The Journal of Neurophysiology Vol. 78 No. 4 October 1997, pp. 2176-2185
Copyright ©1997 by the American Physiological Society

Distribution and Activation of Intracellular Ca²⁺ Stores in Cultured Olfactory Bulb Neurons

Greg C. Carlson, Melissa L. Slawecki, Eric Lancaster, and Asaf Keller

Department of Anatomy and Neurobiology and The Neuroscience Program, University of Maryland School of Medicine, Baltimore, Maryland 21201

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Carlson, Greg C., Melissa L. Slawecki, Eric Lancaster, and Asaf Keller. Distribution and activation of intracellular Ca²⁺ stores in cultured olfactory bulb neurons. J. Neurophysiol. 78: 2176-2185, 1997. The presence and distribution of intracellular Ca²⁺ release pathways in olfactory bulb neurons were studied in dissociated cell cultures. Histochemical techniques and imaging of Ca²⁺ fluxes were used to identify two major intracellular Ca²⁺ release mechanisms: inositol 1,4,5-triphosphate receptor (IP₃R)-mediated release, and ryanodine receptor-mediated release. Cultured neurons were identified by immunocytochemistry for the neuron-specificmarker -tubulin III. Morphometric analyses and immunocytochemistry for glutamic acid-decarboxylase revealed a heterogeneous population of cultured neurons with phenotypes corresponding to both projection (mitral/tufted) and intrinsic (periglomerular/granule) neurons of the in vivo olfactory bulb. Immunocytochemistry for the IP₃R, and labeling with fluorescent-tagged ryanodine, revealed that, irrespective of cell type, almost all cultured neurons express IP₃R and ryanodine binding sites in both somata and dendrites. Functional imaging revealed that intracellular Ca²⁺ fluxes can be generated in the absence of external Ca²⁺, using agonists specific to each of the intracellular release pathways. Local pressure application of glutamate or quisqualate evoked Ca²⁺ fluxes in both somata and dendrites in nominally Ca²⁺ free extracellular solutions, suggesting the presence of IP₃-dependent Ca²⁺ release. These fluxes were blocked by preincubation with thapsigargin and persisted in the presence of the glutamate receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione. Local application of caffeine, a ryanodine receptor agonist, also evoked intracellular Ca²⁺ fluxes in the absence of extracellular Ca²⁺. These Ca²⁺ fluxes were suppressed by preincubation with ryanodine. In all neurons, both IP₃- and ryanodine-dependent release pathways coexisted, suggesting that they interact to modulate intracellular Ca²⁺ concentrations.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Neuronal mechanisms controlling cytoplasmic calcium concentrations ([Ca²⁺]_i) modulate a number of neuronal activities, including neurotransmitter release (Katz 1969), control of membrane excitability (Llinás 1990) and receptor function (Jaffe et al. 1994), differential gene expression (Sheng and Greenberg 1990), and neuronal migration and growth (Komuro and Rakic 1995). Increases in [Ca²⁺]_i occur by extracellular Ca²⁺ influx through transmembrane channels and intracellular release of Ca²⁺ sequestered in intracellular organelles. Because intracellular stores, such as the endoplasmic reticulum (ER), can sequester high concentrations of Ca²⁺, these stores serve not only to buffer free intracellular Ca²⁺, but can produce rapid, local Ca²⁺ transients in response to specific stimuli (Clapham 1995; Simpson et al. 1995). Intracellular stores release Ca²⁺ through at least two pathways, each mediated by a specific family of receptors. In one pathway, Ca²⁺ entry through the plasma membrane activates ryanodine receptors in the ER that mediate Ca²⁺ release from this organelle (Galione 1992). This process often is referred to as Ca²⁺-induced Ca²⁺ release (CICR) (Ehrlich et al. 1994; Henzi and MacDermott 1992; Irving et al. 1992; O'Neil et al. 1990). The other pathway involves activation of plasma membrane receptorssuch as the glutamate-sensitive metabotropic receptorsthat activate, through intermediary G proteins, phospholipase C, leading to the generation of inositol 1,4,5-triphosphate (IP₃). IP₃ then binds to an IP₃ receptor-Ca²⁺ channel complex on the ER, inducing Ca²⁺ release (Putney 1990; Simpson et al. 1995).

Although modulation of [Ca²⁺]_i in general has been shown to regulate numerous neuronal functions, in some cases, these activities may be specifically dependent on Ca²⁺ release from intracellular stores (reviewed in: Simpson et al. 1995). These include the modulation of neuronal excitability (Abdul-Ghani et al. 1996; Brorson et al. 1991; Kawai and Watanabe 1989), presynaptic transmitter release (Blaustein et al. 1980; Peng 1996), long-term depression and potentiation (Behnisch and Reymann 1995; Kasono and Hirano 1995; Kohda et al. 1995; Reyes and Stanton 1996), and neuronal development (Levick et al. 1995; Stewart et al. 1995). Therefore, studies of Ca²⁺-dependent phenomena in neurons must consider the presence of intracellular release pathways.

The present study was designed to identify the types of intracellular Ca²⁺ release mechanisms in cultured neurons of the rat olfactory bulb. Functional imaging of Ca²⁺ fluxes and immunocytochemical techniques revealed that both IP₃- and ryanodine-dependent release mechanisms are expressed in a heterogeneous population of these neurons. Some of these results were published previously in abstract form (Carlson et al. 1996).

	METHODS

Abstract Introduction Methods Results Discussion References

Cell cultures

Dissociated cell cultures were prepared from 3-day-old Wistar rat pups. Olfactory bulbs were dissected into culture medium containing minimal essential medium supplemented with L-alanyl-L-glutamine (MEM  medium, GIBCO), 10% fetal calf serum (GIBCO), 5% horse serum (GIBCO), and 0.6% glucose. The meninges were removed, and the bulbs were dissociated both enzymatically (papain; 20 U/ml, Worthington) and by gentle mechanical trituration. After filtration and centrifugation, cell suspensions were plated on glass coverslips coated with poly-D-lysine and laminin. The coverslips were incubated at 37°C in a humidified incubator saturated with 5% CO₂. After 4 day in culture, the cells were treated with antimitotic agents (20 µM uridine and 5-fluoro-2-deoxyuridine, Sigma) for 24 h. Every 4 days thereafter the culture medium was replaced with a solution containing MEM  medium, 5% fetal calf serum, 5% horse serum, and 0.6% glucose. All experiments were performed with cell cultures ranging in age from 8 to 24 days in vitro.

Immunocytochemistry

Cell cultures were fixed in methanol at 20°C. Unless otherwise indicated, all incubations were performed at room temperature. Several rinses in phosphate-buffered saline (PBS, 0.1 M, pH 7.4) were performed between each of the incubation steps. The following antibodies were used: mouse anti- tubulin-III (1:50; Sigma); sheep anti-glutamic acid decarboxylase (GAD1440-4; 1:2400; gift of Dr. Kopin, NINDS); rabbit anti-IP₃R (AP2A, 10 µg/µl). The IP₃R antibody generously was provided by Dr. Alan Sharp, who has characterized this antiserum (Sharp et al. 1993).

