sign in sign up
<<
Home , Nucleotide

THEJOURNAL OF BIOLMXCAL CHEMISTRY Vol. 269, No. 16, Issue of April 22, pp. 11830-11836, 1994 0 1994 by The American Soeiety for and Molecular , Inc. Printed in USA. Identification of a Nucleotide-selective G--linked That Activates PhospholipaseC*

(Received for publication, October 21, 1993, and in revised form, January 14, 1994)

Eduardo R. Lazarowski and T. Kendall Harden$ From the Department of Pharmacology, Uniuersity of North Carolina, School of Medicine, Chapel Hill,North Carolina 27599

Incubation of C6-2B rat gliomacells withUDP or UTP Although care has been taken toprove that theP,,-purinergic resulted in atime- and concentration-dependent in- receptor that stimulates lipid hydrolysis in man.y tar- crease in the accumulation of inositol . In get tissuesis a single receptor activated by both ATP and UTP contrast, ATP, ADP, and analogs of these nucleotides (8,9), this does not rule out thepossibility that separate recep- known to be effective agonists Pzv-, at Pm-, P,,-, P,,-, and tors for and occur. Indeed, studies meas- P,,-purinergic receptors all had no effect on inositol uring vascular smooth muscle responses (10-12) or metabolic levels inC6-2B cells. nucleotides effects in liver (13) and in neutrophils or HL-60 cells (14, 15) stimulated inositol phosphate accumulation with an or-have led to the proposal that “pyrimidinergic” receptors exist der of potency of UDP > 5-BrUTP > UTP > dTDP > UDP (16).However, none of these havebeen unambiguously defined. glucose. values for UDP, 5-BrUTP,and UTP were 2.3 A large amount of information has accumulated on the physi- 0.5,9 3, and 57 10 respectively.A similar uridine * * m, ological release ofATP and ADP. However, outside of its storage nucleotide selectivity was observed for arachidonic acid release presumably occurring asa consequence ofacti- in high concentrations in platelet storage granules, little is vation of phospholipase 4. Cross-desensitization and known for UTP concerning its potential storage, release, and additivity experiments indicated thatUDP and UTP in- activity as an extracellular molecule. The presence of P,,-pu- teract with the same populationof receptors. The effect rinergic receptors that respond to extracellular UTP supports of uridine nucleotideson inositol phosphate accumula- the idea thatthis is an important extracellular signalingmol- tion was inhibited markedly by pretreatment of cells ecule, and potentially provides avenues for studying thephysi- with pertussis toxin.UDP also caused a nucle- ological regulation of extracellular UTP levels. otide-dependent increase in inositol lipid hydrolysis in The inositol lipid signaling cascade is the major second mes- streptolysin-0-permeabilized cells. Takentogether senger response known to be associated with activation of P,- these results describe the existence of a novel uridine purinergic receptors (3). Both the P,,- and P,,-purinergic re- nucleotide receptor that is not activated by ceptors activate phospholipase C in a broad range of tissues. nucleotides. This receptoris pharmacologically distinct However, this is not the only signaling response regulated by from the previously described P2”- and other P,-puriner-these receptors. For example, activation of a P,-purinergic re- gic receptors, and likely is a member of a new class of ceptor that appears to be distinct from the p,,-purinergic re- receptors for extracellular nucleotides. ceptor that activates phospholipase C results in a decrease in intracellular cyclic AMP accumulation in some target tissues (17-20). It is also likely that there are otherreceptors for ex- Extracellular release of adenine nucleotides from neurons, tracellular adenine and uridinenucleotides that have not been platelets, and other tissues results in activation of a broad identified, and we report here that C6-2B glioma cells express range of physiological responses (1, 2). Considerable progress a novel receptor that possesses a strict selectivity for uridine has been made in the last decade in the pharmacological de- nucleotides. Activation of this receptor results in activation of lineation of the cell surface receptors that are activated by phospholipase C through a guanine nucleotide-dependent per- extracellular ATP and ADP. These P,-purinergic receptors in- tussis toxin-sensitive pathway. The existence of a receptor that clude the P,,-, P,,-, P,,-, P,,-, and P,,-receptors, which are is activated by uridine nucleotides but not by adenine nucle- selectively activated by ATP, ADP, or various analogs of these otides adds support to the idea that extracellular UTP sub- two adenine nucleotides (1-3). The one exception to the adenine serves an important function. nucleotide selectivity of these receptors is the so-called P,,- EXPERIMENTALPROCEDURES , which is not only activated by ATP and ATPyS,’ but also is stimulatedby the pyrimidine, UTP (4-8). Cell Culture and Subcloning“C6-2B and C6 rat glioma cells and HT29 human colonic carcinoma cells were grownin Dulbecco’s modified Eagle’s medium supplemented with DMEM and5% fetal bovine serum * This work was supported by United States Public Health Service at 37 “C in a humidified atmosphere of 5% CO,, 95% air. To isolate Grants HL32322, GM38213,and GM29536. The costsof publication of homogeneous cell populations derived from single cloned