Phospholipase D- and Protein Kinase C Isoenzyme- Dependent ...

0013-7227/98/$03.00/0 Endocrinology Copyright © 1998 by The Endocrine Society

Vol. 139, No. 7 Printed in U.S.A.

Phospholipase D- and Protein Kinase C IsoenzymeDependent Signal Transduction Pathways Activated by the Calcitonin Receptor* FABIO NARO, MARIE PEREZ, SILVIA MIGLIACCIO, DEBORAH L. GALSON, PHILIPPE ORCEL, ANNA TETI†, AND STEVEN R. GOLDRING† Department of Histology and Medical Embryology (F.N., M.P., S.M.), University “La Sapienza,” and Istituto Dermopatico dell’Immacolata (M.P., S.M.), Rome, Italy; Beth Israel Deaconess Medical Center (D.L.G., S.R.G.), New England Baptist Bone and Joint Institute, Harvard Medical School, Boston, Massachusetts; Department of Rheumatology (P.O.), INSERM, Unit 349, Lariboisiere Hospital, Paris, France; Department of Experimental Medicine (A.T.), University of L’Aquila, Italy ABSTRACT The calcitonin receptor expressed by the porcine LLC-PK1 renal tubule cells is a seven-transmembrane domain, G protein-coupled receptor activating adenylyl-cyclase and phospholipase C. Salmon calcitonin stimulated dose- and time-dependent release of the phospholipase D-dependent phosphatidylcholine product [3H]choline with an EC50 5 2.5 6 0.3 3 1028 M, similar to that determined for phosphoinositide metabolism (EC50 5 4.5 6 1.0 3 1028 M). The hormone failed to induce release of [3H]phosphocholine and [3H]glycerophosphocholine, ruling out activation of phosphatydilcholine-specific phospholipase C and phospholipase A. Calcitonin stimulated phosphatidic acid, a product of phospholipase D-dependent phosphatydilcholine hydrolysis. Activation of phospholipase D was confirmed by release of [3H]phosphatydilethanol, a specific and stable product in the presence of a primary alcohol. Activation of calcitonin receptor

C

ALCITONIN (CT), a 32-amino acid peptide, was originally identified as a hormone that lowers extracellular calcium levels. Subsequently, it has been shown to induce biological responses in a number of tissues including the central nervous, cardiovascular, and gastrointestinal systems (1). The recent cloning of calcitonin receptors (CTRs) from several species has provided insights into the unique structural and functional properties of these binding proteins (2). Analysis of the predicted structure of the CTRs reveals that they are members of the seven-transmembrane domain spanning, GTP-binding protein-coupled receptor superfamily, and share homology with the PTH and PTH-related peptide (PTH-PTHrP) receptor, and the receptors for the secretin/glucagon peptides (3, 4). Several CTR isoforms have been identified to date. They are derived from a single gene, by alternative splicing of the messenger RNA (2). Complementary DNAs encoding CTRs from porcine (5), rodent (6), and human (7–9) show splice variation in the first intracel-

Received December 9, 1997. Address all correspondence and requests for reprints to: Anna Teti, Ph.D., Department of Experimental Medicine, via Vetoio, Coppito 2, 67100 L’Aquila, Italy. E-mail: [email protected]. * Supported by grants from Ministero Dell’ Università e Della Ricerca Scientifica e Tecnologica and SmithKline Beecham S.p.A. (to A.T.) and Grants from the United States Public Health Service AR-03564 and NIH DK-46773 (to S.R.G). † Joint senior authors.

induced diacylglycerol formation, with a rapid peak followed by a prolonged increase, due to activation of phospholipase C and of phospholipase D. Consequently, the protein kinase-Ca, but not the d isoenzyme, was cytosol-to-membrane translocated by approximately 50% after 20 min exposure to calcitonin, whereas protein kinase-Cz, which was approximately 40% membrane-linked in unstimulated cells, translocated by approximately 19%. The human calcitonin receptor expressed by BIN-67 ovary tumor cells, although displaying higher affinity for calcitonin, failed to activate phospholipase D and protein kinase-C in response to the hormone. This receptor lacks the G protein binding consensus site due to the presence of a 48-bp cassette encoding for a 16-amino acid insert in the predicted first intracellular loop. This modification is likely to prevent the calcitonin receptor from associating to phospholipase-coupled signaling. (Endocrinology 139: 3241–3248, 1998)

lular (I1) and second extracellular (E2) domains, with 78% identity between rat and human, and 67% between rat and pig CTRs. The predicted native protein sizes have a relative molecular mass of approximately 55 kDa, which may be increased to approximately 70 – 85 kDa by N-linked glycosylation (10). Analysis of the functional properties of the porcine CTR, cloned from the LLC-PK1 renal tubule cells, revealed that it was coupled to at least two signal transduction pathways. Coupling to the activation of adenylyl cyclase/cAMP/ protein kinase A pathway has been described in several cell types (11), and further studies revealed that the same receptor is also coupled to activation of phosphoinositide (PtdIns)dependent phospholipase C (PLC), which results in Ca21 mobilization (12), and protein kinase C (PKC) activation (13). The two signal transduction pathways require the cholera toxin sensitive Gs, and the pertussis toxin-sensitive Gi and possibly the Gq PKC-coupled GTP-binding proteins, respectively. Coupling of CTR to these distinct G proteins leads to opposite biological effects associated with e.g. regulation of the sodium pump (i.e. stimulation by protein kinase A and inhibition by PKC) (13). Furthermore, selective activation of the pathways can occur in a cell cycle-dependent manner because Gs is activated in G2 phase and no longer stimulated in S phase (13, 14). Conversely, in LLC-PK1 CTR couples to Gi in S phase, but not in G2 phase (13, 14).

