The Calcitonin Receptor Stimulates Shc Tyrosine Phosphorylation and Erk1/2 Activation

While it is well established that adenylyl cyclase and phospholipase C-β are two proximal signal effectors for the calcitonin receptor, the more distal signaling pathways are less well characterized. G protein-coupled receptors can activate Erk1/2 by Gs-, Gi-, or Gq-dependent signaling pathways, depending on the specific receptor and cell type examined. Since the calcitonin receptor can couple to all three of these G proteins, the ability of calcitonin to activate Erk1/2 was investigated. Calcitonin induced time- and concentration-dependent increases in Shc tyrosine phosphorylation, Shc-Grb2 association and Erk1/2 phosphorylation and activation in a HEK 293 cell line that stably expresses the rabbit calcitonin receptor C1a isoform. Pertussis toxin, which inactivates Gi, and calphostin C, a protein kinase C inhibitor, each partially inhibited calcitonin-induced Shc tyrosine phosphorylation, Shc-Grb2 association, and Erk1/2 phosphorylation. In contrast, neither forskolin nor H89, a protein kinase A inhibitor, had a significant effect on basal or calcitonin-stimulated Erk1/2 phosphorylation. Our results suggest that the calcitonin receptor induces Shc phosphorylation and Erk1/2 activation in HEK293 cells by parallel Gi- and PKC-dependent mechanisms. The calcitonin-induced elevation of cytosolic free Ca2+ was required for Erk1/2 phosphorylation, since preventing any change in cytosolic free Ca2+ by chelating both cytosolic and extracellular Ca2+ abolished the response. However, the change in Ca2+ that is induced by calcitonin is not sufficient to account for the calcitonin-induced Erk1/2 phosphorylation, since treatment with 100 nm ionomycin or 10 μm thapsigargin, each of which induced elevations of Ca2+ comparable to those induced by calcitonin, induced significantly less Erk1/2 phosphorylation than that induced by calcitonin. Erk1/2 may have important roles as downstream effectors mediating cellular responses to calcitonin stimulation.

adrenergic receptor (AR) and M 2A -muscarinic acetylcholine receptor activate Erk1/2 by a mechanism that involves the sequential tyrosine phosphorylation of the adaptor protein Shc, association of phosphorylated Shc with the Grb2⅐mSOS complex, and activation of Ras (12). The same pathway is used by the receptors for EGF and other growth factors. The G q -coupled receptors ␣ 1B -AR and M 1A -muscarinic acetylcholine receptor can activate Erk1/2 via a PKC-dependent but Ras-independent pathway in some cells (12) and via a Ras-dependent pathway in others (13,14). cAMP, which antagonizes Erk1/2 activation in some cells (15)(16)(17)(18), activates Erk1/2 in PC12 cells (19) and in COS-7 cells (20).
Many GPCRs couple to more than one type of heterotrimeric G protein. For example, the thrombin and lysophosphatidic acid (LPA) receptors couple to G i , G q , and G 12/13 , and the thyrotropin receptor couples to G s , G i , G q , and G 12/13 (21)(22)(23). Interestingly, the LPA receptor-dependent activation of Erk1/2 is virtually abolished by pertussis toxin, indicating that activation is primarily via G i with little contribution from G q or G 12/13 (24). The PTH/PTHrP receptor also couples to at least two different types of heterotrimeric G proteins, G s and G q (25). Recent reports indicate that the PTH/PTHrP receptor can regulate Erk1/2 activity by different pathways, with the specific mechanism and resulting effect on Erk1/2 activity depending on the cellular context (15,26). These observations suggest that the mechanisms by which individual receptors regulate Erk1/2 activity are highly receptor-and cell type-specific.
Since the CTR is capable of coupling to G s , G i , and G q , we examined the ability of CT to regulate Erk1/2 activity. We report here that calcitonin induced Shc tyrosine phosphorylation, Shc-Grb2 association, and Erk1/2 activation in HEK 293 cells that stably express the rabbit osteoclast CTR C1a isoform. Additive effects of pertussis toxin and PKC inhibition suggest that both G i -dependent and pertussis toxin-insensitive PKCdependent mechanisms contribute to the CTR-induced activation of Erk1/2. Erk1/2 are therefore downstream effectors of the CTR which may mediate certain cellular responses to CT.
