Involvement of the Purinergic P2X7 Receptor in the Formationof Multinucleated Giant Cells1
Irma Lemaire,2* Simonetta Falzoni,† Natacha Leduc,* Bin Zhang,* Patrizia Pellegatti,†
Elena Adinolfi,† Paola Chiozzi,† and Francesco Di Virgilio†
Multinucleated giant cells (MGC), a hallmark of chronic inflammatory reactions, remain an enigma of cell biology. There isevidence implicating the purinergic P2X7 receptor in the fusion process leading to MGC. To investigate this, we used HEK 293cells stably transfected with either 1) the full-length rat P2X7 receptor (P2X7 cells), 2) a rat P2X7 receptor lacking the C-terminaldomain (P2X7TC), or 3) a mock vector, and rat alveolar macrophages (MA) expressing the native receptor. P2X7 cells culturedin serum-free medium formed increased numbers of MGC and displayed a higher fusion index compared with mock transfectants.Stimulation of P2X7 pore-forming activity in P2X7 cells by polymyxin B (PMB) further increased significantly the formation ofMGC. Conversely, blockers of P2X-receptors including oxidized ATP, brilliant blue G, and pyridoxal phosphate-6-azophenyl-2�-4�-disulfonic acid inhibited significantly MGC formation in both unstimulated and PMB-stimulated P2X7-transfected cells. Incontrast, cells transfected with the truncated P2X7TC were devoid of pore-forming activity, did not respond to PMB stimulation,and failed to form enhanced numbers of MGC, thus behaving as mock transfectants. As found for P2X7-transfected cells, PMBalso potentiated dose-dependently the formation of multinucleated MA by rat alveolar MA. Pretreatment with oxidized ATPabrogated the PMB stimulatory effects. Together, these data demonstrate unequivocally the participation of P2X7 receptor in theprocess of MGC formation. Our study also provides evidence suggesting that stimulation of the P2X7 receptor pathway in MA maymediate increased formation of MGC during chronic inflammatory reactions. The Journal of Immunology, 2006, 177: 7257–7265.
A n intriguing aspect of macrophage (MA)3 response inseveral situations is the formation of multinucleated gi-ant cells (MGC). MGC are a hallmark of chronic gran-
ulomatous reactions and include Langhans’ giant cells seen in in-fective granulomas as well as foreign body giant cells (FBGC)found in foreign body reactions and sterile inflammation (1–3).Osteoclasts, the major bone-resorbing cells, are also multinucle-ated cells that form from the fusion of hemopoietic progenitors ofthe mononuclear phagocyte lineage (4) as well as from MA (5, 6).The osteoclast that can be described as a specialized mono-cyte/MA polykaryon responsible for normal process of bone re-modeling is also a key effector cell in pathologic processes asso-ciated with excessive bone degradation, such as postmenopausalosteoporosis and periprosthetic granulomatous osteolysis (7). Todate, the basic mechanisms underlying MGC formation are not
understood. Previous studies have demonstrated that MGC can begenerated in vitro from both monocytes (8) and MAs (9–11) as aresult of cell fusion (12). A number of cell surface adhesion mol-ecules such as CD11a (LFA-1) (13) and ICAM-1 (14), as well ascytokines including IFN-�, IL-3, IL-4 (15, 16), IL-13 (17), M-CSF, and GM-CSF (10, 11), all have been shown to be involved inMGC formation. In addition, several other membrane moleculeshave been linked to FBGC formation notably, the mannose recep-tor, �2 integrin (16), Src homology 2 domain-containing tyrosinephosphatase substrate-1 (18, 19), CD44, MA fusion receptor-CD47 (18, 20, 21), and dendritic cell-specific transmembrane pro-tein (22).
Quite interestingly, the P2X7 receptor, an ATP-gated ion chan-nel belonging to the family of P2X purinergic receptors (23), hasbeen implicated in the formation of MGC. P2X7 is highly ex-pressed by cells of hemopoietic origin such as MA, and its acti-vation by ATP triggers the formation of a membrane pore thatallows the bidirectional passage of molecules �900 kDa (23). Fol-lowing brief stimulation with ATP, the P2X7 pore-forming activityis reversible upon removal of the stimulus, whereas longer stim-ulation of P2X7 and opening of the pore leads to irreversible celllysis (24). Several lines of evidence implicate the P2X7 receptor inMGC formation. Thus, in vitro experimentation has demonstratedthat Con A-induced MGC in human monocytes was inhibited byoxidized ATP (oATP), an irreversible blocker of P2X7 function(9). Moreover, blockade of the P2X7 receptor by a specific mAbdirected against the outer domain of human P2X7 prevented fusionin vitro (25). Quite interestingly, in murine MA cell lines, fusionoccurred only between MA displaying high expression of the P2X7
receptor and not among MA with low expression (26). Additionalindirect evidence for a role of the P2X7 receptor in the process ofMGC formation have been provided by the observations thatenhanced MGC formation from monocytes of patients with tuber-culosis and sarcoidosis was abolished by oATP (27) and that
*Department of Cellular and Molecular Medicine, Faculty of Medicine, University ofOttawa, Ottawa, Ontario, Canada; and †Department of Experimental and DiagnosticMedicine, Section of General Pathology and Interdisciplinary Center for the Study ofInflammation, University of Ferrara, Ferrara, Italy
Received for publication March 22, 2006. Accepted for publication August 25, 2006.
The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by grants from the Natural Sciences and EngineeringResearch Council of Canada, the Canadian Institutes of Health Research, the ItalianAssociation for Cancer Research, the Italian Ministry of Education and ScientificResearch, the National Research Council of Italy and the Italian Space Agency. I.L.was recipient of a Canadian Institutes of Health Research-National Research Council(Italy) International Scientific Exchange Award.2 Address correspondence and reprint requests to Dr. Irma Lemaire, Departmentof Cellular and Molecular Medicine, Faculty of Medicine, University of Ottawa,451 Smyth Road, Ottawa, Ontario, Canada K1H 8M5. E-mail address: [email protected] Abbreviations used in this paper: MA, macrophage; MGC, multinucleated giant cell;FBGC, foreign body giant cell; oATP, oxidized ATP; PMB, polymyxin B; BBG,brilliant blue G; PPADS, pyridoxal phosphate-6-azophenyl-2�-4�-disulfonic acid;EtBr, ethidium bromide; BAL, bronchoalveolar lavage; f.i., fusion index.
allopurinol and captopril, two agents with therapeutic effect onsarcoidosis, down-regulated the P2X7 receptor and inhibited MGCformation (28). Expression of the P2X7 receptor has also beenobserved on MA and FBGC during the foreign body reaction toimplanted biomaterials (29). The P2X7 receptor was found to beexpressed on osteoclasts and their precursors in vitro and on hu-man osteoclasts lining bone surface (30, 31). In addition, fusion ofosteoclast precursors to form multinucleated osteoclasts in vitrohas been shown to be inhibited by oATP and a P2X7 receptor mAb(32) and, more recently, by down-regulation of P2X7 expression(33). However, the evidence linking P2X7 to MGC formation hasbeen indirect or circumstantial. In addition, recent work has shownthat mice lacking the P2X7 receptor were able to form multinu-cleated osteoclasts in vitro and in vivo (34), thus bringing intoquestion the role of P2X7 in osteoclasts.
