SEARCH

SEARCH BY CITATION

Keywords:

  • osteocyte;
  • osteoclast;
  • MLO-Y4;
  • RANKL;
  • macrophage colony-stimulating factor;
  • 1,25-dihydroxyvitamin D3

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Osteocytes are terminally differentiated cells of the osteoblast lineage that have become embedded in mineralized matrix and may send signals that regulate bone modeling and remodeling. The hypothesis to be tested in this study is that osteocytes can stimulate and support osteoclast formation and activation. To test this hypothesis, an osteocyte-like cell line called MLO-Y4 and primary murine osteocytes were used in coculture with spleen or marrow cells. MLO-Y4 cells support osteoclast formation in the absence of 1,25-dihydroxyvitamin D3 [1,25(OD)2D3] or any other exogenous osteotropic factor. These cells alone stimulate osteoclast formation to the same extent or greater than adding 1,25(OH)2D3. Coaddition of 1,25(OH)2D3 with MLO-Y4 cells synergistically increased osteoclast formation. Optimal osteoclast formation and pit formation on dentine was observed with 200–1000 MLO-Y4 cells per 0.75-cm2 well. No osteoclast formation was observed with 2T3, OCT-1, or MC3T3-E1 osteoblast cells (1000 cells/well). Conditioned media from the MLO-Y4 cells had no effect on osteoclast formation, indicating that cell contact is necessary. Serial digestions of 2-week-old mouse calvaria yielded populations of cells that support osteoclast formation when cocultured with 1,25(OH)2D3 and marrow, but the population that remained in the bone particles supported the greatest number of osteoclasts with or without 1,25(OH)2D3. To examine the mechanism whereby these cells support osteoclast formation, the MLO-Y4 cells were compared with a series of osteoblast and stromal cells for expression of macrophage colony-stimulating factor (M-CSF), RANKL, and osteoprotegerin (OPG). MLO-Y4 cells express and secrete large amounts of M-CSF. MLO-Y4 cells express RANKL on their surface and their dendritic processes. The ratio of RANKL to OPG mRNA is greatest in the MLO-Y4 cells compared with the other cell types. RANK-Fc and OPG-Fc blocked the formation of osteoclasts by MLO-Y4 cells. These studies suggest that both RANKL and OPG may play a role in osteocyte signaling, OPG and M-CSF as soluble factors and RANKL as a surface molecule that is functional in osteocytes or along their exposed dendritic processes.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

THE OSTEOCYTE is the bone cell that we know least about and, therefore, the function of this cell is the focus of numerous hypotheses. These include the capacity to regulate calcium homeostasis,(1) to respond to mechanical strain,(2–6) and to send signals of bone formation or bone resorption to the bone surface.(7) These functions are proposed to be accomplished through gap junctions,(8) through the secretion of factors,(9) through glutamate receptors,(10) and through the direct dendritic contact with cells on the bone surface such as lining cells and osteoblasts.(11)

Cell-to-cell interactions play a pivotal role in regulation of osteoclast formation and bone resorption. Osteoblasts and/or marrow stromal cells have been implicated as the critical cells responsible for the formation and subsequent activation of osteoclasts (for review see Aubin and Bonnelye(12). Recently, it has been shown that fibroblasts from any tissue source when treated with bone-resorbing factors will support osteoclast formation.(13) It has been hypothesized that osteocytes can initiate resorption through various mechanisms.(14, 15) To test this theory, we have used the osteocyte-like cell line MLO-Y4 that possesses many of the properties of primary osteocytes such as dendritic processes; no expression of the osteoblast-specific antigen known as periostin; low or no ALP expression; and high expression of osteocalcin, CD44, connexin 43, and the osteocyte-like E11 antigen.(16, 17) These cells possess functional gap junctions that are responsive to fluid flow shear stress,(18) can communicate with osteoblasts through gap junctions,(19) and are the only bone cell line known to respond to glucocorticoids with induction of apoptosis.(20)

It has been well known for the last 10–15 years that osteoclast precursors require supporting cells for osteoclast formation. The importance of macrophage colony—stimulating factor (M-CSF) in osteoclast formation has also been determined.(21–23) Critical factors and cell surface molecules involved in this process have been elucidated only recently with the discovery of RANKL and osteoprotegerin (OPG).(24, 25) The osteoclast precursor expresses a receptor known as RANK that signals through the NF-κB pathway. The binding of the cell membrane-bound ligand RANKL activates this receptor. However, a soluble factor (OPG) acting as a “decoy” receptor can bind to RANKL preventing osteoclast formation. The expression of RANKL on the surface of supporting cells occurs when these cells are exposed to bone-resorbing cytokines, hormones, and factors such as interleukins (IL)-1, IL-6, IL-11, parathyroid hormone-related protein, parathyroid hormone, oncostatin M, leukemia inhibitory factor, prostaglandin E2, or 1,25-dihydroxyvitamin D3 [1,25(OH)2D3].(12) These factors up-regulate RANKL to a level capable of overcoming the effects of circulating OPG, thereby resulting in osteoclast formation. Efforts to generate osteoclasts without supporting cells have been accomplished only recently in vitro by using an artificial soluble form of RANKL.(23)

Here, we show that an osteocyte-like cell line MLO-Y4 will support osteoclast formation in the absence of any exogenous bone-resorbing factors and the mechanism whereby this occurs is through M-CSF and RANKL expression.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Materials

Tissue culture media were purchased from Gibco BRL (Grand Island, NY, USA); FBS was from BioWhittaker (Walkersville, MD, USA) or Hyclone Laboratories, Inc. (Logan, UT, USA). Calf serum (CS) was from Hyclone Laboratories, Inc. Rat tail collagen type I was purchased from Becton Dickinson Laboratories (Bedford, MA, USA). Collagenase and leukocyte acid phosphatase kit and peroxidase substrate o-phenylenediamine (OPD) were purchased from Sigma Chemical Co. (St. Louis, MO, USA). Goat immunoglobulin G (IgG) biotinylated anti-horse antibody for a negative control and all kits for the immunohistochemistry, avidin/biotin-blocking kits, ABC-ALP system, Vector Black Substrate kit, and Vectamount were purchased from Vector Laboratories (Burlingame, CA, USA). Neutralizing antibody for mouse M-CSF, recombinant mouse M-CSF, goat IgG biotinylated anti-mouse trance antibody (BAF462), and recombinant mouse Rank-Fc and mouse OPG-Fc (both RANK and OPG coupled to the Fc portion of immunoglobulin) were purchased from R&D Systems (Minneapolis, MN, USA); 1,25(OH)2D3 was from BIOMOL Research Laboratories, Inc. (Plymouth Meeting, PA, USA).