Stable Cell Line and Transient Transfections-HEK 293 cell stably expressing the rabbit CTR C1a (C1a-HEK)(4) were maintained in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum and 50 units/ml penicillin (Sigma) and 50 g/ml streptomycin (Sigma) in 10-cm plates until 80% confluent. Cells were then incubated in DMEM with 0.5% fetal bovine serum for 18 h prior to the treatments. For transient transfection, the HEK 293 cells were plated in 6-well plates until 80 -90% confluent and transfected with 0.4 g/well of CTR C1a in the presence of either 1 g/well empty vector pRK5 or 1 g/well ␤ARKct using LipofectAMINE reagent (Life Technologies). The cells were cultured in DMEM with 0.5% fetal bovine serum for 18 h prior to stimulation with CT.
Cell Stimulation and Preparation of Lysates-Cells were treated with agents as indicated in the figure legends. After treatment, cells were washed once with ice-cold phosphate-buffered saline and lysed in modified RIPA buffer (150 mM NaCl, 50 mM Tris, pH 7.5, 0.25% sodium deoxycholate, 0.1% Nonidet P-40, 1 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride, 1 mM sodium fluoride, 10 g/ml aprotinin, 10 g/ml leupeptin). Cell lysates were centrifuged for 30 s at 12,000 ϫ g to pellet insoluble material. Protein concentrations were assayed with Bradford protein assay reagent (Bio-Rad).
Detection of Shc Tyrosine Phosphorylation and Shc-Grb2 Complex Formation-For analysis of Shc tyrosine phosphorylation and its association with Grb2, 1.5 mg of cell lysate protein were incubated with 5 g of polyclonal anti-Shc antibody plus 50 l of a 50% slurry of protein A-Sepharose (Sigma) with mild agitation for 1 h at 4°C. Immune complexes were washed three times with ice-cold modified RIPA buffer and denatured in Laemmli sample buffer (27). Following resolution by SDS-polyacrylamide gel electrophoresis, proteins were transferred to nitrocellulose membranes which were then cut at the position of the 30-kDa protein marker. The upper part of the membrane was immunoblotted with anti-phosphotyrosine antibody 4G10 (1:1000 dilution) to detect the extent of tyrosine phosphorylation of Shc. The membrane was stripped and reblotted with monoclonal anti-Shc antibody (1:1000 dilution) to determine the amounts of Shc proteins. The lower part of the membrane was immunoblotted with antibody against Grb2 (1:1000 dilution) to detect the presence of co-precipitated Grb2.
Erk1/2 Activity Assay-C1a-HEK cells were grown to 80% confluence, then the medium was replaced by DMEM with 0.5% fetal bovine serum and penicillin/streptomycin for 18 h. The next day, the cells were incubated with or without the MEK inhibitor PD 98059 (50 M) for 30 min prior to the addition of 1 nM CT or 10 ng/ml EGF. After 5-min exposure to the ligands, the cells were harvested in 50 mM ␤-glycerophosphate, pH 7.3, containing 1.5 mM EGTA, 1 mM EDTA, 1 mM dithiothreitol, 1 mM sodium fluoride, 10 g/ml aprotinin, 10 g/ml leupeptin (28). The cells were homogenized by passing through a 27gauge needle five times, and the homogenates centrifuged for 30 min at 12,000 ϫ g. The kinase assay was performed in 25 l of kinase buffer containing 50 mM HEPES, pH 7.4, 10 mM MgCl 2, and 50 M ATP (specific activity 0.8 Ci/mmol). MBP(95-98) substrate peptide was added at a final concentration of 1 g/l. The reaction was initiated by adding 5 g of cellular extracts to the above assay buffer, allowed to proceed for 30 min at room temperature, and terminated by spotting the reaction mixture onto Whatman p81 phosphocellulose paper. Free [␥-32 P]ATP was removed by three extensive washes in 150 mM phosphoric acid. The p81 phosphocellulose paper was rinsed briefly in acetone and the radioactivity incorporated into MBP peptide was measured by scintillation counting.
