New signaling pathway for parathyroid hormone and cyclic AMP action on extracellular-regulated kinase and cell proliferation in bone cells. Checkpoint of modulation by cyclic AMP.

cAMP signaling, activated by extracellular stimuli such as parathyroid hormone, has cell type-specific effects important for cellular proliferation and differentiation in bone cells. Recent evidence of a second enzyme target for cAMP suggests divergent effects on extracellular-regulated kinase (ERK) activity depending on Epac/Rap1/B-Raf signaling. We investigated the molecular mechanism of the dual functionality of cAMP on cell proliferation in clonal bone cell types. MC3T3-E1 and ATDC5, but not MG63, express a 95-kDa isoform of B-Raf. cAMP stimulated Ras-independent and Rap1-dependent ERK phosphorylation and cell proliferation in B-Raf-expressing cells, but inhibited growth in B-Raf-lacking cells. The mitogenic action of cAMP was blocked by the ERK pathway inhibitor PD98059. In B-Raf-transduced MG63 cells, cAMP stimulated ERK activation and cell proliferation. Thus, B-Raf is the dominant molecular switch that permits differential cAMP-dependent regulation of ERK with important implications for cell proliferation in bone cells. These findings might explain the dual functionality of parathyroid hormone on osteoblastic cell proliferation.

RNA Analysis-Total RNA was extracted using guanidinium thiocyanate/phenol/chloroform method as reported by Chomczynski and Sacchi (34). Brain and calvaria were isolated from male ddY mice (Shimizu Experimental Supplies, Kyoto, Japan) and RNA samples were extracted. Northern blot analysis was performed under high stringency conditions as described previously (32). In brief, total RNA (20 g) was electrophoresed in 1.2% agarose-formaldehyde gels, transferred on nylon membrane filters (Hybond Nϩ, Amersham Biosciences, Buckinghamshire, UK), and hybridized with 32 P-labeled cDNA probes. cDNAs encoding PPR and glyceraldehyde-3-phosphate dehydrogenase cloned by polymerase chain reaction (PCR) were used as probes (32). After the final wash, the membrane was exposed to a BAS imaging plate (Fuji Film, Tokyo, Japan), and the relative signal intensity was estimated. For reverse transcription-PCR, a 0.5-g RNA aliquot was reverse transcribed at 37°C for 2 h in a 20-l reaction volume containing 200 units of Superscript II (Invitrogen, Gaithersburg, MD), 4 M random primers, 500 M dNTPs, and 5 mM dithiothreitol. PCR was performed using one-tenth of the reverse transcription reaction volume and 30 pmol of following oligonucleotides. For Epac1, Epac1-F, 5Ј-GCTTCCTCCA-CAAACTCTCA-3Ј, Epac1-R, 5Ј-AACGCTGCCATCACCTCTCT-3Ј (AN: NM_006105); for Epac2, Epac2-F, 5Ј-AGCCTTATCCCATCTTTCTA-3Ј, Epac2-R, 5Ј-CTGACTGTATTCGCCTCCAC-3Ј (AN: NM_007023); for HPRT as an internal control, HPRT-F, 5Ј-GTTGAGAGATCATCTC-CACC-3Ј, HPRT-R, 5Ј-AGCGATGATGAACCAGGTTA-3Ј. PCR was performed using Taq DNA polymerase on the following schedule: denaturation at 94°C for 1 min, annealing at 64°C for 1 min, and extension at 72°C for 2 min on a TAKARA PCR thermal cycler MP system (Shiga, Japan); 30 cycles for the Epac and 20 cycles for the HPRT. Half of each reaction product was loaded onto a 2% agarose gel in 1 ϫ Tris acetate-EDTA buffer. The number of cycles selected for each primer pair produced a linear relationship between the amount of input RNA and resulting PCR products.
cAMP Assay-Cells were washed twice with incubation buffer (␣-MEM containing 0.5 mM 1-methyl-3-isobutylxanthine (IBMX) and 1 mg/ml bovine serum albumin) and incubated for 30 min at 37°C in the same buffer containing various concentrations of test agents. The reaction was terminated with trichloroacetic acid. cAMP was measured by radioimmunoassay (Amersham Biosciences) and the protein concentrations were estimated using a BCA protein assay kit (Pierce).
