P2X7 expression in human fibroblasts - Journal of Cell Science - The ...

297

Journal of Cell Science 112, 297-305 (1999) Printed in Great Britain © The Company of Biologists Limited 1999 JCS9898

Human primary fibroblasts in vitro express a purinergic P2X7 receptor coupled to ion fluxes, microvesicle formation and IL-6 release Anna Solini1, Paola Chiozzi2, Anna Morelli2, Renato Fellin1 and Francesco Di Virgilio2,3,* 1Department

of Clinical and Experimental Medicine, Section of Internal Medicine and 2Department of Experimental and Diagnostic Medicine, Section of General Pathology, University of Ferrara and 3Biotechnology Center, Ferrara, Italy

*Author for correspondence at Department of Experimental and Diagnostic Medicine, Section of General Pathology, Via Borsari, 46, I-44100 Ferrara, Italy (e-mail: [email protected])

Accepted 20 November 1998; published on WWW 13 January 1999

SUMMARY We have investigated reponses to extracellular ATP in human fibroblasts obtained by skin biopsies. Our data show that these cells express a P2X7 purinergic receptor, as judged by (1) RT-PCR with specific primers, (2) reactivity with a specific anti-P2X7 antiserum, (3) activation by the selective P2X agonist benzoylbenzoylATP and (4) stimulation of transmembrane ion fluxes. Stimulation with benzoylbenzoylATP, and to a lesser extent with ATP, also caused striking morphological changes and increased formation of cytoplasmic microvesicles. These changes were fully reversible upon nucleotide removal. Two known blockers of P2X receptors, oxidised ATP and pyridoxalphosphate-6-azophenyl-2′,4′disulfonic acid, inhibited the morphological changes fully

and the ion fluxes partially. The residual rise in intracellular Ca2+ levels and membrane depolarization observed in the presence of the inhibitors were dependent upon activation of a P2Y-type receptor exhibiting a peculiar pharmacological profile, in that CTP was the preferred agonist. ATP stimulation triggered release of the pro-inflammatory cytokine IL-6 in fibroblasts pre-treated with PMA and bacterial endotoxin. These observations reveal a novel pathway for fibroblast activation and for their recruitment in the inflammatory response.

INTRODUCTION

responses, which are mediated by P2Y receptors, some fibroblast lines also undergo changes that are suggestive of expression of P2X receptors, as they are susceptible to the typical ATP-dependent plasma membrane permeabilization diagnostic of P2X7 receptors, while others exhibit transmembrane ion fluxes that are suggestive of the presence of smaller members of the P2X family (Pizzo et al., 1992). However, a molecular identification of the P2X receptor responsible for these responses has not yet been attempted in human fibroblasts. Fibroblasts are non-excitable cells, thus it is unlikely that external nucleotides serve the function of fast mediators of intercellular communication as in the nervous system. Rather they might be involved in the modulation of fibroblast responses to the variety of different microenvironmental situations to which these cells are exposed in the vessel wall or in the interstitial tissue. There is increasing evidence that ATP can accumulate in the external milieu as a consequence of different events: release by secretory exocytosis (e.g. platelets or nerve cells), transport across the plasma membrane via as yet unknown carriers (e.g. endothelial cells), passive release from damaged cells (see Hoyle and Burnstock, 1996). It is well known that cells in the vessel wall are exposed to intense shrear stress forces or actual damage that may release cellular ATP;

Many different cell types express purinergic receptors of the P2X and P2Y subtypes (Harden et al., 1995; Brake and Julius, 1996; North and Barnard, 1997). While a wealth of data has been collected from neuronal cells (Buell et al., 1996; Chen et al., 1995; Lewis et al., 1995; Soto et al., 1996) or leukocytes (Di Virgilio et al., 1989, 1996; Filippini et al., 1990; Humphreys and Dubyak, 1996), comparatively little information is available about human primary fibroblasts, apart from a few anedoctal reports (Fine et al., 1989; Okada et al., 1984). Human skin fibroblasts are an interesting model as they share several features with smooth and striated muscle cells and are directly involved in chronic degenerative diseases such as atherosclerosis and diabetic angiopathy (Shats et al., 1997; Nasrin et al., 1997; Sigari et al., 1997). Furthermore, they can be obtained by a very easy isolation procedure (skin biopsies) from well controlled healthy or diseased subjects. Most data on the effect of extracellular nucleotides on fibroblasts have been obtained in mouse cells, where ATP has a mitogenic effect, either alone or in combination with more classical growth factors, possibly due to stimulation of Ca2+dependent pathways or inhibition of arachidonate metabolism (Gonzales et al., 1989; Huang et al., 1993). Besides these

Key words: Cytoplasmic calcium, Cytokine, Extracellular nucleotide, Inflammation, Golgi apparatus, Fibroblast

