18 U.S.C. Section 1734 solely to indicate this fact. 11 To whom correspondence should be ..... otides (UTP, GTP, ITP, etc.) besides ATP,. This alternative.
Vol. 266, No. 11, Issue of April 15, pp. 8199-8203,1993 Printed in U.S.A.
THEJOURNAL OF BIOLOGICAL CHEMISTRY B 1993 by The American Societyfor Biochemistry and Molecular Biology, Inc.
Oxidized ATP ANIRREVERSIBLEINHIBITOROF
T H E MACROPHAGEPURINERGIC
Pzz RECEPTOR*
(Received for publication, November2, 1992)
Marta Murgial, Stefania Hanaug, Paola Pizzol, Mario Rippag, and Francesco Di VirgilioqII From the $National Research Council Unit for the Study of the Physiology of Mitochondria and Department of Experimental Biomedicine, University . of . Padoua, Padoua, Italy and the Institutes of §Biochemistry and llGeneral Pathology, University of Ferrara, Ferrara, Italy
G protein torelease of Ca2+ from intracellular stores, most the The effects of oxidized ATP (oATP)onresponses triggered by extracellular adenosine 5’-triphosphate potent agonist being the ATP analogue 2-methylthio-ATP; PZxreceptors coincide with ion channels directly gated by the (ATP,) were investigated in the mouse macrophagelike cell line 5774. ATP, induced in this cell line two ligand ( i e . ATP,) and permeable to cations with M , up to kinds of responses mediatedby two different PZpuri- 200, the most potent agonist being a,p-methylene-ATP; PZZ nergic receptors: 1) an early permeabilization of the receptors have been identified with aqueous pores permeable plasma membrane to extracellular hydrophilic mark- to solutesof M , up to900, the most potent agonistbeing ATP ers of M, up to 900 mediated by PZZreceptors; and 2) inits fully dissociated form (ATP4-); P2“ receptorsshare a fast mobilization of Ca2+ from intracellular stores many features of PBYreceptors, except that the preferred mediated by P2y receptors. Low oATP concentrations agonist is reported t o be UTP. On the contrary, although a (100 PM) completely blocked the first response without number of molecules have been proposed asreceptor antagoaffecting thesecond.ATP,-dependent cell swelling, nists, such as reactive blue, suramin, or ANAPPB, none has vacuolization, and lysis were also inhibited. Antago- been shown t o possess high affinity and specificity (Fedan nism developed slowly, as an incubation at 37 “C for and Lamport, 1990; Nakazawa et al., 1990). at least 2 h in the presence of oATP was needed and The 2’,3‘-dialdehyde derivative of ATP, alsoknown as was irreversible, thus suggesting that the inhibitory oxidized ATP (oATP) is routinely used to affinity label nuaction was due to covalent modification of the receptor. cleotide sites in enzymes (Easterbrook-Smith et al., 1976). The rate of hydrolysis of exogenousATP was slightly oATP reacts with unprotonated lysine residues present at the decreased by oATP, indicating a minor blocking effect nucleotide site thus forming covalent bonds (Schiff bases or of this compound on plasma membrane ecto-ATPases dihydromorpholinoderivatives)(Colman,1990).oATPhas in the concentration range tested. These observations been previously used to investigate platelet activation and to suggest that oATP may be a potentially very useful inhibit ATP stimulation of developing chicken skeletalmuscle tool for isolation and characterization of the PZZ puri- (Pearce et al., 1978; Thomas et al., 1991). In the present study, nergic receptor. we have examined the effects of oATP in the macrophagelike cell line 5774. This cell line, which expresses both PZu and PBZreceptors, is very sensitive to the cytotoxic effect of this nucleotide (Steinberg and Silverstein, 1987; Murgia et al., There is increasing evidence thatextracellularATP 1992a); it is thus ideal t o investigate the role of the various (ATP,)’ may serveas a mediator of cell-to-cell communication purinergic receptors in ATP,-dependentcytotoxicity. by interacting with specific cell surface molecules known as Under our experimental conditions, oATP powerfully anPz purinergic