Despite numerous observations that FBGC, Langhans’ giantcells, and osteoclasts display distinct cell morphology, and thattheir respective formation requires a different set of cytokines andreceptors, it is noteworthy that the P2X7 receptor has been impli-cated in the formation of all three types of MGC. It is quite pos-sible that P2X7 is a common molecular step crucial for MGC for-mation during various types of granulomatous reactions. In thisstudy, we took advantage of a heterologous cell system expressingeither the rat P2X7 receptor or a truncated defective receptor, anddemonstrate unequivocally the participation of P2X7 receptor ac-tivity in the cell-to-cell communication process leading to MGCformation. We also provide evidence that enhancement of theP2X7 function in rat alveolar MA by polymyxin B (PMB) stimu-lates cell fusion into MGC.
Materials and MethodsAnimals and reagents
Lung pathogen-free male Wistar rats weighing 225–250 g were purchasedfrom Harlan World. These animals were shipped behind filter barriers andhoused in isolated temperature-controlled quarters under pathogen-freeconditions in an animal isolator unit (Johns Scientific). They were givenstandard laboratory chow and water ad libitum and were used within 2 wk.Approval was obtained from the Animal Care and Use Committee of theUniversity of Ottawa for all procedures. IMDM was purchased from In-vitrogen Life Technologies. Lab-Tek culture chambers were obtained fromNalge Nunc International. PMB, ATP, oATP, poly-L-lysine, brilliant blueG (BBG), and pyridoxal phosphate-6-azophenyl-2�-4�-disulfonic acid(PPADS) were purchased from Sigma-Aldrich. Ethidium bromide (EtBr)was obtained from Molecular Probes. Plasmid DNA of full-length (595 aa)rat P2X7 receptor (P2X7R) and rat P2X7R truncated at the intracellularC-terminal aa 415 were a gift from Dr. G. Buell (Ares-Serono ResearchLaboratories, Geneva, Switzerland). For full-length rat P2X7 receptor, thecDNA was inserted in pcDNA3 vector (ampR) on an EcoRI-EcoR1 frag-ment. For truncated rat P2X7, the cDNA was inserted in pcDNA3 betweenHindIII and EcoR1. G418 sulfate (Geniticin) was provided by Nova Bio-chem. GFP-tagged rat P2X7R and GFP control plasmid were gifts from A.Surprenant (University of Sheffield, U.K.) and R. Rizzuto (University ofFerrara, Italy), respectively.
HEK 293 stably transfected cells and solutions
HEK 293 cells were cultured in DME/Ham’s F-12 medium (F-12)(1:1)(Invitrogen Life Technologies) containing 15% heat-inactivated FCS (Hy-Clone), 100 �/ml penicillin, and 100 �g/ml streptomycin. HEK 293 wild-type cells were transfected as described previously (35) with calcium phos-phate. Briefly, 2.5 � 106 HEK 293 cells were plated in petri dishes andincubated overnight. The next day, plasmid DNA (30 �g) was resuspendedin 450 �l of TE buffer (10 mM Tris, 1 mM EDTA; pH 8), and 50 �l ofCaCl2 (2.5 M) was added dropwise under vortexing to a tube containing500 �l of 2� HBS buffer (280 mM NaCl, 50 mM HEPES, 1 mM Na2
HPO4 (pH 7.12), at 25oC). After a 30-min incubation at room temperature,the cDNA precipitate was added to the dish dropwise and gently swirled.The culture medium was changed on the third day, and on the fourth dayG418 (0.8 mg/ml) was added to selected transfected clones. Followingclone selection, the selective medium contained G418 (0.2 mg/ml) and
stable clones were maintained in culture medium containing G418 at thesame concentration.
Isolation of rat alveolar MA
MA were recovered from normal rats by bronchoalveolar lavage (BAL) asdescribed previously (10). Briefly, the lungs were lavaged with seven 7-mlaliquots of sterile PBS (pH 7.4) (Wisent), and BAL cells were obtained bycentrifugation at 200 � g at 4�C for 5 min. The cells were resuspended inIMDM supplemented with 10% dialysed FBS (Wisent), 0.005% gentami-cin (Schering), and 0.8% HEPES, which will henceforth be referred to asMA complete culture medium (MACM). Cells were counted in a hema-cytometer chamber, and viability (98–100%) was determined by trypanblue exclusion. Differential analysis of lavage cells made by cytocentrifugesmears (Thermo Shandon; 2.5 � 104 cells) and stained with Wright-Giemsa indicated that the BAL cell population was essentially composedof MA (99%) in normal rats.
Culture and generation of MGC
The generation of MGC from rat alveolar MA was performed using ourpreviously described microculture cell system (10, 11). Briefly, 2 � 105
MA were plated in 8-well Lab-Tek tissue culture chambers in 200 �l ofMACM and cultured overnight. The cells were then incubated in the pres-ence or absence of PMB at various concentrations as indicated, and forvarious times (3–5 days) at 37�C in 5% CO2. For the generation of MGCin HEK 293 cell cultures, HEK 293 cells transfected with either the ratP2X7, the rat truncated P2X7 (P2X7TC), or a mock vector, were plated(2 � 105) in 2-well Lab-Tek tissue culture chambers coated with poly-L-lysine in 2 ml of DME/F-12(1:1) containing 10% heat-inactivated FCS.Following overnight incubation, the cells were washed with prewarmedPBS, and cells were incubated in 2 ml of serum-free medium in the pres-ence and absence of PMB at the indicated concentration for various times(24–72 h) at 37oC in 5% CO2. Following incubation of MA and HEK cells,the medium was removed and cells were washed three times with PBS.After drying, the chambers were disassembled, stained with H&E, andmounted with coverslips and Permount (Fisher Scientific) for microscopicevaluation. Ten fields at magnification �330 (or when indicated 20 fieldsat magnification �165) were monitored for the presence of MGC withthree or more nuclei. Images were captured and analyzed with the NorthernEclipse image analysis software. Stained cells were counted with the helpof a software (Man Count) linked to Excel in separate predetermined color-coded classes (1, 2, 3, 4 or more nuclei). From these data, the total numberof cells with �3 nuclei (MGC) and the percentage of MGC were obtained.The fusion index (f.i.) was calculated as follows: f.i. � total number ofnuclei in MGC/total number of nuclei counted � 100.
