JN Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Neurophysiol 79: 695-703, 1998;
0022-3077/98 $5.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Amos, B. J.
Right arrow Articles by Chesler, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Amos, B. J.
Right arrow Articles by Chesler, M.

The Journal of Neurophysiology Vol. 79 No. 2 February 1998, pp. 695-703
Copyright ©1998 by the American Physiological Society

Characterization of an Intracellular Alkaline Shift in Rat Astrocytes Triggered by Metabotropic Glutamate Receptors

Brian J. Amos and Mitchell Chesler

Department of Neurosurgery and Department of Physiology and Neuroscience, New York University Medical Center, New York, New York 10016

    ABSTRACT
Abstract
Introduction
Methods
Results
Discussion
References

Amos, Brian J. and Mitchell Chesler. Characterization of an intracellular alkaline shift in rat astrocytes triggered by metabotropic glutamate receptors. J. Neurophysiol. 79: 695-703, 1998. The modulation of intracellular pH by activation of metabotropic glutamate receptors was investigated in cultured and acutely dissociated rat astrocytes. One minute superfusion of 100 µM (1S,3R)-1-aminocyclopentane-1,3-dicarboxcylic acid (ACPD) evoked an alkaline shift of 0.13 ± 0.013 (mean ± SE) and 0.16 ± 0.03 pH units in cultured (cortical or cerebellar) and acutely dissociated cortical astrocytes, respectively. Alkalinizations were elicited by concentrations of ACPD as low as 1 µM. The ACPD response was mimicked by S-3-hydroxyphenylglycine (3-HPG) and by (s)-4-carboxy-3-hydroxyphenylglycine (4C-3HPG) but was not blocked by alpha -methyl-4-carboxyphenylglycine (MCPG) or (RS)-1-aminoindan-1,5-dicarboxcylic acid (AIDA), features consistent with an mGluR5 receptor-mediated mechanism. The ACPD-evoked alkaline shift was insensitive to amiloride, 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS), and the v-type ATPase inhibitors 7-chloro-4-nitrobenz-2-oxa-1,3-diazol (NBD-Cl), bafilomycin, and concanamycin. The alkaline response persisted in Na+- or Cl--free saline, but was reversibly blocked in bicarbonate-free, N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES)-buffered solutions. A bicarbonate-dependent and Na+-independent alkaline shift could also be elicited by either 3 mM caffeine or 1 µM ionomycin. These data suggest that a rise in cytosolic Ca2+ activity is instrumental in triggering the alkalinizing mechanism and that this response is independent of the classic depolarization-induced alkalinization mediated by electrogenic sodium-bicarbonate cotransport.

    INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References

Neuronal activity is associated with rapid pH shifts in both the extracellular and intracellular space (Chesler 1990; Chesler and Kaila 1992). The intracellular pH (pHi) of astrocytes is especially responsive to the firing of adjacent nerve cells. In vivo recordings of pHi from astrocytes have revealed large, rapid, intracellular alkaline shifts induced by electrical activity or by episodes of spreading depression (Ballanyi et al. 1994; Chesler and Kraig 1987, 1989). These pH shifts were blocked when the activity-dependent depolarization of the astrocyte membrane was prevented (Chesler and Kraig 1989).

A depolarization-induced alkalinization (DIA) in glial cells of the leech was attributed to the action of an electrogenic Na+-HCO-3 cotransporter with a stoichiometry of 1:2 (Deitmer and Szatowski 1990). A number of in vitro studies of mammalian astrocytes have provided evidence for a DIA due to a similar electrogenic mechanism, reliant on Na+ and HCO-3 ions (Brookes and Turner 1994; Grichtchenko and Chesler 1994a,b; Pappas and Ransom 1994). These data appear to explain the in vivo response of astrocytes to neuronal activity. The elevation of extracellular K+ would depolarize the astrocyte membrane and thereby drive the electrogenic inward movement of Na+ and HCO-3.

A number of additional mechanisms may govern astrocyte pH during neural activity. To varying degrees, acid-generating processes can offset the alkalinizing effect of the Na+-HCO-3 cotransporter. These include the metabolic generation of CO2 (Voipio and Kaila 1993) as well as the uptake of glutamate from the extracellular space (Amato et al. 1994). Recent studies by Amos and colleagues have uncovered an additional process that may contribute to the activity-dependent alkalinization of astroctyes. In response to the metabotropic glutamate receptor agonist (1S,3R)-1-aminocyclopentane-1,3-dicarboxcylic acid (ACPD), cultured astrocytes underwent a rapid rise in pHi. By contrast, ACPD caused an intracellular acid shift in neurons (Amos et al. 1996a).