cells, C6-2B this article were defrayed in part by the payment of page charges. This cells were seeded on 150-cm2 dishesat cell densities rangingfrom 10 to article must thereforebe hereby marked “aduertisement”in accordance 1000 cellddish. Attached isolated single cells were identifiedthe follow- with 18 U.S.C. Section 1734 solely to indicate this fact. ing day, and compact clusters (apparently clonal)of dividing cells were $ To whom correspondence should be addressed: Dept. of Pharmacol- isolated using 7-mm diameter cloning cylinders. The C6-2B cell clones ogy, UNC-CH School of Medicine, CB 7365, FLOB, Chapel Hill, NC were transferred to tissue culture flasks andnumbered in the order of 27599-7365. Tel.: 919-966-4816;Fax: 919-966-5640. The abbreviations used are: ATPyS, -5’-0-(3-thiotri- their isolation. Inositol Phosphate Assay-Confluent cells grown on 12-well plastic phosphate); a,p-MeATP,a$-methylene adenosine-5’-triphosphate; 5-BrUTP, 5-bromo-uridine-5‘-triphosphate;2-MeSATP, 2-methylthio- plates (Costar, Cambridge, MA) were incubated for 18 h in 0.5 ml of ; GTPyS, -5‘-0-(3-thiotriphosphate); HPLC, high perfomance liquid chromatography; Ins(1,4,5)P3, inositol 4,5)P4,inositol 1,3,4,5-tetrakisphosphate;DMEM, Dulbecco’s modified 1,4,5-trisphosphate; Ins(1,3,4)P3, inositol 1,3,4-trisphosphate;Ins(l,3, Eagle’s medium. 11830

This is an Open Access article under the CC BY license. Uridine Nucleotide Receptor 11831 inositol-free DMEM containing 10 pCi/ml [3Hlinositol.The cells were washed twice with HEPES-buffered (pH 7.4) DMEM (WDMEM) and preincubated with WDMEM containing 10 m~ LiCl for 15 min before challenge with agonists. In those experiments where desensitization was studied, the cells were preincubated with agonist for 20 min prior to addition of LiCl and rechallenged by agonist. Incubations were ter- 0 CONTROL minated by removal of the medium followed by addition of 5% trichlo- roacetic acid. The trichloroacetic acid extracts were neutralized by (3 x 2 ml) extractions with ethyl ether. Resolution of the individual inositol phosphates onDowex AGl-X8 columns was performed as described previously (21). Quantitation of Individual Inositol Phosphates by HPLC-Samples were prepared as described above except that the cells were labeled with 100 pCi of [3Hlinositol in 0.5 ml of DMEM. Individual inositol phosphates were resolved using a Whatman Partisil 10 strong anion- exchange column (Alltech Applied Science, Deerfield, IL) and an am- monium phosphate gradient as described previously(22). Radioactivity was monitored on-linewith a Radiomatic Flo-one detector (Radiomatic I Instruments, Tampa, FL). 0 10 20 30 Arachidonic Acid Release-Confluent cells grown on 12-well plastic plates were labeled for6 hwith 0.5 pCiof [3Hlarachidonicacid in 0.5 ml TIME, min of DMEM. The cells were washed free of unincorporated radioactivity FIG. 1.Time course of inositol phosphate formation in C6-2B and challenged with agonists for 10 min. Released [3Hlarachidonicacid glioma cells. [3HlInositol-labeled C6-2B cells were challengedwith 300 was measured in the cell supernatant as reported previously (23). p~ carbachol (A),UTP (W), UDP (O),ATP (O),ATPyS (A), or vehicle( 0 ) Assay of Phospholipase CActivity in Permeabilized Cells -[3HlInosi- for the indicated times. Incubations were terminated by addition of 5% tol-prelabeled cells on 12-well plates were rinsed twice with phosphate- trichloroacetic acid and total inositol phosphates were collected and buffered saline and permeabilized by a 5-min incubation with 0.2 IU/ml measured as detailed under “Experimental Procedures.”The amount of streptolysin-0 inassay buffer containing 0.424 m~ CaCl,, 3 m~ MgCl,, [3Hlinositol phosphates present at t = 0 (-2000 cpm) was subtracted 2 m~ EGTA, 115 m~ KCl, 10 m~ HEPES, and 10 m~ LiCl (pH 7.4) as from the indicated values. Data represent the mean value of results described previously(8). Streptolysin-0 containing buffer was removed obtained with triplicate assays that are representative of results from and thecells wereincubated in 0.5 ml of assay buffer containing various three different experiments. Error bars were omitted for the clarity of additions. Assays were terminated by the addition of 0.25 ml of 15% the figure. trichloroacetic acid and [3Hlinositolphosphates were quantitated as described above. TABLEI Materials-Myo-[2-3Hlinositol(20Ci/mmol) and [5,6,7,8,9,11, 12,14,- Agonist-induced stimulation of inositol trisphosphate and 15-3Hlarachidonicacid (200 Ci/mmol) were purchased from American tetrakisphosphate Radiolabelled Chemicals, Inc. (St. Louis, MO). Uridine, UMP,UDP, [3HlInositol prelabeled C6-2B cells were challenged eitherfor 20 s or UTP, 5-BrUTP, UDP-glucose, ADP, ATP, dTTP, GDP, GTP, CDP, CTP, 20 min with 300 p~ ATP, UDP, or carbachol. Individual inositol phos- dUTP, dCTP, dGTP,and dATP were from Boehringer Mannheim. 2-Me- phates were separated by HPLC as detailed under “Experimental Pro- SATP and a,P-MeATP werepurchased from Research Biochemicals Inc. cedures.” Data for levels of Ins(1,4,5)P3,Ins(1,3,4)P3, and Ins(1,3,4,5)P4 (Natick, MA). 2’-Deoxynucleotide diphosphates (dUDP,dTDP, and are presented and are the mean 2 S.D. of three determinations. dCDP) were purchased from Sigma. Pertussis toxin was fromList Bio- Addition Ins(1,4,5)P3Ins(1,3,4)P3Ins(I,3,4,5)P4 logical Laboratories, Inc. (Cumsell, CA). Streptolysin-0 was obtained from Murex DiagnosticsLimited (Dartford, United Kingdom). CPm was obtained from the Center for Disease Control (Atlanta, GA). 