Parathyroid hormone activates PKC-δ and regulates osteoblastic differentiation via a PLC-independent pathway

Bone 38 (2006) 485 – 496 www.elsevier.com/locate/bone

Parathyroid hormone activates PKC-δ and regulates osteoblastic differentiation via a PLC-independent pathway Dehong Yang, Jun Guo, Paola Divieti, F. Richard Bringhurst ⁎ Endocrine Unit, Massachusetts General Hospital and Harvard Medical School, Boston, MA 02114, USA Received 18 May 2005; revised 20 September 2005; accepted 6 October 2005 Available online 1 December 2005

Abstract PTH exerts major effects upon bone by activating PTH/PTHrP receptors (PTH1Rs) expressed on osteoblasts. The PTH1R is capable of engaging multiple signaling pathways in parallel, including Gs/adenylyl cyclase (AC), Gq/phospholipase C/protein kinase C (PLC/PKC) and a distinct mechanism, involving activation of PKC via a PLC-independent pathway, that depends upon ligand determinants within the PTH(29– 34) sequence. The involvement of PLC-dependent vs. PLC-independent PKC activation in PTH action was studied in clonal PTH1Rexpressing murine calvarial osteoblasts (“Wt9”) using two signal-selective analogs, [G1,R19]hPTH(1–28) and [G1,R19]hPTH(1–34). Both analogs lack PLC signaling but differ in their capacity to activate the PLC-independent PKC pathway. Both hPTH(1–34) and [G1,R19]hPTH (1–34), but not [G1,R19]hPTH(1–28), increased differentiation of Wt9 cells during a 16-day alternate-daily treatment protocol. Wt9 cells expressed PKC-βI, -δ, -ε and -ζ, none of which exhibited net translocation to membranes in response to hPTH(1–34) or either analog. hPTH (1–34) induced activation of membrane-associated PKC-δ, however, and a time- and concentration-dependent increase in cytosolic [phosphoThr505]PKC-δ which was maximal within 40 s at 100 nM in both Wt9 cells and primary osteoblasts. This response was mimicked by [G1,R19] hPTH(1–34) but not by [G1,R19]hPTH(1–28). Increased expression of bone sialoprotein (BSP) and osteocalcin (OC) mRNAs induced by PTH (1–34) and [G1,R19]hPTH(1–34) in Wt9 cells was blocked by rottlerin, a PKC-δ inhibitor. We conclude that PTH1Rs activate PKC-δ by a PLC-independent, PTH(29–34)-dependent mechanism that promotes osteoblastic differentiation. © 2005 Elsevier Inc. All rights reserved. Keywords: Parathyroid hormone; Parathyroid hormone receptor; Protein kinase C; Osteoblastic differentiation

Introduction Parathyroid hormone (PTH) is an 84-amino-acid polypeptide hormone that functions as a major regulator of bone metabolism and calcium homeostasis [1,8,19,32]. The responses of bone to PTH are complex but of intense current interest in light of the documented therapeutic efficacy of PTH in treatment of Abbreviations: PTH, parathyroid hormone; PTH1R, PTH/PTHrP receptor; AC, adenylyl cyclase; PLC, phospholipase C; ERK, extracellular signal-related kinase; PKC, protein kinase C; DAG, diacylglycerol; IP3, inositol trisphosphate; PKA, protein kinase A; Epac, exchange protein activated directly by cyclic AMP; GEF, guanine nucleotide exchange factor; PS, phosphatidylserine; FBS, fetal bovine serum; IBMX, isobutylmethylxanthine; TCA, trichloroacetic acid; IP1, inositol monophosphate; IP2, inositol diphosphate; ALP, alkaline phosphatase; Nr4a2, nuclear receptor 4a 2. ⁎ Corresponding author. Fax: +1 617 726 7543. E-mail address: [email protected] (F.R. Bringhurst). 8756-3282/$ - see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2005.10.009

osteoporosis. The net effect of exogenous PTH upon bone mass depends on the mode of hormone administration [6,11,38]. Continuous exposure to high concentrations, without adjunctive antiresorptive therapy, leads to progressive bone loss, whereas intermittently administrated PTH augments bone mass, an effect that preferentially affects trabecular bone. The mechanisms underlying these divergent effects of PTH upon bone mass remain incompletely understood. One possibility is that continuous and intermittent PTH exposures elicit different patterns of signaling within bone cells that express the heterotrimeric G-protein-coupled PTH/PTHrP receptor (PTH1R), which is capable of activating multiple second messenger cascades. Thus, binding of PTH to PTH1Rs leads to Gs-dependent activation of adenylyl cyclase (AC), Gqdependent activation of phospholipase C (PLC), activation of Gi and other responses, including receptor phosphorylation and internalization, that could stimulate nonreceptor tyrosine

486

D. Yang et al. / Bone 38 (2006) 485–496

kinases, extracellular signal-related kinases (ERKs) or other effectors [10,16,26]. Protein kinase C (PKC) may be activated in response to diacylglycerol (DAG) and inositol trisphosphate (IP3)-stimulated increased cytosolic calcium generated via PLC. The ultimate distal response to this complex pattern of receptor signaling may be difficult to predict. For example, with respect to cellular proliferation, cyclic AMP generated via Gs/ AC activation may regulate osteoblastic growth via an inhibitory effect of protein kinase A (PKA) on growth factordependent ERK activation [2,33,35,41,45] or, in cells that express a particular B-Raf isoform, via a stimulatory effect mediated by exchange protein activated directly by cyclic AMP (Epac), a guanine nucleotide exchange factor (cAMP-GEF) that engenders ERK activation [15]. PKC has been found to mediate positive regulation of osteoblastic proliferation by PTH as well [28,45]. PKCs comprise a family of serine-threonine kinases that are physiologically activated by lipid cofactors and serve as important transducers in agonist-induced signaling cascades. The ten known murine PKC isoforms include the conventional PKCs (cPKCs)-α, -βI, -βII, -γ, which can be activated by Ca++, phosphatidylserine (PS), DAG and phorbol ester (i.e., TPA); novel PKCs (nPKCs)-δ, -ε, -η, -θ, which require PS, DAG and phorbol ester and atypical PKCs (aPKCs)-λ, -ζ, which are only dependent on PS. PTH can activate PKCs via a PLC-dependent pathway that involves Gq-dependent formation of DAG and IP3, which stimulates release of Ca++ from intracellular stores [5,9]. Additionally, PTH can stimulate extracellular influx of Ca++ [52]. PKC also may be activated via PTH1Rs by a poorly understood PLCindependent mechanism, as shown by studies in which PKC is activated by amino(N)-truncated PTH(1–34) analogs that cannot activate PLC [20,22,25,45,47,49]. Detailed analysis of the structural determinants within the PTH(1–34) ligand required for PLC-independent PKC activation previously localized the critical residues within the sequence hPTH (29–34) [23,49]. In the present studies, we have employed hPTH(1–34) analogs in which an N-terminal Ser1- N Gly1 mutation is used to eliminate PLC activation via rodent PTH1Rs while preserving AC activity [46]. By further truncating the [Gly1]hPTH analog at the carboxyl(C) terminus (to eliminate the PKC-activating PTH(29–34) sequence) and incorporating a Glu19- N Arg19 substitution previously shown to overcome the lowered binding affinity of such C-truncated analogs [46], we have prepared PTH analogs [G1,R19]hPTH (1–28) and [G1,R19]hPTH(1–34) to assess the roles of PLCdependent vs. PLC-independent PKC activation, respectively, in osteoblasts expressing PTH1Rs. Materials and methods

the Glu19 to Arg19 change preserves PTH1R binding affinity in hPTH(1–28) peptides [46].

