SEARCH

SEARCH BY CITATION

Keywords:

  • Osteopetrosis;
  • Animal Model;
  • Rodent;
  • Bone;
  • Osteoclast

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

A mouse founder with high bone mineral density and an osteopetrotic phenotype was identified in an N-ethyl-N-nitrosourea (ENU) screen. It was found to carry a dominant missense mutation in the Tcirg1 gene that encodes the a3 subunit of the vacuolar type H+-ATPase (V–ATPase), resulting in replacement of a highly conserved amino acid (R740S). The +/R740S mice have normal appearance, size, and weight but exhibit high bone density. Osteoblast parameters are unaffected in bones of +/R740S mice, whereas osteoclast number and marker expression are increased, concomitant with a decrease in the number of apoptotic osteoclasts. Consistent with reduced osteoclast apoptosis, expression of Rankl and Bcl2 is elevated, whereas Casp3 is reduced. Transmission electron microscopy revealed that unlike other known mutations in the a3 subunit of V–ATPase, polarization and ruffled border formation appear normal in +/R740S osteoclasts. However, V–ATPases from +/R740S osteoclast membranes have severely reduced proton transport, whereas ATP hydrolysis is not significantly affected. We show for the first time that a point mutation within the a3 subunit, R740S, which is dominant negative for proton pumping and bone resorption, also uncouples proton pumping from ATP hydrolysis but has no effect on ruffled border formation or polarization of osteoclasts. These results suggest that the V0 complex has proton-pumping-independent functions in mammalian cells. © 2011 American Society for Bone and Mineral Research.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

Healthy bone undergoes constant remodeling, with a regulated balance between formation and resorption. Bone formation is accomplished by osteoblasts that synthesize and mineralize the organic matrix, whereas osteoclasts are responsible for bone resorption. To resorb bone, osteoclasts become multinucleated, polarized, and create a sealed compartment called a resorption lacuna between their ruffled border membrane and the bone surface.1, 2 Vacuolar type H+-ATPases (V–ATPases) containing the a3 subunit localize to the ruffled border membrane inside the sealing zone and actively pump acid onto the bone surface.2 Pathologic conditions such as periodontal disease, inflammatory arthritis, and osteoporosis are characterized by excessive osteoclast activity. In contrast, impaired osteoclast activity or absence of osteoclasts leads to high-bone-mass conditions such as osteopetrosis.3–5

Osteopetrosis encompasses a diverse group of heritable conditions in which there is increased bone mass resulting from defects in the balance of osteoblast and osteoclast activity. Mild, intermediate, and severe forms of osteopetrosis occur in humans. Autosomal dominant osteopetrosis (ADO), the mildest form of osteopetrosis, is further subdivided into two groups. ADOI, caused by gain-of-function mutations in the low-density lipoprotein receptor–related protein 5 (LRP5) gene leading to activation of osteoblasts,6 and ADOII, caused by certain mutations to the chloride channel 7 (ClCN7) gene.7 Intermediate autosomal recessive osteopetrosis (IARO) is also caused by certain mutations in ClCN7, carbonic anhydrase II (CAII), and pleckstrin homology domain–containing family M (with RUN domain) member 1 deficiency (PLEKHM1) genes.8 Autosomal recessive osteopetrosis (ARO) is the most severe form of osteopetrosis with a life expectancy of 4 years without treatment (bone marrow transplant)9 and results from mutations within the following genes: TCIRG1, encoding the V–ATPase a3 subunit,3, 4 ClCN7,10, 11 osteopetrosis-associated transmembrane protein 1 (OSTM1),12 and receptor activator of nuclear factor-κB (RANK) and RANK ligand (RANKL).13, 14 Approximately 50% of patients with ARO have mutations in the TCIRG1 gene.4, 10, 15