receptors (Burnstock, 1990; Bean, 1992; Osip- tagonized (ICso, 30 p ~ ATP-dependent ) plasma membrane chuck and Cahalan, 1992). Classification of P2 receptors is permeabilization, as assessed by lucifer yellow or ethidium still rudimentary, sincelack of selective agonists/antagonists bromideuptake.Maximumprotective effect was obtained has prevented biochemical and molecular characterization. after a 2-h preincubation in the presence of oATP; washing Current data support the existenceof at least three, possibly out the oATP-containing bathing solution did relieve not the four, P2 receptor subclasses. TheyarePzy, PZx, P2z, and, inhibitoryaction.oATP also preventedall morphological according to some authors, P2u (Dubyak, 1991). Pzyreceptors changes and cytotoxic effects due to ATP, stimulation. On belong to the family of Ca2+-mobilizing receptors, linked via the contrary, no effect of oATP pretreatment was observed on intracellular Ca2+mobilization triggered by either ATP or * This work was supported by grants from the National Research UTP. Hydrolytic activityof plasma membrane ecto-ATPases Council (target projects, Biotechnology-Bioinstrumentation(BTBS) was also inhibited by oATP though to a much lesser extent and Clinical Applications of Cancer Research (ACRO), and special project, Biology and Pathologyof Calcium),the Ministry of Scientific than plasma membrane permeabilization. In fact, pretreatment of macrophages with 100 K M oATP for 120 min deResearch (MURST), the Italian Association for CancerResearch (AIRC),and Telethon of Italy. The costs of publication of this article creased hydrolysis of ATP, by only about 30%, while ATP,were defrayed in part by the payment of page charges. This article dependentplasmamembranepermeabilization wascommust therefore be hereby marked “advertisement” in accordance with pletely abolished. These observations indicate that oATP acts 18 U.S.C. Section 1734 solely to indicate this fact. as an antagonist at the PZZbut not at the Ppyreceptor and 11 To whom correspondence should be addressed Inst. of General Pathology, University of Ferrara, Via Borsari, 46, 1-44100 Ferrara, suggest the possible use of this oxidized nucleotide to covalently label the Pzz receptor. Italy. Tel.: 532-291353;Fax: 532-47278. The abbreviations used are: ATP,, extracellularATP;UTP,, MATERIALS ANDMETHODS extracellular UTP; oATP, oxidized ATP; [Ca2+],,cytosolic free Ca2+ concentration; ANAPP3, 3’-0-(3[N-(4-azido-2-nitrophenyl)amino] Cells and Solutions-The 5774 mouse macrophage cell line and the propionylladenosine 5”triphosphate. ATPRB2 5774 variant were grown in spinner cultures in Dulbecco’s
8199
8200
Oxidized ATP-mediated Inhibition of ATP Stimulation
modified Eagle's medium supplemented with 10% heat-inactivated horse serum, penicillin (100 units/d), andstreptomycin (100 pglml). Unless otherwise indicated, experiments were performed in a saline solution containing (in mmol/liter) 125 NaCI, 5 KC], 1 MgSOI, 1 Na2HP04,5.5 glucose, 5 NaHC03, 1 CaCI2, and 20 Hepes (pH 7.4, 37 "C). This saline medium is heretofore also referred to as standard saline. Measurement of Plasma Membrane Permeability Chunges-Increases in plasma membrane permeability were monitored by measuring either lucifer yellow or ethidium bromide uptake, as previously described (Di Virgilio et al., 1988a; Steinberg and Silverstein, 1989). For lucifer yellow uptake, macrophage monolayers, after a variable pretreatment in the presence of oATP, were incubated for 5 min at 37 'C in Ca2+-free,1 mM EGTA-supplemented standard saline containing 3 mM ATP,,250 p M sulfinpyrazone, and 1 mg/ml lucifer yellow. After this incubation time, the monolayers were rinsed, placed in sulfinpyrazone-supplemented, ATP-free medium, and analyzed with an Olympus IMT-2 fluorescence microscope. For ethidium bromide uptake, macrophage suspensions were pretreated for a variable length of time with oATP at37 "Ca t a concentrationof 2 X lo6cells/ ml, washed, and resuspended in a