Cell culture

MLO-Y4, ST2, MC3T3-E1, 2T3, and OCT-1 cells were cultured as described previously.(16) ST2 cells treated with 1,25(OH)2D3 are well known to support osteoclast formation.(26) MC3T3-E1 and OCT-1 cells are used as representative of osteoblasts for the following reasons. MC3T3-E1 cells are a nontransformed cell line established from a normal mouse calvaria(27) and have been well characterized. The OCT-1 cells were established from the same type of transgenic mouse as the MLO-Y4 cells but from newborn mouse calvaria.(28) The osteoblast-like cell line 2T3 was established from transgenic mice expressing the large T-antigen driven by the bone morphogenetic protein (BMP) 2 promoter.(29) OCT-1, 2T3, and MLO-Y4 cells express the large T-antigen. Therefore, any properties of the MLO-Y4 cells that are different from the OCT-1 and 2T3 cells are not because of the expression of this antigen.

Coculture of cells with spleen or marrow cells

MLO-Y4 cells were plated at 50–1000 cells/well in 48-well plates the day before the addition of murine spleen or marrow cells. Spleen or marrow was removed from 4- to 5-week-old mice, a cell suspension prepared and seeded at 5 × 105 spleen cells or 106 marrow cells/well. These cells were cocultured in α-modified essential medium (α-MEM; without deoxyribonucleosides and ribonucleosides; JRH Biosciences, Lenexa, KS, USA) plus 10% FBS (Hyclone Laboratories, Inc.) and treated with or without 10−8 M of 1,25(OH)2D3 for 6 days for marrow coculture or 10 days for spleen coculture. One-half of the media was replaced every 2 days. At the end of culture, the cells are fixed and stained for TRAP using a leukocyte acid phosphatase kit (Sigma Chemical Co.) as described previously.(30) Three or more nuclei TRAP+ cells were counted under a microscope. After counting, cells were removed by toothbrush and stained with 1% toluidine blue, and osteoclast-resorbing lacunae were quantitated as described previously.(30) All of the mice used for these studies were cared for and killed according to the University of Texas Health Science Center, San Antonio (UTHSCSA) Institutional Care and Oversight Committee.

Isolation of primary bone cells

The method used was described by Mikuni-Takagaki(31) with minor modification. Briefly, calvaria or long bones were isolated from 2- to 3-week-old mice and trimmed of soft tissue. The bones were cut into small pieces followed by sequential 0.75% collagenase digestion for five times (fractions 1–5) and alternating treatment with 4 mM of EDTA and collagenase twice (fractions 6–9). The cells from each fraction were collected by centrifugation. The remaining bone pieces were transferred into collagen-coated dishes and cells migrated from the bone pieces after 3–5 days of culture (fraction 10). The cells from each fraction were trypsinized and cocultured with bone marrow cells.

ELISA for M-CSF

M-CSF production in MLO-Y4 cells was determined by a single-site ELISA. Twenty-four-hour serum-free conditioned media were collected from the cultured cells. One hundred microliters of conditioned media were coated onto the high-binding ELISA plates (Corning, Inc., Corning, NY, USA) at 4°C overnight. The bound M-CSF was detected by anti-mouse M-CSF antibody followed by peroxidase-conjugated secondary antibody reaction with OPD substrate. The absorbance was read at 450 nm. The total amount of M-CSF production was converted to nanograms per 105 cells per milliliter using the standard curve.

Reverse transcriptase-polymerase chain reaction

Total RNA was isolated from cultures of cells, fresh tissue homogenate, or bone pieces using RNA-Zol (Biotecx Laboratories, Inc., Houston, TX, USA) according to the manufacturers instructions. cDNA was synthesized from 5 μg of total RNA in a 20-μl reaction mixture containing 1× first-strand buffer, 500 μM of deoxynucleoside triphosphate (dNTP), 10 mM of dithiothreitol (DTT), 500 ng of oligo (dT)12–18 primer, and 200 U Super Script II reverse transcriptase (RT; Gibco BRL, Life Technologies, Baltimore, MD, USA). One microliter of cDNA is amplified using polymerase chain reaction (PCR) in a 50-μl reaction mixture containing 1× PCR buffer, 5 pmol each of 5′ and 3′ primer, 200 μM of dNTP, 1 mM of MgCl2, and 2.5 U of TaqDNA polymerase (Gibco BRL). Amplifications were performed in a DNA Thermal Cycler 480 (Perkin Elmer Cetus, Emeryville, CA, USA) for 40 cycles after the reaction profile: 95°C for 45 s, 57°C for 30 s, and 72°C for 45 s. Mouse-specific primers for RANKL and OPG are kindly provided by Dr. B. Oyajobi (UTHSCSA) and the mouse M-CSF-specific primers used were described by Dr. J. Rubin.(32)

The primers are as follows:

RANKL: forward, 5′-TTTGCAGGACTCGACTCTGGAG-3′; reverse, 5′-TCCCTCCTTTCATCAGGTTATGAG-3′

OPG: forward, 5′-ATCATTGAATGGACAACCCAGG-3′; reverse, 5′-TGCGTGGCTTCTCTGTTTCC-3′

GAPDH: forward, 5′-TTGAAGGGTGGAGCCAAACG-3′; reverse, 5′-ACACATTGGGGGTAGGAACACG-3′

Northern blot analysis

Total RNA was isolated from cultured cells using RNAzol (Tel-Test, Friendswood, TX, USA) followed by enrichment of polyA by using 20 mg of oligo dT columns. Approximately 2 μg of polyA RNA precipitated with 10 μg of t-RNA was loaded in each lane of a 1% agarose formaldehyde gel and transferred to a “super-charge” nytran membrane (Schleicher & Schuell, Keene, NH, USA) using an Schleicher & Schuell turboblotter per manufacturers instructions. The RNA was cross-linked to the filter by UV irradiation (Stratalink; Stratagene, San Diego, CA, USA). The membranes were prehybridized for 1–4 h using freshly boiled and quenched on ice sheared salmon sperm at 150 μg/μl in a hybridization oven. Stratagene solution for hybridization was used (50% formamide [Life Technologies], 5× SSC; 1× PE solution [50 mM of Tris-HCl; pH 7.5; 1% sodium dodecyl sulfate {SDS }; 0.2% polyvinylpyrrlidone-400,000; 0.2% ficol-40,000; 5 mM of EDTA; and 0.2% bovine serum albumin }BSA}], and 150 μg/ml of denatured sheared salmon sperm DNA). The cDNA probes were labeled with P32 using High Prime Mix per manufacturers' instructions (Boehringer Mannheim, Indianapolis, IN, USA) and was added to the prehybridization solution at a concentration of 1.5 × 106 cpm/ml. Hybridization was performed for 16–42 h and then washed at 56°C in 2× SSC and 0.1% SDS for 2–4 h followed by a stringent wash of 0.5% SSC and 0.1% SDS for 15 minutes and a second wash with 0.1% SSC and 0.1% SDS for the same time period. BioMax double-sided film and intensifying screen were used from overnight to 5 days. The RANKL probe is a full-length (2.2 kb) mouse cDNA kindly provided by Dr. Youngwon Choi (Rockefeller University, New York, NY, USA). The OPG probe is a 535-bp (sequence from 133–668) DNA fragment provided by Dr. Babatunde Ojajobi (UTHSCSA). The M-CSF (CSF-1) probe is a mouse 1.3-kb cDNA probe kindly provided by Dr. Nandini Ghosh-Choudhury (UTHSCSA). Mouse GAPDH (1.4 kb fragment) is used as a control. Filters were stripped for reblotting for each probe and analyzed using a PhosphorImager.

