Neurochemical Research, VoL 18, No. 6, 1993, pp. 681-687

Extracellular ATP Stimulates Inositol Phospholipid Turnover and Calcium Influx in Ca Glioma Cells W a n - W a n Lin 1 and D e . M a w C h u a n g 2,3 (Accepted October 19, 1992)

Extracellular ATP caused a dose-dependent accumulation of inositol phosphates and a rise in cytosolic free Ca2+ ([Ca2§ in C6 glioma cells with an ECso of 60 + 4 and 10 +- 5 IxM, respectively. The threshold concentration of ATP (3 g,M) for increasing [Ca2+]i was approximately 10-fold less than that for stimulating phosphoinositide (PI) turnover. The PI response showed a preference for ATP; ADP was about 3-fold less potent than ATP but had a comparable maximal stimulation (11-fold of the control). AMP and adenosine were without effect at concentrations up to 1 mM. ATP-stimulated PI metabolism was found to be partially dependent on extracellular Ca 2§ and Na + but was resistant to tetrodotoxin, saxitoxin, amiloride, ouabain, and inorganic blockers of Ca z+ channels (CC § Mn 2+, La 3+, or Cd2+). In Caa+-free medium, ATP caused only a transient increase in [Ca2+]i as opposed to a sustained [Ca2+]~ increase in normal medium. The ATP-induced elevation of [Ca2+]~ was resistant to Na + depletion and treatment with saxitoxin, verapamil and nisoldipine, hut was attentuated by La 3+. The differences in the characteristics of ATP-caused P1 hydrolysis and [Ca2+]i rise suggest that ATP receptors are independently coupled to phospholipase C and receptor-gated Ca 2§ channels. Because of the robust effect of ATP in stimulating PI turnover and the apparent absence of Pl-purinergie receptors, the C6 glioma cell line provides a useful model for investigating the transmembrane signalling pathway induced by extracellular ATP. The mechanisms underlying the unexpected finding of [Na+]o dependency for ATP-induced PI turnover require further investigation. KEY WORDS: C6 glioma cells; phosphoinositide metabolism; Ca2§ influx; ATP receptors; calcium channels.

INTRODUCTION

ripheral tissues (for review, see 1). ATP has been reported to activate the phosphoinositides (PI)/protein kinase C signaling cascade via P2- purinergic receptors, thus producing a rise in cytosolic free Ca 2+ ([Ca2+]i) in many cell types including HL60 cells, neutrophils, monocytes, mastocytoma cells, cerebral astrocytes, hepatocytes, thyroid cells, erythrocytes, endothelial cells, vascular smooth muscle cells, vas deferens cells, and Ehrlich ascites tumor cells (for review, see 2,3). In addition ATPinduces Ca 2+ -entry in various cell types by the opening of receptor-gated channels (2,3). It is well known that inositol trisphosphate (IP3) produced from the hydrolysis of PI triggers Ca 2+ release from internal stores. However, the Ca 2+ influx mediated by many types of Ca 2+ mobilizing receptors is a more

Extracellular ATP elicits a broad range of physiological responses in the central nervous system and peDepartment of Pharmacology, College of Medicine, National Taiwan University, Taipei, Taiwan, R.O.C. z Biological Psychiatry Branch, National Institute of Mental Health, Bethesda, Maryland 3 Address reprint requests to: De-Maw Chuang, Ph.D., Biological Psychiatry Branch, NIMH, Bldg. 10, Room 3N212, 9000 Rockville Pike, Bethesda, MD 20892. ABBREVIATIONS: PI, phosphoinositide; [Ca2+]i, cytosolic free Ca2§ concentration; PDBu, phorbol 12, 13-dibutyrate; PSS, physiological saline solution; IP, inositol phosphates; IPt, inositol monophosphate, IP2, inositol bisphosphate; IP3, inositol trisphosphate; IP4, inositol (1,3,4,5) tetrakisphosphate; PLC, phospholipase C. 681