Measurement of Cytosolic Free Ca 2ϩ Concentration ([Ca 2ϩ ] i )-C1a-HEK cells were loaded with indo-1 by incubation in conditioned medium with indo-1 AM (2 M) for 30 min at 37°C. Cells were then washed and harvested by brief exposure to nominally Ca 2ϩ -and Mg 2ϩ -free buffer containing trypsin (0.05%) and EDTA (0.5 mM). Conditioned medium was added to inactivate trypsin, following which cells were sedimented and resuspended in HEPES-buffered minimum essential medium. Aliquots of cell suspension were sedimented and resuspended in 2 ml of continuously stirred Na ϩ -HEPES buffer (135 mM NaCl, 5 mM KCl, 1 mM MgCl 2 , 1 mM CaCl 2 , 10 mM glucose, and 20 mM HEPES, pH 7.3, 290 mOsmol/liter) in a fluorimetric cuvette maintained at 37°C. In experiments investigating dependence on extracellular Ca 2ϩ , cells were suspended in nominally Ca 2ϩ -free Na ϩ -HEPES buffer or in Ca 2ϩ -containing Na ϩ -HEPES buffer to which EGTA (5 mM) was added. For most experiments, solutions in the cuvette contained 0.5% fetal bovine serum. Where indicated, cells were loaded with BAPTA-AM (50 M) for 30 min concomitantly with indo-1 AM. Test substances were added directly to the cuvette.
[Ca 2ϩ ] i was monitored using a dual wavelength fluorimeter (model RF-M2004, from Photon Technology International, London, Canada) with excitation wavelength of 355 nm and emission wavelengths of 405 and 485 nm. The system software was used to subtract background fluorescence and calculate the ratio R, which is the fluorescence inten-sity at 405 nm divided by the intensity at 485 nm.
where K d (the dissociation constant for the indo-1-Ca 2ϩ complex) was 250 nM, R min and R max were the values of R at low and saturating concentrations of Ca 2ϩ , respectively, and ␤ was the ratio of the fluorescence at 485 nm measured at low and saturating Ca 2ϩ concentrations (29).
Peak Ca 2ϩ responses were quantified as the maximum elevation of [Ca 2ϩ ] i above basal levels induced by the test substance. Plateau phases of the responses were quantified as the value of [Ca 2ϩ ] i above basal levels, 150 s after addition of the test substance. Data presented are representative traces or mean Ϯ S.E. of the number of separate samples indicated.

Calcitonin Induces Transient and Concentration-dependent
Activation of Erk1/2-To examine the effect of CT on Erk1/2 phosphorylation, C1a-HEK cells were treated with 1 nM CT for increasing periods of time, then lysed and analyzed for phosphorylated Erk1/2 as described under "Experimental Procedures." CT induced rapid phosphorylation of Erk1/2, which was apparent as early as 30 s, reached a maximum at 5 min, and decreased to a low level by 60 min (Fig. 1A, upper panel). Parallel samples blotted with Erk1/2 antibody (lower panel) showed no change in the amount of Erk1/2 protein. The cells express predominantly the 42-kDa Erk2 protein, and both forms are phosphorylated at levels proportional to their expression. The effect of CT on Erk1/2 phosphorylation, measured after 5 min of treatment, was concentration-dependent. A detectable increase in Erk1/2 phosphorylation was induced by as little as 0.001 nM CT and the maximum response occurred at concentrations of 1 nM or higher. The maximum phosphorylation induced by CT was comparable to that induced by 10 ng/ml EGF (Fig. 1B).
To confirm that the increase in Erk1/2 phosphorylation reflected the activation of the enzyme, we directly measured Erk1/2 activity. C1a-HEK cells were serum-starved overnight and subsequently treated with 1 nM CT for 5 min. Erk1/2 activity was determined by measuring [␥-32 P]ATP incorporation into an Erk1/2-specific substrate peptide as described under "Experimental Procedures." The enzymatic activity was expressed as fold increase relative to the activity of untreated cells. CT induced a 2.3-fold increase in Erk1/2 activity, while EGF (10 ng/ml) induced a 2.6-fold increase in Erk1/2 activity (Fig. 2B).