The pharmacological basis of the ACPD-induced alkaline shift and its relationship to the established mechanisms of intracellular pH regulation is unknown. Whether the ACPD response of cultured cells is a feature shared by astrocytes in vivo is also not clear. In this report, we address these issues using both primary cultured and acutely dissociated rat astrocytes. Our data suggest that the response is triggered by the mGluR5 subtype of metabotropic glutamate receptor and is mediated via the release of Ca2+ from intracellular stores. The alkalinizing mechanism is HCO-3 dependent, but is not dependent on external Na+, distinguishing it from the common plasmalemmal acid transport mechanisms. Preliminary accounts of these results have appeared in abstract form (Amos et al. 1996b).

    METHODS
Abstract
Introduction
Methods
Results
Discussion
References

Primary astrocyte culture

Cells were obtained from cortex or cerebellum of 3- to 5-day-old neonatal rats as described previously (Amos and Richards 1996). No differences in results were evident in experiments performed with cortical versus cerebellar astrocytes. Briefly, rats were anesthetized with methoxyfluorane and then decapitated. The brain was rapidly excised and the meninges removed. Cortical or cerebellar chunks were chopped and placed in ice-cold N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES) modified minimum essential medium Eagle's (MEME). Chunks were triturated through progressively finer bore glass pipettes until mechanically dissociated. The cell suspension was then spun at 500 rpm for 5 min and the supernatant discarded. The resuspended cells were plated onto poly-L-lysine coated slips in MEME supplemented with glutamine and 2.5% fetal calf serum and maintained in a 5% CO2-humidified atmosphere at 37°C. Cells were used after 7-14 days in culture.

Cultures contained both astrocytes and neurons. Astrocytes could be distinguished from neurons on the basis of morphology; in low-density cultures they were readily identified by their thin, branched processes that typically emanated from the cell body in a stellate fashion and bore thornlike excrescences. In high-density cultures, astrocytes were flat without processes. Such differences in morphology with different culture densities have been noted previously (Kimelberg 1983). Immunocytochemical staining for glial fibrillary acidic protein confirmed the correlation of these characteristics with astrocytic phenotype. In addition, physiological confirmation of astrocyte phenotype was provided by the presence of depolarization-induced alkalinizations (e.g., Fig. 7), evoked by elevation of bath K+ (Boyarsky et al. 1993; Pappas and Ransom 1994).


View larger version (10K):
[in this window]
[in a new window]
  		FIG. 7. Time course of sodium wash out. A: a depolarization-induced alkalinization (DIA) was elicited with elevated extracellular potassium. After ~3 min in sodium-free solution, the DIA was completely abolished, but recovered after return to normal sodium-containing media. B: a sodium-selective microelectrode placed on the floor of the chamber recorded the time course of sodium wash out and recovery. The sodium concentration fell 2 orders of magnitude in <1 min.

Acute dissociation

Dissociated cells were prepared from the cortex of 13- to 15-day-old rats, based on a protocol by Mintz et al. (1992). Rats were anesthetized and decapitated. The excised brain was cut into 300-µm cortical slices that were incubated in bicarbonate buffered Ringer at room temperature. Slices were transferred to oxygen-equilibrated dissociation solution (see below) containing Sigma type XXIII protease (3 mg/ml) and gently stirred for 13 min at 37°C. The protease solution was then washed off and the slices stirred in a solution containing DNase (bovine type IV, 0.5 mg/ml) and trypsin inhibitor (Boehringer Mannheim, 1 mg/ml) for 4 min. Slices were then mechanically dissociated using glass pipettes of progressively finer bore. Cells were plated onto poly-L-lysine coated slips and left for 1-2 h before use. Acutely dissociated astrocytes had a classical "protoplasmic" architecture, displaying compact arborizations with innumerable fine processes. These cells were readily distinguished from the more numerous neurons that had pyramidal somata and clear basal and apical dendritic arborizations (Tse et al. 1992).

Solutions

Standard bicarbonate-buffered Ringer contained (in mM) 124 NaCl, 26 NaHCO3, 3.0 KCl, 1.0 NaH2PO4, 3.0 CaCl2, 1.5 MgSO4, and 10 glucose. The solution was continually gassed with 95% O2-5% CO2. Gas-impermeant Saran tubing was used to minimize loss of CO2 from the solution lines. In HEPES-buffered solutions, 26 mM HEPES replaced the NaHCO3, and the solution was titrated to pH 7.4 using NaOH, with NaCl adjusted to maintain a constant sodium concentration. In sodium-free solutions N-methyl-D-glucamine and choline bicarbonate replaced NaCl and sodium bicarbonate, respectively. In chloride-free solutions, sodium methanesulfonate and calcium gluconate were substituted for NaCl and CaCl2, respectively. In calcium-free solutions 1 mM ethylene glycol-bis(beta -aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA) was added, and NaHCO3 was increased to 28 mM to balance the acidity of the EGTA. The acute dissociation solution consisted of (in mM) 82 Na2SO4, 30 K2SO4, 10.5 MgCl2, 10 HEPES, and 10 glucose (Mintz et al. 1992). This solution was titrated to pH 7.4 using NaOH and gassed with O2 for 1 h before use.