Cell culture and plasmid transfection F1–14, a conditionally immortalized (temperature-sensitive SV40 T antigen) PTH1R-null osteoblastic cell line [12], was used in these studies to lay the groundwork for future functional studies of mutant PTH1Rs expressed in such cells and also because stable transfection of PTH1R cDNA in PTH1R-null host cells minimizes secondary effects on PTH1R expression. F1–14 cells were co-transfected with a phCMV1 plasmid (Gene Therapy Systems, Inc., San Diego, CA) incorporating the wild-type mouse PTH1R gene and a pMG plasmid containing the hygromycin resistance gene (InvitroGen, San Diego, CA). Transfection was performed using Fugene 6.0, following the instructions provided by the manufacturer (Roche Diagnostics, Indianapolis, IN). After transfection, colonies of cells selected in the presence of 300 μg/ml hygromycin (Mediatech, Inc., Herndon, VA) were isolated and screened for acquisition of a cAMP response to 100 nM PTH(1–34). Unless indicated otherwise, cells were cultured at 33°C in αMEM (Gibco, Grand Island, NY) supplemented with 10% fetal bovine serum (FBS) (Gibco, Grand Island, NY) and 1% penicillin– streptomycin and passaged by trypsinization twice weekly. EW29 clonal LLC-PK1 porcine renal epithelial cells stably transfected with rat PTH1Rs (300,000 per cell) were maintained in αMEM with 5% FBS at 37°C and subcultured weekly by trypsinization as previously described [46]. Normal calvarial osteoblasts were isolated from 2-day-old C57BL/6 neonatal mice by sequential collagenase (types I and II; ratio 1:3) digestion as previously described [12]. Of 5 fractions generated by serial 20-min digestions, fractions 2 to 4 were combined, and the cells were plated at 5 × 105/cm2 and maintained in a humidified atmosphere (95% air/5% CO2) at 37°C using the same growth medium as described above. Animals were maintained in facilities operated by the Center for Comparative Medicine of the Massachusetts General Hospital in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were employed using protocols approved by the institution’s Subcommittee on Research Animal Care.

Radioligand binding PTH1R expression in transfected cells was determined by radioligand binding analysis, as previously described [18]. In brief, confluent monolayers of cells in 24-well plates were washed with 0.5 ml binding buffer (100 mM NaCl, 5 mM KCl, 2 mM CaCl2, 50 mM Tris–HCl, pH 7.8, plus 5% heat-inactivated horse serum) before incubation with 125I-labeled hPTHrP(1–36) amide (100,000–150,000 cpm/well) in 0.5 ml binding buffer overnight at 4°C. Receptor number was ascertained by Scatchard analysis, using nonradioactive hPTH (1–34) as competing ligand.

Cyclic AMP accumulation Cells were plated into 24-well dishes and incubated at 33°C. After reaching confluence, they were washed with assay buffer (135 mM NaCl, 6 mM KCl, 1 mM MgCl2, 2.8 mM glucose, 1.2 mM CaCl2, and 20 mM HEPES, pH 7.4) and incubated for 15 min at 37°C in the same buffer containing 0.1% heatinactivated BSA, 1 mM isobutylmethylxanthine (IBMX, Sigma-Aldrich Inc., St. Louis, MO) and agonist, conditions under which cAMP accumulation was found to be linear with time for 15 min. Buffer then was aspirated quickly, plates were frozen immediately in liquid nitrogen, and the frozen cells subsequently were thawed directly into 50 mM HCl. The cellular cAMP in the acid extracts was measured using a radioimmunoassay kit (New England Nuclear Corp., Boston, MA). Data were expressed as nanomoles per well.

Peptides Phosphatidylinositol hydrolysis Human PTH(1–34)NH2, [G1,R19]hPTH(1–28)NH2 (“G1R19(1–28)”) and 1 19 [G ,R ]hPTH(1–34)NH2 (“G1R19(1–34)”) were synthesized and quantified by amino acid analysis in the Biopolymer Core Laboratory of the Endocrine Unit, Massachusetts General Hospital. The substitution of Ser1 to Gly1 previously was shown to eliminate PLC signaling via rodent PTH1Rs, and

Confluent cells in 6-well (Wt9) or 12-well (EW29) dishes were incubated for 24 h with 4 μCi/ml myo-[3H]inositol (PerkinElmer Life Science, Inc., Boston, MA) in inositol-free DMEM medium (Specialty Media, Phillipsburg, NJ) containing 10% FBS at 37°C. The cells then were incubated in inositol-free

D. Yang et al. / Bone 38 (2006) 485–496

487

For analysis of matrix mineralization, cells plated and treated as above were fixed in 10% neutral formalin on day 16, and the presence of mineralized nodules was assessed by Von Kossa staining. Briefly, the fixed cells were gently washed with distilled water, incubated in 5% silver nitrate, and then exposed to sunlight for 5 min. Alternatively, to determine the calcium content of the cultures, cell monolayers were washed in Ca++- and Mg++-free PBS and then incubated for 3 h in 0.2 ml of 0.6 N HCl. Extracted calcium then was measured spectrophotometrically at 612 nm after reaction with methylthymol blue (SigmaAldrich Inc., St. Louis, MO) [17].

Cell subfractionation

Fig. 1. PTH1R expression in Wt9 cells. Wt9 cells were cultured at 33°C until confluent, then incubated with 125I-labeled [Tyr34]hPTHrP(1–36) amide (100,000–150,000 cpm/well) in the presence of hPTH(1–34) at the indicated concentrations overnight at 4°C before measurement of cell-associated radioactivity. Each point is the mean ± SD of specific binding for duplicate wells, expressed as a percentage of maximal binding in the absence of PTH. The inset shows Scatchard analysis, where the ordinate is the bound/free ratio and the abscissa is specific binding. DMEM containing 10 mM LiCl3 with/without 1000 nM hPTH(1–34) peptide. After 40 min, medium was aspirated, cold 5% trichloroacetic acid (TCA) was added (1 ml/well), plates were incubated on ice for 2 h, extracts were transferred to 15 ml Falcon tubes, wells were washed once with 5% TCA, and the combined acid extracts then were extracted twice with water–saturated diethyl ether before application to columns containing 1 ml AG1-X8 anion exchange resin (Econocolumn, Bio-RA laboratories, Richmond, CA). The columns were washed with 10 ml of 10 mM myo-inositol to remove free [3H]inositol and then with 8 ml Borax/60 mM ammonium formate to elute glycerophosphositides. [3H]inositol monophosphate (IP1), [3H]inositol bisphosphate(IP2), and [3H] inositol triphosphate (IP3) were eluted sequentially with 8 ml 0.2 M ammonium formate–0.1 M formic acid, 10 ml 0.5–0.7 M ammonium formate–0.1 M formic acid, and 8 ml 1.05 M ammonium formate-0.1 M formic acid, respectively. Radioactivity in 4 ml column fractions was measured by liquid scintillation counting. For measurement of total inositol phosphates, elution was performed with 8 ml of 1.05 M ammonium formate–0.1 M formic acid following the initial Borax/ammonium formate wash.

