Dual Regulation of the Parathyroid Hormone (PTH)/ PTH-Related Peptide Receptor Signaling by Protein Kinase C and -Arrestins

Dual Regulation of the Parathyroid Hormone (PTH)/PTH-Related Peptide Receptor Signaling by ProteinKinase C and �-Arrestins

MARIAN CASTRO*, FRANK DICKER*†, JEAN-PIERRE VILARDAGA, CORNELIUS KRASEL,MANFRED BERNHARDT, AND MARTIN J. LOHSE

Institute of Pharmacology and Toxicology, University of Wurzburg, Versbacher Strasse 9, D-97078, Germany

We examined here the role of second messenger-dependentkinases and �-arrestins in short-term regulation of the PTHreceptor (PTHR) signaling. The inhibition of protein kinase C(PKC) in COS-7 cells transiently expressing PTHR, led to anapproximately 2-fold increase in PTH-stimulated inositolphosphate (IP) and cAMP production. The inhibition of pro-tein kinase A increased cAMP production 1.5-fold without af-fecting IP signaling. The effects of PKC inhibition on PTHR-mediated Gq signaling were strongly decreased for a carboxy-terminally truncated PTHR (T480) that is phosphorylationdeficient. PKC inhibition was associated with a decrease inagonist-stimulated PTHR phosphorylation and internali-zation without blocking PTH-dependent mobilization of�-arrestin2 to the plasma membrane. Overexpression of

�-arrestins strongly decreased the PTHR-mediated IP signal,whereas cAMP production was impaired to a much lower ex-tent. The regulation of PTH-stimulated signals by �-arrestinswas impaired for the truncated T480 receptor.

Our data reveal mechanisms at, and distal to, the receptorregulating PTHR-mediated signaling pathways by secondmessenger-dependent kinases. We conclude that regulation ofPTHR-mediated signaling by PKC and �-arrestins are sepa-rable phenomena that both involve the carboxy terminus ofthe receptor. A major role for PKC and �-arrestins in prefer-ential regulation of PTHR-mediated Gq signaling by indepen-dent mechanisms at the receptor level was established.(Endocrinology 143: 3854–3865, 2002)

G PROTEIN-COUPLED RECEPTORS (GPCRs) are thelargest family of signal transducers for extracellular

stimuli. Protein kinases, including second messenger-depen-dent kinases (PKA, PKC), G protein-coupled receptor ki-nases (GRKs), tyrosine kinases, and MAPKs are downstreameffectors of GPCRs. Conversely, upstream regulation ofGPCR signaling involves the interaction of the receptors withthese families of proteins, as well as arrestins (1–4).

Desensitization of GPCRs, that is, the loss of their abilityto respond to an agonist, is a crucial physiological mecha-nism of adaptation to continuous or repeated presence ofstimuli. Rapid (within minutes) events involved in this adap-tive response include phosphorylation of GPCRs or othercomponents within the G protein signaling pathway, afterhomologous or heterologous activation of second messen-ger-dependent kinases (1, 2). In addition, GRKs and arrestinsare involved in the homologous desensitization of GPCRsignaling (3, 5): phosphorylation of agonist-activated GPCRsby GRKs leads to recruitment of �-arrestins, which bind tothe phosphorylated receptors and uncouple the receptorfrom its cognate G proteins. In addition, �-arrestins promotereceptor internalization via clathrin-coated pits (3).

The type I PTH receptor (PTHR) belongs to the class II ofGPCRs (6) and plays a fundamental role in the regulation ofcalcium homeostasis in bone and kidney, as well as in boneformation and resorption (7). Upon activation by agonist

(PTH or PTHrP), PTHR triggers at least two signaling path-ways: Gq/11-mediated PLC� stimulation, leading to inositol-1,4,5-trisphosphate (IP3) production, calcium mobilization,and PKC activation, and Gs-mediated activation of adenylylcyclase, resulting in cAMP production and PKA activation(8–10). The observation of certain cell/tissue specificity inthe PTHR responses adds an additional level of diversity tothe PTHR physiology: whereas PTHR elicits both Gs- andGq-mediated signals in bone and kidney, the PTHR re-sponses are mediated through a Gq pathway in pancreaticinsulinoma cells (11), and predominantly, if not exclusively,through a cAMP-dependent pathway in smooth muscle cells(12).

Second messenger-dependent kinases (13–18), GRKs (19)and arrestins (20) have been implicated in the regulation ofPTHR signaling. The mechanisms of PKC or PKA involvedin desensitization of PTHR remain poorly understood. Indiverse cell types, long-term (within hours) exposure to PTHor to stimulators of second messenger-dependent kinases [i.e.phorbol 12-myristate 13-acetate (PMA), cAMP-analogs orforskolin] resulted in the desensitization of PTH-stimulatedreceptor signaling (13–16). However, a direct effect of recep-tor phosphorylation on desensitization cannot be establishedfrom these studies, as a decrease in receptor number ordown-regulation of other downstream elements of the sig-naling cascade, such as the kinases, can take place underthese conditions. Other studies have shown desensitizationof the PTH-stimulated Ca2�-response after short-term (with-in minutes) pretreatment with PTH or PMA (17, 18). In thesecases, no correlation was established between receptor phos-phorylation and the functional effects observed.

Abbreviations: GPCR, G protein-coupled receptor; GRK, G protein-coupled receptor kinases; HA, hemagglutinin; IP, inositol phosphate;IP3, inositol-1,4,5-trisphosphate; PKA, protein kinase A; PKC, proteinkinase C; PMA, phorbol 12-myristate 13-acetate; PTHR, PTH receptor;SDS, sodium dodecyl sulfate.

0013-7227/02/$15.00/0 Endocrinology 143(10):3854–3865Printed in U.S.A. Copyright © 2002 by The Endocrine Society

doi: 10.1210/en.2002-220232

3854

GRKs, and particularly GRK2, phosphorylate PTHR uponagonist stimulation (19, 21), and they are involved in theregulation of PTHR signaling (19). In addition, agonist-stimulation of PTHR promotes translocation of both �-arrestin1 and �-arrestin2 to the plasma membrane, theirphysical association with the receptor and internalization ofthe receptor/�-arrestin2 complexes (20, 22–24). However,the role of �-arrestins in regulation of the PTHR signalingremains to be fully established.

The aim of this study is to analyze the short-term effectsof PTH-dependent activation of second-messenger depen-dent kinases and of �-arrestins in PTHR signaling. Using thewild-type PTHR and a carboxy-terminally truncated recep-tor, transiently expressed in COS-7 cells, we report that PKCactivation regulates PTHR-mediated Gq and Gs signalingpathways, whereas PKA activation preferentially affectsPTHR-mediated Gs vs. Gq signaling. PKC regulation of PTH-stimulated inositol phosphate (IP) production involvesmechanisms that require the integrity of the carboxy-terminal tail of the receptor. Regulation of PTHR signaling byPKC correlated with effects at the receptor level includingagonist-mediated receptor phosphorylation and internaliza-tion, without compromising �-arrestin2 mobilization to theplasma membrane. In addition, both �-arrestin1 and �-arrestin2 are also able to regulate PTHR signaling upon over-expression in COS-7 cells, preferentially impairing IP vs.cAMP signaling.

Materials and MethodsMaterials

Taq-DNA-polymerase, Fura-2/AM, and 12CA5 monoclonal anti-body were obtained from Roche Diagnostics (Mannheim, Germany).The preparation of �-arrestin1 antibody has been previously described(25). Bisindolylmaleimide I (GF109203X) and staurosporine were ob-tained from Alexis. Pluronic F-127 Protein Grade Detergent and thePKA-inhibitor H89 were from Calbiochem. Human (Nle8,18,Tyr34)-PTH(1–34) (here always referred to as PTH) was purchased from Bachem.Diethylaminoethyl-dextran, protein A-Sepharose CL-4B and human[125I]-(Nle8,18, Tyr34)-PTH-(1–34) used in binding experiments, werefrom Amersham Pharmacia Biotech. [125I]-Labeled human PTHrP-(1–86), used in internalization assays, was kindly provided by Dr. E. Blind(Department of Endocrinology, University of Wurzburg, VershacherStrasse, Germany). The other radiochemicals were obtained from NENLife Science Products (Boston, MA). Cell culture reagents and mediawere from PAN-Systems, except for Roswell Park Memorial Institute1640 medium without l-glutamine and I-inositol, which was obtainedfrom ICN Biomedicals (Eschwege, Germany). Restriction and DNA-modifying enzymes were from New England Biolabs, Inc. (Beverly,MA). The other reagents and chemicals were of analytical grade andobtained either from Sigma (Taufkirchen, Germany) or from Merck(Darmstadt, Germany).