20 s Data Analysis-Unless stated otherwise, experiments wereper- 3240None 2 280 86 2 35 636 2 286 ATP 3363 2 300120 2 29 696 2 12 formed with triplicate determinations that differed ~10%from the UDP 2961 2 293 752 2 53 1819 2 433 mean. The data were consideredstatistically different at p c 0.05 using Carbachol 25162165316625672788240 a paired Student’s t test. 20 min None2594 2119 206 2 93 849 k 316 RESULTS 2938ATP 2122 645 2 16 744 k 23 Incubation of C6-2B rat glioma cells with UDP resulted in a 3146UDP k 609 749 2 43 2433 -c 435 time-dependent increase in intracellularlevels of inositol phos- Carbachol 307523637382229 4205 k 250 phates (Fig. 1).Incubation of these cells with UTPalso elicited an inositol phosphate response that was temporally and quan- producible effect on accumulation of individual [3H]inositol titatively similar to theresponse observed with UDP. Inositol phosphates during a 20-s, 5-min, or 20-min incubation. The phosphate accumulation in the presence of either agonist was lack of effect ofATP was notdue tohydrolysis of this nucleotide essentially maximal within 10 min, an effect that apparently is since greater than 95% of added r3H1ATP (100 p~)was recov- at least in part due homologous to desensitization of the recep- ered unchanged after incubation in thepresence of C6-2B cells tor responding to the undinenucleotides (see below). The time for 20 min at 37 “C (data not shown). Under the same condi- course of accumulation of inositol phosphates in thepresence of tions greater than 95% of added L3H1UTP also was recovered. the undine nucleotides was similar to that observed with the Observation of inositol phosphate responses to UDP and muscarinic cholinergic receptor agonist carbachol; the maximal UTP but not toATP suggested the existence of a receptor that effect observed with carbachol was up to 2-fold greater than had not been described previously. As such, the response to a that with UDP or UTP (Fig. 1).In contrast to the stimulatory wide range of nucleotides and to analogs of UTP, ATP, and ADP effect of the undinenucleotides, incubation of C6-2B cells with were tested (Fig. 3;Table 11). The most potent agonist identified ATP or ATPyS did not result in increases inositolin phosphate was UDP which stimulated inositol phosphate accumulation levels. To further characterize theeffect of undine nucleotides with a of 2.3 * 0.5 p~ (Fig. 3). However, 5-BrUTP = 9 on inositol phosphate production,individual inositol phos- * 3 p~)andUTP = 57 * 10 m) alsowere full agonists. phates in extractsfrom UDP-, carbachol-, and ATP-pretreated Small effects were observed with UDP-glucose but Uh4P or cells were resolved by HPLC. UDP and carbacholcaused uridine had no effect. Similarly, 2”deoxy-UTP had no effect on marked increases in Ins(1,4,5)P3 and its immediate metabolic inositol phosphate accumulation. The pyrimidine dlTP caused products (Table I and Fig. 2). Although these cells were labeled a concentration-dependent activation of phospholipase C, but to high specific radioactivity with [3H]inositol, ATP had no re- this only occurred at concentrations 100-fold greaterthan 11832 Nucleotide Uridine Receptor

e ' TABLEI1 Iri,(I)P ' CONTROL / Effect of nucleotides on inositol phosphate formation in C6-2B cells I".(l.OPZ [3HlInositol-prelabeledcells were incubated for 20 min in the pres- ence of the indicated additions (300 p~ each). Inositol phosphates were measured as described under "Experimental Procedures." The data rep- resent the mean f S.E. fromthree to nine different experiments assayed with triplicate determinations. Addition cpm

None 1994 2 167 UDP 6229 -c 376 5-BrUTP 5725 2 333 UTP 5415 f 162 dTDP 4166 f 458 d5-BrUTP 3839 f 179 dTTP 3568 2 160 UDP-glucose 3317 f 292 dUDP 3210 f 360 dUTP 2127 f 73 UMP 2201 f 28 30 40 50 60 70 80 90 LOO 110 120 130 140 Uridine 1905 f 11 5'-tetraphosphate 2198 12 ELUTIONTIME, minutes f UTP 2',3'-dialdehyde 2077 2 38 FIG.2. Effect of UDP and ATP on accumulation of individual UDP 2',3'-dialdehyde 1983 235 [%Ilinositol phosphates. [3HlInositol-prelabeledC6-2B glioma cells XTP 2074 f 116 were challenged for 5 min with vehicle or with 300 p~ UDP or ATP. ATP 1885 f 26 Individual inositol phosphates were separated by HPLC and identified ATPyS 1903 f 142 by co-elution with authentic standards asindicated. The data are rep- 2MeSATP 2136 f 107 resentative of results obtained from two separate experiments camed 2MeSADP 2008 f 103 out in duplicate. ADP@ 1956 f 130 ADP 2110 f 91

"7 a$-Methylene ATP 2013 f 19 CTP 2098 104 U CDP 1914 f 176 100 GTP 2183 f 77 GTPyS 1933 f 28 GDP 2092 f 121 75 dTTP dATP 1900 e 42 V dUDP dGTP 1959 2 272 0 dCTP 1678 29 .e 50 0 Carbachol 8481 f 922

