Sensing of extracellular cations in CasR-deficient osteoblasts. Evidence for a novel cation-sensing mechanism.

We isolated osteoblastic cell lines from wild-type (CasR(+/+)) and receptor null (CasR(-/-)) mice to investigate whether CasR is present in osteoblasts and accounts for their responses to extracellular cations. Osteoblasts from both CasR(+/+) and CasR(-/-) mice displayed an initial period of cell replication followed by a culture duration-dependent increase in alkaline phosphatase activity, expression of osteocalcin, and mineralization of extracellular matrix. In addition, a panel of extracellular cations, including aluminum and the CasR agonists gadolinium and calcium, stimulated DNA synthesis, activated a transfected serum response element-luciferase reporter construct, and inhibited agonist-induced cAMP in CasR(-/-) osteoblasts. The functional responses to these cations were identical in CasR(+/+) and CasR(-/-) osteoblasts. Thus, the absence of CasR alters neither the maturational profile of isolated osteoblast cultures nor their in vitro responses to extracellular cations. In addition, CasR transcripts could not be detected by reverse transcription-polymerase chain reaction with mouse specific primers in either CasR(+/+) or CasR(-/-) osteoblasts, and immunoblot analysis with a CasR-specific antibody was negative for CasR protein expression in osteoblasts. The presence of a cation-sensing response in osteoblasts from CasR(-/-) mice indicates the existence of a novel osteoblastic extracellular cation-sensing mechanism.

CasR is a G-protein-coupled extracellular calcium-sensing receptor that was originally isolated and cloned from parathyroid (1), kidney (2), and brain (3) cDNA libraries. The major physiological function of CasR is to transduce extracellular calcium regulation of parathyroid hormone secretion in parathyroid glands. This role is supported by the findings that inactivating mutations of CasR cause familial hypocalciuric hypercalcemia and neonatal severe hyperparathyroidism (NSHPT) (4) and that activating mutations result in hypoparathyroidism (5). In addition, targeted ablation of CasR results in hypercalcemia and elevated parathyroid hormone levels in mice (6). The function of CasR in other tissues is less clear. In the kidney, CasR may modulate renal tubular transport, possibly acting as an antagonist to parathyroid hormone actions on renal calcium handling (7). CasR is also located in nerve terminals, where it may transduce calcium-mediated neuro-transmitter release (3). CasR is detected in low abundance in many other tissues and cell types, where its physiologic role is uncertain (7)(8)(9).
Bone is one such tissue in which the expression of CasR and its physiologic role remain uncertain (10 -14). Although CasR is reported to be expressed in bone marrow cells (10,11), there are conflicting data regarding its expression in cells within the osteoblast lineage (13)(14)(15)(16)(17)(18)(19). Some studies have reported the detection of CasR in osteoblasts by Western blot and RT-PCR 1 analyses (13,14); however, recent attempts by others failed to confirm these findings (15)(16)(17)(18)(19). Since cells may reduce CasR expression in culture, it remains unclear whether nontransformed osteoblasts or native osteoblasts express the known CasR. The detection of a functional CasR in bone cells derived from the mouse is potentially confounded by incomplete knowledge of the mouse CasR coding sequence (11,20,21) and the possible presence of another cation-sensing receptor in osteoblasts (17,20). Prior studies by us (20) and others (11,21) have identified only partial segments totaling 1514 bp of the mouse CasR cDNA sequence. Low abundance of CasR as well as nonspecificity of the detection methods may also contribute to the contradictory findings of CasR in osteoblasts (16).
Whereas the variable reports of CasR expression in osteoblasts remain unexplained, all studies to date suggest that osteoblasts display a functional response to extracellular calcium and other cations via a G-protein-coupled receptor-like mechanism that resembles CasR (14,15,17,19,(22)(23)(24)(25)(26)(27). Both DNA synthesis and chemotaxis are stimulated in osteoblasts by a panel of cations, viz. calcium, gadolinium, and neomycin (14,17,25), that act on osteoblasts with relative potencies, apparent affinities, and specificities similar to their activation of CasR (14,17,25). Cations also induce c-fos mRNA expression and SRE-luciferase promoter activity in osteoblasts (17,26). The osteoblastic cation response is coupled to G-proteins as evidenced by [ 32 P]GTP azidoanilide labeling of 42-and 38-kDa G␣-proteins in osteoblast membranes following cation stimulation (27). In addition, the osteoblastic cation-sensing receptor inhibits agonist-induced stimulation of cAMP accumulation as reported for CasR (23).