0364-3190/93/0600-068I$07,00/0 9 1993 Plenum Publishing Corporation

682

complex process and several hypotheses have been proposed (4,5). It has been proposed that inositol phosphates regulate the entry of Ca 2+ into the cell by (i), an IP3-mediated Ca z+ channel in the plasma membrane (6); (ii) a capacitative model in which IP 3 controls a Ca 2+ channel in an endomembrane compartment that indirectly regulates Ca 2+ entry through a negative feedback mechanism (7) or (iii) Ca 2+ entry mediated by inositol (1,3,4,5) tetrakisphosphate (IP4) either by opening Ca 2+ channels in the plasma membrane or by enhancing extracellular Ca2+-dependent refilling of the IP3-sensitive intracellular stores (8). In addition, receptor-gated Ca 2+ entry provides another mechanism by which the cell may increase its [Ca2*]i (5). Recent studies have unravelled some of the mechanisms for the purinergic receptor-mediated rise of [Ca 2+]i; however, the molecular details remian largely unknown. In this study, we used rat C6 glioma ceils to further characterize the ATP-induced PI turnover and the accompanying increase in [Ca2+]i. We demonstrated that ATP-induced PI turnover is partially dependent on extracellular Ca 2+ and Na + . The ATP-induced Ca 2+ influx leads to an enhancement of PI turnover but is not dependent on intracellular Ca 2+ mobilization. Therefore, the C6 glioma cell line provides an attractive model system to further investigate the transmembrane signaling pathway mediated by extracellular ATP and to screen for drugs affecting the ATP receptors.

EXPERIMENTAL PROCEDURE Materials. Dulbecco's modified Eagle medium and fetal calf serum were obtained from Gibco (Grand Island, NY). ATP, ADP, AMP, adenosine, ouabain, amiloride, saxitoxin, verapamil, and phorbol 12, 13-dibutyrate (PDBu) were purchased from Sigma Chemical Co. (St. Louis, MO). Fura-2 AM and BAPTA AM were purchased from Molecular Probes (Eugene, OR). Nisoldipine was a gift from Miles Laboratories (West Haven, CT). Cell Culture. Ca glioma cells obtained from the American Type Culture Collection (Rockville, MD), with a passage number of 39, were grown at 37~ in Dulbecco's modified Eagle medium supplemented with 10% fetal calf serum, penicillin (100 U/ml), and streptomycin (100 Ixg/ml) in the presence of 5% CO2 in a humidified incubator. Cells with additional passage number between 13 and 35 were cultured in 35-mm Petri dishes for the study of Pl hydrolysis or in 9 x 35 mm 2 cover slips for the measurement of cytosolic Caa+. Measurement of PI Hydrolysis. The hydrolysis of PI was expressed as the accumulation of inositol phosphates (IP) in the presence of 20 mM LiCI as described (9). Briefly, C6 cells grown to confluency in 35-mm dishes (about 4 x 106 cells/dish) were labeled overnight with 2.5 ~Ci/dish of ~H-myo-inositol (16.5 Ci/mmol; New England Nuclear, Boston, MA). The labeled cells were washed twice with a physiological saline solution (PSS) (118 mM NaCI, 4.7 mM KCI, 3.0 mM CaCI2, 1.2 mM MgClz, 1.2 mM KH2PO4,0.5 mM EDTA, 10 mM glucose, and 20 mM HEPES [pH 7.4]) and then preincubated at

Lin and Chuang 37~ for 45 rain in 1 ml of PSS containing 20 mM LiC1; at this concentration of lithium, ATP-induced IP formation was maximally amplified. The indicated drugs were added to the Li+-containing PSS and the mixtures were then incubated for 45 rain. The reaction was terminated by addition of ice-cold methanol and the accumulated 3HIP, predominantly inositol monophosphate (IP0, were isolated on a AG-1X8 column (formate form, 100-200 mesh) and eluted with 0.2 M ammonium formate/0.1 N formic acid as originally described by Berridge et al (10). When indicated, inositol bisphosphate (IPa) and IP3 were eluted with 0.4 and 1 M ammonium formate/0.1 N formic acid, respectively. Measurement of Cytosolic Ca2§ Cells grown on a cover slip (9 • 35 mm 2) to confluency (approximately 1 x 10a cells/slip) were loaded with 5 BxMFura-2 AM and pluronic F-127 (0.25%; v/v) in PSS at 37~ for 60 rain. The slip was then placed in a continuously stirred thermostatic curvette maintained at 37~ and the fluorescence was monitored on a PTI Delta Scan spectrofluorometer with dual excitation wavelengths of 340 and 380 nm and an emission wavelength of 510 nm. [Ca2*]~was calculated as described by Grynkiewicz et al (11).

