HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Alnaeeli, M. Articles by Teng, Y.-T. A. Search for Related Content PubMed PubMed Citation Articles by Alnaeeli, M. Articles by Teng, Y.-T. A.

Immune Interactions with CD4⁺ T Cells Promote the Development of Functional Osteoclasts from Murine CD11c⁺ Dendritic Cells¹

Mawadda Alnaeeli^*,, Josef M. Penninger and Yen-Tung Andy Teng^2,*,

^* Department of Microbiology and Immunology, School of Medicine and Dentistry, University of Rochester, Rochester, NY 14620; Department of Microbiology and Immunology, Faculty of Medicine and Dentistry, University of Western Ontario, London, Ontario, Canada; and Institute of Molecular Biotechnology of the Austrian Academy of Sciences, Vienna, Austria

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Dendritic cells (DC) are innate immune effectors and are critically involved in regulating T cell immunity. Osteoclasts (OC) are bone-resorbing cells derived from the monocyte/macrophage lineage in response to receptor activator of NF-B ligand (RANKL). DC and T cells form aggregates in the inflammatory infiltrates at active disease sites in human and in experimental rheumatoid arthritis and periodontitis. We investigated whether DC interactions with T cells in the bone environment can support the development of functional OC. In the present study, we demonstrate that upon proper activation by microbial or protein Ags (namely Actinobacillus actinomycetemcomitans, bovine insulin, and outer membrane protein-1) and during immune interactions with CD4⁺ T cells in vitro, murine BM-derived and splenic CD11c⁺ DC (CD11b^–F4/80^–Ly-6C^–CD31^–) develop into TRAP⁺CT-R⁺cathepsin-k⁺ functional OC in a RANKL/RANK-dependent manner. Rescue and blocking experiments using CD11c⁺ DC derived from Csf-1^–/–op/op mice show that M-CSF is required "before" developing such osteoclastogenic potential upstream of RANKL/RANK signaling, suggesting that immature CD11c⁺ DC can indeed act like OC precursors. In addition, these CD11c⁺ DC-derived OC are capable of inducing bone loss after adoptive transfer in vivo. These data suggest a direct contribution of DC during immune interactions with CD4⁺ T cells to inflammation-induced osteoclastogenesis. Therefore, our findings not only provide further evidence for DC plasticity, but also extend the current paradigm of osteoimmunology.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Inflammatory bone disorders, such as rheumatoid arthritis (RA),³ periodontal disease (PD), and osteomyelitis, manifest imbalanced remodeling processes, resulting in irreversible bone destruction, leading to morbidity, perturbation of life quality, and even potentially life-threatening conditions (1). Bone remodeling is a tightly regulated process by a number of osteogenic cytokines, growth factors, and hormones that exert their effects on bone cells, namely osteoblasts (OB) and osteoclasts (OC). OB and OC function to control bone synthesis and resorption, respectively. OC are derived from the monocytes/macrophages (Mo/MQ) lineage in the presence of receptor activator of NF-B ligand (RANKL; Refs. 1, 2, 3). Functional OC are multinucleated giant cells that express tartrate resistance acid phosphatase (TRAP), calcitonin receptor (CT-R), cathepsin-k, and integrins _v3 and are capable of resorbing bone (4). RANKL induces osteoclastogenesis in the presence of M-CSF (or Csf-1; Ref. 5), where M-CSF signals via c-fms up-regulate RANK expression on OC precursors, promoting their survival and differentiation (6, 7). RANK, the receptor of RANKL, and its antagonist, osteoprotegerin (OPG), have been shown to be the key regulators of bone remodeling and are directly involved in the differentiation, activation, and survival of OC and OC precursors (1, 2, 3, 8). In addition, RANKL/RANK signaling enhances dendritic cell (DC) survival and is indispensable for lymph node organogenesis (9, 10, 11). Genetic mutations of RANKL, RANK and M-CSF demonstrate similar defective phenotypes in OC development with severe osteopetrosis, suggesting that they are essential for osteoclastogenesis and bone remodeling (11, 12, 13, 14). OPG transgenic mice are osteopetrotic with defective OC activity, whereas OPG deficient mice are severely osteoporotic (1, 15). Importantly, our studies and others showed that activated T cells express RANKL and mediate osteoclastogenesis (16, 17), and that blocking RANKL activity via OPG injections results in a significant reduction of bone loss in RA, PD, osteoporosis, cancer-related bone metastasis and type 1 diabetes-associated alveolar bone loss (1, 16, 17, 18, 19, 20).