Expression and localization of rat P2X7 receptor
Expression of the P2X7 receptor and its localization were assessed withHEK 293 cells stably expressing the rat P2X7 coupled to GFP (rat P2X7-GFP). The cells were seeded on 24-mm coverslips in standard saline so-lution (described in the text) and kept at 37�C. Images of cells transfectedwith rat P2X7-GFP were acquired with a Nikon Eclipse TE 300 invertedmicroscope (Nikon) equipped with a thermostated chamber (Bioptechs), a63� immersion objective, and the following filter set: excitation HQ480/40, dichroic Q480LP, and emission HQ510LP. For high-speed acquisitionand processing of fluorescent images, the microscope was equipped withthe following devices: a computer-controlled light shutter, a six-positionfilter wheel, a piezoelectric z-axis focus device, a back illuminated 1000 �800 charge-coupled device camera (Princeton Instruments), and a com-puter with MetaMorph software (Universal Imaging) for image acquisition,two- and three-dimensional visualization and analysis.
Pore-forming activity
P2X7-dependent pore formation and increases in plasma membrane per-meability were measured by monitoring over time the uptake of the fluo-rescent dye EtBr as described previously (36). Briefly, cells (5 � 105/ml)were incubated in a standard saline solution containing 125 mM NaCl, 5mM KCl, 1 mM Mg SO4, 1 mM Na2 HPO4, 5.5 mM glucose, 5 mMNaHCO3, 1 mM CaCl2, and 20 mM HEPES (pH 7.6) in a thermostated(37oC) fluorometric cuvette under constant magnetic stirring with EtBr (20�M). Cells were incubated for 5 min in the presence or absence of PMB(10 �g/ml) and then stimulated with ATP (1 mM). Uptake of the fluores-cent dye was monitored for 25 min. At that time, digitonin (100 �M) wasadded to achieve complete permeabilization of cells (100% fluorescencesignal). Fluorescence was measured at 360 nm/580 nm excitation/emissionwavelengths.
Pore-forming activity was also assessed by determining the percentageof cells that became permeabilized to EtBr, as described previously (37).For this process, HEK 293 cells (4 � 104) were incubated in 200 �l ofcomplete medium in 96-well tissue culture plates for 24 h. The mediumwas then replaced with standard saline solution. EtBr (20 �M) was added,and cells were incubated at 37oC for 5 min in the presence and absence ofPMB. Cells were then stimulated with ATP (0.5 mM) for 5 min at 37oC.After centrifugation (200 � g, 1 min), culture supernatants were replacedwith fresh saline solution, and fluorescence was analyzed using a ZeissAxiovert S1100TV inverted microscope equipped with a rhodamine filterand a 32� objective. Images were captured and analyzed with the NorthernEclipse image analysis software. Data are expressed as percentage of cellsthat become permeabilized to EtBr.
Pharmacological modulation of MGC formation
Cells were cultured in Lab-Tek chambers as described above. For assess-ment of P2X7 antagonists, cells were pretreated for 2 h at 37oC with oATP(100 �M or 500 �M, as indicated), washed in serum-free medium, andcultured as described above for 48–72 h in the presence and absence ofPMB at various concentrations. In some experiments using HEK 293 cells,BBG (5 � 10�9 M) or PPADS (5 � 10�7 M) were added 5 min beforePMB and cells cultured for 48 h.
Statistical analysis
Results are expressed as mean � SEM. Statistical significance was deter-mined using a one-way ANOVA and Bonferroni test ( p � 0.05; Instat;GraphPad).
ResultsEnhanced MGC formation in HEK 293 cells transfected with therat P2X7 receptor
HEK 293 cells transfected with the rat P2X7 receptor and culturedin serum-free medium for intervals up to 72 h undergo fusion intoMGC. The presence of MGC with three nuclei was already appar-ent at 24 h and for up to 72 h. At 48 h, which was the optimal timepoint chosen for most experiments, cell fusion was observed be-
tween MGC and single cells as well as between MGC themselves(Fig. 1, upper left and right panels), indicating that once formed,MGC continue to express at least initially the molecular effector(s)responsible for the fusion process. This is consistent with the no-tion that MGC are involved in further recruitment and fusion ofadditional cells leading to large MGC with numerous nuclei asseen during granulomatous inflammatory reactions. Fluorescencemicroscopy using HEK 293 cells transfected with a rat P2X7
receptor fused with GFP at the C-terminal, further demonstratedthe localization of GFP-tagged P2X7 receptor to the cell membrane(Fig. 1, lower right panel). The levels of surface vs intracellularexpression can be appreciated from the intensity of GFP-P2X7
fluorescence as depicted on a rainbow scale going from blue (lowlevel) to red (high level) (Fig. 1, lower left panel). In contrast,HEK 293 cells transfected with a control GFP-vector alone (Fig. 1,inset, lower left panel) expressed only cytoplasmic GFP withoutmembrane localization.
Although the presence of MGC was found in cultures of mock-transfected cells, cells expressing the P2X7 receptor had a higherf.i. and formed an increased number of MGC (Fig. 2, A and B).Detailed analysis of the number of nuclei contained within theMGC population of P2X7 cells and mock cells further demon-strated that after 48 h incubation, P2X7-expressing cells display1.3- and 2.1-fold increase in MGC harboring three and four nuclei,respectively (Fig. 2C).
