Molecular Mechanism of Basic Calcium Phosphate Crystal-induced ...

5 downloads 0 Views 5MB Size Report
Peter G. Mitchell$$, W. J. Pledgerll, and Herman S . CheungS. From the .... grown in Dulbecco-Vogt modified Eagle's medium supplemented with. 10% (v/v) calf ...... Kelly, K., Cochran, J. H., Stiles, C. D., and Leder, P. (1983) Cell. 47. Davis ...
Vol . 264, No. 24, Issue of August 25, pp 14071-14077, 1989 Printed in U.S.A.

THEJOURNALOF BIOLOGICAL CHEMISTRY

0 1989 by The American Society for Biochemistry and Molecular Biology, Inc

Molecular Mechanismof Basic Calcium Phosphate Crystal-induced Mitogenesis ROLEOFPROTEINKINASE

C* (Received for publication, January 30, 1989)

Peter G. Mitchell$$, W. J. Pledgerll, and HermanS . CheungS From the $Division of Rheumatology, Department of Medicine, Medical College of Wisconsin, Milwaukee, Wisconsin53226 and the TDepartment of Cell Biology, Vanderbilt University School of Medicine, Nashville, Tennessee37232

Synovial tissue hyperplasia in basic calcium phos- phate, and calcium pyrophosphatedihydrate.Clearance of phate deposition diseases has been suggested to develop these crystals occurs by phagocytosis followed by dissolution through the stimulation of cell growth by basic calcium within the synovial cells (3, 4). It has been suggested that phosphate(BCP) crystals deposited injoints.These abnormal proliferation of synovial cells may be due to intracrystals have been used in vitro to stimulate DNA cellular dissolution of the crystals, producing an increased synthesis in quiescent fibroblasts to experimentally cytoplasmic calcium concentration which activates a pathway study this proliferative disease. The stimulation of leading to a mitogenic response (5). BCP crystals have been DNA synthesis, in density-arrested Balb/c 3T3 cells, shown to stimulate mitosis of cultured human skin fibroblasts by BCP crystals was inhibited after down-regulating and/or caninesynovial fibroblasts ina concentration-dependproteinkinaseCactivitywith12-0-tetradecanoylent fashion (6). The mechanismwhereby BCP crystals stimphorbol 13-acetate (TPA). No effect on platelet-deulate cellular proliferation has notbeen described, but it has rived growth factor (PDGF)-stimulated DNA synthesis was observed under the same conditions. The expres- been proposed that BCP crystalsmay stimulate processes in sion of c-myc and c-fos increased in response to BCP common with PDGF stimulation of proliferation (7). PDGF renders quiescent fibroblasts competent to respond to prostimulation in a manner similar to the increase progrowth factor I,necessary duced by stimulation withPDGF. The BCP stimulation gression factors, such as insulin-like of c-fos and c-mycmessages was inhibited60 and 90%, for progression through the cell cycle (8). BCP crystals also as competence factors because the presence respectively,in TPA-pretreated,proteinkinase C- appear to function down-regulated cells. The induction of these tran- of insulin-like growth factor I is necessary for the full mitoscripts by PDGF was unaffected in cells pretreated genic response (7). with TPA. TPA was unable to stimulate c-fos and cStimulation of density-arrested fibroblastswith PDGF elicmyc expression or DNA synthesis following protein its a diverse range of early responses. These include PDGF kinase C down-regulation. Both PDGFand TPA stim- receptor autophosphorylation (9), activation of a phosphatiulated phosphorylation of an 80-kDaprotein, whereas dylinositol phospholipase C (lo), a decrease in the numberof BCP crystals hadno effect on phosphorylation of this high affinity EGF receptors(11-13), a rapid increase in intraprotein. The exposure of density-arrested Balb/c 3T3 cellular calcium (14-17), a n increase in ion fluxes across the cells to BCP crystals had no effect on high affinity epidermal growth factor receptor binding under con- membrane (reviewed in Ref. 18), and the induction of transcription of the nuclear proto-oncogenes c-fos (19-21) and cditions in which PDGF and TPA reduced epidermal growth factor binding. The data suggest that PDGF myc (22, 23). Activation of phospholipase C leads to produccan actto stimulate c-fos and c-myc expression as well tion of diacylglycerol and inositol trisphosphate(10). Inositol as DNA synthesis through a protein kinase C-inde- trisphosphate stimulates therelease of calcium from intracellular sources and diacylglycerol binds to protein kinase C pendent pathway, whereas BCP crystals require at least endogenous levels of protein kinaseC to stimulate (PKC)suchthatthenetresultistheactivation of this these events. calcium- andphospholipid-dependentproteinkinase(reviewed in Ref. 24). An indicator of PKCactivityisthe phosphorylation of a n acidic 80-kDa cytosolic protein (25, 26). Both PDGF and the tumor-promoting phorbol esters Synovial cell hyperplasia is a feature of the chronic syno- have been shown to phosphorylate this protein in densityvitis associated with BCP’ crystal deposition in joint tissues arrested fibroblasts (25). Phorbol esters such as phorbol 12(1, 2). BCP crystals include hydroxyapatite, calcium phos- 0-tetradecanoylphorbol 13-acetate (TPA) can substitute for diacylglycerol and binddirectly to PKC(24). Down-regulation * This work was supported by National Institutes of Health Grants of PKC by chronic treatment with phorbol esterswill effecCA 42713 (to W. J. P.) and AR38421 (to H. S. C.) and Research tively inhibit PDGF-mediated phosphorylationof the 80-kDa Cancer Development Award AM 1065 (to H. S. C.). The costs of publication of this article were defrayed in part by the payment of protein (26, 27). However, a number of other PDGF-stimupage charges. This article must therefore be hereby marked “aduer- lated responses are unaffected by PKC down-regulation in tisement” in accordance with18 U.S.C. Section 1734 solely to indicate fibroblasts. These include phosphorylation of proteins of 22 this fact. and 31 kDa (26), activation of the ribosomal S6 kinase (28), 3 To whom correspondence should be addressed. a reduction in high affinity EGF binding(13,29), EGF recepThe abbreviations used are:BCP, basic calcium phosphate; of DNA synthesis PDGF, platelet-derived growth factor; EGF, epidermal growth factor, tor phosphorylation (29), and the induction been reported that downPKC, protein kinase C; TPA, 12-0-tetradecanoylphorbol13-acetate; (13, 27, 30). In contrast, it has Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid. regulation of PKC in fibroblasts substantially reduces the

