Ionotropic purinergic receptor P2X4 is involved in the regulation of chondrogenesis in chicken micromass cell cultures

UNCORRECTED PROOF Please cite this article in press as: J. Fodor, et al., Ionotropic purinergic receptor P2X4is involved in the regulation of chondrogenesis in chicken micromass cell cultures, Cell Calcium (2009), doi:10.1016/j.ceca.2009.02.004 ARTICLE IN PRESS G Model YCECA10501–10 Cell Calcium xxx (2009) xxx–xxx Contents lists available at ScienceDirect Cell Calcium journal homepage: www.elsevier.com/locate/ceca Ionotropic purinergic receptor P2X4is involved in the regulation of chondrogenesis in chicken micromass cell cultures 1 2 János Fodora,1, Csaba Mattab,1, Tamás Juhászb, Tamás Oláha, Mónika Gönczia, Zsolt Szíjgyártóc, Pál Gergelyc, László Csernocha, Róza Zákányb,∗ 3 4 aDepartment of Physiology, Medical and Health Science Centre, University of Debrecen, Hungary5 bDepartment of Anatomy, Histology and Embryology, Medical and Health Science Centre, University of Debrecen, Nagyerdei krt. 98, H-4032 Debrecen, Hungary6 cCell Biology and Signalling Research Group of the Hungarian Academy of Sciences, Department of Medical Chemistry, Research Centre for Molecular Medicine, Medical and Health Science Centre, University of Debrecen, Hungary 7 8 9 article info10 11 Article history:12 Received 5 August 200813 Received in revised form 2 February 200914 Accepted 9 February 2009 15 Available online xxx16 17 Keywords:18 In vitro cartilage formation19 High density cell culture20 Fura-221 Single cell Ca measurement22 P2X receptors 23 P2Y receptors24 ATP secretion25 Immunocytochemistry26 abstract We have previously demonstrated that elevation of free cytosolic Ca2+ concentration at the time of differentiation of chondroblasts was mainly due to a Ca2+ influx and it was indispensable to cartilage formation in chicken high density mesenchymal cell cultures (HDC) [C. Matta, J. Fodor, Z. Szijgyarto, T. Juhasz, P. Q1 Gergely, L. Csernoch, R. Zakany, Cytosolic free Ca2+ concentration exhibits a characteristic temporal pattern during in vitro cartilage differentiation: a possible regulatory role of calcineurin in Ca-signalling of chondrogenic cells, Cell Calcium 44 (2008) 310–323]. Here, we report that chondrogenic cells secreted ATP and administration of ATP to the culture medium evoked Ca2+ transients exclusively in the presence of extracellular Ca2+ and only on day 3 of culturing, when the final commitment of chondroblasts occurs. Moreover, ATP caused elevated protein expression of the chondrogenic transcription factor Sox9 and stimulated cartilage matrix production. Expression pattern of different types of both ionotropic and metabotropic purinergic receptors was detected. Agonists of metabotropic receptors, ADP and UDP did not evoke any Ca2+ transients andhad noinfluence on cartilageformation, while UTP caused transientelevation of cytosolic Ca2+ concentration in 3-day-old HDC without stimulating matrix production. Suramin, whichblocks all P2X receptorsbut not P2X4didnotimpede the effects of ATP,furthermore,P2X4appeared in the plasma membrane fraction and gave signals with immunocytochemistry only from day 3. In summary, we suggest a role of ionotropic purinergic signalling of P2X4in the generation of ATP-dependent Ca2+ transients of differentiating chondroblasts. © 2009 Published by Elsevier Ltd. 1. Introduction27 Complex regulatory and signalling networks involving28 cell–matrix and cell–cell interactions, including tightly regulated 29 gene expression, mediate the successive stages of proliferation,30 nodule formation and differentiation that produce hyaline carti31 lage [2]. One of the key regulators of these signalling processes32 in chondrogenic cells is Sox9. Since the expression of collagen33 type II and the core protein of aggrecan are controlled by this 34 transcription factor, Sox9 is often referred to as the master gene of 35 chondrogenesis [3,4]. Another important factor in the regulation36 of molecular steps leading to chondrogenic differentiation is the37 transient elevation of the intracellular Ca2+ concentration [1]. The38 tightly regulated level of cytosolic Ca2+ is involved in a number of39 ∗Corresponding author. Fax: +36 52 432 290. E-mail address: r[email protected]te.hu (R. Zákány). 1These two authors contributed equally to the work. signalling processes in a variety of cell types. Among non-excitable 40 cells, the role of intracellular Ca2+ in the differentiation process 41 of keratinocytes [5] and osteoblasts [6] has been established. 42 Cytosolic-free Ca2+ concentration changes are characterized by 43 long-term, high amplitude changes, and by short-term, sponta44 neous, periodic Ca2+ concentration changes, so called oscillations 45 in differentiating mesenchymal stem cells [7]. Moreover, the two 46 different types of Ca2+ concentration changes influence the activity 47 of different transcription factors: oscillations activate CREB, while 48 long-time sustained Ca2+ concentration elevations activate NFAT 49 [8,9]. Both transcription factors have important functions during 50 chondrogenesis [10,11].51 High density cell culture (HDC) established from chondrogenic 52 mesenchymal cells isolated from distal limb buds of 4-day-old 53 chicken embryos is a widely accepted model of in vitro cartilage 54 differentiation [12,13], providing data on the molecular regula55 tion of the differentiation of chondroprogenitor mesenchymal cells 56 to chondroblasts. In this model a spontaneous cartilage forma57 tion occurs; the initial appearance of chondroblasts and cartilage 58 0143-4160/$ – see front matter © 2009 Published by Elsevier Ltd. doi:10.1016/j.ceca.2009.02.004 UNCORRECTED PROOF Please cite this article in press as: J. Fodor, et al., Ionotropic purinergic receptor P2X4is involved in the regulation of chondrogenesis in chicken micromass cell cultures, Cell Calcium (2009), doi:10.1016/j.ceca.2009.02.004 ARTICLE IN PRESS G Model YCECA10501–10 2J. Fodor et al. / Cell Calcium xxx (2009) xxx–xxx specific extracellular matrix molecules takes place on day 3 of cul-59 turing, and the majority of cells differentiate into chondrocytes by60 day 6, when a high amount of cartilage matrix can be detected.61 In our previous work [1] we reported that a tight regulation of62 free cytosolic Ca2+ levels (between 80 and 140 nM) is needed for63 proper chondrogenesis in cells of HDC. We also described a char-64 acteristic temporal pattern of the changes of