The functional properties of the putative osteoblastic CasRlike receptor differ, however, from those of CasR in several ways. The cation specificities for the putative osteoblastic receptor, while overlapping, are not identical to those for CasR (17,19,24,28). For example, the trivalent cation aluminum, which acts by a mechanism in common with gadolinium (26,27), is a potent agonist for the putative osteoblastic cationsensing receptor, but not for CasR (28), whereas neither the CasR agonist magnesium (17,24) nor calcimimetics activate the putative cation receptor in osteoblasts (19). Also, unlike CasR, the osteoblastic receptor does not appear to be coupled to phosphoinositide-specific phospholipase C. Rather, the effects of cations in osteoblasts are mediated by the activation of protein kinase C through unknown signaling pathways (17). These functional differences suggest that a related but molecularly distinct calcium-sensing receptor may be present in osteoblasts (17) and account for the inconsistent findings of CasR expression in osteoblasts.
In an effort to prove the existence of another calcium-sensing receptor, we isolated and cloned the full-length mouse CasR coding sequence and tested for CasR transcripts in osteoblasts using RT-PCR analysis with mouse CasR-specific primers. In addition, we evaluated the ability of a mouse anti-CasR antibody to detect CasR protein in osteoblasts by Western blot analysis. Finally, we determined whether the extracellular cation-sensing response is abolished in osteoblasts derived from mice deficient in CasR. Consistent with the presence of a novel osteoblastic cation-sensing mechanism, we found that cations stimulated to the same degree all measures of cation responses in osteoblasts derived from both wild-type and CasR-deficient mice.

EXPERIMENTAL PROCEDURES
Cloning and Sequencing of Mouse Calcium-sensing Receptor cDNA-We used RT-PCR to isolate the full-length mouse CasR cDNA coding sequence. Total RNAs were prepared from mouse kidney tissue using the TRIzol reagent (Life Technologies, Inc.) as described previously (17). We designed mouse CasR-specific oligonucleotide primers based on the incompletely characterized mouse CasR sequence (11,20,21) as well as additional unpublished sequence information found in GenBank TM (accession number AF068900). For segments in which the mouse sequence was unknown, we substituted primers derived from regions of the rat CasR cDNA sequence that were highly conserved across species (2). Using this approach, we were able to design six primer sets (see Table I) to overlapping segments covering the entire mouse cDNA open reading frame as well as portions of the 5Ј-and 3Ј-untranslated regions. RT-PCR was done using the Titan TM One Tube RT-PCR kit purchased from Roche Molecular Biochemicals. The reverse transcription reaction using 2.0 g of total RNA treated with DNase I (Stratagene, La Jolla, CA) was incubated at 45°C for 60 min, and the template was denatured at 94°C for 2 min. PCR was performed with thermal cycling parameters of 94°C for 30 s, 55°C for 30 s, and 68°C for 45 s for 10 cycles. This was followed by an additional 25 cycles with thermal cycling parameters of 94°C for 30 s, 55°C for 30 s, and 68°C for 45 s plus an additional 5 s with each cycle. The reaction was completed with a final extension at 68°C for 7 min. The resultant PCR products were subcloned into pCRII (Invitrogen, San Diego, CA) and sequenced using M13 forward and reverse primers. Sequence analysis was performed using the GCG software package (Version 8, Genetics Computer Group, Madison, WI).
Isolation of Immortalized Osteoblastic Cell Lines from Normal and CasR Gene Knockout Mouse Calvariae-Heterozygous CasR knockout mice (CasR Ϫ/ϩ ) were obtained form the laboratory of Dr. David Conner (Harvard University, Boston, MA) (6), and C57BL/6J mice expressing the large T antigen of SV40 were purchased from Jackson Laboratories (Bar Harbor, Maine). Mice were maintained and used in accordance with recommendations as described (37) and following guidelines established by the Institutional Animal Care and Use Committee of Duke University. Heterozygous CasR knockout females (CasR Ϫ/ϩ ) were mated with males expressing the large T antigen of SV40. The resulting male and female offspring that were heterozygous for expression of both the large T antigen of SV40 and CasR knockout were mated. Offspring of this mating were genotyped, and those that expressed the large T antigen of SV40 and either CasR ϩ/ϩ (wild type) or CasR Ϫ/Ϫ (homozygous) were selected for isolation of calvarial osteoblasts.
We used modifications of a nonenzymatic method for obtaining the osteoblastic cell lines (31). A fragment of the frontal and/or parietal bone from a single calvaria was aseptically removed from a 3-7-day-old mouse. Suture lines and endosteum were dissected away, and the bone fragment was placed in a culture dish. One or two metal strips were positioned on the endocranial surface and incubated for 3-4 days in Dulbecco's modified Eagle's medium/nutrient mixture F-12 containing 10% (v/v) fetal bovine serum and 1% penicillin/streptomycin until the outgrowth of osteoblasts. The metal strips were removed, and the cells were allowed to grow until ϳ60% confluent. The cells were subcultured and propagated by incubation in ␣-modified essential medium containing 10% fetal bovine serum, 1% penicillin/streptomycin, and 50 g/ml ascorbic acid in a humidified atmosphere of 5% CO 2 and 95% air at 37°C.
In addition, we also used established MC3T3-E1 (29) and clonal TMOb-c (30) osteoblastic cell lines. These cells were grown in ␣-modified essential medium supplemented as described above in a humidified atmosphere of 5% CO 2 and 95% air at 37°C. Cells were subcultured every 3-5 days using 0.001% Pronase.