DC are bone marrow (BM)-derived leukocytes, with key innate immune functions, including antiviral and antimicrobial properties (21). Their wide distribution allows rapid access to sample the environment for Ags and danger signals. As professional APC, DC capture Ags and efficiently prime naive T cells to initiate adaptive immune effector functions or to induce tolerance (21, 22). The origin and developmental pathways of DC remain controversial (23); however, they are currently classified into conventional (including myeloid and lymphoid) and plasmacytoid DC (21, 22, 23, 24, 25, 26, 27). Recent studies reported that: 1) murine DC could be Mo derived (24), suggesting DC and OC may share common progenitors (28) and that Flt3⁺ MQ precursors may differentiate sequentially to OC, DC, and microglia (29); 2) immature DC "transdifferentiate" to OC in RA, suggesting that DC may directly contribute to inflammation-induced osteoclastogenesis (30); and 3) Langerhans cells may give rise to multinucleated TRAP⁺ OC-like cells in the inflammatory Langerhans cell-histiocytosis lesions (31). Furthermore, DC can activate naive T cells in the presence of microbial Ags such as Lactobacillus spp. and Helicobacter pylori (32, 33).

DC are present in active disease sites of RA and PD where they aggregate with T cells in the inflammatory foci (34, 35, 36, 37) and are thought to contribute indirectly to inflammation-induced bone loss in RA (38). Rivollier et al. (30) showed that human blood Mo-derived DC can "transdifferentiate" into OC in the presence of M-CSF and RANKL in vitro, suggesting that DC might directly contribute to osteoclastogenesis. It remains unclear whether DC/T cell interactions support such development and whether DC can indeed directly contribute to inflammation-induced osteoclastogenesis during immune interactions and activation in the bone environment. In addition, the development of OC from committed DC has not yet been described in the murine system.

In the present study, we hypothesized that murine DC carry osteoclastogenic potential and can develop into functional OC during immune interactions with CD4⁺ T cells in the presence of Ag stimulation. We pursued this via in vitro coculture studies using CD11c⁺ DC derived from total BM cultures with GM-CSF and IL-4, followed by MACS sorting to isolate highly purified CD11c⁺CD11b^– DC subset (thus devoid of classical OC precursors; Refs. 4 and 39). Our results show that: 1) through phenotypic and functional analyses, murine CD11c⁺ DC can act like OC precursors and indeed develop into OC during immune interactions with CD4⁺ T cells and microbial or protein Ags in bone environment; 2) M-CSF is essential for the development of such osteoclastogenic potential upstream of RANKL/RANK signaling; and 3) DC-derived OC (DDOC) can induce bone resorption after adoptive transfer onto NOD/SCID mice calvarias in vivo. These data indicate that murine CD11c⁺ DC can develop into bona fide functional OC and that DC/T cells immune interactions can support de novo OC development from local CD11c⁺ DC subset(s).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Mice

Four- to 6-wk-old female BALB/c, B6C3fe Csf-1^–/–op/op, and wild-type (WT) control mice were purchased from The Jackson Laboratory mice and housed under specific pathogen-free conditions in the Animal Facility of the University of Rochester and the University of Western Ontario. All animal protocols were conducted under each institution’s guidelines and approved by the local ethics and animal experimentation committees.