The phosphorylation and activation of Erk1/2 is catalyzed by ERK kinases 1 and 2 (MEK1/2) (9). To further confirm that Erk1/2 phosphorylation correlates with activity, we examined the effect of the MEK-specific inhibitor PD 98059 on Erk activity and phosphorylation. As expected, pretreatment of the cells with 50 M PD 98059 for 30 min almost completely abolished both the phosphorylation ( Fig. 2A) and activation (Fig.  2B) of Erk1/2 induced by CT or by EGF.
The Cyclic AMP-Protein Kinase A Pathway Has Little Effect on Erk1/2 Phosphorylation-GPCR may regulate Erk1/2 activity via one or more of several different pathways involving G s -, G i -, and G q -dependent mechanisms, depending on the receptors and the cellular context (11). Since the CTR is capable of coupling to G s , G i , and G q , we investigated which pathways are involved in CT-induced Erk1/2 activation. Although cAMP generally attenuates growth factor-induced Erk1/2 activation (16 -18), the activation of Erk1/2 by cAMP has been reported (19,20). We therefore examined whether cAMP participated in CT-induced Erk1/2 activation (Fig. 3). Serum-starved C1a-HEK cells were treated with 200 M forskolin for 25 min alone or followed by 1 nM CT for an additional 5 min. Forskolin induced only a slight increase in Erk1/2 phosphorylation (lane 3) and had no apparent effect on CT-induced Erk1/2 phosphorylation (lane 4). Furthermore, the protein kinase A-specific inhibitor H89 (10 M) had little effect on CT-stimulated Erk1/2 phosphorylation (lane 6). Taken together, these data suggest that the cAMP/PKA pathway does not contribute significantly to CT-induced Erk1/2 activation in HEK-293 cells.
Both G i and PKC Mediate CT-induced Shc Tyrosine Phosphorylation and Erk1/2 Phosphorylation-In HEK 293 cells, both PTX-sensitive G i -coupled receptors and PTX-insensitive G q -coupled receptors can activate Erk1/2 via the Shc/Grb2/Ras pathway (13) by which receptor tyrosine kinases such as the EGF receptor activate Erk1/2 (30). To determine whether Shc also plays a role in the mechanism of CT-induced Erk1/2 activation, the tyrosine phosphorylation of Shc and its association with Grb2 were analyzed. C1a-HEK cells were treated with 1 nM CT for various times and lysed as described under "Experimental Procedures." Shc was immunoprecipitated from 1.5 mg of cell lysates and the immunoprecipitated proteins were analyzed for Shc phosphorylation and Grb2 association as described under "Experimental Procedures." As shown in Fig. 4A, 1 nM CT induced the tyrosine phosphorylation of predominantly the 52-kDa Shc, which reached a maximum level at 5-10 min and remained elevated during the time course examined for up to 120 min. The broad band below 52-kDa Shc was a nonspecific interaction of the immunoprecipitation antibody with the immunoblotting antibodies, since it was present in mock immunoprecipitations performed in the absence of cell lysate (not shown). The amount of immunoprecipitated Shc is shown in the middle panel of Fig. 4A. The induction of Shc tyrosine phosphorylation was also concentration-dependent, with the maximum effect induced by CT concentrations Ն1 nM (Fig. 4B). Co-precipitated Grb2 was readily detectable in the samples in which Shc was phosphorylated. As expected, EGF also induced Shc tyrosine phosphorylation and its association with Grb2 (Fig. 4B), although to a much greater extent than did CT.
Since the CTR can couple to both G i and G q , we examined the relative roles of PTX-sensitive and -insensitive G proteins in the CT-induced Shc tyrosine phosphorylation and Erk1/2 phosphorylation. C1a-HEK cells were pretreated with 200 ng/ml PTX for 18 h prior to stimulation with 1 nM CT or 10 M LPA. LPA was used as a positive control since it is reported to activate Erk1/2 via G i coupling (31). Total cell lysates were analyzed for Shc tyrosine phosphorylation, Shc-Grb2 association, and Erk1/2 phosphorylation. As shown in Fig. 5A, PTX partially inhibited CT-and LPA-induced Shc tyrosine phosphorylation and Shc-Grb2 association. PTX completely abolished LPA-stimulated Erk1/2 phosphorylation but only partially inhibited the phosphorylation induced by CT (Fig. 5B).