25 purinergic receptors were tested for comparison. As is illus- trated in Fig. 3, right, ATP and UTP were effectiveagonists in 0 HT29 cells under conditions where, in C6-2B cells, they were 4,/' ' ' ' ' Y/'I ' 1 ' J 0 -7 -6 -5 -4 -3 0 -7 -6 -5 -4 -3 either less potent than UDP (UTP) or were without effect (ATP). UDPwas approximately 100-fold less potent than UTP log [NUCLEOTIDE], M in HT29 cells. FIG.3. Concentration-dependent stimulationof inositol phos- Simultaneous challenge of C6-2B cellswith UDP and carba- phate formation by nucleotides in C62B rat glioma and HT29 human colon carcinoma cells. [3HlInositol-labeledcells were incu- chol resulted in inositol phosphate accumulation that was es- bated for 20 min in thepresence of the indicated agonists. kf?,data for sentially the sum of the accumulations observed in the pres- C6-2B glioma cellsare presented. Individual inositol phosphate concen- ence of either agonist alone (Fig. 4). In contrast, simultaneous tration-response curves represent the mean of three to six different challenge of cells with UTP and UDP resulted in no greater experiments performed with triplicate samples each. The data areplot- ted as thepercentage of the maximal response obtained in each experi- inositol phosphate accumulation than accumulation observed ment with 100 p~ UDP whichranged from 7,000to 21,000 cpm.Right, with either agonist alone, suggesting that these compounds results for HT29 human colon carcinoma cells are presented. The data activate the same receptor. ATP alone had no effect on inositol are the mean of triplicate determinations and are representative of phosphate accumulation and no significant effect of ATP was results obtained in threeexperiments. Error bars were omitted for the observed when combined with UDP, UTP, or carbachol (Fig. 4). clarity of the figure. Cross-desensitization experiments also were carried out to examine whether ATP interacts with the receptor that is acti- stimulatory concentrations of UDP. CTP and GTP had no effect. vated by UDP and UTP. These experiments took advantage of A wide range of compounds previously shownto be agonists at the fact that receptor-stimulated accumulation of total L3H1i- Pm-, PZy-,Pm-, and P,,-purinergic receptors were tested. None nositol phosphates is negligible in the absence of LiCl. Thus, of these compounds stimulated inositol lipid hydrolysis (Table cells were preincubated, i.e. desensitized, with agonist in the 11). For example, the potent P,,-receptor agonist 2-MeSATP, the absence of LiCl and inositol phosphate accumulation in re- potent P,,-receptor agonist a,P-MeATP, the potent P,,-receptor sponse to a second drug challenge was assessed by co-addition agonist 2-MeSADP, and the potent P,,-receptor agonist ATPyS of agonist and LiCl. Preincubation of C6-2B cells with UDP or all had no effect on inositolphosphate accumulation. Similarly, UTP for 20 min resulted in a marked decrease in subsequent the P,-purinergic receptor agonist 2-MeSATP had no effect on responsiveness to both UTP and UDP (Fig. 5) but a much individual inositol phosphates resolved by HPLC after chal- smaller decrease in response to carbachol. In contrast, prein- lenge of cells for 20 s or 20 min (data not shown and Ref. 20). cubation of cells with ATP had no inhibitory effect on subse- Finally, inositol phosphate responses in a cell line, HT29 hu- quent responsiveness to carbachol, UDP, or UTP. Taken to- man colon carcinoma cells, previously shown to express P,,- gether with the additivity experiments, these results indicate Uridine Nucleotide Receptor 11833

il .. 0 -7 -6 -5 -4 +++- +++ log [UDP], M FIG.6. Inhibition of UDP-stimulated inositol phosphate forma- suramin. were 5 FIG.4. Additivity of nucleotide- and carbachol (CARB)-stimu-tion by [3HlInositol-labeled cells preincubated for lated inositol phosphate formation in C6-2B cells. [3HlInositol- min without or with the indicated concentration of suramin and chal- labeled cellswere incubated withthe indicated agonists or combination lenged with UDP foran additional 15 min. Theresults are plotted a8 the of agonists (300 p~ each) for 20 min, trichloroacetic acid (5%) was percentage of the maximal response obtainedwith UDP at each given added, and inositol phosphate formationwas quantitated as indicated suramin concentration. The data represent the mean 2 S.D. of results under "Experimental Procedures." The amount of [3Hlinositol phos- obtained in four different experiments. phates present at t = 0 (52300 cpm) was subtracted from the indicated values. The data are the mean z S.D. of triplicate determinations. *, a noncompetitive inhibition of UDP responses, as judged by differed significantly(p < 0.05) from values determined with carbachol decreases of the maximaleffect of UDP withnegligible changes alone. of the concentrationof agonist producing half-maximal effects (data not shown). OUDP 120 Receptor-mediated regulation of phospholipase C by a broad mUTP range of extracellular stimuli has been shown to involve G- [ZZCARBACHOL . At least two general mechanisms are apparently in- volved, one involving members of the pertussis toxin-insensi- 1 tive G, family of G-proteins (24) and one involving pertussis toxin-sensitive Gi or Go (25-27). C6-2B glioma cells were incu- bated with various concentrations of pertussis toxin and re- sponsiveness to carbachol and UDP were subsequently meas- ured. Pretreatmentof cells with pertussistoxin resulted inonly small (0-20%) decreases incarbachol-stimulated inositol phos- phate accumulation (Fig. 7, A and B). In contrast, UDP- and UTP-stimulated inositol phosphate accumulation wasreduced t by up to 80% by pretreatment of C6-2B cells with maximally t effective concentrations of pertussis toxin. The effect of pertus- Preincubation UDP UTP ATP sis toxin was observed as a decrease in responseto maximally FIG. 5. Desensitization of agonist-stimulated