Cell cultures and reagents

All primary cell cultures were performed in complete RPMI 1640 medium supplemented with 10% heat-inactivated FBS (Invitrogen Life Technologies), 50 µM 2-ME, 100 µg/ml streptomycin, and 100 U/ml penicillin. Cells were incubated at 37°C in a humidified 5% CO₂ incubator (17). The following reagents were purchased from commercial sources: recombinant mouse (m) GM-CSF (Cedarlane Laboratories), anti-mFcRIII/II (CD16/32), PE-conjugated, anti-mCD4, FITC-conjugated anti-mCD11c, PerCPCy5.5-conjugated anti-mCD4, rat anti-mI-A^d, biotin anti-mCD31 (ER-MP12), biotin anti-mLy-6C (ER-MP20), rat anti-mTCR -chain, rat anti-mCD3Abs, rmIL-4, and Con A were all from BD Pharmingen; FITC-conjugated goat anti-hFc- (The Jackson Laboratory); magnetic beads-conjugated anti-mCD11c (Miltenyi Biotec); anti-mM-CSF Ab (R&D Systems); anti-mTRAP Ab (Zymed Laboratories); rabbit anti-mRANK, rabbit anti-mGM-CSFR, goat anti-mCT-R, Texas Red-conjugated bovine anti-goat IgG, and goat anti-murine cathepsin-k Abs (Santa Cruz Biotechnology); rabbit and goat isotype-matched IgGs, rmM-CSF, FITC-conjugated goat anti-human Fc-specific IgG, BSA fraction V (BSA), and LPS from Escherichia coli 0128:B12 (Sigma-Aldrich); biotinylated goat anti-rabbit IgG, streptavidin-7-amino-4-methylcoumarin-3-acetic acid (AMCA), and streptavidin-PE (Vector Laboratories); anti-mCD11b-Cy5, anti-m-integrin ₃ (CD61)-FITC and biotinylated anti-m-integrin _v (CD51), biotinylated goat anti-Armenian hamster, biotin anti-mF4/80, and biotinylated anti-mCD25 Abs (eBioscience); and Alexa Fluor 488 phalloidin (Molecular Probes). Recombinant mRANKL and OPG-Fc fusion proteins were prepared as described previously (11, 17). Quick cell proliferation assay kit (WST-1; BioVision).

Generation of BM-derived CD11c⁺ DC and CD11c⁺DC^–CD4⁺ T cells cocultures

CD11c⁺DC were generated from BM of BALB/c, B6C3fe Csf-1^–/–op/op, and WT control mice using a previously described protocol (40) with modification. Briefly, total BM cells freshly isolated from long bones were cultured with 20 ng/ml rmGM-CSF and 10 ng/ml rmIL-4 to generate DC. On day 7, CD11c⁺ DC were purified by using magnetic bead-conjugated mCD11c mAb and MACS device (Miltenyi Biotec). The resulting population is highly pure (CD11c⁺ DC purity is >99%; see Fig. 1C), typically yielding 10–15% of the total numbers of BM-derived cells from 1-wk culture with GM-CSF and IL-4. The resulting CD11c⁺ DC are CD11b^– (see Fig. 3B), different from the heterogeneous DC population (mainly CD11b⁺) generated using only GM-CSF in the total BM cultures (39, 40). In some experiments, CD11c⁺ DC were generated after depleting Mo and MQ via Ab-dependent complement-mediated lysis. Briefly, BM cells were first incubated with Abs against Mo markers Ly6C (ER-MP20) and CD31 (ER-MP12), and MQ markers F4/80 and CD11b, 3–5 µg/ml each, for 30 min followed by the addition of rabbit complement and incubation at 37°C for 30 min (17, 41, 42). After lysis, remaining Mo/MQ-depleted BM cells were resuspended in conditioned medium with mGM-CSF and mIL-4 for 7 days, and subsequently, CD11c⁺ DC were MACS purified.