cytosolic Ca2+ levels65 during chondrogenic differentiation with a definitive peak on day66 3 of culturing, the day on which chondroprogenitor cells differ-67 entiate to chondroblasts. Although intracellular elements of Ca2+ 68 homeostasis (e.g. RyR, IP3receptor and SERCA) were detected in69 chondrogeniccells, wefailed toshowanyevidenceconcerningtheir70 contributioninevokingCa2+ transientsin differentiatingchondrob-71 lasts. Therefore, the extracellular space has been proved to be the72 source of the elevated cytosolic Ca2+ concentration.73 In the present work, we aimed to determine transmembrane74 proteincandidatesresponsiblefortheCa2+ influxintochondrogenic75 cells. The family of purinergic receptors is ubiquitously present76 in a number of cell types and provides receptors for extracellular77 nucleotides acting as paracrine or autocrine mediators. Purinergic78 receptors have two major types: P1 receptor families are sen79 sitive to adenosine, while P2 receptor families are sensitive to80 ATP, ADP, and UTP. The latter is further divided into two major81 receptor subtypes: P2Y and P2X. Members of the metabotropic82 P2Y subtype are 7 transmembrane domain-containing receptors83 coupled to G proteins and linked to PLC signalling transduction84 pathways that lead to the release of intracellular Ca2+ from inositol-85 1,4,5-trisphosphate (IP3)-sensitive Ca2+ stores. The ionotropic P2X86 receptors are ATP-gated ion channels allowing Ca-influx. Seven87 P2X subunits (P2X1–P2X7) have been described and cloned so far88 [14,15]. On the other hand, eight P2Y isoforms have been described 89 in human tissues: P2Y1, P2Y2, P2Y4, P2Y6, P2Y11 , P2Y12, P2Y13 and90 P2Y14 receptors. Although, based on sequence analysis P2Y3and91 P2Y5were supposed to be members of the P2Y subfamily, fur-92 ther studies revealed that they have different phamacology and are93 involved in non-purinergic signalling pathways as reviewed in [16].94 Thefact thatP2X receptorsaresequentiallyexpressedin embryonic95 rat and mouse skeletal muscle cells and osteoblasts [17-19] raised96 the possibility of the involvement of these channels in the Ca2+ 97 homeostasis of differentiating chondrogenic mesenchymal cells. 98 In this study, we report that cells of HDC responded to extra-99 cellular ATP by elevating their intracellular Ca2+ levels mainly at100 the time of chondroblast formation and administration of ATP to101 the culture medium stimulated chondrogenesis. Receptors respon-102 sible for the elevation of Ca2+ concentration seemed to be members103 of the P2X family, and based on our data, we propose that P2X4 104 receptorscontributetotheelevationofcytosolicCa2+ levelsofchon-105 drogenic cells on day 3 of culturing. Moreover, cells of HDC secreted106 ATP into the culturing medium, which supports our theory that a107 purinergic autocrine regulation is involved in the proper control of108 chondrogenesis.109 2. Materials and methods110 2.1. Cell culture111 High density cell cultures were prepared as described in [1].112 Briefly,distalpartsofthelimbbudsof4-day-oldRosshybridchicken113 embryos (Hamburger–Hamilton stages 22–24 [20])wereremoved 114 and chondrifying micromass cultures of mesenchymal cells were 115 established. 15 or 30␮L droplets of the suspension containing 116 1.5 ×107cells/mL were inoculated on round coverglasses (diam-117 eter: 30 or 10mm; Menzel–Gläser, Menzel GmbH, Braunschweig,118 Germany)placedintoplasticPetridishes(Nunc,Naperville,IL,USA). 119 Cells were allowed to attach to the surface for 2hat 37◦C. Day of120 inoculation is considered as day 0. Colonies were grown in Ham’s 121 F12 medium (Sigma, Budapest, Hungary) supplemented with 10% 122 fetal calf serum (Gibco, Gaithersburg, MD, USA), antibiotics and 123 antimycotics, and were kept at 37◦C in an atmosphere of 95% air 124 and 5% CO2and 80% humidity. The medium was changed on every 125 second day. 126 2.2. Single cell Ca2+ measurements 127 Measurements were performed on different days of culturing 128 using the calcium dependent fluorescent dye Fura-2 as described 129 previously [1]. Fura-2-loaded cells were placed on the stage of 130 an inverted fluorescent microscope (Diaphot, Nikon, Kowasaki, 131 Japan) and viewed using a 40×oil immersion objective. Mea132 surements were performed in normal (137mM NaCl, 5.4mM KCl, 133 0.5mM MgCl2, 1.8mM CaCl2, 11.8mM HEPES, 1 g/L glucose, pH 7.4) 134 or calcium-free (containing 5mM EGTA, without CaCl2) Tyrode’s 135 solution. ATP solution was prepared from normal and Ca2+-free 136 Tyrode’sat180␮Mfinalconcentration.ADP,UDP,UTP(180␮M)and 137 bradykinin (20␮M) were prepared from Ca2+-free Tyrode’s. Before 138 application of Ca2+-free ATP, ADP, UDP and UTP, cells were treated 139 in Ca2+-free Tyrode’s for 150 s. Suramin solution was prepared in 140 Tyrode’s and used at a final concentration of 10 ␮M. Excitation 141 wavelength was altered between 340 and 380nm and fluorescence 142 intensities (F340 and F380) were measured as described previously 143 [1]. Test solutions were directly applied to the cells through a per144 fusion capillary tube (Perfusion PencilTM; AutoMate Scientific, San 145 Francisco,CA,USA)withaninternaldiameterof250␮mata1.5␮L/s 146 rate, using a local perfusion system (Valve BankTM 8 version 2.0, 147 AutoMate Scientific). All measurements were performed at room 148 temperature. Data were statistically analyzed by Student’s t-test. 149 2.3. Preparation of cell extracts 150 Cell cultures were harvested on each day of culturing. Cell pel151 lets were suspended in 100 ␮L of homogenizationbuffer containing 152 50mM Tris–HCl buffer (pH 7.0), 10␮g/mL Gordox, 10␮g/mL leu153 peptin, 1mM phenylmethylsulphonyl (PMSF), 5mM benzamidine, 154 10 ␮g/mL trypsin inhibitor and 0.5% Triton X-100. Samples were 155 snap-frozen in liquid nitrogen, and were stored at −70 ◦C. Samples 156 were sonicated for four times 30s by 50 cycles (Branson Soni157 fier, Danbury, USA). For Western blot analyses, total cell lysates 158 and plasma membrane fractions were used. For isolation of plasma 159 membrane fraction of HDC, sonicated samples were centrifuged at 160 50,000×gfor 90min at 4◦C. Pellet was triturated continuously in 161 50␮L homogenization buffer supplemented with 1% Triton X-100 162 at 4◦C. After 1h of trituration samples were centrifuged again at 163 50,000×gfor 55min at 4◦C, and supernatant containing plasma 164 membrane fraction was used for Western blot analyses. 