RESULTS

ATP-Stimulated PIBreakdown. Prelabeled C6 glioma cells incubated with ATP for 45 rain in the presence of 20 mM LiC1 produced a dose concentration-dependent increase in the accumulation of 3H-IP (Figure 1). The maximal increase occurred at 0.3 mM ATP and was about ll-fold of the control. The enhancement stimulation of the PI response showed a preference for ATP; ADP was about three times less potent than ATP (ECso = 60 ---4-4 g,M for ATP and 174 + 18 p~M for ADP), although comparable maximal responses were observed at 1 mM of either drug. AMP and adenosine were without effect at 1 mM (Figure 1). In the presence of Li +,

0

1400-

c o 0

1200-

r

O ~

;/;J

~ooo-

800600

-

400

-

200

-

c

i

,"O

01

//

oj # i5

: 4

f

i 3

I 2

Concentraf.ion(-Log M) Fig. 1. Dose-dependent stimulation of 3H-IP accumulation by purines. Cells prelabeled with 3H-myo-inositol were exposed to ATP (9 ADP (A), AMP (IN), or adenosine (O) at various concentrations for 45 min and the accumulation of 3H-IP was measured. Values (expressed as % of control) represent the mean _+ SEM of 5 independent triplicate experiments.

Phosphoinositide Turnover and

C A 2+

Influx Induced by ATP

ATP (1 mM) induced a rapid formation of labeled IP3, IP2 and IPG the increase in IP 3 and IP 2 was detected at 15 sec after stimulation (Figure 2A). The percentage of increase for IP3 and IP2 reached a peak at 15 sec and 3 min, respectively; after this time their levels declined (Figures 2A and B). The increase in IPl was time-dependent and peaked between 30 and 45 min. When cells were incubated in the absence of Ca 2§ in PSS and in the presence of 0.5 mM EDTA, the basal PI response was reduced by 44 _+ 8% (n = 3). Under this Ca2+-free condition, the response to ATP (1 raM) expressed as a percentage of the basal value (337 _ 5%) was reduced by 72% compared with that found in normal PSS (1181 _ 147%) (Figure 3A). When PI hydrolysis was studied in Na+-free PSS, in which NaC1 was isoosmolatically replaced by choline chloride or sucrose, we found that the ATP response at 1 mM was significantly attenuated to 62 +_ 2% (n = 4) and 63 _ 7% (n

683

%- ~4oo ,~ ~2oo ~ moo ~ 800 ~cl. -6 ..,z_

6oo 40o:

-~-

200!

A

w

to

o Normal PSS

--~"

8

Ca 2 + - f r e e

1000.

800600400-

~-.

A

300

.,.a tO o

'~ c I I to

IP3 r -1:3------_

15

[]

~5 zoo.

~ ;

IP2

_

t'0 cJ

1O0

o n

I T

0 0

I

I

I

I

15

30

45

6O

200-

NormaIPSS

Na+-free Na+-free (Choline CI) (Sucrose)

Fig. 3. Effects of removal of extracellular Ca 2+ (A) and Na § (B) on ATP-induced PI turnover. Cells were preincubated in Ca 2§ - or Na § free PSS containing 20 mM LiC1 for 45 min then challenged with ATP for another 45 min. In the Ca2+-free experiment, 3 mM CaC12 was omitted from PSS. In the Na+-free experiment, 118 mM NaG was iso-osmolatically replaced with either choline chloride or sucrose. Each value represents the mean _ SEM from 3-5 independent triplicate experiments. *, P <0.05 as compared with the response in normal PSS using the Student's t test.

Time (seconds) ~'-

1000

-

S IP 1

c

o U

800-

6

l

~

o

"6 ~s

600

/

c 0

400

E O

//

~

00

I1.

T

-r"

0

20

I

I

40

6o

Time (minutes) Fig. 2. Time-course of ATP-induced PI hydrolysis in C~ glioma cells. Cells prelabeled with 3H-myo-inositol were stimulated with 1 mM ATP for the indicated time in the presence of lithium. Various forms of 3HIP1 (IPa [D], IP2 [ , ] , and IPl [O]) were separated by anion-exchange chromatography. Data presented are the mean _+ SEM of three independent triplicate experiments.