View larger version (51K): [in this window] [in a new window]   FIGURE 1. BM-derived and MACS-purified CD11c+ cells are bona fide DC. A, Freshly purified BM-derived CD11c+ DC induced proliferation of allogeneic CD4+ T cells in a MLR, and (B) activation of syngeneic naive CD4+ T cells as shown by the expression of CD25 (red) and RANKL (green) expression (orange and white arrows). C, Immunostaining of freshly purified BM-derived CD11c+ DC. FACS dot plots depict the percentage of DC expressing surface CD11c, GM-CSFR, RANK, MHC-II, CT-R, and intracellular cathepsin-k before their coculture. Table, Expression profiles of CD11c+ DC on day 0 and 12 h postcocultures compared with DC+LPS; surface MHC-II (sMHC-II) and intracellularly MHC-II (iMHC-II) indicate surface vs intracellular MHC-II levels, respectively.

View larger version (29K): [in this window] [in a new window]   FIGURE 3. CD11c+ DC can develop into TRAP+ cells with resorptive activity in response to various Ag stimulations. BM-derived CD11c+ DC (105) were cocultured with syngeneic CD4+ T cells in the presence of Aa (10 µg/ml), OMP-1, or BI (20 µg/ml). On day 5, TRAP and resorptive activities were quantified. Note that T cells in DC+T+Aa cocultures were purified from naive mice, while those in DC+T+Omp-1 or BI cultures were purified from Omp-1/CFA and BI/CFA-primed BALB/c mice, respectively. A, Aa, OMP-1, and BI induced similar levels of TRAP and resorptive activities; total number and total surface area of TRAP+ multinucleated cells per unit area (top panel) and resorptive pits (bottom panel) in Aa, OMP-1, and BI cocultures (p > 0.05). B, Freshly isolated CD11c+ DC were stained for CD11b, F4/80, ER-MP12, and ER-MP20 to exclude Mo/MQ contamination. C and D, The levels of TRAP and resorptive pits activities by DDOC, generated from BM with or without Mo/MQ depletion, in CD4+ T cell+Aa cocultures (p 0.5 and p 0.4, respectively). Representative data from three independent experiments are shown.

Mixed lymphocyte cultures

Naive 3 x 10⁵ CD4⁺ T cells purified from C57BL/6 splenocytes were cocultured, in triplicates, with different densities (3 x 10⁵, 3 x 10⁴, 3 x 10³, 3 x 10², and 3 x 10¹) of freshly purified CD11c⁺ DC derived from BM of BALB/c mice as described above in a flat-bottom 96-well plate in a total volume of 100 µl. Plate-bound anti-CD3 Ab (10 µg/ml)-activated CD4⁺ T cells were used as positive control for T cell proliferation, whereas untreated CD4⁺ T cells were used as a negative control. At day 5, 50 µl of WST-1 reagent was added into each well according to the manufacturer’s instructions for colorimetric measurement of cellular proliferation (44). This assay is based on the cleavage of tetrazolium salt WST-1 to formazan by mitochondrial dehydrogenase. OD value was determined 4 h later at 440 nm using Dynatech MR700 Microplate reader (Dynatech).

Quantitation of TRAP⁺ cells and resorptive pits

Cocultured CD11c⁺ DC were fixed by 2% formaldehyde + 0.2% glutaraldehyde in PBS (pH 7.3) for 45 min. Then, cells were incubated with TRAP staining solution (0.2 sodium acetate buffer, 40 mM sodium tartrate, 1.2 mM naphthol AS-MX phosphate, and 1.3 mM fast red violet Luria-Bertani salt). To quantify TRAP signals, images were captured digitally under x400 magnification using a Leica inverted microscope IRBE-DM via a motorized staging facility equipped with a high-resolution Hamamatsu-Orca digital camera (20). Both total numbers and total surface area of (purple-red) TRAP⁺ multinucleated cells (more than or equal to three nuclei) were quantified as described previously (20). Briefly, 12–17 randomly chosen fields (covering 33 mm² = one-third of the total surface areas of the HA well) were used for automated scanning analysis and quantification. The mean of total number TRAP⁺ multinucleated cells per unit area and total surface area of TRAP⁺ multinucleated cells were then calculated after subtracting averaged background signals of the negative control. To quantify the total surface area of resorptive pits in the selected fields, cells were stripped with 1 N NaOH for 16 h, after which the images of eroded bone or HA surfaces were captured as described above.