Plasmid constructs

Hemagglutinin (HA)-tagged receptors were generated by convertingthe sequence EDKEAPTGS (residues 93–101) in the N-terminal region ofthe human PTHR cDNA to YPYDVPDYA, as described for the rat PTHR(26), by PCR reaction. The cDNAs for HA-human PTHR and for aHA-truncated human PTHR lacking the carboxy-terminal tail and hencephosphorylation sites (T480 receptor) were subcloned into pCMV plas-mid as previously described (19). �-Arrestin1 (bovine) cDNA (27) sub-cloned into pcDNA3 and pRK5-�-arrestin2 (bovine) (a generous giftfrom Dr. S. Cotecchia, Institut de Pharmacologie et de Toxicologie,Universite de Lausanne) were used in the second messengers experi-ments. GFP-tagged �-arrestin constructs used in translocation experi-ments were previously described (24, 28).

Cell culture and transfection

COS-7 cells were cultured in DMEM 4.5 g/liter glucose, supple-mented with 10% fetal bovine serum, 2 mm glutamine, 100 U/ml pen-icillin, and 0.1 mg/ml streptomycin at 37 C in a humidified atmospherecontaining 7% CO2. Cells were transiently transfected by the diethyl-aminoethyl-dextran method as described previously (19). For IP andcAMP measurements, phosphorylation, and internalization assays withkinase inhibitors, cells grown on 100-mm dishes were transfected with6 �g of receptor DNA, completed with empty vector to a total of 8 �gof DNA. The control (empty vector) DNA was replaced by �-arrestin2-GFP DNA in experiments studying the effects of staurosporine on �-arrestin2-GFP translocation by confocal microscopy. In transient trans-fections for second messenger experiments with overexpressed�-arrestins, as well as for studies of translocation of GFP-tagged �-arrestins, a �-arrestin:receptor ratio of 3:1 was used, unless mentionedotherwise. HEK293 cells employed for quantification of receptor/�-arrestin1-GFP interaction were cultured in DMEM 1.0 g/liter glucose,supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 0.1mg/ml streptomycin at 37 C in a humidified atmosphere containing 7%CO2. Cells were transfected using a protocol based on the calcium-phosphate method as described previously (29). In all cases, assays werecarried out 48–72 h after transfection.

Binding assays

The binding characteristics of the wild-type PTHR and the truncatedT480 transiently transfected in COS-7 cells were determined in intactcells. Cells were washed with ice-cold DMEM/F12 supplemented with1% BSA and incubated in the same medium for 90 min on ice. After thispreequilibration time, medium was substituted for DMEM/F12 con-taining 1% BSA and [125I]PTH (50,000 cpm/well, 2200 Ci/mmol), andcells were incubated on ice for additional 90 min. Incubations werestopped by placing the cells on ice and rapidly washing with 1 mlice-cold PBS before solubilization with 1 ml 0.8 n NaOH. Apparent Kiand Bmax values were obtained from inhibition curves carried out in thepresence of increasing concentrations of unlabeled PTH (10 pm-1 �m),following the equations for homologous competition binding: KI �IC50-L, where L is the concentration of free radioligand, Bmax � B0*(IC50/L), where B0 is the specific binding. Nonspecific binding was defined byusing 1 �m unlabeled ligand.

Total IP production

The assays were carried out following previously published proce-dures (19). To study the effect of protein kinase inhibitors, myo-[2-3H-(N)]-inositol-loaded cells were preincubated for 1 h at room temperaturewith the different inhibitors or vehicle, at the indicated concentrations,in HEPES buffer A (137 mm NaCl; 5 mm KCl; 1 mm CaCl2; 1 mm MgCl2;and 20 mm HEPES, pH 7.3). Subsequently, stimulation was initiated byplacing the cells in a water bath at 37 C and by adding 10 mm LiCltogether with the appropriate concentration of agonist or vehicle. Cellswere stimulated for 1 h in these conditions, in the absence or continuouspresence of the inhibitors. Incubations were stopped by adding ice-coldHClO4 to the cells and placing the plates on ice.

cAMP accumulation

The measurement of cAMP production was carried out by the RIAmethod (Immunotech, Krefeld, Germany) as described (19). To test theeffects of protein kinase inhibitors on agonist-stimulated cAMP pro-duction, cells were washed in HEPES buffer A and preincubated in thesame buffer with the different inhibitors at the indicated concentrationsfor 1 h at room temperature. Subsequently, cells were placed in a waterbath at 37 C and 0.5 mm 3-isobutyl-1-methylxanthine was added. After5 min, cells were stimulated with the appropriate concentration of ag-onist or vehicle for 15 min in these conditions, in the absence or con-tinuous presence of the inhibitors. Incubations were stopped by addingice-cold HClO4 to the cells and placing the plates on ice.

PKC in vitro phosphorylation assays on cell membranes

Cell membranes were prepared from transiently transfected COS-7cells as described (19). Phosphorylation of the HA-tagged PTHR and

Castro et al. • Regulation of PTH/PTHrP Receptor Signaling Endocrinology, October 2002, 143(10):3854–3865 3855

T480 was carried out with recombinant purified PKC�, kindly providedby Dr. H. Mischak (Institut fur Klinische Molekularbiologie, Munchen,Germany). Membranes from cells expressing the nontagged PTHR wereused as a control. Control of receptor levels in the membrane prepara-tions were performed by determining [125I]PTH specific binding fol-lowing the protocol described above, where serum was substitutedby 0.1% 3[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate.The specific activity of the PKC� preparation was 2.0 nmol/min�mgusing histone as substrate. Receptor-containing cell membranes (2 pmolreceptor/reaction) were incubated with 1 �g total protein of the PKC�preparation in buffer containing 20 mm Tris/HCl, pH 7.4; 2 mm EDTA;6.5 mm MgCl2; 0.65 mm CaCl2; 1 mm dithiothreitol; 8 �m PMA; and 0.1mm [�-32P]-ATP (5 cpm/fmol) in a final volume of 50 �l. The phos-phorylation was carried out for 30 min at 30 C and was stopped byaddition of 500 �l ice-cold 20 mm Tris/HCl, pH 7.4, 2 mm EDTA, andsubsequent centrifugation at 21,000 � g, at 4 C for 30 min. The membranepellet was solubilized in 50 �l of RIPA buffer (1% Nonidet P-40; 0.5%Na-deoxycholate; 0.1% sodium dodecyl sulfate (SDS); 50 mm Tris, pH7.4; 100 mm NaCl; 2 mm EDTA; 50 mm NaF) on a shaking incubator for30 min on ice. The HA-tagged PTHRs were immunoprecipitated with12CA5 (2 �g) monoclonal antibody using protein A-Sepharose. Immu-noprecipitated proteins were analyzed on 8.5% SDS-polyacrylamidegels and visualized by autoradiography. Quantification of phosphory-lation was carried out by PhosphorImager analysis (Molecular Dynam-ics, Inc.).