thermostatted and magnetically stirred fluorometer (Perkin-Elmer LS5) a t a concentration of 5 X 10' cells/ml. Fluorescence changes were monitored at thewavelength pair 360-580 nm. Measurement of [Ca2+Ii-Loading with fura-PIAM and measurement of [Ca2+Iiwere performed essentially as previously described (Di Virgilio et al., 1988b). To prevent fura-2 leakage and sequestration, 250 PM sulfinpyrazone was presentthroughout the loading procedure and [Ca2+Iimeasurement. [Ca2+Ii measurementswere performed in a thermostattedand magnetically stirred fluorometer (Perkin-Elmer LS5) cuvette a t a concentration of 5 X lo5 cells/ml. Measurement of Plasma Membrane Potentiul-Changes in plasma membrane potential were measured with the fluorescent dye bis(l,3diethylthiobarbiturate)trimethineoxonal (bisoxonol) at the wavelength pair 540-580 nm. Measurement of Enzymatic Activity-Lactate dehydrogenase activity was measured according to standardmethods (Bergmeyer, 1963). Hydrolysis of ATP. by plasma membrane ecto-ATPases was measured as described by Ames (1966). Chemicals-oATP was prepared by treating ATP with periodate a5 follows. 300 pmol of ATP disodium salt (Sigma) dissolved in 1 ml of double distilled H20, at pH 5.0, were treated with 600 pmol of solid NaI04 at room temperature in the dark. After 60 min, the solution was subjected to gel filtration on acolumn (34 X 1.7 cm) of Sephadex G-10 fine equilibrated in HZ0 a t 0 "C. The column was eluted with H20, and fractions (1 ml) were collected and analyzed a t 260 nm. Fractions containinghighest oATP concentration, free of NaI03 and NaI04, were used. Control experiments also showed that a 3-h incubation of 5774 macrophages in the presence of NaI04 atconcentrations up to 300 p~ had no inhibitory effect on theP2z receptor. RESULTS
The macrophage P2z receptor is activated by ATP, in the millimolar range, thus allowing cytoplasmic loading of extracellular water-soluble low molecular weight markers such as the highly fluorescent dye lucifer yellow (Fig. 1,panel B ) . Fast opening of the pore by ATP, permits permeabilization of about 80-85% of the cells and near maximal uptake of the external marker within10 min. oATP is not an agonist at the PZZreceptor, as even prolonged incubations with this nucleotide a t a concentration up to 1 mM did not cause detectable increases inplasmamembranepermeability (not shown); furthermore, it appears to be a weak competitive inhibitor, since simultaneous addition of oATP and ATP, did not significantly affect permeabilization (Fig. 1, panel A ) . Macrophages were therefore treated with oATP for increasing incubation times prior to ATP, addition. The length of pretreatment turned out to be crucial, as while a 1-h incubation at 37 "C in the presence of oATP had only a slight protective effect, a 2-h incubation completely inhibited lucifer yellow uptake (Fig. 1,panel A and fluorescence photograph in panel C).Washing out theoATP-containing solutiondid not restore sensibility to ATP,, suggesting that blocking of the P2z recep-
A
Incubation time (h)
FIG.1. Oxidized ATP prevents extracellular ATP-dependent plasma membrane permeabilization. Panel A, macrophage monolayers (2 X 105/well) were treated with 300 p~ oATP at 37 "C in Ca2+-free, 100 p~ EGTA-supplemented saline for the indicated times, exposed to 3 mM ATP, for 10 min, washed, and analyzed by fluorescence microscopy (Olympus IMT-2). 