Neutralization experiments using antibody to M-CSF

The MLO-Y4 cells were cocultured with bone marrow cells as described previously. Four concentrations of anti-mouse M-CSF antibody (0.2, 1.0, 5.0, and 10 μg/ml) were added to the MLO-Y4 cells 2 h before the addition of the bone marrow cells. One-half of the media was replaced every 2 days with fresh antibody and for the final media change the antibody treatment was increased to 100 μg/ml. Normal goat IgG was used as a control. Multinucleated TRAP+ cells were counted for comparison bone marrow treated with 1,25(OH)2D3.

Fluorescent-activated cell sorting analysis

Single cell suspensions were washed twice with PBS containing 5% FCS and 0.1% sodium azide. Nonspecific binding was blocked by incubating cells with PBS containing 5% FCS for 30 minutes at 4°C. Approximately 1–2 × 105 cells were incubated with fluorescein isothiocyanate (FITC) isotype control antibody or RANKL antibody (generous gift from Dr. Nobuyuki Udagawa, Tokyo, Japan) followed by FITC secondary antibody for 30 minutes in the dark. The cells were washed twice and fixed in 0.1% formaldehyde before analysis. Flow cytometry was done using a fluorescence-activated cell-sorting (FACS) Calibur with Simultest analysis software (Becton Dickenson, San Jose, CA, USA).

Immunostaining for RANKL

The MLO-Y4 cells were plated at 1 × 105 in Lab-Tek II Chamber Slides (Nalge Nunc International, Naperville, IL, USA) that were previously coated with collagen as described previously and cultured for 3 days. The media and chambers were removed, the slides were washed 2× with PBS for 5 minutes and fixed in 4% paraformaldehyde for 10 minutes, washed with PBS, and then permeabilized with PBS-0.05% Tween for 3 minutes with agitation. The slides were blocked using the avidin/biotin blocking kit (Vector Laboratories) using the manufacturer's instructions and then blocked with 5% BSA for 30 minutes followed by incubation with the goat IgG biotinylated anti-mouse trance antibody (BAF462) for 1 h. The negative control was goat IgG biotinylated anti-horse Abs (Vector Laboratories). The primary antibodies were used at 0.0025 μg/μl in 1% BSA. After washing in PBS 0.1% Tween, the slides were incubated with the ABC-ALP system (Vector Laboratories), for 30 minutes followed by washing and development using the Vector Black Substrate Kit for 35 minutes. The color reaction was stopped by washing the slides in distilled water. The slides were counterstained with 1% methyl green for 3 minutes, washed in water, dehydrated with graded alcohol, clarified with two washes of xylene, and mounted using VectaMount (Vector Laboratories).

Neutralization of RANK and RANKL using RANK-Fc and OPG-Fc

The MLO-Y4 cells were cocultured with bone marrow cells as described previously. Two concentrations of RANK-Fc were used, 50 ng/ml and 500 ng/ml. In separate experiments, four concentrations of OPG-Fc were used, 12.5, 25, 50, and 100 ng/ml. One-half of the medium was replaced with fresh α-MEM plus RANK-Fc or OPG-Fc every 2 days. The experiment was terminated at 7 days and the cells were fixed, stained, and counted as described previously.

Statistical analysis

Data were analyzed with the one-way ANOVA using the Tukey's multiple comparison or Bonferroni multiple comparison post hoc test.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

To test the hypothesis that osteocytes can support osteoclast formation, MLO-Y4 cells were incubated with murine spleen cells, an enriched source of osteoclast precursors compared with murine bone marrow cells (Fig. 1). MLO-Y4 cells (1000 cells/well) supported osteoclast formation without the addition of 1,25(OH)2D3 when cocultured with spleen cells (5 × 105; significantly different from control, p < 0.001). This activity was enhanced with the addition of 1,25(OH)2D3 that was significantly different from cells without 1,25(OH)2D3 (p < 0.05). Spleen cells alone or treated with 1,25(OH)2D3 will not form osteoclasts.(33)

thumbnail image

Figure FIG. 1.. MLO-Y4 cells when cocultured with spleen cells will support osteoclast formation without the addition of 1,25(OH)2D3. This capacity was further enhanced with the addition of 1,25(OH)2D3. MLO-Y4 cells (103) were cultured with spleen cells (5 × 105) with and without 1,25(OH)2D3. Spleen cells alone or treated with 1,25(OH)2D3 will not form osteoclasts. *Significantly different from control (p < 0.001); #significantly different from the other two groups (p < 0.05).

Download figure to PowerPoint

Next, to determine whether MLO-Y4 cells will support the formation of active osteoclasts, pit formation on dentin slices was examined using coculture with mouse marrow cells. MLO-Y4 cells support the formation and activation of pit-forming osteoclasts (Fig. 2A). Quantitation of pit number per slice (Fig. 2C) and pit area per slice (Fig. 2B) showed that MLO-Y4 cells alone will support pit formation on dentine. The optimal number of MLO-Y4 cells was 500–1000 cells seeded per well. Large areas containing contiguous and overlapping pits were observed, especially with the addition of 1,25(OH)2D3. Coaddition of 1,25(OH)2D3 with MLO-Y4 cells enhanced osteoclast formation above either cells alone or 1,25(OH)2D3 alone. MLO-Y4 cells cocultured alone or treated with 1,25(OH)2D3 do not form osteoclasts (data not shown). When the data were calculated as area of resorption lacunae per TRAP+ multinucleated cell (MNC), an increase in area was observed with increasing number of MLO-Y4 cells (Fig. 2D).

thumbnail image

Figure FIG. 2.. MLO-Y4 cells support the formation of pit-forming osteoclasts. (A) Large areas containing contiguous and overlapping pits were observed especially with the addition of 1,25(OH)2D3. Quantitation of (B) pit area per slice and (C) pit number per slice showed that MLO-Y4 cells alone will support pit formation by TRAP+ MNCs. (D) When calculated as resorption area per TRAP+ MNC, there appeared to be an increase in pit size with an increase in number of MLO-Y4 cells. The optimal number of MLO-Y4 cells appeared to be 500–1000 cells/well. When 1,25(OH)2D3 and MLO-Y4 cells are present in the culture, enhanced osteoclast formation is observed. (B and C) *Significantly different from all other groups (p < 0.001); #significantly different from all groups not treated with 1,25 D3 (p < 0.05). (D) *Significantly different from same group not treated with 1,25(OH)2D3; #significantly different from 0 cells and 500 cells treated with 1,25(OH)2D3.