Genotyping-We used a PCR approach to detect wild-type CasR ϩ/ϩ and CasR Ϫ/Ϫ knockout osteoblasts expressing the large T antigen of SV40. Genomic DNA tissue was extracted and purified from the tail of each mouse or osteoblastic cell culture using a QIAamp blood kit (QIA-GEN Inc., Valencia, CA). To detect the presence of CasR Ϫ/Ϫ , we used the reverse primer CasR2144.R (5Ј-TGAAGCACCTACGGCACCTG-3Ј), specific for the native mouse CasR gene sequence, in combination with a primer designed for upstream elements in exon 4, CasR1956.F (5Ј-TGATGAAGAGTCTTTCTCGG-3Ј), or primer KCM-F (5Ј-TCTTGAT-TCCCACTTTGTGGTTCTA-3Ј) for the inserted neomycin gene sequence used for targeted disruption of exon 4 (32). Progeny containing the SV40 transgene were identified by PCR amplification of an ϳ500-bp product of SV40 from individual genomic DNA using forward primer 5Ј-CAGAGCAGAATTGTGGAGTGG-3Ј and reverse primer 5Ј-GGA-CAAACCACAACTAGAATGCAGTG-3Ј. PCR was performed with thermal cycling parameters of 94°C for 3 min, 94°C for 20 s, 60°C for 20 s, and 72°C for 45 s for 35 cycles followed by a final extension at 72°C for 10 min. Amplification products were resolved by electrophoresis on a 1.5% agarose gel and visualized by ethidium bromide staining.
RT-PCR Analysis of CasR Transcripts in Mouse Osteoblastic Cell Lines-To detect CasR expression, RT-PCR was done using two-step RNA PCR (Perkin-Elmer). In separate reactions, 2.0 g of DNasetreated total RNA was reverse-transcribed into cDNA with the respective reverse primers specified below and Moloney murine leukemia virus reverse transcriptase (Life Technologies, Inc.). Reactions were carried out at 42°C for 60 min followed by 94°C for 5 min and 5°C for 5 min. The products of first strand cDNA synthesis were directly amplified by PCR using AmpliTaq DNA polymerase (Perkin-Elmer) using three separate sets of primers based on the mouse CasR cDNA sequence. PCR was performed with thermal cycling parameters of 94°C for 3 min, 94°C for 1 min, 60°C for 1 min, and 72°C for 2 min for 30 cycles followed by a final extension at 72°C for 10 min. The primer sets used to amplify overlapping regions of the 5Ј-end of mouse CasR included the mouse specific forward primer mCasR216.F (5Ј-AGAGC-CATGGCATGGTTTGG-3Ј) and reverse primer mCasR653.R (5Ј-TGC-TCCCACCACTGCGATGGTTG-3Ј) and the intron-spanning primer set consisting of the forward primer mCasR291.F (5Ј-CAGCGAGCCCAA-AAG AAAGG-3Ј) and reverse primer mCasR1190.R (5Ј-CTTCA GACCG AACCC AATGG-3Ј). We documented that these primer sets were intron-spanning by comparing the sizes of the products derived from PCR of reverse-transcribed kidney RNA and genomic DNA. To amplify a region containing the 3Ј-end of CasR, we used the forward primer mCasR2431.F (5Ј-TCATC TGCAT CATCT GGCTC-3Ј and the mouse specific reverse primer mCasR3537.R (5Ј-TTGGCTTCCTTGGGAA-GACC-3Ј), located within the same exon.
In addition, using a similar protocol (30), the mouse osteocalcin transcript was RT-PCR-amplified using the forward primer mOGϩ8.F (5Ј-CAAGTCCCACACAGCAGCTT-3Ј) and the reverse primer mOGϩ378.R (5Ј-AAAGCCGAGCTGCCAGAGTT-3Ј). Mouse ␤-actin was amplified as a control for the RT-PCRs as described previously (30). Amplification products were resolved by electrophoresis on a 1.5% agarose gel and visualized by ethidium bromide staining.
Assessment of Alkaline Phosphatase Activity-We analyzed alkaline phosphatase in cell layers by colorimetric assay of enzyme activity with the substrate p-nitrophenol phosphate as described previously (26).
Mineralization Assay-The formation of in vitro mineralization nodules was determined by alizarin red S histochemical staining (33). The 14-day cultured cells were fixed for 24 h in a 1:1:1.5 solution of 10% Formalin, methanol, and water; the fixative was removed; and the fixed cells and matrices were stained for 15 min with a 2% (w/v) solution of alizarin red S at pH 4.0. The stained samples were washed three times with water and then air-dried.
