ATP induces Ca2+ release from IP3-sensitive Ca2+ stores exclusively in large DRG neurones

Neuropharmacology and Neurotoxicology NeuroReport 8, 1555–1559 (1997)

1

PURINORECEPTOR-MEDIATED intracellular Ca2+ release was studied in freshly isolated adult mouse dorsal root ganglia (DRG) neurones. The cytoplasmic Ca2+ concentration ([Ca2+]i) was measured using indo-1-based microfluorimetry. The application of 100 mM ATP in Ca2+-free solution triggered an increase in [Ca2+]i in 93% of large DRG neurones but in no small ones. The ATP-induced [Ca2+]i transients in large neurones were inhibited by cells incubation with thapsigargin or by intracellular dialysis with heparin-containing solution. The ATP-triggered increase in [Ca2+]i was not mimicked by adenosine and was blocked by suramin, suggesting the involvement of metabotropic (P2Y) purinoreceptors. We conclude that large (proprioceptive) DRG neurones express P2Y purinoreceptors linked to the inositol 1,4,5-triphosphateCa2+ intracellular signal transduction cascade, whereas small (nociceptive) DRG neurones are devoid of such a mechanism.

1 Key words: Calcium stores; Cytoplasmic calcium; inositol 1,4,5-triphosphate; Metabotropic (P2Y) purinoreceptors; Sensory neurones; Suramin; Thapsigargin

1

1

1

p

ATP induces Ca2+ release from IP3-sensitive Ca2+ stores exclusively in large DRG neurones Natalia Svichar, Anatoly Shmigol, Alexej VerkhratskyCA,1 and Platon Kostyuk Bogomoletz Institute of Physiology and International Center of Molecular Physiology, Bogomoletz St. 4, Kiev-24, GSP 252601, Ukraine. 1Present Address: Max-Delbrück Center for Molecular Medicine, Robert-Rössle Strasse 10, 13122 Berlin-Buch, Germany CA,1

Corresponding Author and Address

Introduction

Materials and Methods

The dorsal root ganglia (DRG) neurones of different modalities can be readily distinguished by the diameter of their somas. The Aa and Ab axons which transmit proprioceptive and tactile information belong to the neurones with large cell bodies, whereas nociceptive, slow conducting Ag and C axons are related to neurones with small somata.1 Numerous experiments have demonstrated that these two classes of DRG neurones differ significantly with respect to their excitable properties2 as well as the mechanisms of calcium homeostasis and signalling. Large and small DRG neurones express different patterns of voltage-gated calcium channels,3,4 and are endowed with distinct sets of neurotransmitter receptors linked to Ca2+ signalling.5,6 Furthermore DRG neurones of different size show different kinetics of depolarization-induced [Ca2+]i transients:7 these are at least partially determined by the fact that large DRG neurones are endowed with Ca2+/caffeine-sensitive intracellular Ca2+ stores, whereas small DRG neurones lack them completely. In the present paper we further substantiate this difference by presenting data that ATP mobilizes Ca2+ from the intracellular stores in large DRG neurones, whereas small neurones do not express such a mechanism.

Experiments were performed on neurones acutely isolated from dorsal root ganglia (DRG) of adult (2- to 3-month-old) mice. Ganglia were enzymatically treated in a Tyrode based isolation medium supplemented by 1 mg/ml protease (Sigma, Type XIV). Tissue was incubated in isolation medium at 35°C for 18–20 min (see Refs. 4,8). After the enzymatic treatment cells were isolated by gentle pipetting in Tyrode solution, and cell suspensions were plated on sterile glass coverslips. Cytosolic free calcium was measured with the Ca2+sensitive fluorescent dye indo-1 as described previously.4,8 Neurones were loaded with indo-1 acetoxymethyl ester (indo-1/AM) by cell incubation with normal physiological Tyrode solution supplemented with 5 mM indo-1/AM (diluted in DMSO) and 0.02% pluronic F-127 detergent for 30 min at 34°C. Indo-1 fluorescence was induced with a single excitation wavelength (360 ± 5 nm), and dual emission (at 400 ± 10 nm and 500 ± 10 nm) was measured for determination of [Ca2+]i. Analogue signals were fed into an IBM compatible PC via TIDA interface (Batelle, Germany). Data acquisition and analysis were controlled by a WinTida software, version 3.0 (HEKA, Germany). The experimental chamber was continuously superfused with Tyrode solution at a rate of 5 ml/min; the