165 2.4. RT-PCR analysis 166 For RT-PCR analysis, cartilage colonies were washed three times 167 with RNase-free physiological sodium chloride, snap-frozen in liq168 uid nitrogen and stored at −70 ◦C. Total RNA was isolated from cells 169 of HDC of various ages using Quiagen RNeasy®Micro Kit accord170 ing to the instructions of the manufacturer (Quiagen, Budapest, 171 Hungary). The assay mixture (20␮L) for reverse transcriptase reac172 tion (Omniscript, Quiagen) contained 500 ng RNA, 0.25 ␮LRNase 173 inhibitor, 0.25␮Loligo (dT), 1␮L dNTP (200␮M), 1␮l M-MLV 174 RT in 1×RT buffer. Amplifications of specific cDNA sequences 175 were performed with specific primers (Integrated DNA Technolo176 gies, Coralville, IA, USA) that were designed based on published 177 chicken nucleotide sequences (for sequences of primer pairs, see 178 Supplementary Material, Table 1). PCR reactions were allowed to 179 proceed in a final volume of 50␮L (containing 2 ␮L forward and 180 UNCORRECTED PROOF Please cite this article in press as: J. Fodor, et al., Ionotropic purinergic receptor P2X4is involved in the regulation of chondrogenesis in chicken micromass cell cultures, Cell Calcium (2009), doi:10.1016/j.ceca.2009.02.004 ARTICLE IN PRESS G Model YCECA10501–10 J. Fodor et al. / Cell Calcium xxx (2009) xxx–xxx 3 reverse primers, 1␮L dNTP [200␮M], and 5 units Promega GoTaq® 181 DNA polymerase in 1×reaction buffer) in a programmable ther-182 mocycler (Eppendorf Mastercycle, Netheler, Hinz GmbH, Hamburg,183 Germany) with the following settings: 2min at 95◦C for initial184 denaturation followed by repeated cycles of denaturation at 94◦C185 for 1min, primer annealing for 60s at an optimized temperature,186 and extension at 72◦Cfor1min30s.After the final cycle, further187 extension was allowed to proceed for another 10 min at 72◦C. PCR188 products were analyzed using a 1.5% ethidium bromide-stained189 agarose gel.190 2.5. Western blot analysis191 Total cell lysates and plasma membrane fractions were exam-192 ined by Western blot. Samples for SDS-PAGE were prepared by193 the addition of 1/5 volume of 5-fold concentrated electrophore-194 sis sample buffer (310 mM Tris–HCl, pH 6.8; 10% SDS, 50% glycerol,195 100 mM DTT, 0.01% bromophenol blue) to cell lysates and boiled196 for 5min. About 50␮g of protein was separated by 7.5% SDS-PAGE197 gel for immunological detection of P2X-receptors. Proteins were 198 transferred electrophoretically to nitrocellulose membranes. After199 blocking in 5% non-fat dry milk in PBS, membranes were incu-200 bated with primary antibodies raised against the carboxy termini201 of P2X-receptors (Alomone Labs, Jerusalem, Israel) and P2Y4recep-202 tor (Sigma, Budapest, Hungary), amino terminus of P2Y1receptor203 (Sigma, Budapest, Hungary) and 3rd intracellular loop of P2Y2 204 receptor(AlomoneLabs,Jerusalem,Israel)overnightat4◦Cin1:200205 dilution. After washing three times for 10min with PBST (PBS sup-206 plemented with 0.1% Tween 20), membranes were incubated with207 a secondary antibody, anti-rabbit IgG (Sigma, Budapest, Hungary) 208 in 1:1000 dilution in PBS containing 5% non-fat dry milk for 1h.209 Signals were detected by enhanced chemiluminescence reaction210 (Amersham Biosciences, Budapest, Hungary).211 2.6. Immunocytochemical staining of P2X receptors212 3-day-old cultures were washed twice with PBS and fixed in213 4% paraformaldehyde for 15min at 4◦C. After washing in PBS, cells214 were permeabilized with 0.1% Triton X-100 in PBS for 30min. Non-215 specific binding sites were blocked by 30min preincubation in 1% 216 bovine serum albumin (BSA) in PBS, followed by incubation with217 the primary antibodies diluted in 1:100 at 4 ◦C overnight. Subse-218 quently, the cultures were washed three times with PBS for 10min, 219 and were incubated with a FITC-conjugated anti-rabbit IgG (Vec-220 tor Laboratories, Burlingame, CA, USA) diluted 1:500 in PBS for 1h.221 Cultures were washed three times with PBS and mounted with222 Vectashield®mounting medium containing DAPI (Vector Labora-223 tories, Burlingame, CA, USA). Control experiments were carried224 out with primary antibodies incubated with their control peptides225 according to the instructions of the manufacturer (data not shown). 226 2.7. Administration of extracellular ATP, ADP, UDP, UTP and227 suramin228 ATP (100 ␮M), ADP, UDP and UTP (180␮M), and suramin229 (10␮M) were administered to cells of HDC on various days of cul230 turing. The nucleotides and suramin were diluted in the culture231 medium.Effectsonmetachromaticcartilagematrixformationwere232 examinedbymetachromaticstainingwithdimethylmethyleneblue233 and toluidine blue as described previously [1].234 2.8. Determination of extracellular ATP in the culture medium235 Concentration of extracellular ATP secreted by cells of high den236 sity cell cultures was determined using Adenosine 5-triphosphate237 (ATP) Bioluminescent Assay Kit (Sigma, Budapest, Hungary). Mea238 surements were carried out according to the instructions of the 239 manufacturer, with minor modifications. Briefly, 20 droplets of the 240 cell suspension (100 ␮L each) were inoculated into Petri dishes 241 (diameter: 200mm, Orange Scientifique, Braine-l’Alleud, Belgium) 242 and were fed with 20mL culture medium. Concentration of ATP 243 secretedbycells ofHDC intotheculturemedium wasdeterminedat 244 approximately the same period of each culturing day. The medium 245 was changed every day following measurements. 50␮L of the cul246 ture medium (pH adjusted to 7.8) was used to determine the 247 amount of ATP in the culture medium in 2 parallel experiments. 248 Background light emission was determined using blanks (both ster249 ile water and Ham’s F12 culture medium). Since the ATP Assay Mix 250 is not stable for a long period, a gradual decrease in the sensitivity 251 may occur. Therefore, a new standard curve was prepared each day 252 prior to measurements (concentrations of ATP standard solutions 253 were as follows: 10−6,10 −7,10 −8,10 −9and 10−10 M). Lumines254 cence of samples was determined using a microwell plate reader 255 (Chameleon, Hidex, Turku, Finland). 256 3. Results 257 3.1. Cells of HDC respond to extracellular ATP by elevating 258 intracellular Ca2+ concentration 259 ATP at a constant concentration of 180␮M was administered to 260 the close proximity of cells of HDC on various days of culturing. 