PMB potentiates both P2X7 pore opening and MGC formationin P2X7-transfected cells
The natural cationic peptide, PMB, has recently been shown topotentiate ATP-induced membrane permeabilization and pore for-mation, a hallmark of P2X7 activation (36). We next investigated
FIGURE 1. Phenotypic characteristics of P2X7-de-rived MGC. HEK 293 cells transfected with the ratP2X7 receptor were cultured in Lab-Tek chambers for48 h as described in Materials and Methods. Cells ex-pressing P2X7 undergo fusion into MGC with randomlydispersed nuclei (up to 10) (upper left panel). Somecells display numerous nuclei arranged in a ring-likefashion (upper right panel). Note the fusion betweenMGC and single cells as well as between MGC them-selves. Fluorescence microscopy demonstrates the lo-calization of GFP-tagged P2X7 receptor at the cellmembrane of adjacent cells (lower right panel) in con-trast to cells transfected with a control GFP-vector alone(inset); the intensity of GFP-P2X7 fluorescence isshown on a rainbow scale going from blue (low level) tored (high level) (lower left panel).
whether activation/promotion of P2X7 would result in further en-hancement of MGC formation. To this aim, HEK 293- P2X7 cellsand HEK 293-mock cells were incubated in the presence of theimpermeant fluorescent dye EtBr (20 �M) and then stimulatedwith ATP (1 mM) in the presence and absence of PMB. Preincu-bation with PMB (10 �g/ml) increased significantly the permeabi-lizing activity of ATP in P2X7-transfected cells as measured bymonitoring the rate of EtBr uptake over a 25-min interval (Fig.3A). In the presence of ATP, the fluorescent dye could be seen inthe nuclei and diffusely throughout the cytoplasm of treated cells
upon microscopic observation. Up to 59.9% of P2X7 cells becamepermeabilized to ATP, and such percentage increased to 79.4%when cells were preincubated for 5 min with PMB (Fig. 3B). Under
FIGURE 2. HEK 293 P2X7-transfected cells form a greater number ofMGC and have a higher f.i. than mock transfectants. HEK 293 cells trans-fected with rat P2X7 receptor or with a mock vector were incubated inLab-Tek chambers for 48 h under similar conditions. A, Total number ofMGC expressed as a percentage of total cells counted in 10 random fields;magnification, �330. B, The f.i. was calculated as described in Materialsand Methods. Data represent mean � SEM of seven experiments. �, Sig-nificantly different from control at p � 0.05. C, The percentage of MGCdisplaying 3, 4, 5, or 6 nuclei was evaluated in mock and P2X7-transfectedcells that display a greater number of cells harboring three or four nuclei.Data represent mean � SEM of at least seven experiments. �, Significantlydifferent from control at p � 0.05.
FIGURE 3. PMB potentiates P2X7 pore-forming activity in P2X7-trans-fected cells. HEK 293-P2X7 and HEK 293-mock cells were incubated in thepresence of 20 �M EtBr and then stimulated with ATP in the presence orabsence of PMB. A, P2X7-transfected cells (5 � 105/ml) were incubated at37oC under constant stirring, and PMB (10 �g/ml) was added 5 min beforeATP (1 mM). Uptake of fluorescence was monitored over time as described inMaterials and Methods. Digitonin (100 �M) was added during the last 5 minto obtain the maximal fluorescence signal. Trace C, without PMB; trace PMB,cells were treated with PMB 5 min before addition of ATP. Data are shown asrate of fluorescence increase in arbitrary units. B, P2X7- and mock-transfectedcells (1 � 105) were incubated in tissue culture plates as described in Materialsand Methods. The medium was changed for standard saline, and cells wereincubated at 37oC as follows: 1) without ATP (C), 2) in the presence of ATP(0.5 mM) for 5 min, 3) with PMB (10 �g/ml) , or 4) with PMB and ATP. Inthis case, PMB was added 5 min before ATP. Fluorescence of cells perme-abilized by ATP was analyzed by microscopy and capture image analysis, anddata are expressed as percentage of cells that become permeabilized to EtBr.Inset, Following incubation with PMB (10 and 50 �g/ml) and ATP, results arealso expressed as percentage of control (P2X7-transfected cells stimulated withATP alone). Data are mean � SEM of four separate experiments. �, Signifi-cantly different from control at p � 0.05. ��, Significantly different from ATP-stimulated cells at p � 0.05.
these conditions, PMB alone had a small effect with 8.6% of cellstaking up EtBr compared with 3.3% for unstimulated cells. Over-all, treatment of P2X7 cells with PMB (10 and 50 �g/ml)augmented the ATP-permeabilizing effect by 140 and 202% (or1.4- and 2-fold), respectively (Fig. 3B, inset). In contrast, HEK293-mock cells did not become permeabilized to EtBr when incu-bated under the same conditions with either ATP, PMB, or a com-bination of both (Fig. 3B).
In parallel experiments, HEK 293-P2X7 cells and HEK 293-mock cells were cultured in serum-free conditions in the presenceand absence of PMB for 48 h and analyzed for MGC formation.Addition of PMB (10 and 50 �g/ml) at the beginning of the cellculture period augmented significantly the total number of MGC inP2X7 cells (14 and 21%, respectively, compared with 10% forunstimulated cells) (Fig. 4A) and the f.i. (34 and 42, respectively,compared with 26 for unstimulated cells) (Fig. 4B). The overallincrease in MGC was 137 and 203% that of control in culturestreated with 10 �g/ml and 50 �g/ml PMB, respectively (Fig. 4A,inset). Enhancement of MGC formation was due mainly to in-creased numbers of MGC harboring three or four nuclei (14 and5%, respectively, compared with 7 and 2% in controls) and up tofive nuclei (Fig. 4C). As observed for pore-forming activity, PMBdid not increase MGC formation in mock transfectants (Fig. 4,A and B).
Enhanced MGC formation in P2X7-transfected cells is inhibitedby P2X7 blockers
Antagonists at the P2X7 receptor were tested for their ability tointerfere with MGC formation in HEK-P2X7 cells. Preincubationwith oATP (500 �M), an irreversible blocker of P2X7 function(38), inhibited MGC formation both in unstimulated cells (77% ofcontrol) and in cells stimulated with PMB (53–57% of control)(Fig. 5A). Similarly, BBG and PPADS, two antagonists of the ratP2X7 receptor known to block P2X7 pore-forming activity (37, 39,40), both decreased the number of MGC and the f.i. in unstimu-lated (71–78 and 72–75% of control, respectively) and PMB-stim-ulated P2X7 cells (54–63 and 60–68% of control, respectively). Incontrast, none of the antagonists inhibited the formation of MGCseen in mock transfectants (Fig. 5B).