A peptide derived from the carboxyl terminus of the ␤-adrenergic receptor kinase 1 (␤ARK1ct) has been reported to block signaling mediated by the ␤␥ subunits of the G i proteins, including the activation of Erk1/2 (13). To demonstrate further that the CTR-mediated Erk1/2 phosphorylation involves G i coupling, HEK 293 cells were transiently transfected with CTR-C1a and pRK5 or CTR-C1a and pRK-␤ARK1ct. As shown in Fig. 5C, ␤ARK1ct reduced the CT-induced Erk1/2 phosphorylation relative to the amount seen in cells transfected with the empty pRK5 vector, indicating the involvement of ␤␥ subunits and supporting the involvement of G i in the CTR-Erk1/2 coupling mechanism.
Since CT stimulates PKC activity in C1a-HEK cells, 2 we sought to determine the involvement of PKC in the process of CT-dependent Erk1/2 phosphorylation. Subconfluent C1a-HEK cells were incubated overnight with or without PTX, then treated with the PKC-specific inhibitor calphostin C for 30 min prior to the addition of 1 nM CT. The cells were harvested and analyzed for Shc phosphorylation and Shc⅐Grb2 complex formation (Fig. 6A) and for Erk1/2 phosphorylation (Fig. 6B). Like PTX, calphostin C only partially blocked the CT-induced Shc phosphorylation and Erk1/2 phosphorylation, while the Shc tyrosine phosphorylation and Erk1/2 phosphorylation induced by 100 nM PMA were largely inhibited by the same concentration of calphostin C. In addition, the effects of PTX and calphostin C on Shc and Erk1/2 phosphorylation were additive. The results suggest that signal transduction from CTR to Shc and Erk1/2 involves both G i -dependent and PKC-dependent mechanisms.
CT-induced Increase in Cytosolic Calcium Is Required for Erk1/2 Phosphorylation-A recent report suggested that the increase in cytosolic Ca 2ϩ is one of the key mediators of both G iand G q -coupled adrenergic receptor-stimulated Erk1/2 phosphorylation in HEK 293 cells (13). On the other hand, the PTH/PTHrP receptor (structurally related to CTR) activated Erk1/2 via a Ca 2ϩ -independent mechanism in Chinese hamster ovary cells (26). Since CT increases [Ca 2ϩ ] i (32), we investigated the role of [Ca 2ϩ ] i in CT-induced Erk1/2 phosphorylation. [Ca 2ϩ ] i levels were manipulated using EGTA and BAPTA to chelate extracellular and intracellular Ca 2ϩ , respectively, the calcium ionophore ionomycin to induce the elevation of [Ca 2ϩ ] i , and thapsigargin to deplete the intracellular Ca 2ϩ stores. Changes in [Ca 2ϩ ] i were monitored by fluorescence spectroscopy as described under "Experimental Procedures." Consistent with earlier reports (32, 33), 1 nM CT induced a biphasic elevation of [Ca 2ϩ ] i , which rose within 30 s from basal levels of 134 Ϯ 19 nM to peaks of 319 Ϯ 37 nM above basal levels, followed by a sustained plateau of 57 Ϯ 5 nM above basal levels (n ϭ 7, Fig. 7B). The earlier reports showed that the initial transient results from the release of Ca 2ϩ from intracellular stores and the sustained phase results from the influx of FIG. 3. Forskolin and H89 do not affect CT induced-Erk1/2 phosphorylation. Subconfluent C1a-HEK cells were serum-starved as described in the legend to Fig. 1, then treated with 200 M forskolin for 25 min alone or followed by 1 nM CT for additional 5 min; or the cells were incubated with 10 M H89 for 25 min alone or followed by 1 nM CT for 5 min. Erk1/2 phosphorylation and total Erk1/2 were determined as described in the legend to Fig. 1.