Preincubation for double labeling of  tubulin-III and GAD was performed for 1 h in PBS containing 5% normal donkey serum. The cultures then were incubated overnight at 4°C in PBS containing the two primary antibodies and donkey serum. The secondary antibody mixture of rhodamine-conjugated donkey anti-mouse and fluorescein-conjugated donkey anti-sheep (both at 1:100; Jackson ImmunoResearch) solution in PBS then was applied to the culturesfor 1 h.

For  tubulin-III/IP₃R double-labeling experiments, cultures were preincubated for 1 h in PBS containing 2.5% normal donkey serum, 2.5% normal goat serum, and 0.2% Triton-X. The cultures were incubated overnight at 4°C in PBS containing the two primary antibodies and donkey serum. The cultures then were incubated in a secondary antibody mixture of biotinylated goat anti-rabbit (1:200; Vector Laboratories) and rhodamine-conjugated donkey anti-mouse (1:100; Jackson ImmunoResearch) for 1 h. A 1 h incubation in fluorescein-conjugated avidin (1:100; Vector Laboratories) diluted in PBS followed. To further increase the IP₃R signal, a 1-h incubation in biotinylated goat anti-avidin (1:100; Vector Laboratories) was performed, and a second incubation in fluorescein-conjugated avidin (1:100; Vector Laboratories) completed the procedure.

A similar procedure was used for GAD/IP₃R double labeling, except that Triton-X was not used, the blocking serum consisted of 2.5% normal donkey serum and 2.5% normal goat serum, anti-GAD was used instead of the  tubulin-III antibody and detected with rhodamine-conjugated donkey anti-sheep secondary antibody (1:100; Jackson ImmunoResearch).

After the final washes, coverslips were mounted on glass slides in Prolong antifade reagent (Molecular Probes) and examined with a confocal scanning light microscope (Zeiss LSM410). Digitized images were stored on a personal computer.

Control experiments were performed by omitting the primary or secondary antibody; staining was not detected in these controls.

Ryanodine staining

Ryanodine receptor binding sites were localized in coverslips of live cells placed in a recording chamber that was mounted on an inverted microscope. The cultures were incubated for 15 min in monosubstituted BODIPY FL-X ryanodine derivative (500 nM; Molecular Probes). Digital images of BODIPY FL-X stained cells and brightfield Nomarski images were collected simultaneously with a confocal scanning light microscope (Olympus FluoView) and stored on a personal computer.

Image analysis

All digitized images were analyzed on a PowerMacintosh 9500, using IPLabs (Signal Analytics) and Photoshop (Adobe). Image manipulations were restricted to linear level adjustments and cropping.

Functional Ca²⁺ imaging

For detecting Ca²⁺ fluxes, cultured cells were incubated with the Ca²⁺ indicator fluo-3 AM (20 µM; Molecular Probes) for 0.75-1 h at 37°C. The cultures were placed in a recording chamber (500 µl volume) mounted on an inverted microscope (Olympus IX70) and perfused with artificial cerebrospinal fluid (ACSF) at 2 ml/min. ACSF contained (in mM) 124 NaCl, 26 NaHCO, 1.2 NaH₂PO₄, 3.2 KCl, 1.2 MgSO₄, 2.4 CaCl₂, and 20 glucose. To prepare nominally Ca²⁺-free ACSF (0 Ca²⁺), the Ca²⁺ was replaced with equimolar Mg²⁺, and 1 mM ethylene glycol-bis(-aminoethyl ether)-N,N,N,N-tetraacetic acid was added.

The following drugs were applied through the perfusate (2ml/min): 10 µM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), 10 µM thapsigargin, and 10 µM ryanodine. Stimulation of visually identified neurons was accomplished by pressure application of agonists through the tip of a large patch pipette (7-9 µm diam) placed within 40 µm of the selected cells. Agonists were applied at the following concentrations: glutamate, 10 or 20 µM; quisqualate, 10 µM; and caffeine, 10 mM, with final concentrations in 0 Ca²⁺ ACSF. All drugs were obtained from Research Biochemicals International, except for glutamate and caffeine, which were obtained from Sigma. Pressure pulses, 30 ms to 3 s in duration, were applied to the back of the pipette with a Picospritzer (General Valve).

Fluorescent signals were collected through either a ×40 (N.A. 1.35) or ×60 (N.A. 1.40) objective (Olympus) using a dichroic mirror and filter set that generate an excitation wavelength centered on 490 nm and limited emission to between 518 and 542 nm. To reduce photobleaching, neutral density filters were used, and a shutter prevented illumination of the samples except during data acquisition.

Fluorescent images were collected with an intensified, cooled charge-coupled device (CCD; Princeton Instruments). Individual frames were obtained by integrating the fluorescent signal on the CCD chip for 10-20 ms, and frames were collected every 250 ms. The images were digitized at 16 bits and captured with the software WinView (Princeton Instruments) on a Pentium based personal computer. Analyses were performed in IPLabs (Signal Analytics) on a Power Macintosh 9500.

Changes in fluorescence were determined by calculating peak fluorescence changes in the form of F/F_i, where F_i is the average fluorescence value of a region of interest over 10 frames directly preceding drug application and F is the difference between the fluorescence value of a region of interest and the baseline signal (F_i). This ratio is reported as percent change; unless otherwise indicated, these ratios were calculated for the somatic region of the cell. Cells were considered to have responded to drug application if the evoked fluorescence signal was >5% relative to baseline. This value was arbitrary, but none of the cells that failed to respond showed a >1% change in fluorescence over baseline. All grouped data are expressed as means ± SE.

Estimates of [Ca²⁺]_i from fluorescent values were calculated as described by Kao (1994), using the divalent cation ionophore4-bromo-A23187 (10 µM, Sigma), in the presence of 4 mM Mn²⁺ in 0 Ca²⁺, and by permibilizing the cells with digitonin (2 mM, Sigma).

	RESULTS

Abstract Introduction Methods Results Discussion References

Characterization of cultured neurons

Dissociated cultures of the olfactory bulb contained a heterogeneous population of cells with respect to size and morphology (Fig. 1A). One class of cells appeared phase-bright in phase contrast microscopy, had a large, clear nucleus with a prominent nucleolus, and extended processes. These features are characteristic of neurons in primary dissociated cultures (Banker and Goslin 1991). To confirm that these cells were neurons, they were stained for the neuron-specific marker  tubulin-III (Sullivan et al. 1986). All cells identified as neurons using phase-contrast microscopy stained positively for  tubulin-III (Fig. 1B). The neurons typically developed on top of a confluent, monolayer of glial cells.