Cell Growth Assay-3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) activity was measured using a colorimetric assay (35). Cells were plated in 96-well plates at a density of 30,000 cells/well. Cells were treated with or without various agents for 4 days. The cells were then gently washed twice with 100 l of phosphate-buffered saline and incubated at 37°C for 2 h after the addition of 100 l of 0.5 mg/ml MTT. Then, 50 l of solubilizing solution containing 20% sodium dodecyl sulfate (SDS), 50% dimethylformamide, 2% acetic acid, and 2.5% of 1 M HCl, pH 4.7, was added to extract the dark blue crystals. After complete extraction, the absorbance was measured on a Bio-Rad Model 550 microplate reader (Bio-Rad, Hemel Hempstead, UK), using a test wavelength of 570 nm and a reference wavelength of 655 nm. Bromodeoxyuridine (BrdUrd) incorporation assay was performed using the colorimetric BrdUrd Cell proliferation Kit (Roche Molecular Biochemicals, Mannheim, Germany) according to the manufacturer's protocol as described previously (36). There was no difference in either assay in the number of dead cells between the cell lines determined by a trypan blue exclusion assay. FCS was reduced to 1% for all treatment conditions.
Establishment of Stable Transformants-The day before transfection, cells were plated on 35-mm dishes at a density of 10 5 cells/ml, and then transfected by FuGENE 6 (Roche Molecular Biochemicals) according to the manufacturer's protocol as described previously (37). The vector containing both the green fluorescent protein and neomycinresistant genes, pEGFP-N1 (CLONTECH, Valencia, CA), and the respective vectors were co-transfected in the cells. Subconfluent cells were trypsinized and plated at low density before selection. Subsequently, cells were selected by culturing them in the presence of 400 g/ml G418 for 3 weeks. Three clones of each mutant from MC4 and MG63 were isolated. In this study, number 004, N17Rap1 MC4; number 005, V12 Rap1 MC4; number 001, N17 Ras MC4; number 021, N17Rap1 MG63; number 008, V12 Rap1 MG63; number 015, B-Raf MG63 were used.
Statistical Analysis-Unless otherwise described, statistical analyses were performed using Student's t test. A p value of less than 0.05 was considered to be significant. Two-way analysis of variance was performed to determine statistical differences between cultures according to time in culture.

RESULTS
Epac, Rap1, and B-Raf Expression-B-Raf exists as a number of isoforms that are expressed primarily in neural and endocrine tissues (26,38). Although the 95-kDa B-Raf isoform is found in both brain and spinal cord, its presence in bone cells has not been systematically explored. We measured B-Raf protein expression in cultures of fibroblast C3H10T1/2, osteoblast MC4, chondrocyte ATDC5, two osteocytic clones MLO-Y4 and MLO-A5, two osteosarcoma cell lines, ROS17/2.8 and MG63, and myoblast C2C12 cells by Western blotting using an antibody specific for B-Raf (Fig. 1C, lower panel). ATDC5, MC4, C3H10T1/2, and the two types of MLO cells had both the 95and 62-kDa B-Raf splice variants, and C2C12, MG63, and ROS17/2.8 cells expressed only the 62-kDa isoform. Additionally, Epac1 and Epac2, two isoforms of Epac, and Rap1 protein expression were investigated (Fig. 1). In vivo bone expressed Epac mRNA (Fig. 1A). Similar Epac mRNA expression was observed in brain and PC12 cells. Epac2 expression was detected in all cells (Fig. 1B). In contrast, Epac1 was not detected in ATDC5, C2C12, and ROS17/2.8, and Rap1 expression was limited and weak in C2C12 and ROS17/2.8 cells. Either expression pattern of 95-kDa B-Raf, Rap1, or Epac, however, failed to show any notable change during maturation (Fig. 1D, right), while ERKs (p44 and p42) were gradually decreased (Fig. 1D, left). Taken together, these results suggest that, among cell types tested, only primary calvaria cells predominantly express the 95-kDa B-Raf isoform while the other limited lines the 62-kDa isoform alone and this allows cell type-specific cAMP signal transduction and diverse proliferative reactions.