The P2X7 receptor C-terminal domain is required for MGCformation
Although the cytosolic C-terminal region of the P2X7 receptor inHEK 293-P2X7 cells has been shown to interact with several pro-teins (41) and is required for pore-forming activity (23), the site ofaction of PMB appears to reside within the extracellular domain(36). Therefore, it is quite possible that the bulky extracellulardomain of P2X7 might be involved in intercellular communicationand MGC formation. To verify this, we studied pore-forming ac-tivity and MGC formation in HEK 293 cells transfected with a ratP2X7 receptor lacking the intracellular C-terminal 180 aa(P2X7TC). We have shown previously that these cells express therat P2X7TC on their plasma membrane as evidenced by their abil-ity to display well-established responses to ATP, including signif-icant [Ca2�]i increase and membrane depolarization (36) that canbe potentiated by PMB (36). As expected, HEK 293-P2X7TC cellslacked pore-forming activity in response to ATP even after prein-cubation with PMB, as determined by monitoring the rate of EtBr
FIGURE 4. PMB potentiates the formation of MGC in P2X7-trans-fected cells. HEK 293-P2X7 and HEK 293-mock cells were incubated inLab-Tek chambers for 48 h, as described in Materials and Methods, in thepresence and absence of PMB (10 �g/ml and 50 �g/ml). PMB was addedat the beginning of the culture. A and B, Enhanced MGC formation wasobserved in P2X7 compared with mock cells. Incubation with PMB furtherincreased the number of MGC in P2X7 cells but not in mock cells asdetermined by the percentage of MGC (A) and the f.i. (B). A, Inset, Dataare also expressed as percentage (%) of control (P2X7-transfected cellsincubated without PMB). Data represent mean � SEM of at least sevenexperiments. Significantly different from control at �, p � 0.05 and ��, p �
0.01. C, Detailed analysis of the number of nuclei present in unstimulated andPMB (10 �g/ml, P10; 50 �g/ml, P50) stimulated P2X7-transfected cells. Datarepresent the mean � SEM of at least four experiments. Significantly differentfrom control at �, p � 0.05, ��, p � 0.02, and ���, p � 0.01.
uptake by cells (Fig. 6A) and the number of fluorescent perme-abilized cells (Fig. 6B). Similarly, cells transfected with the trun-cated receptor did not exhibit higher MGC number and f.i. andfailed to form increased numbers of MGC in response to PMB (10and 50 �g/ml), thus behaving as HEK 293-mock cells (Fig. 6C). Incontrast, parallel cultures of P2X7 cells displayed higher MGCnumbers and f.i. that were further increased following incubationwith PMB.
PMB potentiates MGC formation in rat alveolar MA cultures
Pure populations of rat resident alveolar MA express a functionalP2X7 receptor (37) and represent a useful cell culture system forthe generation of MGC following appropriate stimulation (10, 11).We hypothesized that PMB might also modulate positively P2X7
function in rat alveolar MA expressing the native receptor andpromote MA multinucleation and MGC formation. To test thistheory, MA were incubated with and without PMB at various con-centrations in the presence and absence of oATP for 72 h, and theirf.i. was monitored. Under our culture conditions, unstimulated MAundergo cell fusion and form multinucleated MA at low levels.Addition of PMB at the beginning of cell culture increased in adose-dependent manner the formation of MGC with a maximal f.i.of 3.2 obtained at 5 �g/ml PMB compared with 1.4 for unstimu-
lated cells (up to 2.3-fold stimulation) (Fig. 7A). Pretreatment withoATP blocked the stimulatory effects of PMB at all doses tested.Promotion of MGC formation by PMB was maximal at the 3-daytime point and decreased thereafter (Fig. 7B). Detailed analysis of
FIGURE 5. Inhibition of MGC formation in HEK 293 P2X7-transfectedcells by antagonists at the P2X7 receptor. Unstimulated as well as PMBstimulated (10 �g/ml or 50 �g/ml) P2X7 (A) and mock (B) transfected cellswere either pretreated with oATP (500 �M) or incubated in the presenceand absence of BBG (5 � 10�9 M) or PPADS (5 � 10�7 M) for 48 h inLab-Tek chambers as described in Materials and Methods. Both the per-centage of MGC and the f.i. were determined, and data are expressed aspercentage of control that represents the maximal MGC and f.i. response(100%) in unstimulated cells, or cells stimulated with 10 or 50 �g/ml in theabsence of P2X7 blockers. Data represent the mean � SEM of three ex-periments. �, Significantly different from control at p � 0.01.
FIGURE 6. HEK 293 cells transfected with a truncated P2X7 recep-tor (P2X7TC) lack pore-forming activity and do not exhibit enhancedMGC formation. HEK 293 cells transfected with the full-length ratP2X7R or a P2X7 lacking the C-terminal domain (P2X7TC) were ana-lyzed for their respective ability to form pore (A and B) and MGC (C).A, P2X7 (trace c) and P2X7 TC (trace a) transfected cells (5 � 105/ml)were incubated in the presence of EtBr (20 �M) and then stimulatedwith ATP (1 mM). P2X7TC cells were also stimulated with PMB (10�g/ml) (trace b). Fluorescence increase was measured as described inMaterials and Methods. Data are shown as rate of fluorescence increasein arbitrary units. B, P2X7 and P2X7TC (4 � 104) were incubated instandard saline with EtBr (20 �M) in the presence and absence of ATP(0.5 mM) for 5 min, and the percentage of cells that become perme-abilized by ATP was determined as described in Materials and Meth-ods. Data are means � SEM of three separate experiments. C, P2X7 andP2X7 TC were incubated in Lab-Tek chambers for 48 h, as described inMaterials and Methods, in the presence and absence of PMB (10 and 50�g/ml). MGC formation was determined by evaluating the percentageof MGC and the f.i. Data represent means � SEM of at least six ex-periments. �, Significantly different from control at p � 0.01.
MGC formed in response to PMB showed an increase in the num-ber of multinucleated MA with three or four nuclei that wasblocked by oATP (Fig. 7C). Treatment with oATP also decreasedsignificantly MGC formation by unstimulated MA.
DiscussionIn this study, we used an heterologous cell system expressing theP2X7 receptor to investigate its role in the process of cell-to-cellcommunication leading to MGC formation. This experimental ap-
proach permitted us to single out the participation of the P2X7
receptor in this process from other P2X receptors, cytokines, andmembrane molecules present in cells expressing the native recep-tor. We present evidence that increased numbers of MGC can begenerated from P2X7-transfected cells, indicating that P2X7 recep-tor expression at the cell membrane can provide sufficient stimulusfor induction of such intercellular communication pathways. Ourresults support earlier work that showed a correlation betweenP2X7 expression and giant cell formation (9, 26) and blockade ofthis process by an inactivating P2X7 mAb (25). However, expres-sion of the P2X7 receptor is not a requisite per se for the processof multinucleation, because mock-transfected cells that do not ex-press P2X7 undergo fusion into MGC, albeit to a lower level. It islikely that other plasma membrane molecules expressed by HEK293 cells are involved in triggering cell multinucleation. Our re-sults are consistent with previous findings that mice lacking theP2X7 receptor (P2X7
�/�) retain the capacity to form multinucle-ated osteoclasts (34).
