Deletion of Orai1 Alters Expression of Multiple Genes during Osteoclast and Osteoblast Maturation

Summary

Store-operated Ca²⁺ entry (SOCE) is a major Ca²⁺ influx pathway in most non-excitable cell types and Orai1 was recently identified as an essential pore-subunit of SOCE channels. Here we investigate the physiological role of Orai1 in bone homeostasis using Orai1-deficient mice (Orai1^−/−). Orai1^−/− mice developed osteopenia with decreased bone mineral density and trabecular bone volume. To identify the nature and origin of the bone defect, bone-resorbing osteoclasts and bone-forming osteoblasts from Orai1^−/− mice were examined. Orai1-mediated SOCE was completely abolished in Orai1^−/− osteoclast precursor cells and osteoclastogenesis in vitro from Orai1^−/− mice was impaired due to a defect in cell fusion of pre-osteoclasts. Also, resorption activity in vitro was comparable but the size of pits formed by Orai1^−/− osteoclasts was smaller. We next assessed the role of Orai1 in osteoblast differentiation and function by using a pre-osteoblast cell line, as well as primary osteoblasts from wild-type and Orai1^−/− mice. SOCE in MC3T3-E1 pre-osteoblastic cells was inactivated by lentiviral overexpression of a pore-dead Orai1 mutant. Lack of SOCE in MC3T3-E1 had no effect on alkaline phosphatase staining and expression but substantially inhibited mineralized nodule formation. Consistent with this finding, Orai1-mediated SOCE was markedly reduced in Orai1^−/− osteoblast precursor cells and osteoblastogenesis in vitro from Orai1^−/− stromal cells showed impaired mineral deposition but no change in differentiation. This indicates that Orai1 is involved in the function but not in the differentiation of osteoblasts. Together, these results suggest that Orai1 plays a critical role in bone homeostasis by regulating both osteoblasts and osteoclasts.

Bone is a highly dynamic tissue that undergoes constant remodeling throughout life. This bone homeostasis is maintained by balanced activities of bone-forming osteoblasts and bone-resorbing osteoclasts [1–3]. Bone resorption and formation are determined by proliferation and differentiation of progenitors of the two bone cells. Osteoclasts are differentiated from hematopoietic cells within the monocyte/macrophage lineage. Osteoblasts originate from mesenchymal stem cells. Although these two bone cells orginate from different progenitors, they interact to regulate differentiation and function of one another. Osteoblasts express two cytokines, macrophage colony-stiumlating factor (M-CSF) and receptor activator of NFκB ligand (RANKL, also known as TRANCE/OPGL/ODF), both of which are essential for osteoclastogenesis [1;3]. RANKL from osteoblasts drives osteoclastogenesis by providing an essential signal to osteoclast progenitors through the membrane-anchored receptor RANK in osteoclasts. Osteoblasts also synthesize and secrete osteoprotegerin (OPG), a decoy receptor of RANKL, which blocks the interaction between RANKL and RANK [4;5]. Many bone diseases such as postmenopausal osteoporosis, metastatic osteolytic lesions and bone destruction in rheumatoid arthritis are caused by excessive osteoclastogenesis [6–9], whereas senile osteoporosis mainly results from defects in osteoblastogenesis [10;11]. Therefore, it is critical to define molecular mechanisms underlying differentiation and activation of these cells to develop new therapeutic approaches for the diseases.

A number of signaling pathways are involved in the differentiation of these two major bone cell types. For osteoclast differentiation, binding of RANKL to the RANK receptor leads to recruitment of TNF receptor-associated factor 6 followed by activation of various downstream signaling targets. These include NFκB, MAP kinases (Erk1/2, p38, and c-Jun N-terminal kinase), and c-Fos [3;12]. For osteoblast differentiation, mutiple transcription factors including Runx2, Osterix, and several members of activation protein-1 (AP-1) group of transcription factors such as JunB, Fra-1, and ΔFosB [13–16] were shown to play a pivotal role.

A rise in intracellular Ca²⁺ levels serves as a versatile signal in regulating multiple cellular functions including exocytosis, proliferation, differentiation, growth, and cell death [17;18]. It has long been known that Ca²⁺ signaling plays a critical role in the functions of bone cells [19–25]. Importantly, Takayanagi et al. [21] showed that RANKL evokes Ca²⁺ oscillations in pre-osteoclasts. These Ca²⁺ oscillations result in robust induction of nuclear factor-activated T cells c1 (NFATc1), a Ca²⁺/calcineurin-dependent master regulator of osteoclastogenesis. Extracellular stimuli such as neurotransmitters or hormones activate cell membrane receptors coupled to phospholipase C, resulting in intracellular Ca²⁺ release from the endoplasmic reticulum (ER) followed by extracellular Ca²⁺ influx through store-operated Ca²⁺ entry (SOCE) [26]. SOCE not only refills the depleted ER Ca²⁺stores but also provides a direct Ca²⁺ signal to activate downstream responses [27]. SOCE channels are comprised of Orai subunits, primarily Orai1. Either deletion of the Orai1 gene or overexpression of a pore-dead Orai1 mutant (E106Q) markedly reduces SOCE [28;29]. Clinically, T cells from patients with a severe combined immune deficiency (SCID) syndrome showed lack of SOCE, which is attributed to mutations in the Orai1 gene [30]. We previously demonstrated by a gene silencing approach that Orai1-mediated SOCE plays a critical role in osteoclast differentiation from RAW264.7 cells [31]. In addition, several studies reported the presence of SOCE in osteoblasts [32;33]. Labelle et al. [33] showed that SOCE is involved in osteoblast proliferation, and Robinson et al. [34] recently reported that pharmacological inhibition of SOCE impairs differentiation of human osteoblastic cells. However, the role of SOCE in differentiation and function of primary osteoblasts and osteoclasts lacking an Orai1 gene, and in regulation of the genes associated with their differentiation and function, has not been investigated.

Here we have investigated the physiological role of Orai1 in the differentiation and function of the two major bone cell types by using Orai1^−/− mice. We observed that the ablation of Orai1 in mice results in osteopenia with decreased bone mineral density and trabecular bone volume. This phenotype cannot be due to increased activity of osteoclasts, because we found that osteoclastogenesis in vitro is actually impaired in cells from Orai1^−/− mice. On the other hand, although osteoblast differentiation is not altered, mineralization is substantially impaired in Orai1^−/− osteoblasts due to a defect in expression of collagen type 1. In addition, the regulated expression of a number of key genes involved in the maturation of these two major bone cell types is altered in the absence of Orai1. Thus, Orai1 appears to play a critical role in bone remodeling by regulating intracellular Ca²⁺ signaling that leads to specific gene expression in osteoblasts and osteoclasts.