PPR Expression and cAMP Accumulation Triggered by PTH-(1-34)-Next, we performed Northern blot analysis using specific probes for PPR and examined cAMP accumulation levels triggered by PTH-  in some clonal cells (Fig. 2). PPR mRNA expression was detected in ATDC5, MC3T3-E1, and MC4 cells. The MC4 cells expressed 10-fold more PPR mRNA compared with other lines ( Fig. 2A). As expected, PTH-(1-34) dose-dependently stimulated cAMP accumulation in these cells (Fig. 2B). In MC4, cAMP accumulation levels were ϳ100-fold compared with ATDC5 and MC3T3-E1. On the other hand, the increased cAMP induced by PTH-(1-34) was negligible and barely detectable in MG63, C3H10T1/2, and the two MLO cell lines (data not shown), although high enough to suppress ERK activity and cell proliferation (refer to Figs. 3B and 6A). In development model using primary osteoblastic cells, Northern blot analysis showed that PPR mRNA expression levels were gradually increased up to 21 days and then gradually decreased (Fig. 2C). Similar results were reported in the MC3T3-E1 and ATDC5 developmental culture system (17,39). In contrast, expression of ERK had a peak at the start of culture. Since the Epac pathway is expressed constantly, the ERK level peaked at the start may be an important variable to assure proliferative response toward cAMP signal that increases the cell number, as the main determinant of bone mass. These results suggest that increased intracellular cAMP is perhaps the major PTH-induced signaling mechanism in these osteoblastic cells during cell proliferation.
cAMP Stimulates ERK Phosphorylation in 95-kDa B-Rafexpressing Bone Cells-To determine the consequences of elevated intracellular cAMP on bone cellular ERK activity, we measured ERK activity using antibodies specific for the active phosphorylated ERK. For these experiments, a variety of agents that can increase intracellular levels of cAMP or act as Transcripts encoding Epac1 and Epac2 were analyzed by reverse transcriptase-PCR using specific primer pairs, yielding amplified cDNA products of 304 and 333 bp, respectively. Amplified products were verified by subcloning and sequence analyses. In addition to brain and PC12, Epac1 and Epac2 were expressed in bone and bone-related clonal lines. C, Rap1 and B-Raf expression in bone-like cells. Using PC12 lysates as a positive control, Rap1 (upper) and B-Raf (lower) expression was examined. Cell extracts were loaded detecting for Rap1 (5 g) and for B-Raf (10 g). Rap1 was expressed ubiquitously, but the 95-kDa form of B-Raf was restricted. In C2C12, MG63, and ROS17/2.8, only the 62-kDa form of B-Raf was detected. D, developmental expressions of Rap1, B-Raf, and Epac in primary calvarial osteoblastic cells. Cell lysates were prepared on the indicated day of culture. Rap1, B-Raf, and Epac were not regulated during osteoblast development, while ERKs (p44 and p42) were gradually decreased. Data are representative of four separate experiments. a cAMP analogue, forskolin, dibutyryl-cAMP (Bt 2 cAMP), and 8-(4-chlorphenylthio)-cyclopencyladenosine, were used. Only clonal cells expressing 95-kDa B-Raf, PTH-(1-34), forskolin, Bt 2 cAMP, and 8-(4-chlorphenylthio)-cyclopencyladenosine (8-CPT) increased the activity of ERK 2-12-fold more than in unstimulated cells (Fig. 3, B and C, and data not shown). Maximal activation by 1 nM PTH-(1-34) occurred rapidly, within 5-30 min and its activation was reduced by a higher concentration of the peptide (Fig. 3A). IBMX increases cAMP levels by inhibiting its degradation by cAMP phosphodiesterase. IBMX also potentiated ERK activity (Fig. 3C), indicating that the potentiation was due to increased intracellular cAMP. PTH-(1-34)-induced ERK phosphorylation was completely inhibited by the ERK pathway inhibitor PD98059. When cells were treated with H89, low concentrations of PTH-(1-34)-induced ERK activation were not affected, while increased ERK activation was observed with high concentrations of the peptide (Fig. 3D). These results suggest that PTH stimulates intracellular cAMP accumulation, cAMP then activates Rap1 via GEF molecules, both Epac dependently and PKA independently, which in turn leads to B-Raf activation and results in ERK activation. Therefore, PKA might inhibit cAMP-mediated ERK activation when excess intracellular cAMP accumulates. In our preliminary experiments, calcitonin also stimulated cAMP-ERK signaling mechanisms in two MLO cell lines that express functional calcitonin receptors. 2 Therefore, cAMP-induced ERK activation by other hormones and factors was thought to be a common regulatory pathway in bone cells.