0, cells kept in the presence of 300 p~ oATP duringthe challenge with ATP,; 0,cells washed free of oATP before the challenge with ATP,. Panel B, fluorescence photograph of control macrophages permeabilized by 3 mM ATP.. Panel C, fluorescence photograph of macrophages treated with 300 p M oATP before stimulation with 3 mM ATP.. Note pinocytotic uptake of lucifer yellow in this lattercase. Bar = 25 pm. tor was irreversible under these experimentalconditions (Fig. 1, panel A ) . ATP-dependent changes in plasma membrane permeability can also be monitored by measuring ethidium bromide fluorescence increases. This cationic fluorescent dye easily permeates through the ATP-activated pore, diffuses into thenucleus, binds to DNA, and undergoes a fluorescence increase (Di Virgilio and Steinberg, 1992). As shown in Fig. 2, even under these extreme experimental conditions (Ca2+free saline supplementedwith 1mM EDTA) in which most of the added ATP, was present as the active fully dissociated species ATP:-, challenge with 3 mM ATP, induced an immediate and long lasting ethidium bromide uptake in control (Fig. 2, trace a ) but not in oATP-treated (Fig. 2, trace b ) macrophages. An early consequence of P2z receptor activation is plasma membrane depolarization, one of the most sensitive indexes of change in plasma membrane ion permeability (Fig. 3, upper trace). Pretreatment with oATP fully prevented ATP,-trig-
Oxidized ATP-mediated Inhibition of ATP Stimulation
'ATP
8201
Dig
-
2 rnin
FIG.2. Oxidized ATP prevents ethidium bromide uptake by macrophages pulsed with extracellular ATP. Macrophage suspensions (2 X lo6 cells/ml) were preincubated in standard saline at 37 "C for 2 h in the absence (trace a ) or presence (trace b ) of 300 p~ oATP, washed, and transferred to a fluorometer cuvette containing Ca*+-freesaline supplemented with 1mM EDTA and 20 p~ ethidium bromide. ATP, was added a t a concentration of 3 mM and digitonin (Dig), 100 p M . -/"-l
I
I
-
80 sec
FIG.3. Oxidized ATP inhibits extracellular ATP-induced plasma membrane depolarization.Macrophage suspensions (5 X 105/ml) were pretreated a t 37 "C for 3 h in standard saline in the absence (upper trace) or presence (lower trace) of 300 p~ oATP. They were then washed free of oATP and placed in a fluorometer cuvette containing saline supplemented with 0.5 mM Ca2+and 200 nM bisoxonol. ATP, was 1 mM. Each KC1 addition (small arrows) was 15 mM.
gered depolarization (Fig. 3, lower trace), clearly indicating that P2z-dependent ion fluxeswereblocked in these cells. However, KC1-dependentdepolarization was unaffected, thus a physiological, showing that oATP-treated macrophages had polarized membrane potential. 5774 macrophages usually recoverafter a brief (15-30 min) plasma membrane permeabilization, but longer exposures to ATP, cause irreversible damage and lysis (Di Virgilio et dl., 1988a; Murgia et al., 199213). It is expected that if activation of the P2z receptor and the ensuing changes in membrane permeability areresponsible for the cytotoxic action of ATP,, blocking by oATP should prevent lysis. As shown in Fig. 4, this was indeed the case, as pretreatment with oATP for2 h (oATP I&, 30 PM) completely prevented not onlyrelease of the cytoplasmic enzyme lactic dehydrogenase (Fig. 4, panel A ) but alsomorphological changes (swelling, membrane blebbing, vacuolization) caused in macrophagesby challenge with ATP, for 6 h (Fig. 4, panels B-D). 5774macrophages also express Pay receptors,which are linked to inositol 1,4,5-trisphosphategenerationand Ca2+ mobilization from intracellular stores (Greenberget al., 1988). The mutant macrophage cell line ATPRB2 was selected for resistance to ATP, and shown lack to the P2z but not theP2y receptor (Steinberg and Silverstein, 1987; Greenberg et al., 1988). These mutants are therefore ideally suited to investigate oATP antagonismat thePzu receptor. In the experiment reported in Fig. 5, ATPRBZ macrophage suspensions were
FIG.4. Dependence on the concentration of oxidized ATP of lactic dehydrogenase (LDH) release triggeredby extracellular ATP. Panel A , macrophage monolayers ( 2 X 105/well)were incubated at 37 "C in 24-well Costar dishes in Ca2+-free, 100 p M EGTA-supplemented saline in the presence of increasing oATP concentrations. After 2 h, 3 mM ATP, was added and the incubation carried out for an additional 6 h. Panel B , control macrophages. Panel C, macrophages pulsed with 3 mM ATP, for 6 h. Panel D, macrophages treated with 300 p~ oATP for 2 h before challenge with 3 mM ATP,. Swollen cells are indicated by arrows in panel C. Bar = 25 um.
