Histamine-induced calcium entry in rat cerebellar astrocytes: evidence for capacitative and non-capacitative mechanisms

Abstract

1. We have investigated the effects of histamine on the intracellular calcium concentration ([Ca²⁺]_i) of cultured rat cerebellar astrocytes using fura-2-based Ca²⁺ imaging microscopy.

2. Most of the cells responded to the application of histamine with an increase in [Ca²⁺]_i which was antagonized by the H₁ receptor blocker mepyramine. When histamine was applied for several minutes, the majority of the cells displayed a biphasic Ca²⁺ response consisting of an initial transient peak and a sustained component. In contrast to the initial transient [Ca²⁺]_i response, the sustained, receptor-activated increase in [Ca²⁺]_i was rapidly abolished by chelation of extracellular Ca²⁺ or addition of Ni²⁺, Mn²⁺, Co²⁺ and Zn²⁺, but was unaffected by nifedipine, an antagonist of L-type voltage-activated Ca²⁺ channels. These data indicate that the sustained increase in [Ca²⁺]_i was dependent on Ca²⁺ influx.

3. When intracellular Ca²⁺ stores were emptied by prolonged application of histamine in Ca²⁺-free conditions, Ca²⁺ re-addition after removal of the agonist did not lead to an ‘overshoot’ of [Ca²⁺]_i indicative of store-operated Ca²⁺ influx. However, Ca²⁺ stores were refilled despite the absence of any substantial change in the fura-2 signal.

4. Depletion of intracellular Ca²⁺ stores using cyclopiazonic acid in Ca²⁺-free saline and subsequent re-addition of Ca²⁺ to the saline resulted in an increase in [Ca²⁺]_i that was significantly enhanced in the presence of histamine.

5. The results suggest that besides capacitative mechanisms, a non-capacitative, voltage-independent pathway is involved in histamine-induced Ca²⁺ entry into cultured rat cerebellar astrocytes.

Changes in the intracellular free calcium concentration ([Ca²⁺]_i) mediate a variety of biological responses in both excitable and non-excitable cells. In electrically non-excitable cells, many plasmalemmal receptors are coupled to the phospholipase C (PLC)-inositol-1,4,5-trisphosphate (InsP₃) pathway (Berridge, 1993; Clapham, 1995). Stimulation of these receptors releases Ca²⁺ from intracellular Ca²⁺-storing organelles, such as the endoplasmic reticulum (ER). The signal transduction cascade that links receptor activation to Ca²⁺ release has been thoroughly investigated (for reviews see Irvine, 1992; Berridge, 1993). In short, Ca²⁺ release from intracellular stores is induced by the InsP₃-mediated opening of Ca²⁺ channels present in the membrane of the ER Ca²⁺ stores. The soluble messenger InsP₃ is generated by the PLC-dependent hydrolysis of membrane-bound phosphatidyl-inositol-4,5-bisphosphate (PIP₂). InsP₃-mediated Ca²⁺ release from intracellular stores is typically accompanied by a substantial Ca²⁺ influx from the extracellular space (Meldolesi et al. 1991; Parekh & Penner, 1997; Putney, 1997).

In many cells, InsP₃-mediated depletion of intracellular Ca²⁺ stores stimulates Ca²⁺ influx via a pathway that is commonly referred to as the ‘capacitative’ or ‘store-operated’ Ca²⁺ entry pathway (for review see Putney, 1997). The plasmalemmal Ca²⁺ channels that are part of this pathway are opened in response to a decrease in the Ca²⁺ concentration in the lumen of the ER. Important for the concept of store-operated Ca²⁺ entry is that it is essentially independent of receptor activation. Any means of depleting the intracellular Ca²⁺ pool provides a full and sufficient signal for activation of Ca²⁺ entry, even in the absence of receptor activation or generation of InsP₃. This can be demonstrated using specific membrane-permeant inhibitors of the intracellular Ca²⁺ transport ATPases that mediate the active sequestration of Ca²⁺ into the ER. Application of these inhibitors activates Ca²⁺ entry into the cytoplasm after passive depletion of intracellular Ca²⁺ stores without a concomitant increase in inositol phosphate levels. A weakness of the capacitative model of Ca²⁺ entry is the fact that despite intensive research the molecular basis of the communication between the intracellular Ca²⁺ stores and the plasma membrane Ca²⁺ channels is still unknown (Putney, 1997; Holda et al. 1998).

Although current models of receptor-activated Ca²⁺ entry focus on the capacitative mechanism, there is increasing evidence that the capacitative pathway is not the only pathway through which Ca²⁺ can enter cells in response to receptor activation (for review see Clementi & Meldolesi, 1996; Barritt, 1999). Channels regulated independently of store depletion by second messengers such as InsP₃ (Kuno & Gardner, 1987; Vaca & Kunze, 1995), inositol-1,2,3,4-tetrakisphosphate (InsP₄; Lückhoff & Clapham, 1992), G-proteins (Nie et al. 1998), Ca²⁺ (Loirand et al. 1991), diacylglycerol (DAG; Helliwell & Large, 1997), protein kinase C (PKC; Oike et al. 1993), or arachidonic acid (Shuttleworth & Thompson, 1998; Broad et al. 1999) coexist with store-operated channels. Moreover, many stimuli have been reported to activate non-capacitative Ca²⁺ entry via signalling pathways that still need to be more closely defined (Clementi et al. 1992; Felder et al. 1992; Byron & Taylor, 1993; Mathias et al. 1997).