FACS analysis and scanning IF digital images analysis (SIDIA)

Cells collected from cocultures at various time points were incubated with 5 µg/ml anti-CD16/32 to block Ig binding, then immunostained with isotype control IgG or Ab-conjugated to FITC, PE, Cy5, or biotin as follows: cells were stained with FITC-anti-mCD11c, anti-mGM-CSFR, and anti-mRANK Ab, followed by incubation with 2° biotinylated anti-rabbit-IgG or 3° streptavidin-PE or cy5; anti-mCT-R IgG, followed by 2° FITC-conjugated anti-goat IgG; anti-mI-A^d, followed by 2° biotin goat anti-mIgG and 3° streptavidin-Cy5. For intracellular staining of cathepsin-k, cells were fixed for 30 min with fresh 4% formaldehyde and then permeabilized for 10 min with 0.2% Triton X-100, followed by blocking nonspecific binding by 1% BSA and incubation with anti-m-cathepsin-k IgG and 2° FITC-conjugated anti-goat IgG on ice. Stained cells were scanned in a FACSCalibur flow cytometer (BD Biosciences). Results were analyzed via CellQuest, and dead cells were excluded by propidium iodide staining. CD4⁺ T cell activation was assessed by in situ staining of CD25 with 1° biotinylated anti-mCD25, 2° streptavidin-Cy5, and RANKL with OPG-Fc + FITC-conjugated anti-hFc-IgG. To detect F-actin ring formation, DC+T+Aa coculture of day 5 was intracellularly stained in situ by Alexa-Fluor 488 phalloidin, according to the manufacturer’s instruction, while 0.1 x 10⁶ peritoneal MQ cultured in 25 ng/ml rM-CSF and 30 ng/ml rRANKL for 5 days served as a positive control. For quantitation of IF signals at the single cell level, 0.25 x 10⁶ CD11c⁺ DC were labeled by CFSE before cocultures and then at different time points were stained with: 1) anti-mCT-R and 2° Texas red-conjugated anti-goat IgG and anti-GM-CSFR using biotinylated IgG streptavidin-AMCA as 2° and 3° Abs; and 2) CT-R (Texas Red) and anti-RANK using biotinylated IgG streptavidin-AMCA as 2° and 3° Abs. Stained DC were then placed in 96-well flat-bottom plates for SIDIA. IF images were captured for a series of 12–17 random fields per well, after which the mean fluorescence intensity at single cell level was determined as previously described excluding all CFSE^– cells (<50 pixels; Ref. 20). The sum of FI was calculated by adding the total FI values per well after subtracting the background signals. The results are shown as the mean ± SEM of FI by at least 0.1 x 10⁶ countable cells from three independent experiments.

Adoptive transfer of CD11c⁺ DC onto mouse calvarias and bone resorption

Purified CD11c⁺ DC were stained with CFSE before coculture with CD4⁺ T cells and Aa. On day 2.5, DC were collected and stained with anti-mCD11c mAb plus biotin-conjugated anti-hamster IgG and streptavidin-conjugated Cy5. Then, 2 x 10⁶ CD11c⁺ DC (100 µl of PBS) were injected onto the calvarias of 5- to 6-wk-old female NOD/SCID mice (n = 3). NOD/SCID mice receiving PBS only, 2 x 10⁶ CD11c⁺ DC with or without Aa, or CD4⁺ T cells with or without Aa (n = 3) served as negative controls. All mice were euthanized 4 days postinjection. The injected cells were detected in vivo on mouse skulls when scanned using two sets of fluorescent filters to detect CFSE (465-nm excitation/545-nm emission) and Cy5 (620 nm/670 nm) via a Kodak image station 2000MM. Mouse skulls were fixed in 10% formalin, decalcified in Cal-EX, and paraffin embedded to prepare 4- to 6-µm thick sections for histology and immunohistochemistry (TRAP and CD11c detection; Refs. 17 and 45). Fluorescent microscopy was used to visualize CFSE-labeled DC in histological sections.