View larger version (109K): [in this window] [in a new window]   FIG. 1. Morphology and neurochemistry of cultured olfactory bulb neurons. A: digitized phase-contrast image showing a pyramidal-shaped neuron (p) having an apical dendrite that terminates in a tuft and a multipolar neuron (m). B-E: pseudocolor confocal images of cultured olfactory bulb neurons. B: cells stained for neuronal marker  tubulin-III (red). Bipolar cell (right) also stains for glutamic acid-decarboxylase (GAD; green), whereas presumptive mitral/tufted cell is GAD(). C: almost all neurons, identified by their  tubulin-III staining (red), also stained for inositol 1,4,5-triphosphate receptor (IP3R; green). Nonneuronal cells, presumably astrocytes, were also IP3R(+). D and E: double-labeling for both IP3R (green) and GAD (red) demonstrated that most, but not all (E) IP3R(+) neurons are GAD(+). F: combined Nomarski-optics and fluorescence confocal image demonstrating binding of labeled ryanodine (green) in a cultured neuron. Note that astrocytes are not labeled. Scale bars represent 20 µm.

Previous studies suggest that morphological criteria can be used to identify two classes of neurons in dissociated olfactory bulb cultures: projection (mitral/tufted) and intrinsic (periglomerular/granule) cells (e.g., Trombley and Westbrook 1990). To determine whether these classes of neurons exist in our culture system, quantitative morphometric approaches and immunocytochemical techniques were used to characterize the cultured neurons. The area of the somata of  tubulin-III labeled cells, measured from two-dimensional confocal images, ranged from 38 to 252 µm² (70.3 ± 3.6 µm²). Most neurons were classified as multipolar (58/104; 56%) or bipolar (24%), based on the patterns of processes emanating from their somata. A smaller number of neurons (20%) had a distinct apical dendrite that tapered gradually from its origin at the parent soma and sometimes terminated in a dendritic "tuft" characteristic of mitral/tufted cells (Fig. 1A). Although these neurons resemble mitral and tufted cells, the size of their somata did not differ significantly from that of the multipolar or bitufted cells. It was thus not possible to determine, using morphological criteria alone, whether the different classes of neurons in dissociated cultures correspond to the two major groups of cells in the olfactory bulb in vivo: projection (mitral/tufted) and intrinsic neurons.

Because many types of intrinsic neurons in the olfactory bulbincluding periglomerular and granule cellscontain the neurotransmitter -aminobutyric acid (GABA) (see Shipley and Ennis 1996), we further characterized the cultured neurons by immunocytochemical localization of GAD, the rate-limiting enzyme in GABA synthesis (Fig. 1B). Double labeling for GAD and  tubulin-III revealed that GAD is expressed in the vast majority of cultured neurons (292/306; 95%). The proportion of GAD(+) neurons in our cultures is similar to the ratio of intrinsic to projection neurons in the olfactory bulb in vivo (see Shepherd 1990). Most neurons that did not express GAD had pyramidal shaped somata that were, on average, larger than those of GAD(+) cells (area: 91.0 ± 19.0 µm² vs. 50.2 ± 2.7 µm²; n = 306). Although some neurons with large somata (area 55 µm²) were GAD(+), a significant proportion (25%) of these large neurons were GAD() and therefore classified as presumptive mitral/tufted cells. Therefore, in the Ca²⁺ imaging experiments (described below) we selected both large and small neurons in an attempt to include in the analyses both presumptive projection and intrinsic neurons.

To determine whether different classes of cultured neurons express IP₃R, we immunolabeled cultured neurons using an antibody to IP₃R that was previously characterized by Sharp et al. (1993). IP₃R labeling was present in both glia and in neurons, where labeling was most dense in the somata and proximal dendrites. To determine whether IP₃R is expressed in specific cell types, we double labeled cell cultures using antibodies to both  tubulin-III and IP₃R (Fig. 1C). IP₃R was expressed in 91% of the neurons (94/103). There were no differences in either morphology or size of IP₃R(+) and IP₃R() neurons, suggesting that all classes of cultured olfactory bulb neurons express IP₃R. To further test whether IP₃R is expressed preferentially in specific cell types, double-labeling using antibodies to GAD and IP₃R was performed (Fig. 1, D and E). The vast majority of GAD(+) cells examined (123/125; 98.4%) also stained for IP₃R, indicating that essentially all cultured GABAergic neurons express IP₃R. A small proportion (4/87; 4.6%) of the IP₃R(+) neurons did not stain for GAD, suggesting that excitatory neurons also express IP₃R.

To determine the expression patterns of the ryanodine receptor in cultured neurons, in vitro cell cultures were treated with a monosubstituted BODIPY FL-X ryanodine derivative (500 nM). This resulted in labeling in the somata and dendrites of all neurons examined (Fig. 1F). When BODIPY FL-X ryanodine was coapplied with 10 µM unlabeled ryanodine, no staining was observed. These findings indicate that all types of cultured olfactory bulb neurons express ryanodine-binding sites, presumably representing ryanodine receptors. Unlike the IP₃R, which also was expressed by glia, no ryanodine labeling was detected in the astrocyte layer of the cultures (Fig. 1F).

In summary, all classes of neurons express IP₃R and ryanodine binding sites, and these are distributed in both the somata and dendrites. These findings imply that CICR and IP₃-sensitive intracellular release pathways are present in both the soma and dendrites of these neurons. This hypothesis was tested directly with the use of Ca²⁺ imaging approaches.

Intracellular Ca²⁺ responses

We examined the distribution of intracellular Ca²⁺ release pathways in cultured olfactory bulb neurons using drug application combined with functional Ca²⁺ imaging. To identify changes in intracellular Ca²⁺ concentrations ([Ca²⁺]_i), we bulk loaded the olfactory bulb cultures with the fluorescent Ca²⁺ indicator fluo-3 AM and recorded changes in fluorescence. To generate grouped data, uncorrected fluorescence data were normalized to changes in fluorescence as percent change over baseline (F/F_i). To describe changes in peak responses of an individual cell under different conditions, percent differences in uncorrected fluorescence peak amplitudes were calculated. This allowed comparisons of responses between cells and the use of consistent criteria for determining the presence of evoked Ca²⁺ fluxes. Because Ca²⁺ release from intracellular stores is sensitive to [Ca²⁺]_i, we also performed a semiquantitative calibration (Kao 1994), which revealed resting [Ca²⁺]_i levels that ranged between 65 and 80 nM (74 ± 1 nM; n = 12), values that are consistent with those reported for other types of CNS neurons (Regehr et al. 1989; Shmigol et al. 1994).