298

A. Solini and others

furthermore, during thrombus formation fibroblasts in the vessel wall are also exposed to ATP released from platelets. In the present work we show that human skin fibroblasts express purinergic receptors of the P2X7 subtype and that activation of these receptors is coupled to induction of ion fluxes and extensive cytoplasmic microvesiculation. Furthermore, in fibroblasts pre-treated with PMA and LPS, ATP causes release of the pro-inflammatory cytokine IL-6. MATERIALS AND METHODS Cells and solutions Human skin fibroblasts were grown in DMEM medium supplemented with 10% heat-inactivated fetal calf serum. Mouse microglial cells were grown in RPMI 1640 medium containing 10% heat-inactivated fetal calf serum. Both culture media contained penicillin (100 units/ml) and streptomycin (100 µg/ml). Short-term experiments were performed in saline solution containing 125 mM NaCl, 5 mM KCl, 1 mM MgSO4, 1 mM Na2HPO4, 5.5 mM glucose, 5 mM NaHCO3 and 20 mM Hepes (pH 7.4). We will refer to this saline solution as standard saline. In some experiments, 1 mM CaCl2 was also added (Ca2+-containing standard saline). Cytoplasmic free Ca2+ concentration measurements Changes in cytoplasmic free Ca2+ concentration ([Ca2+]i) were measured with the fluorescent indicator fura-2/AM, as described previously (Falzoni et al., 1995). Briefly, cells were loaded for 15 minutes with 2 µM fura-2/AM and incubated at 37°C in a thermostat-controlled and magnetically stirred fluorometer cuvette (LS50; Perkin Elmer Ltd, Beaconsfield, UK) at a concentration of 106/ml in the presence of 250 µM sulfinpyrazone. Intracellular Ca2+ concentration was determined with the 340/380 excitation ratio at an emission wavelength of 505 nm. Semiquantitative measurement of plasma membrane potential Changes in plasma membrane potential were measured with the fluorescent dye bis[1,3-diethylthiobarbiturate] trimethineoxonal (bisoxonol; Molecular Probes, Inc., Eugene, OR, USA) at the wavelength pair 540-580 nm (Falzoni et al., 1995). Experiments were performed in a spectrofluorometer (LS50, Perkin Elmer) at 37°C at a concentration of 2.5×105 cells/ml. Changes in plasma membrane permeability ATP-dependent increases in plasma membrane permeability were measured with the extracellular fluorescent tracers YO-PRO and Lucifer Yellow (Molecular Probes, Inc.). Cell monolayers or suspensions were incubated for 15 minutes at 37°C in standard saline containing 250 µM sulfinpyrazone and 10 µM YO-PRO or 1 mg/ml Lucifer Yellow, and stimulated with 3 mM ATP. After several washings to remove the extracellular dye, cells were analysed with an inverted fluorescence microscope (Olympus IMT-2; Olympus Optical Co, Ltd., Tokyo, Japan) equipped with a 40× objective and a fluorescein filter.

against the synthetic peptide corresponding to the last 20 amino acids of the P2X7 protein (KIRKEFPKTQGQYSGFKYPY), and was kindly provided by Dr Gary Buell (Glaxo-Wellcome, Geneva) (Surprenant et al., 1996). The primary antibody was used at a dilution of 1:100 in TBS buffer (10 mM Tris-Cl, 150 mM NaCl, pH 8.0). The secondary antibody was a goat anti-rabbit antibody conjugated to alkaline phosphatase. Immunofluorescence Fibroblasts were seeded on glass slides, washed with PBS, and fixed with paraformaldehyde (2% in PBS). After 2 hours, fixed cells were permeabilized with Triton X-100 (1% in PBS) and incubated for 1 hour with the anti-AP-1 monoclonal antibody (1:40 dilution in PBS). Cells were then rinsed three times with PBS and incubated for 30 minutes with an anti-mouse Ig FITC-labelled antibody (1:50 dilution in PBS). At the end of this incubation, slides were rinsed 3 times and analyzed with the fluorescence miscroscope. RT-PCR Total cytoplasmic RNA was extracted from fibroblast cells by the acid guanidinium thiocianate-phenol-chloroform method, as described by Chomczynski and Sacchi (1987). RT-PCR was performed as described by Rappolee et al. (1989). Amplification primers and probe were chosen on the basis of the published sequence of human P2X7 receptor (Rassendren et al., 1997). The 5′ primer was 5′-AGATCGTGGAGAATGGAGTG-3′ and the 3′ primer, 5′-TTCTCGTGGTGTAGTTGTGG-3′; the probe was 5′TCATGCACTACACACCTTCC-3′. Amplification primers for β-actin were: 5′ primer, 5′-TGACGGGGTCACCCACACTGTG-CCCATCTA3′; 3′ primer, 5′-AGTCATAGTCCGCCTAGAAGC-ATTTGCGGT-3′. Oligonucleotides were synthesized by M-Medical Genenco-Life Science (Firenze, Italy). Labelling of the probe and blotting were carried out as described in Dig-labelling and detection protocols (Boehringer Mannheim, Germany). Briefly, RT-PCR products were separated in 1.2% agarose gels and transferred to a positively charged nylon membrane (ICN Biomedicals Inc. Aurora, Ohio, USA) by a vacuum blotter system (BioRad Laboratories, Hercules, CA, USA) for 2 hours. After hybridization, the digoxigenin-labelled P2X7-specific internal oligoprobe was visualized by chemiluminescence detection after incubation with a dilution of anti-digoxigenin Fab conjugated to alkaline phosphatase. IL-6 determination IL-6 was measured with the BioSource Cytoscreen Immunoassay kit (BioSource International Inc., Camarillo, CA, USA). Reagents Nucleotides were HPLC-purified. All other chemicals were reagent grade.