Western Analysis of Mouse CasR Protein Expression-Crude plasma membranes were isolated from culture cells as described (16). The protein content of each sample was determined by the NanoOrange TM protein quantitation kit (Molecular Probes, Inc., Eugene, OR). All membrane protein extracts were stored at Ϫ70°C. Immunoblotting was performed using the mouse anti-CasR monoclonal antibody ADD pro-vided by NPS Pharmaceuticals (Salt Lake City, UT). Mouse monoclonal antibody ADD was raised against the peptide corresponding to amino acids 214 -235 of human CasR (ADDDYGRPGIEKFREEAEERDI) (34). Membrane proteins were dissolved in SDS gel loading buffer (62 mM Tris-HCl (pH 6.8), 2% SDS, 10% glycerol, 5% mercaptoethanol, and 0.1% bromphenol blue), boiled for 10 min before loading, and separated on 6% SDS-polyacrylamide gel. Separated proteins were transferred to nitrocellulose membrane (0.45 m; Bio-Rad) over a 30-min period at room temperature by a semidry blotting system (2.5 mA/cm 2 ; MilliBlot, Marlborough, MA). Immunoblotting was performed by incubating blots for 60 min at room temperature with SuperBlock (TBS) blocking buffer (Pierce). After washing with 1ϫ TBS containing 0.1% Tween 20 (TBST) for 1 h, blots were incubated with the ADD antibody (1:32,000, 100 ng/ml final concentration) overnight at room temperature. Blots were washed with TBST for 60 min and incubated with horseradish peroxidase-conjugated anti-mouse Ig whole antibody (from sheep; Amersham Pharmacia Biotech) for 60 min at room temperature at a dilution of 1:2500. After washing with TBST, immunoreactivity was detected by Western blot chemiluminescence Reagent Plus (NEN Life Science Products).
Cation Effects on Agonist-stimulated cAMP Production-For these studies, cells were plated at an initial density of 100,000 cells/well (six-well plate) and cultured as described above. Quiescent cells were incubated in serum-free medium containing 2 Ci/ml [ 3 H]adenine (Amersham Pharmacia Biotech) for 3 h. Prior to agonist stimulation, cells were incubated for 10 min with 100 M isobutylmethylxanthine in buffer consisting of Hanks' balanced saline solution (without Ca 2ϩ and Mg 2ϩ ) containing 10 mM HEPES (pH 7.4). We stimulated cAMP production by incubation with PGE 1 for 5 min in the presence or absence of various cations that were added 10 min prior. Generation of cAMP was measured by the method of Salomon et al. (35).
DNA Synthesis-For studies that assessed rates of DNA synthesis, subconfluent cells from stock cultures were seeded at an initial density of 20,000 cells/well in 24-well (16-mm) dishes and incubated in ␣-modified essential medium containing 10% fetal bovine serum and 1% penicillin/streptomycin for 2 days prior to the induction of quiescence as described previously (17). Quiescence was achieved in these subconfluent cultures by washing twice with Hanks' balanced saline solution to remove residual serum and incubating for an additional 48 h in serumfree Dulbecco's modified Eagle's medium/nutrient mixture F-12 containing 0.1% bovine serum albumin. The study period was then initiated by changing to fresh serum-free medium containing the various CasR agonists and [ 3 H]thymidine (0.5 Ci/ml; Amersham Pharmacia Biotech). DNA synthesis was assessed 24 h later by the determination of [ 3 H]thymidine incorporation into acid-precipitable material as described previously (17). DNA synthesis rates were determined in control cultures incubated in fresh serum-free stock medium containing 0.1% bovine serum albumin and were compared with those obtained after addition of the respective cation.
Transient Transfection-For these studies, we used the previously described SRE-luciferase plasmid DNA (17,36). The plasmid was prepared using the QIAGEN plasmid midi kit. Transient transfections were preformed as follows. 10 5 cells were plated in the six-well plate and incubated overnight at 37°C. A DNA-liposome complex was prepared by mixing DNA and TransFast TM transfection reagent (1:2 DNA/ TransFast TM transfection reagent; Promega) in Opti-MEM I reduced serum medium (Life Technologies, Inc.) and incubating the mixture at room temperature for 15 min. The mixtures of SRE-luciferase, plasmid DNA, and TransFast TM transfection reagent were diluted with Opti-MEM I reduced serum medium and added to Hanks' balanced saline solution-rinsed cells (final concentration of 5 g of SRE-luciferase plasmid DNA/well). After 1 h of incubation at 37°C, 2 ml of ␣-modified essential medium containing 10% fetal bovine serum and 1% penicillin/ streptomycin was added to the medium overlying the transfected cells and incubated for an additional 2 days prior to treatment. Luciferase activity was assessed in quiescent cells treated with 60 M GdCl 3 , 25 M AlCl 3 , or 5 mM calcium for 8 h prior to harvesting. Controls consisted of untreated quiescent cells for an equivalent time. The luciferase activity in cell extracts was measured using the luciferase assay system (Promega) following the manufacturer's protocol using a BG-luminometer (Gem Biomedical Inc., Hamden, CT).