inositol phos- effective concentrations of UDP with little or no change in phate formation in CS2B cells. [3HlInositol-labeled cellswere pre- apparent potency (Fig. 7B 1. incubated for 20 min with 300 p~ UDP, UTP, or ATP or in the absence of any drug. The cells were rapidly washed and incubated for an addi- The data with pertussis toxin-treated cells suggested that tional 10 min with 300 p~ carbachol (wide strips),300 p~ UDP (open the receptor for UDP coupled to a G-protein to regulate phos- bar),or 300 p~ UTP (narrow stripes)in the presence of 10 m~ LiC1. The pholipase C. This possibility was examined directly by measur- results are plotted as the percent of agonist responses obtained with ing inositollipid hydrolysis in streptolysin-0-permeabilized cells that were preincubated in the absence of any added drug. Data represent the mean S.D. of triplicate determinations. *, p < 0.01. C6-2B cells. GTPyS produced a concentration-dependent in- crease ininositol phosphate formation (Fig.8). Addition of UDP alone had no effect on activity. However, incubation of that UDP and UTP actat a common receptor at which ATP is membranes with 100 p~ UDP in the presence of increasing not an effective . concentrations of GTPyS resulted ain marked stimulationover Reactive blue and suramin have beenproposed to competi- the effect observed withguanine nucleotidealone. The for tively inhibit adenine nucleotide receptors in a number of tis- UDP for augmentation of inositol phosphate formation in the sues (3). Experiments were carried out to examine theeffect of presence of 30 p GTPyS was approximately 20 p (data not reactive blue and suramin on UDP-stimulated inositol phos- shown). phate formation in C6-2B cells. Suramin (3-50 p)caused a The data presented thus far indicatea novelthat receptor for parallel shift to the right of the concentration-effect curve of uridine nucleotides activates phospholipase C in C6-2B glioma UDP without affecting maximal response to UDP (Fig.6; data cells. The possibility that this receptor also activates another not shown). Because suramin concentration inexcess of 50 p second messenger signaling response wasconsidered. Phospho- alsoresulted in inhibition of carbachol-stimulated inositol lipid pools were prelabeled with [3Hlarachidonateas described phosphate formationin C6-2B cells (data notshown), a detailed under "ExperimentalProcedures." Agonist-promoted release of Schild analysis of the effect of suramin on UDP response was [3Hlarachidonic acid, supposedly occurringas a consequence of not performed. Unlike suramin, reactiveblue (3-50 p)caused activation of phospholipase 4,was then studied. Incubationof 11834 Uridine Nucleotide Receptor

,, 0 CARBACHOL 0 CONTROL 120 t A‘ ’ 0 UDP IB

0.0 0.01 1.0 100 0.0 1.0 10 100 1000

[PTXI, ng/ml [UDP], yM FIG. 7. Effect of pertussis toxin (PTX)on inositol phosphate formation.C6-2B cells were labeled for18 h with 5 pCi of [3H]inositolin 0.5 ml of inositol-freeDMEM containing the indicated concentrations of pertussis toxin (panel A) or with 30 ng/d pertussis toxin (panel B).The data are the mean 2 S.D. of triplicate determinations. Panel A, the effects of 300 p~ carbachol, UDP, and UTP on inositol phosphate accumulation were determined. The data are plotted as the percent of maximal stimulation observed with each drug in control cells. Panel B, the response to the indicated concentrations of UDP was quantitated in control cells or cells pretreated for 18 h with 30 ng/ml pertussis toxin. *, p e 0.05.

4/ ’ I I 4- 0 -7 -6 -5 -4 -3 0 1.0 3.0 10 30 100 log [NUCLEOTIDE], M [GTPyS], PM FIG.9. Receptor-stimulated [W]arachidonic acid release in FIG. Guanine nucleotide-dependent inositol lipid hydrolysis 8. C&2B glioma cells. [3HlArachidonic acid-prelabeled cells wereincu- in streptolysin-0-permeabilizedCB2B glioma cells. C6-2B cells were permeabilized with streptolysin-0 and [3Hlinositolphosphate ac- bated for 10 min with the indicated additions. [3HlArachidonic acid cumulation was measured as described under “Experimental Proce- released into the medium was quantitated as described under “Experi- dures.” Assays were in the presence of the indicated concentrations of mental Procedures.” The data are themean * S.D. of results obtained with triplicate determinations and are representative of results ob- GTPyS with (0)or without (0)100 p~ UDP. The data are the mean 2 S.D. of triplicate determinations and the results are representative of tained in four separate experiments. those obtained in three separate experiments. *, p e 0.01. crease in inositol phosphate accumulation (data for 5 clones are cells with UTP, UDP, or carbachol (Fig. 9; data not shown) all presented in Fig. 10, left). The inositol phosphate response to resulted in a concentration-dependent release of [3H]arachi- ATP was negligible in all of these subclones. donic acid. The potency of these agonists was similar to that C6-2B rat glioma cells represent a subclone that was origi- observed in measurements of [3Hlinositol phosphate accumu- nally isolated from C6 glioma cells by de Vellis and Brooker lation. Negligible increases in [3H]arachidonicacid release (28). Inositol phosphate responses to nucleotidesalso were were observed in thepresence of ATP. tested in C6 glioma cellsthat were obtained from the American Experiments were carried out to determine whether the uri- Type