Oxidized ATP-mediated Inhibition
8202
[.""Ii
[Ca2+IinM
A
nM
hTP
2'
of ATP Stimulation
!ATP
d
I ,
-2 40
L-
-640 200
'U TP
'UTP
parental cell line (not shown). At lower agonist concentrations, the nucleotide-induced [Ca2+lirise was too small to allow assessment of a possible inhibition. oATP hasbeen extensively used to label the purine nucleotide-binding siteof enzymes (Easterbrook-Smith et al.,1976); we therefore expected to also find an inhibition of ectoATPase activity. This was indeed the case, albeit to a much lesser extent compared with inhibition of plasma membrane permeabilization. In fact, a 120-min incubation in the presence of 100 PM oATP decreased hydrolysis of ATP, by only 30% (Table I), while under these same conditions, plasma membrane permeabilization was completely abrogated. In addition, even at the highest oATP dose tested (300 p M ) , we were unable to achieve more than 50% inhibition of ectoATPase activity. DISCUSSION
60sep
Almost 30 years ago, Burnstock and co-workers (1964) first suggested a role for ATP as a true extracellular mediator of cell-to-cell communication. This hypothesis has now been 0 confirmed in several biological systems, since more and more cell types appear to be responsive to ATP, (Dubyak, 1991). The wealth of observations on the metabolic and functional responses triggered by ATP,, however, is not supported by a comparable large body of biochemical and molecular data on the membrane receptors because of the lack of selective lirn gands for the various Pz receptor subtypes. oATP is a comrn pound long known as a specific label for the nucleotide site of several ATP-requiring enzymes, since oxidative cleavage of the bond between C-2' and C-3' of the ribose ring yields a 2',3'-dialdehyde derivative that forms covalent bonds with ' unprotonated lysine residues present at theATP-binding site 0 10 30 100 300 (Easterbrook-Smith et al., 1976). With theexception of a few [.ATP] pM studies (Pearce et al., 1978; Thomas et al., 1991), oATP has FIG. 5. Oxidized ATP does not inhibit nucleotide-triggered seldom been used to investigate plasma membrane receptors [Cap+],increases. Macrophages (5 X 106/ml),preincubated at 37 "C for ATP,. In the present report, we have explored the interfor 2 h in the absence (traces a and c) or presence (traces b and d ) of action of oATP with Pzreceptors expressed by the 5774 mouse oATP, were suspended in a fluorometer cuvette and stimulated with macrophages, a cell line that possesses at least two types of either 500 p~ ATP, or UTP, (panel A ) . In p a n e l B, changes in [CaZ'lj Pz receptors (Pzzand Pzy) and isvery sensitive to ATP, are expressed as increases above resting level triggered by either ATP. (Steinberg and Silverstein, 1987; Murgia et al., 1992b). ATP,(W) or UTP. ( 0 ) . resistant 5774 mutant clones ( e g . ATPRB2 cells), which express only Pzyreceptors and are resistantto ATP,, are also incubated at 37 "C in the presence (Fig. 5, trace a ) or absence available. 5774 macrophages undergo complex metabolic (Fig. 5, trace b ) of oATP for 2 h, then challenged with ATP,. changes upon ATP, stimulation and eventually die. Toxic ATP, caused afast [Ca2+Iirise almost exclusively due to release from intracellular stores, which leveled off within 2 TABLEI min and was unaffected by pretreatment with oATP. It is Effect of oATP on the hydrolysis of exogenous ATP becoming increasingly evident that UTPis also a potent Ca2+- Macrophages (2 X 106/well) were pretreated with oATP in serummobilizing agonist in several cell types, although it isunclear free Dulbecco's modified Eagle's medium for the indicated times and whether this occurs via the Ppyor another still poorly char- then incubated in medium containing 120 mM NaC1, 5 mM KCl, 20 acterized "nucleotide" receptor (Dubyak, 1991). As shown in mM Hepes/Tris (pH7.4),2 mM CaC12,and 3 mM MgC12 (final volume, trace c of Fig. 5, UTP,, in contrast to ATP,, also induced a 200 pl). Plateswere transferred at 37 "C and the reaction started by the addition of 2 mM ATP. After 