Receptor phosphorylation in intact cells

Transiently transfected COS-7 cells were washed with 3 � 1 mlphosphate-free carbonate buffer (137 mm NaCl; 5 mm KCl; 1 mm CaCl2;1 mm MgCl2; 4.5 g/liter glucose; and 2.5 g/liter NaHCO3) and incubatedfor 2 h at 37 C in the same buffer supplemented with 100 �Ci/well[32P]orthophosphate. During this period and before stimulation of thecells, inhibitors or vehicle were added for 30 min. Labeled cells wereeither stimulated with 100 nm PTH, 1 �m PMA, 50 �m forskolin orvehicle for 5 min and solubilized in 0.8 ml RIPA buffer as above. TheHA-tagged PTHRs were immunoprecipitated with 12CA5 (10 �g)monoclonal antibody. Immunoprecipitated proteins were resolved andanalyzed as described above.

Determination of receptor internalization kinetics

Transiently transfected COS-7 cells were preincubated for 30 min at37 C with 1 ml DHB buffer (DMEM containing 20 mm HEPES and 0.1%BSA), and incubations were continued for another 30 min in DHB in theabsence (vehicle) or presence of kinase inhibitors at the indicated con-centrations. Internalization kinetics were assayed at room temperatureby replacing the preincubation buffer with 1 ml of DHB buffer contain-ing [125I]-PTHrP-(1–86) (50,000 cpm/well) in the continuous absence orpresence of the inhibitors. Incubations were stopped at the indicatedtimes by placing the cells on ice and rapidly washing with 1 ml ice-coldPBS . The cells were incubated for 2 � 5 min in 0.5 ml acid wash solution(150 mm glycine/50 mm acetic acid, pH 3) to remove the surface-boundradioligand. The supernatants containing the acid-released radioactivitywere collected and 1 ml 0.8 n NaOH was added to the cells to solubilizethe acid-resistant radioactivity. The nonspecific binding was measuredin parallel samples using 10�6 m PTH. The radioactivity was quantifiedin an automatic �-counter (1480 Wizard 3“, Wallac, Inc.). The percentinternalization was calculated after deduction of the respective nonspe-cific value: % internalization � [(cpm acid-resistant)/(cpm acid-resistant� cpm acid-released)] � 100. The curves represent the single-phaseexponential fit of the data. The endocytic rate constant, ke (min�1), wasdetermined from the slope of the line obtained by plotting the amountof internalized radioligand against the integral of the surface-boundradioligand (30).

Visualization of agonist-stimulated mobilization of GFP-tagged �-arrestins by confocal microscopy

COS-7 cells transiently expressing PTHR or T480 together with �-arrestin1-GFP or �-arrestin2-GFP were incubated with 100 nm PTH fordifferent times at 37 C. The incubation was stopped by placing the plateson ice and washing the cells with ice-cold PBS before fixation with 4%

paraformaldehyde for 10 min at room temperature. The effects of kinaseinhibition on PTH-stimulated �-arrestin2-GFP translocation were as-sayed by preincubation of the cells with 3 �m staurosporine, for 30 minat 37 C, before stimulation with 100 nm PTH for 5 min at 37 C. Fixed cellswere observed by confocal scanning in a Leica Corp. True ConfocalScanner (TCS4D) confocal laser microscope using fluorescein filters.Representative sections corresponding to a middle plane of the cells arepresented.

Quantification of agonist-stimulated �-arrestin1-GFPmobilization by Western blot

HEK293 cells transiently transfected with �-arrestin1-GFP and PTHRor T480 were stimulated with 100 nm PTH at 37 C for different times.Cells were place on ice, washed with ice-cold PBS, and detached witha rubber policeman. Cells pelleted by low speed centrifugation werelysed in PBS containing a protease inhibitor mixture (Calbiochem) bysonication. The lysate was centrifuged at 1,000 � g for 10 min to removeunbroken cells and the supernatant was further centrifuged at 20,000 �g for 30 min. The resulting supernatant represents the cytosolic fractionand the pellet the membrane fraction. Membrane proteins were solu-bilized in RIPA buffer (1% Nonidet P-40; 0.5% Na-deoxycholate; 0.1%SDS; 50 mm Tris, pH 7.4; and 150 mm NaCl) supplemented with proteaseinhibitors. Equal amounts of cytosolic and soluble membrane proteinswere resolved on 7.5% SDS-polyacrylamide gels. Proteins were trans-ferred onto an Immobilon P transfer membrane (Millipore Corp., Bed-ford, MA). Membranes were reacted with anti-�-arrestin1 antibody (1:3,000), and antirabbit IgG (1:20,000) coupled to peroxidase andimmunoreactive bands were visualized using chemiluminiscence(Pierce Chemical Co., Rockford, IL). Immunoreactive bands werequantified by densitometry and percent translocation was calculatedas: 100*(�-arrestin1-GFP immunoreactivity)membrane/(�-arrestin1-GFPimmunoreactivity)membrane�cytosol.

ResultsCharacteristics of the PTHR and T480 receptor transientlyexpressed in COS-7 cells

Wild-type PTHR and the carboxy-terminally truncatedT480 receptor transiently expressed in COS-7 cells weretested for their ability to bind [125I]-PTH (Fig. 1A). The re-ceptors displayed an affinity for PTH of Ki (mean � sem, n �3) � 16.02 � 2.65 nm for PTHR and 7.83 � 0.06 nm for T480.Thus, T480 receptor showed an affinity for PTH twice as highas PTHR wild type, whereas the maximal levels of expressionachieved for T480 were 24% lower than those for PTHR(Table 1). This was in agreement with previous works re-porting higher affinity for PTH of this truncated receptor incomparison with wild type (19, 31, 32).

In COS-7 cells transiently expressing PTHR or T480receptors, PTH stimulated IP and cAMP responses in aconcentration-dependent manner (Fig. 1B). EC50 values arereported in Table 1. The EC50 values obtained for PTHR-mediated signaling in our system are comparable to previousdata obtained for transiently transfected PTHRs in COS-7cells (25 � 3 nm and 0.32 � 0.04 nm for IP and cAMP,respectively, for rat PTHR) (33).

Despite a lower expression level, the truncated receptorT480 displayed an increased PTH-stimulated IP and cAMPsignaling in comparison to the wild-type receptor (�2.1-foldand �1.2-fold increase, respectively) (Table 1).

The characteristics of PTHR- and T480-phosphorylationby second messenger-dependent kinases were studied. Inintact cells, heterologous activation of PKA by treatmentwith 50 �m forskolin slightly stimulated PTHR phosphory-lation (1.1 � 0.02-fold of basal, n � 3). However, treatment

3856 Endocrinology, October 2002, 143(10):3854–3865 Castro et al. • Regulation of PTH/PTHrP Receptor Signaling

of the cells with 1 �m PMA promoted PTHR phosphorylationto a much higher extent (2.3 � 0.4-fold of basal, n � 3) (Fig.1C), reaching values similar to that obtained after 5 minstimulation with 100 nm PTH (�2.3 � 0.2-fold of basal) (seeFig. 3A). None of these conditions resulted in phosphoryla-tion of the T480 receptor (not shown). In addition, PTHR butnot T480 was a substrate for in vitro PKC phosphorylation inmembranes prepared from COS-7 cells transiently trans-fected with the receptors (Fig. 1D). Thus, PTHR but not T480

was a substrate for PKC phosphorylation in vitro as well asin intact cells.