Statistical Analysis-Analysis of variance was performed with the Statgraphics software package (Statistical Graphics Corp., Inc., Princeton, NJ).

RESULTS
Establishing Immortalized Mouse Osteoblasts from Wildtype CasR ϩ/ϩ and CasR Ϫ/Ϫ Null Mice-To evaluate a potential functional role of CasR in osteoblasts, we established immortalized osteoblastic cell lines from calvariae derived from wildtype CasR ϩ/ϩ and CasR Ϫ/Ϫ null mice (Figs. 1 and 2). The presence of CasR deficiency was documented by detecting the neomycin resistance gene (neo r ) that disrupts exon 4 of the CasR gene in CasR Ϫ/Ϫ mice (6). We identified the presence of the intact exon 4 of CasR in genomic DNA of wild-type CasR ϩ/ϩ cell lines, but not in CasR Ϫ/Ϫ DNA (Fig. 1A). Conversely, we detected the neo r insert in exon 4 in genomic DNA derived from CasR Ϫ/Ϫ mice, but not in wild-type DNA (Fig. 1B), consistent with the disruption of this exon by the neomycin resistance gene.
Next, we examined whether osteoblasts derived from CasR Ϫ/Ϫ mice display alterations in growth and development (Fig. 2). Similar to our experience with osteoblastic cell lines (29,30), osteoblasts derived from wild-type CasR ϩ/ϩ and CasR Ϫ/Ϫ knockout mice underwent an initial period of rapid cell proliferation characterized by high levels of tritiated thymidine incorporation followed by a significant decrease in cell replication with prolonged culture ( Fig. 2A). At the 4-day time point, the replication rates were slightly lower in CasR Ϫ/Ϫ compared with CasR ϩ/ϩ osteoblasts, but this difference was not present in later cultures. During the period of rapid cell growth during the initial 4 days of culture, both CasR ϩ/ϩ and CasR Ϫ/Ϫ osteoblasts expressed low levels of alkaline phosphatase (Fig.  2B), consistent with their immature, pre-osteoblastic state. Prolonged culture was associated with a significant increase in the expression of alkaline phosphatase activity in both CasR ϩ/ϩ and CasR Ϫ/Ϫ cells, achieving comparable levels by day 14 of culture (Fig. 2B). Similarly, the process of osteoblast maturation in the immortalized cells was marked by the absence of osteocalcin transcripts in 4-day-old cultures, but with high levels of osteocalcin in later cultures of both CasR ϩ/ϩ and CasR Ϫ/Ϫ osteoblasts (Fig. 2C). In addition, alizarin red S-stained mineralization nodules were observed in both CasR ϩ/ϩ and CasR Ϫ/Ϫ osteoblasts by day 14 of culture ( Fig.   FIG. 1. Genotyping of the CasR ؉/؉ and CasR ؊/؊ osteoblastic cell lines. We confirmed the genotype in DNA isolated from these cell lines using primers CasR1956.F and CasR2144.R, designed to amplify a 189-bp fragment from exon 4, and a primer to the neo r sequence (KCM-F) in combination with the CasR primer CasR2144.R to detect the presence of neo r inserted into exon 4. A, detection of wild-type CasR. Using the indicated primer pair, we failed to detect a CasR transcript corresponding to exon 4 by PCR in DNA isolated from CasR Ϫ/Ϫ due to the neomycin resistance gene (neo r ) insert, but detected the predicted 189-base pair product in DNA obtained from CasR ϩ/ϩ mice. B, detection of the neo r insert in DNA derived from CasR Ϫ/Ϫ mice. Using a primer specific for the neomycin resistance gene in combination with a CasRspecific primer, we detected the presence of the neo r insert in DNA isolated from CasR Ϫ/Ϫ mice, but not from CasR ϩ/ϩ mice. Ob, osteoblasts.
2D). These findings indicate that the absence of CasR in mice has no effect on the ex vivo growth and maturation of calvariaderived osteoblasts since both CasR ϩ/ϩ and CasR Ϫ/Ϫ immortalized osteoblastic cell lines retain their capacity to undergo a normal temporal up-regulation of osteoblast-related gene expression. We used these osteoblasts, along with the previously described clonal osteoblastic cell lines MC3T3-E1 (29) and TMOb-c (30), to examine CasR expression and potential function.