FIG. 4. Calcitonin induces tyrosine phosphorylation of Shc and the association of Shc with Grb2
. Subconfluent C1a-HEK cells were serum-starved as described in the legend to Fig. 1 and subsequently treated (A) with 1 nM CT for the indicated times or (B) treated with the indicated concentration of CT or 10 ng/ml EGF for 5 min. Cells were lysed as described in Fig. 1 and Shc was immunoprecipitated from 1.5 mg of total cell lysate as described under "Experimental Procedures." The immune complexes were analyzed for the tyrosine phosphorylation of Shc (upper panel), total Shc content (middle panel), and the presence of coprecipitated Grb2 (lower panel) as described under "Experimental Procedures." extracellular Ca 2ϩ (33). No change in fluorescence was detected when C1a-HEK cells were treated with vehicle alone (Fig. 7A). When cells were loaded with BAPTA and then suspended in nominally Ca 2ϩ -free Na ϩ buffer, the Ca 2ϩ response to CT was virtually abolished (Fig. 7E). Similarly, pretreatment of cells with 5 mM EGTA for 30 min markedly decreased the amplitude of both the peak and plateau phases of the [Ca 2ϩ ] i elevation induced by CT (Fig. 7F). These treatments abolished the CTinduced Erk1/2 phosphorylation (Fig. 9A, lanes 5 and 6), indicating that an increase in [Ca 2ϩ ] i is required for the CT-mediated response. In contrast, preincubation of the cells with EGTA for 30 min had little effect on EGF-induced Erk1/2 phosphorylation (Fig. 9A, lanes 7 and 8).
We next examined the relative contributions of the initial Ca 2ϩ transient and the sustained Ca 2ϩ plateau to CT-stimulated Erk1/2 phosphorylation. Incubation with 5 mM EGTA for 2 min prior to the addition of CT had little effect on the initial transient, but decreased the plateau phase to 12 Ϯ 3 nM above basal levels (n ϭ 7, Fig. 7C). Under these conditions, CTstimulated Erk1/2 phosphorylation was reduced by more than half (Fig. 9A, compare lanes 2 and 3), suggesting that while the CT-induced release of Ca 2ϩ from intracellular stores supports CTR/Erk coupling to some degree, it is not sufficient for the full response. In cells loaded with BAPTA, the initial Ca 2ϩ transient was abolished, while the sustained plateau (53 Ϯ 4 nM above basal levels) was the same as cells not loaded with BAPTA (Fig. 7, D and B, respectively). Under these conditions, BAPTA had little effect on CT-induced Erk1/2 phosphorylation (Fig. 9A, lane 4). Thus, a small sustained increase in [Ca 2ϩ ] i is necessary to support the full Erk1/2 response to CT.
The foregoing experiments do not, however, address the question of whether the receptor-mediated increase in [Ca 2ϩ ] i by itself is sufficient to induce the full receptor-mediated Erk1/2 phosphorylation, as suggested by Della Rocca et al. (13). To examine this question, the effect of CT on Erk1/2 phosphorylation was compared with the effect of ionomycin, a Ca 2ϩ ionophore, and thapsigargin, an inhibitor of the endoplasmic reticular Ca 2ϩ -ATPase that induces a transient increase in [Ca 2ϩ ] i by blocking the re-uptake of Ca 2ϩ into intracellular storage sites. The increase in [Ca 2ϩ ] i induced by 10 nM to 1 M ionomycin were determined (not shown), and the changes in [Ca 2ϩ ] i induced by 100 nM ionomycin (peaks 480 Ϯ 61 nM and plateaus 114 Ϯ 17 nM above basal levels, n ϭ 5, Fig. 8B) approximated the CT-induced [Ca 2ϩ ] i changes. However, the Erk1/2 phosphorylation induced by 100 nM ionomycin (Fig. 9C,  lane 3) was significantly smaller than that induced by 1 nM CT (Fig. 9C, lane 2). (A much higher concentration of ionomycin (3 M) induced Erk1/2 phosphorylation to a similar level as 1 nM CT. 2 ) Similarly, treatment of the cells with 1 M thapsigargin caused a biphasic increase in [Ca 2ϩ ] i that approximated the [Ca 2ϩ ] i response to CT (peaks 357 Ϯ 43 nM and plateaus 74 Ϯ 15 nM above basal levels, n ϭ 5, Fig. 8C), but resulted in a minimal increase in Erk1/2 phosphorylation (Fig. 9C, lane 4). When the cells were treated with 10 M thapsigargin for 25 min and subsequently treated with 1 nM CT for 5 min, CT did not further increase [Ca 2ϩ ] i over the plateau phase induced by thapsigargin (absolute [Ca 2ϩ ] i 314 Ϯ 16 nM, n ϭ 3, Fig. 8D), but Erk1/2 phosphorylation increased to a level similar to that seen in cells treated with CT alone (Fig. 9C, compare lanes 2 and 5).