P2X7

Measurement of enzymatic activity Lactate dehydrogenase activity was measured according to standard methods (Bergmeyer, 1983). Western blotting Cells were lysed in lysis buffer containing 300 mM sucrose, 1 mM K2HPO4, 1 mM MgSO4, 5.5 mM glucose, 20 mM Hepes, pH 7.4, 1 mM benzamidine, 1 mM PMSF, 0.2 µg of DNase and 0.2 µg of RNase by repeated freeze/thawing (three cycles). Proteins were separated on 7.5% SDS-polyacrylamide gels according to Laemmli (1970) and blotted overnight on nitrocellulose paper (Schleicher and Schull Italia Srl, Legnano, Italy). The rabbit polyclonal anti-P2Z/P2X7 serum was raised

Fig. 1. Human fibroblasts express the P2X7 receptor mRNA. Total RNA was extracted as described in Materials and methods and used for RT-PCR. 10 µl of PCR product were loaded in each lane. Detection of the P2X7 fragment was performed by transfer to nylon membrane and hybridization with a specific internal probe. Lane 1, human macrophages; lane 2, human fibroblasts.

P2X7 expression in human fibroblasts RESULTS

kDa

Expression of P2X7 receptor has so far only been reported in immune cells, but very recent evidence suggests that it may also be present in other cell types (for example, see CarioToumaniantz et al., 1998). Fig. 1 shows that P2X7 mRNA is present in human fibroblasts, although to a much lower level than in macrophages, the prototypical cell in which this receptor was originally characterized. P2X7 expression by fibroblasts is also confirmed by using an anti-P2X7 polyclonal antibody (Collo et al., 1997). Fig. 2, lane 1, shows a western blot of human fibroblasts stained with a serum raised against the COOH tail (residues 576-595) of the rat P2X7 receptor. Immunoreactivity of mouse microglial cells, a cell type well known for expressing the P2X7 receptor, is shown in lane 2. In lane 3 we also show lower reactivity to the anti-P2X7 antiserum of a microglial cell clone selected for hypo-expression of P2X7. One of the earliest changes associated to P2X7 receptor activation is plasma membrane depolarization, due to a fast transmembrane cation flux. Fig. 3A, left panel, shows that this response is also present in ATP-stimulated fibroblasts. Surprisingly, while the larger depolarization in response to BzATP was anticipated, given the higher affinity of this ATP analogue for the P2X7 receptor, the fact that CTP was also active was unexpected, as no other nucleotides or nucleosides besides ATP, some ATP analogues and, to a lesser extent, ADP, are active at P2X7. Interestingly, depolarization

A

Fig. 3. Effect of extracellular nucleotides on fibroblast plasma membrane potential. (A) Fibroblasts were incubated at 37°C in Ca2+-containing (left) or Ca2+-free (right) standard saline solution at a concentration of 2.5×105/ml in the presence of 100 nM bisoxonol. ATP and CTP concentrations were 1 mM, BzATP (Bz) was 0.5 mM. Each KCl addition (K+) was 15 mM. (B) Nucleotide dose dependency of plasma membrane depolarisation in Ca2+containing medium. 䊉, ATP; 䊏, BzATP; 䉱, CTP.

299

B

Fig. 2. Human fibroblasts express the P2X7 receptor protein. Cells were lysed in lysis buffer (see Materials and methods) and resuspended at a concentration of 1 mg protein/ml. 30 µg protein were loaded in each lane. Lane 1, fibroblasts; lane 2, microglial cells; lane 3, microglial cells selected for low expression of P2X7. The positions of molecular mass markers are shown.

triggered by CTP was preceeded by a small transient hyperpolarization. Experiments performed in the virtual absence of extracellular Ca2+ helped to solve this paradox, as they showed (Fig. 3A, right panel) that while responses to ATP


[Ca2+]i (nM)

300

Fig. 4. Changes in [Ca2+]i triggered by CTP (a), ATP (b) and BzATP (c). Fibroblasts were incubated in Ca2+-containing standard saline and challenged with increasing concentrations (a1,b1,c1 = 0.1 mM; a2,b2,c2 = 0.5 mM; a3,b3,c3 = 1 mM) of the three nucleotides.