3241

3242

PLD AND PKC ACTIVATION BY THE CALCITONIN RECEPTOR

Full activation of multiple signal transduction pathways is not a common feature of all CTR isoforms. In fact, the BIN-67 human ovary tumor cell line expresses a high copy number of a CTR isoform, which is distinguished from the porcine CTR by the presence of a 48-bp cassette which encodes a 16 aa insert in the predicted first intracellular loop (481) (7). Previous studies by our laboratories and others have shown that this isoform binds CT with high affinity, but fails to stimulate Ca21 mobilization (7, 9). The functional capacity to activate alternate signaling pathways may provide a mechanism by which CT binding results in different biological responses. Therefore, the dissection of multiple intracellular signals elicited by the same receptor is relevant to fully elucidating the mechanism of action of the hormone. This study was aimed at investigating whether CT stimulates novel CTR-dependent signal transfer in LLC-PK1 and BIN-67 cells. To this end, we focused on membrane phosphatydilcholine (PtdCho) metabolism and membrane translocation of PKC isoforms induced by binding of salmon CT (sCT). Our results provide evidence that phospholipase D (PLD) and PKCa, and possibly z, isoenzymes are potential components of CTR(48-)- but not of the CTR(481)-coupled cell signaling. Materials and Methods Materials d-myo-[3H]inositol (specific activity 500 Gbq/mmol), [3H]choline chloride (specific activity 3 Tbq/mmol) were from Du Pont de Nemours (Milano, Italy). [9,10(n)]-[3H]myristic acid (specific activity 2 Tbq/ mmol), Hybond C nitrocellulose membrane and Enhanced Chemiluminescence (ECL) kit were from Amersham Italia (Milano, Italy). Tissue culture grade fatty acid-free BSA was from Boheringer Mannheim (Milano, Italy). DL-propranolol, phospholipid and lipid standards, and chromatography resins were from Sigma Aldrich (Milano, Italy). TLC silica gels were from Merck (Darmstadt, Germany). Anti-PKC isoenzyme antibodies were from Signal Transduction Laboratories (Lexington, KY). Horseradish peroxidase-conjugated secondary antibody was from Santa Cruz Biotechnology, Inc. (Heidelberg, Germany). Tissue culture plastic ware was from Becton Dickinson (Oxnard, CA). All other reagents were of the purest grade from Sigma Chemical Co. (St. Louis, MO).

Cell cultures LLC-PK1 cells are a porcine kidney epithelial cell line previously characterized in our laboratory (15). BIN-67 cells are a human ovary tumor cell line (7). Cells were cultured in DMEM supplemented with 10% FBS and antibiotics and fed once a week. At confluence, cells were split 1:20 by standard trypsin procedures.

Radiolabeling and stimulation [3H]inositolphosphate (InsP) and [3H]choline labeling. Cells were grown in 12-well tissue culture plates. On the second day of culture the medium was substituted with fresh medium containing either 185 kBq/ml d-myo[3H]inositol or 74 kBq [3H]choline chloride and cultured for an additional 48 h. Fatty acid labeling. Cells were plated into 6-cm tissue culture dishes and grown to 80% confluence. Monolayers were then incubated for 3 h with 9.25 kBq/ml [9,10(n)]-[3H]myristic acid. After labeling, the cells at 80% confluence (;2 3 105 cells/dish for InsP and [3H]choline labeling and approximately 3 106 cells/dish for fatty acid labeling) were extensively washed with DMEM supplemented with 20 mm HEPES, pH 7.4, and 0.1% BSA, and stimulated with the agonist. Incubations were stopped by aspirating the medium and adding ice-cold methanol for PtdCho, or 10% trichloroacetic for InsP determi-

Endo • 1998 Vol 139 • No 7

nation (16). For soluble PtdCho metabolites, stimulation was terminated by direct addition of 1 vol of ice-cold methanol to the incubation medium (17).

Measurement of InsP accumulation [3H]InsP-labeled mono and bisphosphate were extracted, separated by ion exchange chromatography (Dowex 138 –200) and measured as previously reported (18).

Analysis of water-soluble PtdCho metabolites Cells were scraped and extracted for 1 h at 4 C before adding chloroform and water to a final ratio of CHCl3/MeOH/H2O 5 1:1:0.9. To separate labeled glycerophosphocholine, phosphocholine and choline, the aqueous-methanolic phase was chromatographed on Dowex 50WH1 according to a modification of the procedure described by Cook and Wakelam (17, 19).