The molecular basis of the diversity of Ca²⁺ entry pathways and the relation of store depletion-sensitive to store depletion-insensitive, Ca²⁺-permeant channels are unclear. However, it is likely that multiple Ca²⁺ influx channels are involved in Ca²⁺ entry initiated by the activation of PLC-coupled receptors.

Glial cells are known to express a wide variety of receptors for neurotransmitters and hormones, the majority of which are coupled to cytosolic [Ca²⁺] increases via the PLC-InsP₃ pathway (Verkhratsky et al. 1998; Deitmer et al. 1998). Although capacitative Ca²⁺ entry seems to be operational in different types of glial cells (Hildebrandt & Hildebrandt, 1997; Wu et al. 1997; Hartmann & Verkhratsky, 1998; Rzigalinski et al. 1999), the precise relationship between receptor activation, Ca²⁺ store depletion, and activation of Ca²⁺ entry has not been explored.

In the present study, we investigated the interrelations between histamine receptor activation, Ca²⁺ release from intracellular stores and Ca²⁺ entry across the plasma membrane in cultured rat cerebellar astrocytes. Histamine is a neurotransmitter widely distributed in the mammalian central nervous system. The actions of histamine are mediated by three subtypes of receptor, H₁, H₂ and H₃, where H₃ receptors are autoreceptors present on histamine-releasing neurones (for review see Hill et al. 1997). At the cellular level, H₁ and H₂ receptors have been identified not only on neurones, but also on astrocytes and blood vessels. Some authors postulate that glial cells might be a major target of central histamine release (Inagaki & Wada, 1994). Hence, the effects of histamine on different glial cells have been extensively studied. Histamine has been shown to stimulate phosphoinositide turnover in cultured rat cerebral type 2 astrocytes (Kondou et al. 1991) and human U373 MG astrocytoma cells (Arias-Montaño et al. 1994). H₁ receptor-mediated increases in [Ca²⁺]_i have been reported both in vitro, e.g. for C6 glioma cells (Weiger et al. 1997), cultured rat cerebral astrocytes (Inagaki et al. 1991; Shao & McCarthy, 1993) or human UC-11MG astrocytes (Lucherini & Gruenstein, 1992), and in situ (Bernstein et al. 1996; Kirischuk et al. 1996). Most of the studies focused on whether histamine was able to produce an effect in the cells under investigation, without further characterizing the response. Histamine-induced Ca²⁺ entry has been described by some authors (Arbonés et al. 1990; Fukui et al. 1991; Lucherini & Gruenstein, 1992). However, the mechanisms underlying histamine-induced Ca²⁺ entry in glial cells have not been explored to date. In this study, we were able to demonstrate that besides capacitative mechanisms, a store-independent, voltage-insensitive pathway is involved in histamine-induced Ca²⁺ entry into cultured rat cerebellar astrocytes. Preliminary results have been communicated to the German Neurobiology Conference (Jung & Deitmer, 1999).

METHODS

Cell culture

Astrocyte-enriched primary cultures were prepared from cerebellar hemispheres of newborn rats (postnatal day (P)1–2) by a method similar to that described by Fischer (1984). Briefly, rats were killed by decapitation (according to the German Tierschutzgesetz) and the cerebella removed and placed in an isotonic salt solution (mm: NaCl 120, KCl 5.4, MgCl₂ 0.8, Tris-HCl 25 and D-glucose 15, pH 7.2). After the meninges had been removed, the cerebella were dissociated by trypsinization (0.1 % trypsin) and mechanical trituration. The cells were incubated with deoxyribonuclease, centrifuged, resuspended in medium A (Dulbecco's modified Eagle's medium supplemented with 0.1 % bovine serum albumin, 10 μg l⁻¹ epidermal growth factor, 10 mg l⁻¹ insulin, 1 g l⁻¹ streptomycin, 1000 units l⁻¹ penicillin and 99 mg l⁻¹ transferrin), plated in culture flasks coated with poly-D-lysine, and incubated at 37°C in a humidified atmosphere with 7 % CO₂. After the cell layer had reached confluency, oligodendrocytes and macrophages were removed by shaking the culture flasks. The remaining cells were detached from the bottom of the culture flasks by exposure to 0.1 % trypsin, diluted in culture medium, reseeded on glass coverslips coated with poly-D-lysine (0.001 %) and incubated at 37°C in a humidified incubator with 7 % CO₂. Trypsinization steps were stopped by adding fetal calf serum (10 %)-containing solution. All experiments were performed at room temperature (approximately 20–22°C) between 3 and 17 days after plating of the astrocytes. Cultures showed > 95 % staining for the astrocyte-specific marker glial fibrillary acidic protein (GFAP).