= 5) of the control response in choline chloride- and sucrose-replaced medium, respectively. In contrast, the basal PI response was not significantly affected by removal of external Na + (Figure 3B). Figure 4 shows that depletion of Na + resulted in a marked inhibition of the PI response to ATP in the range of 100 txM to 1 raM. Interestingly, the response to lower concentrations of ATP (30 IxM) was unaffected by external Na + depletion. The ATP (1 mM)-induced PI response was unchanged by pretreatment for 15 min with inhibitors of voltagegated Na + channels (3 ~M tetrodotoxin or 1 ~M saxitoxin), Na+/Ca 2+ and Na+/H + exchanger (1 mM amiloride), Na +, K+-ATPase (100 IxM ouabain), and inorganic blockers of Ca2+ channels (1 mM Co2+, Mn2§ L a 3+, or C d 2+) (data not shown).

Effects of ATP on Cytosolic Free Ca 2§ in C6 Glioma Cells. Figure 5 illustrates the typical changes in [Ca2+]i

684

Lin and Chuang 2000

ATP (1 mM) was still able to induce a relatively weak [CaZ+]i increase (Figure 6). In Ca2+ -free PSS containing 5 mM EGTA, only the initial Ca2+-transient was triggered by ATP and the maximal [Ca2+]~ increase by 100 ~M ATP was inhibited by 57 -+ 6%. Short-term (10 min) exposure of cells with 500 nM PDBu further attenuated this transient Ca2+ spike. Organic blockers of Ca2+ channels verapamil and nisoldipine failed to inhibit the response to ATP. On the contrary, La 3+ (1 mM) attenuated this response by about 50%.

-

<

~D O0 ~- uu DO

1000"

NORMAL PSS

o~

.

0

/ /

j -5

~

.

'

, -'3

-4

-2

ATPCONCENTRATION(IogM) Fig. 4. Effects of Na § depletion on the dose-dependent stimulation of PI turnover by ATP. Values represent the mean _ SEM from three independent triplicate experiments. *, P<0.05 as compared with the response in normal PSS using the Student's t test.

DISCUSSION Extracellular ATP is known to induce a variety of responses by acting on purinergic receptors in several cell types (1,2). Purinergic receptors have been classified into two groups based on agonist selectivity (2): ATP- (and/or ADP)- preferring P2-purinoceptors and adenosine-preferring Pl-purinoceptors. In this study, we found that exogenous ATP triggered PI turnover and increased intracellular free Ca2+ in C6 glioma cells. The ECso of ATP for PI hydrolysis metabolism and [Ca2+]i increase were 60 + 4 and 10 + 5 ~M, respectively. Considering that the rank order of potency for PI turnover was ATP > ADP > > AMP or adenosine, it may be inferred that ATP-induced response in C6 glioma cells is due to an action on P2- purinergic receptors which are coupled to phospholipase C (PLC). We have shown that this ATP-mediated PI response in C6 glioma cells was insensitive to pertussis toxin (9),

observed in Fura 2-loaded cells. Both transient and sustained [Ca2+]i increases were triggered by stimulation with ATP. The net maximal increase of the initial phase of [Ca2+]i elicited by ATP was dose-dependent in the range of I ~M to 1 raM, with a half-maximal stimulation concentration (ECso) being 10 _+ 5 IxM. Figure 6 and Table I show the effects of various pharmacological manipulations on the [Ca2+]i increase induced by 100 I~M ATP. Unlike PI turnover, the [Ca2+]~ increase elicited by ATP was unaffected by isoosmotic substitution for Na + in PSS. Saxitoxin (1 ~M) pretreatment also had no effect (data not shown). In cells pretreated with BAPTA to chelate intracellular free Ca2+,

40O

A SO0

&

5oo-

- 400

B

200 400'

200 t

ATP 13~ kTP

8001

C"

(to ,u~

v

300

+

BOO

200

500F

t.

~

600 <~]

400

400

200~

200

,oo

(seconds) kTP t30MM) 0

kTP(Ioo)JN)

100-

-6

-5

-4

-3

~2

Log ATP (M)

I00

[seconds] Fig. 5. Dose-dependent increase in [Ca2§ by ATP. Traces shown in (A) are typical responses to ATP at various concentrations. The maximal net increase of the initial phase of [Ca2§ shown in (B) are mean _ SEM from 4-12 independent experiments.

Phosphoinositide Turnover and CA 2+ Influx Induced by ATP

400l Ho-lree

~ o t ~u-froo [5 uN EGT~

1

~ 27 7~176

200

~4DO I Li [I Id4115IIn prelreatlent Na-lree

~

eooI ~ lip[I00~141

400

I1PIO l ON~ Fig. 6. Effects of pharmacologicalmanipulations on [Ca2+]~rise induced by ATP. In Na+-free conditions, 118 mM NaC1was replaced by an equal molarityof choline chloride. The experimentwas repeated three times with similar results.