Download figure to PowerPoint

Plastic surfaces were compared with dentin surfaces with respect to MLO-Y4 support of osteoclast formation (Fig. 3). Quantitation of TRAP+ MNCs showed that ∼500–1000 MLO-Y4 cells seeded in 48-well plates gave optimum osteoclast support on both plastic and dentin. However, greater numbers of TRAP+ MNCs were observed on dentin compared with plastic, but the TRAP+ MNCs appeared larger and more spread out on the plastic surface compared with dentin, suggesting that on plastic these cells may fuse to a greater extent. Addition of 1,25(OH)2D3 significantly enhanced osteoclast formation but the effect on dentin was more dramatic with more than a twofold increase in TRAP+ MNCs (Fig. 3D).

thumbnail image

Figure FIG. 3.. MLO-Y4 cells support the formation of TRAP+ MNCs by bone marrow cells on both plastic and dentine. Photographs of TRAP+ MNCs on (A) plastic and (C) dentin. Quantitation of TRAP+ MNCs showed that ∼500 MLO-Y4 cells gave optimum support of osteoclast formation on (B) plastic and (D) dentin. Greater numbers of TRAP+ MNCs were observed on (D) dentin compared with (B) plastic, (A) but the TRAP+ MNCs appeared larger and more spread out on the plastic surface. 1,25(OH)2D3, (B) did not greatly enhance osteoclast formation stimulated by MLO-Y4 cells cultured on plastic, (D) but significantly enhanced the effects of MLO-Y4 cells cultured on dentin. *Significantly different from the rest of the groups but not between the two (p < 0.01); ±significantly different from control and group 50 without 1,25(OH)2D3 (p < 0.001); #significantly different from the corresponding cell number without 1,25(OH)2D3 (p < 0.001 for group 0 and group 50; p < 0.05 for group 500 and group 1000).

Download figure to PowerPoint

MLO-Y4 cells were compared with the osteoblast cell lines OCT1 and 2T3 with respect to support of osteoclast formation and to determine if MLO-Y4 conditioned media would support osteoclast formation (Fig. 4A). The support cells were seeded at 1000 cells/well for 24 h before the addition of 106 bone marrow cells. The cells were cultured with and without the addition of 1,25(OH)2D3. MLO-Y4 cells significantly supported the formation of TRAP+ MNCs without the addition of 1,25(OH)2D3 and this activity was enhanced further with the addition of 1,25(OH)2D3. Neither osteoblast cell line supported osteoclast formation. The addition of MLO-Y4 conditioned media had no effect, suggesting that cell-to-cell contact is necessary.

thumbnail image

Figure FIG. 4.. Comparison of MLO-Y4 cells to the osteoblast cell lines OCT1 and 2T3 and primary osteocytes with respect to support of osteoclast formation from marrow cells. MLO-Y4 cells significantly support the formation of TRAP+ MNC without the addition of 1,25(OH)2D3. This activity was enhanced further with the addition of 1,25(OH)2D3. (A) Neither osteoblast cell line supported osteoclast formation. The addition of MLO-Y4 conditioned media (Y4-C.M.) had no effect, suggesting that cell-to-cell contact is necessary. *Significantly different from all of the other treatment groups (p < 0.001); #significantly different from all of the other groups without 1,25(OH)2D3 treatment (p < 0.001). (B) Primary cells were isolated from murine long bones and the early (F1–2), middle (F3–5), and late (F6–9) collagenase-digested fractions and cells migrating from the bone pieces were compared for support of osteoclast formation. The cells that migrated from the bone pieces significantly supported osteoclast formation without the addition of 1,25(OH)2D3, which was greatly enhanced with the addition of 1,25(OH)2D3 (BM, bone marrow cells). +Significantly different from all other 1,25(OH)2D3 minus groups; *significantly different from all other groups; #significantly different from the plus 1,25(OH)2D3 groups (p < 0.001).

Download figure to PowerPoint

Previously, it has been shown that freshly isolated chick osteocytes will support osteoclast formation.(14) This has not been performed with mammalian cells. Primary cells were isolated from murine long bones and the early (F1–2), middle (F3–5), and late (F6–9) collagenase-digested fractions were collected. The remaining bone pieces were cultured and cells migrating from the bone pieces were isolated (F10). These cells were incubated with murine bone marrow with and without the addition of 1,25(OH)2D3. The collagenase-digested fractions did not significantly support osteoclast formation without the addition of 1,25(OH)2D3; however, the cells that migrated from the bone pieces (F10) significantly supported osteoclast formation without the addition of 1,25(OH)2D3, which was greatly enhanced with the addition of 1,25(OH)2D3 (Fig. 4B).

Because M-CSF has been shown to support the formation of osteoclast precursors, it was next determined if MLO-Y4 cells express or secrete this factor. MLO-Y4 cells express mRNA for both secreted and membrane-bound M-CSF as determined by RT-PCR (Fig. 5A). No obvious differences could be observed between MLO-Y4 cells and 2T3, OCT-1, MC-3T3, ST2, ST2 plus 1,25(OH)2D3, or spleen cells as determined by RT-PCR. Northern blot analysis was performed comparing MLO-Y4 cells with 2T3 osteoblast cells. After normalizing the M-CSF to GAPDH, the MLO-Y4 cells expressed 21.2 times greater mRNA than 2T3 cells (data not shown). The MLO-Y4 cells secrete large amounts of M-CSF protein into their conditioned media compared with marrow cells treated with or without 1,25(OH)2D3 (Fig. 5B). In an attempt to determine if the M-CSF produced by the MLO-Y4 was necessary to support osteoclast formation, neutralization experiments were performed using anti-mouse M-CSF. Only at very high concentrations of antibody could an effect be observed on osteoclast formation (Fig. 6). It was not possible to add sufficient amounts of neutralizing antibody to block all M-CSF activity produced by the MLO-Y4 cells. According to the manufacturer's instructions, the ND50 neutralization dose for 2.5 ng/ml of recombinant murine M-CSF was 0.2–0.6 μg/ml.

thumbnail image

Figure FIG. 5.. MLO-Y4 cells secrete large amounts of M-CSF. (A) MLO-Y4 cells express mRNA for both secreted and membrane-bound M-CSF as determined by RT-PCR. No obvious differences could be observed between MLO-Y4 cells and 2T3, OCT-1, MC-3T3-E1, ST2, ST2 plus 1,25(OH)2D3, or spleen cells. However, by Northern analysis, MLO-Y4 cells express 21.2 times more message for M-CSF than 2T3 cells (data not shown). (B) MLO-Y4 cells secrete large amounts of M-CSF protein into their conditioned media compared with bone marrow cells or spleen cells with or without 1,25(OH)2D3 treatment (BM, bone marrow cells). *Significantly different from all of the other groups (p < 0.001); +significantly different from each other (p < 0.001).