We first tested the response of cultured olfactory bulb neurons to glutamatethe primary excitatory neurotransmitter in the bulb (Ennis et al. 1996; Trombley and Westbrook 1990). After a brief pressure application of glutamate (30-300 ms, 10 or 20 µM) near the soma, olfactory bulb neurons in normal ACSF responded with a robust increase in [Ca²⁺]_i in both their somata (mean peak F/F_i =159 ± 26%; n = 6) and dendrites (mean peak F/F_i =89 ± 9%; n = 7 dendrites from 3 cells; Fig. 2C, left). In all cells, the initial response in Ca²⁺-containing ACSF was immediate, occurring within the first 250 ms. The total duration and rate of decay of these responses showed a variety of kinetics, both between cells and among trials from the same cell. Often there was a long-lasting plateau in the response, as shown in Fig. 2C. Other types of responses were relatively rapid and monophasic or long-lasting with multiple peaks. In the presence of Ca²⁺-containing ACSF, the origin of these responses could not be determined because glutamate can activate a number of different receptors. Therefore, the Ca²⁺ fluxes in Ca²⁺-containing ACSF could be mediated by a number of Ca²⁺ sources. These include activation of Ca²⁺-permeable NMDA receptors or other trans-membrane ionotropic receptors, leading to voltage-dependent calcium channel gating, as well as release from intracellular Ca²⁺ stores.

View larger version (20K): [in this window] [in a new window]   FIG. 2. A: digitized phase-contrast image of a cultured neuron. Scale bar represents 10 µm. B: pseudocolor image of peak Ca2+ transient (expressed as percent change in fluorescence of fluo-3) evoked by glutamate application in 0 Ca2+. Somatic and dendritic regions from which fluorescence measurements were calculated are delineated in white. C: plots of changes in fluorescence as a function of time calculated for soma and dendrite of cell shown in A and B. Glutamate (10 µM) was pressure applied for 300 ms at times indicated (), in normal artificial cerebrospinal fluid (ACSF; left) and 0 Ca2+ (right). * Time point used to generate color image on left. EGTA, ethylene glycol-bis(-aminoethyl ether)-N,N,NN-tetraacetic acid.

To test if cultured olfactory neurons can generate Ca²⁺ fluxes that are dependent only on intracellular Ca²⁺ stores, glutamate was applied after 5 min of superfusion with nominally Ca²⁺-free ACSF (0 Ca²⁺). In averaged grouped data of normalized responses, only a slightly smaller change in fluorescence was seen in 0 Ca²⁺ compared with responses in normal ACSF (mean peak at the soma F/F_i = 152 ± 25%; n = 6; in dendrites 85 ± 15%; n = 7; Fig. 2C, right). This similarity in F/F_i values in 0 Ca²⁺ and in normal ACSF did not imply that nearly the same levels of [Ca²⁺ ]_i were reached under these two conditions because F/F_i normalizes the change in fluorescence to the initial baseline, and the baseline fluorescence was always lower in 0 Ca²⁺. This was represented when the peak fluorescence responses in normal ACSF and in 0 Ca²⁺ were compared for each individual cell: the peak response was always less in 0 Ca²⁺ (83 ± 6% of the ACSF response).

These results indicate that olfactory bulb neurons can generate Ca²⁺ fluxes in the absence of external Ca²⁺, suggesting that these Ca²⁺ fluxes originate from intracellular Ca²⁺ stores.

IP₃-sensitive stores

To test for the presence of IP₃-dependent intracellular Ca²⁺ release, we used quisqualate, an agonist of the type 1 and 5 metabotropic glutamate receptors (mGluR_1,5) (Aramori and Nakanishi 1992; Sladeczek et al. 1985; Tanabe et al. 1992). Activation of mGluR_1,5 has been shown to evoke the production of IP₃ in neurons through a G-protein-mediated process linked to phospholipase C (Kirischuk et al. 1995; Seymour-Laurent and Barish 1995; Verkhratsky and Kettenmann 1996). Therefore, quisqualate can be usedto assay Ca²⁺ release from IP₃-sensitive Ca²⁺ stores. Because quisqualate is also a potent and selective agonist of the ionotropic -amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) glutamate receptors (Mayer and Westbrook 1987), as well as a mGluR_1,5 agonist, Ca²⁺ transients evoked in normal ACSF could arise from both transmembrane sources and intracellular stores. This allowed us to use two approaches to isolate IP₃-dependent Ca²⁺ release from intracellular stores. First, by lowering external [Ca²⁺] to limit transmembrane Ca²⁺ influx, and second, by blocking the AMPA receptor-mediated responses to suppress iontopic receptor activation and the concomitant changes in membrane potential, thereby limiting Ca²⁺ fluxes through both Ca²⁺-permeable AMPA channels and voltage-activated Ca²⁺ channels.

In normal ACSF, quisqualate application produced a variable rise of [Ca²⁺]_i in all neurons (mean peak F/F_i = 198 ± 16%; n = 40), similar to that seen after glutamate application (Figs. 3A and 4A). In 0 Ca²⁺, intracellular Ca²⁺ fluxes occurred in nearly all neurons tested (35/40; mean peak response F/F_i = 77 ± 12%; Figs. 3B and 4B). Consistent with the assumption that this response was due to activation of IP₃ receptors via mGluR_1,5 activation, similar results also were obtained in normal ACSF when the AMPA receptor antagonist CNQX was applied to suppress the ionotropic actions of quisqualate, as shown in Fig. 3D (peak F/F_i = 56 ± 21%; n = 6). Similar to the responses after glutamate application, the quisqualate-evoked Ca²⁺ fluxes occurred simultaneously in both dendrites and somata. Because Ca²⁺ diffusion is restricted (Allbritton et al. 1992), this finding suggests that IP₃-sensitive stores are distributed in both these compartments (Fig. 2).

View larger version (7K): [in this window] [in a new window]   FIG. 3. Release of Ca2+ from intracellular stores evoked by IP3R activation after quisqualate application. Changes in fluorescence plotted as a function of time calculated for soma of a cultured neuron. Quisqualate (10 µM) was pressure applied for 300 ms (). In normal ACSF (A), fluorescence changes represent Ca2+ transients through both plasma membrane and from intracellular stores. Intracellular release component is revealed in 0 Ca2+ (B) and also is demonstrated by effects of -amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX; D).

View larger version (63K): [in this window] [in a new window]   FIG. 4. Release of Ca2+ from IP3-sensitive intracellular stores in neurons and glia. A: plots of changes in fluorescence as a function of time calculated for somata of a cultured neuron and an astrocyte. These cells are shown in B, where neuron (N) is stained for both  tubulin-III (red) and IP3R (green), whereas glia cell (G) stains only for IP3R. Immunocytochemical localization was performed at end of imaging experiment. Quisqualate (10 µM) was pressure applied for 300 ms (). Responses in glia follow neuronal responses with a latency of ~1 s in both normal ACSF and 0 Ca2+. Changes in raw fluorescence are depicted in pseudocolor images below traces; times indicate intervals after quisqualate application.