Mouse CasR cDNA Sequence-Knowledge of the entire mouse CasR coding sequence is an important prerequisite for determining if mouse osteoblastic cell lines express a functional full-length CasR transcript. To amplify the full-length mouse CasR coding sequence, we used various regions from the partial sequence of mouse CasR cDNA that have been published (2,11,20,21) and devised an RT-PCR approach using six separate primer sets (Table I). We successfully amplified from mouse kidney a 3569-bp cDNA that contained a 3237-bp open reading frame, 222 bp of the 5Ј-untranslated region, and 110 bp of the 3Ј-untranslated region (Fig. 3). Mouse CasR is predicted to encode a protein of 1079 amino acids. The previously identified structural domains of CasR (1-3) are conserved in the mouse CasR sequence, including a large amino-terminal domain, a seven-membrane spanning domain, and a 216-amino acid C terminus (Fig. 3). The respective nucleotide and deduced amino acid sequences are 94 and 98% identical to rat, 87 and 93% identical to human, and 86 and 92% identical to bovine CasR. Although the overall sequence and the predicted structure of mouse CasR were highly homologous to CasR from other species (Fig. 3), we identified sequences in the 5Ј-and 3Ј-untranslated regions that were unique for the mouse CasR sequence. The greatest disparities in the nucleotide sequences between mice and other species were between 216 and 235 bp in the 5Ј-end and 3518 and 3537 bp in the 3Ј-end of mouse CasR. From these regions, we designed mouse-specific oligonucleotides as well as intron-spanning primers to perform RT-PCR

FIG. 2. Characterization of temporal maturational sequence in osteoblasts derived from wild-type CasR ؉/؉
and CasR ؊/؊ knockout mouse calvariae. A, DNA synthesis; B, alkaline phosphatase (ALP) activity; C, osteocalcin transcript expression; D, histochemical staining of mineralization nodules. CasR Ϫ/Ϫ null and wild-type CasR ϩ/ϩ osteoblastic cell lines were cultured for up to 14 days, and the phenotype was characterized as described under "Experimental Procedures." Mouse specific primers were used to RT-PCR-amplify osteocalcin (oc), and ␤-actin served as a control for relative mRNA abundance. Staining with alizarin red S was performed to assess mineralized nodule formation as described under "Experimental Procedures." Both osteoblastic cell lines derived from wild-type CasR ϩ/ϩ and CasR Ϫ/Ϫ knockout mice displayed a similar pattern of culture duration-dependent maturation. Numeric values represent means Ϯ S.E. of at least three separate determinations. Values sharing the same superscript are not significantly different at p Ͻ 0.001. Ob, osteoblasts.  RT-PCR Analysis of CasR Expression in Mouse Osteoblasts-We used information from the complete mouse CasR sequence to examine if these mouse-derived osteoblastic cell lines express CasR transcripts. Using three different primer sets spanning the majority of the CasR cDNA sequence, we successfully amplified the predicted 438-, 900-, and 1107-bp reverse transcription-dependent products in positive control mouse kidney RNA (Fig. 4A). In contrast, we failed to amplify CasR in the mouse MC3T3-E1, TMOb-c, CasR ϩ/ϩ , and CasR Ϫ/Ϫ osteoblastic cell lines (Fig. 4A). Thus, CasR transcripts were not detected by RT-PCR with mouse specific primers in any of the osteoblastic cell lines, but were detected in tissues known to express CasR.
Western Blot Analysis of CasR Protein Expression-Next, we performed Western blot analysis to determine if the CasR protein is expressed in osteoblastic cell lines (Fig. 4B). For these studies, we used the mouse neuronal cell line AtT-20, which is known to express a functional CasR, as a positive control (21). We used the highly specific ADD monoclonal antibody to CasR (16,34). We have previously shown that this antibody, which is generated from a synthetic peptide derived from a region that is 100% conserved between human, rat, and mouse, detects CasR in tissues derived from these species (16). This antibody detected bands of ϳ140 and 165 kDa in membrane protein preparations from the mouse AtT-20 cells, representing glycosylated CasR monomers. In contrast, immunoblot analysis with the ADD antibody of membranes from the mouse osteoblastic cell lines MC3T3-E1 and TMOb-c as well as wild-type CasR ϩ/ϩ and CasR Ϫ/Ϫ knockout osteoblasts cultured for 4 and 14 days failed to identify these bands, despite loading 5-fold greater amounts of protein/lane (Fig. 4B). We did observe a weak band of ϳ115 kDa in osteoblasts. This band appears to be nonspecific since it was present in cells derived from both CasR Ϫ/Ϫ and CasR ϩ/ϩ mice and was smaller than the CasR products in the positive control AtT-20 cells.
Functional Response of Wild-type CasR ϩ/ϩ and CasR Ϫ/Ϫ Knockout Osteoblasts to Extracellular Cations-Finally, we compared the functional response of osteoblasts derived from CasR ϩ/ϩ and CasR Ϫ/Ϫ mice with a panel of extracellular cations. We have previously shown that extracellular cations inhibit agonist-induced cAMP generation, stimulate DNA synthesis, and induce c-fos promoter transcription in osteoblasts (23). If this response is due to a receptor other than CasR, then cation sensing should be preserved in CasR Ϫ/Ϫ knockout osteoblasts. We found that treatment of quiescent osteoblasts from both wild-type CasR ϩ/ϩ and CasR Ϫ/Ϫ knockout mice produced identical increments in cAMP production over control values in response to PGE 1 (Fig. 5A). Moreover, we found that Gd 3ϩ (60 M), Al 3ϩ (25 M), and Ca 2ϩ (5 mM) also inhibited PGE 1 -stimulated cAMP production in osteoblasts derived from both wildtype CasR ϩ/ϩ and CasR Ϫ/Ϫ knockout mice. The cation-mediated inhibition was identical in both wild-type CasR ϩ/ϩ and CasR Ϫ/Ϫ knockout osteoblasts. Cations had no effect on basal unstimulated cAMP levels in either cell line (data not shown).