Table I. Effects of Various PharmacologicalAgents on ATPInduced Net [Ca2+]~Increase Pretreatment

n

Net [CaZ+]i increase (nM)

None Na+-free Caa+-free Ca2+-free, PDBu (500 nM, 10 rain) Verapamil (10 ~M, 15 min) Nisoldipine (i0 IxM, 15 rain) La3+ (1 raM, 10 rain) PDBu (500 nM, 24 hr)

12 3 6 3 3 3 5 3

421 -+26 435 -+40 180 -+26* 45-+5* 497-+70 476_+69 209-+26* 400-+5

Experimental conditions were as described in the Experimental Procedures except that cells were pretreatedwith an agent, as indicated, before stimulation with ATP (100 IxM). Each value represents the mean -- SEM from at least 3 experiments. *, P < 0.05 compared with the control response using the Student's t test. The program of Bonferroni T was also used at an overall level of 0.05 to test the pairwise differences among the means. The pairs that are statistically significantly different using this procedure include untreated control versus CaZ+-free;untreated control versus CaZ+-freeplus PDBu; untreated control versus La3+; Ca2+-freeversus CaZ+-freeplus PDBu and La3+ versus Ca2+-freeplus PDBu.

suggesting that ATP stimulates PLC via a pertussis toxininsensitive G protein. Moreover, in agreement with results for receptor-mediated PI breakdown, ATP-induced PI turnover is partially blocked by short-term treatment with PDBu, suggesting that a specific protein participates in the ATP-induced activation of PLC and is phosphorylated by protein kinase C (9). In the absence of free extracellular Ca 2+, a very rapid and transient increase in [Ca2+]i was induced by ATP and this transient

685

Ca 2+ spike was inhibited by PDBu, consistent with the role of IP 3 in Ca 2+ mobilization. In addition to the intracellular Ca 2+ release, ATP appears to trigger extracellular Ca 2+ influx. The finding that ATP elicited a [Ca2+]i increase, even under conditions when intracellular Ca 2+ was rapidly chelated by BAPTA, supports such a notion. The pathways involved in the activation of Ca 2+ influx are unlikely to be L-type Ca z+ channels, or either Na+/Ca 2§ or Na+/H + exhange because of the lack of effect of organic blockers of Ca z§ channels, Na+-depletion and amiloride. The observation that the [CaZ+]i increase was inhibited 50% by La 3+ can be explained by its inhibitory effect on both voltagesensitive and voltage-insensitive Ca 2§ influx pathways. We have recentIy reported that [Ca2+]~ in C6 glioma cells is unchanged by high K + (50 raM) treatment, suggesting that voltage-sensitive Ca 2+ channels are not expressed in these cells (12). To examine whether ATP-induced Ca 2+ influx is a consequence of stimulation of PI turnover, the characteristics of ATP-induced Ca z+ influx and PI turnover were compared. Moreover, since iF4 production is stimulated by ATP in DDTIMF 2 smooth muscle cells (13) and plays an important role in receptor-mediated Ca z§ entry in some cells (4,14), one needs to consider a possible role for IP 4 in ATP-induced Ca 2+ influx. However, this possibility seems to be unlikely for several reasons. First, the stimulatory effects of ATP on the [Ca2+]~ increase occurred at concentrations as low as 3 IxM, whereas stimulation of PI turnover was detected only at higher concentrations (~ 30 IxM). Second, La 3+ at a concentration that inhibited the [Ca2+]i rise, did not block the PI response. Third, PI turnover, but not Ca 2+ signals, was inhibited by Na+-depletion. Fourth, long-term pretreatment of cells with PDBu decreased the PI response to ATP (9) but failed to change the [Ca2+]i increase. Thus, the underlying mechanism for the Ca 2§ influx p r o duced by A T P probably involves receptor-gated ion channels rather than ATP-induced IP4 formation. Voltage clamp studies have demonstrated that ATP activates a cation-permeable current that reverses near 0 mV in vascular and visceral smooth muscIe cells (15-17), skeletal muscle cells (18), artial cells (19), and neurons (20,21). It may be that, similar to glutamate receptors in astrocytes (22), ATP receptors in C6 glioma cells are coupled to PI turnover and receptor-gated channels. It is unknown at this time whether both responses are mediated by the same receptor or distinct receptor subtypes. An increase in [Ca2+]i directly activates PLC in numerous systems. The effects of ATP on PI metabolism are partially dependent on extracellular [Ca2+]o , suggesting that extracellular Ca 2+ influx is facititatory but