These results suggest that the small sustained elevation of [Ca 2ϩ ] i induced by CT or by thapsigargin synergizes with other CT-stimulated signaling events to produce CTR-dependent Erk1/2 activation. To examine whether PKC-and Ca 2ϩ -dependent mechanisms are required simultaneously to achieve the full response, the cells were treated with calphostin C and EGTA. As shown in Fig. 9B, the combined treatment reduced Erk1/2 phosphorylation to a minimum level. DISCUSSION This study demonstrates that CT activates Erk1/2 in HEK 293 cells that express the C1a isoform of the rabbit CTR in a time-and concentration-dependent manner. The CT-induced activation of Erk1/2 is largely independent of cAMP/PKA-dependent pathways. Rather, in these cells the CTR activates the Shc/Grb2/Erk1/2 pathway via both G i -and PKC-coupled mech-anisms. The CTR-stimulated sustained increase in [Ca 2ϩ ] i is necessary to support the full CT-dependent Erk1/2 phosphorylation, but the concurrent activity of other signaling events (e.g. PKC activation) is required for the maximal CT-induced response.
Recent results from us and others have shown that signaling via the CTR involves the activation of at least two G proteins, G s and G q (2)(3)(4). In the present study, we demonstrate that the CTR can also activate G i since the CTR-stimulated events described here are partly PTX-sensitive. The CTR stimulates prolonged Shc tyrosine phosphorylation, as reported for some other GPCR, such as the thrombin and endothelin receptors, and for the insulin receptor or growth hormone receptor (34 -37). This prolonged Shc tyrosine phosphorylation is in contrast to the transient effects observed with the G i -coupled ␣ 2A -AR (31) or the G q -coupled ␣ 1B -AR (38). On the other hand, the CTR-induced Erk1/2 phosphorylation is transient. While CT induces prolonged Shc phosphorylation, it is likely that activated Erk1/2 trigger a negative feedback loop in which MAP kinase phosphatase(s) are activated and in turn dephosphorylate and inactivate Erk1/2, since treatment with the phosphatase inhibitor sodium orthovanadate prolonged CT-stimulated Erk1/2 phosphorylation. 2 The difference in the time course of phosphorylation of the two proteins is the subject of further investigation.
The finding that PTX and calphostin C have additive effects on both Shc and Erk phosphorylation is important for several reasons. It suggests that the G i -dependent and PKC-dependent mechanisms that are activated by the CTR each can activate the Shc/Grb2/Erk1/2 pathway independently of the other, and that the G i -and PKC-dependent mechanisms converge at or above the level of Shc in HEK 293 cells. Furthermore, it demonstrates for the first time that a single GPCR, the CTR, can activate Erk1/2 through parallel independent signaling pathways in the same cells.