V–ATPases are proton pumps composed of at least 14 subunits organized into two functional domains, a peripheral V1 domain containing 8 subunits and an integral V0 domain containing 5 subunits, including the a subunit. ATP hydrolysis occurs in the V1 domain, which drives proton transport through the V0 domain. The a subunit has four isoforms–a1, a2, a3, and a4–each encoded by separate genes. While a1, a2, and a3 are expressed ubiquitously, the a4 isoform is specifically expressed in kidney and inner ear.16 While ubiquitous, the expression of a3 is 100-fold greater in osteoclasts than in other cell types, and it is enriched at the osteoclast plasma membrane, where it is responsible for the extracellular acidification required for bone resorption.3–5, 17

While humans homozygous for mutations within TCIRG1 are severely osteopetrotic, their heterozygous parents have no detectable phenotype. Although initial studies pointed to the possibility that dominant mutations in TCIRG1 might underlie some cases of osteopetrosis, subsequent work demonstrated that these patients also had a genomic deletion in TCIRG1, indicating that these individuals are compound heterozygotes.18 Mice homozygous for a naturally occurring Tcirg1 truncation (oc/oc)19 and mice completely lacking Tcirg1 (Tcirg1–/–)20 also have severe osteopetrosis and die 3 to 5 weeks after birth, and similar to humans, their heterozygous counterparts have no reported phenotype.

In an N-ethyl-N-nitrosourea (ENU) mutagenesis screen for dominant mutations, we identified a mouse founder with high bone mineral density (BMD) and a missense mutation to Tcirg1 that results in the substitution of a highly conserved arginine by a serine at residue 740 (R740S) of a3. We report for the first time among human or mouse a3 mutations, a dominant-negative Tcirg1 mutation that almost completely abolishes proton pumping, resulting in osteopetrosis in heterozygous mice.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

Mice and ENU mutagenesis

C57BL/6J (B6) males, C3H/HeJ (C3) males and females, and FVB females were purchased from the Jackson Laboratories (Bar Harbor, ME, USA) at 6 to 8 weeks of age. Male C57BL/6J mice received three intraperitoneal injections of ENU 1 week apart at a dose of 85 mg/kg, as described previously.21 ENU-mutagenized males were bred to C3H/HeJ females, and the offspring were C3;B6 F1 hybrid pups known as generation 1 (G1). G1 mice were screened, bred to C3H/HeJ mice, and the G2 mice (C3;CgN2) produced were tested for heritability and used for genetic mapping. Lines were maintained by breeding to C3H/HeJ mice, producing G3 (C3;CgN3) and G4 (C3;CgN4) mice. To produce larger litters, G3 mice were bred to FVB females to produce FVB;C3 F1 (FVB;C3CgN3) mice. Mice carrying the R740S mutation bred onto the C3 or FVB backgrounds are referred to as +/R740S mice. All experimental procedures received approval from the local animal care committees and were conducted in accordance with the guidelines of the Canadian Council on Animal Care.

Genetic mapping

DNA was extracted from tail tissue followed by a genome scan using polymerase chain reaction (PCR) amplification of microsatellite markers using standard procedures, as described previously.22

In-life screening of mutants with BMD anomalies

Mice were screened at 8 and 10 weeks of age by dual-energy X-ray absorptiometry (DXA; PIXImus, Lunar Corp., Madison, WI, USA) to measure whole-body bone mineral content (BMC) and bone mineral density (BMD).

Histomorphometry

Briefly, mice (6 mice/genotype) were euthanized at 2 months of age after receiving intraperitoneal injections of 30 mg/kg of calcein at 7 and 3 days before euthanization to label actively mineralizing tissue. The humeri, femurs, tibias, lumbar vertebrae, and calvaria were excised, cleaned of soft tissue, fixed overnight in 3.7% paraformaldehyde, and rinsed in PBS, and undecalcified specimens were embedded in polymethyl methacrylate (PMMA). Then 5-µm sections were stained with tartrate-resistant acid phosphatase (TRACP), hematoxylin and eosin (H&E), and von Kossa to detect osteoclasts, osteoblasts, and mineralized and unmineralized tissue. Some 5-µm sections of PMMA-embedded tissue were left unstained to measure the distance between and the length of the two fluorochrome labels.