and benzoylbenzoylATP (BzATP) were potentiated as expected, CTP-dependent depolarization was totally abrogated, thus indicating that CTP caused depolarization not by directly activating a depolarizing flux, but rather by causing an increase in the intracellular Ca2+ concentration. The pharmacological sequence for plasma membrane depolarization in Ca2+-containing standard saline was: BzATP>CTP>>ATP=ATPγS=2 methylthio ATP. Fig. 3B shows ATP, BzATP and CTP dose-dependency for plasma membrane depolarization in Ca2+-containing medium. BzATP and CTP showed an EC50 of 120 and 100 µM, respectively. ATP EC50 was difficult to determine under these conditions as the response to this nucleotide was very small in the presence of Ca2+. Fig. 4 shows Ca2+ changes triggered by increasing concentrations of CTP (a), ATP (b) and BzATP (c) in the presence of extracellular Ca2+. Among the three nucleotides, the most powerful agonist was CTP, followed by ATP, while BzATP was a very weak stimulus at all concentrations tested. In the absence of any extracellular Ca2+, the response to CTP and ATP was decreased by about 30% while that to BzATP was completely abolished (not shown). One of the main drawbacks in the study of P2 receptors is the lack of selective antagonists. However, two compounds that over the last few years have turned out to be useful are the dialdehyde ATP derivative, oxidized ATP (oATP) (Murgia et al., 1993), and the aldehyde compound pyridoxalphosphate-6azophenyl-2′,4′-disulfonic acid (PPADS) (Lambrecht et al., 1992). These reagents react with lysine residues in the vicinity of plasma membrane ATP-binding sites and have been shown to inhibit P2X7-dependent responses. Fig. 5 shows that these two compounds also inhibit depolarization and [Ca2+]i increases triggered in fibroblasts by BzATP and ATP,

respectively. Inhibition of plasma membrane depolarization by oATP or PPADS was evaluated using BzATP as a stimulus, as this ATP analogue is a better depolarizing agent than ATP. On the contrary, inhibition of [Ca2+]i changes was evaluated with reference to the maximal increase caused by ATP, as BzATP caused a very small increase in [Ca2+]i. Confirming previous observations performed in our laboratory (O. R. Baricordi and F. Di Virgilio, unpublished), oATP was a much better inhibitor than PPADS on a purely P2X7-mediated response, such as depolarization, while both oATP and PPADS partially blocked (about 50%) the [Ca2+]i increase, a response that was dependent on activation of both P2Y and P2X7 receptors. The hallmark of P2X7 receptor activation is an ATPdependent reversible permeabilization of the plasma membrane to solutes of molecular mass 2 mM) was there a significant level of cell death (not shown). ATP and CTP had no cytotoxic activity at any concentrations tested. However, although overt signs of ATP-dependent cytotoxicity were lacking, it was difficult to rule out the possibility that YO-PRO uptake was the consequence of a nonspecific increase in plasma membrane permeability as a result of the prolonged alteration of intracellular ion homeostasis.

Fig. 5. Inhibition of depolarization and [Ca2+]i increase by oATP and PPADS. Fibroblasts were incubated in Ca2+-containing standard saline solution for 2 hours at 37°C in the absence (controls) or presence of 300 µM PPADS or oATP, rinsed, and challenged with 1 mM BzATP (depolarization; left) or ATP ([Ca2+]i increases; right). Data are means ± s.d. of triplicate determinations from one representative experiment repeated three times with similar results.

P2X7 expression in human fibroblasts

301

Fig. 6. P2X7 activation causes delayed YO-PRO uptake. Fibroblasts were incubated for 2 hours at 37°C in Ca2+-containing standard saline supplemented with 10 µM YO-PRO, in the absence of added nucleotides (A and D, controls), or in the presence of 1 mM ATP (B,E) or 1 mM BzATP (C,F). A-C, fluorescence microscopy; D-E, phase-contrast microscopy. Bars, 25 µm.

While on the one hand P2X7 receptor stimulation did not induce overt signs of cell injury, on the other it caused a large increase in cytoplasmic microvesicle formation (Fig. 7D). This effect was clearly visible with BzATP, even after relatively short incubations, while was only occasionally observed with ATP (Fig. 7B), and never with CTP. CTP, on the contrary, stimulated an intriguing rearrangement of the monolayer, where the fibroblasts assumed a spindle-like morphology and generated intersecting cellular cords (Fig. 7C). Within 15-30 minutes of BzATP removal, microvesicles disappeared, and fibroblasts resumed a normal morphology (not shown). The ability of extracellular nucleotides to trigger cytoplasmic vesicle formation is not a new phenomenon, as this is one of the most extensively documented early changes observed upon

stimulation of macrophages with extracellular nucleotides (Cohn and Parks, 1967; Di Virgilio and Steinberg, 1993). However, in fibroblasts this phenomenon was rather peculiar as vesicles were of small size and had a perinuclear location. Thus we further investigated the nature of these vesicles. Extracellular nucleotides are known to increase the rate of pinocytosis; however, we found that upon BzATP stimulation uptake of Texas Red-labelled ovalbumin was only marginally increased (about 10% with respect to unstimulated controls). Since the perinuclear location of the vesicles suggests an expansion of the Golgi apparatus, we labelled BzATPstimulated fibroblasts with a monoclonal antibody directed against a specific Golgi marker (γ-adaptin, AP-1) (Ahle et al., 1988). As shown in Fig. 8, fibroblasts incubated in the presence