Nigericin and ionomycin were dissolved from a 1-mM stock in ethanol. The agents ACPD, S-3-hydroxyphenylglycine (3-HPG), alpha -methyl-4-carboxyphenylglycine (MCPG), (RS)-1-aminoindan-1,5-dicarboxcylic acid (AIDA), and (s)-4-carboxy-3-hydroxyphenylglycine (4C-3HPG) were dissolved in water as 50-mM stocks with one equivalent of NaOH. 2',7'-bis(carboxyethyl)-5-(6)-carboxyfluorescein acetoxymethyl ester (BCECF-AM), ethylisopropyl amiloride, and 7-chloro-4-nitrobenz-2-oxa-1,3-diazol (NBD-Cl) were dissolved in dimethyl sulfoxide as stock solutions. Caffeine, 4-acetamido-4'-isothyocyanostilbene-2,2'-disulfonic acid (SITS), and 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS) were added directly to the superfusion solution.

Measurement and data analysis

Intracellular pH was measured using the fluorophore BCECF. Cells were loaded with 2 µM of the acetoxy-methyl ester form of the dye for 15 min at room temperature. The coverslip served as the floor of a submersion chamber mounted on the stage of a Zeiss Axiovert inverted microscope equipped for epifluorescence. Alternate excitation with 490- and 440-nm light (at 100 Hz) was achieved with a 75-W xenon lamp alternately directed toward 490- or 440-nm excitation filters (Omega Optical) using a chopper-based system (Photon Technology International). Emissions above 515 nm were collected by a photomultiplier via a ×40 oil immersion objective. Data points consisted of the ratios of the 490- to 440-nm-induced emissions averaged over 1-s intervals. To calibrate the emission ratios, a curve was generated (Boyarsky et al. 1988) using the high-K+ (150 mM) and nigericin (3 µM) containing solutions (Thomas et al. 1979) buffered with piperazine-N,N'-bis(2-ethanesulfonic acid) or HEPES-buffered solutions over the pH range 6.0-8.0.

Statistics were expressed as means ± SE. Because responses to ACPD exhibited variability when repeated on the same cell, unpaired comparisons were usually made against a large control population using a Student's t-test. In some early cultures, responses to ACPD were atypically large. In making unpaired comparisons from these experiments, a set of control responses from within this data set was utilized. In instances where responses in the same cell tended to be consistent, paired comparisons were made using a Student's t-test.

    RESULTS
Abstract
Introduction
Methods
Results
Discussion
References

Response of astrocytes to ACPD

Bath application of 100 µM ACPD evoked a rapid alkalinization in cultured astrocytes (Fig. 1A). Alkalinizing responses could be elicited by concentrations of ACPD as low as 1 µM. Repeated application of the agonist often produced responses of inconsistent amplitude (either larger or smaller) with successive trials. Because of this variable amplitude, dose-response studies were not performed on individual cells. The alkalinization, which began immediately upon application of ACPD, was sustained for the duration of the ACPD exposure, when applied for 1 min. The mean increase in pHi induced by 1-min exposure to 100 µM ACPD was 0.13 ± 0.013 pH units (mean ± SE; n = 23 cells) from a mean baseline pHi of 6.75 ± 0.10. When ACPD exposure was carried out for 3 min (Fig. 1B), the alkalinization gradually declined by an average of 35 ± 4% (n = 4 cells). Longer exposures were not studied.


View larger version (13K):
[in this window]
[in a new window]
  		FIG. 1. (1S,3R)-1-Aminocyclopentane-1,3-dicarboxcylic acid (ACPD)-evoked alkalinizations in astrocytes. A: brief superfusion of 100 µM ACPD elicited a reversible alkaline shift in a 7-day cultured, cortical astrocyte. B: a more sustained application of 100 µM ACPD to a 10-day cultured, cerebellar astrocyte. C: reversible alkaline shift elicited by 100 µM ACPD in a 12-day cultured, cerebellar astrocyte. Cells were exposed to 1 mM dibutyrl c-AMP for 4 h before recording. D: reversible alkaline shift elicited by 100 µM ACPD in an acutely dissociated cortical astrocyte.

The alkalinizing effect of ACPD was noted in most, but not in all astroctye cultures. Response to the agonist was not correlated with any obvious morphological feature or history of the cells. Thus similar ACPD-induced alkaline shifts were noted both in flat, process-free and in highly branched or stellate astrocytes. In addition, similar alkaline shifts (0.10-0.15 pH units in 3 cells) were observed in astrocytes that were "rounded" by a 4- to 6-h exposure to 1 mM dibutyrl-c-AMP (Kimelberg et al. 1979), as shown in Fig. 1C. To determine whether the alkalinizing effect of ACPD was a feature induced by primary culture, we also carried out similar experiments on astrocytes acutely dissociated from rat cortex. In seven acutely dissociated cells, 100 µM ACPD induced a mean alkaline shift of 0.16 ± 0.03 pH units (Fig. 1D).