Cells plated into 150-mm dishes were cultured in αMEM with 10% FBS at 37°C for 5 days or until 95% confluence prior to preliminary incubation in serum-free medium for 24 h and subsequent addition of agonists. At appropriate times, cells were washed twice with cold PBS before addition of 500 μl of hypotonic Buffer A (50 mM Tris–HCl, pH 7.4; 2 mM EDTA; 2 mM EGTA; 10 mM β-mercaptoethanol; 4 μl/ml of protease inhibitor cocktail (Sigma-Aldrich Inc., St. Louis, MO) and 1 mM sodium orthovanadate). After allowing cells to swell on ice for 1 h, they were homogenized (Pellet Pestle Motor), and the resulting lysates were spun at 600×g for 20 min at 4°C to precipitate the nuclei and unbroken cells. Supernatants then were centrifuged at 50,000 × g for 30 min at 4°C to separate cysotol and membranes. The membrane pellets were resuspended in Buffer A containing 1% Triton-X100, sonicated and recentrifuged at 9000 × g for 20 min to obtain extracted membrane proteins.

PKC isoform immunoprecipitation and kinase assays Cell lysates (15–35 μg/50 μl) were pre-cleared by incubating with 10 μl of a 50% slurry of protein A Sepharose beads (Amersham Pharmacia Biotech, Piscataway, NJ) at 4°C for 10 min on a rocker apparatus. The beads were removed by centrifugation, and the supernatant was incubated with anti-PKC-δ antibody (50 μg lysate protein/1 μg antibody) for 2 h at 4°C. Incubation with fresh protein A Sepharose beads then was conducted as above, and the centrifuged beads were washed three times with cold PBS. PKC kinase activity was measured directly on the beads by using a PKC activity assay kit (Upstate, Lake Placid, NY). Immunoprecipitated PKC isoform was incubated with specific substrate peptide MARCKS (QKRPSQRSKYL), Ca++, Mg++, [γ-32P]ATP and

Assays of cellular differentiation Cells were plated in 24-well plates coated with type I collagen (BD Bioscience, MA) and incubated at 37°C until 90% confluent, whereupon they were transferred to 39°C. After a further 5 days, the medium was switched to αMEM supplemented with 10% FBS, 100 mM β-glycerophosphate and 50 μg/ml ascorbic acid containing fresh peptide (100 nM) or vehicle alone and refed thereafter with the same medium every 48 h. On day 13, cells in some wells were fixed with 10% neutral formalin, washed with PBS, and incubated in alkaline phosphatase (ALP) staining solution (0.1 mg/ml Naphthol AS-MX phosphate (Sigma-Aldrich Inc., St. Louis, MO), 5 μl/ml N,N-dimethylformamide, 0.1 M Tris–HCl, pH 8.5, and 0.6 mg/ml Fast Blue BB Salt) for 30 min before washing with distilled water and visualization of ALP by light microscopy. ALP activity also was measured enzymatically. For this purpose, cells incubated as above were washed twice with PBS, harvested with Tris– Triton solution (50 mM Tris–HCl, pH 7.6, plus 0.1% Triton X-100), scraped, and sonicated. Aliquots of the resulting lysates were assayed for ALP (hydrolysis of p-nitrophenol phosphate) for 15 min at 37°C using a commercial kit (Sigma-Aldrich Inc., St. Louis, MO). Activity was normalized to lysate protein content, as measured using the Bradford reagent (SigmaAldrich Inc., St. Louis, MO).

Fig. 2. Dose-dependent activation of adenylyl cyclase by PTH(1–34), G1R19 (1– 28), and G1R19 (1–34). Confluent monolayers of Wt9 cells were incubated for 15 min at 37°C in cAMP assay buffer containing hPTH(1–34) ( ), G1R19 (1– 28) (▴), G1R19 (1–34) (▾) (10−11 to 10−6M), or vehicle before cAMP measurement. Values are mean ± SD of triplicate wells. The experiment was repeated three times with similar results.

▪

488

D. Yang et al. / Bone 38 (2006) 485–496

liquid activator (containing DAG/PS) at 30°C for 20 min before addition of 2× SDS sample buffer (125 mM Tris–HCl pH 6.8, 4% w/v SDS, 20% glycerol, 100 nM DTT, 0.02% w/v bromophenol blue), boiling (3 min) and application of 40 μl of the resulting solution to P81 anion-exchange paper. The P81 papers then were washed with 0.75% phosphoric acid (×3) and acetone (×1) before measurement of adsorbed radioactivity by liquid scintillation counting. PKC activity was defined as the amount of [γ-32P]ATP incorporated into the MARCKS substrate in 1 min/mg of total protein lysate used for immunoprecipitation (pmol/mg/min). Equivalent amounts of PKC-δ were present in each reaction, as verified by Western blotting (described below).

Western blots Proteins (8 μg/well) were separated by 8% SDS-PAGE and transferred to Hybond nitrocellulose membranes (Amersham Life Science, Inc., Arlington Heights, IL) using standard procedures. Membranes were blocked for 30 min at room temperature in Tris-buffered saline, pH 7.6, plus 0.05% Tween-20 (TBST) with 5% powdered nonfat milk solution, immunostained overnight at 4°C with antibodies directed against specific PKCs (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) or phospho-PKCs (Cell Signaling Technology, Beverly, MA), washed further in TBST, and then incubated for 60 min with a 1:2000 dilution of antirabbit immunoglobulin G [IgG] second antibody conjugated to horseradish peroxidase [HRP] (Santa Cruz Biotechnology Inc., Santa Cruz, CA). Immunoreactive bands were detected by enhanced chemiluminescence assay (PerkinElmer Life Sciences, Boston, MA) conducted according to the manufacturer’s instructions. Band intensities were quantitated by densitometry (AlphaImager™ 2200, Alpha Innotech Corporation, San Leandro, CA). For graphical presentation, band intensities were normalized to the level of the same protein measured in control cells, the level in which was set at 1.0.

Measurement of mRNA regulation by real-time RT-PCR Wt9 cells cultured in collagen I-coated plates were transferred from 33°C to 37°C upon reaching 80% confluence. After a further 5 days, the medium was switched to αMEM supplemented with 10% FBS, 100 mM β-glycerophosphate and 50μg/ml ascorbic acid. Cells then were refed every other day for a further 6 days before being washed twice with serum-free medium and cultured for another 24 h in αMEM medium containing 1% FBS, 100 mM β-glycerophosphate, and 50 μg/ml ascorbic acid. PTH peptide then was added, in the presence of 1 μM rottlerin (Acros Organics, supplied by Fisher Co.) or DSMO vehicle alone (added 1 h previously), and incubations were continued for a further 4 h before extraction of total RNA using an RNeasy Mini Kit (Qiagen Science,

MD). Expression of BSP, OC, and ALP was measured by two-step real-time RTPCR. Briefly, the first strand of cDNA was synthesized according to the manufacturer’s instructions using a SuperScript First-Strand kit (Invitrogen Life Technologies, Carlsbad, CA). For each gene, two specific PCR primers (BSP/ fw, 5′-AGGGAACTGACCAGTGTTGG-3′; BSP/re, 5′-ACTCAACGGTGCTGCTTTT-3′. OC/fw, 5′-AAGCAGGAGGGCAATAAGGT-3′; OC/re, 5′-GCGGTCTTCAAGCCATACTG-3′. ALP/fw, 5′-GCTGATATGAGATGTCCTT-3′; ALP/re, 5′-GCACTGCCACTGCCTACT-3′. GAPDH/fw, 5′-TGTCGTGGAGTCTACTGGTG-3′; GAPDH/re, 5′-GCATTGCTGACAATCTTGAG-3′) were designed and synthesized by Sigma-Genosys (Woodlands, TX). The PCR reactions (94°C, 20 s; 60°C, 20 s; 72°C, 20 s) were performed on an Opticon OPTICON® 2 (CFD-3220; MJ Research, Inc., Waltham, MA) using a QuantiTect™ SYBR Green PCR kit (Qiagen, Valencia, CA). Gene expression was normalized to that of GAPDH and then expressed as fold over control.