14071

14072

Molecular Mechanism of BCP Crystal-induced Mitogenesis

PDGF-induced induction of ornithine decarboxylase activity (31) and thePDGF-stimulated increase in the c-myc (27, 3234) and c-fos (32, 35) messages. The purpose of this work wasto investigate the mechanisms of PDGF and BCP crystal-stimulated mitogenesis in relation to PKCactivity. Our data shows a dependence on the presence of PKC activity for BCP crystal-induced mitogenesis. In addition, substantial reductions in both c-fos and c-myc induction were found after BCP crystal stimulation of PKCdeficient cells. However, neither EGF receptor down-regulation nor PKC activation could be demonstrated after BCP stimulation, suggesting that BCPcrystals were acting through basal levels of PKC activity in Balb/c 3T3 cells. In contrast, PDGF caused activation of PKC, although PKC down-regulation had no effect on either PDGF-stimulated mitogenicity or c-fos or c-myc induction, thus suggesting that PDGFcould act via a PKC-independent pathway.

gates, sterilized by heating at 200 "C for 90 min, weighed, and suspended in medium. The suspensions were sonicated before use. Materials-Highly purified PDGF was prepared as described (43). EGF was purified and iodinated according t o the methods of Savage and Cohen (44) and Hunter and Greenwood (45), respectively. The pfos-1 and pSVc-myc-1 plasmids were kindly provided by I. Verma (Salk Institute, San Diego, CA) (40) and R. A. Weinberg (MassachusettsInstitute of Technology, Cambridge, MA)(41), respectively. [3H]Thymidine and [a-32P]dCTPwere from Amersham Corp. Probes were randomly primed with kits from Boehringer Mannheim. TPA was obtained from Sigma. Carrier-free 32Pi(8 mCi/ml) was purchased from Amersham Corp. RESULTS