Solutions

During the experiments, the cells were continuously superfused (flow approximately 12.5 ml min⁻¹) with an extracellular solution containing (mm): NaCl 145, KCl 5, CaCl₂ 2, MgCl₂ 1, D-glucose 10 and Hepes 10 (pH adjusted to 7.4 with NaOH). Ca²⁺-free extracellular solution was prepared by replacing CaCl₂ with equimolar amounts of MgCl₂ and by adding 0.5 or 1 mm EGTA.

Histamine dihydrochloride, mepyramine, cyclopiazonic acid, 4-bromo-A23187, nifedipine, LaCl₃, MnCl₂, CoCl₂ and NiCl₂ were purchased from Sigma. U73122 was obtained from Biomol, SKF96365 from Calbiochem, thioperamide maleate and tiotidine from Tocris, and fura-2 AM from Molecular Probes.

Imaging and data analysis

Intracellular calcium was monitored using the fluorescent Ca²⁺ indicator fura-2 AM. Cultures were loaded in extracellular solution (for composition see above) supplemented with 2–3 μm fura-2 AM for 30–60 min at room temperature. Cultures were then rinsed with extracellular solution and allowed to de-esterify for at least 30 min before use. Measurements of [Ca²⁺]_i of single cells were performed using an inverted fluorescence microscope (Diaphot, Nikon, Tokyo, Japan) equipped with a dual excitation fluorometric imaging system (PTI, Wedel, Germany). The illumination was generated by a 75 W xenon bulb. The excitation wavelength was alternated between 340 and 380 nm, and the emission fluorescence of selected areas within the astrocytic monolayer was passed through a 480 nm long-pass filter and recorded with a video camera (SIT C-2400, Hamamatsu, Garching, Germany). Monochromator settings, chopper frequency and complete data acquisition were controlled by ImageMaster software (PTI). The sampling rate was 0.2 or 0.5 Hz. At the end of each experiment, cells were permeabilized using the calcium ionophore 4-bromo-A23187 (20 μm) and the background fluorescence determined after exposure of the cells to an extracellular solution containing 2 mm Mn²⁺ instead of Ca²⁺ in order to quench the fura-2 fluorescence. Background fluorescence was subtracted from the raw signals at each excitation wavelength before calculation of the fluorescence ratio R =F₃₄₀/F₃₈₀.

Results are given either as the ratio of the fluorescence intensities at the different wavelengths (340 nm/380 nm) or as the approximate [Ca²⁺]_i, calculated from the 340 nm/380 nm fluorescence values as described by Grynkiewicz et al. (1985) using the following equation:

The values R_min and R_max (fluorescence ratios of free and Ca²⁺-bound fura-2) and the constants S_f2 and S_b2 (fluorescence of free and Ca²⁺-bound fura-2 at 380 nm) were determined using an in vivo calibration procedure as follows. Cells were permeabilized using the calcium ionophore 4-bromo-A23187 (20 μm) in order to allow [Ca²⁺]_i to equilibrate with [Ca²⁺]_o. R_min and S_f2, and R_max and S_b2 were obtained in the absence (0 mm Ca²⁺, 10 mm EGTA) or presence (10 mm Ca²⁺) of extracellular calcium, respectively. Correction for background fluorescence was made at the end of each calibration experiment as described above. The dissociation constant K_d of the fura-2-Ca²⁺ complex was assumed to be 225 nm in the cytosolic environment (Grynkiewicz et al. 1985). In some sets of experiments the fura-2 ratio responses were not calibrated, but compared relative to each other (see Figs 4B, 5, 6 and 7).

Figure 4. [Ca²⁺]_i increases induced by Ca²⁺ removal/Ca²⁺ re-addition protocols are unaffected by prior application of histamine.

A, cytosolic Ca²⁺ changes upon removal (0 mm Ca²⁺, 0.5 mm EGTA) and re-addition (2 mm Ca²⁺) of external Ca²⁺ with (dotted trace) or without (continuous trace) prior application of histamine (HA, 100 μm). Each trace is the mean of 19 experiments. The experimental protocol is illustrated by representative traces in the insets above (left, with histamine; right, without histamine). Histamine was applied either for 2 min or for 6 min. B, Ca²⁺ transients monitored as fura-2 ratio induced by histamine in the absence (left) and in the presence (right) of external Ca²⁺, showing that the Ca²⁺ stores refilled after re-addition of external Ca²⁺.

Figure 5. Manganese reversibly inhibits Ca²⁺ entry.

Ca²⁺ responses monitored as fura-2 ratio. The sustained component evoked by histamine and by re-addition of external Ca²⁺ in the presence of cyclopiazonic acid (CPA) was reversibly reduced by 1 mm Mn²⁺.

Figure 6. Capacitative and non-capacitative Ca²⁺ entry in store-depleted cells (1).

Capacitative Ca²⁺ entry following repeated removal and re-addition of external Ca²⁺ to cells treated with CPA (10 μm). A, two consecutive responses to Ca²⁺ re-addition after treatment with CPA in a Ca²⁺-free solution. B, similar protocol to that in A, but with the first Ca²⁺ step in the presence of histamine. C, statistical analysis of all experiments. The amplitudes of the cytosolic [Ca²⁺] increases upon the second Ca²⁺ re-addition were normalized to the second [Ca²⁺]_i increase, which was taken as 100 %. When compared to control (1st entry), histamine produced a significant (* P < 0.01) enhancement of the amplitude of the first [Ca²⁺]_i increase.