Statistics Each experiment was repeated at least three times. The statistical analysis was carried out using GraphPad Prism software (GraphPad Software Inc., San Diego, CA). The significance of differences was determined based on analysis of variance (ANOVA). When significant differences were detected, group means were compared using Bonferroni’s multiple comparison test.

Results Isolation of clonal osteoblasts stably expressing wild-type murine PTH1Rs To prepare osteoblasts constitutively expressing murine PTH1Rs, clonal, conditionally immortalized, PTH1R-null, osteoblastic F1–14 cells, previously isolated from calvarial bones of fetal mice lacking both alleles of the PTH1R gene (i.e., PTHR(−/−)) [12] were co-transfected with plasmids encoding wild-type PTH1R cDNA and the hygromycin resistance gene. Forty-eight hours after transfection, hygromycin was added to the medium, and 51 colonies surviving this selection were isolated 3 week later. Upon screening for cAMP responsiveness to hPTH(1–34), 23 subclones were found to express functional

Fig. 3. PLC activation by hPTH(1–34), G1R19 (1–28) and G1R19 (1–34). Confluent monolayers of Wt9 cells (A) or EW29 cells (B) were labeled with [3H]myo-inositol for 24 h at 37°C, and PTH peptides were added for 40 min before isolation of labeled inositol phosphates. (A) Wt9 cells incubated with 1000 nM PTH(1–34) or vehicle alone for 40 min. The cellular content of radioactive IP1, IP2, and IP3 then was determined. (B) Total [3H]inositol phosphates (“Total IPs”) were measured in EW29 cells after exposure for 40 min to hPTH(1–34) ( ),G1R19 (1–28) (▴) or G1R19 (1–34) (▾) at the indicated concentrations. Data are expressed as mean ± SE of three independent experiments. *P b 0.01 vs. control.

▪

D. Yang et al. / Bone 38 (2006) 485–496

489

(Fig. 1), exhibited PTH1R density within the presumed physiologic range and was selected for further study after confirming that it could differentiate to form mineralized nodules under appropriate conditions (see Fig. 4B). Signaling properties of hPTH(1–34), G1R19(1–28) and G1R19(1–34) Substitution of Gly for the N-terminal Ser in hPTH(1–34) previously was shown to abolish PLC activation via rat PTH1Rs in LLC-PK1 porcine renal epithelial cells [46]. Further, substitution of Arg19 for Glu19 partially mitigates the loss of apparent binding affinity and potency for AC activation that otherwise attends removal of the sequence hPTH(29–34) [46], a region capable of activating PKC in a PLC-independent manner via PTH1Rs [4,22,46,47]. To characterize their utility as probes of specific PTH1R signaling pathways, we compared signaling via PTH1Rs of G1R19(1–28) and G1R19(1–34) with that of hPTH(1–34). Concentration-dependent activation of cAMP generation over 15 min in Wt9 cells by each of these three peptides (10−11 to 10−6 M) is shown in Fig. 2. In this system, hPTH(1–34)

Fig. 4. Effects of PTH(1–34), G1R19(1–28) and G1R19(1–34) on differentiation of osteoblastic cells. Wt9 cells were plated on collagen I-coated 24-well plates at 33°C and transferred to 39°C upon reaching 90% confluence. Cells then were maintained in αMEM containing 10% FBS, 100 nM β-glycerolphosphate and 50 μg/ml ascorbic acid and refed every 48 h thereafter with medium containing fresh peptide (100 nM) or vehicle alone. (A) ALP was detected by histochemical staining and measured enzymatically (on day 13). (B) Calcium accumulation was assessed by von Kossa staining and by measurement of calcium mass (on day 16). Images in the top panel depict duplicate wells for each treatment. Graphical data are presented as mean ± SE of three independent experiments. *P b 0.01 vs. control.

PTH1Rs, Scatchard analysis of which demonstrated a range of PTH1R expression from 40,000–120,000 sites/cell. Although PTH1R density on normal bone cells is not known with certainty and may vary from one cell type to another, available data suggest that physiologic expression of PTH1Rs by osteoblasts may be in the range of 10,000–100,000 per cell [7,27,43]. Subclone Wt9, shown to express about 80,000 receptors/cell

Fig. 5. Effect of PTH(1–34) on the activation of PKC-δ. Confluent Wt9 cells were incubated in serum-free medium for 24 h before addition of hPTH(1–34) (100 nM, 2 min), 8BrcAMP (100 μM, 2 min), TPA (100 nM, 15 min), or vehicle alone. Membrane (A and B) and cytosolic (A) fractions were isolated. The cell lysates were incubated with antibody to PKC-δ and then with Protein A sepharose beads for 2 h serially at 4°C. PKC-δ activity associated with the washed beads then was measured. Data are expressed as mean ± SE of three determinations. *P b 0.01 vs. control.

490

D. Yang et al. / Bone 38 (2006) 485–496

showed an EC50 of 1.67 ± 0.83 nM, whereas that of G1R19(1– 28) was approximately 5-fold higher (8.17 ± 3.5 nM), as previously reported in LLC-PK1 cells that stably express approximately 300,000 transfected rat PTH1Rs per cell [46]. Extension of the C-terminus to position 34, as in G1R19(1–34), restored the EC50 for AC to a value (1.76 ± 0.74 nM), comparable to that of hPTH(1–34). All three peptides demonstrated full agonism, however, and no differences in maximal cAMP accumulation were observed at peptide concentrations at or above 100 nM (Fig. 2). Activation of phosphatidylinositol hydrolysis by hPTH(1– 34) via transfected PTH1Rs in Wt9 cells was observed (Fig. 3A) but was too modest in amplitude to enable reliable comparison of the potencies of different PTH analogs (average increases in total inositol phosphates of 50–80% over basal, consistently observed only in IP1 and IP2 fractions). As an alternative, we tested the mutant PTH analogs for PLC activation using clonal LLC-PK1 cells (EW29) that express 300,000 rat PTH1Rs per cell and manifest a substantial increase in PLC activity in response to hPTH(1–34). In these highly responsive cells, neither G1R19(1–28) nor G1R19(1– 34) could elicit significant PLC activation, even at concentrations as high as 1000 nM (Fig. 3B). Role of PLC-independent signaling in osteoblast differentiation To assess the involvement of PLC-independent signaling pathway(s) in regulating osteoblast differentiation, we administered PTH(1–34), G1R19(1–28), or G1R19(1–34) to osteoblasts intermittently (once every 48 h) at the concentration of 100 nM. At day 13, ALP activity was investigated by histochemical staining and enzymatic measurement, which showed that PTH(1–34) could increase ALP activity, whereas G1R19(1–28) was inactive and G1R19(1–34) caused an increase that was not significant (Fig. 4A). At day 16, both von Kossa staining and calcium mass measurements indicated that G1R19(1–34), but not G1R19(1–28) could increase this measure of osteoblastic differentiation, albeit to a lesser extent than hPTH(1–34) (Fig. 4B).