Pharmacology of the ACPD response

In a previous report, the ACPD-evoked alkalinization of cultured astrocytes was mimicked by the specific group I mGluR agonist 3-HPG, whereas group II and group III agonists were ineffective (Amos et al. 1996a). We noted similar alkaline shifts induced by 250 µM 3-HPG in both cultured and acutely dissociated astrocytes (Fig. 2A). These data suggest that the response is mediated by either the mGluR1 or mGluR5 subtype within group I. To further examine this possibility, we tested the efficacy of the general mGluR antagonist MCPG. This agent generally blocks both mGluR1 and mGluR5 responses (Kingston et al. 1995) but in some reports was ineffective against mGluR5a and mGluR5b receptors (Joly et al. 1995). As shown in Fig. 2B, 500 µM MCPG failed to prevent or reduce the ACPD-evoked alkalinization. The mean alkaline shift in MCPG (0.09 ± 0.02 pH units, n = 7 cells) was not significantly different from control responses (P = 0.07). Furthermore, AIDA (100 µM), which inhibits mGluR1a-mediated phosphoinositide hydrolysis in baby hamster kidney cells (Pellicciari et al. 1995), failed to block the ACPD-induced alkalinization (Fig. 2C). Among four cells, the mean response to ACPD in the presence of AIDA was 0.15 ± 0.05 pH units, which was not significantly different from controls (P = 0.33). Additional experiments utilized the mGluR1 antagonist 4C-3HPG. At high concentrations, this agent has shown agonist activity at mGluR5 receptors (Brabet et al. 1995) and may therefore serve to differentiate between mGluR1 and mGluR5 responses. In six astrocytes, 200 µM 4C-3HPG elicited a rapid, reversible alkalinization (0.10-0.25 pH units) consistent with an mGluR5 receptor-mediated response (Fig. 2D).


View larger version (12K):
[in this window]
[in a new window]
  		FIG. 2. Pharmacology of the ACPD response. A: the group I mGluR agonist S-3-hydroxyphenylglycine (3-HPG; 250 µM) evoked a reversible alkaline shift in a cortical astrocyte. B: application of 50 µM ACPD elicited an alkaline shift in a cortical astrocyte in the presence of the general mGluR antagonist alpha -methyl-4-carboxyphenylglycine (MCPG; 500 µM). C: application of 25 µM ACPD elicited an alkaline shift in a cultured cortical astrocyte in the presence of the mGluR1 antagonist (RS)-1-aminoindan-1,5-dicarboxcylic acid (AIDA; 100 µM). D: superfusion of (s)-4-carboxy-3-hydroxyphenylglycine (4C-3HPG; 200 µM) elicited an alkaline shift in a cultured cerebellar astrocyte.

Effect of acid transport inhibitors on the ACPD response

Four plasmalemmal acid transport mechanisms have been identified on astrocytes. Acid extruders include Na+-H+ antiport, electrogenic Na+-HCO-3 symport and a v-type ATPase. In addition, Cl--HCO-3 antiport has been identified, which may serve to acidify the cytosol (for review, see Deitmer and Rose 1996). Classic inhibitors of these mechanisms had no effect on the ACPD-induced alkaline shift. As shown in Fig. 3A, application of 1 mM amiloride, which effectively blocks Na+-H+ antiport in astrocytes (Pappas and Ransom 1993; Pizzonia et al. 1996), did not block the response to 100 µM ACPD. In seven cells, the mean alkaline shift in the presence of amiloride was not significantly different (P = 0.06) from the control population. In paired trials on three astrocytes (Fig. 3A), there was also no difference in the size of the response in the presence of amiloride versus controls (P = 0.5). Stilbene inhibitors of HCO-3 transport were similarly ineffective. Neither 500 µM SITS (n = 2) nor 500 µM DIDS (n = 7) blocked the alkaline shift. The ACPD response in 500 µM DIDS (0.19 ± 0.02; n = 4 cells) was not significantly different (P = 0.06) from a set of control responses (0.24 ± 0.002 pH units, n = 6 cells). In addition, in paired trials performed on three astrocytes, the ACPD-evoked alkaline shifts in 1 mM DIDS were not significantly different(P = 0.37) from control responses (Fig. 3B).


View larger version (15K):
[in this window]
[in a new window]
  		FIG. 3. Classic inhibitors of secondary acid transport do not block the ACPD response. A: ACPD-evoked alkaline shift persisted after a 5-min incubation in 1 mM amiloride (cortical astrocyte, 50 µM ACPD). B: ACPD-evoked alkaline shift persisted after a 10 min incubation in 1 mM 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS; cortical astrocyte, 50 µM ACPD).