261 Fig. 1 (A–E) shows that administration of ATP could induce a tran262 sientincreasein intracellularCa2+ levelsincells of aCa2+ containing 263 bathing solution. Note that both the amplitude of the average 264 response (maximal increase in intracellular Ca2+ concentration) 265 and the time of exposure to ATP needed to induce the transient 266 exhibited a differentiation-dependent pattern (see also Fig. 1F–G). 267 The shortest exposure of ATP needed to evoke a Ca2+ transient was 268 characteristic to culturing days 3 and 4, and at the same time Ca2+ 269 transients with the highest amplitude (179 and 165nM, respec270 tively) were also recorded on these days. On the first culturing day 271 ATP, even though administered for a long period (120 s), could not 272 evoke any changes in the intracellular Ca2+ concentration (Fig. 1A), 273 furthermore, we could hardly detect characteristic peaks in 2-day274 old cells either (Fig. 1B). By day 6, the amplitude of the transients 275 decreasedandonlya prolongedapplicationof ATPcouldevokesuch 276 responses (Fig. 1E). 277 Striking differences were also observed regarding the number of 278 cells responding to ATP (Fig. 1F and G). While most of the cells (90%) 279 respondedtoATPin3-day-oldcultures,essentiallynonedidatday1 280 of culturing. Administration of ATP on other days could also induce 281 calcium transients, but the proportion of cells that responded was 282 hardly comparable with that on day 3 (Fig. 1F). 283 To establish whether metabotropic or ionotropic purinergic 284 receptors were responsible for these effects, ATP was administered 285 to cells in a Tyrode’s solution lacking free Ca2+ (Fig. 2A and B). In the 286 30 cells examined, no response was detected on either days of cul287 turing in the absence of extracellular Ca2+. This observation firmly 288 supportedour theory that influx ofextracellularCa2+ wasneeded to 289 evoke the effect of extracellular ATP and the receptor of ATP could 290 be a member of the ionotropic purinergic receptor family (P2X), but 291 did not exclude the role of metabotropic purinergic receptors and 292 intracellular Ca2+ stores. 293 To find candidates among P2X receptors, their non-specific 294 antagonist suramin was tested on cells of HDC. It is known that 295 suramin inhibits all P2X receptors except P2X4and P2X6[16].296 Cells treated with suramin (10␮M) showed no significant alter297 ation in the intracellular Ca2+ level following the administration 298 of ATP (Fig. 2C). Repetitive administration of ATP could induce 299 UNCORRECTED PROOF Please cite this article in press as: J. Fodor, et al., Ionotropic purinergic receptor P2X4is involved in the regulation of chondrogenesis in chicken micromass cell cultures, Cell Calcium (2009), doi:10.1016/j.ceca.2009.02.004 ARTICLE IN PRESS G Model YCECA10501–10 4J. Fodor et al. / Cell Calcium xxx (2009) xxx–xxx Fig. 1. Effect of 180␮M ATP on the cytosolic Ca2+ levels of Fura-2-loaded cells of HDC. (A–E) Ca2+ transients evoked by administration of ATP in cells on different days of culturing. Representative records of 5 independent experiments. Lines indicate the application of ATP. (F) Ca2+ transients were measured in the presence of 1.8mM external calcium. Ratio of cells responding to ATP on each day of culturing. Representative data of 5 independent experiments. (G) Changes in the peak amplitude of calcium transients detected on different days of culturing. Numbers indicate the proportion of cells responding to ATP. Representative data of 5 independent measurements. repetitive and transient elevations in intracellular Ca2+ concentra-300 tion in 3-day-old HDC (Fig. 2D). Relatively short periods of washing 301 (approximately 100s) were required to allow the cells to recover302 from the 30-sec-long exposures of ATP. This observation, together303 with the lack of significant desensitization (Fig. 2D) raised the pos-304 sibility of the presence and function of P2X4and/or P2X6receptor305 subtypes.306 Although the above results clearly suggested that P2X recep307 tors play the decisive role, we also carried out experiments to 308 obtain data on the function of metabotropic P2Y receptors. First309 we intended to examine whether intracellular Ca2+ stores are310 presentandcontainreleasableCa2+ bytheactivationofIP3pathway.311 Since bradykinin receptors are known to activate this pathway and312 are described as being expressed by chondrocytes [21], therefore313 bradykinin was administered to 3-day-old cells at a concentration314 of 20␮M. A slight elevation (30nM) of free cytosolic Ca2+ concen-315 tration was observed in 60% of cells proving the presence and active316 functioning of IP3signalling (Fig. 3A).317 ADP, UDP and UTP are non-specific agonists of metabotropic318 purinergic receptors (P2Y). These compounds were administered319 to cells of HDC at a concentration of 180␮M on day 3 of culturing320 (Fig. 3B–D). Slight elevation of cytosolic Ca2+ was detected only in321 50% of cells measured during the administration of UTP. The aver322 age amplitude of UTP-evoked Ca2+ transients was 57nM. On the323 other hand, administration of ADP and UDP did not result in any 324 significant Ca2+ transients.325 3.2. Chondrogenic mesenchymal cells express various P2X and 326 P2Y receptor subtypes during differentiation 327 To identify the presence and expression pattern of various 328 purinergicreceptorsduringchondrogenicdifferentiationofchicken 329 mesenchymal cells, RT-PCR reactions were performed. mRNA 330 sequences of chicken P2X receptors, but not of P2X6(not yet pub331 lished) as well as P2Y1, P2Y3, and P2Y5receptors were downloaded 332 from GenBank and specific primer pairs for each mRNA sequence 333 were designed for amplification (see Supplementary Material). 334 Amplimers of expected sizes were identified for all the avail335 able mRNAs, except for P2X2, where only very weak signals were 336 detected(Fig.4A). mRNAexpressionofP2X1and P2X7receptorsub337 types followed a peak-like pattern during differentiation with the 338 highest expression levels on days 3 or 4, respectively. P2X3receptor 339 subtype mRNA exhibited a rather variable expression profile: the 340 strongest bands were detected between days 1 and 3 of culturing. 341 P2X4and P2X5receptor subtypes showed the strongest expres342 sion levels. Both receptors expressed markedly on day 1 then the 343 signal became gradually weaker. mRNAs of P2Y1, P2Y3and P2Y5344 showed constant expression levels throughout the culturing period 345 (Fig. 6B). 