PTH(1–34) effect on PKC-δ/ζ activation To analyze activation of PKC(s) by PTH1Rs in Wt9 cells, we first performed Western blots of total cell lysate proteins using PKC isoform-specific antibodies. Of the 10 isoforms sought by this technique, only PKCs βI, δ, ζ, and ε were visualized in Wt9 cells and its parent F1–14 cells (data not shown). Among these isoforms, PKC-δ is known to increase cell differentiation in several systems and has been found to respond to PTH in rat osteosarcoma cells [14,21,29,31,34], whereas the atypical PKC family member, PKC-ζ, has been shown to be unresponsive to PTH(1–34) stimulation [13]. To assess the effect of PTH on activation of the PKC-δ and PKC-ζ isoforms, Wt9 cells were lysed 2 min after exposure to 100 nM hPTH(1–34) and cytosol and solubilized membrane fractions were prepared. Isoform-specific antibodies were used to immunoprecipitate the PKCs from each fraction by adsorption to protein A-Sepharose beads, whereupon kinase assays were performed on the beads using the MARCKS peptide substrate. Preliminary Western blots of SDS-solubilized immunoprecipitates confirmed the presence of the expected PKC isoforms, including their phosphorylated forms (data not shown). As shown in Fig. 5A, hPTH(1–34) evoked significant activation of PKC-δ in the membrane fraction but not in the cytosolic fraction. PKC-δ activity in the membrane fraction was stimulated also by TPA, as expected, but not by 8Br-cAMP (Fig. 5B). In analogous experiments using antibodies to PKC-ζ, no activation was observed in response to hPTH(1–34), TPA or 8Br-cAMP (data not shown). Role of PLC-dependent vs. PLC-independent pathways in PKC-δ activation As noted above (Fig. 2), G1R19(1–28) and G1R19(1–34) both are capable of maximally activating cAMP accumulation via murine PTH1Rs but, unlike hPTH(1–34), neither can activate PLC at concentrations as high as 1000 nM (Fig. 3).

Fig. 6. Effects of PTH(1–34), G1R19(1–28) and G1R19(1–34) on PKC activity. Confluent Wt9 cells were cultured in serum-free medium at 37°C for 24 h before incubation for 2 min with 100 nM of PTH(1–34), G1R19(1–28) or G1R19(1–34). Measurement of PKC-δ activity in protein A-sepharose bead-associated immunoprecipitates was performed and depicted as in Fig. 5. The PKC-δ activity is expressed as mean ± SE of three separate experiments. *P b 0.01 vs. control.

D. Yang et al. / Bone 38 (2006) 485–496

These two analogs would be expected to differ, however, in their ability to activate PKC(s) via a PLC-independent pathway, a property of PTH analogs that retain the hPTH (29–32) sequence [22], which is present in G1R19(1–34) but not in G1R19(1–28). To address the potential involvement of PLC-independent mechanisms in activation of PKC-δ, Wt9 cells were exposed to 100 nM of G1R19(1–28) or G1R19(1–34) for 2 min, and membrane PKC activity subsequently was measured in immunoprecipitates prepared using antibody to PKC-δ. As shown in Fig. 6A, G1R19(1–34), like hPTH(1–34), activated

491

PKC-δ in Wt9 cell membranes, whereas G1R19(1–28) was inactive. No activation was observed in cytosolic immunoprecipitates with any of these peptides. The failure of G1R19(1–28) to activate membrane PKC-δ at 100 nM seemed not to be due to its 5-fold lower potency, as previously determined in the cAMP assay (Fig. 2) because even 1000 nM of G1R19(1–28) was incapable of activating membrane PKC-δ (Fig. 6B). These results suggested that PTH activates PKC-δ in Wt9 cells via a PLC-independent mechanism that does not involve cAMP production (which is fully stimulated by the G1R19(1–28)).

Fig. 7. Effect of PTH(1–34) on translocation and phosphorylation of PKC-δ in Wt9 cells. Confluent Wt9 cells were cultured in serum-free medium at 37°C for 24 h before addition of PTH(1–34) (100 nM) for the indicated time intervals (A and B) or for 2 min at the indicated PTH concentrations (C and D). TPA (100 nM) was present for 15 min and 8Br-cAMP (100 μM) for 2 min. Western blots of subcellular fractions (membranes, A and C; or cytosol, B and D), prepared as described in Materials and methods, were probed with antibodies of the indicated specificity and bands were visualized by chemiluminescence. Measurement of β-actin expression was employed to control for protein loading. Western blot results for PKC-ζ and [P-Thr410]PKC-ζ, which were not regulated by PTH, are shown as controls (left panels). Graphs depict the densitometric results for β-actin, PKC-δ, and [P-Thr505]PKC-δ band intensities, each normalized to its respective level in control cells and expressed as mean ± SE, from three independent experiments. *P b 0.05 vs. control.

492

D. Yang et al. / Bone 38 (2006) 485–496

Effect of PTH(1–34) on PKC translocation and phosphorylation Human PTH(1–34), added at 100 nM to serum-starved Wt9 cells, caused no detectable translocation of PKCs βI, δ, ζ, or ε between cytosolic and membrane fractions at times between 40 s and 30 min. TPA did induce translocation from cytosol to membranes of PKC-βI, δ, and ε (but not of PKC-ζ), as expected. The results for PKC-δ and PKC-ζ are shown in Fig. 7. Using available antibodies to phosphorylated forms of PKC-δ (P-Thr505) and PKC-ζ P-Thr410 ), we did observe transient phosphorylation of PKC-δ between 40 s and 10 min in response to 100 nM hPTH(1–34), but this occurred in the cytosolic fraction and not, as with TPA, in the membrane fraction (Figs. 7A and B). No change in phospho-PKC-ζ was observed in

response to hPTH(1–34) in either membranes or cytosol (Figs. 7A and B, left panels). The concentration dependence of the PTH effect on PKC-δ phosphorylation was examined at 2 min after peptide addition to Wt9 cells (Figs. 7C and D). Again, no net translocation of PKC isoforms between cytosol and membranes could be detected in response to hPTH(1–34) at concentrations ranging from 1 to 1000 nM. The effect of TPA to translocate PKC-δ (but not PKCζ) again was seen, and no response was observed with the cAMP analog 8Br-cAMP. The increase in cytosolic phosphoPKC-δ induced by hPTH(1–34) was evident at concentrations between 10 and 1000 nM, but no increase in phospho-PKC-ζ was observed, even at the highest concentration (1000 nM) tested (Figs. 7C and D, left panels). The PTH effect on phosphoPKC-δ was not mimicked by 8Br-cAMP. Similar but less

Fig. 8. Effect of PTH(1–34) on translocation and phosphorylation of PKC-δ in primary calvarial osteoblasts. The same experimental protocol as in Fig. 7 was employed using confluent primary osteoblasts from neonatal C57BL/6 mice that had been cultured in serum-free medium at 37°C for 24 h before addition of PTH(1– 34), TPA, 8Br-cAMP, isolation of subcellular fractions and measurement of protein expression by Western blotting, as described in Fig. 7. *P b 0.05 vs. control.