Culture Collection. In contrast to the lack of response dine nucleotide-selective response could be observed in cells observed with C6-2B cells, ATP and ATPyS markedly stimu- expanded from single cells cloned from the C6-2B cell line as lated inositol phosphate accumulation in C6 glioma cells (Fig. described under “Experimental Procedures.” Eight different 10, right), and the pharmacological specificity forenhancement subclones were isolated and theirinositol phosphate responses of inositol lipid hydrolysis in C6 glioma cells (ATP = UTP > measured. Seven of these subclones (Clones1,2,3,4,6,7, and ATPyS >> UDP) was similar to that observed in HT29 human 8) were morphologically similar to the parent cells that were colon carcinoma cells (Fig. 3, right) and for P,,-purinergic re- previously studied in detail. Clone 5 was markedly different ceptors in many target cells. Carbachol, whichmarkedly stimu- morphologically. These cells were muchsmaller and possessed lated inositol phosphate accumulation in C6-2B glioma cells many long processes. However, all eight of the subclones ex- (Fig. 11, had no reproducibleeffect on accumulation in C6 pressed receptors that responded to UDP with a marked in- glioma cells (Fig. 10, right). Uridine Nucleotide Receptor 11835 resulted in enhancedactivity of the ana- -f/ I T,' 1 C62B cells C6 ATCC cells logs. These results suggest thatcompounds such as 5-BrUDP or 0 ATP 5-MeUDP will be very potent agonists at this receptor. .UTP Specific antagonists for receptors for extracellular nucleoti- des have not been identified. However, suramin and reactive blue, which have a broad range of pharmacological activities A CARBACHOL previously have been shown to competitively block P,-puriner- gic receptors in a number of tissues (3).Suramin also blocks the receptor for uridine nucleotides on C6-2B glioma cells, al- though at higher concentrations an apparently nonspecific in- hibition of the inositol lipid hydrolysis occurred. Effects of extracellular uridine and nucleotides have been reported previously. These responses range from relax- ation or contraction of various smooth muscle preparations to inhibition of 0, uptake, stimulation of glucose output, and en- I I -4/ ' 4/ ' hancement of superoxide formation (16). Difficulties inherent 0 -6 -5 -4 -3 0 -6 -5 -4 -3 in defining receptors for extracellular nucleotides on heteroge- -log [AGONIST], M neous tissue preparations have compromised classification of these responses, but it is likely that many of these effects are FIG. 10. Effect of UDP and ATP on inositol phosphate forma- mediated by P,,-purinergic receptors. That is, a receptor that is tion in subclones of C6-2Bcells and in C6 glioma cells obtained from the American Type Culture Collection.Left, the effect of UDP activated by both UTP and ATP is involved. However, PZu- (filled symbols)and ATP (open symbols) on inositol phosphate forma- purinergic receptors apparently do not completely account for tion was determined with C6-2B cell populations expanded from five the response to uridine nucleotides in peripheral tissues and different cloned single cells as detailed under "Experimental Proce- dures." The data are the mean of triplicate determinations and are Seifert and Schultz (16) haveproposed the existence of a sepa- representative of results fromtwo different experiments with each rate class of pyrimidinergic receptors. The pharmacological se- clone. Right, the effect of ATP, UTP, UDP, ATP+, and carbachol on lectivities of responses in several vascular smooth muscle inositol phosphate accumulation was determined in C6 glioma cells preparations were not the sameas those of the P,,-purinergic obtained from the American weCulture Collection. receptor (10-12), and in severalcases the response also can be distinguished from that observed in C6-2B glioma cells since DISCUSSION UMP, uridine, and CTP all were shown to produce vascular A receptor that has notbeen described previously exists on responses similar to those of UTP or UDP (29-33). Support for C6-2B rat glioma cells. This receptor, which stimulates inositol the existence of a receptor for UTP thatis not activated by ATP lipid hydrolysis, shows a strict selectivity for uridine nucleoti- also has come from studies inwhich responses to UTP andATP des and is activated only nominally or not at all by ATP, ADP, in a heterogeneous tissue could be differentially blocked by or analogs of these adenine nucleotides. This receptor can be pretreatment withsupposed antagonists andby experiments in clearly differentiated from the P,,-purinergic receptor, which is which cross-desensitization did not occur between the two found on many target tissues. The C6-2B cell receptor is acti- nucleotides (10, 11, 34). vated by UDP, and UTP is a less potent full agonist. ATP and The data obtained with C6-2B glioma cells places the exist- ATPyS have little or no effect. In contrast, ATP, ATPyS, and ence of a selective receptor for uridine nucleotides on firm UTP are themost potent agonists at P,,-purinergic receptors; ground. With the pharmacological specificity defined here in UDP has been shown to be much less potent than UTP(Fig. 3, hand, it will be important to more clearly delineate thephysi- right, and Ref. 5) or to be without effect (8) at these receptors. ological role of this receptor in various tissues. These results