346 Western blot analysis showed a different expression profile for 347 the different P2X receptor subtypes (Fig. 4B). We could not detect 348 the P2X2receptor subtype, and no signals were visible for P2X3349 and P2X6, either in total lysates or in isolated plasma membrane 350 UNCORRECTED PROOF Please cite this article in press as: J. Fodor, et al., Ionotropic purinergic receptor P2X4is involved in the regulation of chondrogenesis in chicken micromass cell cultures, Cell Calcium (2009), doi:10.1016/j.ceca.2009.02.004 ARTICLE IN PRESS G Model YCECA10501–10 J. Fodor et al. / Cell Calcium xxx (2009) xxx–xxx 5 Fig. 2. Responses of cells of HDC to administration of ATP on day 3 of culturing. (A) Record showing the lack of ATP-evoked Ca2+ transients in the absence of external calcium. (B) Basal intracellular Ca2+ levels and the peak amplitude of ATP-evoked Ca2+ transients in the presence and absence of external calcium. Numbers in parentheses show the number of cells measured. Data represent mean±standard error of the mean of intracellular Ca2+ levels of cells assayed in 5 independent experiments. Asterisk indicates significant (*P<0.01) increase in peak amplitude of ATP-evoked Ca2+ transients as compared to the respective control. (C) Effect of the P2X antagonist suramin (10␮M) on ATP-evoked calcium transients in the presence of external calcium. (D) Calcium transients evoked by repeated administration of ATP in the presence of external calcium showing the lack of desensitization of P2X receptors in differentiating chondrocytes. Representative record of 5 independent experiments is presented in panels A, C or D. Lines in panels A, C and D indicate the application of ATP. Preceding the application of Ca-free ATP cells were treated in Ca-free Tyrode’s for 150 s. fractions (data not shown). Protein expression of P2X1subtype in351 total lysates followed a similar profile to the mRNA expression, but352 in the plasma membrane fractions strong bands were detected on353 days 1–3, thereafter the protein levels markedly decreased. In con-354 trast with the results of RT-PCR reactions, protein levels of P2X5 355 receptor subtype were hardly detectable in total cell lysates, but356 in plasma membrane fractions of day 4 was characterized by a 357 stronger signal.358 For P2X7, a profile showing a variable expression pattern was359 observed in total lysates with the strongest bands on days 2 and360 4, respectively, however, in the plasma membrane fractions a361 peak-like pattern with strongest bands on days 2, 3 and 4 was 362 observed. 363 Nevertheless, protein expression of P2X4receptor subtype 364 proved to be the most interesting (Fig. 4B). While in total cell lysates 365 it showed a rather variable profile, in isolated plasma membrane 366 fractions it first appeared on day 3 with a strong band, and by 367 days 4 and 6 its expression rapidly diminished. It is important to 368 note that the vast majority of chondrogenic mesenchymal cells 369 responded to ATP on this day of culturing, which also coincides 370 with the day of differentiation characterized by elevated cytoso371 lic Ca2+ levels reported earlier [1]. Presence of P2X1, P2X4and 372 Fig. 3. Functional characterization of P2Y receptors in cells of HDC on day 3 of culturing. (A) 20␮M bradykinin-evoked Ca2+ transients measured in the absence of external calcium. Line indicates the application of bradykinin. (B) Effect of 180 ␮M ADP on Ca2+ transients measured in the absence of external calcium. Line indicates the application of ADP. (C) Effect of 180␮M UDP on Ca2+ transients measured in the absence of external calcium. Line indicates the application of UDP. (D) 180␮M UTP-evoked Ca2+ transients in the absence of external calcium. Line indicates the application of UTP. Representative records of 3 independent experiments. UNCORRECTED PROOF Please cite this article in press as: J. Fodor, et al., Ionotropic purinergic receptor P2X4is involved in the regulation of chondrogenesis in chicken micromass cell cultures, Cell Calcium (2009), doi:10.1016/j.ceca.2009.02.004 ARTICLE IN PRESS G Model YCECA10501–10 6J. Fodor et al. / Cell Calcium xxx (2009) xxx–xxx Fig. 4. Expression pattern of P2X receptor subtypes in cells of HDC on different days of culturing. (A) mRNA expression pattern of various P2X receptors was detected by RT-PCR reactions. P2X receptor subtypes (but not P2X6) were amplified using specific primers and detected at expected sizes. GAPDH was used as a control. (B) Western blot analysis of P2X receptor proteins in cells of HDC. Total protein and membrane fraction samples were used (50␮g in each lane) to examine the protein expression level. Representative data each of 3 independent experiments, performed in triplicates. (C) Immunocytochemical staining of 3-day-old HDC demonstrating presence of P2X1, P2X4 and P2X7receptors. Arrows indicate the accumulation of immunopositive signal in the plasma membrane of chondrogenic cells. Images were recorded from 1␮m thick optical slices. Original magnification was 40×. Representative images of 3 independent experiments, each performed in triplicates. P2X7receptorswasalsoprovedbyimmunocytochemicalstainingof373 HDC. A membrane-bound localization was clearly visible for P2X4 374 receptors (Fig. 4C). Furthermore, presence of P2X4in cartilagineous375 primordiain developinglimbsof chicken embryoswas also demon-376 stratedatadevelopmentalstage(8-day-oldembryo)corresponding377 to approximately 4-day-old HDC (data not shown). 378 We also detected the expression of metabotropic purinergic379 receptors in HDC. As P2Y3and P2Y5receptors are not regarded as380 functional members of this family of receptors [16], we investigated 381 P2Y1, P2Y2, and P2Y4protein in total cell lysates and plasma mem-382 brane fractions. On day 1, the protein of P2Y1was not expressed383 by cells of HDC, then it was present at a constant level both in384 total cell lysates and in plasma membrane fractions until day 6, 385 when it showed a small decline. The P2Y2receptor protein was386 found to be expressed in a peak like pattern in total lysates with387 strongestsignals ondays2-4.However,wedetectedaconstantlevel 388 of expression in the plasma membrane fraction with the exception 389 of day 6, when the signal became weaker. We could only detect390 specific signals for P2Y4receptor in the plasma membrane frac-391 tion with a stronger band on day 2, but no immunopositivity was392 observed in total lysates (Fig. 6C).393 3.3. Administration of extracellular ATP on day of differentiation 394 increases matrix production 395 In order to support our hypothesis that the entrance of extra396 cellular Ca2+ into chondrogenic mesenchymal cells is via P2X 397 receptors, especially via P2X4subtype, further experiments were 398 performed. ATP was administered at various concentrations tocells 399 of high density cultures on day 3 of culturing. At the concentration 400 of 100 ␮M, extensive matrix production occurred by day 6 (Fig. 5A) 401 demonstrated by both DMMB and TB stainings. mRNA levels of col402 lagen II and the core protein of aggrecan also reflected the slightly 403 higher rate of matrix production under the effect of ATP (Fig. 5B). 