Micro–computed tomography (µCT) of femurs and vertebrae

The distal metaphysis of the left femurs and fourth lumbar vertebrae were scanned with a Skyscan 1172 µCT instrument (Skyscan, Kontich, Belgium) at the Université du Québec à Montréal. Two-dimensional (2D) images were used to generate 3D reconstructions of the bones from three +/R740S mice and three +/+ littermates at 2 months of age. Morphometric parameters were calculated with 3D Creator software supplied with the instrument.

Serum measurements

Serum was collected from 2-month-old mice (3 mice/genotype) and used to measure total calcium and alkaline phosphatase (IDEXX Laboratories, Burlington, Ontario, Canada).

Transmission electron microscope (TEM) imaging

Femurs from 2-month-old mice (3 mice/genotype) were fixed in 4% paraformaldehyde–1% glutaraldehyde in 0.1 M sodium cacodylate buffer for 4 hours at room temperature, stored overnight at 4oC, decalcified in 1.9% gluteraldehyde and 0.15 M EDTA for 1 week, postfixed in 1% osmium tetroxide and 1.5% potassium ferrocyanide in 0.1 M sodium cacodylate buffer, and then embedded in Spurr's resin and processed as described previously.23 Samples were viewed with a Hitachi H-7000 electron microscope (Toronto, ON, Canada) with a Hamamatsu ORCA-HR digital camera (Hammamitsu City, Japan) in the Microscope Imaging Laboratory (MIL) at the University of Toronto.

Terminal deoxynucleotidyl transferase–mediated deoxyuridine triphosphate–biotin nick end labeling (TUNEL) staining of sections of femur

Femurs from mice at 2 months of age (3 mice/genotype) were excised, cleaned of soft tissue, fixed overnight in 3.7% paraformaldehyde, rinsed in PBS, decalcified as earlier, embedded in paraffin, and sectioned. The 5-µm-thick sections were deparaffinised in xylenes, rehydrated in decreasing concentrations of ethanol, and used for colorimetric detection of apoptotic cells with the fragEL DNA Fragmentation Kit according to the manufacturer's directions (Calbiochem, EMD Biosciences, Burlington, ON, Canada).

Stromal cell cultures

Bone marrow stromal cells were collected from femurs of 2-month-old male mice and cultured for 21 days. The cultures then were stained for alkaline phosphatase and von Kossa to identify mature osteoblasts and mineralized matrix, as described previously.24, 25 For gene expression experiments, stromal cells were cultured for 8 days, and RNA was collected using TRIzol reagent (Invitrogen, Carlsbad, CA, USA). Three separate experiments were performed, each on pools of cells from three mice per genotype.

Conventional and real-time quantitative PCR

Total RNA was extracted from bone metaphyses using TRIzol reagent, and 3 µg was treated with DNase I (Invitrogen) at 1 unit of DNase I/µg of RNA and then reverse transcribed using Revert Aid H First Strand Kit (Fermentas, Burlington, ON, Canada). PCR reactions were performed in 50-µL volumes using HotStarTaq polymerase (Qiagen, Valencia, CA, USA). PCR products were visualized using GeneSnap software Version 4.00.00 (SynGene, Cambridge, UK). For quantitative reverse-transcriptase polymerase chain reaction (qRT-PCR) gene expression analysis, RNA was isolated as described earlier. Then 3 µg of RNA were treated with DNase I (Invitrogen), and cDNA was synthesized using Maxima Reverse Transcriptase (Fermentas) according to the manufacturer's instructions. qRT-PCR was performed using the Quanti-Tect SYBR Green PCR Kit (Qiagen) in an Mx3000P QPCR system (Stratagene, Mississauga, ON, Canada). Specific primer sequences used in the reactions are summarized in Supplemental Table S1. All data were normalized to L32. Three separate experiments were performed, each on pools of cells from three mice per genotype.