Figure 7. Capacitative and non-capacitative Ca²⁺ entry in store-depleted cells (2).

Capacitative Ca²⁺ entry following repeated removal and re-addition of external Ca²⁺ to cells treated with CPA (10 μm). A, two consecutive responses to Ca²⁺ re-addition after treatment with CPA in a Ca²⁺-free solution. B, similar protocol to that in A, but with the second Ca²⁺ step in the presence of histamine. C, statistical analysis of all experiments. The amplitudes of the cytosolic [Ca²⁺] increases upon the second Ca²⁺ re-addition were normalized to the first [Ca²⁺]_i increase of the protocol which was taken as 100 %. When compared to control (2nd entry), histamine produced a significant (*P < 0.001) enhancement of the amplitude of the second [Ca²⁺]_i increase.

Statistics

All data are given as means ±s.e.m. The statistical significance of differences between mean values was assessed using Student's t test. Differences were regarded as statistically significant for P < 0.05.

RESULTS

Histamine-induced [Ca²⁺]_i increase in cultured rat cerebellar astrocytes

Ratiometric imaging with the Ca²⁺-sensitive dye fura-2 was used to monitor [Ca²⁺]_i in cultured rat cerebellar astrocytes. The mean resting [Ca²⁺]_i was 31 ± 1 nm (n = 415 experiments). Upon application of 100 μm histamine, approximately 90 % of the cells tested exhibited an increase in [Ca²⁺]_i as shown in Fig. 1A. The [Ca²⁺]_i responses triggered by 100 μm histamine varied depending on the duration of agonist application. Short applications of about 5 s resulted in a transient elevation of [Ca²⁺]_i (Fig. 1A, left panel, n = 264). A fast initial increase in [Ca²⁺]_i was followed by a slower decline to the resting level.

Figure 1. Histamine-induced [Ca²⁺]_i increases in cultured astrocytes.

A, response patterns obtained with varying durations of agonist application: transient response to short application of 100 μm histamine (HA, 5 s, left panel); biphasic (middle panel) or oscillatory responses (right panel) upon longer applications of histamine. The panels show data from three different cells. B, concentration dependence of the histamine-induced [Ca²⁺]_i increases. Because response amplitudes decline upon repetitive stimulation with the same agonist concentration, the dose-response relation was determined using the following protocol. Two applications of a single test concentration of histamine (between 0.3 and 1000 μm) were separated by an application of 100 μm histamine. The duration of each application was 5 s, the interval between successive histamine applications 3 min. Peak amplitudes of the responses to the test concentrations were normalized to the response to 100 μm histamine and averaged. A fit to the dose-response curve yielded an EC₅₀ value of 6.4 ± 0.8 μm and a maximally effective histamine concentration of 100 μm. Error bars indicate s.e.m.; the number of experiments (n) is indicated above the error bars.

When the application of histamine was prolonged (2–6 min), three different response patterns could be distinguished. The majority of the cells (66 %) showed a biphasic Ca²⁺ response consisting of an initial, large transient component and a smaller sustained component (Fig. 1A, middle panel). A monophasic, transient [Ca²⁺]_i increase was exhibited by 28 % of the cells, where [Ca²⁺]_i, after reaching a peak, monotonically returned to prestimulation levels, even in the continued presence of the agonist (data not shown). Finally, in 6 % of the cells, the initial [Ca²⁺]_i peak was followed by [Ca²⁺]_i oscillations (Fig. 1A, right panel). Mean peak amplitudes of histamine-induced [Ca²⁺]_i increases were similar, irrespective of the duration of the histamine application or the response pattern exhibited. They ranged from 30 nm to 8.4 μm, being on average 1.2 ± 0.1 μm (n = 270).

Histamine elicited [Ca²⁺]_i elevations in a concentration-dependent manner as shown in Fig. 1B The threshold for histamine-induced [Ca²⁺]_i increases was approximately 1 μm and the responses saturated at 100 μm. A fit to the dose-response curve yielded an EC₅₀ value of 6.4 ± 0.8 μm and a Hill coefficient of 1.83. The latter does not necessarily indicate a 1:2 receptor:ligand stoichiometry, since several transduction steps are involved between agonist binding and the observed [Ca²⁺]_i increase.

Ca²⁺ release versus Ca²⁺ entry

The majority of the astrocytes responded to prolonged application (2–6 min) of 100 μm histamine like the cell shown in the middle panel of Fig. 1A. The response pattern was biphasic and characterized by an initial peak that evolved into a slowly declining plateau. The sustained component of the response persisted until receptor occupancy was terminated by removing the agonist from the extracellular solution.