404 Although mRNA expression level of Sox9 did not change (Fig. 5B), 405 protein expression of this transcription factor became higher as 406 a result of the administration of ATP (Fig. 5C). Treatment of HDC 407 with ATP on days 2 and 4 of culturing did not alter the cartilage 408 matrix production (data not shown). These results indicate that 409 ATP has a positive effect on both cartilage matrix production and 410 chondroblast differentiation when it is applied at the time of final 411 commitment of chondroprogenitor cells and ATP does not exert any 412 effect on premature or mature chondroblasts. 413 UNCORRECTED PROOF Please cite this article in press as: J. Fodor, et al., Ionotropic purinergic receptor P2X4is involved in the regulation of chondrogenesis in chicken micromass cell cultures, Cell Calcium (2009), doi:10.1016/j.ceca.2009.02.004 ARTICLE IN PRESS G Model YCECA10501–10 J. Fodor et al. / Cell Calcium xxx (2009) xxx–xxx 7 Fig. 5. Effects of ATP (100 ␮M) and suramin (10 ␮M) on cartilage development of chondrifying micromass cultures. Both chemicals were administered on day 3 of culturing. (A) Metachromatic cartilage areas in 6-day-old high density colonies visualized with DMMB dissolved in 3% acetic acid. Optical density (OD625) of samples containing toluidine blue extracted with 8% HCl dissolved in absolute ethanol. Data are mean values of each experimental group out of 5 measurements. Standard errors of the means were within±8%. Asterisks indicate significant (*P<0.01) increase in optical density of extracted toluidine blue as compared to the respective control. (B) mRNA expression of collagen type II, core protein of aggrecan and sox9 after treatment with ATP and/or suramin. GAPDH was used as a control. Representative data of 3 independent experiments, each performed in triplicates. (C) Effect of ATP and/or suramin on the protein expression of Sox9. Representative data of 3 independent experiments, each performed in triplicates. Administration of the non-specific P2X receptor antagonist 414 suramin alone or combination with ATP did not cause any signifi-415 cant alteration in the amount of cartilage matrix produced by the416 end of the 6-day-old culturing period (Fig. 5A). Moreover, the ATP417 stimulated Sox9 expression was not affected by suramin treatment, 418 further supporting our theory that P2X4receptor could be involved419 inthetransmissionofthechondrogenesispromotingeffectofextra-420 cellular ATP (Fig. 5B and C). 421 3.4. Administration of ADP, UDP and UTP to the culture medium422 has no effect on cartilage formation423 We examined the administration of the nucleotides on car-424 tilage matrix production of HDC to elucidate a putative role of425 metabotropic purinergic receptors. The nucleotides applied at a426 concentration of 180␮M into the culturing medium on day 3 did427 not exert any effect on the amount of cartilage matrix produced by428 theend of the 6-day-longculturing period asrevealedbymetachro-429 matic staining (Fig. 6A).430 3.5. Cells of high density cultures secrete ATP into the culture 431 medium 432 The demonstration of the effectiveness of administration of ATP433 on matrix production raises the question whether the chondro434 genic mesenchymal cells secrete ATP into the culture medium as 435 an autocrine mediator to promote and facilitate their own differen436 tiation. To investigate this, the culture medium was removed from437 the cells of high density cultures on each day of culturing and ATP438 assays were performed. We found that on each day of culturing a 439 small amount of ATP was detectable in the culture medium in the440 range of 2–10nM, which is comparable to data measured in culture 441 medium of other non-excitable cells [22].442 4. Discussion 443 In vitro chondrogenesis is a dynamic, multistep process regu444 lated by a variety of molecular processes, many of which involve 445 activation and deactivation of protein kinases and phosphatases 446 sensitivetochangesof intracellularCa2+ levels.Inchickenhigh den447 sity mesenchymal cell cultures, chondrogenic mesenchymal cells 448 differentiate into chondroblasts and then to chondrocytes during a 449 6-day-long culturing period. The majority of chondroblasts, char450 acterized by the ability of production of a cartilage specific ECM, 451 appear from culturing day 3. 452 We have previously demonstrated that cytoplasmic-free Ca2+ 453 concentration of chondrogenic cells exhibited a characteristic tran454 sient elevation on day 3 of culturing. This has been found to be 455 indispensable to proper differentiation and the essential role of the 456 influx of extracellular Ca2+ has been documented [1]. Intracellular 457 Ca2+ stores have been shown to contain releasable Ca2+, but the rate 458 of leak was low and free cytoplasmic Ca2+ concentration became 459 only slightly higher in the absence of extracellular Ca2+. Moreover, 460 RyR and IP3receptors have been found to be expressed weakly and 461 stimulation of RyR did not result in the elevation of cytoplasmic 462 Ca2+. Our data have underlined the role of Ca2+ influx from extra463 cellular space in the generation of the cytoplasmic Ca2+ peak. The 464 intracellular stores seemed to be contributing to the maintenance 465 of cytosolic basal Ca2+ concentration [1].466 In the present study, we report the possible involvement of 467 P2X and P2Y, ligand-gated purinergic receptors, in the regulation 468 of the Ca2+ homeostasis of chondrogenic cells particularly during 469 UNCORRECTED PROOF Please cite this article in press as: J. Fodor, et al., Ionotropic purinergic receptor P2X4is involved in the regulation of chondrogenesis in chicken micromass cell cultures, Cell Calcium (2009), doi:10.1016/j.ceca.2009.02.004 ARTICLE IN PRESS G Model YCECA10501–10 8J. Fodor et al. / Cell Calcium xxx (2009) xxx–xxx Fig. 6. (A) Metachromatic