Materials and Methods

Reagents and antibodies

M-CSF and mouse soluble RANKL were obtained from R & D Systems (Minneapolis, MN, USA) and PeproTech EC (London, UK), respectively. Anti-ATP6v0d2 antibody (ab87059) was purchased from Abcam (Cambridge, MA, USA).

Mice

Orai1^−/− mice were obtained from Dr. Jean-Pierre Kinet (Harvard Medical School) and described previously [35]. Orai1^−/− mice in C57BL/6 background were then outbred for at least two generations with ICR mice in order to improve the survival of knockout animals. All mice used in these experiments were 6–13 weeks old and genotyping was performed on genomic DNA extracted from tails. Mice were housed in the animal facility of the National Institute of Environmental Health Sciences (NIEHS) according to the Association for the Assessment and Accreditation of Laboratory Animal Care Guidelines. All studies were approved by the Animal Care and Use Committee at the NIEHS.

Histology and microcomputed tomography (μCT)

Long bones were fixed, decalcified, and paraffin-embedded and stained with either H&E or for TRAP. μCT (μCT40, Scanco medical, Bassersdorf, Switzerland) was performed on femurs (15 μm isotropic voxel resolution). Cortical analysis was completed on an area 0.5 mm thick that was 5.25 mm proximal to the distal growth plate. For trabecular analysis, a three-dimensional cylindrical volume 0.750 mm in the z direction (axial) was selected and 0.98 mm in diameter that was 0.750 mm proximal from the distal growth plate was selected.

Bone mineral density

Bone mineral density of femurs from 6–13 week-old mice was determined using a Lunar PIXImus II densitometer (GE Healthcare, Piscataway, NJ, USA). The PIXImus was calibrated using a phantom of known BMD before use.

Cell culture

MC3T3-E1 cells (subclone 14) were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA) and were maintained in α-MEM (no ascorbic acid, Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (Invitrogen, Carlsbad, CA, USA), 100 U/ml penicillin G and 100 μg/ml streptomycin (Invitrogen, Carlsbad, CA, USA) at 37 °C with 5% CO₂. Osteoblastic differentiation of MC3T3-E1 cells was induced by 50 μg/ml ascorbic acid (Sigma, St. Louis, MO, USA) and 10 mM β-glycerophosphate (Sigma, St. Louis, MO, USA).

Lentiviral transfection

Recombinant lentiviruses were produced as previously described [31]. Briefly, either pCDH511B-1 pCDH-CMV-MCS-EF1-copGFP (System Biosciences, CA, USA) or pCDH-CMV-human Orai1 E106Q-EGFP-EF1-puro lentiviral plasmids were co-transfected with psPAX2 and pMD2.G into HEK293T/17 cells (ATCC, VA, USA) using LipofectAMINE Plus (Invitrogen, Carlsbad, CA, USA). The medium was replaced after 24 h, and viral supernatants were harvested 2 d post-transfection and stored at −80 °C until use. MC3T3-E1 cells were plated at a density of 200,000 cells/well of a 6-well plate and infected with 120 MOI of viral supernatant containing 5 μg/ml polybrene (Sigma, St. Louis, MO, USA). The media was replaced the following day and cultured for further experiments.

Osteoclast differentiation in vitro

Bone marrow cells were flushed out from femurs and tibia and cultured in α-MEM containing 10% FBS and M-CSF (30 ng/ml) overnight. Nonadherent cells were harvested and cultured in the presence of M-CSF for 3 d. After 3 d, adherent cells were used as bone marrow-derived macrophages (BMMs). The BMMs were plated on a 48-well plate at a density of 15,000 cells/well and cultured in the presence of 100 ng/ml RANKL and M-CSF (30 ng/ml) for 4 d. The medium was replaced every 2 d. The cultured cells were stained for tartrate-resistant acid phosphatase (TRAP) staining using an Acid Phosphatase Leukocyte kit (#387A, Sigma, St. Louis, MO USA). TRAP staining was performed in triplicate wells for each condition according to the manufacturer’s instruction. In order to measure secreted TRAP during osteoclastogenesis, cells were plated on a 96-well plate at a density of 6,000 cells/well and cultured in the presence or absence of RANKL for 4 d (media was replaced on day 2). On day 4, 25 μl of the media were mixed with 75 μl of TRAP solution (Acid Phosphatase Leukocyte kit #387, Sigma, St. Louis, MO, USA). The mixture was incubated at 37°C for 2 h and the absorbance was read at 540 nm (adapted from a protocol of TRAP staining kit, B-Bridge international, Inc., Cupertino, CA, USA).

In vitro resorption pit assay

BMMs were seeded on the 16-well BD BioCoat® Osteologic® calcium phosphate-coated slides (BD Biosciences, San Jose, CA, USA) at a density of 6,000 cells/well and osteoclastogenesis was induced by 100 ng/ml RANKL for 7 d. The medium was replaced every 2 d. Cells were detached with sodium hypochlorite solution (Sigma, St. Louis, MO, USA) and rinsed with distilled water. The plates were stained with 5% silver nitrate as instructed in Technical Bulletin #444 of BD Biosciences (http://www.bdbiosciences.com/external_files/dl/doc/tech_bulletin/live/web_enabled/TB444.pdf). The number of pits per well was determined in quadruplicate wells. For measurement of pit size, regions of interest (ROIs) for each pit were drawn and analyzed by MetaMorph imaging software (Molecular Devices, Inc., Sunnyvale, CA, USA).

Osteoblast differentiation in vitro

For osteoblast differentiation from mesenchymal stromal cells, bone marrow cells from long bones were cultured in α-MEM (no ascorbic acid, Invitrogen, Carlsbad, CA, USA) containing 10% FBS for 9 d [36]. Cells were replated and cultured until they became confluent. Osteoblast differentiation was induced in osteogenic medium (50 μg/ml ascorbic acid, 10 mM β-glycerophosphate, and 10 nM dexamethasone). After 7 d, cells were fixed with 4% paraformaldehyde and stained for alkaline phosphatase (ALP, Sigma, St. Louis, MO, USA). Bone nodule formation was identified with 2% Alizarin Red S (Sigma, St. Louis, MO, USA) staining on d 14. For quantification, Alizarin Red staining from the cell matrix was released with 10% cetylpyridinium chloride in 10 mM sodium phosphate for 15 m. The concentration of released Alizarin Red was determined by measuring the absorbance at 562 nm [37;38]. Mineral deposition was alternatively measured by von Kossa staining (5% silver nitrate, Sigma, St. Louis, MO, USA) on d 21.