Activation of second-messenger kinases upon PTH-stimulation regulates PTHR IP and cAMP signaling

The effects of activation of second-messenger kinasesupon agonist stimulation on PTHR signaling were studied byusing three different kinase inhibitors: H89 and bisindolyl-

TABLE 1. Ligand binding and signaling characteristics of the wild-type PTHR and truncated T480 receptor transiently expressed inCOS-7 cells

Bmax (sites/cell � 106) Ki (nM) Max. response EC50 (nM)

PTHR125I-PTH binding (n � 3) 3.75 � 0.03 16.02 � 2.65Inositol phosphate production (n � 3) 3.8 (fold of basal) 44.4 � 9.4cAMP production (n � 5) 223.7 � 7.8 (pmol/well) 1.1 � 0.3

T480125I-PTH binding (n � 3) 2.86 � 0.28 7.83 � 0.06Inositol phosphate production (n � 3) 7.6 (fold of basal) 18.7 � 7.4cAMP production (n � 5) 268.3 � 26.8 (pmol/well) 0.8 � 0.3

FIG. 1. Characteristics of PTHR and T480 receptor transiently expressed in COS-7 cells. A, Competition by PTH for the binding of [125I]PTHto intact COS-7 cells transiently expressing PTHR or T480. Data are the mean � SEM of three independent experiments performed in triplicate.B, Concentration-response effects of PTH on IP and cAMP signaling mediated by PTHR and T480. COS-7 cells transiently expressing PTHRor T480 were stimulated with increasing concentrations of PTH for 1 h (IP production) or 15 min (cAMP accumulation), at 37 C. Dotted linesshow the percent of maximal response achieved at PTH concentrations of 100 nM (IP) and 10 nM (cAMP) chosen in single-concentrationexperiments. Basal values were subtracted from agonist-stimulated values and expressed as % of control for each receptor. Data are the mean �SEM of three (IP measurements) or five (cAMP measurements) independent experiments performed in duplicate. C, Phosphorylation of PTHRin intact cells upon heterologous activation of PKC and PKA. COS-7 cells transiently expressing HA-tagged PTHR were treated with 1 �M PMAor 50 �M forskolin for 5 min. Phosphorylation of anti-HA-immunoprecipitated receptors was quantified by PhosphorImager analysis. Basalphosphorylation was set to 100%. Bars represent the mean � SEM of three independent experiments. D, Cell membranes from COS-7 cellstransiently expressing HA-tagged PTHR or T480, or nontagged PTHR as a control, were prepared for in vitro PKC� phosphorylation assaysas described in Materials and Methods. Receptors were immunoprecipitated, analyzed on 8.5% SDS-polyacrylamide gels, and visualized byautoradiography. The picture shows a representative gel from one of three independent experiments.

Castro et al. • Regulation of PTH/PTHrP Receptor Signaling Endocrinology, October 2002, 143(10):3854–3865 3857

maleimide I (GF109203X), which are selective for PKA andPKC, respectively, and staurosporine, which inhibits a broadspectrum of second messenger-dependent kinases.

Treatment of cells expressing PTHR with kinase inhibitorssignificantly increased PTH-mediated IP and cAMP re-sponses (Fig. 2, A and B). We used concentrations of 100 nmand 10 nm PTH to promote IP and cAMP production, re-spectively, to achieve levels of receptor occupancy that leadto a similar extent of maximal response for each pathway(�85% of the maximum) (Fig. 1B). PTHR responses undercontrol conditions were 4-fold increase of basal IP productionand 23.9 � 2.7 pmol cAMP/well (Fig. 2, A and B). Exposureof the cells to 3 �m staurosporine resulted in a similar effecton both PTHR-mediated IP and cAMP responses: a 2.5-foldincrease in PTH (100 nm)-induced IP signal and a 2.3-foldincrease in PTH (10 nm)-induced cAMP accumulation wereobserved (Fig. 2, A and B). Selective inhibition of PKC by 30�m GF109203X resulted in a 1.8-fold increase in PTHR-mediated IP signal and a 1.7-fold increase in the cAMP signal.A differential regulation of both pathways was observedafter selective inhibition of PKA by 10 �m H89, which af-fected the cAMP signal (1.5-fold increase in PTH (10 nm)-induced cAMP accumulation) but not the IP signal (Fig. 2, Aand B). None of these treatments had effects on basal IP andcAMP levels (data not shown).

The carboxy-terminal tail of PTHR is an importantdeterminant for the PKC-mediated regulation ofthe IP signaling

As mentioned above, the phosphorylation-deficient T480receptor showed an increased agonist-stimulated IP re-sponse in comparison to the wild-type receptor (7.2-foldincrease of basal IP production and 25.3 � 5.5 pmol cAMP/well under control conditions). Inhibition of PKC byGF109203X or staurosporine potentiated the IP signaling ofthe T480 receptor to a lower extent than that of the PTHR (1.3and 1.4-fold increase for T480 vs. 1.8- and 2.5-fold increase forPTHR, respectively) (Fig. 2A). Truncation of the carboxy-terminal tail, however, did not impair the observed poten-tiation of the cAMP response by the different kinase inhib-itors: H89 (1.3-fold increase for T480 vs. 1.5-fold increase forPTHR), GF109203X (2.0-fold increase for T480 vs. 1.7-foldincrease for PTHR) or staurosporine (2.5-fold increase forT480 vs. 2.3-fold increase for PTHR) (Fig. 2B).

Concentration-response curves for PTH-stimulated IP andcAMP signals in control cells or cells treated with 3 �mstaurosporine confirmed the higher potentiation of IP signalsmediated by PTHR vs. T480 receptor (Fig. 2C), as well as thesimilar behavior of both receptors in terms of staurosporineeffects on cAMP signals (Fig. 2D).

FIG. 2. Regulation of PTH-induced IP and cAMP signaling by PKA and PKC. COS-7 cells transiently expressing PTHR or T480 werepreincubated for 1 h at room temperature with vehicle (control), H89 (10 �M), GF109203X (30 �M) or staurosporine (3 �M), and PTH (100 nM,1 h)-stimulated IP production (A) or PTH (10 nM, 15 min)-stimulated cAMP accumulation (B) was determined as described in Materials andMethods. Basal values were subtracted from agonist-stimulated values and expressed as % of control for each receptor. Bars represent themean � SEM of three to five independent experiments performed in duplicate (* means significantly different with respect to control, P � 0.05,** P � 0.01, Mann Whitney test). C, Concentration-response curves for PTH-stimulated IP production in cells expressing PTHR or T480, incontrol conditions or upon treatment with staurosporine (3 �M). PTHR-mediated maximal effect (1 �M PTH) under control conditions was setto 100%. D, Idem for PTH-stimulated cAMP production. Data represent means � SEM of two independent experiments performed in duplicate.

3858 Endocrinology, October 2002, 143(10):3854–3865 Castro et al. • Regulation of PTH/PTHrP Receptor Signaling

Mechanisms contributing to the regulation of PTHRsignaling at the receptor level by PKC

The effects of kinase inhibitors on PTH-stimulated PTHRphosphorylation were tested in intact cells. Exposure of thecells to the PKC inhibitor GF109203X (30 �m) lowered thebasal levels of PTHR phosphorylation by 25%, and, signifi-cantly, the PTH (100 nm)-stimulated phosphorylation ofPTHR (by 50%) (Fig. 3). In contrast, the PKA inhibitor H89(10 �m) did not influence basal or PTH-mediated phosphor-ylation of the PTHR. These data show a correlation betweenPKC-dependent phosphorylation of PTHR upon agoniststimulation and PKC-mediated regulation of the PTHR sig-naling at the receptor level.

Given that internalization of GPCRs is one of the possiblemechanisms that regulate receptor signal responsiveness, wethen examined whether PKC inhibition altered PTHR inter-nalization. As shown in Fig. 4A, GF109203X and staurospor-ine significantly decreased the extent of PTHR internaliza-tion in transiently transfected COS-7 cells (72.0 � 1.4% and66.1 � 1.6% of control, respectively), whereas PKC inhibition

did not modify the endocytic rate constant of PTHR inter-nalization: ke(control) � 0.09 � 0.02 min�1; ke(GF109203X) �0.09 � 0.03 min�1 (Fig. 4B). PKC-dependent internalizationof the PTHR requires the integrity of the carboxy-terminaltail of the receptor, as the T480 receptor showed a deficientinternalization, which was PKC-insensitive (Fig. 4B). Neitherthe extent nor the rate constant of internalization of T480were affected by GF109203X (ke (control) � 0.14 � 0.01min�1; ke(GF109203X) � 0.12 � 0.01 min�1) (Fig. 4B). Thus,selective inhibition of PTHR phosphorylation and internal-ization by PKC inhibitors was concomitant to the observedregulation of PTHR-mediated IP signals by PKC at the re-ceptor level.