Kinetics of BCP Stimulation of DNA Synthesis in Balblc 3T3 Cells-BCP crystals have been shown to stimulate DNA synthesis in human foreskin fibroblasts and canine synovial fibroblasts (6). We investigated the mitogenic response of Balb/c 3T3 cells to BCPcrystals. Exposure of quiescent Balb/ EXPERIMENTALPROCEDURES c 3T3 fibroblasts to medium containing various concentraCell Culture-stock cultures of Balb/c 3T3 cells (clone A31) were tions of BCP crystals and 5% platelet-poor plasma caused a grown in Dulbecco-Vogt modified Eagle's medium supplemented with dose-dependent increase in DNA synthesis as determined by 10% (v/v) calf Serum, 4 mM L-glutamine, penicillin at 50 units/ml, [3H]thymidine incorporation in cells harvested after 48 h of and streptomycin at 50 pg/ml in humidified 5% COz, 95% air at 37 "C. For all experimental procedures,cells were grown to confluency incubation (Fig. 1).The stimulation of DNA synthesis plain serum-containing medium and used 2-3 days after growth cessa- teaued between 20 and 50 pg/ml of BCP crystals; therefore, 50 pg/ml of BCP crystals was used in subsequentexperiments. tion. EGF Binding-The procedure used was essentially that of Olashaw Similar data were obtained after a 24-h incubation. The et al. (13) in which cells in 35-mm culture dishes were transferred to quantity of BCP crystals required for stimulation of DNA 0.5 mlof binding medium (Hanks' balanced salt solution supplemented with 0.2% bovine serum albumin and 30 mM Hepes, pH 7.4) synthesisin quiescent Balb/c 3T3 cells is similar to that required to stimulatesynovial cells and human foreskin fibrocontaining '251-labeledEGF (1-2 ng/ml). The cultures were incubated at 4 "C for 2.5 h, rinsed three times with bindingmedium, and blasts (6). solubilized in 0.1 N NaOH containing 1%NaDodSO,. The amountof The time courses of stimulation of DNA synthesis by PDGF cell-bound radioactivity was determined using a y counter and was and BCP crystals were compared in density-arrested Balb/c corrected for nonspecific binding occurring in the presence of EGF 3T3 cells. Fig.2 shows a timecourse comparing [3H]thymidine (1 d n l ) . DNA Synthesis Assay-Cells in 96-well plates were incubated with incorporation of PDGF-stimulated fibroblasts with that of BCP-stimulated fibroblasts. In general, the stimulation of ['Hlthymidine (5 pCi/ml),rinsed with phosphate-bufferedsaline, treated twice for 10 min with methanol, and then rinsed with water. DNA synthesis by BCP crystals was found to vary between The bottom of each well was then punched out, and theincorporated 50 and 90% of that obtained with PDGF (data notshown and [3H]thymidine was measured by scintillation counting. Fig. 3). In this experiment (Fig. 2 ) , the BCPcrystals were not Preparation of RNA-RNA was prepared by scraping the cells of these data revealed from 100-mm plates and washing in phosphate-buffered saline fol- as stimulatory as PDGF. Extrapolation lowed by disruption of the cells in guanidinium isothiocyanate. The similar kinetics of entry into S phase,although the cells RNA was then precipitated in 4 M LiCl as described by Cathala et al. stimulated with BCP crystals lagged by 1-2 h as observed (36). previously (6). Thus, in Balb/c 3T3 cells PDGF and BCP RNABlotAnalysis-RNA (5 pg/lane) was electrophoresed on crystals bring about the initiation of DNA synthesis with