Optical imaging of intracellular Ca²⁺ concentrations

Cells were seeded on coverslips and cultured overnight. BMMs and bone marrow stromal cells (BMSCs) were loaded with 5 μM Fura-2/AM Ca²⁺ indicator dyes (Invitrogen, Carlsbad, CA, USA) for 45 m at room temperature (MC3T3-E1 at 37 °C). Relative intracellular Ca²⁺ concentrations are reported as the ratio of fluorescence emitted when cells were excited alternately at 340 and 380 nm wavelengths of light. Peak ratio values of Fura-2 from BMMs, BMSCs, and MC3T3-E1 cells in response to thapsigargin (2 μM) were consistently between 1 and 2 while R_max, determined in situ, averaged about 5 for BMMs and MC3T3-E1, and 3.5 for BMSC, indicating that the indicator was less than half-saturated and thus yielded ratios that are proportional to Ca²⁺concentrations.

RNA isolation and quantitative real time-PCR (qRT-PCR)

BMMs from mouse long bones were plated at a density of 250,000 cells/well of a 6-well plate. After culture for 24 h, cells were treated with 100 ng/ml RANKL for 3 d. MC3T3-E1 cells were plated at a density of 200,000 cells/well of a 6-well plate and cultured in osteogenic medium for 7 d. BMSCs were plated at a density of 100,000 cells/well of a 6-well plate and cultured in osteogenic medium for 7 d. Total RNA was isolated and reverse transcribed to cDNA using an Omniscript RT kit (Qiagen, Valencia, CA, USA) according to manufacturer’s protocol. qRT-PCR was performed using a SYBR Green PCR Master Mix (Applied Biosystems, Carlsbad, CA, USA) with the ABI Prism 7000 Instrument (Applied Biosystems, Carlsbad, CA, USA). The mRNA levels were normalized to GAPDH. The sequence of primer pairs for murine ATP6v0d2, cathepsin K, DC-STAMP, GAPDH, NFATc1, and TRAP was previously described [31]. Sequences of other primer pairs are as follows.

• ALP sense 5′-ATCTTTGGTCTGGCTCCCATG-3′ and antisense 5′-TTTCCCGTTCACCGTCCAC-3′ [39];

• cFms sense 5′-CCTCCTCTGGTCCTGCTGCTGG-3′ and antisense 5′-GCTCACACATCGCAGGGTCACC-3′ [40];

• Col1α1 sense 5′-CCTGGTAAAGATGGTGCC-3′ and antisense 5′-CACCAGGTTCACCTTTCGCACC-3′ [41];

• Col1α2 sense 5′-TGGTCCTCTGGGCATCTCAGGC-3′ and antisense 5′-GGTGAACCTGCTGTTGCCCTCA-3′ [41];

• Fra-1 sense 5′-CCCCCGCAAGCTCAGGCACAGAC-3′ and antisense 5′-GCAGATGGGGCGATGGGCTTCC-3′ [16];

• NFATc2 sense 5′-TGGCCCGCGACATCTACCCT-3′ and antisense 5′-TGGTAGAAGGCGTGCGGCTT-3′ [42];

• OPG sense 5′-GCACATTTGGCCTCCTGCTAATTC-3′ and antisense 5′-ACTCTCGGCATTCACTTTGGTCCC-3′ [43];

• Orai1 5′-CTCAAAGCTTCCAGCCGGAC-3′ and antisense 5′-GGGAGCGGTAGAAGTGAACA-3′;

• Orai2 primer pairs were purchased from SABiosciences (PPM39942B)

• Osteocalcin sense 5′-GCAATAAGGTAGTGAACAGACTCC-3′ and antisense 5′-GTTTGTAGGCGGTCTTCAAGC-3′ [39];

• Osterix sense 5′-CTGGGGAAAGGAGGCACAAAGAAG-3′ and antisense 5′-GGGTTAAGGGGAGCAAAGTCAGAT-3′ [43];

• RANK sense 5′-TCATCTCTGTGGTAGTAGTGGCTG-3′ and antisense 5′-TTAGGAGCAGTGAACCAGTCGAAG-3′ [43];

• RANKL sense 5′-CAGCCATTTGCACACCTCACCATC-3′ and antisense 5′-TTTCGTGCTCCCTCCTTTCATCAG-3′ [43];

• Runx2 sense 5′-GAACCAAGAAGGCACAGACA-3′ and antisense 5′-AACTGCCTGGGGTCTGAAAA-3′ [43].

Western blot analysis

Cells were plated on a 6-well plate at the same density used in RNA isolation. Cells were lysed in RIPA buffer and protein concentrations were measured using a Bio-Rad Dc protein assay kit (Bio-Rad, Hercules, CA, USA). Thirty μg of proteins were separated on 4 – 20% gradient SDS-PAGE (Bio-Rad, Hercules, CA, USA) and transferred to a PVDF membrane. The membrane was incubated with primary antibodies at 4 °C overnight and incubated with a HRP-conjugated secondary antibody for 45 m, and developed using an enhanced chemiluminescence system.

CFU-F and CFU-ALP assay

Bone marrow cells were plated at a density of 1 × 10⁷ cells/well in a 6-well plate and cultured for 14 d in α-MEM (no ascorbic acid, Invitrogen, Carlsbad, CA, USA ) supplemented with 10% FBS. Cells were fixed with 4% formaldehyde and stained with H&E. Colonies containing more than 30 or more cells were counted as CFU-F. For CFU-ALP, colonies were further cultured in osteogenic medium for 7 d and subjected to ALP staining.

Cell viability

Cell viability was determined using a cell proliferation reagent WST-1 for osteoclasts (Roche Applied Science, Bavaria, Germany) and a MTT kit for osteoblasts (Cayman Chemical Company, Ann Arbor, MI, USA) according to manufacturer’s protocol.

Statistical analyses

Statistical significance was determined by use of ANOVA or Student’s t test. All statistical analyses were carried out using GraphPad Prism Software (GraphPad, La Jolla, CA, USA). P values less than 0.05 were considered statistically significant.