Download figure to PowerPoint

thumbnail image

Figure FIG. 6.. Neutralization experiments with anti-mouse M-CSF. Neutralizing antibody to M-CSF at 100 μg/ml and 10 μg/ml significantly but not completely blocked 1,25(OH)2D3 stimulation of osteoclast formation by murine bone marrow. However, no effect was observed on the MLO-Y4 support of osteoclast formation. *Significantly different from the IgG control and nontreated group (p < 0.001).

Download figure to PowerPoint

It is well known that RANKL is necessary for osteoclast formation and it has been proposed that the ratio of RANKL to OPG determines whether cells will support resorption or formation.(34) Horwood and coworkers(34) used this approach to show the importance of the relative abundance of RANKL compared with the levels of OPG in the induction of osteoclastogenesis. Figure 7A shows a comparison of RANKL and OPG mRNA expression in a series of cell lines using RT-PCR. The ratio of RANKL to OPG in the MLO-Y4 cells is higher than the other cell types. MLO-Y4 cells appear to express relatively greater amounts of mRNA for RANKL than any of the other cells lines, including ST2 cells treated with 1,25(OH)2D3. MLO-Y4 cells also express OPG mRNA. Obviously, the ratio of RANKL and OPG cannot explain the support of osteoclast formation by MLO-Y4 cells because the OCT-1 did not express mRNA for OPG but did express RANKL mRNA. Northern analysis was also performed using MLO-Y4 cells and osteoblast 2T3 cells (Fig. 7B). The relative ratio of RANKL and OPG in MLO-Y4 cells and 2T3 cells shows that MLO-Y4 cells express 40 times more RANKL mRNA and almost 10 times less OPG mRNA as 2T3 cells. If RANKL is normalized to 1 in the 2T3 cells, the value for RANKL in MLO-Y4 cells is 39.54. If OPG is normalized to 1 in the 2T3 cells, the OPG value in MLO-Y4 cells is 0.086, giving a RANKL/OPG ratio of 77.47.

thumbnail image

Figure FIG. 7.. Comparison of RANKL and OPG mRNA expression in a series of cell lines using RT-PCR and Northern analysis. The PCR product for RANKL is 479 bp and for OPG the PCR product for RANKL is 536 bp. (A) The ratio of RANKL to OPG is higher in the MLO-Y4 cells compared with the other cell types. MLO-Y4 cells appear to express relatively greater amounts of mRNA for RANKL than any of the other cells lines, including ST2 cells treated with 1,25(OH)2D3. (B) Relative ratio of RANKL and OPG in MLO-Y4 cells and 2T3 cells as shown by Northern analysis. If RANKL is normalized to 1 in the 2T3 cells, the value for RANKL in MLO-Y4 cells is 39.54. If OPG is normalized to 1 in the 2T3 cells, the OPG value in MLO-Y4 cells is 0.086, giving a RANKL/OPG ratio of 77.47.

Download figure to PowerPoint

It is well known that mRNA expression does not necessarily correlate with protein expression; therefore, we examined the expression of RANKL on the surface of MLO-Y4 cells. Using anti-RANKL antibody, the cells were sorted using FACS analysis. A significant proportion of the cells expressed the protein on their surface (Fig. 8). Immunohistochemical staining also showed that these cells were positive for RANKL. In fact, the staining with this antibody made the dendritic processes more visible (Fig. 9).

thumbnail image

Figure FIG. 8.. MLO-Y4 cells express RANKL protein on their surface as determined by FACS analysis. The first peak contains the cells incubated with control antibody and the second peak contains the cells incubated with the anti-RANKL antibody.

Download figure to PowerPoint

thumbnail image

Figure FIG. 9.. Immunostaining of the MLO-Y4 cells for the expression of RANKL was clearly positive as shown at lower magnification (20×) and enhanced the visibility of the dendritic processes as shown at higher magnification (40×). The nuclei were counterstained using methyl green.

Download figure to PowerPoint

The RANK receptor coupled to the Fc portion of the immunoglobulin molecule has been shown to block the effects of RANKL on osteoclast formation.(35, 36) Experiments were performed using RANK-Fc in cultures of MLO-Y4 cells with murine marrow cells to verify that RANKL protein expressed on the cell was responsible for osteoclast formation (Fig. 10). RANK-Fc (500 μg/ml) effectively blocked the support of osteoclast formation by MLO-Y4 cells alone and coincubated with 1,25(OH)2D3. These results were validated further using OPG-Fc (Fig. 11). OPG-Fc significantly blocked MLO-Y4 support of osteoclast formation at 12.5 ng/ml and completely blocked MLO-Y4 support of osteoclast formation at 100 ng/ml.

thumbnail image

Figure FIG. 10.. MLO-Y4 support of osteoclast formation is blocked by RANK-Fc. RANK-Fc at 500 ng/ml significantly inhibited the support of osteoclast formation by MLO-Y4 cells with or without 1,25(OH)2D3 treatment (BM, bone marrow cells). (A) Co-cultures plus 1,25(OH)2D3; (B) co-cultures without 1,25(OH)2D3; (C) bone marrow alone plus 1,25(OH)2D3. The value of *p < 0.001 for MLO-Y4 cocultures with bone marrow plus 1,25(OH)2D3; p < 0.05 for bone marrow compared with bone marrow plus 1,25(OH)2D3 and bone marrow plus MLO-Y4 cells.

Download figure to PowerPoint

thumbnail image

Figure FIG. 11.. MLO-Y4 support of osteoclast formation is blocked by OPG-Fc. OPG-Fc at 12.5 ng/ml significantly inhibited the support of osteoclast formation by MLO-Y4 cells with and without 1,25(OH)2D3 treatment. OPG-Fc at 100 ng/ml completely blocked the formation of osteoclast in all three conditions (BM, bone marrow cells). (A) Co-cultures plus 1,25(OH)2D3; (B) co-cultures without 1,25(OH)2D3; (C) bone marrow alone plus 1,25(OH)2D3. The value of *p < 0.001 for MLO-Y4 cocultures with bone marrow plus 1,25(OH)2D3; p < 0.05 for bone marrow compared with bone marrow plus 1,25(OH)2D3 to bone marrow plus MLO-Y4 cells.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Here, we show that MLO-Y4 osteocyte-like cells will support osteoclast formation and activation in the absence of 1,25(OH)2D3 or any other exogenous osteotropic factor. This capacity has not been described for any other cell type except for chick osteocytes.(14) As few as 500 MLO-Y4 cells were more potent than the optimal concentration of 1,25(OH)2D3 in supporting osteoclast formation and activation. Other cell lines such as the mouse bone marrow-derived stromal cell line ST2 support osteoclast formation only in the presence of 1,25(OH)2D3 or dexamethasone(26) because of induction of RANKL expression.(24)