Several different types of experiments including those examin- together with the large amountof existing information on P2"- ing additivity of responses or cross-desensitization also con- purinergic receptors indicate that there areat least two differ- firmed that the receptor on the C6-2B glioma cells does not ent receptors that respond to extracellular uracil nucleotides. recognize adenine nucleotides. Relatively little is known about the regulation of the storage A full delineation of the structure-activity relationships for andrelease of these nucleotides, althoughplatelet storage nucleotide-activated inositolphosphate formation in C6-2B cells granules are a rich source of UTP and UTP is released upon was limited by the lack of availability of a wide range of py- platelet activation(16, 35). Resting levels of extracellular UTP rimidine nucleotides. UDP was the most potent and effective of in solid tissues are in the range of 1 x M (16, 36). A multi- the nucleotides studied. Thepresence of a y-phosphate, i.e. UTP, plicity and widespread existence of receptors for uridine nucle- resulted ina reduced potency of approximately 30-fold. Replace- otides and the presence of nucleotides at concentrations ca- ment of the y-phosphate of UTP withglucose (UDP-glucose) re- pable of activating thesereceptors suggests that the regulated sulted in furtherdecreases of both potency and efficacy. How- release of uridine nucleotides will prove to be a physiologically ever, increased potency relative to UTP was observed with important process. substitution of bromo at position 5 of the pyrimidine ring of UTP The second messenger responses regulated by the receptor (5-BrUTP). In contrast, substitutionof a keto group for the -NH, for uridine nucleotides on C6-2B glioma cells followed the gen- at position 4 (CTP, CDP) resulted in total loss of activity. 2'- eral properties observed for many receptors includingthe mus- Deoxynucleotides have no effect on P,,-purinergic receptors (8), carinic cholinergic receptor on the same cells. Streptolysin-0- and similarly, the receptor on C6-2B cells was not activated by permeabilized cells were used to illustrate directly that this dUTP at any concentration tested up to1 m~. However, as was receptor promotes guanine nucleotide-dependent activation of observed with 5-BrUTP, 5-bromo substitution of dUTP (2'-de- phospholipase C. The large sensitivity of the intact cell re- oxy-5-BrUTP) resulted in an increase activity.in Some activity sponse to inactivation by pertussis toxin places these receptors (-50% of maximal stimulation observed with UDP) was ob- into the subclass of receptors that couple to phospholipase C served with d"P and dTDP, which can be considered as through Gi or Go (25-27). Several recent reports suggest that 5-methyl analogs of dUTP and dUDP, respectively. Thus, sub- activation of phospholipase C in this pathwayoccurs as a con- stitution at position 5 of the pyrimidine ring with bromo or -CH, sequence of release of &-subunits from the involved G-protein 11836 Uridine Nucleotide Receptor (3740). Our assumption is that theactivation of arachidonate Raven Press, New York release by this uridine nucleotide receptor is phospholipase 2. Gordon, J. L. (1986) Biochem. J. 233,309-319 3. Dubyak, G. R., and El-Moatassim, C. (1993) Am. J. Physiol. 286, C577-42606 &-mediated and occurs secondarily to activation of phospho- 4. Dubyak, G. R., Cowen, D. S.,and Meuller, L. M. (1988) J. Eiol. Chern. 263, lipase C with subsequent Ca2+mobilization and activation of 18108-18117 protein C; however, this has not been formally proven. 5. Fine, J., Cole, P., and Davidson, J. S. (1989) Biochem. J. 263,371-376 6. Okajima, F., Kioichi, J. S., Nazarea, M., Sho, K, and Kondo, Y. (1989) J. Bid. In addition to the uridine nucleotide-selective receptor, we Chern. 264, 13029-13037 have observed previously that activation of a P,-purinergic 7. Stutchfield, J., and Cockmtt, S. (1990) FEBS Lett. 262,256-258 receptor on C6-2B glioma cells results in inhibition of adenyl- 8. Brown, H. A., Lazarowski, E. R., Boucher,R. C., and Harden, T. K (1991) Mol. Pharmacol. 40,648455 ylcyclase and a decrease in intracellularcyclic AMP levels (20). 9. PfeilschiRer, J. (1990) Biochem. J. 272.469-472 This receptor does not activate phospholipase C and thus is 10. von Kugelgen, I., Haussinger, D., and Starke, K. (1987) Naunyn-Schmiede- apparently different from the P,,-purinergic receptods) found berg's Arch. Phannacol. 338, 556-560 11. von Kugelgen, I., and Starke, K. (1990) Naunyn-Schmiedebeg's Arch. Phar- on many target tissues which activates phospholipase C but macol. 341,538-546 does not regulateadenylylcyclase activity. Lin and Chuang(41) 12. Ralevic, V., and Burnstock, G. (1991) Circ. Res. 69,1583-1590 have reported thatATP stimulates inositol phosphate andCa" 13. Haussinger, D., Stehle, T., and Gerok, W. (1987) Eur: J. Biochem. 