To test whether the histamine-induced [Ca²⁺]_i response resulted from Ca²⁺ influx across the plasma membrane or from intracellular Ca²⁺ release, we removed extracellular Ca²⁺ prior to, or during, histamine application (Fig. 2A and B). As in other non-excitable cells (Matsumoto et al. 1986; Kotlikoff et al. 1987), the transient [Ca²⁺]_i increase in response to short applications of histamine, as well as the initial peak component of the biphasic [Ca²⁺]_i increase observed upon prolonged application of the agonist, were insensitive to the removal of extracellular Ca²⁺ (Fig. 2B) and, thus, were probably due to release of Ca²⁺ from intracellular stores. As has been described for various cell types including astrocytes from different brain regions (Arbonés et al. 1988; Fukui et al. 1991; Kondou et al. 1991; Marmy et al. 1993), this histamine-induced Ca²⁺ release was mediated by H₁ receptors that are coupled to the PLC-InsP₃ pathway (Table 1). We also studied the effects of selective inhibitors of the different histamine receptor subtypes on the plateau component of the histamine-induced Ca²⁺ response. The sustained [Ca²⁺]_i rise observed in the continued presence of histamine was fully inhibited by the H₁ receptor antagonist mepyramine (2 μm; Fig. 3A), whereas neither the H₂ receptor blocker tiotidine (10 μm) nor the H₃ receptor antagonist thioperamide (1 μm) had any measurable effect on either the transient or the sustained component of the histamine-induced [Ca²⁺]_i response (Table 1). Hence, both the sustained [Ca²⁺]_i plateau and the transient component are exclusively dependent on H₁ receptor activation.

Figure 2. The sustained component of the [Ca²⁺]_i response to prolonged application of histamine is due to Ca²⁺ entry.

A, effect of removal and re-addition of extracellular Ca²⁺ during the sustained phase of the histamine-induced [Ca²⁺]_i response. B, response to a prolonged application of histamine in Ca²⁺-free extracellular solution. Re-addition of Ca²⁺ in the continuous presence of the agonist induced a sustained increase in [Ca²⁺]_i. C, reversible inhibition of the plateau phase of the histamine-induced [Ca²⁺]_i increase by Ni²⁺ (1 mm). D, effect of nifedipine (10 μm), an inhibitor of L-type voltage-activated Ca²⁺ channels, on the histamine-induced Ca²⁺ plateau.

Table 1.

Pharmacology of the histamineinduced Ca²⁺ influx

Substance	Concentration (μm)	Inhibition of HAinduced Ca2+ release	n	Inhibition of HAinduced Ca2+ influx	n	Inhibition of CPAinduced Ca2+ influx	n
Mn2+	1000	−	9	+	8	+	23
Ni2+	1000	n.d.	—	+	11	+	19
Co2+	1000	n.d.	—	+	—	+	7
Zn2+	1000	n.d.	—	+	12	+	16
Nifedipine	10	n.d.	—	−	8	n.d.	—
Mepyramine	1–5	+	14	+	4	n.d.	—
Tiotidine	10	−	12	−	12	n.d.	—
Thioperamide	1	−	12	−	9	n.d.	—
SKF96365	100	n.d.	—	+	6	n.d.	—
U73122	5	+	11	+	10	n.d.	—

The table summarizes the effects of various substances on the release and on the influx component of the histamine (HA)-induced Ca²⁺ response. + indicates full inhibition of the response whereas − denotes no inhibition; n.d., not determined. CPA, cyclopiazonic acid.

Figure 3. Pharmacological characterization of the sustained component of the histamine-induced [Ca²⁺]_i response.

A, inhibition of the plateau component of the histamine-induced [Ca²⁺]_i increase by the H₁ receptor antagonist mepyramine (2 μm). B, abolition of the plateau component of the histamine-induced [Ca²⁺]_i increase by the phospholipase C inhibitor U73122 (5 μm).

Extracellular Ca²⁺ is required for the generation and maintenance of the sustained [Ca²⁺]_i plateau during agonist stimulation (Fig. 2A and B). When external Ca²⁺ was removed during the plateau phase of the response, [Ca²⁺]_i returned to resting levels (n = 5,Fig. 2A), even in the continued presence of the agonist. After re-addition of Ca²⁺ to the extracellular solution, [Ca²⁺]_i returned to the plateau level. When histamine was administered in the absence of extracellular Ca²⁺, the cells exhibited only the transient [Ca²⁺]_i peak without a plateau phase (n = 11,Fig. 2B). A sustained increase in [Ca²⁺]_i developed only after re-addition of extracellular Ca²⁺.

Our data suggest that the biphasic [Ca²⁺]_i responses induced by 100 μm histamine can be separated into an initial transient component that is primarily due to Ca²⁺ release, and a plateau component for which Ca²⁺ entry across the plasma membrane is necessary.

The requirement of continued PLC activity for histamine-induced Ca²⁺ entry is suggested by the ability of the PLC inhibitor U73122 (5 μm) to block the Ca²⁺ influx component (n = 10,Fig. 3B and Table 1).

Pharmacology of Ca²⁺ entry

To further investigate whether the sustained component of the [Ca²⁺]_i increase in response to prolonged application of histamine was indeed due to Ca²⁺ influx, we tested several polyvalent cations for their ability to block the plateau component of the response. In the presence of 1 mm Ni²⁺ (n = 11), 1 mm Co²⁺ (n = 8) or 1 mm Zn²⁺ (n = 12), the histamine-induced [Ca²⁺]_i plateau was completely and reversibly suppressed (Fig. 2C and Table 1). La³⁺ at concentrations of 0.1–3 mm also reduced the sustained component of the histamine-induced [Ca²⁺]_i response, although somewhat less reliably than the other metal ions tested (n = 20, data not shown). Since polyvalent cations are known to block many different Ca²⁺ entry pathways, the above results strongly suggest that the plateau component of the histamine-induced [Ca²⁺]_i response is mediated by calcium influx from the extracellular solution. SKF96365 (100 μm), a blocker of voltage-activated as well as receptor-mediated Ca²⁺ influx, also reversibly inhibited the histamine-induced Ca²⁺ entry (n = 6,Table 1).

