Extracellular osmolarity modulates G protein-coupled receptor-dependent ATP release from 1321N1 astrocytoma cells

Abstract

We previously reported that ATP release from 1321N1 human astrocytoma cells could be stimulated either by activation of G protein-coupled receptors (GPCR) or by hypotonic stress. Cheema et al. (Cheema TA, Ward CE, Fisher SK. J Pharmacol Exp Ther 315: 755-763, 2005) have demonstrated that thrombin activation of protease-activated receptor 1 (PAR1) in 1321N1 cells and primary astrocytes acts synergistically with hypotonic stress to gate the opening of volume-sensitive organic osmolyte and anion channels (VSOAC) and that hypertonic stress strongly inhibits PAR1 gating of VSOAC. We tested the hypothesis that a VSOAC-type permeability might comprise a GPCR-regulated pathway for ATP export by determining whether PAR1-sensitive ATP release from 1321N1 cells is similarly potentiated by hypotonicity but suppressed by hypertonic conditions. Strong hypotonic stress by itself elicited ATP release and positively modulated the response to thrombin. Thrombin-dependent ATP release was also potentiated by mild hypotonic stress that by itself did not stimulate ATP export. Notably, PAR1-sensitive ATP export was greatly inhibited in hypertonic medium. Neither the potency nor efficacy of thrombin as an activator of proximal PAR1 signaling was affected by hypotonicity or hypertonicity. 1,9-Dideoxyforskolin and carbenoxolone similarly attenuated PAR1-dependent ATP release and suppressed the PAR1-independent ATP elicited by strong hypotonic stress. Probenecid attenuated PAR1-stimulated ATP release under isotonic but not mild hypotonic conditions and had no effect on PAR1-independent release stimulated by strong hypotonicity. PAR1-dependent ATP export under all osmotic conditions required concurrent signaling by Ca(2+) mobilization and Rho-GTPase activation. In contrast, PAR1-independent ATP release triggered by strong hypotonicity required neither of these intracellular signals. Thus, we provide the new finding that GPCR-regulated ATP release from 1321N1 astrocytoma cells is remarkably sensitive to both positive and negative modulation by extracellular osmolarity. This supports a model wherein GPCR stimulation and osmotic stress converge on an ATP release pathway in astrocytes that exhibits several features of VSOAC-type channels.

Fig. 1.

Kinetics of basal and thrombin-stimulated ATP release from 1321N1 cells in isotonic or hypertonic media. Shown are time courses of ATP release in the presence (■) versus absence (☐) of 10 nM thrombin in 1321N1 cells bathed in 320 mosM isotonic (A) or 380 mosM hypertonic (B) basal saline solution (BSS). Values are means ± SE (n = 4). *P < 0.05.

Fig. 2.

Basal and thrombin-stimulated ATP release from 1321N1 cells is inversely correlated with extracellular osmolarity. A: extracellular ATP at 10 min following transfer to BSS with the indicated osmolarities in the absence (open bars) or presence (filled bars) of 10 nM thrombin. Values are means ± SE (n = 4). *P < 0.05. B: basal and thrombin-activated ATP release in isotonic 320 mosM saline or 380 mosM hypertonic BSS generated by addition of 60 mM mannitol or 30 mM NaCl. Values are means ± SE (n = 8, 320 mosM; n = 4, 380 mosM). *P < 0.05.

Fig. 3.

Concentration-response relationships for thrombin-stimulated ATP release and Ca²⁺ mobilization in isotonic, hypotonic, or hypertonic media. A: thrombin (0–15 nM) stimulated ATP release in 1321N1 astrocytes incubated in BSS of 250 mosM (▼), 320 mosM (■), and 380 mosM (▲). Values are means ± SE; n = 3. B: fura-2-loaded 1321N1 cells were suspended in BSS with the indicated osmolarity and stimulated with 0–15 nM thrombin. Peak changes in cytosolic Ca²⁺ concentration ([Ca²⁺]) were determined. Values are means ± SE; n = 3.

Fig. 4.

Differential inhibitory effects of BAPTA and Clostridial toxin B on ATP release stimulated by thrombin versus strong hypotonic stress. A: basal and 10 nM thrombin-stimulated ATP release from 1321N1 cells loaded with (open bars) or without (filled bars) BAPTA in BSS with the indicated osmolarity. Values are means ± SE (n = 4). *P < 0.05. B: basal and 10 nM thrombin-stimulated ATP release from 1321N1 cells treated with (open bars) or without (filled bars) toxin B (ToxB) in media with the indicated osmolarity. Values are means ± SE (n = 3). *P < 0.05.

Fig. 5.

Concentration-inhibition relationships for the effects of dideoxyforskolin (ddF) or carbenoxolone (CBX) on ATP release by thrombin versus strong hypotonic stress. A and C: 1321N1 monolayers bathed in isotonic 320 mosM or modestly hypotonic 250 mosM BSS were stimulated for 10 min with 10 nM thrombin in the presence of 0–300 μM ddF (A) or 0–300 μM CBX (C). B and D: 1321N1 monolayers were stimulated by transfer to strongly hypotonic 215 mosM BSS for 10 min in the presence of 0–300 μM ddF (B) or 0–300 μM CBX (D). In A–D, ATP release in the presence of ddF or CBX was normalized to the maximal release in the absence of inhibitors. Values are means ± SE (n = 4). *P < 0.05, significant differences from the normalized maximal release in the absence of inhibitors.

Fig. 6.

Concentration-inhibition relationships for the effects of probenecid (PB) on ATP release by thrombin versus strong hypotonic stress. A: 1321N1 monolayers bathed in isotonic 320 mosM (■) or modestly hypotonic 250 mosM (□) BSS were stimulated for 10 min with 10 nM thrombin in the presence of 0–3 mM PB. ATP release in the presence of PB was normalized to the maximal release in the absence of PB. B: 1321N1 monolayers were stimulated by transfer to strongly hypotonic 215 mosM BSS (●) for 10 min in the presence of 0–3 mM PB. ATP release in the presence of PB was normalized to the maximal release in the absence of PB. C: 1321N1 monolayers were bathed in BSS with the indicated osmolarity and then incubated for 10 min with no other additions (open bars), with 10 nM thrombin alone (filled bars), or with 10 nM thrombin plus 2 mM PB (hatched bars). Values in A–C are means ± SE (n = 4). *P < 0.05, significant differences from the normalized maximal release in the absence of PB. D: fura-2-loaded 1321N1 cells were suspended in BSS with or without 2.5 mM PB and then stimulated with 10 nM thrombin. Peak changes in cytosolic [Ca²⁺] were determined. Values are the average ± range; n = 2.

Fig. 7.

The maxi-anion channel inhibitor Gd³⁺ does not inhibit thrombin-dependent or hypotonic stress induced ATP release from 1321N1 cells. A: changes in extracellular [ATP] in unstimulated cells (☐) versus cells stimulated with 10 nM thrombin in the absence (■) or presence (●) of 50 μM Gd³⁺ were recorded online every 4 min. B: changes in extracellular [ATP] in cells bathed in isotonic (320 mosM) (☐) versus hypotonic (215 mosM) BSS in the absence (■) or presence (●) 50 μM Gd³⁺ were recorded online every 4 min. Values are means ± SE (n = 7).