MLO-Y4 cell support of osteoclast formation was shown to depend on RANKL expression. The ratio of RANKL to OPG mRNA was higher in MLO-Y4 cells than any other cell type examined. RANKL was first cloned by Anderson(36) and Wong(37) during a search for apoptosis regulatory genes and was found predominantly expressed on activated T cells.(37) Suda and coworkers(38, 39) had proposed that a cell membrane factor on osteoblasts and stromal cells was responsible for 1,25(OH)2D3-induced osteoclast formation, which they called osteoclast differentiation factor. On cloning, this factor was found identical to RANKL.(40) OPG, the decoy receptor for RANKL, is identical to the osteoclastogenesis inhibitory factor identified by Tsuda and coworkers.(40) Mice lacking OPG exhibit severe osteoporosis(41, 42) whereas mice overexpressing OPG are severely osteopetrotic.(25) Prostaglandin E2(43) and hydrocortisone(44) decrease, whereas transforming growth factor β increases expression of OPG.(45) Others have compared expression of these two opposing factors and proposed that the balance of expression between these RANKL and OPG determines whether resorption or formation occurs.(34) Future studies will be to examine the regulation of expression and therefore the ratio of RANKL to OPG in MLO-Y4 cells in response to mechanical strain.

M-CSF is required for osteoclast formation. Kodama proposed that M-CSF is necessary for the proliferation and differentiation of osteoclast precursors.(21) Like RANKL, M-CSF is increased in murine osteoblasts on treatment with 1,25(OH)2D3.(32) In these studies, it was observed that MLO-Y4 cells express mRNA for membrane and secreted M-CSF and that these cells secrete very large amounts of this factor that could not be blocked completely by large amounts of neutralizing antibody. These findings raise the possibility that M-CSF could be a secretory factor made by osteocytes to signal and support proliferation of osteoclast precursors. Of interest is the finding of Rubin and coworkers that application of hydrostatic pressure to murine bone marrow will decrease M-CSF and also decrease osteoclast formation.(46)

It has been proposed that damaged or apoptotic osteocytes can send signals to initiate bone resorption.(7) Apoptotic osteocytes are highly localized to regions of microcracks and subsequently are removed by resorbing osteoclasts.(47–49) Intracortical remodeling in the adult skeleton removes and replaces bone that has sustained microdamage. Microdamage is defined as linear microcracks formed in fatigue-loaded bone. Others have shown that the capacity of bone to repair microdamage appears to be dependent on osteocyte viability, not death.(50) These studies suggest that damaged but viable osteocytes may also release signals that lead to rapid repair of microdamage. These studies suggest that viable osteocytes can support osteoclast formation and activation. In the present studies, immunohistochemical staining for RANKL enhanced the visibility of the dendritic processes showing RANKL expression on these processes and suggesting that these processes can support osteoclast formation. It has been shown that metatarsal bone stripped of the periosteum will support osteoclast formation without the addition of other factors.(51–53) This suggests that exposed osteocytes or their exposed dendritic processes are sufficient to support osteoclast formation without the induction of apoptosis.

In previous studies by Tanaka and coworkers,(14) viable isolated chick osteocytes supported osteoclast formation but through the production of unknown soluble factors. These investigators added highly purified chick osteocytes to cultures of mouse spleen hemopoietic blast cells from 5-fluorouracil-treated mice. In the present studies, cell contact was necessary for the formation of osteoclasts, suggesting that the MLO-Y4 cells do not produce soluble RANKL or other soluble factors. MLO-Y4 conditioned media did not have significant effects on osteoclast formation. This could be because of the fact that this cell line may only represent a subset of osteocyte-like cells or that mammalian osteocytes differ from avian osteocytes.

Our studies raise the question as to how a cell located in the bone matrix could influence osteoclast formation on the bone surface. Because the dendritic processes of the MLO-Y4 cells stained positive for RANKL protein, this suggests that osteocytic processes reaching the bone surface could support osteoclast formation and activation. In vivo, RANKL is expressed on actively synthesizing osteoblast cells on the bone surface and on osteocytes near the bone surface but not deep in the mineralized bone.(54) It is not clear under what circumstances viable osteocytes would come into contact with osteoclast precursors. It has been shown that osteocyte dendritic processes come in contact with osteoblasts and lining cells on the bone surface.(8, 55) Lining cells retract to allow osteoclasts access to the bone surface. Under these conditions, the dendritic processes certainly would come into direct contact with early and mature osteoclasts. Recently, Kamioka and colleagues(56) have shown that although osteocytes widely send their processes into the osteoblast layer, some of these processes actually reach the vascular-facing surface of the osteoblast layer. These observations suggest that osteocytes can communicate with cells such as osteoclast precursors in the marrow. Future studies will be performed to determine if only dendritic processes of the MLO-Y4 cells are sufficient for osteoclast formation.

In summary, like authentic osteocytes isolated from chickens, MLO-Y4 cells also support osteoclast formation in the absence of added osteotropic factors. The mechanism responsible is RANKL expression on the surface of the cells and secretion of high concentrations of M-CSF. This contrasts with results previously reported and confirmed in this study that osteoblasts and stromal cells do not support osteoclastogenesis by themselves but must be treated with “factors” that enhance RANKL expression. Another important finding is that RANKL is expressed on the dendritic processes of MLOY4 cells and this provides a potential means for osteocytes buried within bone to interact and stimulate the differentiation of osteoclast precursors at the bone surface. The relative contribution of osteocytes in osteoclastogenesis currently is unknown but these studies suggest that under certain conditions, osteocytes but not osteoblasts or stromal cells are expressing a sufficient amount of RANKL to maintain osteoclastogenesis. The effects of mechanical strain on osteocyte-induced osteoclastogenesis currently are unknown but one could speculate that under conditions of immobilization, lack of gravity, remodeling, and microdamage RANKL/M-CSF would be elevated in osteocytes. This is the underlying hypothesis for numerous future studies.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