Given that �-arrestin2 is involved in the internalizationprocess of PTHR (22, 23), we investigated the effects of PKCinhibition on the recruitment of �-arrestin2-GFP to theplasma membrane upon agonist stimulation of PTHR in

FIG. 3. A, Effects of kinase inhibitors on PTH-mediated phosphory-lation of PTHR in intact cells. [32P]Orthophosphate loaded COS-7cells transiently transfected with PTHR were preincubated with ve-hicle (control), GF109203X (30 �M), or H89 (10 �M) for 30 min at 37C. Agonist-stimulation was started by adding PTH (100 nM) for 5 minat 37 C. Phosphorylation of anti-HA-immunoprecipitated receptorswas quantified by PhosphorImager analysis and basal (nonstimu-lated) phosphorylation was set to 100% (lower panel). Bars representthe mean � SEM of three to five independent experiments (* meanssignificantly different with respect to basal phosphorylation P � 0.05,** P � 0.01, # means significantly different with respect to PTH-stimulated phosphorylation in control conditions P � 0.05, MannWhitney test). B, Visualization of a gel from a representative exper-iment.

FIG. 4. A, Effects of PKC inhibition on internalization of PTHR.COS-7 cells transiently transfected with PTHR were preincubatedwith vehicle, GF109203X (30 �M) or staurosporine (3 �M) for 30 minat 37 C. After that, PTH-induced internalization of PTHR was mea-sured at room temperature as described in Materials and Methods, inthe continuous absence or presence of the kinase inhibitors. B, Effectsof PKC inhibition on the kinetics of internalization of PTHR and T480.COS-7 cells transient transfected with PTHR or T480 were preincu-bated with vehicle or GF109203X (30 �M) for 30 min at 37 C. Afterthat, PTH-induced internalization of the receptors was measured atdifferent time points as described in Materials and Methods, in thecontinuous absence or presence of the kinase inhibitors. Internalizedradioligand is expressed as percent of total specific radioligand bound.Inset, Kinetics of the binding of radiolabelled [125I]PTH to these cells.Data represent the mean � SEM of three independent experimentsperformed in duplicate (* means significantly different with respectto control, P � 0.05, Mann Whitney test).

Castro et al. • Regulation of PTH/PTHrP Receptor Signaling Endocrinology, October 2002, 143(10):3854–3865 3859

COS-7 cells (Fig. 5). �-Arrestin2-GFP showed a homoge-neous cytosolic distribution in nonstimulated cells both inthe absence (panel a) or presence (panel c) of staurosporine.Addition of 100 nm PTH to the cells expressing PTHR pro-duced a rapid translocation of �-arrestin2-GFP to the plasmamembrane (panel b) that was not inhibited by treatment ofthe cells with staurosporine (panel d). Thus, the observedeffects of PKC inhibition on PTHR internalization are not dueto an impaired translocation of �-arrestin2.

Altogether, our data relate the effects of PKC on PTHR-mediated IP pathway at the receptor level with receptorphosphorylation and internalization events independent of�-arrestin2.

�-Arrestin1 and �-arrestin2 differentially desensitize PTH-mediated PTHR signaling

To further elucidate the mechanisms of regulation of bothPTHR signaling pathways, we analyzed the contribution of�-arrestin-dependent mechanisms to the short-term regula-tion of PTHR signaling. Figure 6A showed that the overex-pression of �-arrestin 1 reduced the PTHR-mediated IP re-sponse by 56% and led to a smaller, although significant,reduction of the cAMP response (by 19%). A similar patternwas observed after overexpression of �-arrestin 2, whichsignificantly reduced PTHR-mediated signaling responsesby 86% for IPs and by 18% for cAMP (Fig. 6B). Concentration-response curves reflected the preferential inhibition of thePTHR-mediated IP signals vs. cAMP signals, in cells over-expressing �-arrestin2 (Fig. 6, C and D).

Overexpression of �-arrestins resulted in a predominantinhibition of T480-mediated IP signals vs. cAMP signals aswell (Fig. 6B). Although the overall pattern of effects ofoverexpression of �-arrestins on PTHR or T480 signaling wassimilar, the magnitude of the impairment observed for T480signaling was reduced in comparison to that observed forPTHR signaling (Fig. 6, A and B). This was highlighted in the

case of �-arrestin1 overexpression (Fig. 6A). In the case of�-arrestin2, strong differences in the sensitivity of PTHR andT480 signaling to �-arrestin2 were revealed by lowering�-arrestin2 overexpression levels (Fig. 6E).

These data indicate that �-arrestin1 or �-arrestin2 wereable to regulate PTHR signaling in COS-7 cells with a pref-erential impairment of IP vs. cAMP signaling, and that thetruncation of the carboxy-terminal tail of the receptor com-promised the ability of �-arrestins to regulate PTHRsignaling.

The interaction of �-arrestins with both receptors was fur-ther analyzed by visualization of GFP-tagged �-arrestintranslocation to the plasma membrane. In the absence ofagonist, both �-arrestin1-GFP and �-arrestin2-GFP showed ahomogeneous distribution throughout the cytosol in COS-7cells expressing PTHR or T480, as observed by confocal mi-croscopy (Fig. 7A, panels a, e, i, m). After PTH stimulationfor 5 min, both �-arrestins translocated to the plasma mem-brane in cells expressing PTHR (Fig. 7A, panels b and j).Longer (60 min) exposition to PTH resulted in the localiza-tion of �-arrestins in intracellular vesicles, presumably re-flecting endocytosis of receptor-�-arrestin complexes (Fig.7A, panels d and l).

Recruitment of GFP-tagged �-arrestins to the plasmamembrane was less efficient in cells expressing the T480receptor (Fig. 7A, panels e–h and m–p). In addition, theinteraction of �-arrestins with T480 seemed to be transitory,as certain redistribution of �-arrestins in the cytosol wasobserved at long time (60 min) exposition to the agonist (Fig.7A, panels h and p). This was especially notorious in the caseof �-arrestin1-GFP.

These results were confirmed by quantification of recep-tor/�-arrestin1-GFP interactions by Western blot of cytosoland membranes from unstimulated or PTH-stimulated cells(Fig. 7B). These experiments were carried out in HEK293transiently expressing �-arrestin1-GFP and PTHR or T480, as

FIG. 5. Visualization of PTH-induced �-arrestin2-GFPtranslocation to the plasma membrane. COS-7 cells tran-siently coexpressing PTHR and �-arrestin2-GFP were pre-treated with vehicle (a, b) or staurosporine 3 �M (c, d) for30 min at 37 C. Unstimulated cells (basal) or cells stimu-lated by 100 nM PTH for 5 min in the continuous absenceor presence of staurosporine were fixed and examined byconfocal scanning fluorescence microscopy.

3860 Endocrinology, October 2002, 143(10):3854–3865 Castro et al. • Regulation of PTH/PTHrP Receptor Signaling

these cells were more suitable than COS-7 cells for subcel-lular fractionation and further detection of �-arrestin1 usingour anti-�-arrestin1 antibody. No �-arrestin1 immunoreac-tivity was detected in membranes from unstimulated cellsexpressing PTHR or T480. Upon PTH (100 nm)-stimulationfor 5–10 min, a high �-arrestin 1 immunoreactivity was as-sociated with the membrane fraction in cells expressingPTHR (40% of total �-arrestin1-GFP immunoreactivity at 10

min). PTH-mediated �-arrestin1-GFP recruitment to theplasma membrane was less efficient in cells expressing T480(�14% of total �-arrestin1-GFP immunoreactivity at 10 min).For both receptors, the fraction of �-arrestin1-GFP associatedwith the plasma membrane diminished at later times (20–40min) (Fig. 7B). These data show the contribution of thecarboxy-terminal tail of the receptor to the interaction of thePTHR with �-arrestins.