Assays of PLD activity by phosphatidylethanol (PtdEtOH) production, and diacylglycerol (DAG) and phosphatidic acid (PA) determinations PLD activity was evaluated measuring PtdEtOH formation from [3H]myristic acid labeled PtdCho, which is considered a direct and specific assay for PLD (20). After washing as described above, cells were incubated with 1% ethanol for 10 min, and sCT was added for the times indicated. Cells were scraped from the culture dishes in ice-cold methanol, and lipids were extracted by the method of Bligh and Dyer (21): the monophase was vortexed, 1 vol of chloroform, 1 vol of 1 m KCl and HCl to 1 n final concentration were added, and the resulting aqueous and organic phases separated by centrifugation for 15 min at 3,000 3 g. The lower phase was washed with synthetic upper phase (22), dried down under a stream of N2, stored at 280 C, and analyzed by TLC. DAG was separated by chromatography on 20 3 20 cm silica gel 60 plates, activated at 110 C for 1 h before use. The solvent system was petroleum ether/diethyl ether/acetic acid (80:20:1, vol/vol) (23) (DAG RF 5 0.15– 0.20). For the analysis of PA and [3H]PtdEtOH lipid extracts were applied, along with standards, to 20 3 10-cm silica gel 60 plates, and separated using the solvent system ethyl acetate/iso-octane/acetic acid/water (13:2:3:10, vol/vol) (PA RF 5 0.30). [3H]PtdEtOH was identified by comigration with unlabeled PtdEtOH prepared according to Baldini et al. (22).

Detection of PKC For immunoblotting, 90% confluent cells were incubated in serumfree DMEM containing 0.2% BSA for 1 h, then treated with sCT. After the indicated time periods, the medium was removed, and cells were washed twice with ice-cold PBS. Cells were then scraped into 1 ml of ice-cold homogenization buffer [20 mm Tris-HCl, pH 7.5, 1 mm EDTA, 1 mm EGTA, 2 mm dithiothreitol, 1 mm phenylmethanesulfonyl fluoride, 10 mm benzamidine, 25 mg/ml leupeptin and 6 mg/ml aprotinin]. All subsequent steps were carried out at 4 C. The cells were lysed by ultrasonication and spun for 1 h at 100,000 3 g. Supernatants were used as a source of cytosolic protein. Pellets were resonicated in 1 ml of the same buffer containing 1% Triton X-100 and centrifuged for 1 h at 100,000 3 g, yielding the solubilized membrane fraction. The cell fractions were subjected to SDS-PAGE (8% acrylamide gel), and proteins were transferred onto nitrocellulose membranes. Filters were then incubated at 4 C overnight with monoclonal antibodies against PKCa, d, and z isoenzymes (diluted 1:1000) and then with antimouse horseradish peroxidase-conjugated secondary antibody (diluted 1:5000), which was detected by ECL. Cellular distribution of PKC isoenzymes was evaluated by standard indirect immunofluorescence.

Statistics Data are presented as average 6 se. Statistical analysis was performed by ANOVA. P , 0.05 was conventionally considered to indicate statistical significance.

PLD AND PKC ACTIVATION BY THE CALCITONIN RECEPTOR Results Phospholipid metabolism

To determine the effect of sCT on PtdIns metabolism, LLC-PK1 cells were prelabeled with d-myo-[3H]inositol and stimulated for 30 min with different concentrations of the hormone, in the presence of 10 mm of the inositol monophosphatase inhibitor LiCl. Under these conditions, sCT induced a dose-dependent accumulation of InsP112 starting above 10210 m and reaching a maximum at 1026 m, with an EC50 5 4.5 6 1.0 3 1028 m (mean 6 se, n 5 3, Fig. 1A). This was in agreement with what we have previously reported (11).

3243

Hydrolysis of PtdCho was first investigated in LLC-PK1 cells by the analysis of water soluble PtdCho metabolites to discriminate among possible PLC, PLD, or PLA activities (16). Treatment with sCT induced a clearly detectable increase of free [3H]choline in the extracellular medium compared with unstimulated controls (Fig. 1B) but failed to raise the levels of [3H]phosphocholine and [3H]glycerophosphocholine measured in parallel and within the same cells (data not shown). These results suggest that only PtdCho-PLD was activated by sCT treatment, whereas neither PtdCho-PLC nor PLA activity were affected (19). The increase of PLDdependent extracellular free [3H]choline was time and concentration dependent (Figs. 1B and 2A). The effect started above 1029 m and reached a plateau at 1026 m sCT, with an EC50 5 2.5 6 0.5 3 1028 m (mean 6 se, n 5 3, Fig. 1B). The other product of PLD-dependent PtdCho hydrolysis is PA (23). We investigated PA generation in [3H]myristic acid labeled cells after stimulation with 1026 m sCT. As shown in Fig. 2B, the hormone induced a time-dependent increase of PA generation, with a maximum (2-fold) achieved within 5 min, followed by a sharp decrease probably due to conversion to DAG (see below). Activation of PLD

In the presence of EtOH, activation of PLD can be measured in [3H]myristic acid-labeled cells by monitoring the formation of [3H]PtdEtOH. This is a specific and stable product of the characteristic PLD-dependent transphosphatidylation of PtdCho in the presence of a primary alcohol (24). 1026 m sCT induced a time-dependent accumulation of [3H]PtdETOH, which was detectable within a few seconds and reached a plateau in 5 min (Fig. 3). These results are consistent with those observed for free [3H]choline and PA. DAG generation

FIG. 1. Dependency of [3H]choline and [3H]InsP production on sCT concentration. A, For [3H]InsP determination, LLC-PK1 cells, labeled with 185 kBq of D-myo-[3H]inositol for 48 h, were challenged for 30 min with different concentrations of sCT in the presence of 10 mM LiCl, then [3H]InsP112 accumulation was measured. B, For [3H]choline determination, LLC-PK1 cells were labeled with [3H]choline chloride (74 kBq/ml) for 48 h and stimulated with increasing concentrations of sCT for 10 min. Total free [3H]choline was then measured. Results are plotted as percentage vs. unstimulated samples. Data represent the mean 6 SE of three independent experiments.