RESULTS

Orai1^−/− mice are osteopenic

To investigate the possible role of Orai1 in skeletal formation, we first analyzed the bone characteristics of Ora1^−/− mice. Section of femurs from Orai1^−/− mice stained with H&E revealed reduced number of trabeculae compared to wild-type (WT) littermates (Fig. 1A). A Lunar PIXImus densitometer demonstrated significantly lower bone mineral density (BMD) of femurs from both male and female Orai1^−/− mice (Fig. 1B). Consistently, a microcomputed tomography (μCT) study revealed substantial decrease in bone mass of Orai1^−/− femurs [Fig. 1C, axial (left panel) and coronal section (right panel)]. The number and thickness of trabeculae were significantly reduced whereas trabecular separation was increased in Orai1^−/− femurs compared to that of WT littermates (Fig. 1D). In addition, μCT scan exhibited a consistent slight decrease in cortical bone volume over total bone volume in Orai1^−/− femurs compared to WT littermates, but it was not statistically significant (data not shown). These observations indicate that deletion of Orai1 gene in mice results in osteopenia with decreased bone mineral density and trabecular bone volume. A similar conclusion was reached by Robinson et al. [34] with a different Orai1^−/− mouse model, although the functions of isolated osteoclasts and osteoblasts from these mice were not investigated.

Figure 1.

Ablation of Orai1 gene in mice results in osteopenia. (A) Histological sections of the femurs from wild-type (WT) and Orai1−/− mice (7-week females). The sections were stained with haematoxylin and eosin (H&E). The Orai1−/− femurs show reduced volume of trabeculae (magnification, x4). (B) Bone mineral density (BMD) of the femurs of WT and Orai1−/− mice was measured using a Lunar PIXImus II densitometer. The BMD from both males (6–13 weeks) and females (7–12 weeks) is significantly reduced in Orai1−/− mice (*** p < 0.001). For males, both WT and Orai1−/− groups contained mice of the following ages: one 6 week, two 7 week, two 8 week, two 9 week, and one each 10, 11, 12, 13 week. For females, both groups contained: one 7 week, one 8 week, one 10 week, two 11 week and one 12 week. (C) Microcomputed tomography of the femurs from WT and Orai1−/− mice (8-week males). Both axial (left panel) and coronal section (right panel) revealed lower bone mass in the femur of Orai1−/− mice compared to WT littermates. (D) Three-dimensional trabecular structural parameters in the distal femur obtained from data illustrated by the example in Panel C. Data represent means ± SE. n =3 (* p < 0.05). WT group included 8 week male, 7 week male, and 7 week female. Mutant group included 8 week male, and two 7 week females.

We considered that the osteopenia in Orai1^−/− mice could result from either increased activity of bone-resorbing osteoclasts or decreased mineralization from osteoblasts. We first investigated the number and function of osteoclasts. Bone sections of femurs stained with tartrate-resistant acid phosphatase (TRAP, an osteoclast biomarker) showed that the number of TRAP-positive (TRAP⁺) cells was comparable in WT and Orai1^−/− mice (Fig. 2A). Subsequently we determined whether deletion of Orai1 affects osteoclastogenesis in vitro using BMMs isolated from long bones of Orai1^−/− mice. We confirmed deletion of Orai1 gene in Orai1^−/− BMMs by quantitative real-time PCR (qRT-PCR) analysis (Supplemental Figure 1A). We have so far been unable to reliably detect murine Orai1 protein by Western blot analysis. However, we found by Ca²⁺ imaging analysis that SOCE was completely abolished in Orai1^−/− BMMs (Supplemental Figure 1B), consistent with a lack of functional Orai1 protein. This also indicates that other Orai forms (Orai2 and 3) do not contribute significantly to SOCE in these cells.

Figure 2.

Deletion of Orai1 leads to defective osteoclastogenesis in vitro. (A) TRAP staining of the femurs from WT and Orai1^−/− mice (7-week old) showed no significant difference in the number of TRAP-positive osteoclasts. (B) Bone marrow derived macrophages/monocytes (BMMs) isolated from long bones (femurs and tibias) were differentiated into mature multinucleated osteoclasts on plastic dishes in the presence of M-CSF and RANKL for 4 d and stained for TRAP, an osteoclast marker. Note that the size of osteoclasts generated from Orai1^−/−BMMs is substantially smaller than that of WT BMMs. (C) TRAP secreted in the media during osteoclastogenesis in vitro was measured after 4 d of differentiation. Data are representative of three independent experiments with similar results. Statistical significance was evaluated by an ANOVA followed by Newman-Keuls Multiple Comparison Test. Error bars denote standard error between samples performed in triplicate (*** p < 0.001). (D) WST-1 cell proliferation assay of osteoclasts cultured with or without RANKL for 4 d. Data are representative of two independent experiments with similar results. Error bars denote standard error between samples performed in triplicate (*<0.05, ** p < 0.01, n.s.: non-significant). (E) BMMs were grown on the Ca²⁺-phosphate coated plate in the presence of M-CSF and RANKL for 7 d. The medium was replaced every 2 d. Cells were detached, stained with 5% silver nitrate, and photographed under a light microscope.