In a previous study of cultured astrocytes, the recovery of intracellular pH from an acid load was found to have a component sensitive to bafilomycin, suggesting the presence of a plasmalemmal v-type H+-ATPase (Pappas and Ransom 1993). We tested three inhibitors of v-type H+-ATPases to determine whether a similar mechanism could be responsible for the ACPD-evoked alkaline shift. At 500 µM, the general inhibitor NBD-Cl (Forgac 1989) caused an immediate acidification of the cultured astrocytes (mean of -0.35 ± 0.06 pH units with an initial rate of -0.0026 ± 0.0004 pH units/s,n = 4). A similar rapid acid shift induced by NBD-Cl has also been noted in an abstract by Volk et al. (1996). After stabilization of pHi, application of 100 µM ACPD (Fig. 4) induced a mean alkaline shift 0.14 ± 0.04 pH units (n = 4), which was not significantly different from the control ACPD response (P = 0.29). Due to its high cost (and therefore limited availability), bafilomycin A1 (Bowman et al. 1988) was tested by preincubating cells in 1 µM bafilomycin for 15 min. Preincubation effectively blocked the ATPase in other preparations (Swallow et al. 1990). In four cells preincubated with bafilomycin, the ACPD-induced alkalinization was not significantly different from paired controls (P = 0.32), nor was the baseline pHi lower. Concanamycin A is a more potent and readily obtained inhibitor of v-type H+-ATPases (Drose et al. 1993). In three astrocytes, superfusion of 1 µM concanamycin A failed to inhibit the ACPD-evoked alkaline shift and had no effect on baseline pHi. Last, we examined the f-type H+-ATPase inhibitor N,N'-dicyclohexylcarbodiimide (50 µM). This agent caused a delayed acidification (-0.0002 ± 6 × 10-5 pH units/s at 102 ± 7.8 s, n = 3) before the ACPD challenge, but after 4 min exposure, also failed to inhibit the ACPD-induced alkaline shift (not shown).


View larger version (7K):
[in this window]
[in a new window]
  		FIG. 4. Application of the general v-type ATPase inhibitor, 7-chloro-4-nitrobenz-2-oxa-1,3-diazol (NBD-Cl, 50 µM) to a cultured cortical astrocyte. NBD-Cl caused an immediate acidification, but did not block the alkalinization evoked by 100 µM ACPD.

Bicarbonate dependence of the ACPD-induced alkalinization

In a previous investigation of cultured astrocytes, ACPD evoked a small acidification in HCO-3-free, HEPES-buffered saline (Brune and Deitmer 1995). Subsequently, ACPD was found to evoke an alkaline shift in astrocytes studied in HCO-3-buffered Ringer. Cells studied in HEPES again produced only acid shifts (Amos et al. 1996a).

The most likely explanation of these observations is that the metabotropic alkaline shift requires HCO-3 ions. This was tested in single astrocytes exposed to transitions between HCO-3 and HEPES-buffered media as shown in Fig. 5. The switch from HCO-3 to HEPES and back to HCO-3 Ringer produced alkaline and acid transients, because of the respective wash out and reentry of CO2 (Roos and Boron 1981; Thomas 1984). As shown in Fig. 5, ACPD evoked a rapid alkaline shift in the HCO-3 Ringer, which was reversibly blocked on switching to the nominally HCO-3-free HEPES saline. In 12 astrocytes a similar block of the ACPD response was noted after transition between HCO-3 and HEPES buffers. In contrast with the previous study in which ACPD elicited an acid shift in HEPES-buffered media (Amos et al. 1996a), no change in pHi occurred when ACPD was applied in HEPES-buffered Ringer.


View larger version (9K):
[in this window]
[in a new window]
  		FIG. 5. ACPD-evoked alkaline shift is bicarbonate dependent. A control alkalinization evoked by 100 µM ACPD is shown in normal solution. Replacement of control saline with bicarbonate-free, N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES)-buffered solution resulted in an alkaline transient due to wash out of CO2. After stabilization of pHi in the HEPES media, a 2nd ACPD application failed to elicit a response. The response recovered after return to bicarbonate-buffered solution.

ACPD-evoked alkaline shift in Na+- and Cl--free solutions

The effect of Na+ substitution was tested in CO2/HCO-3-buffered Ringer using N-methyl-D-glucamine chloride and choline bicarbonate in place of sodium chloride and sodium bicarbonate, respectively. Transition to the sodium-free Ringer caused a rapid acidification as shown in Fig. 6. Similar effects have been described previously (Boyarsky et al. 1993) and are presumably due to reversal of the Na+/H+ antiporter and/or the Na+-HCO-3 cotransporter. Prolonged exposure to Na+-free solutions was avoided because the pHi often continued to fall in this media, and because changes in intracellular Ca2+ (due reversal of Na+-Ca2+ exchange) would be expected to alter the pHi responses (see below). In a total of eight cells, 100 µM ACPD continued to elicit a pronounced alkaline shift (Fig. 6) after as long as 7 min in zero media Na+ (after a mean of 302 ± 32 s, range 220-430 s). In an unpaired comparison, the alkaline shift induced by 100 µM ACPD in Na+-free solution(0.26 ± 0.01 pH units, n = 6 cells) was not significantly different from a set of control responses (0.24 ± 0.002 pH units n = 6 cells) obtained in normal saline (P = 0.36). In two experiments on later cultures (where responses were smaller) the ACPD-evoked alkaline shifts were similarly unaffected in Na+-free saline.