D. Yang et al. / Bone 38 (2006) 485–496

striking results were observed in cultures of primary calvarial osteoblasts (Fig. 8), which were shown to exhibit cAMP responsiveness to hPTH(1–34), as expected (data not shown). Role of PLC-dependent vs. PLC-independent pathways in PKC-δ phosphorylation Like hPTH(1–34), G1R19(1–34) increased the abundance of phospho-PKC-δ in cytosol (Fig. 9, right panels) without inducing a net translocation of total PKC-δ protein between cytosol and membrane fractions (data not shown). G1R19(1– 28), on the other hand, had no effect. Neither of these PTH analogs altered the phosphorylation level or net translocation of PKC-ζ (Fig. 9). Role of PKC-δ in osteoblast gene regulation by PTH To investigate the relationship between PKC-δ activation and regulation of osteoblastic differentiation by PTH, gene expression of osteoblastic differentiation markers was measured. Expression of BSP and OC mRNAs (but, interestingly, not that of the earlier marker of osteoblastic differentiation, ALP) was increased 2-fold (Fig. 10A) by treatment for 4 h with PTH(1–34) or G1R19(1–34) but not G1R19(1–28). These increases were totally inhibited by the PKC-δ inhibitor, rottlerin (Fig. 10B). Expression of another gene known to be upregulated by PTH in osteoblasts via a PKA-dependent mechanism [48], Nuclear receptor 4a 2 (Nr4a2), also was increased by PTH(1–34) and by both analogs in a manner that was not inhibited by rottlerin treatment (see legend, Fig. 10; some data not shown). Thus, whereas rottlerin may not be

493

completely specific for PKC-δ, these results are consistent with involvement of PKC-δ in these responses to PTH, and the Nr4a2 results argue against nonspecific toxicity of the drug in this system. Discussion In this paper, we have employed mutant analogs of hPTH(1– 34) with selectively defective signaling properties to probe the roles of these signaling pathways in PTH1R action in osteoblasts. As reviewed in the Introduction, it is known that PTH1Rs activate PKCs via at least two mechanisms, one of which is associated with activation of PLC and increased cytosolic Ca++ and requires that the PTH ligand possess an intact N-terminus [46]. The other route of PKC activation is less well understood but maps to a different region of the PTH(1– 34) ligand [the sequence PTH(29–34)], is stimulated by Ntruncated PTH peptides incapable of activating PLC [4,22,45,47,49,50], and, by implication, likely involves different effectors. To distinguish the roles of these PLC-dependent and PLC-independent modes of PKC activation via PTH1Rs in intact osteoblasts, we designed two analogs, G1R19(1–34) and G1R19(1–28). The G1R19(1–28) was expected to retain the ability to activate AC but to lack PLC activity (on the basis of the Ser1- N Gly1 change) [46] and, as well, to be incapable of PLC-independent PKC activation, due to the absence of the PTH(29–34) sequence [22]. G1R19(1–34) was expected to be similarly incapable of activating PLC but to retain AC activity as well as PTH(29–34)-dependent, PLC-independent PKC activation. The results of the present study are entirely

Fig. 9. Effects of PTH(1–34), G1R19(1–28) and G1R19 (1–34) on PKC-δ and PKC-ζ phosphorylation. Confluent Wt9 cells were cultured in serum-free medium at 37°C for 24 h before incubation for 2 min with 10 nM or 100 nM of PTH(1–34), G1R19 (1–28), or G1R19 (1–34). TPA (100 nM) was added for 15 min. Cell fractionation and Western blotting with antibodies to phospho-PKC-δ (upper lanes) or phospho-PKC-ζ (lower lanes) were performed. Graphs depict the results, expressed as mean ± SE, of the percentage of expression observed in controls, of phospho-PKC-δ/ζ band intensities from three independent experiments. *P b 0.05 vs. control.

494

D. Yang et al. / Bone 38 (2006) 485–496

Fig. 10. Effect of rottlerin upon osteoblastic gene regulation by PTH. Wt9 cells in collagen I-coated plates were cultured in αMEM supplemented with 10% FBS, 100 mM β-glycerophosphate and 50 μg/ml ascorbic acid for 5 days and then switched to medium containing only 1% FBS for 24 h. Rottlerin (1 μM; panel B) or DMSO vehicle alone (panel A) then was added 1 h before addition of the indicated PTH peptides (100 nM), incubation for a further 4 h and subsequent RNA extraction. Expression of ALP, BSP, and OC was measured by two-step real-time RT-PCR, as described in Materials and methods, and expressed relative to that in control cultures not treated with PTH peptides. Experiments were repeated 3 times with similar results. In control experiments, regulation by PTH(1–34) of the expression of Nr4a2 was unaffected by rottlerin (control = 1, PTH = 3.4 ± 0.2, rottlerin = 0.9 ± 0.2, PTH + rottlerin = 3.5 ± 0.4; N.S.) *P b 0.01.

consistent with these expectations. Further, we have demonstrated activation of a specific PKC isoform (PKCδ) via the PLC-independent, PTH(29–34)-dependent signaling pathway, in that this response was observed with G1R19(1–34) but not with the closely related C-truncated analog, G1R19(1–28). Neither of these analogs activated PLC. A previous report by Erclik and Mitchell also pointed to involvement of PKC-δ in PTH regulation of IGFBP5 expression in UMR 106-01 osteosarcoma cells, a response that was inhibited by a dominant-negative PKC-δ mutant and was mimicked by PTH(3–34), suggesting mediation by the C-terminal portion of PTH(1–34) [14]. Further analysis of PTH1R regulation of differentiation of the Wt9 clonal osteoblasts studied here indicated that the PLC-independent mechanism engaged by PTH(29–34) also augments differentiation, as measured by production of mineralized matrix, in a manner that apparently cannot be mimicked by PTH1Rdependent generation of cAMP alone, of which the G1R19(1–28) analog is fully capable at the concentrations employed. This result is also consistent with previous descriptions of biologic actions in bone cells of PTHdependent PKC activation or of N-truncated PTH analogs

[3,14,25,36,37,40,42,44,45,51]. A direct involvement of PTH-dependent PKC-δ activation in the accelerated differentiation of Wt9 cells induced by 2 weeks of alternate-day PTH treatment cannot be proven, as available maneuvers to inhibit PKC-δ expression or activity cannot be readily applied over such an extended period of culture. We did examine the involvement of PKC-δ in the acute upregulation by PTH and G1R19(1–34) (but not by G1R19(1–28)) of BSP and OC in these cells and found these responses to be blocked by the relatively selective PKC-δ inhibitor rottlerin. By Western blot analysis, we detected expression in Wt9 cells of PKC isoforms from each major class: PKC-βI (conventional), PKC-δ and PKC-ε (novel) and PKC-ζ (atypical). These findings are similar to those reported in normal calvarial osteoblasts [39] and in UMR106-01 osteosarcoma cells [13], except that we did not consistently observe expression of PKC-α or PKC-η. In contrast to previous studies reporting PTH-induced membrane translocation of PKC isoforms in osteoblastic cells [13,14], we did not observe detectable translocation from cytosol to membrane fractions of any of the expressed PKC isoforms following PTH addition to either Wt9 cells or primary murine calvarial osteoblasts. This