Osteoclastogenesis in vitro of Orai1^−/− BMMs is impaired

We next investigated the differentiation potential of BMMs via an in vitro osteoclastogenesis assay. BMMs were cultured in the presence of M-CSF and RANKL for 4 d. The differentiated WT osteoclasts were stained for TRAP (stained in dark purple, Fig. 2B). Osteoclastogenesis in vitro was impaired in BMMs from Orai1^−/− mice (Fig. 2B). TRAP⁺-multinucleated cells (MNCs) were counted under a light microscope. The number of TRAP⁺ MNCs per well was significantly smaller in BMMs from Orai1^−/− mice compared to WT controls (WT 85.3 ± 10.6, KO 21.0 ± 2.6, *** p<0.001). In addition, the size of Orai1^−/− osteoclasts is markedly smaller than WT osteoclasts (data not shown), suggesting defective multinucleation of Orai1^−/− BMMs during osteoclastogenesis. We also measured TRAP proteins secreted from osteoclasts into the media during osteoclastogenesis. RANKL highly induced the secretion of TRAP from WT osteoclasts (Fig. 2C). Consistent with the defective osteoclast differentiation as shown in Fig. 2B, the amount of secreted TRAP during RANKL-induced osteoclastogenesis was dramatically decreased from Orai1^−/− osteoclasts (Fig. 2C). We next examined whether Orai1 deficiency affects proliferation or survival of osteoclasts using a WST-1 assay. BMMs were cultured with either M-CSF only or combination of M-CSF and RANKL for 4 d. We found that treatment of RANKL reduced the proliferation of both WT and Orai1^−/− BMMs, implying that RANKL induces differentiation rather than proliferation of BMMs at 4 d (Fig. 2D). However, the slower rate of proliferation during RANKL-induced osteoclastogenesis was not significantly different between WT and Orai1^−/− osteoclasts, indicating that the inhibition of osteoclast formation from Orai1^−/− BMMs is not caused by a difference in cell proliferation or survival (Fig. 2D). We also examined the resorptive activity of Orai1^−/− BMMs by using Ca²⁺-phosphate coated plates (Fig. 2E). The total area of pits (pixels per plate) formed by Orai1^−/− osteoclasts was comparable to that for WT osteoclasts (WT 136,061 ± 7,237, KO 136,489 ± 17,705) whereas the average size of pits (pixels per pit) was significantly smaller for Orai1^−/− osteoclasts (WT 241.1 ± 13.6, KO 173.0 ± 10.9, ** p<0.01). These observations suggest that once osteoclasts were formed from Orai1^−/− BMMs, the rate of resorption is similar to WT osteoclasts. However, it may indicate defects in fusion of pre-osteoclasts lacking Orai1-mediated SOCE.

Changes in osteoclastic genes in Orai1^−/− osteoclasts

To determine possible molecular players that lead to the defect in osteoclastogenesis, we assessed the expression of osteoclast-specific markers in Orai1^−/− osteoclasts by qRT-PCR. RANKL-mediated induction of TRAP, an early osteoclast marker, was significantly reduced in Orai1^−/− osteoclasts (Fig. 3A), consistent with previous observations of defective formation of TRAP⁺-MNC and defective secretion of TRAP from Orai1^−/− BMMs (Fig. 2B and 2C, respectively). Induction of cathepsin K, a late osteoclast marker, was also markedly decreased in Orai1^−/− osteoclasts (Fig. 3A). Two osteoclastic fusion genes, d2 isoform of vacuolar ATPase Vo domain (ATP6v0d2) and dendritic cell-specific transmembrane protein (DC-STAMP), were investigated. Both genes are highly induced during RANKL-mediated osteoclastogenesis and are essential for cell fusion of pre-osteoclasts [44;45]. As shown in Fig. 3A, RANKL-mediated induction of the two osteoclastic fusion genes was markedly abrogated in Orai1^−/− osteoclasts, reflecting the observed smaller size of TRAP⁺ MNCs and pits formed by Orai1^−/− osteoclasts. Western blot analyses confirmed that the induction of ATP6v0d2 proteins in response to RANKL was substantially reduced in Orai1^−/− osteoclasts (Fig. 3B). These observations are consistent with our previous study with RAW264.7 cells, which showed that Orai1-mediated Ca²⁺ entry plays a critical role in the cell-cell fusion of pre-osteoclasts [31]. qRT-PCR analysis revealed that expression levels of both RANK and M-CSF receptors (cFms) in Orai1^−/− osteoclasts were comparable to that of WT osteoclasts, suggesting that responsiveness of pre-osteoclasts to the two major cytokines is not altered in Orai1^−/− osteoclasts (Fig. 3A). However, expression of cFms was significantly elevated in unstimulated Orai1^−/− cells, such that RANKL caused a decrease in expression. The basis for this unexpected result is not known. Also unexpectedly, and in contrast to our earlier finding with RAW264.7 cells [31], the induction of NFATc1 message (Fig. 3A) as well as protein (not shown) in response to RANKL was not suppressed in Orai1^−/− osteoclasts.

Figure 3.

Changes in osteoclastic genes in Orai1^−/− osteoclasts. (A) BMMs were cultured in the presence or absence of RANKL for 3 d. Total RNA was isolated and qRT-PCR analysis of osteoclast differentiation markers was performed. The relative expression levels were normalized to GAPDH. Note that whereas the expression of NFATc1 is not changed, the induction of two cell fusion factors, ATP6v0d2 and DC-STAMP, is significantly suppressed in Orai1^−/− osteoclasts. Statistical significance was evaluated by an ANOVA followed by Newman-Keuls Multiple Comparison Test. Data are means ± SE of triplicate samples from three independent experiments with similar results (**P < 0.01, ***P < 0.001, n.s.: non-significant). (B) Western blot analysis of ATP6v0d2. Cell lysates were harvested at 3 d of differentiation in the presence of RANKL. Data are representative of three independent experiments with similar results.