cAMP Stimulates ERK Phosphorylation and Cell Prolifera-tion in 95-kDa B-Raf-expressing Cells-In neuronal cells, cAMP signaling has an important role in cell differentiation and survival through a Rap1-B-Raf expression-dependent mechanism (24). Activated ERK provides a mitogenic and a differentiating signal in many cell types (22,23). To determine whether cAMP stimulates cell proliferation, we examined cell proliferation of bone cells assessed by MTT and BrdUrd incorporation assays. MTT activity and BrdUrd incorporation were directly correlated with the counted cell number of MC4 and MG63 cells in our preliminary experiments, thus growth was estimated using the MTT method. There was no correlation, however, between MTT activity and BrdUrd incorporation in ATDC5 and gene-transduced cells. Therefore, the BrdUrd method was used to measure cell proliferation (data not shown). PTH-(1-34), Bt 2 cAMP, and forskolin stimulated BrdUrd incorporation in two B-Raf-expressing bone cell lines in a low concentration range (ATDC5 and MC4; Fig. 4, A and B).
With increased ERK activity, IBMX also potentiated cell proliferation in basal and cAMP-stimulated conditions (Fig. 4C), indicating that growth potentiation was due to the increase in intracellular cAMP. The ERK pathway inhibitor PD98059 inhibited cAMP-triggered ERK activation and cell proliferation (Figs. 3D and 4D). Basal cell proliferation of MC4 was not affected by 1 M PD98059. The PKA inhibitor H89 (1 M) did not affect low concentrations cAMP-induced cell growth (Fig.  4D), indicating that if the PKA signaling pathway was inhibited, the mitogenic action of cAMP was not blocked because the signaling pathway of the Epac remained active. In contrast to the low concentration effects of cAMP, high concentration cAMP-induced cell proliferation was normalized to control levels. H89 did not affect the low concentration cAMP stimulation, whereas it potentiated the high concentration cAMP-mediated Regulation by Rap1 in 95-kDa B-Raf-expressing Cells-Because Rap1 is a transducer of cAMP-mediated regulation of ERK, we next tested the hypothesis that the cAMP effect on ERK activity and MC4 proliferation is the result of Rap1 activation. We established MC4 overexpressing dominant negative N17Rap1 or the constitutively active V12Rap1 mutants, and mutants of Ras were generated and several independently isolated clones analyzed for Rap1 expression by Western blotting using HA antibody. All clones demonstrated elevated Rap1 or Ras expression compared with vector-transfected control lines (data not shown). ERK activation by PTH-(1-34), Bt 2 cAMP, and forskolin was observed in Ras mutant clones, whereas it was completely blocked in N17Rap1-transduced MC4 cells, and accelerated in V12Rap1-transduced cells (Fig. 5A). Thus, cAMP actions on ERK were mediated via Rap1 activation and activated Rap1 induced ERK activation. We next tested the hypothesis that the mitogenic effects of cAMP were the result of Rap1 activation. The increased cAMP-induced cell growth was blocked in N17Rap1 but not in N17Ras clones, and accelerated growth was observed in V12Rap1 (Fig. 5B), demonstrating that Rap1 functions to stimulate ERK activation and cell proliferation in B-Raf-expressing cells.
cAMP Suppresses ERK Phosphorylation and Cell Proliferation in 95-kDa B-Raf-lacking Cells-Ras-dependent signaling activates ERK, but can be blocked by cAMP-dependent activation of Rap1 in many cell types (19). Thus, cAMP should decrease ERK activity through a negative Rap1 effect on Ras signaling in 95-kDa B-Raf-lacking cells. Consistent with this idea, cAMP reduced ERK activity in B-Raf-lacking cells. In MG63 and ROS17/2.8 cells, PTH-(1-34), forskolin, and Bt 2 cAMP suppressed ERK activation ( Fig. 3B and data not shown). To determine the effect of cAMP-mediated down-regulation of ERK activity on cell proliferation in B-Raf-lacking cells, MTT activity was measured in MG63 cultures exposed to treatments that modify ERK activity. Consistent with our prior observation that cAMP decreased ERK activity, cell proliferation was inhibited by cAMP and its inhibition was not blocked by H89 in either a low or high concentration of forskolin (Fig. 6, A and B, and data not shown), suggesting that PKA does not participate in the modulation of cell growth in B-Raf-lacking skeletal cells. PD98059 only blocked proliferation of MG63 cells to 70% of control levels (Fig. 6B), suggesting that ERK is an important mitogenic signal in MG63 cells.