Taken together, these observations demonstrate that sCT can activate PLC- and PLD-dependent signaling pathways in LLC-PK1 cells. Both pathways are known to lead to DAG generation. In fact, hydrolysis of PtdIns by PLC directly produces DAG, whereas PA derived from PLD-dependent PtdCho hydrolysis can be converted into DAG by a specific phosphatase (22). To investigate the pattern of sCT-induced DAG formation, LLC-PK1 cells were prelabeled with [3H]myristic acid and stimulated with 1026 m sCT for different times. As shown in Fig. 4, the hormone induced a rapid DAG increase occurring within a few seconds, followed by a slower accumulation, reaching a maximum after 5 min, and subsequently declining. The existence of this biphasic sCT response is consistent with a different origin of the two DAG phases, as we have already reported for vasopressin in myogenic cells (16). To confirm the PLD origin of one of the two phases, before stimulation with sCT, cells were pretreated for 15 min with 100 mm propranolol, a specific PA-phosphatase inhibitor (25). Under these conditions, analysis of DAG (Fig. 5) showed that the late phase of sCT-dependent DAG generation, measured after 5 min of incubation, was completely abolished by this agent, suggesting that this component of DAG increase is entirely due to the conversion of PA into DAG. In contrast, the first phase of DAG generation was not

3244


Endo • 1998 Vol 139 • No 7

FIG. 3. sCT-dependent generation of PtdEtOH. LLC-PK1 cells were incubated for 3 h with 9.25 kBq/ml [3H]myristic acid, extensively washed, and incubated at 37 C for 10 min with 1% EtOH dissolved in DMEM11% BSA, before stimulation with 1026 M sCT for the times indicated and measurement of PtdEtOH. Results are plotted as percentage increase vs. unstimulated samples. Data represent the mean 6 SE of three independent experiments. Inset, Similar experiment conducted with BIN-67 cells treated for the indicated times with 1027 M PMA. Results are plotted as percentage increase vs. unstimulated samples. Data are from one representative out of three independent triplicate experiments.

FIG. 2. Time course of sCT-dependent release of free [3H]choline and PA. A, LLC-PK1 cells were labeled with [3H]choline chloride (74 kBq/ ml) for 48 h and stimulated with 1026 M sCT for 10 min, then the total free [3H]choline was measured. Results are plotted as percentage vs. unstimulated samples. Data represent the mean 6 SE of three independent experiments. B, LLC-PK1 cells were incubated for 3 h with 9.25 kBq/ml [3H]myristic acid, extensively washed and incubated at 37 C with 1026 M sCT for the times indicated on the abscissa. PA was then analyzed. Results are plotted as percentage increase vs. unstimulated samples. Data are from one representative out of three independent triplicate experiments.

inhibited by propranolol, suggesting that its origin is from PLC activity (Fig. 5). It should be noted that pretreatment with propranolol increased PLC-dependent DAG generation. This effect was probably due to stimulation of PLC activity by the accumulated PA as we and others have previously reported (16, 26). Similar results were obtained with alprenolol (cpm, basal: 2858 6 85; 1026 m sCT, 5 min: 4236 6 124; 100 mm alprenolol 1 1026 m sCT, 5 min: 2728 6 185; mean 6 se, n 5 3 P , 0.02).

FIG. 4. sCT-dependent generation of DAG. LLC-PK1 cells were incubated for 3 h with 9.25 kBq/ml [3H]myristic acid, extensively washed, and incubated at 37 C with 1026 M sCT for the times indicated, then DAG was analyzed. Results are plotted as percentage increase vs. unstimulated samples. Data represent the mean 6 SE of three independent experiments.

Activation of PKC isoforms

Because DAG is known to activate PKC by translocation from the cytosol to the membrane fraction, we next inves-


FIG. 5. Effect of propranolol on sCT-dependent generation of DAG. LLC-PK1 cells were incubated for 3 h with 9.25 kBq/ml [3H]myristic acid, extensively washed, and preincubated at 37 C for 15 min with 100 mM propranolol dissolved in DMEM11% BSA, before stimulation with 1026 M sCT for 10 sec or 5 min, DAG generation was then analyzed. Results are plotted as percentage increase vs. unstimulated samples treated or not with propranolol. The data shown in the figure are the mean 1 SE of three independent experiments.

tigated whether this occurs in sCT-treated LLC-PK1 cells. Immunoblotting analysis using PKC isoform specific antibodies was used to identify the isoenzymes potentially involved. Figures 6A and 7 show analysis of the Ca21/DAGdependent PKCa, the DAG-dependent, Ca21-independent PKCd, and the atypical PKCz isoenzymes in cells treated with 1026 m sCT for the indicated time periods. We observed that while PKCa and d were totally present in the soluble fraction in untreated cells, the z-isoenzyme was partially membrane linked, indicating a potential constitutive activation of this enzyme. In cells treated with sCT, we observed a time-dependent membrane translocation of the PKCa isoform. Potential activation of this isoenzyme was indicated by a similar translocation induced by 1027 m phorbol 13-myristate 12acetate (PMA), a phorbol ester that is a potent PKC activator in short-term treatments (time 0, cytosol 94 6 4%, membrane 6 6 4%; 5 min PMA, cytosol 29 6 6%, membrane 71 6 6%, n 5 3). However, sCT-dependent PKCa membrane translocation had a slower pattern compared with PMA because lower levels of translocation were observed after 5 min with sCT (11 6 6%, n 5 3) relative to PMA (66 6 10%, n 5 3, P , 0.01). Our data are consistent with a sustained stimulation of PKCa via PLD activation. After sCT treatment, PKCz showed a modest (19 6 9%) but clearly detectable increase in the particulate fraction, correlated with a simultaneous and comparable decrease in the cytosol (Fig. 6A and 7). sCT did not induce any significant effect on the PKCd isoform (Fig. 6A and 7). This was not due to compartmentalization of this isoform which, similar to the a and the z isoenzymes, was found by immunofluorescence to be diffusely distributed