FIG. 6. Effects of overexpression of �-arrestins on PTH-induced signaling.A, COS-7 cells transiently expressingPTHR or T480 were cotransfected withempty vector (control) or �-arrestin1.Cells were assessed for PTH (100 nM,1 h)-stimulated IP and PTH (10 nM, 15min)-stimulated cAMP responses asdescribed in Materials and Methods.B, Idem for cells coexpressing �-arrestin2. Basal values were sub-tracted from agonist-stimulated val-ues and expressed as percent of con-trol values for each receptor. Barsrepresent the mean � SEM of threeindependent experiments per-formed in triplicate (IP measure-ments) or seven independent exper-iments performed in duplicate(cAMP measurements). C, Concen-tration- response curves for PTH-stimulated IP production mediatedby PTHR and T480 in control cellsor cells coexpressing �-arrestin2.PTHR-mediated maximal effect (1 �MPTH) in control cells was set to100%. D, Concentration-responsecurves for PTH-stimulated cAMP pro-duction mediated by PTHR and T480in control cells or cells coexpressing�-arrestin2. E, PTH (100 nM, 1 h)-stim-ulated IP production in cells express-ing PTHR or T480 cotransfected with�-arrestin2 at different �-arrestin2:re-ceptor DNA ratios. PTH-induced IPresponse in control cells (cotrans-fected with empty vector) was set to100% in each case. Bars representthe mean � SEM of three indepen-dent experiments performed in du-plicate. (* means significantly dif-ferent with respect to control, P �0.05, ** P � 0.01, Mann Whitneytest).

Castro et al. • Regulation of PTH/PTHrP Receptor Signaling Endocrinology, October 2002, 143(10):3854–3865 3861

Altogether, the effects of �-arrestins on the signaling ofPTHR and T480 receptor resembled the different ability ofthe two receptors to interact with �-arrestins.

Discussion

Here we investigate the involvement of second messenger-dependent kinases and �-arrestins in the short-term regula-tion of PTHR signaling. Because PTHR stimulates both Gs

and Gq signaling pathways, a major aim of this study was toexplore whether common mechanisms were involved in theregulation of these two pathways.

Cross-talk between Gs and Gq signaling pathways in theregulation of PTHR signaling

Our results show a cross-talk between the Gs and the Gqpathways at the level of regulation of PTHR signaling by

FIG. 7. Analysis of the interaction of �-arrestins with PTHR or T480 upon agonist-stimulation. A, Time-course of PTH-stimulated translocationof �-arrestin1-GFP and �-arrestin2-GFP to the plasma membrane in COS-7 cells transiently coexpressing PTHR or T480. Unstimulated cellsor cells stimulated with 100 nM PTH for the indicated times at 37 C were fixed and examined by confocal scanning fluorescence microscopy.B, Immunoblot analysis of �-arrestin1-GFP translocation in response to 100 nM PTH in HEK293 cells transiently coexpressing �-arrestin1-GFPand PTHR or T480. Equal amounts of cytosolic and solubilized membrane proteins were separated by SDS-PAGE and �-arrestin1-GFP wasvisualized by immunoblot analysis with a specific �-arrestin1 antibody. Data represent the mean � SEM of seven (PTHR) or three (T480)independent experiments. Inner panel shows a representative Western blot.

3862 Endocrinology, October 2002, 143(10):3854–3865 Castro et al. • Regulation of PTH/PTHrP Receptor Signaling

PKC because PKC inhibition by GF109203X up-regulatedboth IP and cAMP signals to a similar extent. PKA, however,differentially regulated the Gs signaling pathway, becauseselective inhibition of PKA by H89 resulted in up-regulationof the cAMP production without changes of the IP signal. Inagreement with our data, short-term regulation of PTHR-mediated calcium and cAMP signals by PKC has been re-ported (18), whereas significant regulation of the IP or cal-cium signals by cAMP analogs or forskolin could not beobserved in different cell types (15, 17, 18). Considering thisregulatory cross-talk, the selective activation of Gs or Gqpathways by PTHR in certain cell types (11, 12) could leadto particularities in the regulation of the PTHR signaling indifferent tissues. In fact, regarding vascular tissue, wherePTHR selectively activates the Gs signaling pathway, a recentstudy reported the lack of homologous desensitization of thevasodilatory responses to PTHrP in patients with hyper-parathyroidism (34).

Mechanisms at, and distal to, the receptor differentiallycontribute to the regulation of PTHR-mediated IP andcAMP signals

Desensitization mechanisms operating both at, and distalto, the receptor level are known to regulate the signaling ofother Gs- and Gq-coupled receptors (35). The dissociation ofkinase effects on PTHR signaling at, and distal to, the re-ceptor level was achieved in this study by analyzing thesignaling of the carboxy-terminally truncated receptor, T480receptor. This receptor, which lacks all but the juxtamem-brane 16 amino acids of the carboxy-terminal tail of thePTHR, was phosphorylation-deficient upon stimulation byPTH (19), in agreement with other studies pointing out thecarboxy-terminal tail of PTHR as the domain containing thesites of PTH-stimulated receptor phosphorylation (21, 36,37). Our approach allows us to quantify, for the first time toour knowledge, the contribution of mechanisms distal to thereceptor to the short-term desensitization of PTHR signaling.

The results revealed differences in the mechanisms of reg-ulation of PTHR-mediated signals by kinases operating onthe Gs and Gq pathways. The regulation of IP signals medi-ated by PKC was mainly exerted by mechanisms operatingat the receptor level, with certain contribution of effects distalto the receptor. This was reflected by the higher potentiationof the IP signal mediated by PTHR in comparison to T480signals. In contrast, regulation of cAMP signals by PKA orPKC was exerted mainly by mechanisms operating at levelsdown-stream from the receptor, without involvement of re-ceptor phosphorylation events. This was shown by the sim-ilar effects of kinase inhibitors on the cAMP signals mediatedby PTHR or T480. Possible targets for regulation of PTHRsignaling by kinases at levels downstream from the receptorinclude PKC effects on PLC� (38, 39) and PKA and PKCeffects on adenylyl cyclase (40–42).

Regulation of PTHR-mediated IP signaling by PKC isconcomitant to receptor phosphorylation and internalizationevents independent of �-arrestin mobilization

PKC regulation of IP signals at the receptor level correlatedwith PKC-mediated phosphorylation and internalization of

PTHR. In addition, the internalization of the phosphoryla-tion-deficient T480 receptor occurred to a lower extent thanthat of the wild-type, and was no more sensitive to PKCinhibition. Thus, the observed PKC effects on the Gq pathwaycould be mainly attributed to the C-terminus and its phos-phorylation regulating internalization, with small contribu-tion of mechanisms downstream from the receptor. The factthat the inhibition of PTHR internalization has greater effectson the PTHR Gq pathway than on Gs-mediated signals wasexpected, considering the more efficient coupling of the re-ceptor to the Gs pathway.

The requirement of PKC activation for PTHR internal-ization is a controversial question, as opposite results havebeen reported in different cell lines (20, 22, 43). In agree-ment with a recent study performed in HEK293 cells (20),our data support the existence of a PKC-dependent PTHRinternalization pathway in COS-7 cells. This pathway ap-pears either independent of �-arrestin2 or subsequent to�-arrestin2 recruitment to the plasma membrane. In thisway, recent studies in �-arrestin knockout cells suggesta novel mechanism of GPCR internalization through adynamin- and clathrin-dependent pathway that is inde-pendent of �-arrestins (44).

Interaction of both �-arrestin1 and �-arrestin2 with PTHRinvolves the carboxy terminus of the receptor and leads topreferential desensitization of the Gq signaling pathway

Here we report for the first time the functional conse-quences of the interaction of PTHR with �-arrestin1. Giventhat PTHR and �-arrestin1 are endogenously expressed inosteoblastic cells (45), a physiological role for �-arrestin1 asregulator of PTHR signaling appears likely.

Overexpression of �-arrestins rescued the internalizationdeficiency of the T480 receptor in HEK293 cells, whereas ithad smaller effects on the internalization of the wild-typereceptor (23). These effects on the internalization of bothreceptors do not correlate with the effects on the signalingreported here. The smaller sensitivity of T480 signaling vs.PTHR signaling to overexpression of �-arrestins correlates,however, with the reduced ability of T480 to mobilize �-arrestins to the plasma membrane. Thus, whereas the effectsof PKC on PTHR signaling correlated with internalizationevents, the regulation of PTHR signaling by �-arrestinsseems to be associated with recruitment of �-arrestins to thereceptor rather than receptor internalization.