formaldehyde gels and transferred nitrocellulose to filters as described previously (37). The filters were hybridized with inserts purified from plasmids carrying sequences of the desired gene. The inserts were labeled to a specific activity greaterthan 10' cpm/pg usingthe random primer method (38). Protein Phosphorylation-Phosphorylation of proteins in response to various factors was carried out using a modification of the procedure published by Rozengurt et al. (25). Cells were incubated with carrier-free 32P1(200 pCi/ml) in phosphate-free minimum Eagle's medium containing 0.5% platelet-poor plasma, for 3 h. Stimulating agents were added directly to this media and left for specified times. The cultures were then washed with ice-cold Tris saline (0.15 M NaCI, 20 M Tris, pH 7.5) and treated with 5% trichloroacetic acid for 20 min on ice. After washing twice more with ice-cold Tris saline, the precipitated proteins were then dissolved in 200 p l of reticulocyte saline buffer (10 mM Tris, pH 7.4, 10 mM NaC1, 1.5 mM MgCM containing I.% Nonidet P-40. Portions of each sample were trichloracetic acid-precipitated, and sample volumes containing equal counts (generally 50,000 cpm) were loaded and run on 20% polyacrylamide gels according to Laemmli (39). The gels were then dried and exposed to film overnight. Plasmids-The jos insert used as a probe in this work was a 1.0kilobase PstI-v-jos fragment from the pfos-1 plasmid (40). The myc probe was a 4.8-kilobase XbaI-BamHI fragment from the pSVc-myc1plasmid which carries mouse myc sequences (41). Preparation of BCP Crystals-BCP crystals were prepared as described (42). Theywere crushed and sieved to yield 10-20-pm aggre-

pg BCP/ml

FIG. 1. The effects of different concentrations of BCP crystals on ['Hlthymidine incorporation in Balb/c 3T3 cells. Cells in 96-well plates in medium containing 5 platelet-poor plasma and [3H]thymidine were treated withvarying concentrations of BCP crystals for 48 h. Plates were fixed and individual wells counted as described under "Experimental Procedures." Values are the meansof triplicate determination.