3245

FIG. 6. Immunoblotting analysis of sCT-induced translocation of PKC isoenzymes. A, Cells were treated with 1026 M sCT for different periods of time (0 to 609). Cells were homogenized and separated into cytosol (c) and membrane (m) fractions. Equal amounts of cell protein were subjected to 8% SDS-PAGE, then transferred onto nitrocellulose paper. Western blotting analysis was performed using monoclonal antibodies against PKCa, d, and z, followed by horseradish peroxidase-conjugated antimouse antibody. Bands were detected by ECL reagents. A, LLCPK1. B, BIN-67. These experiments were repeated three times with similar results.

both in the nucleus and in the cytoplasm of unstimulated LLC-PK1 cells (data not shown). BIN-67 cells

We next addressed the question of whether activation of PLD- and PKC-dependent signal transduction pathways is a common feature of structurally distinct CTR isoforms. To accomplish this, we investigated [3H]choline, [3H]phosphocholine, and [3H]glycerophosphocholine production, as well as PKCa, d, and z membrane translocation, in the human BIN-67 ovarian carcinoma cell line expressing the CTR48(1) isoform (Table 1). In contrast to results with the LLC-PK1 cells, treatment of BIN-67 cells with either salmon or human CTs failed to activate PLD (Table 1) or induce membrane translocation of the PKC isoenzymes a, d, and z (Fig. 6B). In the same cells, parallel experiments showed that the PLD pathway was activated by 1027 m PMA (Fig. 3, inset), and that a PLC-coupled receptor agonist, bradykinin (1026 m), stimulated by 2-fold both InsP formation and [Ca21]i increase (not shown), thus indicating that normal intracellular signals were operative. Discussion

In this study, we demonstrated that the porcine CTR activates a novel signal transduction pathway associated with increased PLD activity, and that downstream signals from the CT/CTR complex lead to activation of the PKCa, and possibly z isoenzymes. The capacity of CTR to couple to multiple signal transduction pathways, including the PLD, could provide an explanation for the diverse biological activities of CT. CT-induced activation of adenylyl cyclase was

3246


FIG. 7. Densitometric analysis of sCT-induced translocation of PKC isoenzymes. Scanning densitometry of Western blots was performed using a laser densitometer (Ultrascan, XL; Pharmacia LKB, Milano, Italy). The sum of cytosol and membrane fractions for each different time point is considered as 100%. Changes in the intensity of the bands for both cytosol and membrane fractions are expressed as percent of this total. Data are mean 6 SE from three independent experiments.

the first signaling mechanism recognized for the CTR, and activation of this pathway is, for example, responsible for regulation of specific cell functions, such as stimulation of the sodium pump (13), or arrest of cell motility (27–29). It recently became apparent that not all of the CTR-mediated cell functions can be explained by effects on increased cAMP levels. For example, stimulation of Ca21 signaling has been shown to be responsible for osteoclast retraction, which, together with the arrest of cell motility by cAMP, impairs bone resorption (29, 30). The predicted structure of the LLC-PK1 cell CTR exhibits several unique features that characterize this subfamily of receptors, including a short cytosolic loop that is not similar to corresponding regions (between helices V and VI) of other adenylyl cyclase-coupled receptors. This region, as well as the proximal portion of the carboxy-terminal end, does contain a potential consensus sequence for G protein coupling.

Endo • 1998 Vol 139 • No 7

In preliminary studies, we have used a chimeric IGF II/CTR receptor to demonstrate that these nucleotide sequences do encode amino acid cassettes that are able to interact with distinct G proteins (31, Orcel et al. manuscript in preparation). Thus, these unusual structural features may account for coupling of the CTR to different G proteins and for stimulation of multiple signal transduction pathways. Phospholipase-dependent breakdown of phospholipids represents one of the major contributors to hormone-mediated cell regulation. CT-mediated activation of PtdIns-PLC leading to inositol trisphosphate-induced Ca21 mobilization has been demonstrated in LLC-PK1 cells and in cells transfected with the porcine CTR (11, 12, 32). The ultimate product of PtdIns metabolism is DAG, a membrane lipid that binds DAG-dependent PKC, activating the catalytic domain of the enzyme. DAG formation, however, is not an unique feature of PtdIns turnover because it may also be derived from the metabolism of PtdCho (24). This is a substrate for several phospholipases and in this study we report that its metabolism is stimulated in LLC-PK1 cells by treatment with sCT in a dose- and time-dependent manner. Analysis of specific products of PtdCho breakdown suggested that CTR activates PLD rather than PtdCho specific-PLC or PLA. This is further confirmed by ethanol transphosphatidylation, an event dependent upon specific PLD activation (24). sCT-induced PLD activation markedly increased PA, a molecule that has been implicated in signal transduction pathways related to the induction of mitogenesis (33), for example by activating MAP kinase (34), possibly explaining the cell cycle-dependent effect of sCT on LLC-PK1 cells (13, 14). In addition, it has recently been reported that PA directly activates PKCz, an event independent of both Ca21 and DAG (22). Here we report a modest PKCz translocation, which could be due to sCT-induced PLD activation, and subsequent PA formation. Our results also demonstrate that DAG production is regulated with a distinct temporal pattern, whether derived by PtdIns or PtdCho metabolism. Examination of the timedependent effects of sCT, and the inhibition of sustained DAG production by propranolol (25), indicate that activation of PtdIns-PLC is likely to induce a rapid, transient increase in DAG formation, which is over within a few seconds, whereas PLD activation stimulates a second, sustained increment of DAG production lasting several minutes. This pattern may explain the prolonged activation of the PKCa isoenzyme observed in sCT-treated cells. PKCa is a classic Ca21/DAG-dependent PKC isoform, which is highly expressed in LLC-PK1 cells. CT induces a slow and sustained PKCa membrane translocation. This is at variance with the pattern observed in cells treated with PMA, which induced a rapid activation of the isoenzyme. These results suggest a possible role played by PLD in the sCT-induced sustained PKCa activation because PLD-dependent DAG generation is quantitatively predominant and generated for a prolonged time compared with the DAG produced by PLC activation. Among the many biological roles for PKC, the enzyme has been found to be involved in the modulation of the CTR function and response to the hormone (35–37). Our data provide insights indicating that PKCa and z are potential components of the CTR(48-)-coupled signaling. In contrast, despite the generation of DAG induced by sCT, translocation