The reduced affinity of T480 for �-arrestins indicates thatthe interaction of �-arrestins with PTHR involves the carboxy-terminal tail of the receptor. However, as shown for otherGPCRs (46), additional receptor domains seem to contributeto the interaction with �-arrestins, as the lower sensitivity ofT480 receptor signaling to �-arrestin2 in our study was over-come by increasing the levels of �-arrestin2 overexpression.

Our data reveal that �-arrestins preferentially desensitizedthe PTHR-stimulated Gq vs. Gs pathway. This does not seemto be due to the different time-frame of measurement of bothsignals. A strong translocation of �-arrestins to the plasmamembrane was already visible upon 5 min of agonist stim-ulation of the cells, and it remains stable at least for 1 h. Theefficient coupling of PTHR to the Gs pathway suggests a

Castro et al. • Regulation of PTH/PTHrP Receptor Signaling Endocrinology, October 2002, 143(10):3854–3865 3863

higher affinity of the receptor for Gs than for Gq, which couldfacilitate the preferential disruption of the PTHR-Gq vs. -Gscoupling by �-arrestins. Thus, a competition of �-arrestinsfor domains on the PTHR also involved in interaction withG proteins could be postulated as the mechanism underlyingthe selective desensitization of PTHR-mediated IP signalingby �-arrestins. Other authors have shown impairment ofcAMP signaling in COS-7 cells upon overexpression of a�-arrestin2-GFP construct (20). However, effects on IP sig-naling were not determined in parallel in this study, whichmakes a comparison to the effects observed in our systemdifficult. Our data highlight a differential desensitization ofthe two PTHR signaling pathways by �-arrestins and suggestthe possibility of selective desensitization of a certain path-way leaving intact other receptor functions.

In conclusion, we have demonstrated different mecha-nisms of regulation of the PTHR Gq and Gs signaling path-ways upon agonist-stimulation of the receptor in COS-7 cells.Mechanisms operating at, and distal to, the receptor con-tribute to a different extent to the short-term feedback reg-ulation of PTHR IP and cAMP signaling by agonist-activatedsecond messenger-dependent kinases. PKC- and �-arrestin-mediated effects involving the carboxy-terminal tail of thereceptor preferentially regulate the Gq vs. Gs signaling path-way. Whereas both PKC and �-arrestin2 are implicated inPTHR internalization, PKC inhibition does not impair ago-nist-stimulated �-arrestin2 recruitment to the plasma mem-brane. We conclude that regulation of PTHR signaling byPKC and �-arrestins are separable phenomena which bothinvolve the carboxy-terminal tail of the receptor.

Acknowledgments

We thank Dr. E. Blind, Dr. S. Cotecchia, and Dr. H. Mischak for kindlyproviding material required to perform this study, as well as Dr. K. Nipand Dr. U. Quitterer for critically reading the manuscript and helpfuldiscussion.

Received February 27, 2002. Accepted June 27, 2002.Address all correspondence and requests for reprints to: Martin J. Lohse,

Institute of Pharmacology and Toxicology, University of Wurzburg, Vers-bacher Strasse 9, D-97078, Germany. E-mail: [email protected].

This work was supported by grants from the Deutsche Forschungs-gemeinschaft, and the Fonds der Chemischen Industrie (grants to M.J.L.)and fellowships from the Alexander von Humboldt Foundation andMinisterio de Ciencia y Tecnologıa, Spain (to M.C.).

* M.C. and F.D. contributed equally to the work.† Present address: Department of Gene Therapy, Roche Pharma Re-

search Oncology, Nonnenwald 2, D-82372 Penzberg, Germany.

References

1. Ferguson SS, J Zhang, LS Barak, MG Caron 1998 Molecular mechanisms ofG protein-coupled receptor desensitization and resensitization. Life Sci 62:1561–1565

2. Chuang TT, Iacovelli L, Sallese M, De Blasi A 1996 G protein-coupled re-ceptors: heterologous regulation of homologous desensitization and its im-plications. Trends Pharmacol Sci 17:416–421

3. Krupnick JG, Benovic JL 1998 The role of receptor kinases and arrestins in Gprotein-coupled receptor regulation. Annu Rev Pharmacol Toxicol 38:289–319

4. Liebmann C, Bohmer FD 2000 Signal transduction pathways of G protein-coupled receptors and their cross-talk with receptor tyrosine kinases: lessonsfrom bradikinin signaling. Curr Med Chem 7:931–943

5. Bunemann M, Hosey MM 1999 G-protein coupled receptor kinases as mod-ulators of G-protein signalling. J Physiol 517:5–23

6. Juppner H, Abou-Samra AB, Freeman M, Kong XF, Schipani E, Richards J,Kolakowski Jr LF, Hock J, Potts Jr JT, Kronenberg HM, Segre GV 1991 A G

protein-linked receptor for parathyroid hormone and parathyroid hormone-related peptide. Science 254:1024–1026

7. Rosenblatt M, Kronenberg HM, Potts Jr JT 1989 Parathyroid hormone: phys-iology, chemistry, biosynthesis, secretion, metabolism, and mode of action. In:de Groot LJ, ed. Endocrinology. Philadelphia: Saunders; 848–891

8. Abou-Samra AB, Juppner H, Force T, Freeman MW, Kong XF, Schipni E,Urena P, Richards J, Bonventre JV, Potts JTJ, Kronenberg HM, Segre GV 1992Expression cloning of a common receptor for parathyroid hormone and para-thyroid hormone-related peptide from rat osteoblast-like cells: a single recep-tor stimulates intracellular accumulation of both cAMP and inositol trisphos-phates and increases intracellular free calcium. Proc Natl Acad Sci USA 89:2732–2736

9. Pines M, Fukayama S, Costas K, Meurer E, Goldsmith PK, Xu X, MuallemS, Behar V, Chorev M, Rosenblatt M, Tashjian Jr AH, Suva LJ 1996 Inositol1-, 4-, 5-trisphosphate-dependent Ca2� signaling by the recombinant humanPTH/PTHrP receptor stably expressed in a human kidney cell line. Bone18:381–389

10. Offermanns S, Iida-Klein A, Segre GV, Simon MI 1996 G�q family memberscouple parathyroid hormone (PTH)/PTH-related peptide and calcitonin re-ceptors to phospholipase C in COS-7 cells. Mol Endocrinol 10:566–574

11. Gaich G, Orloff JJ, Atillasoy EJ, Burtis WJ, Ganz MB, Stewart AF 1993Amino-terminal parathyroid hormone-related protein: specific binding andcytosolic calcium responses in rat insulinoma cells. Endocrinology 132:1402–1409

12. Maeda S, Wu S, Juppner H, Green J, Aragay AM, Fagin JA, Clemens TL 1996Cell-specific signal transduction of parathyroid hormone (PTH)-related pro-tein through stably expressed recombinant PTH/PTHrP receptors in vascularsmooth muscle cells. Endocrinology 137:3154–3162

13. Parnalete N, Garcia JC, Betts CR, Martin KJ 1990 Inhibitors of proteinkinase-C modulate desensitization of the parathyroid hormone receptor-adenylate cyclase system in opossum kidney cells. Endocrinology 126:407–413

14. Mitchell J, Goltzman D 1990 Mechanisms of homologous and heterologousregulation of parathyroid hormone receptors in the rat osteosarcoma cell lineUMR-106. Endocrinology 126:2650–2660

15. Fukayama S, Tashjian AH, Bringhurst FR 1992 Mechanisms of desensitizationto parathyroid hormone in human osteoblast-like SaOS-2 cells. Endocrinology131:1757–1769

16. Guo J, Liu B-Y, Bringhurst FR 1997 Mechanisms of homologous and heter-ologous desensitization of PTH/PTHrP receptor signaling in LLC-PK1 cells.Am J Physiol 273:E383–E393