Isolation of osteoclast and kidney membranes

Membranes were prepared from +/R740S and +/+ osteoclasts and kidney as described previously.26, 27 Protein concentrations were measured by the bicinchoninic acid (BCA) protein assay (Pierce, Nepean, ON, Canada).

Proton transport assay

Proton transport was assayed as described previously26, 27 using osteoclast membranes (40 µg) and kidney membranes (20 µg). Proton transport was initiated by addition of ATP (1.5 mM) and monitored by measuring uptake of acridine orange with a dual-wavelength Hitachi F-2500 fluorescence spectrophotometer (Pleasanton, CA, USA) with excitation at A492 and emission at A540. The initial rate of proton transport (A492–540/min) was derived from the slope generated by the first 30 seconds of the acidification assay. Three separate experiments were performed that included three pooled mice per genotype.

ATP hydrolysis assay

Concanamycin A–sensitive ATPase activity of osteoclast membranes (40 µg) was measured as the rate of ADP-dependent NADH oxidation using a coupled spectrophotometric assay.28 Three separate experiments were performed that included three pooled mice per genotype.

Immunoblotting

Thirty micrograms of osteoclast membranes underwent immunobloting with the following antibodies: V–ATPase subunits a3 and E (from Dr Beth Lee). Images were captured using BioRad ChemiDoc Gel Docking System and Quantity One software (BioRad, Hercules, CA, USA).

Statistical analysis

Results are plotted as the mean ± SD of independent experiments or biologic samples; sample size is as indicated in the figures and tables. Statistical analysis was performed using a two-tailed Student's t test, and p values were considered significant at p < .05 or as noted.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

A dominant missense mutation in Tcirg1 causes osteopetrosis in +/R470S mice

In a mouse ENU mutagenesis screen for dominant bone phenotypes, a founder was identified with high BMD. The phenotype is heritable with high penetrance, and positional cloning and sequencing reveal that these mice carry a missense mutation on chromosome 19, in exon 18, nucleotide 2316 (C to A) of Tcirg1 that results in the substitution of a highly conserved arginine by a serine at residue 740 (R740S) in the V–ATPase a3 subunit (Fig. 1A). Both male and female mice heterozygous for R740S (+/R740S) have a similar lifespan and size as their wild-type (+/+) littermates, but their BMD and bone mineral content (BMC) are increased significantly relative to +/+ mice at all ages measured (Fig. 1B, C). Micro–computed tomographic (µCT) imaging of the distal femur (Fig. 2A and Table 1) and vertebrae (not shown) reveal that +/R740S mice have increased trabecular bone mass and decreased bone marrow space compared with +/+ mice. Serum parameters, including calcium and alkaline phosphatase levels, are not significantly different in +/R740S versus +/+ mice (Table 1).

thumbnail image

Figure 1. Mice heterozygous for a3 R740S have a dominant osteopetrotic phenotype. (A) Sequencing of the Tcirg1 gene (Gen Bank NM_ 016921) revealed a C to A transition only in mice with high BMD (+/R740S). (B) Bone mineral density (BMD) and (C) bone mineral content (BMC) were significantly elevated in +/R740S mice throughout their life span. Each data point includes at least six mice/genotype. Results are displayed as mean ± SD, with differences significant as indicated by an asterisk.

Download figure to PowerPoint

thumbnail image

Figure 2. Despite the increase in trabecular bone, +/R740S mice have a normal number of osteoblasts and normal bone formation. (A) Three-dimensional (left), 2D (middle), and cross-sectional (right) reconstruction of femurs by µCT reveals increased trabecular bone in 2-month-old +/R740S mice versus +/+ mice. (B) von Kossa staining of femur sections reveal no changes in bone morphology for +/R740S compared with +/+ femurs. (C) H&E staining of femur sections shows active osteoblast surfaces (arrows point to osteoblasts) in both genotypes (×20 magnification). (D) Normal osteoid seams are noted in the H&E-stained femur sections as well as similar growth plate width in +/R740S sections compared with +/+ sections (×5 magnification). (E) Double fluorescent calcein labels in sections of femurs show similar mineral apposition rates in +/R740S mice relative to +/+ mice. (F, G) Alkaline phosphatase and von Kossa staining of stromal cell cultures from +/+ and +/R740S mice (×1.5 magnification). No differences are noted in CFU-F (total number of colonies, including fibroblast colonies), CFU-ALP (alkaline phosphatase–positive colonies), or CFU-O (alkaline phosphatase–positive colonies with mineralized matrix) from +/R740S and +/+ stromal cells.