Statistical analysis

Statistical analysis was performed using the two-sided Student t test. Differences between groups were considered statistically significant with >95% confidence when p value was <0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Characterization of BM-derived CD11c⁺ DC

We combined BM-derived DC generation (by GM-CSF and IL-4) with MACS cell-sorting protocols to produce highly purified CD11c⁺ DC. We characterized the Ag-presenting capability of freshly purified CD11c⁺ DC and found that they can efficiently induce: 1) proliferation of allogeneic naive CD4⁺ T cells from C57BL/6 (H-2^b) in mixed lymphocytes cultures (Fig. 1A); and 2) activation of naive syngeneic CD4⁺ T cells with Aa-Ags as determined by CD25 and RANKL expression (Fig. 1B) and ELISA using culture supernatants (data not shown). Furthermore, freshly purified CD11c⁺ DC expressed no detectable CD80/CD86 and low cell surface MHC-II (sMHC-II: 1–15%) with the majority of MHC-II localized intracellularly (iMHC-II = 40–60%; day 0 in Fig. 1C, table), indicating their immature phenotype. Based on FACS analyses, day 0 CD11c⁺ DC exhibited the following phenotype: GM-CSFR⁺RANK⁺MHC-II^lowCT-R^–cathepsin-k^– (Fig. 1C). Importantly, these DC quickly up-regulated surface MHC-II and the costimulatory molecules CD80 and CD86 when activated by sonicated Aa-Ags or E. coli-LPS (Fig. 1C, table). Collectively, these data indicate that CD11c⁺ cells used in our study are indeed bona fide DC.

CD11c⁺ DC develop TRAP and bone resorptive activities during cocultures with CD4⁺ T cells and foreign Ags in a RANKL-dependent manner

To explore the osteoclastogenic potential of DC, BALB/c BM-derived MACS-purified CD11c⁺DC were incubated with naive syngeneic CD4⁺ T cells and sonicated Ags of Aa, a key human periodontal pathogen (17, 20) on dentine slices, bone discs (46), or HA-coated 48-well plates for 2–10 days. Total splenocytes plus Con A cocultures were used as positive control for TRAP and resorptive activity (43); in addition, freshly prepared TRAP⁺ multinucleated BM cells were used to validate the positive control (Fig. 2C). Interestingly, CD11c⁺ DC were able to fuse and become TRAP⁺ multinucleated giant cells, some of which manifested dendrites and maintained interaction with T cells (Fig. 2G). Since T cells do not express TRAP as determined by TRAP staining of CD4⁺ T cells cocultured with and without Aa or Con A, all TRAP⁺ cells must be DC derived. TRAP activity on day 5 was assessed by the total number and the total surface area of TRAP⁺ cells (Fig. 2A). Ninety to 95% of TRAP⁺ cells included in the quantitative analysis were multinucleated (more than or equal to three nuclei). Note that titration of cells to as low as 5 x 10⁴ CD11c⁺ DC and 5 x 10⁴ CD4⁺ T cells also yielded significant TRAP and resorptive pit activities, compared with negative controls (data not shown). Furthermore, TRAP expression peaked on day 5 and required stimulation by Aa, as DC cultured alone, DC plus Aa, or DC plus T cells without Aa, did not develop significant TRAP activity (p 0.03; Fig. 2A). Surprisingly, as early as day 2, total surface area and numbers of TRAP⁺ cells were already significantly higher than control samples (data not shown).