Interestingly, expression of the P2X7 receptor is associated witha greater ability to form MGC, as evidenced by increased numbersof MGC and higher f.i. in P2X7-transfected cells compared withmock cells. Furthermore, we demonstrate that the natural cationicpeptide PMB augments P2X7 pore-forming activity and enhancesMGC formation selectively in P2X7-transfected cells but not inmock transfectants. These results confirm previous reports thatPMB modulate selectively P2X7 receptor function (36) and presentthe first evidence that PMB can enhance P2X7-dependent MGCformation. Under our culture conditions, PMB (10 �g/ml), whichis not toxic (36), caused by itself a small but significant increase inthe number of P2X7 cells permeabilized to EtBr but had no effecton mock cells. Whether this reflects a direct stimulatory effect ofPMB on P2X7 or is related to our experimental culture conditionsis not known and would require further investigation. A directeffect of PMB, even small, on P2X7 cells would be consistent withour observations of enhanced MGC formation in these cells in theabsence of exogenous stimulation with ATP. It should be pointedout, however, that the most striking effect of PMB on P2X7 pore-forming activity is to potentiate the effect of ATP at the P2X7
receptor and increase its sensitivity to lower ATP concentration(36). This could be the case during MGC formation becauseP2X7-transfected cells release in the culture medium ATP atlevels sufficient to promote activation of P2X7 in the presenceof PMB (42, 43).
Although the process of MGC formation requires 24 to 48 h, itis likely that PMB action is rapid because the peptide would berapidly metabolized under normal culture conditions. This impliesthat the PMB effect could be related to brief activation of P2X7. Insupport of this, potentiation of ATP-induced cell permeabilizationwas not observed in P2X7-transfected cells incubated overnightwith PMB (our unpublished results). MGC formation is a processthat requires concerted cytoskeletal reorganization and regulatedplasma membrane disruption. It is tempting to speculate that MGCformation may be associated with the reversible pseudoapoptoticsignaling pathway induced by brief activation of P2X7 receptors(44). In support of this, experimental evidence has demonstratedthat intercellular fusion leading to multinucleation may occurthrough mechanisms that are independent of apoptosis (45, 46). Incontrast, overexpression of P2X7 has been associated with celldeath (24), and PMB has been shown to augment P2X7-dependentcytotoxicity (36). Therefore, the possibility remains that MGC for-mation results from the fusion of cells programmed for death. Fur-ther experimentation is currently underway to address these issues.
FIGURE 7. PMB increases MGC formation in rat alveolar MA. A, MA(2 � 105) cultured in Lab-Tek chambers were pretreated or not with oATP(100 �M) for 2 h. The medium was then changed, and cells were incubatedin the presence and absence of PMB at various concentrations (0–10 �g/ml) for 72 h. The f.i. was determined as described in Materials and Meth-ods. Data represent means � SEM of three experiments. �, Significantlydifferent from control at p � 0.05. B, MA were incubated in the presenceand absence of PMB (10 �g/ml) for 3, 4, and 5 days. Incubation with PMBincreased the number of multinucleated MA as determined by the percent-age of MGC (%) and the f.i. Data represent means � SEM of three ex-periments. �, Significantly different from control at p � 0.01. C, The per-centage of multinucleated MA with three or four nuclei was evaluated inunstimulated and PMB (10 �g/ml) stimulated MA pretreated or not withoATP, as described above. Data represent means � SEM of three exper-iments. �, Significantly different from control following incubation withPMB at p � 0.05. ��, Significantly different from control following pre-treatment with oATP at p � 0.05.
One question that has been left unanswered with regard to theP2X7 receptor is whether a functional receptor is needed for mem-brane fusion leading to MGC. It has been proposed that in partnercells expressing P2X7, activation of the receptor could generate a“fusion pore” that would bridge the cytoplasm of the adjacent cellsand drive the eventual fusion (25). In this study, we demonstratethat cells transfected with a P2X7 receptor truncated at theC-terminal domain and which fail to induce pore formation inresponse to ATP, also lose the ability to form increased numbers ofMGC, even in the presence of PMB thought to act at the extra-cellular domain (36). Our data provide direct evidence that MGCformation requires an integral C-terminal domain and a functionalreceptor. In accordance with this result, previous work has shownthat the formation of multinucleated osteoclasts was lost in cellsthat were resistant to ATP but was recovered in cells that regainedP2X7 pore-forming activity (33), suggesting that P2X7 might beinvolved in the mechanics of cell fusion or, alternatively, in asignaling pathway proximal to this event. However, it is difficult tolink MGC formation to the pore-forming activity in our study be-cause the cytosolic C-terminal region of the P2X7 receptor hasbeen shown to be important for many P2X7-related functions be-sides pore formation including membrane blebbing, and apoptosis.Little is known about how the functional domains of the C-termi-nal region contribute to the numerous properties ascribed to P2X7.As mentioned earlier, the process of multinucleation requires cy-toskeletal rearrangements and regulated membrane disruption, andthe C-terminal tail of P2X7 interacts with several proteins thatcould potentially drive such a process including actin, actinin, and�2 integrin (41).
It is noteworthy that the observations made with an heterologouscell system expressing no other P2X receptors than P2X7, can betranslated in rat alveolar MA expressing the native receptor. Ratalveolar MA have the potential to form MGC in vivo (47) and invitro (10, 11, 18) and exhibit a functional P2X7 receptor that, uponstimulation with ATP, mediates pore formation and activates theproinflammatory IL-13 IL-6 cytokine cascade (37). We providefurther evidence that activation of P2X7 by PMB in these cellsincreased their ability to form MGC. As discussed for P2X7-trans-fected cells, extracellular ATP released by alveolar MA under ourculture conditions, even at low levels, may act in concert withPMB to promote P2X7 receptor activation and MGC formation.In accordance with earlier findings (36), our data point to PMBas a selective tool to increase P2X7-mediated function in ratalveolar MA.
Altogether, our observations provide direct evidence that ex-pression and activation of the P2X7 receptor is associated withincreased cell-to-cell fusion and enhanced formation of MGC.Thus, P2X7-mediated MGC may represent a separate, amplifica-tion pathway aimed at augmenting the process of MA multinucle-ation during host response to danger signals and chronic granulo-matous reactions. In support of such an assumption, activation ofP2X7 in MA has been linked to several responses relevant to in-flammation including NF-�B, cytokine and NO release, superox-ide production, and cytotoxicity (48). It is likely that intercellularcommunication is crucial in coordinating MA responses duringinflammatory reactions, and our results point to the P2X7 receptoras an efficient system for the promotion and maintenance of inter-cellular communication during such reactions. This is directly rel-evant to recent views that chronic inflammation is the product ofpathologic cell-cell interactions, occurring within a unique and de-fined biological compartment such as the lung, gut, and synovium.Understanding the mechanisms underlying the participation of theP2X7 receptor in cell-to-cell communication, may provide a basis
for the development of strategies aimed at interfering with patho-logic MA-MA interactions in diseases.