Download figure to PowerPoint

Table 1. Analysis of +/R740S and Wild Type Femurs Using µCT and Histomorphometric Analyses
 +/++/R740SP value
  • BV/TV, bone volume/tissue volume, Tb.N, trabecular number, Tb.Th, trabecular thickness, Tb.S, trabecular space, MAR, mineral apposition rate, MS/BS, mineralized surface/bone surface, Ob.S/BS, osteoblast surface/bone surface, RS/BS, resorptive surface/bone surface, Oc.S/BS, osteoclast surface/bone surface, N.Oc, number of osteoclasts. Femurs from 2 month old male mice (n = 3/genotype) were analyzed for both static and dynamic histomorphometric parameters, with the exception of Ob.S/BS, where 6 mice/genotype were analyzed. Serum from 2 month old male mice (n = 3/genotype) was analyzed.

  • *

    p < .05.

µCT imaging
 BV/TV (%)18.6 ± 1.5343.13 ± 13.3*0.034
 Tb.N (µm−1)2.68 ± 0.2187.55 ± 1.35*0.004
 Tb.Th (µm)0.0693 ± 0.0030.0563 ± 0.0080.062
 Tb.S (µm)0.193 ± 0.010.0817 ± 0.012*0.0002
Histomorphometry
 MAR (µm/day)0.255 ± 0.0110.297 ± 0.03750.069
 MS/BS (%)2.85 ± 0.0823.61 ± 0.03570.22
 Ob.S/BS (%)0.257 ± 0.1170.245 ± 0.08650.423
 RS/BS (%)0.120 ± 0.010.012 ± 0.003*0.044
 N.Oc23.5 ± 10.843.3 ± 11.3*0.047
 Oc.S/BS (%)4.32 ± 0.336.63 ± 0.23*0.021
Serum
 Calcium2.65 ± 0.092.57 ± 0.220.58
 Alkaline phosphatase149.3 ± 9.1137.7 ± 12.80.27

+/R740S mice have normal osteoblast parameters

Osteoblasts (Fig. 2C), osteoblast surface/bone surface (Ob.S/BS; Table 1), osteoid seams, and growth plates (Fig. 2B, D) appear comparable in +/R740S and +/+ femurs. The mineralized surface/bone surface (MS/BS; Table 1) and mineral apposition rate (MAR; Fig. 2E and Table 1) are not significantly different in +/R740S compared with +/+ femurs. Quantification of differentiation time course, osteoblast number, and activity confirm no significant difference in +/+ and +/R740S stromal cell cultures (Fig. 2F, G).

+/R740S osteoclasts are increased in number, are morphologically indistinguishable from wild-type osteoclasts, and do not undergo apoptosis

Osteoclast surface per bone surface (Oc.S/BS) is significantly increased and the number of osteoclasts is almost double in +/R740S relative to +/+ femurs (Fig. 3A, Table 1). Despite the increase in osteoclast number, fewer +/R740S osteoclasts are associated with resorption lacunae than +/+ osteoclasts with a 10-fold decrease in resorptive surface/bone surface (RS/BS; Fig. 3A, Table 1). The ruffled border is the unique membrane structure that covers the resorption lacunae and contains the V–ATPases and chloride channels required to lower the pH of the extracellular resorption compartment.1, 2 A common feature in many cases of osteopetrosis resulting from loss-of-function mutations in genes required for osteoclast function is that osteoclasts form and attach to the bone surface with a sealing zone, but they do not form a ruffled border. Therefore, we next used transmission electron microscopy (TEM) to assess whether osteoclast polarization and ruffled border formation are affected by the R740S mutation. Both +/R740S and +/+ osteoclasts are polarized, with abundant vacuoles, a sealing zone, and strikingly, both have well-developed ruffled border membranes (Fig. 3B).