© Rapid Science Publishers

Vol 8 No 7 6 May 1997

1555

N. Svichar et al. 1

11

11

11

fast application system allowed a complete change of the external solution in < 100 ms. All experiments were performed at 33–35°C. All data are given as mean ± s.d. The [Ca2+]i values were calculated using the equation from Ref. 9. Values for KdB, Rmin and Rmax obtained utilizing the ionomycin-based intracellular calibration procedure8 were 1092 nM, 0.05 and 2.1, respectively. The Tyrode solution was composed of (in mM): NaCl 140; KCl 4; CaCl2 2; MgCl2 2; HEPES/NaOH 5; glucose 10; pH 7.4. To produce high-potassium extracellular solutions the [K+] was adjusted by iso-osmotic replacement of Na+. For intracellular dialysis the whole-cell patch clamp configuration was employed. Recording pipettes were fabricated from borosilicate capillaries (Hilgenberg, Malsfeld, Germany) and had resistances of 3–5 MV when filled with intrapipette solution. Current signals were amplified with conventional electronics (D3900A patch clamp amplifier, Dagan Corp., USA) filtered at 2 kHz and sampled at 5–10 kHz by TIDA interface connected to a PC system, which also served as a stimulus generator. The intrapipette solution contained (in mM): KCl 130, MgCl2 2, HEPES/KOH 10, ATPNa2 2.5, Indo-1 K5 0.2, pH 7.2. Indo-1/AM, Indo-1 K5 and pluronic F-127 were obtained from Molecular Probes, Eugene, OR, USA, and all other chemicals were from Sigma Chemical Co. (USA).

Results ATP-triggers Ca2+ release from internal stores only in large DRG neurones: We performed our experiments on two groups of DRG neurones with soma diameters 20–25 mm (small neurones) and 40–45 mm (large neurones). To discriminate the ATP-triggered Ca2+ release from the internal stores over trans-

membrane Ca2+ influx, which may accompany the stimulation of ionotropic purinoreceprtors, we applied ATP in Ca2+-free extracellular solution. External application of Ca2+ free solution supplemented with 100 mM ATP induced prompt elevation in [Ca2+]i in majority of large DRG neurones (67 out of 72 cells, Fig. 1A). Depolarization-induced [Ca2+]i elevation (produced by 5 s application of solution containing 50 mM KCl and 2 mM Ca2+) significantly increased the amplitude of the subsequent ATPinduced [Ca2+]i transient: the initial amplitude of ATP-triggered [Ca2+]i transient was 162 ± 47 nM whereas ATP applied 60 s after the end of KCl depolarization elevated [Ca2+]i to 257 ± 32 nM (n = 44). That increase in the amplitude of ATP-triggered [Ca2+]i transients after KCl challenge presumably indicates that a proportion of Ca2+ ions entering the cell upon the depolarization was trapped by the endoplasmic reticulum stores, increasing their releasable Ca2+ content. In contrast to large cells, none of the small DRG neurones (n = 29) responded with an increase in [Ca2+]i to ATP applied in the Ca2+-free solution either to the resting cell or 60 s after cell depolarization (Fig. 1B). ATP-induced [Ca2+]i signalling is due to Ca2+ release from inositol 1,4,5-triphosphate-sensitive intracellular stores: In order to clarify the mechanisms of ATPinduced [Ca2+]i signalling in large DRG neurones we used several experimental protocols aimed to determine the intracellular signalling pathway activated by ATP and to determine the receptor type mediating the ATP effects. As mentioned above, ATP triggered [Ca2+]i transients in large DRG neurones bathed in Ca2+-free

11

11

1p

FIG. 1. Effect of ATP on [Ca2+]i in large (A) and small (B) dorsal root ganglion neurones. ATP (100 mM) was applied in Ca2+-free solution to ensure the intracellular origin of the [Ca2+]i transients. Between ATP applications cells were bathed in normal Tyrode (2 mM Ca2+) solution. The 50 mM KCl applications (marked by an arrow) lasted for 4 s.