To confirm that Ca²⁺ fluxes in the presence of 0 Ca²⁺ represented release from intracellular stores, thapsigargin was bath-applied. Thapsigargin causes depletion of intracellular stores by inactivating the ATPase responsible for loading Ca²⁺ stores against the Ca²⁺ concentration gradient (Thastrup et al. 1990). Pretreatment of the cultures with 10 µM thapsigargin for 20-30 min completely suppressed quisqualate-induced responses in 0 Ca²⁺, whereas Ca²⁺ fluxes still could be evoked by quisqualate in normal ACSF (n = 3).

Each cell from which Ca²⁺ imaging data were obtained also was included in anatomic analyses to determine its size and morphology, and, whenever possible, its GAD and IP₃R immunoactivity (e.g., Fig. 4B). There was no correlation between the amplitude and kinetics of the responses of individual neurons to glutamate or to quisqualate and their characterization as presumptive projection or intrinsic neurons.

Responses in glia

In nonneuronal cells, glutamate or quisqualate application rarely evoked Ca²⁺ fluxes (4 cells in 53 image fields), despite the presence of putative astrocytes in each field, adjacent to neurons in the field that were activated in 0 Ca²⁺, and despite the fact that nonneuronal cells stained positive for IP₃R (Figs. 1C and 4B). The failure to evoke Ca²⁺ fluxes in these putative glial cells is consistent with the reported absence of mGluR_1,5 activity in cultures of cortical astrocytes (Miller et al. 1995; but see Milani et al. 1989). When nonneuronal cells did respond, as shown in Fig. 4, they did so with a robust monophasic response, limited to a discreet phase-dark area of the cell corresponding to the nucleus. This response occurred in both the presence and absence of external Ca²⁺. Unlike the neuronal responses, the onset of Ca²⁺ fluxes in nonneuronal cells was not immediate, but was delayed by ~1 s after the application of glutamate and quisqualate (Fig. 4).

Ryanodine-sensitive stores

A second mechanism of intracellular Ca²⁺ release in neurons is CICR from ryanodine-sensitive stores (Henzi and MacDermott 1992; Irving et al. 1992). Caffeine has been shown to activate these stores while suppressing IP₃-induced Ca²⁺ release (Henzi and MacDermott 1992; Irving et al. 1992; Parker and Ivorra 1990) and therefore is a commonly used agonist to isolate the presence of ryanodine-sensitive stores (Ehrlich et al. 1994; Irving et al. 1992; O'Neil et al. 1990). To select viable neurons, caffeine was applied to cells that showed a quisqualate or glutamate response. Caffeine application triggered a monophasic rise in [Ca²⁺]_i (Fig. 5A), and this response was evoked in almost all cells tested (n = 12/13, peak F/F_i = 113 ± 25%). As illustrated in Fig. 5A, caffeine-induced Ca²⁺ fluxes also were evoked in 0 Ca²⁺ with kinetics similar to responses evoked in normal ACSF, but the amplitude of the responses in 0 Ca²⁺ were smaller (n = 9/11, peak F/F_i = 74 ± 19%; P = 0.049, single factor analysis of variance).

View larger version (14K): [in this window] [in a new window]   FIG. 5. Release of Ca2+ from ryanodine-sensitive intracellular stores by caffeine application. Plots of changes in fluorescence as a function of time calculated for soma of a cultured neuron. Caffeine (10 mM) was pressure applied for 3 s at times indicated. A: Ca2+ fluxes evoked in normal ACSF remain in 0 Ca2+ but are blocked by 20-min application of ryanodine. Inset: digitized pseudocolor image of peak fluorescence response in 0 Ca2+. B: suppression of response by ryanodine (middle) is partially reversible after 40-min perfusion with normal ACSF. Inset: phase-contrast image of cell from which recordings were obtained. A decrease in fluoresence intensity (seen in A and to a lesser extent in B) often occcured during caffeine application. Scale bar represents 10 µm.

To confirm that the caffeine-induced response was due to activation of ryanodine-sensitive stores, ryanodine was bath-applied in normal ACSF to suppress release from ryanodine-sensitive stores (Irving et al. 1992). Although ryanodine is an agonist of ryanodine receptors at low concentrations (nanomolar), it acts as an antagonist at higher concentrations (Ehrlich et al. 1994; Henzi and MacDermott 1992). As shown in Fig. 5B, bath application of 10 µM ryanodine for 20 min completely suppressed caffeine-induced Ca²⁺ fluxes in both normal ACSF and 0 Ca²⁺ (n = 4). As evidence that Ca²⁺ stores in these cells remained viable after this treatment, partial restoration of this response was produced after a 40-min washout of ryanodine in normal ACSF (Fig. 5B). Caffeine application failed to evoke Ca²⁺ fluxes in all of the nonneuronal cells examined, suggesting that these cells may lack a comparable ryanodine-sensitive release mechanism.

As with the glutamate- and quisqualate-mediated responses, caffeine application evoked Ca²⁺ fluxes in both the soma and proximal dendrites of these neurons. Furthermore, every cell that responded to caffeine also showed IP₃-sensitive release (n = 12). There was no correlation between the responses of individual neurons to various antagonists and their characterization as presumptive projection or intrinsic neurons. Thus both forms of intracellular Ca²⁺ release are present in a heterogeneous population of cultured olfactory bulb neurons and are colocalized in the soma and the dendrites.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

The aim of this study was to identify the mechanisms of Ca²⁺ release from intracellular stores in cultured olfactory bulb neurons. As demonstrated above, olfactory bulb cultures contain a heterogeneous population of neurons, including presumptive projection and intrinsic neurons. Consistent with the relative proportions of these cell types in vivo, the vast majority of the neurons are GABAergic, presumably representing periglomerular and granule cells. These findings are in agreement with previous descriptions of cultured olfactory bulb neurons (Trombley and Westbrook 1990).

The results demonstrate that an IP₃ and a CICR agonist can both generate Ca²⁺ release from intracellular stores in the somata and dendrites of a heterogeneous population of neurons. Although the presence of CICR was not directly tested, the ability of caffeine to generate release from these stores and of ryanodine to antagonize release at micromolar concentrations is a defining pharmacological characteristic of the receptors involved in CICR (Ehrlich et al. 1994; Henzi and MacDermott 1992; Irving et al. 1992; O'Neil et al. 1990). The changes in [Ca²⁺] recorded in the somatic and dendritic regions of the cells are likely to represent Ca²⁺ release from localized intracellular stores, because Ca²⁺ diffusion is known to be highly restricted (Allbritton et al. 1992). This suggests that both IP₃- and ryanodine-dependent Ca²⁺ release mechanisms exist in both the somata and dendrites of cultured olfactory bulb neurons. Support for the presence of these Ca²⁺ release pathways also is derived from immunocytochemical findings of IP₃R expression and ryanodine binding in almost all of the neurons examined, irrespective of their morphology, size, or GAD immunoreactivity. Thus we demonstrate that cultured olfactory bulb neurons support both IP₃- and ryanodine-sensitive Ca²⁺ release and that these two sources of Ca²⁺ are colocalized in both somatic and dendritic compartments.