1 h, supernatants were withdrawn sustained [Caz+li increase after the initial fast transient, a and assayed for inorganic phosphate content. Data are means of typical kinetics of receptors that activate both Ca2+release duplicate samplesfrom a single experiment repeated in three different from stores and Ca2+influx through receptor-activated chan- occasions with similar results. Inhibition is expressed as percent nels (Di Virgilio et al., 1987; Pietrobon et al., 1990). Neither decrease of maximal rate of inorganic phosphateformation (24 nmol/ in the the early nor the late [Ca2+]irise was blocked by oATP. In lo6 cells/min)measuredincontrolmacrophagesincubated these experiments, resting [Ca2+Iiwas about twice as high as absence of oATP and in the presence of 2 mM ATP.. Inhibition the levelpreviously reported inthese and other cells (Di Pretreatment Virgilio et al., 1988a; Greenberg et al., 1988), an effect likely 10pM" 30pM" 50pM" 100pM' 300pMO dependent on the prolonged incubation in standard saline at rnin ?6 37 "C. The results of a number of nucleotide-triggered [Ca2+]i 0 0 14 0 2 6 measurements performed in ATPRB2 macrophages preincu10 15 22 30 10 13 bated with increasing oATP concentrations areshown in Fig. 60 10 13 30 20 30 15 23 28 22 38 90 5, panel B. oATP pretreatment was also without effect on 120 10 13 26 28 43 Cap+mobilization triggered by ATP,/UTP, concentrations as a oATP concentration. low as 30 PM in both the ATPRB2 macrophages and the5774
a30:l:: :
'
Oxidized ATP-mediated Inhibition of ATP Stimulation responses to ATP, have been described in several other cell types, although the membrane receptors andintracellular mechanisms involved are poorly characterized. As previously reported by other investigators (Thomas et al., 1991), oATP is unable to antagonize ATP, stimulation when the two nucleotides are added together. The inhibitory effect develops only after a prolonged (1-2 h) incubation at 37 "C and is irreversible, an indication that oATP covalently modifies the receptor. Under these conditions, effects due to activation of PZzreceptors (permeabilization to lucifer yellow, swelling, and lysis) are fully antagonized at oATP concentrationsas low as 100 p ~ On . the contrary, mobilization of intracellular Ca2+, an effect linked to activation of the PPY receptor, is unaffected. This observation could be explained by either low affinity of the P2y receptor for oATP or lack of an unprotonated lysine at thePPyreceptor nucleotide-binding site. There is, however, a third possible explanation, i.e. that the receptor responsible for Ca2+ release from intracellular purinergic but a stores in 5774 macrophages is not a P ~ Y nucleotide receptor responsive to several extracellular nucleotides (UTP, GTP, ITP,etc.) besides ATP,. This alternative is intriguing, since accruing evidence suggests that other extracellular nucleotides besides ATP,, in particular UTP,, might have important signaling roles, although it is not at all clear whether theseeffects are mediated via the same receptor activated by ATP, (purinergic) or via a different receptor (pyrimidinergic? nucleotide receptor?) (Seifert and Schultz, 1989; O'Connor, 1992). The different pattern of [Ca2+Ii changes triggered by ATP, and UTP, (see Fig. 5) might suggest that in ATPRB2 macrophages, purine and pyrimidine nucleotides activate two different Ca2+-mobilizingreceptors, but further investigation is clearly needed to solve this problem. In this respect, it is not unlikely that theuse of oxidized UTP, which should inhibit ATP,-triggered Ca2+release from intracellular stores without effects on ATP,-dependent plasma membrane permeabilization to extracellular markers, . , could be of great help. oATP also decreased the rate of hydrolysis of ATP, by intact macrophages. An effect on plasma membrane ectoATPases was not unexpected, as oATP has been extensively used to label the nucleotide-binding site of ATP-dependent enzymes (Easterbrook-Smith et al., 1976;Colman, 1990);however, ecto-ATPaseinhibition was much smaller thanthe inhibition of plasma membrane permeabilization. This result suggests the possible use of oATP toselectively abrogate PPZdependent plasma membrane permeability changes with a minimal decrease of ecto-ATPase activity. oATP selectivity for the PPZreceptor suggests this oxidized nucleotide is a potentially very useful tool for isolation and investigation of the physiological role of this receptor. Although long known (Cockcroft and Gomperts, 1979), no real