Orai1-mediated SOCE is critical for mineralization of osteoblasts

These findings provide new information on the role of Orai1 and by inference SOCE in the differentiation and function of bone-resorbing osteoclasts. However, our data also indicate that defects in osteoclastogenesis in vitro and resorption activity in Orai1^−/− osteoclasts are unlikely to explain the osteopenia of Orai1^−/− mice. Thus we next investigated the possibility that osteoblast function may be impaired in Orai1^−/− mice. In order to determine whether lack of Orai1-mediated SOCE affects differentiation and function of osteoblasts, we first used a murine pre-osteoblast cell line, MC3T3-E1. Cells were infected with lentivirus expressing a pore-dead Orai1 mutant (Orai1 E106Q) to completely suppress Orai1-mediated SOCE [28;29]. Ca²⁺ imaging analysis confirmed essentially complete inhibition of SOCE in MC3T3-E1 cells infected with Orai1 E106Q lentivirus compared to that of GFP control (Fig. 4A). Differentiation of these cells was induced by ascorbic acid and β-glycerophosphate for 7 d at which time they were stained for alkaline phosphatase (ALP), an early osteoblast differentiation marker. Intensity of ALP staining in osteoblasts infected with Orai1 E106Q lentivirus was comparable to that of cells infected with GFP lentivirus (Fig. 4B), suggesting that SOCE may not be involved in early differentiation of osteoblasts. In order to determine whether the lack of SOCE affects the activity of osteoblasts, extracellular mineral deposition was analyzed. Alizarin Red (AR) staining after 14 d of induction revealed a substantial decrease in mineralized extracellular matrix formed by osteoblasts expressing Orai1 E106Q compared to control (Fig. 4C, left panel). Subsequently, the AR staining was released by 10 % cetylpyridinium chloride and quantified by spectrophotometry. Consistent with the AR staining, the amount of released dye was significantly less in osteoblasts infected with Orai1 E106Q lentivirus compared to control, indicating impaired mineral deposition by osteoblasts when SOCE is abolished (Fig. 4C, right panel). The infected cells were further cultured for 21 d and bone nodule formation was determined by von Kossa staining. Consistently, bone nodule formation by osteoblasts infected with Orai1 E106Q lentivirus was markedly reduced compared to GFP control (Fig. 4D). Taken together, these results indicate that Orai1-mediated SOCE plays a critical role in the mineral deposition of osteoblasts. To determine whether the impaired function of osteoblasts infected with Orai1 E106Q lentivirus is not due to a reduction in proliferative potential, cell proliferation (or survival) was measured at 7 d (Fig. 4E, left panel) and 14 d (Fig. 4E, right panel) of induction using a MTT assay. Osteogenic induction of MC3T3-E1 cells with ascorbic acid and β-glycerophosphate enhanced cell proliferation (or survival) at 7 d (Fig. 4E. left panel) and 14 d (Fig. 4E, right panel) compared to vehicle controls. However, the enhancement in cell proliferation at both 7 d and 14 d was similar in cells infected with either GFP or Orai1 E106Q lentivirus, suggesting the defective mineralization in Orai1 E106Q-infected osteoblasts was not caused by a difference in cell proliferation (or survival). In an attempt to identify possible mechanisms underlying the reduced bone-forming activity of Orai1 E106Q-infected osteoblasts, we analyzed the expression levels of osteoblast marker genes by qRT-PCR after 7 d of osteogenic induction. As expected, we found that osteogenic medium significantly induced the expression of most osteoblastic marker genes. Notably, the induced expression of bone matrix markers, osteocalcin, collagen type 1 alpha1 and 2 (col1α1 and col1α2), was significantly reduced in osteoblasts infected with Orai1 E106Q lentivirus, whereas that of ALP and runx2 (early differentiation markers) in osteoblasts lacking SOCE was comparable to GFP controls (Fig. 5). Interestingly, the expression of Fra-1 was markedly reduced in osteoblasts lacking SOCE (Fig. 5). Fra-1 has been shown to be critical for bone matrix formation [16]. We also examined the expression levels of NFATc1 and c2, Ca²⁺-sensitive transcript factors known to be involved in osteoblast differentiation [46]. However, there was no difference in the expression levels of these factors (Fig. 5). In addition, expression of osterix, an osteoblastic transcription factor that is known to interact with NFATc1 [46], was not changed in osteoblasts derived from MC3T3-E1 lacking SOCE (Fig. 5). These results indicate that early differentiation is not altered but bone matrix formation is impaired in osteoblasts expressing the Orai1 pore-dead mutant.

Figure 4.

Inhibition of Orai1-mediatede SOCE leads to defective osteoblast function. (A) MC3T3-E1 cells were transduced with lentiviruses that overexpress either GFP or a pore- dead mutant Orai1 (Orai1 E106Q-GFP). Optical imaging of intracellular Ca²⁺ concentrations showed that SOCE is completely abolished in cells expressing the Orai1 E106Q mutant compared to the GFP control. (B) Infected MC3T3-E1 cells were differentiated in the presence of ascorbic acid (AA) and β-glycerophosphate (GP) for 7 d and stained for alkaline phosphatase (ALP). ALP activity is unchanged in cells infected with the Orai1 E106Q mutant compared to control. (C) MC3T3-E1 cells infected with either GFP or Orai1 E106Q-GFP lentivirus were differentiated for 14 d and deposition of extracellular matrix was analyzed by alizarin red staining (shown in dark red). The stained matrix was photographed (left panel) and the alizarin red staining was quantified as described in Materials and Methods (right panel). Data are representative of three independent experiments with similar results (**P < 0.01, ***P < 0.001). (D) Mineralized nodule formation was assessed by von Kossa staining after 21 d of induction. Data are representative of three independent experiments with similar results. (E) MTT cell proliferation assay of osteoblasts cultured for 7 d (left panel) and 14 d (right panel). Data are representative of three independent experiments with similar results. Error bars denote standard error between samples performed in triplicate (*** p < 0.001, n.s.: non-significant).

Figure 5.

Transcriptional regulation is altered in osteoblasts lacking SOCE. MC3T3-E1 cells infected with either GFP or Orai1 E106Q-GFP lentivirus were cultured in the presence of ascorbic acid (AA) and β-glycerophosphate (GP) for 7 d. Total RNA was isolated and qRT-PCR analysis of osteoblast differentiation markers was performed. Expression levels were normalized to GAPDH. Data are representative of three independent experiments with similar results. Error bars denote standard error between samples performed in triplicate (* < 0.05, ** < 0.01, *** p < 0.001, n.s.: non-significant).