1556 Vol 8 No 7 6 May 1997

ATP-induced Ca2+ release in sensory neurones

1

1

1

1

1

p

FIG. 2. Inhibition of endoplasmic reticulum Ca2+ pumps or IP3-gated Ca2+ release channels suppressed the ATP-driven [Ca2+]i transients in large DRG neurones. (A) Incubation of DRG neurone with 50 nM thapsigargin for 10 min inhibited the ATP-induced [Ca2+]i elevation. (B) Intracellular perfusion of the DRG neurone with intrapipette solution enriched with 20 mg/ml heparin markedly decreased the amplitude of ATP-triggered [Ca2+]i transient. The control [Ca2+]i transient was taken from the indo-1/AM loaded cell; the second [Ca2+]i transient was recorded 10 min after the beginning of intracellular dialysis with heparin-containing solution. ATP was applied in Ca2+-free solution throughout.

solution, suggesting that Ca2+ came from the intracellular sources. To clarify the nature of intracellular Ca2+ storage compartment involved in the generation of the ATP-triggered [Ca2+]i increase we used thapsigargin, a selective inhibitor of endoplasmic reticulum Ca2+ pumps.4 Incubation of DRG neurones for 10 min with 50 nM thapsigargin led to an almost complete (by 92 ± 4% compared with control) and irreversible inhibition of ATP-induced [Ca2+]i elevation (n = 9; Fig. 2A). As the second tool, aimed to characterize the intracellular mechanism of ATPdriven Ca2+ mobilization, we used heparin, believed to be a specific blocker of inositol 1,4,5-triphosphategated Ca2+ release channels.10 In these experiments we initially recorded the control [Ca2+]i transient from indo-1/AM loaded cells. Afterwards, cells were approached by a patch pipette and the whole-cell mode was established. As demonstrated in Fig. 2B, after 10 min of cell dialysis with the intracellular solution supplemented with 20 mg/ml heparin the ATP-triggered [Ca2+]i elevation was significantly suppressed (137 ± 31 nM in the control vs 34 ± 9 nM after 10 min of cell dialysis with heparin-containing solution; n = 6). In contrast, 10 min of dialysis with the normal intrapipette solution did not affect the ATP-triggered [Ca2+]i increase considerably ([Ca2+]i transient amplitudes were 128 ± 37 nM in the control and 131 ± 24 nM after 10 min of intracellular dialysis, n = 5, data not shown). As ATP effects may be conveyed through the activation of either P1 (adenosine) or P2 purino-

receptors11 we tested the ability of adenosine to alter [Ca2+]i in large DRG neurones. The external application of adenosine (100 mM) did not change the [Ca2+]i significantly (n = 12, Fig. 3A). In contrast, the antagonist of P2 purinoreceptors suramin12 (200 mM, 5 min incubation) completely inhibited ATP-induced [Ca2+]i transients (n = 5, Fig. 3B).