3247

TABLE 1. Time course of sCT-dependent release of free [3H]-choline, [3H]-phosphocholine, and [3H]-glycerophosphocholine in BIN-67 cells Time of treatment (1026 M sCT)

0 3 5 15

min min min min

[3H]-choline (% of increase)

[3H]-phosphocholine (% of increase)

[3H]-glycerophosphocholine (% of increase)

100 108 6 4 121 6 16 105 6 7

100 103 6 3 97 6 8 98 6 3

100 100 6 5 93 6 11 98 6 14

BIN-67 cells were labelled with [3H]-choline chloride (74 kBq/ml) for 48 h and stimulated with 1026 M sCT for 10 min, then the total free [3H]-choline, [3H]-phosphocholine and [3H]-glycerophosphocholine were measured as described in Materials and Methods. Results are expressed as percentage vs. unstimulated samples. Data represent the mean 6 SE of three independent experiments.

of the DAG-dependent PKCd was not detected. To explain this phenomenon, we examined the cellular distribution of this isoenzyme by immunofluorescence but failed to detect any compartmentalization that could prevent its activation by surface receptor-coupled signals. Our results, however, were similar to those observed with angiotensin II, thrombin, and bradykinin receptors, which are known to generate DAG and translocate several PKC isoforms, including a and z, but fail to translocate PKCd in kidney proximal tubule cells, platelets, and fibroblasts, respectively (38 – 40). The porcine CTR has a predicted structure that substantially differs from the structure of the human CTR cloned from the ovarian carcinoma cell line, BIN-67 (7). The latter shows higher affinity for sCT relative to the porcine CTR. Although the two receptors are 75% identical, comparison of the predicted structures reveals a substantial difference, where the ovarian CTR contains a 48-bp cassette encoding a 16-amino acid insert located between the predicted first and second transmembrane domains. Recent reports have demonstrated that this insert greatly modifies the signaling properties of the receptor. In fact, low cAMP production has been demonstrated in response to CT, and PtdIns metabolism mediated by PLC is impaired in cells expressing the ovarian CTR receptor, with consequent block of Ca21 mobilization (7). In our hands, treatment of BIN-67 cells with either sCT or human CT failed to activate PLA-, PLC-, and PLD-induced PtdCho signals, as well as PKC activation of representatives of the classic (a), novel (d), and atypical (z) PKC isoenzymes. We cannot exclude that, similar to what observed for the PTH/PTHrp receptor (41, 42), the different responses to CT depend on cell type. However, in this study we demonstrated that PtdIns metabolism and PLD signaling were operative in the BIN-67 cells, as indicated by the bradykinindependent stimulation of InsP formation and [Ca21]i increase, and by the PMA-induced stimulation of PtdEtOH production, ruling out impairment of these pathways as mechanisms preventing PtdIns and Pthd-Cho metabolism in response to the hormone. Several lines of evidence indicate that the 48-bp cassette plays an inhibitory role in the signal transduction mechanism coupled to CTR. For example, in transfection experiments using the BHK-570 cell line, the CTR(48-), but not the CTR(481), isoform induce Ca21 signaling (9). Similarly, preliminary data obtained in our laboratory using 48(1) and 48(-) CTRs transiently transfected in COS-7 cells confirmed that the 48(1) isoform shows a response to CT strongly attenuated relative to the 48(-) isoform (43). Taken together, these observations provide a reasonable evidence that the lack of activation of intracellular signals is a property of the CTR(481) and not a cell-type specific phe-

nomenon. Analysis of the predicted amino acid sequence of the first intracellular domain, including the 16-amino acid insert, reveals that this domain lacks the consensus site for G protein binding and would, therefore, be predicted not to directly couple to these signaling molecules (7). The 16 amino acid insert may also modify the spatial conformation of the CTR that, although permitting binding of the hormone to the receptor, prevents its association with the phospholipasecoupled signal transduction pathways. In conclusion, we have demonstrated that the porcine CTR(48-) is coupled to novel PLD-signal transduction events leading to sustained DAG production and PKCa and possibly z activation in LLC-PK1 cells. Further studies are necessary to define the downstream events associated with the activation of these pathways. The failure of the CTR48(1) isoform to activate the phospholipase dependent pathways provides a novel mechanism by which structurally distinct CTR isoforms could produce specific cell responses by selective activation of different signals. Understanding of the molecular basis of this phenomenon may help to establish treatment strategies that will permit more effective use of CT. Acknowledgments We thank Walter Nardone, Evelina Tirone, and Sabrina Pagliei for their excellent technical help and Dr. Sergio Adamo for critically reading the manuscript.