View larger version (11K):
[in this window]
[in a new window]
  		FIG. 6. ACPD-evoked alkaline shift does not require external sodium. Following a control response elicited by 100 µM ACPD, the saline was changed to a sodium-free solution, causing a rapid acidification. After 220 s, a 2nd ACPD application elicited a similar alkaline response.

Several steps were taken to ensure that extracellular Na+ was adequately washed out of the bath. First, we tested whether the electrogenic Na+-HCO-3 cotransporter was blocked by similar transitions to Na+-free Ringer. As shown in Fig. 7A, elevation of external K+ produced the expected DIA in normal Ringer, but failed to do so after a few minutes in Na+-free solution. In total, the DIA was blocked in five cells at 3-11 min after transition to Na+-free solution. To further test the wash out of bath Na+, measurements were made using Na+-selective microelectrodes with tips positioned on the floor of the experimental chamber. Figure 7B displays a transition from 150 to 1.5 mM Na+, which was 99% complete within 40 s.

In principle, a Cl-/HCO-3 antiporter could alkalinize the cytosol if ACPD induced a rise in intracellular Cl- activity causing efflux of Cl- in exchange for HCO-3. To address this possibility, cultured astrocytes were bathed in Cl--free bicarbonate Ringer for >= 10 min to deplete intracellular Cl- and to prevent possible influx of Cl-. In both brain slices (Ballanyi et al. 1987) and in cultured astrocytes (Bevensee 1997), glial intracellular chloride has been shown to be rapidly depleted by similar exposure to Cl--free media. On transition from normal to Cl--free saline, an immediate alkalinization occurred (presumably due to reversal of a Cl-/HCO-3 antiporter), which was followed by a variable recovery back toward baseline pHi after return to normal media. In seven cells, after 9-13 min in Cl--free Ringer, application of 100 µM ACPD elicited a rapid alkaline shift (0.13 ± 0.03 pH units, n = 7; Fig. 8), which was not different from the responses in normal saline (P = 0.35).


View larger version (8K):
[in this window]
[in a new window]
  		FIG. 8. ACPD-evoked alkaline shift persists in the absence of external chloride. Transition to chloride-free media caused a reversible alkalinization. After stabilization of the pH, superfusion of 100 µM ACPD elicited a typical rapid alkaline shift. A similar ACPD response was evoked after return to normal saline.

Role of calcium in the ACPD-evoked alkalinization

In an earlier report, it was suggested that elevation of astrocyte cytosolic Ca2+ was instrumental in triggering the ACPD-related alkaline shift, because the response was inhibited in cells loaded with 1,2,-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA) (Amos et al. 1996b). To further evaluate this hypothesis, we investigated the Ca2+ dependence of the response by a variety of means.

The ACPD-evoked alkaline shifts were unchanged in Ca2+-free Ringer containing 1 mM EGTA, although in a few preliminary experiments the responses were reduced (Amos et al. 1996b). In eight of nine cells, an alkaline shift could be repeatedly elicited by 100 µM ACPD in the absence of external Ca2+. The mean response in Ca2+-free Ringer (0.10 ± 0.01 pH units) was not significantly different from controls (P = 0.09). Thus the ACPD-induced alkaline shift does not directly depend on the entry of Ca2+ across the plasma membrane.

Exposure to caffeine has been shown to elevate cytosolic Ca2+ in cultured astrocytes (Golovina et al. 1996). In 13 of 23 cultured astrocytes, exposure to 3 mM caffeine induced an alkalinization (0.1 ± 0.02 pH units) that could be evoked repeatedly (Fig. 9A). In 6 of the 23 cells, caffeine induced an acidification, and in 4 cells, no response was seen. It is notable that caffeine may not consistently raise cytosolic Ca2+ in astrocyte cultures (Langley and Pearce 1994), which may explain the absence of an alkalinization in some cases.


View larger version (13K):
[in this window]
[in a new window]
  		FIG. 9. Caffeine mimics the ACPD-evoked alkaline shift. A: superfusion of 3 mM caffeine elicited repeatable alkaline shifts in a cultured cortical astrocyte. B: the alkaline shift evoked by 3 mM caffeine persisted in the absence of extracellular sodium.

To compare the ionic dependence of the caffeine-evoked pH shift with the ACPD response, we tested the effect of either HCO-3 or Na+-free Ringer. The Na+-free media was tested on four astrocytes in which caffeine evoked an alkaline shift. In three of four cells, the alkalinization persisted in Na+-free solution (Fig. 9B), although the increases in pH were diminished (~0.05 pH units) relative to the paired control responses in Na+-containing media (~0.1 pH units). The bicarbonate dependence of the caffeine response was tested in three astrocytes. In all of these cells, caffeine elicited an alkaline shift in normal media, which was reversibly blocked following transition to bicarbonate-free, HEPES-buffered media (Fig. 10).