Several reports have described the expression of voltage-activated Ca²⁺ channels in astrocytes (for review see Sontheimer, 1994). We tested whether histamine-induced Ca²⁺ entry was mediated by voltage-gated Ca²⁺ channels. Depolarization-induced [Ca²⁺]_i increases could only rarely be observed in this cell type (< 10 % of cells). In the few cases in which a rise in [Ca²⁺]_i developed in response to depolarization by elevated extracellular potassium concentration (50 or 100 mm K⁺), it could be completely inhibited by the L-type Ca²⁺ channel blocker nifedipine. As shown in Fig. 2D, nifedipine (10 μm) did not affect the [Ca²⁺]_i plateau (n = 8). These data suggest that the agonist-induced Ca²⁺ entry is not mediated by voltage-sensitive Ca²⁺ channels.

Capacitative versus non-capacitative mechanisms

We used classical Ca²⁺ removal/Ca²⁺ re-addition protocols to evaluate the role of Ca²⁺ store depletion in mediating Ca²⁺ entry in the prolonged presence of histamine. In the absence of extracellular Ca²⁺, cells were exposed to histamine to deplete the agonist-sensitive intracellular Ca²⁺ store. After removal of the agonist, Ca²⁺ was re-added to the extracellular saline solution, and a concomitant [Ca²⁺]_i increase was observed (Fig. 4A, left inset). This increase in [Ca²⁺]_i was compared to the [Ca²⁺]_i increases in cells exposed to Ca²⁺-free saline alone (Fig. 4A, right inset). Control experiments confirmed that prolonged application of histamine in Ca²⁺-free saline (2–6 min) emptied Ca²⁺ stores completely, because subsequent histamine applications in Ca²⁺-free saline were unable to elicit [Ca²⁺]_i increases (data not shown). It should be pointed out that Ca²⁺ removal alone did not deplete internal stores, because histamine-induced [Ca²⁺]_i increases could still be elicited after cells had been exposed to Ca²⁺-free saline for more than 15 min (data not shown). Statistical analysis of the [Ca²⁺]_i elevations upon Ca²⁺ re-addition with or without prior depletion of the stores following agonist application did not reveal any significant differences in the amplitude and kinetics of the [Ca²⁺]_i increase (Fig. 4A, n = 19). However, [Ca²⁺]_i increases upon histamine application could again be observed after re-addition of Ca²⁺ to the saline (Fig. 4B). Hence, Ca²⁺ uptake into the empty stores must have occurred in the absence of any overshoot in the fura-2 signal.

We used cyclopiazonic acid (CPA, 10 μm), an inhibitor of the ER Ca²⁺-ATPase, as a tool to further investigate the Ca²⁺ entry pathway stimulated in cultured rat cerebellar astrocytes upon prolonged application of histamine. CPA, by preventing the reuptake of Ca²⁺ into the ER, leads to a gradual depletion of the intracellular Ca²⁺ stores and, thus, should activate a store-operated Ca²⁺ entry pathway. In Ca²⁺-free saline, CPA elicited a transient increase in [Ca²⁺]_i (Figs 5–7). Upon re-addition of Ca²⁺ to the extracellular saline, a large increase in [Ca²⁺]_i could be observed suggesting that, after emptying of the intracellular Ca²⁺ stores by CPA, Ca²⁺ entry across the plasma membrane was stimulated as would be expected from the capacitative calcium entry model. Ca²⁺ entry during prolonged agonist exposure as well as following Ca²⁺ re-addition in the presence of CPA was similarly inhibited by 1 mm Mn²⁺ (Fig. 5 and Table 1). There was no significant quenching of the fluorescence signal observed in the presence of Mn²⁺, as also supported by the reversibility of the Mn²⁺ effect, indicating that Mn²⁺ itself does not permeate these Ca²⁺ entry pathways.