686

not absolutely essential for the activation of PLC coupled to ATP receptors. The ATP-evoked 3H-IP accumulation found in the absence of extracellular Caz+ (28% of the response in normal PSS) may result from an initial PLCmediated breakdown of phosphatidylinositol 4,5-bisphosphate, because IP3-mediated Ca2+ mobilization is still observed in Ca2+-free medium. It is noteworthy that in the case of ATP-induced PI turnover, stimulation only occurred when [CaZ+]i was increased to 400 nM by the presence of 30 o~M of this agonist. At 10 IxM, although the net [CaZ+]i was increased up to 228 _+ 87 nM, no significant increase in IP accumulation was observed. Recently, we found that ionomycin-induced PI breakdown was detected only when cytosolic Ca2§ was increased to approximately 1100 nM by 10 txM of this ionophore (12). Thus, PLC is more sensitive to activation by the [Ca2+]i increase elicited by extracellular ATP than by the Ca2+ ionophore. This preference may be explained by synergistic activation of PLC when G protein-coupled PLC was stimulated by the corresponding receptor agonist in the presence of slightly elevated [Ca2+]i. The superactive state of PLC involving these dual mechanisms has been documented (23,24). Unexpectedly, the absence of extraceltular Na § attenuated the efficacy of ATP with respect to PI breakdown, while not reducing the effects of ATP on stimulation of [Ca2+]i. This dependency of PI hydrolysis on [Na+]o is not a general phenomenon for Ca2+ -mobilizing receptor agonists. For example, the PI response to endothelin-1 in C6 glioma cells was unaffected by depletion of extracellular Na + (25). Our preliminary results using NG108-15 neuroblastoma cells also indicate an unusual dependency on [Na+]o for ATP-induced PI hydrolysis. Voltage-dependent Na § channel activators and the Na + ionophore (monensin) have been reported to promote PI hydrolysis in different tissues (26,27), possibly due to a rise in [CaZ+]i resulting from activation of Na+/Caz+ exchange (28), membrane depolarization (29), and/or Na § influx (29). Consistent with this notion is the observation that the plasma membrane channels coupled to P2 -purinoceptors are permeable to monovalent ions in rat mast cells (30), mouse macrophages (31), and a variety of transformed cells (32,33). In C6 glioma cells, we found that monensin also induced the stimulation of PI turnover which was dependent on extracellular Na + and partially inhibited by amiloride (W.W. Lin, unpublished observations). The ineffectiveness of Na + channel blockers and amiloride in blocking ATP-induced PI turnover in C6 glioma cells speaks against the involvement of pathways such as activation of voltage-dependent Na + channel, Na+/I-I+ exchange, and Na+/Ca 2+ exchange. Taken together, these results suggest a role for Na § entry in

Lin and Chuang

ATP-induced PI hydrolysis in C6 glioma cells. However, the underlying mechanism(s) for this Na+-dependent PI turnover caused by ATP remains to be explored. In addition to ATP-induced stimulation of PI turnover and Ca2+ influx reported in this study, we confirmed the previous finding that ATP inhibited agoniststimulated adenylate cyclase activity via action of P2purinoceptors in this cell line. However, in contrast to ATP, we found that adenosine failed to affect basal PI turnover (Figure 1) or cAMP levels (data not shown) at concentrations up to 1 mM. Moreover, adenosine at 1 mM neither affected ATP-stimulated PI metabolism (data not shown) nor inhibited isoproterenol-elicited cAMP formation (34). Taken together, these results indicate that adenosine-preferring Pl-purinoceptors are not expressed in C6 glioma cells. Since activation of Pl-purinoceptors is known to inhibit or stimulate agonist-induced PI breakdown (35,36) and adenylate cyclase activity (2), the co-presence of P1- and P2-purinoceptors might interfere with the investigation of signalling pathways coupled to P2-purinoceptors. In support of this notion, in FRTL-5 thyroid cells, adenosine via A1 receptors enhanced P2-purinoceptor-mediatedPI turnover (37). Thus, the absence of Pl-purinoceptors and the robust activity of ATP is stimulating PI hydrolysis in C6 glioma cells would make this cell line an ideal model for studying the signal transduction mediated by P2-purinceptors, for classifying the P2 receptor subtypes, and for screening new drugs acting on the P2 receptors. In summary, our results demonstrate that functional receptors activated by extracellular ATP in C6 glioma cells are linked to PI metabolism and Ca2+ influx. The Ca2+ entry is mediated by ATP receptor-gated channels and may further amplify the PI signaling pathway. The dependency of [Na+]o for the ATP-induced PI turnover but not Ca2+ entry, provides a useful tool to further investigate molecular mechanisms underlying the PI response.