Molecular Mechanism of BCP Crystal-induced Mitogenesis

14073

plasma in density-arrested Balb/c 3T3 cells. However, treatment of cells with 200 nM TPA for 20 h,inconditions 140nonpermissivefor cell cycle traverse, prevented TPA and E .. . a BCP crystal stimulation of mitogenesis. In contrast, as ex’?u 120pected, pre-exposure of cells to TPA had no effect on the 0 c 100 PDGF mitogenic response. The data inFig. 3 clearly demon._ strate that pretreatment with TPA abolished the proliferative 80responseinduced by theBCPcrystalsandthusstrongly suggest that BCP crystals promote mitogenesis through a mechanism that has a strict dependence on the presence of PKC. Role of Protein Kinase C in the BCP Crystal Activation of .... , I , , , , , , , , . , ,, c-fos and c-myc-The addition of PDGF or TPA to densityarrested quiescent Balb/c 3T3cells has been shown to rapidly 0 48 12 16 20 24 28 32 36 40 44 48 induce the transcriptionof c-fos and c-myc. In order to deterHours After Addition mine whether PKC wasrequired in early events of BCP FIG. 2. Growth curve of BCP crystal-stimulated and PDGF- crystal-induced mitogenesis, the following experiments were stimulated cultures. Cells in 96-well plates in medium containing performed. We compared the induction of c-fos RNA and c5% platelet-poor plasma and [3H]thymidine(5 pCi/ml) were treated with either PDGF (70 ng/ml) or BCP crystals (50 pg/ml). At various myc RNA by PDGF, TPA, and BCP crystals. Stimulationof times, individual wells were fixed by the addition of 60 p1 of 1 M quiescent fibroblastsby PDGF produced a transient increase ascorbic acid. After 24 h, the plates were rinsedand wells were counted in the c-fos message that peaked at 30 min after stimulation as described under“Experimental Procedures.” Values are the means (Fig. 4B). The levels of this transcript then rapidly declined of triplicate determinations. PDGF; +, BCP. to basal levels. The c-myc message peaked at about 1 h after PDGF stimulation and thendeclined slowly over a period of several hours (Fig. 5B). Figs. 4 and 5 also show the relative levels of the c-fos and c-myc messages after stimulation with either TPA or BCP crystals as compared to PDGF stimulation. BCP crystals and TPA were as effective as PDGF in the induction of c-fos and c-myc mRNA. The kineticsof induction of c-fos and c-myc mRNA were the same for PDGF, TPA, and BCP crystal stimulation, with the exception that the TPA-stimulated c-myc message rapidly declined to basallevels. In similar experiments cells were treated with200 nM TPA to produce PKC-deficient cells before stimulation by PDGF, TPA, and BCPcrystals. Figs. 4 and 5 display Northern blot quantitation of c-fos and c-myc induction by PDGF, BCP crystals, and TPA in TPA-pretreated cells. As can be seen in FIG. 3. Effect of protein kinase C down-regulation on Figs. 4A and 5A, TPA pretreatmentcompletely abolished the PDGF-, TPA-, and BCP crystal-stimulated [‘Hlthymidine in- induction of the c-fos and c-myc messages by an additional no on the ability of PDGF corporation. Control cells in 35-mm plates were incubated in me- treatment with TPA but had effect dium containing 5% platelet-poor plasma (ppp) and [3H]thymidine to increase thesemessages. The induction of c-fos and c-myc with either PDGF (70 ng/ml), TPA (200 nM), or BCP crystals (50 by BCP crystals (Figs. 4C and 5C) was reduced by 60 and pg/ml). TPA-pretreated cells were incubated for 20 h in medium containing 200 nM TPA and 0.25% platelet-poor plasma.At 20 h, the go%, respectively, in PKC-deficient cells. Thus, PKC downmediumwasremoved and replacedwithmedium containing 5% regulation dramatically inhibited both c-fos and c-myc mesplatelet-poorplasmaand[3H]thymidine,and the cultures were sage induction by BCP crystals. Thisreduction in the mRNA treated with the same concentrations of either PDGF, TPA,or BCP level of these two proto-oncogenes correlated with the inhicrystals as in control cells. After 24 h of stimulation, the plates were bition of BCP crystal-induced DNA synthesis in TPA-prewashed and fixed as described under “Experimental Procedures.” The cells were solubilized in1ml of 0.1 M NaOH containing 1%NaDodSOl treated cells (Fig. 3). In order to investigate the role of cytoplasmic Ca2+ in the and 100 pl were counted in a scintillation counter. no pretreatment; induction of c-fos, we used the calcium ionophore A23187 to @, TPA pretreatment. increaseintracellular calcium concentrations. Wedid not observe any increase in c-fos message after the addition of similar kinetics in a manner dependent on plasma suppleA23187 to density-arrested Balb/c 3T3 cells (data not shown). mentation of the medium. BCP Crystals Do Not Alter EGF Binding-Olashaw et al. Protein Kinase C Is Required for BCP Crystal Mitogenic(13) and Wharton et al. (11)have shown in Balb/c 3T3 cells ity-To gain insight into the mechanism of BCP crystaldecrease high affinityEGF binding. stimulated mitogenesis, we examined the potential involve- that both PDGF and TPA It has been suggested that this occurs via activation of PKC ment of PKC. It has beenshownpreviously that chronic treatment of fibroblasts with phorbol esters causes a loss of (24) and the phosphorylation of the EGF receptor (47, 48). PKC activity (46) with concomitant loss of TPA responsive- Because PKC activitywas apparently required for BCP crysness (25,26,46). Although it hasbeen shown that PKC is not tals toinduce c-fos and c-myc and DNA synthesis, we wanted required for PDGF mitogenicity (13, 27, 30), we not present to determine if BCP crystals altered EGF binding. To test evidence that, unlike PDGF, the mitogenic response of Balb/ whether BCP crystals had an effect on EGF receptors, we c 3T3 cells to BCP crystals is dependent on PKC activity. measured EGF binding at various times after stimulation of Fig. 3 shows that PDGF, TPA, and BCP crystals all stimu- quiescent cultures with BCPcrystals. Fig. 6 shows that BCP lated DNA synthesisinthepresence of 5% platelet-poor crystals did not decrease EGF bindingbeyond that caused by -