cartilage areas in 6-day-old HDC visualized with DMMB dissolved in 3% acetic acid. Optical density (OD625) of samples containing toluidine blue extracted from HDC with 8% HCl dissolved in absolute ethanol. Data are mean values of each experimental group out of 4–4 parallel samples of 3 independent measurements. Standard errors of the means were within ±8%. (B) mRNA expression pattern of various P2Y receptors detected by RT-PCR reactions. P2Y receptor subtypes were amplified using specific primers and detected at expected sizes. GAPDH was used as a control. Representative data of 3 independent experiments, each performed in triplicates. (C) Western blot analysis of P2Y1P2Y2and P2Y4receptor proteins in cells of HDC. Total protein and membrane fraction samples were used (50␮g in each lane) to examine the protein expression level. Representative data of 3 independent experiments, each performed in triplicates. their differentiation. Purinoreceptors are known to be expressed in470 embryonic tissues [23] and are probably involved in the differentia471 tion process of excitable [24] and non-excitable [25] cells. However, 472 no data are available concerning the possible involvement of such473 processes in the differentiation of chondroblasts from mesenchy-474 mal cells. During single cell measurements using Fura-2-loaded475 cells, ATP, an agonist of purinergic receptors, was administered on476 various days of culturing. We found that the cells of HDC responded477 to ATP by characteristic Ca2+ transients. We also found that most of478 the cells only responded to ATP on day 3 of culturing, at the time of 479 differentiation of chondroblasts. The phenomenon that some cells480 showed response on other days than day 3 can be explained by481 considering the fact that the cells of HDC exhibit some heterogene-482 ity in their stage of differentiation, and though the vast majority483 differentiates on day 3 of culturing, there are some cells, which 484 could reach this stage of development somewhat earlier or later.485 We also showed that probably the members of the ionotropic P2X 486 receptor subfamily can be accounted for the influx of extracellular487 Ca2+.488 We also tested the effect of P2Y receptor agonists on Ca2+ tran489 sients in cells of 3-day-old HDC. When ATP was administered to the 490 cellsinaTyrode’ssolutionlackingfreeCa2+,ATPfailedtoelevatethe 491 intracellular Ca2+ concentration. ADP, the agonist of P2Y1, and UDP, 492 theligand ofP2Y6receptor, neither evokedanysignificant elevation 493 in the free cytoplasmic Ca2+ concentration, nor did they influence 494 cartilage formation of HDC. However, the agonist of P2Y2and P2Y4495 receptors, UTP caused a transient elevation of cytosolic-free Ca2+ in 496 50% of cells investigated. RT-PCR and Western blot analyses proved 497 the presence of P2Y1, P2Y2, P2Y3, P2Y4and P2Y5receptors in cells 498 of HDC. 499 All the P2X receptor mRNAs investigated, except that of P2X2,500 were expressed by cells of HDC showing variable expression pro501 files. The phenomenon of age-dependent expression suggests the 502 involvement of purinergic signalling in the mediation of chondro503 genic differentiation. Our findings that P2X receptor subtypes are 504 expressed by differentiating chondrogenic cells in a differentia505 tion stage-dependent manner are comparable to studies conducted 506 on hematopoietic cell lines [26]. At the protein level, cells of HDC 507 UNCORRECTED PROOF Please cite this article in press as: J. Fodor, et al., Ionotropic purinergic receptor P2X4is involved in the regulation of chondrogenesis in chicken micromass cell cultures, Cell Calcium (2009), doi:10.1016/j.ceca.2009.02.004 ARTICLE IN PRESS G Model YCECA10501–10 J. Fodor et al. / Cell Calcium xxx (2009) xxx–xxx 9 expressed only receptor subtypes P2X1, P2X4, P2X5and P2X7and508 the expression profiles in total cell lysates and plasma membrane509 fractions were different. This difference was exceptionally inter-510 esting in the case of P2X4: this receptor started to appear on511 day 3 in the plasma membrane fraction with a very strong sig-512 nal, whereas weaker signals were detected on days 4 and 6. The513 characteristic expression profile of P2X4raised the possibility that514 this ligand-gated receptor could be an important channel through515 which extracellular Ca2+ enter the cytosol and contribute to the516 elevatedCa2+ levelneededforthechondrogenicdifferentiationpro-517 cess. Although other P2X receptors were also present in plasma518 membrane fractions, ATP-evoked Ca2+ transients were not elimi-519 nated when suramin and ATP were applied simultaneously. Since520 suramin is not a P2X4antagonist but inhibits other P2X type521 purinergic receptors expressed by cells of HDC, our data suggest522 that these receptors may contribute to the maintenance of the basal523 cytosolic Ca2+ concentration.524 The ATP-mediated function of P2X4receptor during the differ-525 entiation process was proved by the addition of extracellular ATP526 to the culture medium. It resulted in an increase of the expres-527 sion of Sox9, the master transcription factor of chondrogenesis. We 528 detected higher amount of metachromatic cartilage matrix pro-529 duced in ATP-treated HDC by the end of the 6-day-long culturing530 period. This effect was achieved exclusively when ATP was admin-531 istered on day 3 of culturing. When ATP was added prior to (day 3)532 or after the differentiation period of chondrogenic cells (day 4 of533 culturing), it did not result in any significant effect on the matrix534 formation of HDC. This observation underlines the importance of535 ATP in the facilitation of cartilage differentiation and may rule out536 its role in the stimulation of matrix production of mature cartilage.537 Although application of metabotropic P2Y receptor agonists to the 538 culture medium on day 3 caused Ca2+ transients similar to those539 generated by ATP, but did not alter the amount of cartilage matrix540 produced by the end of the 6-day-long culturing period. Therefore541 we suggest that metabotropic P2Y receptors rather contribute to542 the maintenance of basal cytosolic Ca2+ concentration in cells of543 HDC.544 Although suramin did not eliminate the ATP-evoked Ca2+ tran-545 sients of chondrogenic cells, the increased metachromatic cartilage 546 matrix production caused by ATP was completely diminished. This 547 effect does not seem to be exerted via the inhibition of chondrogen-548 esis, since the expression of Sox9, was not reduced by suramin, and549 the mRNA expression levels of neither collagen type II nor aggre-550 can were affected. Suramin has been reported to inhibit hyaluronic551 acid synthesis of fibroblasts [27], and hyaluronic acid is responsi-552 ble for holding aggregates of aggrecan together in cartilage matrix. 