D. Yang et al. / Bone 38 (2006) 485–496

was especially remarkable in the case of PKC-δ, which we found by kinase assays of specific immunoextracts to be activated exclusively in the membrane fraction within minutes of PTH addition. Equally surprising was the finding that PTH led to increased amounts of the Thr505 -phosphorylated form of PKC-δ in the cytosolic, rather than the plasma membrane fraction. Phosphorylation of this threonine in the activation loop generally is associated with, and even a prerequisite for, full PKC activity [30]. On the other hand, the molecular mechanisms of PKC activation are complex and may differ among isoforms [24,30]. Thus, PKC-δ, unlike conventional isoforms, is known to be capable of activity in the absence of phosphorylation at Thr505 and may be activated by other mechanisms, including tyrosine phosphorylation by growth factor receptors or nonreceptor tyrosine kinases [24]. In preliminary experiments, however, we have not observed PTH-dependent increases in tyrosine phosphorylation of PKC-δ (unpublished data). Another possibility relates to the fact that membrane-bound PKCs are much more sensitive to dephosphorylation than are soluble, cytosolic forms and that inactive but Thr505-phosphorylated forms may then accumulate in the cytosol following agonist stimulation [30]. Thus, we cannot exclude the possibility that rapid shuttling between cytosol and membranes and activation of membrane-associated PKC via mechanisms not revealed by the Thr505-specific antibody could have occurred without a net accumulation of enzyme molecules in the membranes. Clearly, further investigation will be required to determine the molecular and cellular mechanism(s) underlying this apparently paradoxical dissociation of increased Thr505 -phosphorylated forms of PKC-δ in the cytosol from the temporally associated activation of PKC-δ in the plasma membranes of these osteoblasts. It seems likely, however, that both responses result from the same proximal mechanism, as both are dependent upon activation by the PTH(29–34) sequence. In conclusion, our results point to a potentially significant role for PLC-independent PKC activation, mediated by interaction of the PTH(29–34) sequence with PTH1Rs, in PTH regulation of osteoblast differentiation. Further work is needed to assess the molecular mechanism(s) of this signaling process and the possible importance of this pathway, including the role of PKC-δ in particular, in actions of PTH in vivo.

Acknowledgments This work was supported by the National Institutes of Health award DK11794, DK02889 and DK65032. We thank Dr. Henry Kronenberg for critical review of the manuscript. References [1] Adami S, Mian M, Bertoldo F, Rossini M, Jayawerra P, O’Riordan JL, et al. Regulation of calcium-parathyroid hormone feedback in primary hyperparathyroidism: effects of bisphosphonate treatment. Clin Endocrinol (Oxf) 1990;33:391–7. [2] Angelo G, Lee H, Weiner RI. cAMP-dependent protein kinase inhibits

[3]

[4]

[5]

[6]

[7]

[8] [9]

[10]

[11] [12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

495

the mitogenic action of vascular endothelial growth factor and fibroblast growth factor in capillary endothelial cells by blocking Raf activation. J Cell Biochem 1997;67:353–66. Arao M, Yamaguchi T, Sugimoto T, Fukase M, Chihara K. Protein kinase C is crucial for the stimulation of sodium-dependent phosphate transport by parathyroid hormone-related peptide in osteoblast-like cells. Biochem Biophys Res Commun 1994;199:1216–22. Azarani A, Goltzman D, Orlowski J. Structurally diverse N-terminal peptides of parathyroid hormone (PTH) and PTH-related peptide (PTHRP) inhibit the Na+/H+ exchanger NHE3 isoform by binding to the PTH/ PTHRP receptor type I and activating distinct signaling pathways. J Biol Chem 1996;271:14931–6. Babich M, Choi H, Johnson RM, King KL, Alford GE, Nissenson RA. Thrombin and parathyroid hormone mobilize intracellular calcium in rat osteosarcoma cells by distinct pathways. Endocrinology 1991;129: 1463–70. Bellido T, Ali AA, Plotkin LI, Fu Q, Gubrij I, Roberson PK, et al. Proteasomal degradation of Runx2 shortens parathyroid hormone-induced anti-apoptotic signaling in osteoblasts. A putative explanation for why intermittent administration is needed for bone anabolism. J Biol Chem 2003;278:50259–72. Bos MP, van der Meer JM, Feyen JH, Herrmann-Erlee MP. Expression of the parathyroid hormone receptor and correlation with other osteoblastic parameters in fetal rat osteoblasts. Calcif Tissue Int 1996; 58:95–100. Breslau NA, Pak CY. Hypoparathyroidism. Metabolism 1979; 28: 1261–76. Civitelli R, Agnusdei D, Nardi P, Zacchei F, Avioli LV, Gennari C. Effects of one-year treatment with estrogens on bone mass, intestinal calcium absorption, and 25-hydroxyvitamin D-1 alpha-hydroxylase reserve in postmenopausal osteoporosis. Calcif Tissue Int 1988;42:77–86. Daaka Y, Luttrell LM, Lefkowitz RJ. Switching of the coupling of the beta2-adrenergic receptor to different G proteins by protein kinase A. Nature 1997;390:88–91. Dempster DW, Cosman F, Parisien M, Shen V, Lindsay R. Anabolic actions of parathyroid hormone on bone. Endocr Rev 1993;14:690–709. Divieti P, Lanske B, Kronenberg HM, Bringhurst FR. Conditionally immortalized murine osteoblasts lacking the type 1 PTH/PTHrP receptor. J Bone Miner Res 1998;13:1835–45. Dossing DA, Radeff JM, Sanders J, Lee SK, Hsieh MR, Stern PH. Parathyroid hormone stimulates translocation of protein kinase C isozymes in UMR-106 osteoblastic osteosarcoma cells. Bone 2001; 29:223–30. Erclik MS, Mitchell J. The role of protein kinase C-delta in PTH stimulation of IGF-binding protein-5 mRNA in UMR-106-01 cells. Am J Physiol Endocrinol Metab 2002;282:E534–41. Fujita T, Meguro T, Fukuyama R, Nakamuta H, Koida M. New signaling pathway for parathyroid hormone and cyclic AMP action on extracellularregulated kinase and cell proliferation in bone cells. Checkpoint of modulation by cyclic AMP. J Biol Chem 2002;277:22191–200. Gentili C, Morelli S, Boland R, de Boland AR. Parathyroid hormone activation of map kinase in rat duodenal cells is mediated by 3′,5′-cyclic AMP and Ca(2+). Biochim Biophys Acta 2001;1540:201–12. Gindler EM, King JD. Rapid colorimetric determination of calcium in biologic fluids with methylthymol blue. Am J Clin Pathol 1972; 58:376–82. Guo J, Iida-Klein A, Huang X, Abou-Samra AB, Segre GV, Bringhurst FR. Parathyroid hormone (PTH)/PTH-related peptide receptor density modulates activation of phospholipase C and phosphate transport by PTH in LLC-PK1 cells. Endocrinology 1995;136:3884–91. Hampl H, Steinmuller T, Frohling P, Naoum C, Leder K, Stabell U, et al. Long-term results of total parathyroidectomy without autotransplantation in patients with and without renal failure. Miner Electrolyte Metab 1999;25:161–70. Hollnagel A, Schroder D, Gross G. Domain-specific gene activation by parathyroid hormone in osteoblastic ROS17/2.8 cells. J Biol Chem 1996;271:21870–7. Jimenez SA, Gaidarova S, Saitta B, Sandorfi N, Herrich DJ, Rosenbloom

496

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30] [31]

[32]

[33]

[34]

[35]

[36]

