulb. 2. Prominent differences in the intraglomerular dendritic arborization and their relationship to olfactory nerve terminals. Neuroscience 76: 775-786, 1997.[Medline]
• KRATSKIN, I., KENIGFEST, N., REPÉRANT, J., RIO, J. P., VESSELKIN, N. Presumptive GABAergic feedback input to the frog olfactory bulb: a double labelling study with retrograde axonal tracing and GABA immunohistochemistry. Brain Res. 581: 171-174, 1992.[Medline]
• KUNZE, W.A.A., SHAFTON, A. D., KEMM, R. E., MCKENZIE, J. S. Olfactory bulb output neurons excited from a basal forebrain magnocellular nucleus. Brain Res. 583: 327-331, 1992.[Medline]
• LAURENT, G., DAVIDOWITZ, H. Encoding of olfactory information with oscillating neural assemblies. Science 265: 1872-1875, 1994.[Abstract/Free Full Text]
• LAURIE, D. J., SEEBURG, P. H., WISDEN, W. The distribution of 13 GABA_A receptor subunit messenger RNAs in the rat brain. 2. Olfactory bulb and cerebellum. J. Neurosci. 12: 1063-1076, 1992.[Abstract]
• LE JEUNE, H., JOURDAN, F. Cholinergic innervation of olfactory glomeruli in the ratan ultrastructural immunocytochemical study. J. Comp. Neurol. 336: 279-292, 1993.[Medline]
• MACRIDES, F., DAVIS, B. J. The olfactory bulb. In: Chemical Neuroanatomy, edited by and P. C. Emson New York: Raven, 1983, p. 391-426.
• MCLEAN, J. H., SHIPLEY, M. T. Serotonergic afferents to the rat olfactory bulb. I. Origins and laminar specificity of serotoninergic inputs in the adult rat. J. Neurosci. 7: 3029-3039, 1987.[Abstract]
• NEHER, E. Correction for liquid junction potentials in patch clamp experiments. In: Methods in Enzymology. Ion Channels, edited by B. Rudy, and L. E. Iverson San Diego: Academic, 1992, vol. 207, p. 123-131.
• PINCHING, A. J., POWELL, T.P.S. The neuron types of the glomerular layer of the olfactory bulb. J. Cell Sci. 9: 305-345, 1971a.[Abstract/Free Full Text]
• PINCHING, A. J., POWELL, T.P.S. The neuropil of the periglomerular region of the olfactory bulb. J. Cell Sci. 9: 379-409, 1971b.[Abstract/Free Full Text]
• PINCHING, A. J., POWELL, T.P.S. The termination of centrifugal fibres in the glomerular layer of the olfactory bulb. J. Cell Sci. 10: 621-635, 1972.[Abstract/Free Full Text]
• PUOPOLO, M., BELLUZZI, O. Sodium current in periglomerular (PG) cells of rat olfactory bulb in vitro. Neuroreport 7: 1846-1850, 1996.[Medline]
• RALL, W. Core conductor theory and cable properties of neurons. In: Handbook of Physiology. The Nervous System. Cellular Biology of Neurons. Bethesda, MD: Am. Physiol. Soc., 1977, sect. 1, vol I, part 1, p. 39-97.
• RAMON Y CAJAL, S. Origen y terminacion de las fibras nerviosas olfatorias. Gas. Sanit. Barcelona III: 1-21, 1890.
• REEVES, R. B. Role of body temperature in determining the acid-base state in vertebrates. Federation Proc. 28: 1204-1208, 1969.[Medline]
• SHIPLEY, M. T., ENNIS, M. Functional organisation of olfactory system. J. Neurobiol. 30: 123-176, 1996.[Medline]
• SIKLÓS, L., RICKMANN, M., JOÓ, F., FREEMAN, W. J., WOLFF, J. R. Chloride is preferentially accumulated in a subpopulation of dendrites and periglomerular cells of the main olfactory bulb in adult rats. Neuroscience 64: 165-172, 1995.[Medline]
• TROMBLEY, P. Q., SHEPHERD, G. M. Synaptic transmission and modulation in the olfactory bulb. Curr. Opin. Neurobiol. 3: 540-547, 1993.[Medline]
• WEHR, M., LAURENT, G. Odour encoding by temporal sequences of firing in oscillating neural assemblies. Nature. 384: 162-166, 1996.[Medline]
• WHITE, E. L. Synaptic organization of the mammalian olfactory glomerulus: new findings including an intraspecific variation. Brain Res. 60: 299-313, 1973.[Medline]
• WHITTINGTON, A. M., TRAUB, R. D., JEFFERYS, J.G.R. Synchronized oscillations in interneuron networks driven by metabotropic glutamate receptor activation. Nature 373: 612-615, 1995.[Medline]
• YOKOI, M., MORI, K., NAKANISHI, S. Refinement of odor molecule tuning by dendrodendritic synaptic inhibition in the olfactory bulb. Proc. Natl. Acad. Sci. USA 92: 3371-3375, 1995.[Abstract/Free Full Text]
• ZABORSZKY, L., CARLSEN, J., BRASHEAR, H. R., HEIMER, L. Cholinergic and GABAergic afferents to the olfactory bulb in the rat with special emphasis on the projection neurons in the nucleus of the horizontal limb of the diagonal tract. J. Comp. Neurol. 243: 488-509, 1986.[Medline]

0022-3077/98 $5.00 Copyright ©1998 The American Physiological Society