553 The reduced amount of hyaluronic acid may cause increased loss of554 aggrecan during metachromatic staining procedures of HDC, caus-555 ing virtual reduction of the detected amount of cartilage matrix.556 We also showed that cells of HDC secreted ATP into the culture557 medium. This ATP may act as an autocrine mediator to facilitate and558 promote their own differentiation. Chondrogenic cells secreted ATP559 throughout the culturing period, and they responded to the extra560 cellularly administered ATP with a peak-like elevation of ic. Ca2+ 561 concentration only at the time of differentiation. This fact further562 supports the purinergic concept in the control of chondrogenesis.563 Our data on ATP concentrations secreted into the culture medium 564 seem significantly less than the concentrations applied for matrix565 production assays. It is important to emphasize that ATP assays566 were performed in 20mL of culture medium. Furthermore, the vol567 ume of the cell culture itself and the volume of the culture medium 568 differbya factorofapproximately1000. Therefore,the ATPsecreted 569 by the cells could have reached much higher concentrations at the570 close proximity of cells. Thus the detected concentrations of ATP571 secreted by chondrogenic cells should be in the range in which P2X 572 receptors respond to this ligand [28,29].573 In summary, our observations provide the first evidence on 574 the possibility of a purinergic autoregulation of chondrogenesis. 575 Purinergic receptors, members of the Ca2+ tool kit used by cells of 576 HDCcan beone of the keyelements in the regulation ofthe elevated 577 cytosolic Ca2+ levels during cartilage differentiation in vitro and in 578 vivo.579 Acknowledgements 580 The authors thank Mrs. Krisztina Bíró and Mrs. Ibolya Varga for 581 their excellent assistance. We also thank Ádám Jóna for his work 582 in performing immunohistochemical stainings of P2X4receptor on 583 paraffin-embedded chicken embryos. This work was supported by 584 grants from the Hungarian Science Research Fund (OTKA K60620 585 and OTKA T49151) and from the Hungarian Ministry of Health (ETT 586 083/2006). R.Z. is supported by a Mecenatura grant (29/ME-02) 587 from the Medical and Health Science Centre, University of Debre588 cen, Hungary. 589 Appendix A. Supplementary data 590 Supplementary data associated with this article can be found, in 591 the online version, at doi:10.1016/j.ceca.2009.02.004.592 References 593 [1] C. Matta, J. Fodor, Z. Szijgyarto, T. Juhasz, P. Gergely, L. Csernoch, R. Zakany, 594 CytosolicfreeCa2+ concentrationexhibitsacharacteristictemporalpatterndur595 ing in vitro cartilage differentiation: a possible regulatory role of calcineurin in 596 Ca-signalling of chondrogenic cells, Cell Calcium 44 (2008) 310–323. 597 [2] M.B. Goldring, K. Tsuchimochi, K. Ijiri, The control of chondrogenesis, J. Cell 598 Biochem. 97 (2006) 33–44. 599 [3] W.M. Kulyk, J.L. Franklin, L.M. Hoffman, Sox9 expression during chondrogene600 sis in micromass cultures of embryonic limb mesenchyme, Exp. Cell Res. 255 601 (2000) 327–332. 602 [4] V. Lefebvre, W. Huang, V.R. Harley, P.N. Goodfellow, C.B. de, SOX9 is a potent 603 activator of the chondrocyte-specific enhancer of the pro alpha1(II) collagen 604 gene, Mol. Cell Biol. 17 (1997) 2336–2346. 605 [5] C.L. Tu, Y. Oda, L. Komuves, D.D. Bikle, The role of the calcium-sensing receptor 606 in epidermal differentiation, Cell Calcium 35 (2004) 265–273. 607 [6] M. Zayzafoon, Calcium/calmodulin signaling controls osteoblast growth and 608 differentiation, J. Cell Biochem. 97 (2006) 56–70. 609 [7] S. Kawano, K. Otsu, S. Shoji, K. Yamagata, M. Hiraoka, Ca(2+) oscillations regu610 lated by Na(+)-Ca(2+) exchanger and plasma membrane Ca(2+) pump induce 611 fluctuationsof membranecurrentsandpotentialsinhumanmesenchymalstem 612 cells, Cell Calcium 34 (2003) 145–156. 613 [8] W.G. Fisher, P.C. Yang, R.K. Medikonduri, M.S. Jafri, NFAT and NFkappaB acti614 vation in T lymphocytes: a model of differential activation of gene expression, 615 Ann. Biomed. Eng. 34 (2006) 1712–1728. 616 [9] S. Chawla, H. Bading, CREB/CBP and SRE-interacting transcriptional regulators 617 are fast on-off switches: duration of calcium transients specifies the magnitude 618 of transcriptional responses, J. Neurochem. 79 (2001) 849–858. 619 [10] S. Piera-Velazquez, D.F. Hawkins, M.K. Whitecavage, D.C. Colter, D.G. Stokes, S.A. 620 Jimenez, Regulation of the human SOX9 promoter by Sp1 and CREB, Exp. Cell 621 Res. 313 (2007) 1069–1079. 622 [11] M. Tomita, M.I. Reinhold, J.D. Molkentin, M.C. Naski, Calcineurin and NFAT4 623 induce chondrogenesis, J. Biol. Chem. 277 (2002) 42214–42218. 624 [12] R. Zakany, E. Bako, S. Felszeghy, K. Hollo, M. Balazs, H. Bardos, P. Gergely, L. 625 Modis, Okadaic acid-induced inhibition of protein phosphatase 2A enhances 626 chondrogenesis in chicken limb bud micromass cell cultures, Anat. Embryol. 627 (Berl.) 203 (2001) 23–34. 628 [13] P.B. Ahrens, M. Solursh, R.S. Reiter, Stage-related capacity for limb chondroge629 nesis in cell culture, Dev. Biol. 60 (1977) 69–82. 630 [14] F.Soto,M. Garcia-Guzman, W.Stuhmer,Cloned ligand-gatedchannelsactivated 631 by extracellular ATP (P2X receptors), J. Membr. Biol. 160 (1997) 91–100. 632 [15] R.A. North, A. Surprenant, Pharmacology of cloned P2X receptors, Annu. Rev. 633 Pharmacol. Toxicol. 40 (2000) 563–580. 634 [16] V. Ralevic, G. Burnstock, Receptors for purines and pyrimidines, Pharmacol. Rev. 635 50 (1998) 413–492. 636 [17] I.R. Orriss, G.E. Knight, S. Ranasinghe, G. Burnstock, T.R. Arnett, Osteoblast 637 responses to nucleotides increase during differentiation, Bone 39 (2006) 638 300–309. 639 [18] C. Collet, C. Strube, L. Csernoch, N. Mallouk, C. Ojeda, B. Allard, V. Jacquemond, 640 Effects of extracellular ATP on freshly isolated mouse skeletal muscle cells dur641 ing pre-natal and post-natal development, Pflugers Arch. 443 (2002) 771–778. 642 [19] M. Ryten, A. Hoebertz, G. Burnstock, Sequential expression of three receptor 643 subtypes for extracellular ATP in developing rat skeletal muscle, Dev. Dyn. 221 644 (2001) 331–341. 645