The functional imaging methods used in this study limited us to a qualitative description of both these release pathways. Fluo-3 responds to changes in [Ca²⁺]_i nonlinearly, and it is difficult to control for differences in [fluo-3]_i that would affect calibration of [Ca²⁺]_i, especially at higher [Ca²⁺]_i. Furthermore, when peak [Ca²⁺]_i levels were estimated they often indicated levels that could saturate fluo-3, rendering it impossible to accurately estimate the amplitudes and kinetics of evoked Ca²⁺ fluxes (Kao 1994). The results do show that many of these response are large, generating >10-fold changes in [Ca²⁺]_i.

We found no correlation between response amplitudes or kinetics and the size or morphology of the cells. The variability of the responses may be attributed to a number of sources. These include variations in the effective concentration of the pressure-applied agonists, differences in intracellular fluo-3 concentrations, and the physiological status of the cells.

IP₃-mediated Ca²⁺ release

The presence of IP₃-dependent stores was demonstrated by the ability of mGluR_1,5 receptors to generate intracellular Ca²⁺ fluxes in nominally Ca²⁺-free extracellular solutions. In the in vivo olfactory bulb, IP₃-dependent Ca²⁺ release may be evoked by activation of various metabotropic receptors. These include serotonin receptors type 2, muscarinic receptors type 1, ₁-adrenoceptors (reviewed in Shipley et al. 1996) as well as mGluR (van den Pol 1995). Thus the intracellular Ca²⁺ release demonstrated here may be activated in vivo by olfactory nerve inputs, intrinsic glutamatergic activity, and several of the afferent systems projecting to the olfactory bulb.

Our finding that IP₃-sensitive Ca²⁺ stores are expressed in a heterogeneous population of neurons is consistent with other studies of cultured olfactory bulb neurons. Tani et al. (1992) reported that cultured mouse olfactory bulb interneurons respond to serotonin with an increase in [Ca²⁺]_i in the absence of extracellular Ca²⁺. Similarly, putative mitral/tufted cells produce Ca²⁺ fluxes in 0 Ca²⁺ after activation of mGluRs (Geiling and Schild 1996; van den Pol 1995). Geiling and Schild (1996) report that in cultures of the Xenopus laevis tadpole olfactory bulb, glutamate-dependent Ca²⁺ release from intracellular stores is present only in projection neurons. In contrast, our study in the rat demonstrates that IP₃-dependent release occurs in all types of cells. The disparity in the results may reflect a species difference or different culture conditions. Alternatively, these differences may be due to the developmental stage of the cells studied-in contrast to the Xenopus tadpole data, our findings are based on studies of cultures prepared from postnatal rats. Support for this hypothesis is derived from a recent study in our laboratory that demonstrated a developmental shift in IP₃R expression in the rat olfactory bulb in vivo (Slawecki et al. 1996). Specifically, we find that during the first postnatal week, IP₃R is expressed exclusively in mitral cells (consistent with Dent et al. 1996) and that IP₃R expression in intrinsic neurons gradually develops during the second and third postnatal weeks concomitant with a reduction in IP₃R expression in mitral cells. By the fourth postnatal week, an adult pattern of IP₃R expression develops in which the receptor is expressed preferentially in GABAergic periglomerular cells and in a distinct population of interneurons in the granule cell layer (Slawecki et al. 1996).

If the IP₃R in vivo is expressed preferentially in subclasses of olfactory bulb neurons, why do essentially all cultured bulbar neurons express IP₃R-mediated Ca²⁺ release? One possibility is that the developmental cues found in vivo may be lacking in culture. Although olfactory bulb cultures can produce a diverse population of cell types with many of the characteristic neurotransmitters, receptors, and synaptic interconnections described in the intact olfactory bulb (present study and see Kim 1972; Trombley 1994; Trombley and Westbrook 1990; van den Pol 1995), they lack the appropriate peripheral inputs that may function to regulate the development of these cells into their mature in vivo phenotype. This possibility suggests that expression of IP₃R is activity dependent. The developmental shifts in IP₃R expression described above (Slawecki et al. 1996) support this hypothesis. This hypothesis also is supported by reports of activity-dependent down regulation of IP₃R expression (Wojcikiewicz 1995; Wojcikiewicz and Oberdorf 1996).

Calcium-induced calcium release

The ability of ryanodine to block caffeine-evoked Ca²⁺ fluxes in both normal ACSF and in the absence of external Ca²⁺ suggests that caffeine triggers Ca²⁺ release from ryanodine-sensitive intracellular stores. However, peak responses in Ca²⁺-free ACSF averaged only 60 ± 24% of the caffeine induced responses in normal ACSF, and a few cells did not respond at all in 0 Ca²⁺. It is possible that caffeine may activate another source of Ca²⁺ in the presence of extracellular Ca²⁺. Alternatively, these results are consistent with findings that Ca²⁺ release from ryanodine-sensitive stores is sensitive to Ca²⁺ depletion and cytoplasmic [Ca²⁺] (Shmigol et al. 1994). Thus incubation in 0 Ca²⁺ may decrease the [Ca²⁺]_i available in the CICR pool for caffeine-dependent release or suppress the responses by modulating the sensitivity or activity of ryanodine receptors.

Activation of ryanodine-dependent CICR in bulbar neurons in vivo may be triggered by a number of mechanisms that evoke increases in [Ca²⁺]_i. These mechanisms are likely to include both trans-membrane Ca²⁺ fluxes and Ca²⁺ release from IP₃-dependent stores. It is therefore possible that some portion of the IP₃-mediated responses described in the present study may include CICR, triggered by Ca²⁺ efflux from IP₃-sensitive stores. Because both IP₃- and ryanodine-dependent stores are modulated by [Ca²⁺]_i in a biphasic manner (Simpson et al. 1995), Ca²⁺ release from each of these stores can activate, facilitate, or suppress release from the other. Therefore, coactivation of these stores along with transmembrane Ca²⁺ fluxes may interact to produce complex variations in [Ca²⁺]_i responses to both ionotropic and metabotropic receptor activation.

	ACKNOWLEDGEMENTS

  The authors are grateful to Drs. Mordecai P. Blaustein, Mathew Ennis, Joseph P. Kao, and Daniel Weinreich for valuable comments on this manuscript.

  This work was supported in part by National Institute of Neurological Disorders and Stroke Grants NS-31078 and NS-35360.