View larger version (9K):
[in this window]
[in a new window]
  		FIG. 10. Caffeine-evoked alkaline shift is bicarbonate-dependent. Sequential transition from bicarbonate- to HEPES-buffered media repeatedly blocked the alkaline shift elicited by 3 mM caffeine in a cultured cerebellar astrocyte.

The calcium ionophore ionomycin provided a second means of increasing cytosolic calcium. In 11 astrocytes, superfusion of 1 µM ionomycin induced an alkaline shift of 0.10 ± 0.01 pH units in normal, HCO-3-buffered media. In eight cells, the ionomycin-induced alkalinization was blocked after transition to HEPES-buffered media, and recovered on return to HCO-3 buffer (Fig. 11). In three cells, however, we noted a persistent, albeit smaller (mean of 0.07 pH units) alkalinization in HEPES-buffered solution. Like the ACPD and caffeine responses, the ionomycin-evoked alkaline shift persisted in the absence of extracellular sodium (Fig. 12). In an unpaired comparison of responses from five astrocytes, superfusion of 1 µM ionomycin in sodium-free saline evoked alkaline shifts (0.18 ± 0.02 pH units) that were slightly larger than controls (P = 0.03). In three paired trials on single astrocytes, the ionomycin-evoked alkaline shift was not significantly different from the control responses (P = 0.11).


View larger version (10K):
[in this window]
[in a new window]
  		FIG. 11. Ionomycin elicits an alkaline shift that is dependent on bicarbonate. Superfusion of 1 µM ionomycin caused a rapid, reversible alkalinization of a cultured cortical astrocyte in normal saline. The response was reversibly blocked in bicarbonate-free, HEPES-buffered media.


View larger version (7K):
[in this window]
[in a new window]
  		FIG. 12. Ionomycin-evoked alkaline shift does not require extracellular sodium. After ~4 min in sodium-free saline, ionomycin (1 µM) evoked a reversible alkalinization in a cultured cortical astrocyte. A similar response was elicited by ionomycin after return to normal extracellular sodium.

    DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References

Astrocytes undergo significant increases in intracellular pH during neuronal activity (Ballanyi et al. 1994; Chesler and Kraig 1987, 1989). This alkaline shift is triggered by membrane depolarization that drives the inward, electrogenic cotransport of Na+ and HCO-3 ions (Deitmer and Rose 1996). In cultured astrocytes, Amos and colleagues noted a rapid alkaline shift triggered by mGluR receptors (Amos et al. 1996a). Here we report that this metabotropic alkalinization is a feature of both cultured and acutely dissociated astrocytes and demonstrate that its mechanism of action is distinct from the depolarization-induced alkaline shift of the Na+-HCO-3 cotransporter.

Earlier pharmacological data suggested that the alkaline response to ACPD is mediated by group I metabotropic glutamate receptors because it could be evoked by HPG, but not by group II or group III agonists (Amos et al. 1996a). The present results agree with this interpretation and suggest a differentiation between mGluR1 and mGluR5 subtypes. Failure of MCPG to block the response may be consistent with a few reports in which this antagonist failed to block mGluR5-mediated effects (Joly et al. 1995). In addition, because AIDA may be considered an effective antagonist at mGluR1 receptors (Pellicciari et al. 1995), its failure to block the response argues against the involvement of this subtype. At high concentrations, 4C-3HPG acts as an antagonist at mGluR1 receptors but as an agonist at the mGluR5 subtype (Brabet et al. 1995). The fact that 4C-3HPG mimicked the ACPD-induced alkaline shift would therefore argue in favor of an mGluR5-mediated mechanism.

Activation of group I mGluR receptors is coupled to the liberation of Ca2+ from internal stores via the elevation of inositol triphosphate (IP3) (Abe et al. 1992; Kawabata et al. 1996; Pin and Duvoisin 1995). The present data suggest that a rise in internal Ca2+ is instrumental in triggering the ACPD-evoked alkalinization. Whereas the ACPD response is most likely mediated by release of Ca2+ from IP3-sensitive stores, the Na+-independent, and HCO-3-dependent alkaline shifts induced by caffeine suggest that ryanodine receptor-coupled Ca2+ stores can serve a similar role. Our results therefore indicate that the alkalinizing mechanism can be triggered by a number of processes that lead to an increase in cytosolic Ca2+ activity.

Ionomycin also induced an alkaline shift that was independent of Na+ ions and could be reversibly abolished by withdrawal of HCO-3. In a few examples, however, ionomycin caused a rise in pHi in the nominal absence of HCO-3. At similar concentrations, this ionophore has been reported to alkalinize other cells in HCO-3-free media (Asem et al. 1992). To some degree ionomycin may exchange Ca2+ for H+ across the plasma membrane (Kauffman et al. 1980). But because the intracellular buffering power is greater in the presence of HCO-3, and the alkalinizations were usually abolished in HEPES media, it appears unlikely that simple Ca2+-H+ exchange by the ionophore can account for the alkaline shifts observed in the HCO-3 Ringer.