References 1. Deftos LJ, Favus MJ ed 1996 Primer on the metabolic bone diseases and disorders of mineral metabolism. Philadelphia, NY, Lipincott-Raven Press, pp 82– 87 2. Goldring SR 1996 The structure and molecular biology of calcitonin receptor. In: Bilezikian JP, Raisz LG, Rodan GA (eds) Principles of Bone Biology. NY, Academic Press, pp 461– 470 3. Juppner H, Abou-Samra A, Freeman M, Kong XF, Schipani E, Richards J, Kolakowski LFj, Hock J, Potts Jr JT, Kronenberg HM, Segre GV 1991 A G-protein-linked receptor for parathyroid hormone and parathyroid hormonerelated peptide. Science 254:1024 –1026 4. Ishihara T, Nakamura S, Kazino Y, Takahashi T, Takahashi K, Nagata S 1991 Molecular cloning and expression of a cDNA encoding the secretin receptor. EMBO J 10:1635–1641 5. Lin YH, Harris TL, Flannery MS, Aruffo A, Kaji EH, Gorn AH, Kolakowski LFj, Lodish HF, Goldring SR 1991 Expression cloning of an adenylate cyclasecoupled calcitonin receptor. Science 254:1022–1024 6. Sexton PM, Houssami S, Hilton JM, O’Keefle LM, Center RJ, Gillespie MT, Darcy P, Findlay DM 1993 Identification of brain isoforms of the rat calcitonin receptor. Mol Endocrinol 7:815– 821 7. Gorn AH, Lin HY, Yamin M, Auron PE, Flannery MR, Tapp DR, Manning CA, Lodish HF, Krane SM, Goldring SR 1992 Cloning, characterization, and expression of a human calcitonin receptor from an ovarian carcinoma cell line. J Clin Invest 90:1726 –1735 8. Kuestner RE, Ebrod R, Grant FJ, Hagen FS, Kuijper JL, Matthwes SL, O’Hara PJ, Sheppard PO, Stroop SD, Thompson DL, Whitmore TE, Findlay DM, Houssami S, Sexton PM, Moore EE 1994 Cloning and characterization of an abundant subtype of the human calcitonin receptor. Mol Pharmacol 46:246 –255

3248


9. Moore EE, Kuestner RE, Stroop SD, Grant FJ, Matthewes SL, Brady CL, Sexton PM, Findlay DM 1995 Functionally different isoforms of the human calcitonin receptor result from alternative splicing of the gene transcript. Mol Endocrinol 9:959 –968 10. Quiza M, Dowton M, Perry KJ, Sexton PM 1997 Electrophoretic mobility and glycosilation characteristics of heterogeneously expressed calcitonin receptors. Endocrinology 138:530 –539 11. Force T, Bonventre JV, Flannery MR, Gorn AH, Yamin M, Goldring SR 1992 A cloned porcine renal calcitonin receptor couples to adenylyl cyclase and phospholipase C. Am J Physiol 262:F1110 –F1115 12. Teti A, Paniccia R, Goldring SR 1995 Calcitonin increases cytosolic free calcium concentration via capacitative calcium influx. J Biol Chem 270:16666 –16670 13. Chakraborty M, Chatterjee D, Gorelick FS, Baron R 1994 Cell cycle dependent and kinase specific regulation of the apical Na/H exchanger and the Na, K-ATPase in the kidney cell line LLC-PK1 by calcitonin. Proc Natl Acad Sci USA 91:2115–2119 14. Chakraborty M, Chatterjee D, Kellokumpu S, Rasmussen H, Baron R 1991 Cell cycle-dependent coupling of the calcitonin receptor to different G proteins. Science 251:1078 –108 15. Goldring SR, Dayer JN, Ausiello DA, Krane SM 1978 A cell strain cultured from porcine kidney increases cyclic AMP content upon exposure to calcitonin or vasopressin. Biochem and Biophys Res Commun 83:434 – 440 16. Naro F, Donchenko V, Minotti S, Zolla S, Molinaro M, Adamo S 1997 Role of phospholipase C and D signalling pathways in vasopressin-dependent myogenic differentiation. J Cell Physiol 17:34 – 42 17. Cook SJ, Wakelam MJO 1989 Analysis of the water-soluble products of phosphatidylcholine breakdown by ion-exchange chromatography. Biochem J 263:581–587 18. Teti A, Naro F, Molinaro M, Adamo S 1993 Differential response of myogenic cell lines to AVP: modifications of phosphoinositide breakdown, cytosolic free calcium concentration and intracellular pH, and dependency on protein kinase C activity. Am J Physiol 265:C113–C121 19. Nervi C, Benedetti L, Donchenko V, Naro F, Di Noi A, Molinaro M, Adamo S 1995 Regulation of skeletal muscle differentiation by arginine-vasopressin. In: Waxmann S (ed) Differentiation Therapy, Rome, Ares Serono Symposia Publ., pp 193–197 20. Heller M 1978 Phospholipase D. Adv Lipid Research 16:267–326 21. Bligh EG, Dyer WJ 1959 A rapid method of total lipid extraction and purification. Can J Biochem Physiol 37:911–917 22. Baldini PM, Zannetti A, Donchenko V, Dini L, Luly P 1992 Insulin effect on isolated rat hepatocytes: Diacylglycerol-phosphatidic acid interrelationship. Biochem Biophys Acta 1992 1137:208 –214 23. Bocckino SB, Blackmore PF, Wilson PB, Exton JH 1987 Phosphatidate accumulation in hormone-treated hepatocytes via a phospholipase D mechanism. J Biol Chem 262:15309 –15315 24. Exton JH 1994 Phosphatidylcholine breakdown and signal transduction. Biochem Biophys Acta 1212:26 – 42 25. Pappu AS, Hauser G 1983 Propranolol-induced inhibition of rat brain cytoplasmic phosphatidate phosphohydrolase. Neurochem Res 8:1565–1575 26. Qian A, Drewes LR 1991 Cross-talk between receptor-regulated phospholipase D and phospholipase C in brain. FASEB J 5:315–319 27. Chambers TJ, McSheehy PMJ, Thomson BM, Fuller K 1985 The effects of