302


Fig. 7. P2X7 receptor activation causes cytoplasmic microvesiculation. Fibroblasts were plated as described in Fig. 6 and stimulated with 1 mM ATP (B) or 1 mM CTP (C) for 6 hours, or with 0.5 mM BzATP for 2 hours (D). (A) Controls. Phase-contrast microscopy. Bars, 25 µm.

of BzATP show an increased staining of the Golgi apparatus (cf. Fig. 8A,B), suggesting that microvesicles originate, at least in part, from the Golgi. Although the physiological function of the P2X7 is as yet unknown, in mononuclear phagocytes this receptor has been associated with cytokine release (Perregaux and Gabel, 1994; Ferrari et al., 1996). Fibroblasts release several cytokines and pro-inflammatory factors, including IL-6 (Baumann and Kushner, 1998), thus we investigated whether extracellular ATP could promote secretion of this cytokine. Extracellular nucleotides by themselves were unable to stimulate IL-6 secretion (not shown); however, addition of ATP or BzATP, but not CTP, to fibroblasts pre-treated with bacterial endotoxin and PMA almost doubled IL-6 release (Fig. 9), a behaviour reminiscent of the effect of ATP on IL-1β release from human and mouse macrophages and microglial cells (Perregaux and Gabel, 1994; Ferrari et al., 1996). DISCUSSION Purinergic P2X receptors are becoming one of the most interesting families of ligand-gated ion channels. Seven P2X receptor subunits have been cloned and sequenced so far, and their amino acid sequence suggests a rather unusual membrane topology in that these molecules are understood to span the membrane twice, with both the N and C termini on the cytoplasmic side of the plasma membrane (Harden et al., 1995; Brake and Julius, 1996; North and Barnard, 1997; Buell et al.,

1996). The seven members of the P2X subfamily show a tissueand differentiation-dependent expression, suggesting an involvement in specialized cell functions such as neurotransmission, cell proliferation, cell-to-cell communication, cytokine release and cytotoxic reactions (Buell et al., 1996; Chen et al., 1995; Lewis et al., 1995; Soto et al., 1996; Di Virgilio et al., 1996). In this study we show that human skin-derived fibroblasts express the P2X7 receptor, although its permeability properties appear to be slightly different with respect to the better characterized macrophage P2X7 receptor. Stimulation of fibroblast P2X7 triggered a fast plasma membrane depolarization and a large Ca2+ influx. However, the increase in plasma membrane permeability, a typical response consequent to P2X7 activation, was very small and delayed. We were able to detect some uptake of an extracellular fluorescent dye only 2 hours after addition of BzATP. Although even after such a long incubation fibroblasts did not show any overt sign of cytotoxicity, nor released lactic dehydrogenase, it is difficult to link this delayed dye uptake unambiguously to activation of P2X7. Interpretation of fibroblast responses to nucleotides is further complicated by expression of a previously undescribed receptor responsive to CTP that also causes a fast depolarization of the plasma membrane, albeit with a mechanism different from ATP, as in the case of the CTP receptor depolarization is secondary to the increase in cytoplasmic Ca2+ levels. It is not completely unusual for CTP to be an agonist at P2 receptors, as activatory reponses have been described in mouse macrophages and in oocytes injected

P2X7 expression in human fibroblasts

303

Fig. 8. Staining of BzATP-treated fibroblasts with a monoclonal antibody directed against an antigen selectively expressed in the Golgi. Fibroblasts were plated on glass coverslips (5×104/coverslip) and further incubated at 37°C. (A,D) Controls; no BzATP was present. (B,C,E,F) 0.5 mM BzATP was added for 1 hour. For immunofluorescence, samples were processed as described in Materials and methods. In C and F only the secondary FITC-labelled antibody, and not the primary anti-AP-1, was added. A-C, fluorescence microscopy; D-F, phase-contrast microscopy. Bars, 25 µm.

with a P2Y receptor cloned from Xenopus (Greenberg et al., 1988; Burnstock, 1997), and a member of the P2X subfamily, P2X4, is also weakly responsive to CTP (Garcia-Guzman et al., 1997). Nonetheless it is surprising that CTP appeared to be by far the most potent among the nucleotides tested in our study. While in other cells expressing P2X7 sustained stimulation with ATP causes swelling, blebbing and eventually death, in human fibroblasts the most striking morphological alteration consisted in an expansion of the cytoplasmic vesicular compartment. There is a precedent for such an alteration in mouse macrophages, in which ATP causes a dramatic but reversible increase in the content of large cytoplasmic phase-

lucent vesicles (Cohn and Parks, 1967; Di Virgilio and Steinberg, 1993). In fibroblasts microvesiculation occurs, rather than macrovesiculation. These microvesicles do not arise from stimulation of pinocytosis, but rather from an expansion of the Golgi compartment, although further experiments are clearly needed to fully characterize this process. For fibroblasts, as for other cell types, the question arises as to the physiological significance of ATP-gated ion channels. Until a few years ago it was doubtful that significant amounts of ATP were ever released in the extracellular milieu, but now it is clear that non-lytic ATP release is an event far more common than previously believed (von Kugelgen and Starke, 1991; Pearson and Gordon, 1979; Meyers et al., 1982; Reisin