	FOOTNOTES

  Address for reprint requests: A. Keller, Dept. of Anatomy and Neurobiology, University of Maryland School of Medicine, 685 W. Baltimore St., Baltimore, MD 21201. E-mail: akeller{at}umabnet.ab.umd.edu

  Received 28 January 1997; accepted in final form 29 April 1997.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• ABDUL-GHANI, M. A., VALIANTE, T. A., CARLEN, P. L., PENNEFATHER, P. S. Metabotropic glutamate receptors coupled to IP₃ production mediate inhibition of IAHP in rat dentate granule neurons. J. Neurophysiol. 76: 2691-2700, 1996.[Abstract/Free Full Text]
• ALLBRITTON, N. L., MEYER, T., STRYER, L. Range of messenger action of calcium ion and inositol 1,4,5,-triphosphate. Science 258: 1812-1815, 1992.[Abstract/Free Full Text]
• ARAMORI, I., NAKANISHI, S. Signal transduction and pharmacological characteristics of a metabotropic glutamate receptor, mGluR1, in transfected CHO cells. Neuron 8: 757-765, 1992.[Medline]
• BANKER, G. AND GOSLIN, K. Culturing Nerve Cells. Cambridge: The MIT Press, 1991.
• BEHNISCH, T., REYMANN, K. G. Thapsigargin blocks long-term potentiation induced by weak, but not strong tetanisation in rat hippocampal CA1 neurons. Neurosci. Lett. 192: 185-188, 1995.[Medline]
• BLAUSTEIN, M. P., RATZLAFF, R. W., SCHWEITZER, E. S. Control of intracellular calcium in presynaptic nerve terminals. Federation Proc. 39: 2790-2795, 1980.[Medline]
• BRORSON, J. R., BLEAKMAN, D., GIBBONS, S. J., MILLER, R. J. The properties of intracellular calcium stores in cultured rat cerebellar neurons. J. Neurosci. 11: 4024-4043, 1991.[Abstract]
• CARLSON, G. C., SLAWECKI, M. L., KELLER, A. The distribution and activation of Ca²⁺ stores in cultured olfactory bulb neurons. Soc. Neurosci. Abstr. 22: 2020, 1996.
• CLAPHAM, D. E. Calcium signalling. Neuron 80: 259-268, 1995.
• DENT, M. A. R., RAISMAN, G., LAI, F. A. Expression of type 1 inositol 1,4,5-trisphosphate receptor during axogenesis and synaptic contact in the central and peripheral nervous system of developing rat. Development 122: 1029-1039, 1996.[Abstract]
• EHRLICH, B. E., KAFTAN, E., BEZPROZVANNAYA, S., BEZPROZVANNY, I. The pharmacology of intracellular Ca²⁺ -release channels. Trends Pharmacol. Sci. 15: 145-149, 1994.[Medline]
• ENNIS, M., ZIMMER, L. A., SHIPLEY, M. T. Olfactory nerve stimulation activates rat mitral cells via NMDA and non-NMDA receptors in vitro. NeuroReport 7: 989-992, 1996.[Medline]
• GALIONE, A. Ca²⁺-induced Ca²⁺ release and its modulation by ADP-ribose. Trends Pharmacol. Sci. 13: 304-306, 1992.[Medline]
• GEILING, H., SCHILD, D. Glutamate-mediated release of Ca²⁺ in mitral cells of the olfactory bulb. J. Neurophysiol. 76: 563-570, 1996.[Abstract/Free Full Text]
• HENZI, V., MACDERMOTT, A. B. Characteristics and function of Ca²⁺ and inositol 1,4,5-releasable stores of Ca²⁺ in neurons. Neuroscience 46: 251-273, 1992.[Medline]
• IRVING, A. J., COLLINGRIDGE, G. L., SCHOFIELD, J. G. Interactions between Ca²⁺ mobilizing mechanisms in cultured rat cerebellar granule cells. J. Physiol. Lond. 456: 667-680, 1992.[Abstract/Free Full Text]
• JAFFE, D. B., ROSS, W. N., LISMAN, J. E., LASSERROSS, N., MIYAKAWA, H., JOHNSTON, D. A model for dendritic Ca²⁺ accumulation in hippocampal pyramidal neurons based on fluorescence imaging measurements. J. Neurophysiol. 71: 1065-1077, 1994.[Abstract/Free Full Text]
• KAO, J. P. Practical aspects of measuring [Ca²⁺] with fluorescent indicators. Methods Cell Biol. 40: 1994.
• KASONO, K., HIRANO, T. Involvement of inositol trisphosphate in cerebellar long-term depression. NeuroReport 6: 569-572, 1995.[Medline]
• KATZ, B. The Release of Neural Transmitter Substances. Liverpool: Liverpool UP, 1969.
• KAWAI, T., WATANABE, M. Effects of ryanodine on the spike after-hyperpolarization in sympathetic neurons of the rat superior cervical ganglion. Pflügers Arch. 413: 470-475, 1989.[Medline]
• KIM, S. U. Light and electron microscope study of neurons and synapses in neonatal mouse olfactory bulb cultured in vitro. Exp. Neurol. 36: 336-349, 1972.[Medline]
• KIRISCHUK, S., SCHERER, J., MÖLLER, T., VERKHRATSKY, A., KETTENMANN, H. Subcellular heterogeneity of voltage-gated Ca²⁺ channels in cells of the oligodendrocyte lineage. Glia 13: 1-12, 1995.[Medline]
• KOHDA, K., INOUE, T., MIKOSHIBA, K. Ca²⁺ release from Ca²⁺ stores, particularly from ryanodine-sensitive Ca²⁺ stores, is required for the induction of LTD in cultured cerebellar purkinje cells. J. Neurophysiol. 74: 2184-2188, 1995.[Abstract/Free Full Text]
• KOMURO, H., RAKIC, P. Dynamics of granule cell migration: a confocal microscopic study in acute cerebellar slice preparations. J. Neurosci. 15: 1110-1120, 1995.[Abstract]
• LEVICK, V., COFFEY, H., DMELLO, S. R. Opposing effects of thapsigargin on the survival of developing cerebellar granule neurons in culture. Brain Res. 676: 325-335, 1995.[Medline]
• LLINÁS, R. R. The role of the intrinsic electrophysiological properties of central neurones in oscillation and resonance. In: Cell to Cell Signalling: From Experiments to Theoretical Models., . New York: Academic, 1990, p. 3-16
• MAYER, M. L., WESTBROOK, G. L. The physiology of excitatory amino acids in the vertebrate central nervous system. Prog. Neurobiol. 28: 197-276, 1987.[Medline]
• MILANI, D., FACCI, L., GUIDOLIN, D., LEON, A., SKAPER, S. D. Activation of polyphosphoinositide metabolism as a signal-transducing system coupled to excitatory amino acid receptors in astroglial cells. Glia 2: 161-169, 1989.[Medline]