D. Yang et al. / Bone 38 (2006) 485–496 JC, et al. Role of protein kinase C-delta in the regulation of collagen gene expression in scleroderma fibroblasts. J Clin Invest 2001;108:1395–403. Jouishomme H, Whitfield JF, Chakravarthy B, Durkin JP, Gagnon L, Isaacs RJ, et al. The protein kinase-C activation domain of the parathyroid hormone. Endocrinology 1992;130:53–60. Jouishomme H, Whitfield JF, Gagnon L, Maclean S, Isaacs R, Chakravarthy B, et al. Further definition of the protein kinase C activation domain of the parathyroid hormone. J Bone Miner Res 1994;9:943–9. Kikkawa U, Matsuzaki H, Yamamoto T. Protein kinase C delta (PKC delta): activation mechanisms and functions. J Biochem (Tokyo) 2002;132:831–9. Kim TY, Vargas V, Mayer H, Somjen D, Kaye AM. Selective anabolic effects of muteins of mid-region PTH fragments on skeletal tissues of prepubertal rats. Bone 2002;30:78–84. Lederer ED, Sohi SS, McLeish KR. Parathyroid hormone stimulates extracellular signal-regulated kinase (ERK) activity through two independent signal transduction pathways: role of ERK in sodium-phosphate cotransport. J Am Soc Nephrol 2000;11:222–31. McCauley LK, Koh AJ, Beecher CA, Cui Y, Rosol TJ, Franceschi RT. PTH/PTHrP receptor is temporally regulated during osteoblast differentiation and is associated with collagen synthesis. J Cell Biochem 1996;61:638–47. Miao D, Tong XK, Chan GK, Panda D, McPherson PS, Goltzman D. Parathyroid hormone-related peptide stimulates osteogenic cell proliferation through protein kinase C activation of the Ras/mitogen-activated protein kinase signaling pathway. J Biol Chem 2001;276:32204–13. Mischak H, Pierce JH, Goodnight J, Kazanietz MG, Blumberg PM, Mushinski JF. Phorbol ester-induced myeloid differentiation is mediated by protein kinase C-alpha and -delta and not by protein kinase C-beta II, -epsilon, -zeta, and -eta. J Biol Chem 1993;268:20110–5. Newton AC. Regulation of the ABC kinases by phosphorylation: protein kinase C as a paradigm. Biochem J 2003;370:361–71. Ohba M, Ishino K, Kashiwagi M, Kawabe S, Chida K, Huh NH, et al. Induction of differentiation in normal human keratinocytes by adenovirusmediated introduction of the eta and delta isoforms of protein kinase C. Mol Cell Biol 1998;18:5199–207. Parfitt AM. The actions of parathyroid hormone on bone: relation to bone remodeling and turnover, calcium homeostasis, and metabolic bone diseases: II. PTH and bone cells: bone turnover and plasma calcium regulation. Metabolism 1976;25:909–55. Partridge NC, Opie AL, Opie RT, Martin TJ. Inhibitory effects of parathyroid hormone on growth of osteogenic sarcoma cells. Calcif Tissue Int 1985;37:519–25. Pessino A, Passalacqua M, Sparatore B, Patrone M, Melloni E, Pontremoli S. Antisense oligodeoxynucleotide inhibition of delta protein kinase C expression accelerates induced differentiation of murine erythroleukaemia cells. Biochem J 1995;312(Pt 2):549–54. Piiper A, Gebhardt R, Kronenberger B, Giannini CD, Elez R, Zeuzem S. Pertussis toxin inhibits cholecystokinin- and epidermal growth factorinduced mitogen-activated protein kinase activation by disinhibition of the cAMP signaling pathway and inhibition of c-Raf-1. Mol Pharmacol 2000;58:608–13. Radeff JM, Nagy Z, Stern PH. Involvement of PKC-beta in PTH, TNF-alpha, and IL-1 beta effects on IL-6 promoter in osteoblastic cells and on PTH-stimulated bone resorption. Exp Cell Res 2001; 268:179–88.

[37] Reid IR, Lowe C, Cornish J, Gray DH, Skinner SJ. Adenylate cyclase blockers dissociate PTH-stimulated bone resorption from cAMP production. Am J Physiol 1990;258:E708–14. [38] Rubin MR, Bilezikian JP. The anabolic effects of parathyroid hormone therapy. Clin Geriatr Med 2003;19:415–32. [39] Sanders JL, Stern PH. Expression and phorbol ester-induced downregulation of protein kinase C isozymes in osteoblasts. J Bone Miner Res 1996;11:1862–72. [40] Schluter KD, Hellstern H, Wingender E, Mayer H. The central part of parathyroid hormone stimulates thymidine incorporation of chondrocytes. J Biol Chem 1989;264:11087–92. [41] Schmitt JM, Stork PJ. Cyclic AMP-mediated inhibition of cell growth requires the small G protein Rap1. Mol Cell Biol 2001;21:3671–83. [42] Schwartz Z, Semba S, Graves D, Dean DD, Sylvia VL, Boyan BD. Rapid and long-term effects of PTH(1–34) on growth plate chondrocytes are mediated through two different pathways in a cell-maturation-dependent manner. Bone 1997;21:249–59. [43] Shukunami C, Shigeno C, Atsumi T, Ishizeki K, Suzuki F, Hiraki Y. Chondrogenic differentiation of clonal mouse embryonic cell line ATDC5 in vitro: differentiation-dependent gene expression of parathyroid hormone (PTH)/PTH-related peptide receptor. J Cell Biol 1996;133:457–68. [44] Sprague SM, Popovtzer MM, Dranitzki-Elhalel M, Wald H. Parathyroid hormone-induced calcium efflux from cultured bone is mediated by protein kinase C translocation. Am J Physiol 1996;271:F1139–46. [45] Swarthout JT, Doggett TA, Lemker JL, Partridge NC. Stimulation of extracellular signal-regulated kinases and proliferation in rat osteoblastic cells by parathyroid hormone is protein kinase C-dependent. J Biol Chem 2001;276:7586–92. [46] Takasu H, Gardella TJ, Luck MD, Potts Jr JT, Bringhurst FR. Aminoterminal modifications of human parathyroid hormone (PTH) selectively alter phospholipase C signaling via the type 1 PTH receptor: implications for design of signal-specific PTH ligands. Biochemistry 1999; 38: 13453–60. [47] Takasu H, Guo J, Bringhurst FR. Dual signaling and ligand selectivity of the human PTH/PTHrP receptor. J Bone Miner Res 1999;14:11–20. [48] Tetradis S, Bezouglaia O, Tsingotjidou A. Parathyroid hormone induces expression of the nuclear orphan receptor Nurr1 in bone cells. Endocrinology 2001;142:663–70. [49] Whitfield JF, Isaacs RJ, Chakravarthy B, Maclean S, Morley P, Willick G, et al. Stimulation of protein kinase C activity in cells expressing human parathyroid hormone receptors by C- and N-terminally truncated fragments of parathyroid hormone 1–34. J Bone Miner Res 2001; 16:441–7. [50] Wu S, Pirola CJ, Green J, Yamaguchi DT, Okano K, Jueppner H, et al. Effects of N-terminal, midregion, and C-terminal parathyroid hormonerelated peptides on adenosine 3′,5′-monophosphate and cytoplasmic free calcium in rat aortic smooth muscle cells and UMR-106 osteoblast-like cells. Endocrinology 1993;133:2437–44. [51] Wu Y, Kumar R. Parathyroid hormone regulates transforming growth factor beta1 and beta2 synthesis in osteoblasts via divergent signaling pathways. J Bone Miner Res 2000;15:879–84. [52] Yamaguchi DT, Hahn TJ, Iida-Klein A, Kleeman CR, Muallem S. Parathyroid hormone-activated calcium channels in an osteoblast-like clonal osteosarcoma cell line. cAMP-dependent and cAMP-independent calcium channels. J Biol Chem 1987;262:7711–8.