Our results confirm the recent report of Della Rocca and colleagues (13) that both G i -and G q -coupled receptors activate Erk1/2 through the Shc/Grb2 pathway in HEK cells. However, our results differ in some respects. First, inhibition of PKC partially blocks the CTR-induced Erk1/2 phosphorylation, but has no effect on the responses elicited by the G i -coupled ␣ 1B -AR or the G q -coupled ␣ 2A -AR. This difference may be due to the receptors, the G protein isoforms or other intermediate effec- tors that are activated by the CTR in our study and by the ␣ 1B -AR and ␣ 2A -AR in the other. Second, these authors found that 10 M A23187 stimulated Erk1/2 phosphorylation to the same degree as the adrenergic receptors and that Erk phosphorylation was mostly inhibited by BAPTA. On this basis, they proposed that the receptor-induced increase in [Ca 2ϩ ] i leads to activation of tyrosine kinases and subsequently of the Shc/Ras/ Erk cascade. In contrast, we found that treatment with 100 nM ionomycin, which induced a [Ca 2ϩ ] i increase that was comparable to the CT-dependent increase in [Ca 2ϩ ] i , resulted in much lower Erk1/2 phosphorylation than was seen following CT treatment. Moreover, BAPTA did not inhibit the Erk1/2 phosphorylation induced by CT. Although the absence of data on the actual changes in [Ca 2ϩ ] i induced by the adrenergic receptors makes it impossible to fully compare these results, our results clearly demonstrate that the maximal phosphorylation of Shc and Erk1/2 in response to CT requires a sustained [Ca 2ϩ ] i response that is at least 40 -50 nM above basal levels, and that CT can stimulate Erk1/2 phosphorylation when the initial Ca 2ϩ transient is completely suppressed by BAPTA. Furthermore, the changes in [Ca 2ϩ ] i that are induced by stimulation of the CTR are not sufficient to cause the maximal phosphoryla-tion of Shc and Erk 1/2 in the absence of other signaling events.
The fact that CT-stimulated Erk1/2 phosphorylation requires sustained [Ca 2ϩ ] i elevation of 40 -50 nM above basal levels is in keeping with the recent report on differential activation of transcription factors by a small Ca 2ϩ plateau versus a large Ca 2ϩ transient in B-lymphocytes (39). In many situations, particularly in cells that express voltage-sensitive Ca 2ϩ channels, an influx of extracellular Ca 2ϩ is both necessary and sufficient for activation of Erk1/2 (40,41). However, in C1a-HEK cells, the increase in [Ca 2ϩ ] i alone is not sufficient to account for the full CT-induced increase in Erk1/2 phosphorylation, since 100 nM ionomycin or 1 M thapsigargin each induced changes in [Ca 2ϩ ] i comparable to those induced by CT, but stimulated little Erk1/2 phosphorylation. Furthermore, although challenging the thapsigargin-treated cells with CT failed to induce any additional change in [Ca 2ϩ ] i , the level of Erk1/2 activation was much higher in cells treated with both thapsigargin and CT than in cells treated only with thapsigargin. Thus, both the increase in [Ca 2ϩ ] i and the activation of other signaling effectors, including PKC, are required for the full CTR-dependent Erk1/2 response.
Although the main focus of investigations of Erk1/2 function has been on their role in mediating the mitogenic effects of diverse ligands for receptor tyrosine kinases and GPCR, these proteins were originally identified as enzymes that phosphorylate microtubule-associated proteins and thereby regulate cytoskeletal stability (42)(43)(44). It is now well established that Erk1/2 are involved in far more than just mitogenic signaling (10). The ability of the CTR to activate the Shc to Erk1/2 pathway by two independent mechanisms implies that Erk1/2 may play an important role in mediating the morphological and/or biochemical effects of CT on osteoclast.
The duration of Erk activation has been proposed to determine the cell fate in PC 12 cells (45). Nerve growth factor stimulates sustained Erk activation which leads to the outgrowth of neurites and cessation of cell division, whereas EGF induces transient Erk activation which correlates with cell proliferation. If the model of PC 12 cells can be extended to cells expressing CTR, CT-induced transient Erk1/2 activation may be responsible for the proliferative effect of CT on a human prostate cancer cell line (46). The effect of CT on the proliferation of T47D and MCF-7, two other carcinoma cell lines that express the CTR, is currently under investigation.
In conclusion, we have demonstrated that CT induces Shc tyrosine phosphorylation, Shc-Grb2 association and Erk1/2 activation in C1a-HEK by parallel independent pathways involving G i and PKC. Both pathways require the elevation of [Ca 2ϩ ] i , but the change in [Ca 2ϩ ] i that occurs in response to treatment with CT cannot by itself induce full Erk1/2 phosphorylation.