• MILLER, S., KESSLAK, J. P., ROMANO, C., COTMAN, C. W. Roles of metabotropic glutamate in brain plasticity and pathology. Ann. NY Acad. Sci. 10: 460-474, 1995.
• O'NEIL, S. C., DONOSO, P., EISNER, D. A. The role of [Ca²⁺]_i and Ca²⁺ sensitization in the caffeine contracture of rat myocytes: measurement of [Ca²⁺]_i and [caffeine]_i. J. Physiol. (Lond.) 425: 55-70, 1990.[Abstract/Free Full Text]
• PARKER, I., IVORRA, I. Caffeine inhibits inositol triphosphate-mediated liberation of intracellular calcium in xenopus oocytes. J. Physiol. (Lond.) 433: 229-240, 1990.[Abstract/Free Full Text]
• PENG, Y. Y. Ryanodine-sensitive component of calcium transients evoked by nerve firing at presynaptic nerve terminals. J. Neurosci. 16: 6703-6712, 1996.[Abstract/Free Full Text]
• PUTNEY, J. W. Capacitative calcium entry revisited. Cell Calcium 11: 611-624, 1990.[Medline]
• REGEHR, W. G., CONNOR, J. A., TANK, D. W. Optical imaging of calcium accumulation in hippocampal pyramidal cells during synaptic activation. Nature 341: 533-536, 1989.[Medline]
• REYES, M., STANTON, P. K. Induction of hippocampal long-term depression requires release of Ca²⁺ from separate presynaptic and postsynaptic intracellular stores. J. Neurosci. 16: 5951-5960, 1996.[Abstract/Free Full Text]
• SEYMOUR-LAURENT, K. J., BARISH, M. E. Inositol 1,4,5-trisphosphate and ryanodine receptor distributions and patterns of acetylcholine- and caffeine-induced calcium release in cultured mouse hippocampal neurons. J. Neurosci. 15: 2592-2608, 1995.[Abstract]
• SHARP, A. H., DAWSON, T. M., ROSS, C. A., FOTUHI, M., MOUREY, R. J., SNYDER, S. H. Inositol 1,4,5-triphosphate receptors: immunohistochemical localization to discrete areas of rat central nervous system. Neuroscience 53: 927-942, 1993.[Medline]
• SHENG, M., GREENBERG, M. E. The regulation and function of c-fos and other immediate early genes in the nervous system. Neuron 4: 477-485, 1990.[Medline]
• SHEPHERD, G. M. Olfactory bulb. In: The Synaptic Organization of the Brain, edited by and G. M. Shepherd . New York: Oxford Univ. Press, 1990, p. 133-169
• SHIPLEY, M. T., ENNIS, M. Functional organization of olfactory system. J. Neurobiol. 30: 123-176, 1996.[Medline]
• SHIPLEY, M. T., MCLEAN, J. H., ZIMMER, L. A., ENNIS, M. The olfactory system. In: Handbook of Chemical Neuroanatomy, Vol. 12: Integrated Systems of the CNS, Part III, edited by A. Björklund, T. Hökfelt, and L. W. Swanson . Amsterdam: Elsevier, 1996, p. 467-571
• SHMIGOL, A., KIRISCHUK, S., KOSTYUK, P., VERKHRATSKY, A. Different properties of caffeine-sensitive Ca²⁺ stores in peripheral and central mammalian neurones. Pflügers Arch. 426: 174-176, 1994.[Medline]
• SIMPSON, P. B., CHALLISS, R. A. J., NAHORSKI, S. R. Neuronal Ca²⁺ stores: activation and function. Trends Neurosci. 18: 299-306, 1995.[Medline]
• SLADECZEK, F., PIN, J. P., RECASENS, J. B., WEISS, S. Glutamate stimulates inositol phosphate formation in striatal neurons. Nature 317: 717-719, 1985.[Medline]
• SLAWECKI, M. L., CARLSON, G. C., AND KELLER, A. Differential distribution of IP₃ receptors in rat olfactory bulb. J. Comp. Neurol. in press.
• STEWART, R., ERSKINE, L., MCCAIG, C. D. Calcium channel subtypes and intracellular calcium stores modulate electric field-stimulated and-oriented nerve growth. Dev. Biol. 171: 340-351, 1995.[Medline]
• SULLIVAN, K. F., HAVERCROFT, J. C., MACHLIN, P. S., CLEVELAND, D. W. Sequence and expression of the chicken beta^5- and beta^4- tubulin genes define a pair of divergent beta-tubulins with complementary patterns of expression. Mol. Cell. Biol. 6: 4409-4418, 1986.[Abstract/Free Full Text]
• TANABE, Y., MASU, M., ISHII, T., SHIGEMOTO, R., NAKANISHI, S. A family of metabotropic glutamate receptors. Neuron 8: 169-179, 1992.[Medline]
• TANI, A., YASHIHARA, Y., MORI, K. Increase in cytoplasmic free Ca²⁺ elicited by noradrenalin and serotonin in cultured local interneurons of mouse olfactory bulb. Neuroscience 49: 193-199, 1992.[Medline]
• THASTRUP, O., CULLEN, P. J., DROBAK, B. K., HANLEY, M. R., DAWSON, A. P. Thapsigargin, a tumor promotor, discharges intracellular Ca²⁺ stores by specific inhibition of the endoplasmic reticulum Ca²⁺ ATPase. Proc. Natl. Acad. Sci. USA 87: 2466-2470, 1990.[Abstract/Free Full Text]
• TROMBLEY, P. Q. Noradrenergic modulation of synaptic transmission between olfactory bulb neurons in culture: implications to olfactory learning. Brain Res. Bull. 35: 473-484, 1994.[Medline]
• TROMBLEY, P. Q., WESTBROOK, G. L. Excitatory synaptic transmission in cultures of rat olfactory bulb. J. Neurophysiol. 64: 598-606, 1990.[Abstract/Free Full Text]
• VAN DEN POL, A. Presynaptic metabotropic glutamate receptors in adult and developing neurons: autoexcitation in the olfactory bulb. J. Comp. Neurol. 253-271: 1995.
• VERKHRATSKY, A., KETTENMANN, H. Calcium signalling in glial cells. Trends Neurosci. 19: 346-352, 1996.[Medline]
• WOJCIKIEWICZ, R. J. H. Type I, II, and III inositol 1,4,5-triphosphate receptors are unequally susceptible to down-regulation and are expressed in markedly different proportions in different cell types. J. Biol. Chem. 270: 11678-11683, 1995.[Abstract/Free Full Text]
• WOJCIKIEWICZ, R. J. H., OBERDORF, J. A. Degradation of inositol 1,4,5-trisphosphate receptors during cell stimulation is a specific process mediated by cysteine protease activity. J. Biol. Chem. 271: 16652-16655, 1996.[Abstract/Free Full Text]

0022-3077/97 $5.00 Copyright ©1997 The American Physiological Society