In order to discern whether histamine-induced Ca²⁺ entry was entirely mediated by a store-dependent pathway, or whether store-independent Ca²⁺ entry mechanisms were involved, we tested the effect of histamine on CPA-induced Ca²⁺ entry. The following protocol was used. In control cells, intracellular Ca²⁺ stores were emptied by application of CPA in Ca²⁺-free saline. In the continued presence of the (sarco-)endoplasmic reticulum Ca²⁺-ATPase (SERCA) inhibitor, Ca²⁺ was added, removed and returned to the extracellular saline, and the amplitudes of the concomitant [Ca²⁺]_i increases compared. As can be seen in Figs 6A and 7A, after depletion of the intracellular Ca²⁺ stores by CPA in Ca²⁺-free saline, a large [Ca²⁺]_i increase was evoked upon re-addition of Ca²⁺ to the saline. This increase in [Ca²⁺]_i was abolished by chelation of extracellular Ca²⁺ and reoccurred upon addition of Ca²⁺ to the saline. The amplitude of the second rise was, however, significantly smaller than that of the first increase and reduced to a mean [Ca²⁺]_i increase of 89 ± 0.9 % of the amplitude of the first rise (P < 0.001,n = 50). To test whether histamine was able to enhance this [Ca²⁺]_i increase, the protocol was slightly modified such that either the first (Fig. 6B) or the second (Fig. 7B) Ca²⁺ entry phase occurred in the presence of both CPA (10 μm) and histamine (100 μm). Although histamine induced no further Ca²⁺ release from intracellular stores (confirming that the stores were empty), the amplitude of the first [Ca²⁺]_i increase was significantly larger by 16 % than the first, control, [Ca²⁺]_i increase without histamine (n = 43,P < 0.01; Fig. 6B and C). When histamine was present during the second re-addition of external Ca²⁺, the evoked [Ca²⁺]_i rise was significantly enhanced as compared to the second [Ca²⁺]_i increase without histamine (P < 0.001,n = 45; Fig. 7B and C). These experiments suggest that the effects of CPA and histamine are additive, and that in the presence of histamine, additional, store-independent, Ca²⁺ entry pathways are activated. The effect of histamine on Ca²⁺ entry was abolished by the H₁ receptor antagonist mepyramine (2 μm, n = 36, data not shown).

DISCUSSION

In the present study, we were able to demonstrate that prolonged application of histamine results in biphasic calcium signalling in the majority of astrocytes (Fig. 1A, middle panel). The responses were mediated by the H₁ receptor subtype and consisted of an initial, large transient due to Ca²⁺ mobilization from internal storage organelles via the PLC-InsP₃ pathway, followed by a decrease to a prolonged elevated plateau that was mediated by calcium influx across the plasma membrane.

Stimulation of Ca²⁺ influx across the plasma membrane is a key element in transmembrane signalling because it links activation of various receptors on the cell surface to the control of cell function. Three main categories of Ca²⁺-permeable channels have been reported to mediate Ca²⁺ entry from the extracellular space (for review see Barritt, 1999): voltage-operated Ca²⁺ channels, ligand-gated non-specific cation channels (ionotropic receptors) and receptor-activated Ca²⁺ channels that are not part of the receptor molecule. Of all plasma membrane Ca²⁺ channels, the receptor-activated Ca²⁺ channels are the most poorly understood. This incomplete knowledge may be partly because a large number of channel subtypes belong to this category (Parekh & Penner, 1997). Thus, receptor-activated Ca²⁺ channels are further subdivided into store-operated or capacitative Ca²⁺ channels and a number of Ca²⁺-permeable channel types that are regulated independently of store depletion.

There are only a few studies that have investigated Ca²⁺ entry following receptor activation in glial cells. The existence of store-operated channels in glial cells has been demonstrated recently (Fischer et al. 1997; Wu et al. 1997; Hartmann & Verkhratsky, 1998; Rzigalinski et al. 1999). However, experimental data on the properties and regulation of capacitative Ca²⁺ entry mechanisms in glial cells are rather limited. There is one report discussing the importance of intracellular ATP levels for the maintenance of a sustained [Ca²⁺]_i plateau (Wu et al. 1997). More recently, 5,6-epoxyeicostrienoic acid has been suggested to be a component of the calcium influx factor coupling store depletion to Ca²⁺ entry in cultured cortical astrocytes (Rzigalinski et al. 1999). The involvement of non-capacitative Ca²⁺ entry mechanisms has been suggested only for the hyposmolarity-induced Ca²⁺ response in cultured rat astrocytes (Fischer et al. 1997). In this case, nimodipine-sensitive, voltage-dependent Ca²⁺ channels are likely to play a role in Ca²⁺ influx across the plasma membrane.

In the present study, several lines of evidence suggest that voltage-operated Ca²⁺ channels are not involved in histamine-induced Ca²⁺ entry in cultured rat cerebellar astrocytes. Depolarization of astrocytes by raising the extracellular K⁺ concentration only rarely (< 10 %) resulted in a measurable increase in [Ca²⁺]_i. Furthermore, nifedipine, an inhibitor of voltage-dependent L-type Ca²⁺ channels, had no effect on the histamine-induced [Ca²⁺]_i plateau, whereas it was able to block depolarization-induced [Ca²⁺]_i increases. Hence, only capacitative and/or non-capacitative, non-voltage-operated Ca²⁺ entry pathways could account for the Ca²⁺ influx activated by histamine in cultured rat cerebellar astrocytes.