Regulation by Rap1 in MG63-In B-Raf-lacking MG63 cells, cAMP suppressed ERK activation. Because Rap1 is activated by cAMP, we tested whether the inhibitory effects of cAMP on ERK activity and cell proliferation were the result of Rap1 activation. We established MG63 cells that overexpress mutant Rap1 proteins. Clones were generated and several independently isolated clones were analyzed for HA-Rap1 proteins. All clones had elevated Rap1 expression compared with vector- transduced control lines (data not shown). To determine how overexpression of mutant Rap1 would affect ERK activity and proliferation, the clones were analyzed by phospho-ERK West-ern blot analysis and BrdUrd incorporation assay. Basal levels of ERK activity were decreased in V12Rap1-transduced cells (Fig. 6C). Increased ERK activation, however, was observed in Dual Functionality of PTH on Cell Proliferation 22196 clones containing the N17Rap1 mutation. These results suggest that Rap1 suppresses ERK activity. To elucidate whether N17Rap1-induced ERK activation results in increased cell proliferation, cell growth of the clones was analyzed (Fig. 6D). In N17Rap1 clones, increased cell proliferation correlated with increased ERK activity and suppressed growth in V12Rap1, directly demonstrating that Rap1 functions to inhibit ERK activation and cell proliferation in B-Raf-lacking MG63 cells.
Conversion to B-Raf-expressing MG63 Cells-As Rap1 appears to mediate the cAMP growth inhibitory signal for B-Raflacking MG63 cells, we determined whether the critical difference in Rap1/cAMP signaling was dependent on the presence or absence of B-Raf. For these experiments, we generated stable MG63 clones expressing the 95-kDa isoform of B-Raf. Basal levels of ERK activity were similar in vector-and B-Raf-transduced cells (Fig. 6C). ERK activation by forskolin, however, was observed only in B-Raf-transduced clones. Thus, introduction of the B-Raf protein alone is sufficient to convert cAMP from a negative to a positive regulator of ERK. Finally, we determined whether the increased levels of ERK activity resulting from B-Raf overexpression were associated with increased cell proliferation. Basal growth levels were increased in B-Raf clones compared with vector-transduced control cells, and dramatically increased cell proliferation by forskolin were observed in B-Raf-transduced clones (Fig. 6D). These data indicate that one functional outcome of the molecular switch provided by B-Raf is increased ERK activation and cell proliferation in MG63 cells. DISCUSSION cAMP Signaling Pathway to Induce PTH-induced Proliferation in Bone Cells-In a wide variety of tissues (40), cAMP signaling is known to activate the ERK signaling pathway, thereby up-regulating cell proliferation and/or viability. As confirmed with cells of bone origin expressing PPR in this study, PTH action on PPR generates a cAMP signal and stimulates cell proliferation. Because activation of the classic target of cAMP, PKA, is inhibitory to cell proliferation (18 -21), and consistent with the inhibition of cell proliferation observed with a high concentration of forskolin (Fig. 4D), this suggests that the cAMP signal activates a hitherto unidentified cascade upstream of the ERK junction to stimulate cell proliferation. Our most striking observation was that, in mouse bone cells, the cAMP signal was propagated through a new route to the ERK junction to control cell proliferation, not via other classic pathways (e.g. Ras-Raf-1, PKA-CREB). The functional outcome, stimulated or inhibited proliferation of bone cells, solely depended on the cell type-specific expression of the ratio of two B-Raf splicing variants. In this new route, cAMP directly and specifically activates Epac (27), a cAMP-GEF, which then acts on Rap1, an antagonist of Ras-dependent signaling (41)(42)(43), and blocks Ras-dependent activation of Raf-1 in the presence or absence of the short form of B-Raf (41, 44 -46). In cells that express the Raf isoform, B-Raf, cAMP is known to activate ERK via the activation of Rap1 (24,(47)(48)(49)(50)(51)(52). Thus, two splice variants of B-Rafs function as a molecular switch in the activation or inhibition of the MEK-ERK pathway, and in our study, cAMPmediated inhibition of proliferation of MG63 cells was due to the predominant expression of 62-kDa B-Raf. This is the first report to show that molecules in this pathway are constitutively expressed in bone clonal cell lines and possibly in bone tissue in vivo and that these molecules play a major role in directing the cell reaction to the cAMP signal that is generated by PPR stimulation.