thumbnail image

Figure 3. +/R740S bones contain more osteoclasts that display ruffled borders and have reduced apoptosis. (A) Increased number of TRACP+ (red) cells in sections of femurs from +/+ (left) and +/R740S (right) mice; ×10 magnification (top) and ×60 magnification of a single osteoclast sitting on top of a resorption surface (bottom). (B) Transmission electron micrograph images of osteoclasts from the metaphyses of +/+ and +/R740s bones. Attachment to bone, polarization, and ruffled border are evident in the +/R740S osteoclast image. (C) Osteoclast apoptosis is significantly decreased in +/R740S femur sections (N = 6 sections/genotype). (D) Increased expression of an antiapoptotic gene Bcl2 and decreased expression of Casp3, an apoptotic gene, in +/R740S bones. (All markers were normalized to Gapdh, and ratios of the +/+ bones were set at 100%.) Gene expression experiments were conducted three times with three mice per genotype (N = 3 sections/genotype). RB = ruffled border; V = vacuoles; P = polarization; M = mitochondria; SZ = sealing zone.

Download figure to PowerPoint

To determine whether the increased number of osteoclasts found in +/R740S bones is due to increased survival, osteoclast apoptosis was quantified and found to be decreased 10-fold in +/R740S relative to +/+ mouse femurs (Fig. 3C). Consistent with this observation, there is a significant increase in Bcl2 gene expression, an antiapoptotic marker, along with a significant decrease in Casp3, a marker for apoptosis, in +/R740S relative to +/+ bones (Fig. 3D).

Consistent with the increased number of osteoclasts, RNA expression within bone of all osteoclast markers tested, such as tartrate-resistant acid phosphatase (Tracp), calcitonin receptor (CalR), dendritic cell–specific transmembrane protein (Dc-stamp), osteoclast-associated receptor (Oscar) and V–ATPase subunits a3 and d2, are significantly elevated (Fig. 4A, B). Moreover, factors secreted by stromal cells and osteoblasts that are necessary for osteoclast differentiation and survival, such as receptor activator of nuclear factor-κB ligand (Rankl) and macrophage colony-stimulating factor (M-csf), are also significantly increased, whereas osteoprotegerin (Opg) levels are not affected (Fig. 4C). Expression of M-csf, Rankl, and Opg are key determinants of osteoclastogenesis and bone resorption, as well as of osteoclast survival.29 These results demonstrate that this significantly elevated Rankl and M-csf expression is most likely responsible for the higher osteoclast number, as well as for the decrease in the number of apoptotic cells.

thumbnail image

Figure 4. +/R740S osteoclasts have elevated expression of differentiation markers. (A) Conventional PCR reveals that +/R740S bones contain increased levels of osteoclast differentiation markers (Tracp, Oscar, and Dc-stamp), calcitonin receptor (CalR), V–ATPase subunits a3 and d2, as well as Rankl and Opg. (B) Quantification of expression of the osteoclast differentiation markers. (C) Real-time quantitative PCR reveals increased expression of Rankl, Opg, and M-csf, secreted by osteoblasts and stromal cells, which are markers that promote osteoclast differentiation and survival. All conventional PCR markers were normalized to Gapdh, and ratios of the +/+ bones were set at 100% and all quantitative PCR markers are relative to L32. Gene expression experiments were conducted three times with three mice per genotype. Results are displayed as mean ± SD, with significant differences between +/+ and +/R740S indicated by an asterisk.