ACKNOWLEDGMENT This work was supported in part by a research grant from the National Science Council (NSC 81-0412-B002-28) of the Republic of China.

REFERENCES 1. Gordon, J. L. 1986. Extracellular ATP: effects, sources and fate. Biochem. J, 233:309-319. 2. Burnstock, G. 1990. Purinergic mechanisms. Ann. N.Y. Acad. Sci. 603:1-17.

Phosphoinositide Turnover and CA 2§ Influx Induced by ATP 3. E1-Moatassim, C., Dornand, J., and Mani, J. C. 1992. Extracellular ATP and cell signalling. Biochim. Biophys. Acta 1134:3145. 4. Berridge, M. C., and Irvine, R. F. 1989. Inositol phosphate and cell signalling. Nature 341:197-205. 5. Hallam, T. J., and Rink, T. J. 1989. Receptor-mediated Ca2+ entry: diversity of function and mechanism. Trends Pharmacol. Sci. 10:8-10. 6. Kuno, N., and Gardner, P. 1987. Ion channels activated by inositol 1,4,5-trisphosphate in plasma membrane of human T-lymphocytes. Nature 326:301-304. 7. Putney, J. W. 1986. A model for receptor-regulated calcium entry. Cell Calcium 7:1-12. 8. Putney, J. W. 1987. Formation and actions of calcium-mobilizing messenger, inositol 1,4,5-triphosphate. Am. J. Physiol. 252:G149157. 9. Lin, W. -W., Lee, C. Y., and Chuang, D. -M. 1990. Comparative studies of phosphoinositide hydrolysis induced by endothelin-related peptides in cultured cerebellar astroeytes, C~-glioma and cerebellar granule cells. Biochem. Biophys. Res. Commun. 168:512-519. 10. Berridge, M. J., Downes, C. P., and Hanley, M. R. 1982. Lithium amplifies agonist-dependent phosphatidyl-inositol responses in brain and salivary glands. Biochem. J. 206:587-595. 11. Grynkiewicz, G., Poenie, M. and Tsien, R. Y. 1985. A new generation of Ca2+ indicators with greatly improved fluorescence properties. J. Biol. Chem. 260:3440---3450. 12. Lin, W. -W., and Chuang, D. -M. 1992. Potentiation by Caa+ ionophores and inhibition by extracellular KC1 of endothelin-induced phosphoinositide turnover in Ca glioma cells. Neurochem. Int. 21:293-301. 13. Hoiting, B., Molleman, A., Nelemans, A., and Hertog, A. D. t990. P2-purinoceptor-activated membrane currents and inositol tetrakisphosphate formation are blocked by suramin. Eur. J. Pharmacol. 181:12%131. 14. Downes, C. P. 1988. Inositol phosphates: a family of signal molecules? Trends Neurosci. 11:336-339. 15. Benham, C. D., and Tsien, R. W. 1987. A novel receptor-operated Ca2+-permeable channel activated by ATP in smooth muscle. Nature 328:275-278. 16. Nakazawa, K., and Matsuki, N. 1987. Adenosine triphosphateactivated inward current in isolated smooth muscle from rat vas deferens. Pflugers Arch. 409:644-646. 17. Friel, D. D. 1988. An ATP-sensitive conductance in single smooth muscle cells from the rat vas deferens. J. Physiol. 401:361-380. 18. Hume, R. I., and Hong, M. G. 1986. Excitatory action of ATP on embryonic chick muscle. J. Neurosci. 6:681-690. 19. Friel, D. D., and Bean, B. P. 1988. Two ATP-activated conductances in bullfrog atrial cells. J. Gen. Physiol. 91:1-27. 20. Y,2-ishtal,O. A., Marchenko, S. M., and Pidoplichko, V. I. 1983. Receptor for ATP in the membrane of mammalian sensory neurons. Neurosci. Lett. 35:41-45. 21. Bean, B. P. 1990. ATP-activated channels in rat and bullfrog sensory neurons: concentration dependence and kinetics. J. Neurosci. 10:1-10. 22. Glaum, S. R., Holzwarth, J. A., and Miller, R. J. 1990. Glutamate receptors activate Ca2. mobilization and Ca2§ influx into astrocytes. Proc. Natl. Acad. Sci. USA 87:3454-3458.