17. Fujimori A, Miyauchi A, Hruska KA, Martin KJ, Avioli LV, Civitelli R 1993Desensitization of calcium messenger system in parathyroid hormone-stim-ulated opossum kidney cells. Am J Physiol 264:E918–E924

18. Lee SK, Stern PH 1994 Studies on the mechanism of desensitization of theparathyroid hormone-stimulated calcium signal in UMR-106 cells: reversal ofdesensitization by alkaline phosphatase but not by protein kinase C down-regulation. J Bone Miner Res 9:781–789

19. Dicker F, Quitterer U, Winstel R, Honold K, Lohse MJ 1999 Phosphorylation-independent inhibition of parathyroid hormone receptor signaling by G pro-tein-coupled receptor kinases. Proc Natl Acad Sci USA 96:5476–5481

20. Ferrari SL, Bisello A 2001 Cellular distribution of constitutively active mutantparathyroid hormone (PTH)/PTH-related protein receptors and regulation ofcyclic adensine 3�, 5�-monophosphate signaling by �-arrestin2. Mol Endocrinol15:149–163

21. Malecz N, Bambino T, Bencsik M, Nissenson RA 1998 Identification ofphosphorylation sites in the G protein-coupled receptor for parathyroid hor-mone. Receptor phosphorylation is not required for agonist-induced internal-ization. Mol Endocrinol 12:1846–1856

22. Ferrari SL, Behar V, Chorev M, Rosenblatt M, Bisello A 1999 Endocytosis ofligand-human parathyroid hormone receptor 1 complexes is protein kinaseC-dependent and involves �-arrestin2. J Biol Chem 274:29968–29975

23. Vilardaga J-P, Krasel C, Chauvin S, Bambino T, Lohse MJ, Nissenson RA2002 Internalization determinants of the parathyroid hormone receptor dif-ferentially regulate �-arrestin/receptor association. J Biol Chem 277:8121–8129

24. Vilardaga JP, Frank M, Krasel C, Dees C, Nissenson RA, Lohse MJ 2001Differential conformational requirements for activation of G proteins andregulatory proteins, arrestin and GRK in the G protein-coupled receptor forparathyroid hormone (PTH)/PTH related protein. J Biol Chem 276:33435–33443

25. Sohlemann P, Hekman M, Puzicha M, Buchen C, Lohse MJ 1995 Binding ofpurified recombinant �-arrestin to guanine-nucleotide-binding-protein-coupled receptors. Eur J Biochem 232:464–472

26. Lee C, Gardella TJ, Abou-Samra A-B, Nussbaum SR, Segre GV, Potts Jr JT,Kronenberg HM, Juppner H 1994 Role of the extracellular regions of theparathyroid hormone (PTH)/PTH-related peptide receptor in hormone bind-ing. Endocrinology 135:1488–1495

27. Lohse MJ, Benovic JL, Codina J, Caron MG, Lefkowitz RJ 1990 �-Arrestin:a protein that regulates �-adrenergic receptor function. Science 248:1547–1550

28. Groarke DA, Wilson S, Krasel C, Milligan G 1999 Visualization of agonist-induced association and trafficking of green fluorescent protein-tagged formsof both �-arrestin-1 and the thyrotropin-releasing hormone receptor-1. J BiolChem 274:23263–23269

3864 Endocrinology, October 2002, 143(10):3854–3865 Castro et al. • Regulation of PTH/PTHrP Receptor Signaling

29. Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA,Struhl K 1997 Current protocols in molecular biology. New York: Wiley

30. Lund KA, Opresko LK, Starbuck C, Walsh BJ, Wiley HS 1990 Quantitativeanalysis of the endocytic system involved in hormone-induced receptor in-ternalization. J Biol Chem 265:15713–15723

31. Huang Z, Chen Y, Pratt S, Chen T-H, Bambino T, Shoback DM, NissensonRA 1995 Mutational analysis of the cytoplasmic tail of the G protein-coupledreceptor for parathyroid hormone (PTH) and PTH-related protein: effects onreceptor expression and signaling. Mol Endocrinol 9:1240–1249

32. Iida-Klein A, Guo J, Xie LY, Juppner H, Potts Jr JT, Kronenberg HM, Brin-ghurst FR, Abou-Samra AB, Segre GV 1995 Truncation of the carboxyl ter-minal region of the rat parathyroid hormone (PTH)/PTH-related peptidereceptor enhances PTH stimulation of adenylyl cyclase but not phospholipaseC. J Biol Chem 270:8458–8465

33. Iida-Klein A, Guo J, Takemura M, Drake MT, Potts Jr JT, Abou-Samra A,Bringhurst R, Segre GV 1997 Mutations in the second cytoplasmic loop of therat parathyroid hormone (PTH)/PTH-related protein receptor result in selec-tive loss of PTH-stimulated phospholipase C activity. J Biol Chem 272:6882–6889

34. Chadwick DR, Harrison BJ, Chan P, Chong L, Peatchell P 2000 Vasoactiveand proliferative effects of parathyroid hormone and parathyroid hormone-related peptide on human vascular smooth muscle. Br J Surg 87:1529–1533

35. McArdle CA, Davidson JS, Willars GB 1999 The tail of the gonadotrophin-releasing hormone receptor: desensitization at, and distal to, G protein-coupled receptor. Mol Cell Endocrinol 151:129–136

36. Qian F, Leung A, Abou-Samra A 1998 Agonist-dependent phosphorylation ofthe parathyroid hormone/parathyroid hormone-related peptide receptor. Bio-chemistry 37:6240–6246

37. Blind E, Bambino T, Huang Z, Bliziotes M, Nissenson RA 1996 Phosphor-

ylation of the cytoplasmic tail of the PTH/PTHrP receptor. J Bone Miner Res11:578–586

38. Flitz TM, Cunningham ML, Stanig KJ, Paterson A, Harden TK 1999 Phos-phorylation by protein kinase C decreases catalytic activity of avian phos-pholipase C-�. Biochem J 338:257–264

39. Yue C, Ku C-Y, Liu M, Simon MI, Sanborn BM 2000 Molecular mechanismof the inhibition of phospholipase C �3 by protein kinase C. J Biol Chem275:30220–30225

40. Iwami G, Kawabe J, Ebina T, Cannon PJ, Homcy CJ, Ishikawa Y 1995Regulation of adenylyl cyclase by protein kinase A. J Biol Chem 270:12481–12484

41. Chen Y, Harry A, Li J, Smit MJ, Bai X, Magnusson R, Pieroni JP, Weng G,Iyengar R 1997 Adenylyl cyclase 6 is selectively regulated by protein kinaseA phosphorylation in a region involved in G�s stimulation. Proc Natl Acad SciUSA 94:14100–14104

42. Lai H-L, Yang T-H, Messing RO, Ching Y-H, Lin S-C, Chern Y 1997 Proteinkinase C inhibits adenylyl cyclase type VI activity during desensitization of theA2a-adenosine receptor-mediated cAMP response. J Biol Chem 272:4970–4977

43. Tawfeek HAW, Che J, Qian F, Abou-Samra AB 2001 Parathyroid hormonereceptor internalization is independent of protein kinase A and phospholipaseC activation. Am J Physiol Endocrinol Metab 281:E545–E557

44. Paing MM, Stutts AB, Kohout TA, Lefkowitz RJ, Trejo J 2002 �-arrestinsregulate protease-activated receptor-1 desensitization but not internalizationor down-regulation. J Biol Chem 277:1292–1300

45. Bliziotes M, Murtagh J, Wiren K 1996 �-adrenergic receptor kinase-like ac-tivity and �-arrestin are expressed in osteoblastic cells. J Bone Miner Res11:820–826

46. Cen B, Xiong Y, Ma L, Pei G 2001 Direct and differential interaction of�-arrestins with the intracellular domains of different opioid receptors. MolPharmacol 59:758–764

Castro et al. • Regulation of PTH/PTHrP Receptor Signaling Endocrinology, October 2002, 143(10):3854–3865 3865