167,65-71 14. Seifert, R., Wenzel, K., Eckstein, F., and Schultz, G. (1989) Eur J. Biochem. accumulation in C6 glioma cells, and we have confirmed their 181,277-285 results with C6 glioma cells obtained from the American Type 15. Seifert, R., Burde, R., and Schultz, G. (1989) Biochem. J. 269,813-819 Culture Collection (Fig. right). In contrast, we have been 16. Seifert, R., and Schultz, G. (1989) lkd Pharmacol. Sci. 10,365-369 10, 17. Okajima, F., lbkumitsu, Y.,Kondo, Y., and Ui, M. (1987)J. Eiol. Chem. 282, unable to observe stimulation of inositol lipid hydrolysis by 13483-13490 ATP or its analogs in C6-2B glioma cells, which are a subclone 18. Sato, K, Okajima, F., and Kondo, Y. (1992) Biochem. J. 283,281-287 19. Yamada, M., Hamamori, Y., Akita, H., and Yokoyama, M. (1992) Circ. Res. 70, originally isolated from the RGC6 glioma cell line by deVellis 477-485 and Brooker (28). We also isolated eight additional subclones 20. Boyer, J. L., Lazarowski, E. R., Chen, X.-H., and Harden, T. K. (1993) J. from C6-2B glioma cells. All of these expressed a phospholipase Pharmacol. Exp. Thec 287, 1140-1146 C-linked receptor that was activated by UDP but not by ATP. 21. Nakahata, N., and Harden, T. K. (1987) Eiochem. J. 241,337444 22. Hughes, P. J., Hughes, A. R., Putney. J. W., and Shears, S. B. (1989) J. Eiol. Earlier studieson the P-adrenergic receptorkyclicAMP system Chem. 264, 19871-19878 had revealed differences among the signalingproperties of the 23. Lazarowski, E. R., Boucher, R. C., and Harden, T. K. (1994) Am. J. Physiol. 268, C40W415 various C6 cell (including C6-2B cells) subclones (42451,and 24. Harden, T. K. (1992) in Advances in Second Messenger and Phosphoprotein although carbachol markedlystimulated inositol phosphate ac- Research (Putney, J. W., Jr., ed) pp. 11-33, Raven Press, Ltd., New York cumulation in C6-2B glioma cells, no reproducible effect of mus- 25. Harden, T.K. (1989) in Inositol Lipids in Cell Signalling (Michell, R. H., Drummond, A. G., and Downes, C. P., eds) pp. 113-134, Academic Press, carinic receptor stimulation was observed in C6 cells obtained London from the American Type Culture Collection. 26. Martin, T.F. J. (1989) in Inositol Lipids in Cell Signalling (Michell, R. H., In summary, a phospholipase C-activating receptor for UDP Drummond, A. H., and Downes, C. P., eds) pp. 81-107, Academic Press, London and UTP hasbeen identified on C6-2B rat glioma cells. Based 27. Meldrum, E., Parker, P. J.,and Carozzi,A. (1991)Eiochirn.Biophys. Acta 1092, on the properties of second messenger responses, this receptor 49-71 must be structurally related to thesuperfamily of seven trans- 28. de Vellis, J., and Brooker, G. (1973) in !lissw Culture of the Nervous System (Sato, G., ed) pp. 231-245, Plenum, New York membrane-spanning G-protein-linked receptor proteins. The 29. MacDonald, G., Assef, R., Cui&, A., and Lo, E. (1984)Clin. Exp. Pharmacol. recently cloned P2,- (46) andP,,,-purinergic receptors (47) have & Physiol. 11,381-384 sequences that predict seven transmembrane span- 30. Hardebo, J. E., Hanko, J., and Owman, C. (1983) Gen. Phannacol. 14,133-134 31. Hardebo, J. E., Kahrstrom, J., Owman, C., and Salford, L. G. (1987) Blood ning motifs. However, they express a low Vessels 24, 150-155 with each other and suggest the existence of two new sub- 32. Shirasawa, Y., White, R. P., and Robertson, J. T. (1983) Stroke 14, 347-355 classes of G-protein-linked receptors. Our anticipation is that 33. Urquilla, P. R. (1978) Stroke 9, 132-136 34. Sakai, K., Akima, M., and Mataushita, H. (1979) Jpn. J. Pharrnacol. 29,243- the receptor described here will fall structurally into a class of 251 receptors with structures containing only low sequence homol- 35. Goetz, U.,DaPrada, M., and Pletacher,A. (1971) J. Pharmacol. Exp. Ther: 178, 210-215 ogy to PZypurinergic receptors and perhaps somewhat higher 36. Keppler, D., Rudiger, J., and Decker, K (1970)Anal. Biochem. 38, 105-114 homology to theso-called P,,-purinergic receptods) thatis ac- 37. Camps, M., Hou, C., Sidiropoulos, D., Stock, J. B., Jakobs, K. H., and Gier- tivated by both UTP andATP. Since only uridine nucleotides are schik, P. (1992) Eur: J. Biochem. 206,821-831 38. Boyer, J. L.,Waldo, G. L., and Harden, T.K. (1992) J. Biol. Chem. 267, agonists at the receptor studied here,we propose that it be called 25451-25456 a uridine nucleotide receptor. Molecular cloning of this uridine 39. Blank, J. L., Brattain, K. A,, and Exton, J. H. (1992) J. Bid. Chem. 257, nucleotide receptor will be required to establish whether is it a 23069-23075 member of a new class of receptor proteins or is a second member 40. Katz, A,, Wu, D., and Simon, M. I. (1992) Nature 360,686-689 41., Lin, W.-W., and Chuang, D.". (1993) Neurochem. Res. 16,681487 of the subfamily of P,,-purinergic receptor-like proteins. 42. de Vellis, J., and Brooker, G. (1974) Science 186, 1221-1223 43. Homburger, V., Lucas, M., Cantau, B., Barabe, J., Penit, J., and Bockaert, J. Acknowledgements-We are gratefulto Johnny Obie and Jim Putney (1980) J. Biol. Chem. 265.1043f5-10444 for HPLC analyses of inositol phosphates, to Drs. Jose L. Boyer and 44. Fishman, P. H., Mallorga, P., and Tallman, J. F. (1981) Mol. Phnrmacol. 20, Robert Nicholasfor helpful comments, andto John OTuel for typing the 310-318 45. Frederich, R. M., Waldo,G. L., Harden, T. K., and Perkins, J. P. (1983)J. Cyclic manuscript. Nucleotide Res. 9, 103-118 46. Lustig, K. D., Shiau, A. K, Brake, A. J., and Julius, D. (1993) Pmc. Natl. Acad. REFERENCES Scz. U.S. A. 90, 5113-5117 1. Burnstock, G. (1978) in Receptors for Drugs and Hormones: A 47. Webb, T. E., Simon, J., Krishek, B. J., Bateson, A. N.. Smart, T. G.,King, B. F., Multidisciplinary Approach (Straub, R. W., and Bolis, L., eds) pp. 107-118, Burnstock, G., and Barnard, E. A. (1993) FEES Lett. 324,219-225