304


Fig. 9. ATP causes IL-6 release from PMA and LPS-treated fibroblasts. Fibroblasts (3×104/well) were plated in serum-free RPMI medium for 3 hours and challenged with the various stimulants. PMA and bacterial lipopolisaccharide (LPS) were used at a concentration of 100 nM and 1 µg/ml, respectively. In the experiments in which nucleotides were also present, fibroblasts were incubated in the presence of PMA and LPS for 2 hours, then the nucleotide was added and incubation carried out for 1 residual hour. After this time, supernatants were removed and assayed for IL-6 content by ELISA. Data are means of duplicate determinations from one experiment representative of three others.

et al., 1994), not forgetting that every injured cell will also discard its cytoplasmic ATP content, usually in the 5-10 mM range. Thus, it has to be expected that cells are continuously exposed to varying extracellular ATP concentrations that, by acting at membrane receptors with widely different affinities, will elicit complex cellular responses. In this way P2 receptors may endow fibroblasts with an array of very sensitive sensors capable of detecting minute alterations in the metabolic activity or integrity of neighbouring tissues. Fibroblasts are well known not only for being the major producers of extracellular matrix, but also a cell type actively involved in synthesis of inflammatory mediators and tissue repair; thus we consider the observation that ATP triggers IL6 release might be of physiological interest. IL-6 is one of the most important pro-inflammatory cytokines, secreted from mononuclear phagocytes and antigen-presenting cells as well as fibroblasts. Furthermore, IL-6 may be one of the causative factors of the diverse hematologic alterations observed in diabetic patients (e.g. elevation of plasma fibrinogen), as indicated by enhanced secretion of IL-6 and TNF from human monocytes under hyperglycemic conditions (Morohoshi et al., 1996). However, triggering factors for IL-6 release are incompletely understood, although it is clear that in fibroblasts

IL-6 release is the result of the synergistic action of several inflammatory mediators. At sites of inflammation the concentration of extracellular ATP is likely to be significantly increased as a consequence of cell injury. A well-documented example occurs in rheumatoid arthritis, where ATP is a common constituent of the synovial fluid (Ryan et al., 1991). Synovial cells have been shown to express P2Y receptor (Loredo and Benton, 1998), thus ATP might be one component of the ‘inflammatory milieu’ of the arthritic joint. It is likely that this process also occurs in other anatomical sites and during different disease processes. In particular, there are sites at which fibroblasts may be exposed to massive amounts of ATP. For example, at sites of endothelial cell lesions in the vasculature, platelet aggregation and activation will release massive amounts of nucleotides that will act at the fibroblasts of the subendothelial layer. Nucleotides, in synergy with other factors released from platelets and blood leukocytes (growth factors, cytokines, prostaglandins), may enhance stimulation of fibroblasts, thus leading to a second wave of release of pro-inflammatory molecules such as IL-6. Incidentally, among the various NF-κB isoforms involved in IL-6 gene expression, the most active is the unusual p65 homodimer, and it is therefore intriguing that Ferrari et al. (1997) have recently shown that in mouse microglial cells the p65 homodimer is the NF-κB isoform selectively induced by P2X7 receptor activation (Ferrari et al., 1997). Ability of fibroblast P2X7 receptor to trigger IL-6 release may suggest additional targets for pharmacological intervention in a number of diseases in which fibroblasts are implicated, such as atherosclerosis, degenerative joint diseases and diabetic angiopathy. This work was supported by the Italian Ministry for Scientific Research (40% and 60%), the Italian National Research Council (Target Project on Biotechnology), the Italian Association for Cancer Research, the X AIDS Project, the II Tubercolosis Project, Telethon of Italy, and by a special grant from Azienda Ospedaliera Arcispedale S. Anna, Ferrara, Italy. The authors are grateful to Dr Stefania Hanau for providing HPLC-purified nucleotides.

REFERENCES Ahle, S., A. Mann, A., Eichelsbacher, U. and Ungenwickell, E. (1988). Structural relationship between clathrin assembly proteins from the Golgi and the plasma membrane. EMBO J. 7, 919-929. Baumann, H. and Kushner, I. (1998). Production of interleukin-6 by synovial fibroblasts in rheumatoid arthritis. Am. J. Pathol. 152, 641-644. Bergmeyer, H. U. (1983). Methods of Enzymatic Analysis, vol. III, pp. 118133. Academic Press, London. Brake, A. J. and Julius, D. (1996). Signalling by extracellular nucleotides. Annu. Rev. Cell Dev. Biol. 12, 519-541. Buell, G., Collo, G. and Rassendren, F. (1996). P2X receptors: an emerging channel family. Eur. J. Neurosci. 8, 2221-2228. Burnstock, G. (1997). The past,present and future of purine nucleotides as signalling molecules. Neuropharmacology 36, 1127-1139. Cario-Toumaniantz C., Loirand, G., Ladoux, A. and Pacaud, P. (1998). P2X7 receptor activation-induced contraction and lysis in human saphenous vein smooth muscle. Circ. Res. 83, 196-203. Chen, C-C., Akopian, A. N., Sivilotti, L., Colquhoun, D., Burnstock, G. and Wood, J. N. (1995). A P2X purinoceptor expressed by a subset of sensory neurons. Nature 377, 428-431. Chomczynski, P. and Sacchi, N. (1987). Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156-159. Cohn, Z. A. and Parks, E. (1967). The regulation of pinocytosis in mouse