Relationship of New Signaling Pathway to Other Known Pathways-Quite recently, Swarthout et al. (53) reported that in the UMR106 rat osteosarcoma cell line and calvarial osteoblasts, subnanomolar PTH-(1-34) increased ERK activity in a manner sensitive to PKC inhibitor (GF109203) and MEK in- hibitor (PD98059) but resistant to PKA inhibitor (H-89), suggesting the participation of the PKC pathway in the activation of ERK. During the course of our study (data not shown), we also tested whether PKC lies upstream along the pathway leading to ERK activation. However, we were unable to inhibit cAMP-induced ERK activation with PKC inhibitors (calphostin and staurosporine) although they prevented cell proliferation. These differences may reflect the significant degree of hetero- geneity in downstream signaling via G-protein-coupled receptors between different cell types, as shown in Fig. 1, or they may simply reflect the different experimental conditions used. Further studies are needed to conclusively determine the role of PKC in the mechanism of action of PTH. To be stressed, however, is our observation that primary calvarial cells with presumably less homogeneity than clonal cell lines express 95-kDa B-Raf constitutively and predominantly, and this expression pattern remained unchanged during 1 month or more of culture in vitro, in which time the osteoblasts proliferated, matured, and mineralized as previously reported (32). It is highly possible that the expression pattern of given proteins are destined to change depending on variables in the in vivo environment. However, if the constant expression pattern of 95-kDa B-Raf in primary cells is physiological, our observations of varying B-Raf isoforms in clonal cell lines may indicate the abnormality and/or diversity of these clonal cell lines. To evaluate the physiological meaning of our results thus, our next steps should be a detailed analysis of this new signaling pathway in bone tissue in vivo as well as evaluations of PTH signaling in transgenic mice in which the key molecule(s) of this new pathway are depleted or overexpressed.
Another area of interest pertains to growth modulation by cAMP-mediated classical PKA signaling. The potential downstream signaling targets of PKA in bone are the cAMP-response element-binding protein (CREB) (56) and AP1 (57,58). PKA phosphorylates and activates CREB, and PTH activates AP1 family members such as Fos/Jun. Although these PKAassociated molecules are involved in bone development, recent evidence suggests a role for AP1 and CREB in differentiation rather than in proliferation (59). Our data indicate that the inhibition of PKA does not block cAMP-induced mitogenic action in cells of bone origin, regardless of whether the cells express or lack 95-kDa B-Raf. H-89, however, affected the growth inhibition produced by a high concentration of cAMP, suggesting that in the presence of high concentrations of cAMP, PKA may function in 95-kDa B-Raf-expressing cells to limit excessive growth.
The Role of the New Pathway in PPR-regulated Control of Bone Growth-The dual functionality of PTH action on bone metabolism has been previously noted in three cases. 1) The pattern of internal production or external dosing of PTH influences whether it is catabolic or anabolic in clinical and experimental situations (54,55). 2) PTH has either catabolic or anabolic effects on mineralization in cultured osteoblastic cell lines, depending upon the timing and duration of action of PTH during the cell maturation stage (12)(13)(14)(15)(16)(17)32). 3) Finally, PPR stimulation has been shown to have either catabolic or anabolic effects during development, depending upon which cortical or trabecular bone is involved (12). This study describes a fourth example of the dual functionality of PTH on bone metabolism. The anabolic effects previously described may arise from the synchronized maturation of cells and from mineralization promoted by extended cell viability. It is possible that the Epac/ Rap1/B-Raf pathway identified in our study as a key determinant of the effects of PTH signaling can explain the dual functionality of PTH that has been previously described, because cell type-specific expression of members of the pathway could drive a diverse functional outcome. We observed a constant expression pattern of a predominant variant in a primary cell line throughout its differentiation and maturation, while expression levels of PPR and ERK changed to some extent, but not in a manner sufficient to explain the dual functionality of the PTH signal. Further studies as described above are expected to more fully explain the mechanisms for the dual functionality of PTH signaling.
In conclusion, we have presented evidence indicating that normal bone cells of mouse calvaria as well as a variety of bone-related clonal cell lines constitutively express a signal transduction pathway including Epac, Rap1, and B-Rafs upstream of the MEK-ERK cascade. This pathway integrates the cAMP signal to promote or inhibit cell proliferation in a PKAand Ras-independent manner. The expression pattern of the members of the pathway, especially the B-Raf splice variants (95 and 62 kDa), varies in a cell type-specific manner in clonal cell lines and possibly in normal osteoblasts, and it determines whether the final effect of cAMP is to increase or decrease cell proliferation.