Defective mineralization of Orai1^−/− osteoblasts

To investigate the role of Orai1 in the differentiation and function of primary osteoblasts derived from bone marrow stromal cells (BMSCs), BMSCs were obtained from 9 d culture of whole bone marrow cells in the absence of osteogenic medium [36]. qRT-PCR analysis confirmed ablation of Orai1 gene in Orai1^−/− BMSC (Fig. 6A). To confirm a lack of SOCE in Orai1^−/− BMSC, Ca²⁺ imaging analysis was performed, which showed a substantial but not complete abolishment of SOCE in Orai1^−/− BMSCs compared to WT BMSCs (Fig. 6B). Low concentration of Gd³⁺ inhibited not only SOCE in WT BMSCs, but also residual SOCE in Orai1^−/− BMSCs, indicating contribution of other isoforms of Orai (Orai2 and/or Orai3) in the activation of SOCE in Orai1^−/−BMSCs (Fig. 6B). Leak control with no depletion of ER Ca²⁺ stores confirmed the presence of residual SOCE in Orai1^−/− stromal cells (Fig. 6B). Indeed, qRT-PCR showed substantial levels of Orai2 and 3 mRNA in Orai1^−/− BMSC (Orai2 in Fig. 7 and data not shown for Orai3). We next examined the osteoblastic differentiation potential of Orai1^−/− BMSCs in vitro. The BMSCs were cultured in osteogenic medium (50 μM ascorbic acid, 10 mM β-glycerophosphate, and 10 nM dexamethasone) for 7 d and subjected to ALP staining. Intensity of ALP staining of Orai1^−/− osteoblasts was similar to that of WT osteoblasts (Fig. 6C) which is consistent with the lack of difference in ALP staining of MC3T3-E1 cells infected with Orai1 E106Q lentivirus (Fig 4B). Cells were further cultured in osteogenic medium for 14 d and analyzed for the effect of Orai1 deficiency on mineral deposition by AR staining. We confirmed that osteogenic medium highly induced mineral deposition of WT stromal cells evidenced by strong AR staining (Fig. 6D, bottom panel) compared to vehicle controls (Fig. 6D, top panel). Importantly, the intensity of AR staining was significantly lower in Orai1^−/− osteoblasts (Fig. 6D). Spectrophotometric analysis of released dye from the AR-positive mineral deposition consistently showed a significant decrease in mineral deposition in Orai1^−/− osteoblasts compared to WT osteoblasts (Fig. 6E). In addition, von Kossa staining at 21 d of induction revealed substantial reduction in bone nodule formation of Orai1^−/− osteoblasts (Fig. 6F). Consistent with the observed defect in mineral deposition in MC3T3-E1 cells infected with a dominant-negative Orai1 mutant, these results indicate that Orai1-mediated SOCE also plays a critical role in mineral deposition of primary osteoblasts derived from BMSCs. We next performed a colony-forming unit fibroblast (CFU-F) assay to measure any difference in the number of bone marrow-derived mesenchymal progenitors in the mutant mice [47;48]. We observed no significant difference in the number of CFU-F between WT and KO mice (Fig. 6G, Supplemental Fig. 3A). Also the number of CFU-ALP was comparable between WT and Orai1^−/− mice (Supplemental Fig. 3B), suggesting the number of mesenchymal progenitors is similar in WT and mutant mice. To determine whether the loss of Orai1 affects the proliferative potential of primary osteoblasts, we measured cell proliferation (or survival) of BMSCs using a MTT assay. We found no difference in cell proliferation between WT and Orai1^−/− osteoblasts at both 7 d (Fig. 6H, left panel) and 21 d (Fig. 6H, right panel) of induction by osteogenic medium. To identify candidate molecules responsible for the observed defects in primary Orai1^−/− osteoblasts, the expression levels of key osteoblastic marker genes at 7 d of induction was investigated by qRT-PCR (Fig. 7). We confirmed the loss of Orai1 gene and unaffected expression of Orai2 in differentiated Orai1^−/− osteoblasts (Fig. 7). Consistent with the observation from MC3T3-E1 lacking SOCE, the expression levels of col1α2 and Fra-1 were decreased in Orai1^−/− osteoblasts compared to WT osteoblasts (Fig. 7). Both OPG and RANKL expression were also decreased in mutant osteoblasts. Consistent with MC3T3-E1 lacking SOCE, we found no difference in the expression levels of NFATc1 between WT and Orai1^−/− osteoblasts (Fig. 7). The expression of ALP and osteocalcin was increased in mutant osteoblasts (Fig. 7), likely as a compensatory action of cells against defective mineralization. However, there was no difference in runx2 expression between two groups (Fig. 7), again consistent with the data from MC3T3-E1 infected with Orai1 E106Q mutant virus as shown in Fig. 5. Since the effect of dexamethasone on the differentiation of osteoblasts is controversial [49–51], we cultured BMSCs in osteogenic medium without dexamethasone for 7 d, and ALP staining and qRT-PCR of the osteoblast marker genes were assessed (Supplemental Fig. 2). Although dexamethasone slightly suppressed the intensity of ALP staining (Supplemental Fig. 2A), changes in the expression levels of osteoblastic markers were consistent with those for cells cultured with dexamethasone (Supplemental Fig. 2B). The expression levels of Col1α2 and Fra-1 were significantly reduced in Orai1^−/− osteoblasts cultured in osteogenic medium without dexamethasone (Supplemental Fig. 2B). In addition, von Kossa staining of osteoblasts cultured in the absence of dexamethasone also showed decreased mineral deposition of Orai1^−/− osteoblasts (Supplemental Fig. 2C). Taken together, these observations demonstrate that Orai1 plays a pivotal role in the mineralization function of osteoblasts by regulating expression of bone matrix proteins.

Figure 6.

Mineralization is impaired in Orai1^−/− osteoblasts. (A) Total RNA was isolated from bone-marrow-derived stromal cells (BMSCs) and qRT-PCR analysis was performed to measure the levels of murine Orai1 mRNA. (B) Optical imaging of intracellular Ca²⁺ concentrations showed markedly reduced SOCE in Orai1^−/− BMSCs compared to that of WT. Leak control without ER store depletion showed further reduction of SOCE in Orai1^−/− BMSC, indicating involvement of other SOCE channels such as Orai2 and/or Orai3. (C) Osteoblast differentiation of BMSCs was induced by osteogenic medium (50 μg/ml ascorbic acid, 10 mM β-glycerophosphate, and 10 nM dexamethasone) and stained for ALP after 7 d of induction. (D) BMSCs were cultured in either vehicle (Veh) or osteogenic medium (OSM) for 14 d and deposition of extracellular matrix was analyzed by alizarin red staining (shown in dark red). The stained matrix was photographed and the alizarin red staining was quantified as described in Materials and Methods (E). Data are representative of three independent experiments with similar results. (F) Mineralized nodule formation was determined by von Kossa staining after 21 d of differentiation of BMSCs. Data are representative of three independent experiments with similar results. (G) CFU-F of BMSCs was measured. (H) MTT cell proliferation assay of osteoblasts at 7 d (left panel) and 21 d (right panel) of differentiation. Error bars denote standard error between samples performed in triplicate (*** p < 0.001, n.s.: non-significant).

Figure 7.

Deletion of Orai1 alters expression of multiple genes during osteoblast maturation. BMSCs were cultured in the presence of osteogenic medium for 7 d. Total RNA was isolated and qRT-PCR analysis of osteoblast differentiation markers was performed. Note that Orai1^−/− osteoblasts culture in osteogenic medium without dexamethasone showed similar transcriptional profile of the markers (Supplemental Fig. 2). Student’s t test was used for statistical analysis. Data are representative of two independent experiments with similar results. Error bars denote standard error between samples performed in triplicate (* < 0.05, ** < 0.01, *** p < 0.001, n.s.: non-significant).