P2X7 expression in human fibroblasts macrophages. III. The induction of vesicle formation by nucleosides and nucleotides. J. Exp. Med. 125, 457-466. Collo, G., Neidhart, S., Kawashima, E., Kosco-Vilbois, M., North, R. A. and Buell, G. (1997). Tissue distribution of the P2X7 receptor. Neuropharmacology 36, 1277-1283. Di Virgilio, F., Bronte, V., Collavo, D. and Zanovello, P. (1989). Responses of mouse lymphocytes to extracellular adenosine 5′-triphosphate (ATP). Lymphocytes with cytotoxic activity are resistant to the permeabilizing effect of ATP. J. Immunol. 143, 1955-1960. Di Virgilio, F. and Steinberg, T. H. (1993). ATP permeabilization. In Cell and Tissue Culture: Laboratory Procedures (ed. A. Doyle, J. B. Griffiths and D. G. Newell), pp. 5B: 4.1-4.8. Wiley, London. Di Virgilio, F. (1995). The P2Z purinoceptor: an intriguing role in immunity, inflammation and cell death. Immunol. Today 16, 524-528. Di Virgilio, F., Ferrari, D., Chiozzi, P., Falzoni, S., Sanz, J. M., Dal Susino, M., Mutini, C., Hanau, S. and Baricordi, O. R. (1996). Purinoceptor function in the immune system. Drug Dev. Res. 39, 319-329. Falzoni, S., Munerati, M., Ferrari, D., Spisani, S., Moretti, S. and Di Virgilio, F. (1995). The purinergic P2Z receptor of human macrophage cells. Characterization and possible physiological role. J. Clin. Invest. 95, 12071216. Ferrari, D., Villalba, M., Chiozzi, P., Falzoni, S., Ricciardi-Castagnoli, P. and Di Virgilio, F. (1996). Mouse microglial cells express a plasma membrane pore gated by extracellular ATP. J. Immunol. 156, 1531-1539. Ferrari, D., Wasselborg, S., Bauer, M. K. A. and Schulze-Osthoff, K. (1997). Extracellular ATP activates transcription factor NF-κB through the P2Z purinoceptor by selectively targeting NF-κB p65 (RelA). J. Cell Biol. 139, 1635-1643. Filippini, A., Taffs, R. E., Agui, T. and Sitkovsky, M. V. (1990). EctoATPase activity in cytolytic T lymphocytes: protection from the cytolytic effects of extracellular ATP. J. Biol. Chem. 265, 334-340. Fine, J., Cole, P. and Davidson, J. S. (1989). Extracellular nucleotides stimulate receptor-mediated calcium mobilization and inositol phosphate production in human fibroblasts. Biochem. J. 263, 371-376. Garcia-Guzman, M., Soto, F., Gomez-Hernandez, J. M., Lund, P. E. and Stuhmer, W. (1997). Characterization of recombinant human P2X4 receptor reveals pharmacological differences to the rat homologue. Mol. Pharmacol. 51, 109-118. Gonzales, F. A., Rozengurt, E. and Heppel, L. A. (1989). Extracellular ATP induces the release of calcium from intracellular stores without the activation of protein kinase C in Swiss 3T6 mouse fibroblasts. Proc. Natl. Acad. Sci. USA 86, 4530-4534. Greenberg, S., Di Virgilio, F., Steinberg, T. H. and Silverstein, S. C. (1988). Extracellular nucleotides mediate Ca2+ fluxes in J774 macrophages by two distinct mechanisms. J. Biol. Chem. 263, 10337-10343. Harden, T. K., Boyer, J. L. and Nicholas, R. A. (1995). P2-purinergic receptors: subtype-associated signalling, responses and structure. Annu. Rev. Pharmacol. Toxicol. 35, 541-579. Hoyle, C. H. V. and Burnstock, G. (1996). Purines. In Principles of Medical Biology (ed. E. E. Bittar and N. Bittar), pp. 49-75. JAI Press, Greenwich, Huang, N.-N., Wang, D.-J. and Heppel, L. A. (1993). Stimulation of aged human lung fibroblasts by extracellular ATP via suppression of arachidonate metabolism. J. Biol. Chem. 268, 10789-10795. Humphreys, B. D. and Dubyak, G. R. (1996). Induction of the P2Z/ P2X7 nucleotide receptor and associated phospholipase D activity by lipopolysaccharide and IFN-γ in the human THP-1 monocytic cell line. J. Immunol. 157, 5627-5637. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685. Lambrecht, G., Friebe, T., Grimm, U., Windscheif, U., Bungardt, E., Hilderbrandt, C., Baumert, H., Spatz-Kumbel, G. and Mutschler, E. (1992). PPADS, a novel functionally selective antagonist of P2 purinoceptor-mediated responses. Eur. J. Pharmacol. 217, 217-219. Lewis, C., Neidhart, S., Holy, C., North, R. A., Buell, G. and Surprenant,