We acknowledge the technical assistance of Marie Harris with the Northern blot analysis. We also acknowledge the technical assistance of Fabio Jiminez for the FACS analysis. This work was supported by the National Institutes of Health (NIH) grants AR-46798 and AR42372.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES
  • 1
    Rubinacci A, Villa I, Dondi Benelli F, Borgo E, Ferretti M, Palumbo C, Marotti G 1998 Osteocyte-bone lining cell system at the origin of steady ionic current in damaged amphibian bone. Calcif Tissue Int 63: 331339.
  • 2
    Klein-Nulend J, van der Plas A, Semeins CM, Ajubi NE, Frangos JA, Nijweide PJ, Burger EH 1995 Sensitivity of osteocytes to biomechanical stress in vitro. FASEB J 9: 441445.
  • 3
    Lanyon LE 1992 Control of bone architecture by functional load bearing. J Bone Miner Res 7: S2;S369S375.
  • 4
    Marcus R 1996 Mechanisms of exercise effects on bone. In: BilezikianJP, RaiszLG, RodanGA (eds.) Principles of Bone Biology. Academic Press, San Diego, CA, USA, pp. 11351146.
  • 5
    Turner CH, Forwood MR, Otter MW 1994 Mechanotransduction in bone: Do bone cells act as sensors of fluid flow? FASEB J 8: 875878.
  • 6
    Weinbaum S, Cowin SC, Zeng Y 1994 A model for the excitation of osteocytes by mechanical loading-induced bone fluid shear stresses. J Biomech 27: 339360.
  • 7
    Lanyon LE 1993 Osteocytes, strain detection, bone modeling and remodeling. Calcif Tissue Int 53(Suppl 1):S102S106. discussion S106–7
  • 8
    Doty SB 1981 Morphological evidence of gap junctions between bone cells. Calcif Tissue Int 33: 509512.
  • 9
    Ajubi NE, Klein-Nulend J, Alblas MJ, Burger EH, Nijweide PJ 1999 Signal transduction pathways involved in fluid flow-induced PGE2 production by cultured osteocytes. Am J Physiol 276: E171E178.
  • 10
    Mason DJ, Suva LJ, Genever PG, Patton AJ, Steuckle S, Hillam RA, Skerry TM 1997 Mechanically regulated expression of a neural glutamate transporter in bone: A role for excitatory amino acids as osteotropic agents? Bone 20: 199205.
  • 11
    Martin RB 2000 Toward a unifying theory of bone remodeling. Bone 26: 16.
  • 12
    Aubin JE, Bonnelye E 2000 Osteoprotegerin and its ligand: A new paradigm for regulation of osteoclastogenesis and bone resorption. Medscape Womens Health 5: 5.
  • 13
    Quinn JM, Horwood NJ, Elliott J, Gillespie MT, Martin TJ 2000 Fibroblastic stromal cells express receptor activator of NF-kappa B ligand and support osteoclast differentiation. J Bone Miner Res 15: 1459166.
  • 14
    Tanaka K, Yamaguchi Y, Hakeda Y 1995 Isolated chick osteocytes stimulate formation and bone-resorbing activity of osteoclast-like cells. J Bone Miner Metab 13: 6170.
  • 15
    Noble BS, Reeve J 2000 Osteocyte function, osteocyte death and bone fracture resistance. Mol Cell Endocrinol 159: 713.
  • 16
    Kato Y, Windle JJ, Koop BA, Mundy GR, Bonewald LF 1997 Establishment of an osteocyte-like cell line, MLO-Y4. J Bone Miner Res 12: 20142023.
  • 17
    Bonewald LF, Zhao S, Zhang Y, Harris S 2000 Expression of the osteocyte-specific antigen. E11, in MLO-Y4 cells and mineralizing osteoblasts. J Bone Miner Res 15: S1;S502.
  • 18
    Cheng B, Zhao S, Luo J, Sprague E, Bonewald LF, Jiang JX 2001 Expression of functional gap junctions and regulation by fluid flow in osteocyte-like MLO-Y4 cells. J Bone Miner Res 16: 249259.
  • 19
    Yellowley CE, Li Z, Zhou Z, Jacobs CR, Donahue HJ 2000 Functional gap junctions between osteocytic and osteoblastic cells. J Bone Miner Res 15: 209217.
  • 20
    Plotkin LI, Weinstein RS, Parfitt AM, Roberson PK, Manolagas SC, Bellido T 1999 Prevention of osteocyte and osteoblast apoptosis by bisphosphonates and calcitonin. J Clin Invest 104: 13631374.
  • 21
    Kodama H, Nose M, Niida S, Yamasaki A 1991 Essential role of macrophage colony-stimulating factor in the osteoclast differentiation supported by stromal cells. J Exp Med 173: 12911294.
  • 22
    Takahashi N, Udagawa N, Akatsu T, Tanaka H, Shionome M, Suda T 1991 Role of colony-stimulating factors in osteoclast development. J Bone Miner Res 6: 977985.
  • 23
    Quinn JM, Elliott J, Gillespie MT, Martin TJ 1998 A combination of osteoclast differentiation factor and macrophage-colony stimulating factor is sufficient for both human and mouse osteoclast formation in vitro. Endocrinology 139: 44244427.
  • 24
    Yasuda H, Shima N, Nakagawa N, Yamaguchi K, Kinosaki M, Mochizuki S, Tomoyasu A, Yano K, Goto M, Murakami A, Tsuda E, Morinaga T, Higashio K, Udagawa N, Takahashi N, Suda T 1998 Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci USA 95: 35973602.
  • 25
    Simonet WS, Lacey DL, Dunstan CR, Kelley M, Chang MS, Luthy R, Nguyen HQ, Wooden S, Bennett L, Boone T, Shimamoto G, DeRose M, Elliott R, Colombero A, Tan HL, Trail G, Sullivan J, Davy E, Bucay N, Renshaw-Gegg L, Hughes TM, Hill D, Pattison W, Campbell P, Boyle WJ 1997 Osteoprotegerin: A novel secreted protein involved in the regulation of bone density. Cell 89: 309319.
  • 26
    Udagawa N, Takahashi N, Akatsu T, Sasaki T, Yamaguchi A, Kodama H, Martin TJ, Suda T 1989 The bone marrow-derived stromal cell lines MC3T3-G2/PA6 and ST2 support osteoclast-like cell differentiation in cocultures with mouse spleen cells. Endocrinology 125: 18051813.
  • 27
    Kodama H, Amagai Y, Sudo H, Kasai S, Yamamoto S 1982 Establishment of a clonal osteogenic cell line from newborn mouse calvaria. Japanese Journal of Oral Biology 23: 899893.
  • 28
    Chen D, Chen H, Feng JQ, Windle JJ, Koop BA, Harris MA, Bonewald LF, Boyce BF, Wozney JM, Mundy GR, Harris SE 1995 Osteoblastic cell lines derived from a transgenic mouse containing the osteocalcin promoter driving SV40 T-antigen. Mol Cell Diff 3: 193212.
  • 29
    Ghosh-Choudhury N, Windle JJ, Koop BA, Harris MA, Guerrero DL, Wozney JM, Mundy GR, Harris SE 1996 Immortalized murine osteoblasts derived from BMP 2-T-antigen expressing transgenic mice. Endocrinology 137: 331339.