DisclosuresThe authors have no financial conflict of interest.
References1. Adams, D. O. 1976. The granulomatous inflammatory response: a review.
Am. J. Pathol. 84: 164–191.2. Chambers, T. J. 1978. Multinucleated giant cells. J. Pathol. 126: 125–148.3. Anderson, J. M. 2000. Multinucleated giant cells. Curr. Opin. Hematol. 7:
40–47.4. Burger, E. H., J. W. Van der Meer, J. S. Van de Gevel, J. C. Gribnau,
G. W. Thesingh, and R. Van Furth. 1982. In vitro formation of osteoclasts fromlong term cultures of bone marrow mononuclear phagocytes. J. Exp. Med. 156:1604–1614.
5. Udagawa, N., N. Takahashi, T. Akatsu, H. Tanaka, T. Sasaki, T. Nishihara,T. Koga, T. J. Martin, and T. Suda. 1990. Origin of osteoclasts: mature mono-cytes and macrophages are capable of differentiating into osteoclasts under suit-able microenvironment prepared by bone marrow-derived stromal cells. Proc.Natl. Acad. Sci. USA 87: 7260–7264.
6. Itonaga, I., A. Sabokbar, D. W. Murray, and N. A. Athanasou. 2000. Effect ofosteoprotegerin and osteoprotegerin ligand on osteoclast formation by arthro-plasty membrane derived macrophages. Ann. Rheum. Dis. 59: 26–31.
7. Ingham, E., and J. Fisher. 2005. The role of macrophages in osteolysis of the totaljoint replacement. Biomaterials 26: 1217–1286.
8. Most, J., L. Spotl, G. Mayr, A. Gasser, A. Sarti, and M. P. Dierich. 1997. For-mation of multinucleated giant cells in vitro is dependent on the stage of mono-cyte to macrophage maturation. Blood 89: 662–671.
9. Falzoni, S., M. Munerati, D. Ferrari, S. Spisani, S. Moretti, and F. Di Virgilio.1995. The purinergic P2z receptor of human macrophage cells: characterizationand possible physiological role. J. Clin. Invest. 95: 1207–1216.
10. Lemaire, I., H. Yang, W. Lauzon, and N. Gendron. 1996. M-CSF and GM-CSFpromote alveolar macrophage differentiation into multinucleated giant cells withdistinct phenotypes. J. Leukocyte Biol. 60: 509–518.
11. Lemaire, I., H. Yang, V. Lafond, J. Dornand, T. Commes, and M. F. Cantin.1996. Differential effects of macrophage- and granulocyte-macrophage colony-stimulating factors on cytokine gene expression during rat alveolar macrophagedifferentiation into multinucleated giant cells (MGC). J. Immunol. 157:5118–5125.
12. Murch, A. R., M. D. Grounds, C. A. Marshall, and J. M. Papadimitriou. 1982.Direct evidence that inflammatory multinucleated giant cells form by fusion.J. Pathol. 137: 177–180.
13. Most, J., H. P. Neumayer, and M. P. Dierich. 1990. Cytokine-induced generationof multinucleated giant cells in vitro requires interferon-� and expression ofLFA-1. Eur. J. Immunol. 20: 1661–1667.
14. Fais, S., V. L. Burgio, M. Silvesti, M. R. Capobianchi, A. Pacchiarotti, andF. Pallone. 1994. Multinucleated giant cells generation induced by IFN-�:changes in the expression and distribution of the intercellular adhesion mole-cule-1 during macrophages fusion and multinucleated giant cell formation. Lab.Invest. 71: 737–744.
15. McInnes, A., and D. M. Rennick. 1988. Interleukin 4 induces cultured mono-cytes/macrophages to form giant multinucleated cells. J. Exp. Med. 167:598–611.
16. McNally, A. K., and J. M. Anderson. 2002. �1 and �2 integrins mediate adhesionduring macrophage fusion and multinucleated foreign body giant cell formation.Am. J. Pathol. 160: 621–630.
17. Byrd, T. F. 1998. Multinucleated giant cell formation induced by IFN-�/IL-3 isassociated with restriction of virulent Mycobacterium tuberculosis cell to cellinvasion in human monocyte monolayers. Cell. Immunol. 188: 89–96.
18. Saginario, C., H. Sterling, C. Beckers, R. Kobayashi, M. Solimena, E. Ullu, andA. Vignery. 1998. MFR, a putative receptor mediating the fusion of macro-phages. Mol. Cell. Biol. 18: 6213–6223.
19. Kyriakides, T. R., M. J. Foster, G. E. Keeney, A. Tsai, C. M. Giachelli,I. Clark-Lewis, B. J. Rollins, and P. Bornstein. 2004. The CC chemokine ligand,CCL2/MCP1, participates in macrophage fusion and foreign body giant cell for-mation. Am. J. Pathol. 165: 2157–2166.
20. Bonnema, H., E. R. Popa, M. M. Van Timmeren, P. B. Van Wachem,L. F. M. H. de Leij, and M. J. Van Luyn. 2003. Distribution pattern of themembrane glycoprotein CD44 during the foreign-body reaction to a degradablebiomaterial in rats and mice. J. Biomed. Mater. Res. 64: 502–508 (Pt. A).
21. Vignery, A. 2005. Macrophage fusion: the making of osteoclasts and giant cells.J. Exp. Med. 202: 337–340.
22. Yagi, M., T. Miyamoto, Y. Sawatani, K. Iwamoto, N. Hosogane, N. Fujita,K. Morita, K. Ninomiya, T. Suzuki, K. Miyamoto, et al. 2005. DC-STAMP isessential for cell-cell fusion in osteoclasts and foreign body giant cells. J. Exp.Med. 202: 345–351.
23. Rassendren, F., G. N. Buell, C. Virginio, G. Collo, R. A. North, andA. Surprenant. 1997. The permeabilizing ATP receptor (P2X7): cloning and ex-pression of human cDNA. J. Biol. Chem. 272: 5482–5486.
24. Steinberg, T. H., A. S. Newman, J. H. Swanson, and S. C. Silverstein. 1987.ATP4-permeabilizes the plasma membrane of mouse macrophages to fluorescentdyes. J. Biol. Chem. 262: 8884–8888.
25. Falzoni, S., P. Chiozzi, D. Ferrari, G. Buell, and F. Di Virgilio. 2000. P2X7
receptor and polykarion formation. Mol. Biol. Cell 11: 3169–3176.