Download figure to PowerPoint

The +/R740S mutation uncouples ATP hydrolysis from proton transport

V–ATPases are evolutionarily conserved, and mutational analysis has demonstrated that R735 in yeast (R740 in mouse) is essential for proton transport.30 We therefore quantified the ability of +/R740S osteoclast membranes to pump protons and hydrolyze ATP. Immunoblots reveal that osteoclast membranes from +/R740S mice express both the a3 subunit (V0) and the E subunit (V1; Fig. 5D).The initial rate of concanamycin A–sensitive proton transport is comparable between +/R740S and +/+ kidney membranes (Fig. 5A), in which a4 rather than a3 is the predominant V–ATPase isoform.31 In contrast, there is a 90% reduction in concanamycin A–sensitive proton transport in +/R740S osteoclast membranes (Fig. 5B). Nevertheless, there is no statistical difference found in the rates of concanamycin A–sensitive ATP hydrolysis between +/R740S and +/+ osteoclast membranes (Fig. 5C).

thumbnail image

Figure 5. The +/R740S mutation is dominant negative for V–ATPase activity and uncouples ATP hydrolysis from proton transport. (A) Proton transport in kidney is not affected, whereas (B) osteoclast membrane proton transport is decreased by 90% in +/R740S relative to +/+ mice. (C) Concanamycin A–sensitive ATP hydrolysis is not statistically different between +/R740S and +/+ osteoclast membranes. (D) Immunoblotting of osteoclast membranes (30 µg) shows expression of the V0 subunit a3 and the V1 subunit E.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

We describe the first identified dominant-negative Tcirg1 point mutation that abrogates V-ATPase proton pumping and bone resorption while also uncoupling the proton pumping from ATP hydrolysis in mammalian cells. Unexpectedly, and in contrast to cases where Tcirg1 is truncated or absent (Tcirg1–/– and oc/oc mice), osteoclasts from +/R740S mice are polarized and display ruffled borders, suggesting that V0 complexes containing the a3 subunit have functions independent of proton translocation.

Similarly to oc/oc mice,32 +/R740S mice have normal osteoblast parameters (with Ob.S/BS, MS/BS, and MAR in vivo and stromal cell differentiation time course and endpoints in vitro not affected by the mutation) and significantly increased osteoclast parameters (N.Oc and Oc.S/BS). An increase in osteoclast number in vivo is common to certain osteopetrotic conditions in mice and humans, including mutations in the a3 subunit of V–ATPase and some ClC7 mutations,3, 33, 34 and has been attributed to compensatory mechanisms resulting from increased bone density or decreased resorption ability of the osteoclasts present.35 Karsdal and colleagues also have suggested that resorption may trigger apoptosis such that when resorption is decreased, apoptosis is decreased.35The latter is consistent with the 10-fold decrease in apoptosis, decreased expression of Casp3, and increased expression of Bcl2 in +/R740S bones. Bcl2, in addition to being antiapoptotic, also may be critical for osteoclast maturation.36 PCR analysis of RNA from +/R740S bones also demonstrated elevated Rankl and M-csf expression relative to +/+ mice. A similar upregulation of Rankl and M-csf expression has been observed in the bone marrow of the oc/oc mice.37 Rankl is a known osteoclast survival factor, and therefore, elevated Rankl levels in +/R740S bones could be responsible for the decrease in osteoclast apoptosis.29 Taken together, our data suggest that both an increase in Rankl and Bcl2 expression and a decrease in osteoclast apoptosis contribute to the increase in osteoclast number seen in +/R740 and certain other osteopetrotic models in order to compensate for their loss of function. This increase in osteoclast number and osteoclast differentiation markers also suggests that osteoclast differentiation per se is unaffected by the R740S mutation in vivo, similar to findings from Tcirg1–/– and oc/oc mice.20, 38