We also found that CasR Ϫ/Ϫ osteoblasts retained their ability to respond to extracellular cations by increasing DNA synthesis, although the magnitude of this response was smaller than we previously observed in MC3T3-E1 osteoblasts (17). This may be due to the transformation of our immortalized osteoblasts, which in turn leads to higher basal replication rates. Nevertheless, treatment of quiescent wild-type CasR ϩ/ϩ and CasR Ϫ/Ϫ knockout osteoblasts with Gd 3ϩ , Ca 2ϩ , or Al 3ϩ at doses shown to activate the putative osteoblastic cation receptor (17) stimulated DNA synthesis over control values (Fig. 5B).
Another characteristic of the extracellular cation-sensing response in osteoblasts is the rapid induction of c-fos promoter transcription (17,26). To investigate if this response is preserved in osteoblasts from CasR Ϫ/Ϫ mice, we examined the effects of our panel of cation agonists on transcriptional activation of a c-fos promoter-reporter construct containing SRE in both the CasR Ϫ/Ϫ and CasR ϩ/ϩ cell lines. Gd 3ϩ (60 M), Al 3ϩ (25 M), and Ca 2ϩ (5 mM) displayed similar effects to increase luciferase activity compared with unstimulated control cells in both CasR Ϫ/Ϫ and CasR ϩ/ϩ cells containing the SRE-luciferase construct (Fig. 5C). DISCUSSION We provide evidence for the presence of a novel cation-sensing mechanism in osteoblasts. The preservation of cation-sensing responses in osteoblasts from CasR-deficient mice provides the most compelling evidence for another cation-sensing mech- FIG. 4. Expression of the CasR gene in mouse osteoblasts. A, RT-PCR amplification of mouse CasR mRNA using 5Ј-and 3Ј-region mouse specific primer sets. Total RNAs (2 g) were obtained from mouse kidney (mKidney), mouse osteoblastic cell lines MC3T3-E1 and TMOb-c, and osteoblastic cell lines isolated from wild-type (CasR ϩ/ϩ Ob) and CasR null (CasR Ϫ/Ϫ Ob) mice. Mouse ␤-actin was amplified as a control. CasR was detected in mouse kidney, but not in any of the osteoblastic cell lines. B, Western blot analysis of CasR expression. 100 g of crude membrane proteins were derived from MC3T3-E1 and TMOb-c osteoblasts as well as from osteoblasts isolated from wild-type (CasR ϩ/ϩ Ob) and CasR null (CasR Ϫ/Ϫ Ob) mice. CasR ϩ/ϩ and CasR Ϫ/Ϫ cells was cultured for either 4 or 14 days. Membrane proteins were subjected to 6% SDS-polyacrylamide gel electrophoresis. As a positive control, 20 g of crude membrane protein from the mouse AtT-20 cell line was used. CasR protein was detected by immunoblot analysis with the ADD antibody as described under "Experimental Procedures." Arrows demonstrate that the anticipated bands representing differentially glycosylated monomeric CasR protein were found in the CasR-expressing AtT-20 cell line, but not in the osteoblastic cell lines.
anism. Indeed, cations affected to the same degree all measures of cation responses, including stimulation of cell replication and SRE-luciferase activity as well as inhibition of agonistinduced cAMP, in osteoblasts derived from both wild-type and CasR-deficient mice. Although the response of osteoblasts to extracellular cations functionally resembles that of CasR, our findings suggest that this response is mediated by a distinct extracellular cation-sensing mechanism.
We used a variety of approaches to determine if a functional CasR is expressed in mouse osteoblasts. We isolated immortalized osteoblastic cell lines from CasR Ϫ/Ϫ and CasR ϩ/ϩ mice ( Figs. 1 and 2) and ascertained the mouse CasR cDNA (Fig. 3). We tested if a CasR transcript is present in these osteoblasts by RT-PCR analysis (Fig. 4A) using new information obtained from the mouse CasR coding sequence (Fig. 3). Using mouse CasR-specific primers obviates potential concerns regarding cross-reactivity and stringency that may have confounded prior evaluations (14,17). We detected reverse transcription-dependent products of the correct size in mouse kidney mRNA, but not in MC3T3-E1 osteoblasts, which display a functional response to extracellular cations (17). In addition, we failed to detect CasR by RT-PCR in any of the other mouse osteoblastic cell lines that we tested (Fig. 4A). Thus, using a sensitive and specific detection technique, we were unable to establish the presence of CasR transcripts in osteoblasts.