If ionomycin raises cytosolic Ca2+ via direct inward transport of Ca2+ across the plasma membrane (however, see Morgan and Jacob 1994), then physiological influx of Ca2+ across the plasma membrane might also trigger a rise in pHi. Thus, during neural activity, in addition to the DIA mediated by electrogenic Na+-HCO-3 cotransport, a second component of alkalinization might arise due to entry of Ca2+ across voltage-gated channels (MacVicar et al. 1991). However, in the present experiments, a DIA was not seen in the absence of Na+ (Fig. 7), although the resulting depolarization would be expected to open voltage-gated Ca2+ channels. Given the large fall in pHi that occurred on withdrawal of Na+, and the sensitivity of Ca2+ channels to H+ (Iijima et al. 1986; Tombaugh and Somjen 1997), it is possible that this conductance pathway was not sufficiently activated. At present, however, the relationship between voltage-gated Ca2+ entry and astrocyte pH remains to be clarified.

The mechanism by which mGluR activation and Ca2+ elevation trigger alkalinization of astrocytes is not known. The dependence on HCO-3 ions is a principal feature of the ACPD response. Because the alkaline shift is blocked by neither stilbenes nor amiloride, and persists in the absence of Na+ and Cl-, classic plasmalemmal acid transporters can be excluded, such as Cl--HCO-3 exchange, Na+-H+ antiport, Na+-coupled Cl--HCO-3 exchange, and Na+-HCO-3 cotransport.

It is notable that depolarization-induced alkaline shifts in glia have been reported in Na+-free media (Grichtchenko and Chesler 1994b). In addition, in some cultured cells, recovery from acid loading has been observed in the absence of Na+ (Wuttke and Walz 1990). Pappas and Ransom (1993) reported a component of acid extrusion that was blocked by bafilomycin, a specific v-type H+-ATPase inhibitor (Bowman et al. 1988). However, the ACPD-evoked alkaline shift was not blocked by a variety of v-type ATPase inhibitors, including bafilomycin, concanamycin, or NBD-Cl.

These pharmacological data suggest a mechanism that does not involve a typical v-type ATPase. However, to exclude this class of mechanism based solely on sensitivity to these particular agents may be premature. If it is assumed that the ACPD response is due to a transmembrane net acid efflux, then an ATP-coupled mechanism must be entertained, because the response requires a source of energy other than the Na+ gradient. In this context, it is notable that some v-type H+-ATPases may be stimulated by HCO-3 ions (Kimelberg and Bourke 1973), a feature consistent with the HCO-3 dependence of the ACPD response.

An alternate possibility is that the ACPD response is due to organellar transport or metabolic processes confined to the cytosol. In reactive hippocampal astrocytes, a cytosolic mechanism was proposed for a component of a DIA that was independent of Na+ and apparently unrelated to acid extrusion (Grichtchenko and Chesler 1994b). Thus an important step in the further characterization of the ACPD-induced alkalosis will be to determine whether the cytosolic alkalinization is mediated by acid extrusion across the plasma membrane.

The mGluR-mediated alkalinization was noted in both cultured and dissociated astrocytes. It is therefore likely that these responses are a normal feature of astrocyte physiology. A rise in astrocyte pH triggered by local release of glutamate may augment the DIA mediated by sodium bicarbonate cotransport. Alternatively, various stimuli that elevate cytosolic calcium might result in an alkalinization independent of astrocyte depolarization. The functional consequences of a rise in cytosolic pH are manifold. These include an increase in the glycolytic rate (Trivedi and Danforth 1966), enhanced gap junctional coupling (Spray et al. 1981) and glutamate uptake (Billups and Attwell 1996; Judd et al. 1996), as well as modulation of astrocyte proliferation (Pappas et al. 1994). The factors that govern the intracellular pH of astrocytes can therefore be expected to have a significant impact on the behavior of these cells in both the normal and pathological setting.

    ACKNOWLEDGEMENTS

  This work was supported by National Institute of Neurological Disorders and Stroke Grants NS-32123 and NS-34906.

    FOOTNOTES

  Address for reprint requests: M. Chesler, Dept. of Physiology and Neuroscience, New York University Medical Center, 550 First Ave., New York, NY 10016.

  Received 1 August 1997; accepted in final form 1 October 1997 .

    REFERENCES
Abstract
Introduction
Methods
Results
Discussion
References
0022-3077/98 $5.00 Copyright ©1998 The American Physiological Society



This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
D. Willoughby, N. Masada, A. J. Crossthwaite, A. Ciruela, and D. M. F. Cooper
Localized Na+/H+ Exchanger 1 Expression Protects Ca2+-regulated Adenylyl Cyclases from Changes in Intracellular pH
J. Biol. Chem., September 2, 2005; 280(35): 30864 - 30872.
[Abstract] [Full Text] [PDF]