Our experiments suggest that multiple Ca²⁺ influx pathways are involved in Ca²⁺ entry initiated by the activation of PLC-coupled H₁ receptors. There is clearly a capacitative component, which depends on the filling state of InsP₃-sensitive Ca²⁺ stores, as well as an additional component, which might not only be dependent on store depletion. The involvement of a capacitative component in histamine-induced Ca²⁺ entry into cultured rat cerebellar astrocytes can be concluded from experiments using cyclopiazonic acid (CPA), a selective inhibitor of the Ca²⁺ pumps of intracellular storage organelles. When cytosolic Ca²⁺ stores were emptied by CPA in the absence of extracellular Ca²⁺, a pronounced [Ca²⁺]_i elevation could be observed upon addition of Ca²⁺ to the extracellular solution, which suggests the existence of capacitative entry mechanisms in cultured rat cerebellar astrocytes (Figs 5–7). However, in the presence of histamine, the [Ca²⁺]_i increase upon re-addition of external Ca²⁺ was increased even beyond the level observed when the capacitative pathway was activated by CPA alone. Hence, it can be concluded that in cultured rat cerebellar astrocytes, histamine-induced calcium entry cannot be mimicked by store depletion alone and is likely to involve a non-capacitative component. On the other hand, the additional effect of histamine was smaller than would have been expected if histamine-induced Ca²⁺ entry was entirely mediated by non-capacitative mechanisms. This might argue in favour of the activation by CPA and histamine of a Ca²⁺ entry pathway dependent on store depletion.

Broad et al. (1999) reported that, after stimulation with maximal agonist concentrations, about 25 % of the Ca²⁺ entry into rat A7r5 smooth muscle cells stimulated by vasopressin occurs via non-capacitative pathways. At lower agonist concentrations, however, the non-capacitative pathway provides the major route for Ca²⁺ entry, accounting for up to 90 % of the total influx. Similar suggestions have been made by Shuttleworth & Thompson (1998) for the human embryonic kidney (HEK293) cell line, who reported that while capacitative Ca²⁺ entry can be clearly demonstrated under conditions of maximal, or near-maximal, agonist stimulation, it is uncertain whether such a mechanism can operate during the oscillatory [Ca²⁺]_i signals that are frequently seen at low, more physiological agonist concentrations. Thus, it will be interesting for future studies to investigate the contribution of the non-capacitative Ca²⁺ entry component in cultured rat cerebellar astrocytes stimulated with submaximal histamine concentrations.

In agreement with observations on Ca²⁺ entry in A7r5 smooth muscle cells (Broad et al. 1999), chinese hamster ovary cells expressing the platelet-derived growth factor receptor (Mathias et al. 1997) and HEK293 cells stably expressing human Trp3 (Zhu et al. 1998), our results suggest that the activation of PLC is necessary for receptors to stimulate both the capacitative and the non-capacitative Ca²⁺ entry pathway. This can be concluded from the fact that the histamine-induced [Ca²⁺]_i plateau was abolished in astrocytes treated with the phospholipase inhibitor U73122. The mechanisms involved seem to differ from those reported by Clementi et al. (1992) and Shuttleworth & Thompson (1998) who demonstrated that receptor-activated non-capacitative Ca²⁺ entry is independent of the simultaneous activation of the PLC-InsP₃ pathway in rat PC12 cells and HEK293 cells stably transfected with the human m3 muscarinic receptor, respectively.

Interestingly, no [Ca²⁺]_i increase above control levels could be observed upon addition of Ca²⁺ to the bathing solution after depletion of intracellular Ca²⁺ stores with histamine under Ca²⁺-free conditions and subsequent removal of the agonist (Fig. 4). Byron et al. (1992) describe a similar finding in human foreskin fibroblasts stimulated with bradykinin. The authors report that refilling of intracellular Ca²⁺ stores requires extracellular Ca²⁺, but may occur without a detectable elevation of [Ca²⁺]_i after inactivation or inhibition of the non-capacitative Ca²⁺ entry pathway. As in the case of human fibroblasts, refilling of the intracellular Ca²⁺ stores in cultured rat cerebellar astrocytes was found to occur in the absence of the agonist and with no observable elevation in [Ca²⁺]_i. On the one hand, this finding provides further evidence for the existence of a capacitative Ca²⁺ entry route in the astrocytes, because, for the stores to refill with calcium, the receptors obviously do not need to remain activated. On the other hand, it raises the question of how Ca²⁺ enters the cell without a change in fura-2 fluorescence. One possible mechanism is postulated by Mogami et al. (1997) for pancreatic acinar cells, where recharging of the Ca²⁺ stores occurred via ‘tunnels’, unseen by cytosolic Ca²⁺ indicators. In this context, it has to be kept in mind that imaging intracellular Ca²⁺ concentration with fluorescent Ca²⁺ indicators such as fura-2 reveals only net Ca²⁺ fluxes. Hence, our results do not necessarily favour refilling of Ca²⁺ stores by a route that bypasses the cytoplasm; it may rather indicate a fast sequestration of entering Ca²⁺ into Ca²⁺ stores located close to the plasma membrane. In other words, refilling of internal Ca²⁺ stores is ‘silent’ with respect to changes in cytosolic Ca²⁺. Ca²⁺ entry will only become visible if, at the same time, Ca²⁺ uptake into intracellular storage organelles is reduced, or if Ca²⁺ release from these sites is stimulated. This would explain the pronounced [Ca²⁺]_i elevations due to Ca²⁺ entry observed in the presence of CPA and/or histamine. It remains to be elucidated how the agonist-induced Ca²⁺ entry might be related to the filling state of the intracellular Ca²⁺ stores.