687

23. Diamant, S., and Atlas, D. 1989. Potentiation of [3H]inositol phosphate formation by receptor activation and membrane depolarization in brain cortical slices (1). Brain Res. 503:55-61. 24. Kendall, D. A., and Nahorski, S. R. 1984. Inositol phospholipid hydrolysis in rat cerebral cortical slices. II. Calcium requirement. J. Neurochem. 42:1388-1394. 25. Lin, W. W., Kiang, J. G., and Chuang, D. -M. 1992. Pharmacological characterization of endothelin-stimulated phosphoinositide breakdown and cytosolic free Ca2+ rise in C6 glioma cells. J. Neurosci. 12:1077-1085. 26. Gusovsky, F., Hollingsworth, E. B., and Daly, J. W. 1987. Stimulation of phosphoinositide breakdown in brain synaptoneurosomes by agents that activate sodium influx: antagonism by tetrodotoxin, saxitoxin and cadmium. Mol. Pharmaeol. 32:479487. 27. McDonough, P. M., Ooldstein, D., and Brown, J. H. 1988. Elevation of cytoplasmic calcium concentration stimulates hydrolysis of phosphatidylinositol bisphosphate in chick heart cells: effect of sodium channel activators. Mol. Pharmacol. 33:310-315. 28. Benuck, M., Reith, M. E. A., and Lajtha, A. 1989. Evidence for the involvement of Na+/Ca ++ exchange in the stimulation of inositol phospholipid hydrolysis by sodium channel activation and depolarization. Eur. J. Pharmacol. 159:187-190. 29. Guiramand, J., Vignes, M., Mayat, E., Lebrun, F., Sassetti, I., and Recasens, M. 1991. A specific transduction mechanism for the glutamate action on phosphoinositide metabolism via the quisqualate metabotropic receptor in rat brain synaptoneurosomes: I. External Na ~ requirement. J. Neurochem. 57:1488-1500. 30. Dahlquist, R. 1974. Relationship of uptake of sodium and "%alcium to ATP-induced histamine release from rat mast cells. Acta Pharmacol. Toxicol. 35:11-22. 31. Steinberg, T. H., and Silverstein, S. C. 1987. Extracellular ATP4promotes cation fluxes in the J774 mouse macrophage cell line. J. Biol. Chem. 262:3118-3122. 32. Weisman, G. A., Dunn, S. D., De, B. K., Kitagawa, T., and Friedberg, I. 1984. Cellular responses to external ATP which precede an increase in nucleotide permeability in transformed cells. J. Cell Physiol. 119:211-219. 33. Gonzalez, F. A., Wang, D., Huang, N., and Heppel, L. A. 1989. Two distinct receptors for ATP can be distinguished in Swiss 3T6 mouse fibroblasts by their desensitization. Biochem. Biophys. Res. Commun. t64:706-713. 34. Pianet, J., Merle, M., and Labouesse, J. 1989. ADP and, indirectly, ATP are potent inhibitors of cAMP production in intact isoproterenol-stimulated C6 glioma cells. Biochem. Biophys. Res. Commun. 163:1150-1157. 35. Linden, J., and Delahunty, T. M. 1989. Receptors that inhibit phosphoinositide breakdown. Trends Pharmacol. Sci. 10:114--120. 36. El-Etr, M., Matin, P., Tence, M., Delumean, J. C. Cordier, J., Glowinski, J., and Premont, J. 1992. 2-Chloroadenosine potentiates the al-adrenergic activation of phospholipase C through a mechanism involving arachidonic acid and glutamate in striatal astrocytes. J. Neurosci. 12:1363-1369. 37. Okajima, F., Sato, K., Nazarea, M., Sho, K., and Kondo, Y. 1989. A permissive role of pertussis toxin substrate G-protein in P2-purinergic stimulation of phosphoinositide turnover and arachidonate release in FRTL-5 thyroid cells. J. Biol. Chem. 264:1302913037.