Discussion During recent years ATP has been recognized as a neurotransmitter,13,14 and evidence has mounted that its action is mediated by a broad family of purinoreceptors which are abundantly expressed in the neural cells.11,15 The purinoreceptor family is represented by two distinct subclasses, P1 (adenosine receptors) and P2 (nucleotide receptors). The latter are subclassified on ionotropic (P2X and P2Z) and metabotropic (P2Y, P2U, P2T and P2D) receptors. Molecular cloning of the P2 receptors extended this classification, demonstrating that P2 receptors are represented by at least two distinct gene families. The P2X receptors comprise several subtypes (P2X1–6) of ligand-gated ionic channels with a unique two transmembrane domain topology; members of the P2X family differ in their ion selectivity and gating properties.16 Cloned P2Z receptors were identified as the P2X7 subtype; they are represented by a large transmembrane pore which is activated by a tetra-anionic form of ATP (ATP4–) and may pass molecules with Vol 8 No 7 6 May 1997

1557

N. Svichar et al. 1

11

11

11

11

11

1p

FIG. 3. ATP-induced [Ca2+]i transients in large DRG neurones are mediated through P2 metabotropic purinoreceptors. (A) ATP-induced [Ca2+]i transients recorded from the large DRG neurone. (B) From the same neurone [Ca2+]i was monitored upon application of 100 mM adenosine (ADO). (C) Inhibition of the ATP-mediated [Ca2+]i transients by suramin. The cell was incubated with 200 mM suramin for 5 min prior to ATP administration. ATP and adenosine were applied in Ca2+-free solution.

mol. wt up to 1 kDa.17 Cloned metabotropic receptors have been classified as P2Y family, being represented by seven members (P2Y1–P2Y7). The P2Y1 receptor is pharmacologically identical to the P2Y subtype, P2Y2–to the P2U subtype, and P2Y3 probably corresponds to the P2T receptor; P2Y4–7 are not yet assigned to known subtypes.18 Members of the P2X and P2Y families are expressed throughout the brain19,20 including sensory neurones. DRG neurones express mRNAs for P2X1–6 receptors and the administration of ATP was reported to initiate an increase in [Ca2+]i in both small and large neurones.5 In the present study, however, we demonstrate that the mechanisms of ATP-induced 1558 Vol 8 No 7 6 May 1997

Ca2+ mobilization differ between small and large neurones. The application of ATP in the absence of extracellular Ca2+ initiated an increase in [Ca2+]i only in large DRG neurones, but had no effect on [Ca2+]i in small neurones. In large neurones ATP obviously activates Ca2+ release from the endoplasmic reticulum storage compartment, as indicated by its sensitivity to thapsigargin. This Ca2+ release involves the activation of IP3-gated Ca2+ release channels, since intracellular perfusion with heparin suppressed the ATP-driven [Ca2+]i increase. The effects of ATP on [Ca2+]i in large DRG neurones were not mimicked by adenosine but were inhibited by suramin, suggesting that ATP-triggered [Ca2+]i transients are

ATP-induced Ca2+ release in sensory neurones

1

1

mediated through P2 metabotropic purinoreceptors. We conclude, therefore, that large DRG neurones are endowed with metabotropic purinoreceptors of the P2Y family which act through the IP3–Ca2+ intracellular signal transduction chain. Conversely, small nociceptive DRG neurones lack the P2Y/IP3–Ca2+ signalling mechanism. Our data are in good agreement with results of molecular cloning experiments, which demonstrated that P2Y1 mRNA was exclusively concentrated in sensory neurones bearing axons of large diameter.21 In contrast, small nociceptive DRG neurones are distinguished by their specific expression of P2X3 receptor transcripts21,22 as well as P2X1,2,4,5, and P2X6 mRNAs. All these receptors might participate in generation of cytoplasmic Ca2+ signals by either allowing Ca2+ fluxes through the receptor pore or depolarizing membrane over the voltage-gated Ca2+ channel threshold. Thus we may assume that the Ca2+ signalling in nociceptive and proprioceptive sensory neurones relies on different pathways: in small nociceptive neurones it mainly depends on plasmalemmal Ca2+ influx, whereas in large proprioceptive neurones Ca2+ release via both Ca2+/ caffeine-sensitive and IP3-sensitive routes participate in shaping the cytoplasmic Ca2+ signal.