305

A. (1995). Coexpression of P2X2 and P2X3 receptor subunits can account for ATP-gated currents in sensory neurons. Nature 377, 432-435. Loredo, G. A. and Benton, H. P. (1998). ATP and UTP activate calciummobilizing P2U-like receptors and acts synergistically with interleukin-1 to stimulate prostaglandin release from human rheumatoid synovial cells. Arthritis Rheum. 41, 246-255. Meyers, K. M., Holmsen, H. and Seachord, C. L. (1982). Comparative study of platelet dense granules constituents. Am. J. Physiol. 243, R454-R461. Morohoshi, M., Fujisawa, K., Uchimura, I. and Numano, F. (1996). Glucose-dependent interleukin-6 and tumor necrosis factor production by human peripheral blood monocytes in vitro. Diabetes, 45, 954-959. Murgia, M., Hanau, S., Pizzo, P., Rippa, M. and Di Virgilio, F. (1993). Oxidized ATP: an irreversible inhibitor of the macrophage purinergic P2Z receptor. J. Biol. Chem. 268, 8199-8203 Nasrin, N., Mimish, L. A., Manogaran, P. S., Kunhi, M., Sigut, D., AlSedairy, S. and Hannan, M. A. (1997). Cellular radiosensitivity, radioresistant DNA synthesis, and defect in radioinduction of p53 in fibroblasts from atherosclerosis patients. Arterioscler. Thromb. Vasc. Biol. 17, 947-953. North, R. A. and Barnard, E. A. (1997). Nucleotide receptors. Curr. Opin. Neurobiol. 7, 346-357. Okada, Y., Yada, T., Ohno-Shoshaku, T., Oiki, S., Ueda, S. and Machida, K. (1984). Exogenous ATP induces electrical membrane responses in fibroblasts. Exp. Cell Res. 152, 552-557. Pearson, J. D. and Gordon, J. L. (1979). Vascular endothelium and smooth muscle cells in culture selectively release adenine nucleotides. Nature 281, 384-386. Perregaux, D. and Gabel, C. A. (1994). Interleukin-1β maturation and release in response to ATP and nigericin. J. Biol. Chem. 269, 15195-15203. Pizzo, P., Murgia, M., Zambon, A., Zanovello, P., Bronte, V., Pietrobon, D. and Di Virgilio, F. (1992). Role of P2Z purinergic receptors in ATPmediated killing of tumor necrosis factor (TNF)-sensitive and TNF-resistant L929 fibroblasts. J. Immunol. 149, 3372-3378. Rappolee, D. A., Wang, A., Mark, D. and Werb, Z. (1989). Novel method for studying mRNA phenotypes in single or small numbers of cells. J. Cell Biochem. 39, 1-11. Rassendren, F., Buell, G., Virginio, C., Collo, G., North, R. A. and Surprenant, A. (1997). The permeabilising ATP receptor, P2X7. J. Biol. Chem. 272, 5482-5486. Reisin, I. L., Prat, A. G., Abraham, E. H., Amara, J. F., Gregory, R. J., Ausiello, D. A. and Cantiello, H. F. (1994). The cystic fibrosis transmembrane conductance regulator is a dual ATP and chloride channel. J. Biol. Chem. 269, 20584-20591. Ryan, L. M., Rachow, J. W. and McCarty, D. J. (1991). Synovial fluid ATP: a potential substrate for the production of inorganic pyrophosphate. J. Rheumatol. 18, 716-720. Shats, E. A., Nair, C. H. and Dhall, D. P. (1997). Interaction of endothelial cells and fibroblasts with modified fibrin networks: role in atherosclerosis. Atherosclerosis 129, 9-15. Sigari, F., Lee, C., Witzum, J. L. and Reaven, P. D. (1997). Fibroblasts that overexpress 15-lipoxygenase generate bioactive and minimally modified LDL. Arterioscl. Thromb. Vasc. Biol. 17, 3639-3645. Soto, F., Garcia-Guzman, M., Gomez-Hernandez, J. M., Hollman, M., Karschin, C. and Stuhmer, W. (1996). P2X4: an ATP-activated ionotropic receptor cloned from rat brain. Proc. Natl. Acad. Sci. USA 93, 3684-3688. Steinberg, T. H., Newman, A. S., Swanson, J. A. and Silverstein, S. C. (1987). ATP4− permeabilizes the plasma membrane of mouse macrophages to fluorescent dyes. J. Biol. Chem. 262, 8884-8888. Surprenant, A., Rassendren, F., Kawashima, E., North, R. A. and Buell, G. (1996). The cytolytic P2Z receptor for extracellular ATP identified as a P2X receptor (P2X7). Science, 272, 735-738. von Kugelgen, I. and Starke, K. (1991). Release of noradrenaline and ATP by electrical stimulation and nicotine in guinea-pig vas deferens. NaunynSchmiedeberg’s Arch. Pharmacol. 344, 419-420.