  • 30
    Oreffo RO, Marshall GJ, Kirchen M, Garcia C, Gallwitz WE, Chavez J, Mundy GR, Bonewald LF 1993 Characterization of a cell line derived from a human giant cell tumor that stimulates osteoclastic bone resorption. Clin Orthop 296: 229241.
  • 31
    Mikuni-Takagaki Y, Kakai Y, Satoyoshi M, Kawano E, Suzuki Y, Kawase T, Saito S 1995 Matrix mineralization and the differentiation of osteocyte-like cells in culture. J Bone Miner Res 10: 231242.
  • 32
    Rubin J, Fan X, Thornton D, Bryant R, Biskobing D 1996 Regulation of murine osteoblast macrophage colony-stimulating factor production by 1,25(OH)2D3. Calcif Tissue Int 59: 291296.
  • 33
    Takahashi N, Akatsu T, Udagawa N, Sasaki T, Yamaguchi A, Moseley JM, Martin TJ, Suda T 1988 Osteoblastic cells are involved in osteoclast formation. Endocrinology 123: 26002602.
  • 34
    Horwood NJ, Elliott J, Martin TJ, Gillespie MT 1998 Osteotropic agents regulate the expression of osteoclast differentiation factor and osteoprotegerin in osteoblastic stromal cells. Endocrinology 139: 47434746.
  • 35
    Nakagawa N, Kinosaki M, Yamaguchi K, Shima N, Yasuda H, Yano K, Morinaga T, Higashio K 1998 RANK is the essential signaling receptor for osteoclast differentiation factor in osteoclastogenesis. Biochem Biophys Res Commun 253: 395400.
  • 36
    Anderson DM, Maraskovsky E, Billingsley WL, Dougall WC, Tometsko ME, Roux ER, Teepe MC, DuBose RF, Cosman D, Galibert L 1997 A homologue of the TNF receptor and its ligand enhance T-cell growth and dendritic-cell function. Nature 390: 175179.
  • 37
    Wong BR, Josien R, Lee SY, Sauter B, Li HL, Steinman RM, Choi Y 1997 TRANCE (tumor necrosis factor [TNF]-related activation-induced cytokine), a new TNF family member predominantly expressed in T cells, is a dendritic cell-specific survival factor. J Exp Med 186: 20752080.
  • 38
    Suda T, Takahashi N, Martin TJ 1992 Modulation of osteoclast differentiation. Endocr Rev 13: 6680.
  • 39
    Suda T, Udagawa N, Nakamura I, Miyaura C, Takahashi N 1995 Modulation of osteoclast differentiation by local factors. Bone 17(Suppl 2):87S91S.
  • 40
    Yasuda H, Shima N, Nakagawa N, Mochizuki SI, Yano K, Fujise N, Sato Y, Goto M, Yamaguchi K, Kuriyama M, Kanno T, Murakami A, Tsuda E, Morinaga T, Higashio K 1998 Identity of osteoclastogenesis inhibitory factor (OCIF) and osteoprotegerin (OPG): A mechanism by which OPG/OCIF inhibits osteoclastogenesis in vitro. Endocrinology 139: 13291337.
  • 41
    Bucay N, Sarosi I, Dunstan CR, Morony S, Tarpley J, Capparelli C, Scully S, Tan HL, Xu W, Lacey DL, Boyle WJ, Simonet WS 1998 osteoprotegerin-deficient mice develop early onset osteoporosis and arterial calcification. Genes Dev 12: 12601268.
  • 42
    Mizuno A, Amizuka N, Irie K, Murakami A, Fujise N, Kanno T, Sato Y, Nakagawa N, Yasuda H, Mochizuki S, Gomibuchi T, Yano K, Shima N, Washida N, Tsuda E, Morinaga T, Higashio K, Ozawa H 1998 Severe osteoporosis in mice lacking osteoclastogenesis inhibitory factor/osteoprotegerin. Biochem Biophys Res Commun 247: 610615.
  • 43
    Brandstrom H, Jonsson KB, Ohlsson C, Vidal O, Ljunghall S, Ljunggren O 1998 Regulation of osteoprotegerin mRNA levels by prostaglandin E2 in human bone marrow stroma cells. Biochem Biophys Res Commun 247: 338341.
  • 44
    Vidal NO, Brandstrom H, Jonsson KB, Ohlsson C 1998 Osteoprotegerin mRNA is expressed in primary human osteoblast-like cells: Down-regulation by glucocorticoids. J Endocrinol 159: 191195.
  • 45
    Takai H, Kanematsu M, Yano K, Tsuda E, Higashio K, Ikeda K, Watanabe K, Yamada Y 1998 Transforming growth factor-beta stimulates the production of osteoprotegerin/osteoclastogenesis inhibitory factor by bone marrow stromal cells. J Biol Chem 273: 2709127096.
  • 46
    Rubin J, Biskobing D, Fan X, Rubin C, McLeod K, Taylor WR 1997 Pressure regulates osteoclast formation and MCSF expression in marrow culture. J Cell Physiol 170: 8187.
  • 47
    Verborgt O, Gibson GJ, Schaffler MB 2000 Loss of osteocyte integrity in association with microdamage and bone remodeling after fatigue in vivo. J Bone Miner Res 15: 6067.
  • 48
    Mori S, Burr DB 1993 Increased intracortical remodeling following fatigue damage. Bone 14: 103109.
  • 49
    Burr DB, Turner CH, Naick P, Forwood MR, Ambrosius W, Hasan MS, Pidaparti R 1998 Does microdamage accumulation affect the mechanical properties of bone? J Biomech 31: 337345.
  • 50
    Tomkinson A, Reeve J, Shaw RW, Noble BS 1997 The death of osteocytes via apoptosis accompanies estrogen withdrawal in human bone. J Clin Endocrinol Metab 82: 31283135.
  • 51
    Burger EH, van der Meer JW, Nijweide PJ 1984 Osteoclast formation from mononuclear phagocytes: role of bone-forming cells. J Cell Biol 99: 19016.
  • 52
    Schneider GB, Relfson M 1988 A bone marrow fraction enriched for granulocyte-macrophage progenitors gives rise to osteoclasts in vitro. Bone 9: 303308.
  • 53
    Braidman IP, Anderson DC 1993 Role of bone matrix in osteoclast recruitment in cultured fetal rat calvariae. J Bone Miner Res 8: 231238.
  • 54
    Kartsogiannis V, Zhou H, Horwood NJ, Thomas RJ, Hards DK, Quinn JM, Niforas P, Ng KW, Martin TJ, Gillespie MT 1999 Localization of RANKL (receptor activator of NF kappa B ligand) mRNA and protein in skeletal and extraskeletal tissues. Bone 25: 525534.
  • 55
    Shapiro F 1997 Variable conformation of GAP junctions linking bone cells: A transmission electron microscopic study of linear, stacked linear, curvilinear, oval, and annular junctions. Calcif Tissue Int 61: 285293.
  • 56
    Kamioka H, Honjo T, Takano-Yamamoto T 2001 A three-dimensional distribution of osteocyte processes revealed by the combination of confocal laser scanning microscopy and differential interference contrast microscopy. Bone 28: 145149.