We also failed to detect the known CasR in osteoblasts by Western blot analysis using a highly specific monoclonal anti-body to CasR (ADD) (34). Immunoblot analysis with ADD under reducing conditions typically identifies two predominant CasR species, including an ϳ165-kDa fully glycosylated CasR and a lower molecular mass band (ϳ140 kDa) representing a partially glycosylated product (16). We detected the correct size bands in CasR-expressing AtT-20 pituitary cells (Fig. 4B) (21), but failed to detect these bands in any of the osteoblastic cell lines (Fig. 4B). Although negative immunoblot analysis does not preclude the presence of CasR in low abundance, such a possibility seems unlikely, given our failure to amplify CasR transcripts in osteoblasts by standard RT-PCR (Fig. 4A). Rather, our inability to detect CasR expression in osteoblastic cell lines by either RT-PCR or Western blot analyses and prior evidence for a functional response to extracellular cations in these cells (17) are consistent with the presence of a molecularly distinct, but functionally similar cation-sensing receptor in cultured osteoblastic cell lines.
In an effort to define a role for CasR in transducing the response to extracellular cations observed in osteoblasts, we tested the ability of osteoblasts from CasR-deficient mice to respond to a specific panel of extracellular cations previously shown to activate the putative osteoblastic cation receptor (17). We found that the previously described functional responses attributed to CasR are retained in osteoblasts derived from CasR Ϫ/Ϫ mice (Fig. 5). In this regard, stimulation of DNA synthesis and SRE-promoter activity as well as inhibition of agonist-induced cAMP accumulation were all present in osteo- blasts derived from CasR knockout mice. Moreover, stimulation of DNA synthesis and SRE-promoter activity and inhibition of agonist-induced cAMP by calcium, gadolinium, and aluminum were similar in wild-type and CasR Ϫ/Ϫ osteoblasts (Fig. 5). In addition, the maturational profile, including the up-regulation of alkaline phosphatase, osteocalcin expression, and mineralization, also was not altered in the CasR Ϫ/Ϫ osteoblasts (Fig. 2). These findings add further support for the existence of a novel cation-sensing mechanism in osteoblasts that is distinct from CasR.
Our failure to find evidence to support the presence of a CasR transcript or protein in osteoblasts by RT-PCR and immunoblot analysis is consistent with at least four separate laboratories that have reported unsuccessful attempts to detect CasR expression and/or function in osteoblasts (15)(16)(17)(18)(19). Indeed, to date, no studies have identified in osteoblasts the ϳ165and ϳ140-kDa products that are characteristic of the functional CasR protein. In contrast to our findings, however, other reports have purported to detect CasR in osteoblasts and have attributed the functional response to extracellular calcium in osteoblasts to the presence of CasR (10,11,13,14). Possible explanations for these discrepancies include differences in the specificity of the anti-CasR antibodies, nonspecificity of the RT-PCR results, variable expression of the receptor in cells from different species, the presence of aberrant glycosylation in osteoblasts, an alternatively spliced form of CasR, and the loss of CasR expression in cell cultures. Still other explanations for these discrepancies are possible, but our failure to detect CasR in osteoblasts that respond to extracellular cations and the presence of a cation-sensing response in mutant osteoblasts that lack CasR provide strong support for the presence of a novel osteoblastic cation-sensing mechanism that is functionally related, but molecularly distinct from CasR.
Although our current data advance the notion that a novel cation-sensing receptor may account for the osteoblastic cationsensing response, the identity of the putative osteoblastic cation receptor remains unknown. Nevertheless, we confirmed several important features that distinguish the osteoblastic cation-sensing response from CasR. In this regard, our studies show that aluminum is a potent agonist for the putative osteoblastic CasR-like receptor in CasR Ϫ/Ϫ mice, whereas micromolar concentrations of the trivalent cation aluminum do not stimulate CasR (28). We also failed to observe increments in intracellular calcium in response to cations in CasR Ϫ/Ϫ osteoblasts, consistent with the observation that the osteoblastic CasR-like receptor is not coupled to phosphoinositide-specific phospholipase C (17). Other data suggest that the osteoblast response to aluminum and other impermeant extracellular cations is mediated by a G-protein-coupled receptor-like mechanism rather than direct activation of G-proteins (17). Additional work is needed, however, to define the signal transduction pathways that link extracellular cations to the observed functional responses. Recently, we have identified several partial sequences that are related to the family of calcium-sensing receptors and pheromone receptors, which are potential candidates for the osteoblastic CasR protein (20). Thus far, we have been unable to establish that any of these receptors account for the observed response to extracellular cations in osteoblasts.
In conclusion, our studies indicate that a functionally related, but distinct gene product likely mediates the cationsensing response in cultured osteoblasts. Ultimate proof of this novel osteoblastic CasR-like receptor and its role in regulating bone remodeling will require, however, its isolation and clon-ing. At present, however, our finding that osteoblasts derived from CasR Ϫ/Ϫ mice retain a functional response to extracellular cations provides justification for the further pursuit of the putative osteoblastic extracellular cation-sensing mechanism.