28. 29. 30. 31.

32. 33. 34. 35. 36. 37.

38. 39. 40. 41. 42.

43.

Endo • 1998 Vol 139 • No 7

calcium-regulating hormones and prostaglandins on bone resorption by osteoclasts disaggregated from neonatal rabbit bones. Endocrinology 116:234 –239 Sexton PM, Adam WR, Hoseley SM, Martin TJ, Mendelshon FAO 1987 Localization and characterization of renal calcitonin receptors by in vitro autoradiography. Kidney Int 32:862– 868 Su Y, Chakraborty M, Nathanson MH, Baron R 1992 Differential effects of the 39,59-cyclic adenosine monophosphate and protein kinase C pathways on the response of isolated rat osteoclasts to calcitonin. Endocrinology 131:1497–1502 Moonga BS, Alan AS, Bevis PJ, Avaldi F, Sorcini R, Huang CL, Zaidi M 1992 Regulation of cytosolic free calcium in isolated rat osteoclats by calcitonin. J Endocrinol 132:241–249 Chabre O, Conklin BR, Lin HY, Lodish HF, Wilson E, Ives HE, Catanzariti L, Hemmings BA, Bourne HR 1992 A recombinant calcitonin receptor independently stimulates 39,59-cyclic adenosine monophosphate and Ca21/ inositol phosphate signaling pathways. Mol Endocrinol 6:551–556 Orcel P, Takahashi K, Okamoto T, Krane SM, Goldring SR, Nishimoto I 1995 Identification and characterization of two Gs protein coupling sequences in the calcitonin receptor. Bone 16:28 (Abstract) Siddiqui RA, Yang Y-C 1995 Interleukin-11 induces phosphatidic acid formation and activates MAP kinase 3T3–L1 cells. Cell Signal 7:247–259 Limatola C, Barabino B, Nista A, Santoni A 1997 Interleukin 1-b-induced protein kinase C-z activation is mimicked by exogenous phospholipase D. Biochem J 321:497–501 Wada S, Martin TJ, Findlay DM 1995 Homologous regulation of the calcitonin receptor in mouse osteoclast-like cells and human breast cancer T47D cells. Endocrinology 136:2611–2621 Findlay DM, Michelangeli VP, Moseley JM, Martin TJ 1981 Calcitonin binding and deregulation by two cultured human breast cancer cell lines (MCF7 and T47D). J Biochem 196:513–520 Findlay DM, Houssami S, Christopoulos G, Sexton PM 1996 Homologous regulation of the rat C1a calcitonin receptor (CTR) in nonosteoclastic cells is independent of CTR messenger ribobucleic acid changes and cyclic adenosine 39-59-monophosphate-dependent protein kinase activation. Endocrinology 137:4576 – 4585 Karim Z, Defontaine N, Paillard M, Poggioli J 1995 Protein kinase C isoforms in rat kidney proximal tubule: acute effect of angiotensin II. Am J Physiol 269:C134 –C140 Baldassare JJ, Henderson PA, Burns D, Loomis C, Fisher GJ 1992 Translocation of protein kinase C isoenzymes in thrombin-stimulated human platelets. Correlation with 1:2-diacylglycerol. J Biol Chem 267:15585–15590 Tippmer S, Quitterer U, Kolm V, Faussner A, Roscher A, Mosthaf L, MullerEsterl W, Haring H 1994 Bradykinin induces translocation of the protein kinase C isoforms a, e, and z. Eur J Biochem 225:297–304 Massfelder T, Helwig J-J, Stewart AF 1996 Parathyroid hormone-related protein as a cardiovascular regulatory protein. Endocrinology 137:3151–3153 Hoffermanns S, Iida-Klein A, Segre GV, Simon MI 1996 G alpha q family members couple parathyroid hormone (PTH)/PTH-related peptide and calcitonin receptors to phospholipase C in COS-7 cells. Mol Endocrinol 10:566 –574 Perez M, Naro F, Galson DL, Orcel P, Migliaccio S, Teti A, Goldring SR 1997 The calcitonin receptor and Ca21/diacylglycerol-dependent cell signals. J Bone Miner Res 11:F387 (Abstract)