As raised earlier, osteopetrosis can be subdivided into two broad groups: one is characterized by a lack of osteoclast differentiation (as with RANKL and RANK mutations13, 14, 39, 40), and the other one is characterized by osteoclasts present in normal or even elevated numbers but with reduced or absent resorption capacity (mutations or deletions to TCIRG1 of V–ATPase, ClCN7, CAII, PLEKHM1, and OSTM1 genes). These osteoclasts either have a disrupted or absent ruffled border and in oc/oc mice are not polarized. The +/R740S mouse is unique: osteoclasts are polarized and have a well-developed ruffled border membrane. One possible explanation for this difference could be the residual (∼10%) proton translocation observed in +/R740S osteoclast membranes. However, this residual pumping activity may be attributable to Na/H antiport activity rather than V–ATPase, as was observed in humans with a3 mutations.41 Alternatively, the difference in phenotype may be attributable to the fact that, similar to mutations made to the yeast R735 residue,42 the R740S mutation affects proton transport but not V–ATPase subunit expression or V0 assembly. Polarization and/or ruffled border formation may require the presence of the V0 but not its proton-translocation activity. Evidence suggesting that the V0 complex has proton-pumping-independent functions comes from specific mutations in a isoforms that affect synaptic vesicle fusion in Drosophila43 and insulin secretion in mice44; in both cases, the primary defect is vesicular fusion, not acidification. One explanation for this proton-pumping-independent role comes from data showing that vesicular fusion is facilitated by two V0 complexes coming together in a head-to-head configuration bringing two vesicles together.45, 46 Alternatively, the a subunit may act as a scaffold for other proteins essential to vesicle formation and budding. This idea is supported by work in renal epithelial cells showing that the a2 isoform interacts directly with two proteins essential to vesicle formation and budding (ARF and ARNO47, 48). Further support for this hypothesis comes from d2, the other osteoclast-enriched V–ATPase V0 subunit; deleting d2 in mice does not affect osteoclast V–ATPase activity or resorptive capacity but rather decreases osteoclast fusion efficiency, an observation that led the authors to suggest that d2 functions as part of a “fusicon” rather than in a proton-pumping complex.27

In summary, we provide for the first time evidence that a missense point mutation, R740S, of the a3 subunit of the V-ATPase complex is dominant negative for proton pumping and bone resorption while also uncoupling the proton pumping function from the ATP hydrolysis function of the V-ATPase in mammalian cells. We also demonstrate, in contrast to the case when the a3 subunit is ablated or truncated (oc/oc mice), that the R740S mutation in a3 does not affect polarization or ruffled border formation in osteoclasts of the +/R740S mice. This shows mechanistically that polarization and ruffled border formation require the presence of the V0 subunit of the V-ATPase but not its proton-translocation activity and lends support to the hypothesis that the V0 complex has proton-pumping-independent functions.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

MF Manolson and JE Aubin contributed equally to this work.

We thank LB Kelsey and I Vukobradovic (Centre for Modeling Human Disease) for assistance with maintaining the mice. We also thank members of the Aubin Lab for assistance with dissection and helpful discussion; K Lee (Manolson Lab) for making the recombinant RANKL; W Li (Henderson Lab) for µCT and histomorphometric analyses; F Serraf (Dental Research Institute Histology Lab, Faculty of Dentistry, University of Toronto) for paraffin embedding, cutting, and staining femur sections; the Microscope Imaging Laboratory at University of Toronto for sample preparation for transmission electron microscopy; and Dr B Lee (Ohio State University) for V–ATPase a3 and E antibodies. This work was supported by grants from the Canadian Institutes of Health Research (FRN 69198 to JEA and FRN MOP-79322 to MFM and JEA) and the Quebec Transgenic Network (JEH), as well as scholarship support from the Canadian Arthritis Network (IV and RAZ), the CIHR/Osteoporosis Society (RAZ), the CIHR Strategic Training Program Cell Signaling in Mucosal Inflammation and Pain (NO), and the Faculty of Dentistry (University of Toronto) Harron Fund (NO).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References
  10. Supporting Information

Additional Supporting information may be found in the online version of this article.

FilenameFormatSizeDescription
jbmr_355_sm_suppTable1.doc28KSupplemental Table 1. List of primers used for conventional and real time PCR.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.