Conclusion 1

1

Sensory neurones of different modalities are endowed with different sets of purinoreceptors. The neurones with large cell bodies involved in transmission of proprioceptive information express P2Y metabotropic

purinoreceptors linked to intracellular Ca2+ release via the IP3 signalling pathway, whereas small nociceptive neurones lack this mechanism. The possible role of P2Y purinoreceptors in proprioceptive signal transmission remains to be elucidated. References 1. Scott SA. Sensory neurones: Diversity, development, and plasticity. New York, Oxford: Oxford University Press, 1992: 267. 2. Harper AA and Lawson SN. J Physiol Lond 359, 47–63 (1985). 3. Scroggs RS and Fox AP. J Physiol Lond 445, 639–658 (1992). 4. Shmigol A, Kostyuk P and Verkhratsky A. Neuroscience 65, 1109–1118 (1995). 5. Bowie D, Feltz P and Schlichter R. Neuroscience 58, 141–149 (1994). 6. Tang T, Stevens BA and Cox BM. J Neurosci Res 44, 338–343 (1996). 7. Shmigol A, Kostyuk P and Verkhratsky A. NeuroReport 5, 2073–2076 (1994). 8. Usachev Y, Shmigol A, Pronchuk N, et al. Neuroscience 57, 845–859 (1993). 9. Grynkiewicz G, Poenie M and Tsien RY. J Biol Chem 260, 3440–3450 (1985). 10. Michelangeli F, Mezna M, Tovey S and Sayers LG. Neuropharmacology 34, 1111–1122 (1995). 11. Fredholm BB, Abbracchio MP, Burnstock G, et al. Pharmacol Rev 46, 143–156 (1994). 12. Chen ZP, Levy A and Lightman SL. Brain Res 641, 249–56 (1994). 13. Chen ZP, Levy A and Lightman SL. J Neuroendocrinol 7, 83–96 (1995). 14. Edwards FA. Curr Opin Neurobiol 4, 347–352 (1994). 15. Conigrave AD and Jiang L. Cell Calcium 17, 111–119 (1995). 16. Buell G, Collo G and Rassendren F. Eur J Neurosci 8, 2221–2228 (1996). 17. Surprenant A, Rassendren F, Kawashima E, et al. Science 272, 735–738 (1996). 18. Burnstock G and King BF. Drag Develop Res 38, 67–71 (1996). 19. Collo G, North RA, Kawashima E, et al. J Neurosci 16, 2495–2507 (1996). 20. Seguela P, Haghighi A, Soghomonian JJ and Cooper E. J Neurosci 16, 448–455 (1996). 21. Nakamura F and Strittmatter SM. Proc Natl Acad Sci USA 93, 10465–10470 (1996). 22. Chen CC, Akopian AN, Sivilotti L, et al. Nature 377, 428–431 (1995). ACKNOWLEDGEMENTS: This research was supported in part by INTAS (International association for the Promotion of the Cooperation with Scientists from the New Independent States of the former Soviet Union) grant to P.K. and by Wellcome Trust Collaborative Research Grant to A.V. The authors thank Ludmila Grigorovitch for excellent technical assistance.

Received 28 January 1997; accepted 6 March 1997

General Summary The axons of sensory neurones of the dorsal root ganglia of different size carry different sensory information: neurones with large somata are coupled with thick axons conducting mainly proprioceptive information whereas neurones with small somata are responsible for conveying the sense of pain (nociceptive information). As was shown previously by us and other groups the large and small dorsal root ganglion (DRG) neurones differ with respect of mechanisms of [Ca2+]i homeostasis and Ca2+ signalling. In particular Ca2+-induced Ca2+ release is absent in nociceptive, but present in proprioceptive neurones. In the present study we further substantiated this difference by showing that whereas large (proprioceptive) DRG neurones are endowed with metabotropic P2Y purinoreceptors which triggers the inositol 1,4,5-triphosphate-induced Ca2+ release from the internal stores, the small (nociceptive) neurones are completely devoid of such a mechanism.

1

p Vol 8 No 7 6 May 1997

1559