26. Chiozzi, P., J. M. Sanz, D. Ferrari, S. Falzoni, A. Aleotti, G. N. Buell, G. Collo,and F. Di Virgilio. 1997. Spontaneous cell fusion in macrophages cultures ex-pressing high levels of the P2Z/P2X7 receptor. J. Cell Biol. 138: 697–706.
27. Mizuno, K., H. Okamoto, and T. Horio. 2001. Heightened ability of monocytesfrom sarcoidosis patients to form multi-nucleated giant cells in vitro by super-natants of concanavalin A-stimulated mononuclear cells. Clin. Exp. Immunol.126: 151–156.
28. Mizuno, K., H. Okamoto, and T. Horio. 2004. Inhibitory influences of xanthineoxidase inhibitor and angiotensin I-converting enzyme inhibitor on multinucle-ated giant cell formation from monocytes by downregulation of adhesion mole-cules and purinergic receptors. Br. J. Dermatol. 150: 205–210.
29. Luttikhuizen, D. T., M. C. Harmsen, L. F. de Leiji, and M. J. Van Luyn. 2004.Expression of P2 receptors at sites of chronic inflammation. Cell Tissue Res. 317:289–298.
30. Modderman, W. E., A. F. Weiderman, J. Vrijheid-Lammers, A. M. Wassenaar,and P. J. Nijweide. 1994. Permeabilization of cells of hematopoietic origin byextracellular ATP4: elimination of osteoclasts, macrophages and their precursorsfrom isolated bone cell populations and fetal bone rudiments. Calcif. Tissue Int.55: 141–150.
31. Gartland, A., K. A. Buckley, R. A. Hipskind, W. B. Bowler, and J. A. Gallagher.2003. P2 receptors in bone: modulation of osteoclasts formation and activity viaP2X7 activation. Crit. Rev. Eukaryotic Gene Expression 13: 237–242.
32. Gartland, A., K. A. Buckley, W. B. Bowler, and J. A. Gallagher. 2003. Blockadeof the pore-forming P2X7 receptor inhibits formation of multinucleated humanosteoclasts in vitro. Calcif. Tissue Int. 73: 361–369.
33. Hiken, J. F., and T. H. Steinberg. 2004. ATP downregulates P2X7 and inhibitsosteoclast formation in RAW cells. Am. J. Physiol. 287: C403–C412.
34. Ke, H. Z., H. Qi, A. F. Weidema, Q. Zhang, N. Panupinthu, D. T. Crawford,W. A. Grasser, V. M. Paralkar, M. Li, L. P. Audoly, et al. 2003. Deletion of theP2X7 nucleotide receptor reveals its regulatory roles in bone formation and re-sorption. Mol. Endocrinol. 17: 1356–1367.
35. Morelli, A., P. Chiozzi, A. Chiesa, D. Ferrari, J. M. Sanz, S. Falzoni, P. Pinton,R. Rizzuto, M. F. Olson, and F. Di Virgilio. 2003. Extracellular ATP causesROCK1-dependent bleb formation in P2X7-transfected HEK 293 cells. Mol. Biol.Cell 14: 2655–2664.
36. Ferrari, D., C. Pizzirani, E. Adinolfi, S. Forchap, B. Sitta, L. Turchet, S. Falzoni,M. Minelli, R. Baricordi, and F. Di Virgilio. 2004. The antibiotic polymyxin Bmodulates P2X7 receptor function. J. Immunol. 173: 4652–4660.
37. Lemaire, I., and N. Leduc. 2003. Purinergic P2X7 receptor function in lung al-veolar macrophages: pharmacologic characterization and bidirectional regulationby Th1 and Th2 cytokines. Drug Dev. Res. 59: 118–127.
38. Murgia, M., S. Hanau, P. Pizzo, M. Rippa, and F. Di Virgilio. 1993. OxidizedATP: an irreversible inhibitor of the macrophage purinergic P2z receptor. J. Biol.Chem. 268: 8199–8203.
39. Jiang, L. H., A. B. MacKenzie, R. A. North, and A. Surprenant. 2000. Brilliantblue G selectively blocks ATP-gated rat P2X7 receptors. Mol. Pharmacol. 58:82–88.
40. Chen, Y. W., D. L. Donnelly-Roberts, M. T. Namovic, G. A. Gintant, B. F. Cox,M. F. Jarvis, and R. R. Harris. 2005. Pharmacological characterization of P2X7
receptors in rat peritoneal cells. Inflamm. Res. 54: 119–126.41. Kim, M., L. H. Jiang, H. L. Wilson, R. A. North, and A. Surprenant. 2001.
Proteomic and functional evidence for a P2X7 receptor signaling complex.EMBO J. 20: 6347–6358.
42. Pellegatti, P., S. Falzoni, P. Pinton, R. Rizzuto, and F. Di Virgilio. 2005. A novelrecombinant plasma membrane-targeted luciferase reveals a new pathway forATP secretion. Mol. Biol. Cell 16: 3659–3665.
43. Adinolfi, E., M. G. Callegari, D. Ferrari, C. Bolognesi, M. Minelli,M. R. Wieckowski, P. Pinton, R. Rizzuto, and F. Di Virgilio. 2005. Basal acti-vation of the P2X7 ATP receptor elevates mitochondrial calcium and potential,increases cellular ATP levels, and promotes serum-independent growth. Mol.Biol. Cell 16: 3260–3272.
44. MacKenzie, A. B., M. T. Young, E. Adinolfi, and A. Surprenant. 2005. Pseudo-apoptosis induced by brief activation of ATP-gated P2X7 receptors. J. Biol.Chem. 280: 33968–33976.
45. van den Eijnde, S. M., M. J. B. van den Hoff, C. P. M. Reutelingsperger,W. L. van Heerde, M. E. R. Henfling, C. Vermeij, B. Schutte, M. Borgers, andF. C. S. Ramaekers. 2001. Transient expression of phosphatidylserine at cell-cellcontact areas is required for myotube formation. J. Cell Sci. 114: 3631–3642.
46. Das, M., B. Xu, L. Lin, S. Chakrabarti, V. Shivaswamy, and N. S. Rote. 2004.Phosphatidylserine efflux and intercellular fusion in a BeWo model of humanvillous cytotrophoblast. Placenta 25: 396–407.
47. Lemaire, I. 1991. Selective differences in macrophage populations and monokineproduction in resolving pulmonary granuloma and fibrosis. Am. J. Pathol. 138:487–495.
48. Di Virgilio, F., P. Chiozzi, D. Ferrari, S. Falzoni, J. M. Sanz, A. Morelli,M. Torboli, G. Bolognesi, and O. R. Baricordi. 2001. Nucleotide receptors: anemerging family of regulatory molecules in blood cells. Blood 97: 587–600.