View larger version (49K): [in this window] [in a new window]   FIGURE 2. CD11c+DC develop TRAP and resorptive activities in a RANKL-dependent manner during cocultures with CD4+ T cells and Aa-Ags in vitro. A total of 0.5 x 106 BALB/c BM-derived MACS-purified CD11c+ DC was cultured with Aa (10 µg/ml) sonicated Ags with or without 0.5 x 106 syngeneic CD4+ T cells or sRANKL (30 ng/ml). On day 5, cells were fixed and stained for TRAP. Note that 90–95% of TRAP+ cells included in the quantification are multinucleated (average three to five nuclei; data not shown). A, DC expressed significantly higher levels of TRAP (p 0.03) after coculture with naive syngeneic CD4+ T cells and Aa when compared with negative controls. B, TRAP expression was abolished by OPG (p 0.01) and anti-RANK Ab (p 0.01) compared with DC+T cells+Aa and significantly increased in DC+RANKL (30 ng/ml) + Aa cultures (p 0.008) compared with DC+RANKL alone. C, TRAP expression was reduced significantly when DC/T cell interaction was blocked by anti-MHC-II Ab (p 0.008) or anti-TCR/CD3 Abs (p 0.02) compared with DC+T cells+Aa. To assess OC function, cells were stripped with 1 N NaOH on day 5, and the total surface area of HA resorption was quantified. D, DDOC induced significant bone resorption in T cells+Aa coculture compared with negative controls (p 0.03). E, Both OPG and anti-RANK Ab blocked resorptive activity (p 0.02 and p 0.007, respectively); RANKL restored this ability in the presence of Aa stimulation (p 0.01) compared with DC+RANKL alone. F, Blocking DC/T cell interaction by anti-MHC-II Ab or anti-TCR/CD3 Abs abolished resorptive pits activity (p 0.009 and p 0.01, respectively) compared with DC+T cells+Aa. G, Representative images of TRAP activities from DC+T cells+Aa, DC+ RANKL+Aa, and DC+T cells+Aa+OPG cocultures and their resorptive pits on HA at day 5 under different magnifications (x100, x150, and x400). Bottom panel, Multinucleated DDOC at day 5 of DC+T+Aa coculture (two panels in bottom left corner; x400) and DDOC in "resorptive tracks" on bone discs (two panels in bottom right corner; x150), respectively. H, Representative fluorescent images of F-actin ring formation by DDOC at day 5 of DC+T+Aa coculture (left) and MQ+M-CSF (25 ng/ml) + RANKL (30 ng/ml) as positive control (right) detected by Alexa-fluor 488 phalloidin (under x400). I, BM-derived and MACS-purified splenic CD11c+ DC express comparable levels of TRAP, indicating their osteoclastogenic potential. The results are shown as the total number per unit area and total surface area (mm2) of TRAP+ multinucleated cells or resorptive pits from 12 to 17 randomly selected fields per well (mean ± SEM) from five independent experiments using total splenocytes + Con A as the positive control. The results of splenic CD11c+ DC resorptive pits assay mirrored the TRAP (data not shown here).

Regardless of the bone substrates used (i.e., HA, bone discs, and dentine slices), we found that DDOC induced significant resorptive pits compared with controls. However, DDOC appeared to form localized resorption pits on HA-coated wells rather than the migratory resorption tracks typical for dentine and bone discs (i.e., bone disc in Fig. 2G). In addition, both TRAP and resorptive activities were significantly higher in DC+T+Aa cocultures as compared with controls, regardless of DC:T cells ratios used (from 1:1–10; data not shown). It is noteworthy that there was no significant difference regarding DC and T cells viability (see Table I), as determined by propidium iodide staining where there was no significant cell death, suggesting that phenotypic and functional changes on CD11c⁺ DC are not associated with potential phagocytosis of apoptotic cells in the cocultures. Furthermore, actin-ring formation characteristic of OC activity was detected in DDOC using phalloidin staining (Fig. 2H). Interestingly, the results of parallel analyses of the MACS-purified splenic CD11c⁺ DC showed that they also developed into TRAP⁺ multinucleated cells (Fig. 2I), supporting the notion that the OC differentiation potential studied here is applicable to other CD11c⁺ DC, not just those derived from BM cultures containing GM-CSF.