DISCUSSION

Intracellular Ca²⁺ signaling has long been known to involve differentiation and function of the two major bone cell types [20;23–25]. Recent studies utilizing pharmacological inhibition of SOCE or gene silencing of Orai1 revealed an important role for SOCE in osteoclastogenesis [31;52]. We previously showed that knockdown of Orai1 in RAW264.7 cells impairs osteoclastogenesis by inhibiting transactivation of NFATc1, a Ca²⁺-dependent transcription factor [31]. In the present study, osteoclastogenesis in vitro is also impaired in Orai1^−/− BMMs. However, induction of NFATc1 by RANKL in Orai1^−/− osteoclasts was comparable to that of WT osteoclasts (Fig. 3A). Since TG-induced SOCE was completely abolished in Orai1^−/− BMMs, these results suggest that one or more Ca²⁺ channels other than store-operated channels are also likely involved in osteoclastogenesis induced by RANKL. Indeed, previous studies have suggested that a role for other Ca²⁺ influx channels in osteoclastogenesis [53–56].

The decrease in bone mass in Orai1^−/− mice cannot result from decreased differentiation of osteoclasts, which would result in an increase in bone mass due to a defect in bone resorption and so we investigated the possible role of the bone-forming osteoblasts in the osteopenic phenotype of Orai1^−/− mice. The presence of SOCE has previously been shown in primary and osteoblastic cell lines [32;33;57]. Robinson et al. [34] recently reported that a pharmacological inhibitor of SOCE reduced the differentiation and function of human osteoblastic cells. However, our results, utilizing both an osteoblastic cell line and Orai1-deficient primary osteoblast cells, indicate that Orai1-mediated SOCE is involved in the function but not in the differentiation of osteoblasts. This discrepancy may be due to the use of different species (human vs. mouse), or different strategies to interfere with SOCE (pharmacological inhibition vs. gene deletion). MC3T3-E1 cells expressing a pore-dead dominant-negative Orai1 mutant showed defects in mineralization despite the fact that osteoblast differentiation was not affected. Consistently, osteoblasts differentiated from Orai1^−/− BMSCs showed impaired mineralization accompanied by reduced expression of co1α2. Orai1 deficiency also did not affect osteoblast differentiation. Interestingly, in both MC3T3-E1 and primary osteoblasts, the expression of Fra-1 is reduced by inhibition of Orai1. Fra-1 is a component of the dimeric transcription factor, AP-1 and conditional knockout of Fra-1 (Fra-1^+/+) in mice results in low-turnover osteopenia [16]. Similar to Orai1^−/− mice, osteoclasts in vivo were normal but expression of colα2 was decreased in long bones of Fra-1^+/+ mice. This suggests that Orai1-mediated Ca²⁺ signaling in osteoblasts may regulate downstream targets including transcription factors. Further study is needed to identify those targets. However, we cannot rule out the possibility that Orai1-mediated SOCE may play a role in the secretion of collagen via exocytosis. Synaptotagmin (Syt) is a Ca²⁺- dependent regulator of exocytosis, and. Syt VII-deficient mice showed osteopenia with defects in bone resorption and formation [58]. Secretion of cathepsin K from osteoclasts and bone matrix proteins from osteoblasts was impaired in Syt VII-deficient mice. Orai1-mediated Ca²⁺ signaling may be involved in the activation of synaptotagmin to control exocytosis in bone cells. However, further work is needed to clarify the role of Orai1 in exocytosis of bone cells. Since other isoforms of Orai appear to be expressed in osteoblasts, and since some residual SOCE was seen in Orai1^−/− stromal cells (Fig. 7 and Fig. 6B, respectively), it is also possible that other isoforms of Orai (Orai2 and 3) and/or other Ca²⁺ entry channels may contribute to osteoblastogenesis. Baldi et al showed that TRPC3-like proteins contributed to SOCE in an osteoblastic cell [59]. Several studies have reported that L-type voltage-sensitive channels can regulate intracellular Ca²⁺ signaling in osteoblasts [23;60;61].

The results of this study strongly implicate Orai1-mediated Ca²⁺ entry as a major signaling mechanism regulating the differentiation and/or functions of the two bone cell types. Two independently generated Orai1^−/− mouse models result in small homozygous pups [35;62], a phenotype largely attributed to impaired skeletal muscle development [63]. The results of the current study, and those of Robinson et al. [34] suggest that effects of Orai1 deficiency on bone may contribute as well. It appears that Orai1^−/− mice develop low turnover osteopenia due to defects in osteoblasts and osteoclasts. It has been demonstrated that deletion of specific genes often results in low-turnover osteopenia due to impaired function of both osteoblasts and osteoclasts. Genetic ablation of Abl, CCR1, JunB, or Sh3bp2 in mice results in low-turnover osteopenia [13;64–66]. Similar to Orai1^−/− mice, Abl-deficient mice developed osteopenia due to defects in osteoblast maturation [64].

However, identifying the precise underlying molecular mechanisms will require further study. Several key questions remain. What are the in vivo stimuli for Orai1 activation in osteoclasts and osteoblasts? What are the immediate downstream targets of Orai1 activation that leads to specific gene expression in bone cells? Is either the expression or activation of Orai1 changed in pathological conditions of bone homeostasis including postmenopausal osteoporosis and rheumatoid arthritis? Do other isoforms of Orai (Orai2 and/or Orai3) contribute to bone remodeling? From the clinical perspective, do patients with the SCID syndrome in which Orai1 gene is mutated have a bone phenotype [30]? Although a defect in dental enamel calcification was reported in patients with Orai1 mutation [67], to our knowledge the skeletal phenotype of SCID patients has not yet been extensively investigated. In order to unravel the precise role of Orai1 in each bone cell type, lineage-specific conditional knockout animals will be necessary.

In summary, we have demonstrated an osteopenic phenotype of Orai1^−/− mice and we have delineated a physiological role of Orai1 that controls both osteoblasts and osteoclasts in this mouse model. We elucidated that SOCE is important for cell fusion of pre-osteoclasts and show that Orai1-mediated SOCE plays a crucial role in the mineralization of osteoblasts by regulating expression of bone matrix protein. The implications of these findings to the management of bone diseases must also await future research. As a key regulator of bone remodeling and mineralization, one or more of the Orai channels could be potential therapeutic targets for treatment of bone diseases. Additionally, blockers of store-operated channels are currently in development for use as immune suppressants (www.calcimedica.com). It will be important to be aware of the potential long-term effects of such agents on bone.