8203
physiological significance has ever emerged for the plasma membrane-permeabilizing effects of ATP,. We andothers have proposed that ATP, could function as a cytotoxic molecule in a number of physiological and pathological conditions and that the P2z receptor might be a "suicide receptor" that mediates ATP, cytotoxic activity (Di Virgilio et al., 1990; Filippini et al., 1990), butthis hypothesis has beenvery difficult to prove because of the lack of selective blockers of the PZzreceptor. The present observations support the identification of Pzz as the receptor mediating ATP, cytotoxic effects in macrophages and offer a tool to furtherexplore the role of this nucleotide in other cytotoxic reactions (e.g. cellmediated cytotoxicity). In this respect, oATP has two interesting features: 1) it covalently and irreversibly modifies the receptor; and 2) it is free of untoward side effects for incubations up to several hours. Thus,it should be possible to investigate responses of macrophages (and possibly other cell types) permanently modified in their PZzreceptors in in vitro and in vivo conditions that do not require the continuous presence of bathing oATP. REFERENCES Ames, B. N. (1966) Methods Enzymol. 8,115-117 Bean, B. P. (1992) Trends Pharmacol. Sci. 13,87-90 Bergmeyer, H. U. (1983) Methods of Enzymatcc Analysis, Vol. 111, pp. 118-133, Academic Press Ltd., London Bumstock, G. (1990) Ann. N. Y. Acad. Sei. 603.1-17 Bumstock. G.. CamDbell. G.. Bennet. M. & Holman. M. E. (1964) . . Int. J. NeuropLrmacol. 3 , 163-166 Cockcroft, S. & Gomperts, B. D. (1979) Nature 279,541-542 Colman, R. F. (1990) in The Enzymes (Sigman, D. S., and Boyer, P. D., eds) Vol. 19, pp. 283-323, Academic Press, San Diego Di Virgilio, F. & Steinberg, T. H. (1993) Protocols in Cell & Tissue Culture, John Wiley & Sons Ltd., West Sussex, UnitedKingdom, in press Di Virgilio, F., Lew, D. P., Andersson, T. & Pozzan, T. (1987) J. Biol. Chem. 262. - -, 4.574-4.579 - - . - - - .Di Virgilio, F., Meyer, B. C., Greenberg, S. & Silverstein, S. C. (1988a) J. Cell Biol. 106,657-666 Di Virgilio, F., Fasolato, C. & Steinberg, T. H. (1988b) Biochem. J. 2 5 6 , 959QG!2 1 "
Di Virgilio, F., Pizzo, P., Bronte, V., Zanovello, P. & Collavo, D. (1990)Immuml. Today 11,274-277 Dubyak, G. R. (1991) Am. J. Respir. Cell Mol. Biol. 4,295-300 Easterbrook-Smith.. B.. ,Wallace. J. C. & Keech. D. B. (1976) Eur. J. Biochem. 62,125-130 Fedan, J. S. & Lamport, S. J. (1990) Ann. N . Y. Acad. Sei. 6 0 3 , 182-197 Filippini, A,, Taffs, R. E., Agui, T. & Sitkovsky, M. V. (1990) J. Biol. Chem. 266,334-340 Greenberg, S., Di Virgilio, F., Steinberg, T. H. & Silverstein, S. C. (1988) J. Biol. Chem. 2 6 3 , 10337-10343 Murgia, M., Pizzo, P., Zanovello, P., Zambon,A. & Di Virgilio, F. (1992a) Altern. Laboratory Animals 20,66-70 Murgia, M., Pizzo, P., Steinberg, T. H. & Di Virgilio, F. (1992b) Biochem. J. 288,897-901 Nakazawa, K., Fujimori, K., Tanaka, A. & Inoue, K. (1990) Br. J. Pharmocol. 1 0 1 , 224-226 OConnor, S. E. (1992) Life Sci. 60,1657-1664 Osipchuck, Y. & Cahalan, M. (1992) Nature 369,241-244 Pearce, P. H., Wright, J. M., Egan, C. M. & Scrutton, M. C. (1978) Eur. J. Biochem. 88,543-554 Pietrobon, D., Di Virgilio, F. & Pozzan, T. (1990) Eur. J. Biochem. 1 9 3 , 599G99
" I
Seifert, R. & Schultz, G. (1989) Trends Pharmacol. Sci. 10,365-369 Steinberg, T.H. & Silverstein, S. C. (1987) J. Biol. Chem. 262,3118-3122 Steinberg, T.H. & Silverstein, S. C. (1989) Methods Cell Biol. 31,45-61 Thomas, S. A., Zawisa, M. J., Lin, X. & Hume, R. I. (1991) Br. J. Pharmacol. 103,1963-1969