Coculturing CD11c⁺ DC with T cells and Con A resulted in significant TRAP and resorptive activity (data not shown). Thus, to further confirm the role of T cell activation in the development of DC into OC, we tested the effects of anti-MHC-II (I-A^d) or -CD3/TCR blocking Abs in these cocultures. Results showed that both Abs were able to abolish TRAP (p = 0.008 and p = 0.02; Fig. 2C) as well as resorptive activities (p = 0.009 and p = 0.01; Fig. 2F). Furthermore, at least 25–50% of TRAP⁺ multinucleated DDOC/field had direct contact with T cells at the time of analyses (i.e., day 5; data not shown), which is consistent with the above results. Taken together, our data indicate that T cell activation is critically involved in the development of TRAP and resorptive activities by CD11c⁺ DC via RANKL production.

OC development from CD11c⁺ DC is not Aa specific

To assess whether this phenomenon is Aa specific, we cocultured CD11c⁺ DC with CD4⁺ T cells (primed with BI or OMP-1; Refs. 41 and 47) and protein Ags BI or OMP-1 (20 µg/ml). The results showed that, although TRAP activity appears to be slightly enhanced in OMP-1 and BI cocultures compared with Aa (Fig. 3A), both TRAP and resorptive activities were comparable to those detected in DC+T cells+Aa cocultures (Fig. 3A). Moreover, when sonicated Ags of Porphyromonas gingivalis (17) or E. coli were used they induced TRAP and resorptive activities comparable to those induced by Aa in this coculture system (data not shown). Taken together, our data suggest that this process is not unique to Aa but can be induced by different Ags.

We next evaluated any possible presence of Mo/MQ contaminants to further validate our findings. By FACS, we confirmed that MACS-purified CD11C⁺ DC are CD11b^–F4/80^–ER-MP12^–ER-MP20^– (Fig. 3B; note that ER-MP12 = CD31 and ER-MP20 = Ly-6C), excluding the presence of: 1) Mo/MQ contaminants; and 2) CD11b⁺ cells (i.e., OC precursors) (4). This is likely due to the use of IL-4 in BM cultures, which in turn inhibit Mo/MQ development (39). To further assess the possible contribution of Mo/MQ to the generation of CD11c⁺ DC studied here, mature Mo/MQ were depleted from BM cells, followed by CD11c⁺ DC purification by using anti-F4/80, CD11b, ER-MP12, and ER-MP20 mAbs and complements (41, 42). As expected, CD11c⁺ DC derived from Mo/MQ-depleted or nondepleted BM cultures expressed similar levels of TRAP and resorptive activities by day 5 (Fig. 3, C and D). Thus, CD11c⁺ DC are capable of developing into functional OC regardless of whether they are derived from Mo/MQ. Furthermore, the frequency of TRAP⁺ bone resorbing DDOC was the same whether DC were generated from Mo/MQ-depleted or nondepleted BM (data not shown). It is unlikely that the purified CD11c⁺ DC studied here contain DC/OC progenitors or early Mo precursors because: 1) addition of IL-4 and GM-CSF suppresses Mo/MQ and promotes DC development in BM cultures, respectively (39); and 2) CD11c is not expressed on early DC/OC progenitors (48, 49). These data indicate that contamination by Mo/MQ cannot be responsible for the reported phenomenon; however, this possibility cannot be entirely excluded.

Activated CD11c⁺ DC express functional OC phenotype

The results of TRAP and resorptive pit assays prompted us to further characterize DDOC phenotype. Since CT-R signifies OC function, we used anti-CT-R Ab to detect CT-R expression on DDOC, which has been shown to resemble CT-R detection via radiolabeled CT-R (50). We also assessed the expression of other OC and DC surface markers on these cells. Our results show that by day 3, a significant percentage of CD11c⁺ cells up-regulated the expression of CT-R (<158%). In contrast, GM-CSFR expression was down-regulated (9321%). At day 5 of the coculture, CT-R expression peaked (93–96%), and GM-CSFR expression leveled off to 25% (Fig. 4A). Thus, during activation and RANKL-RANK signaling, CD11c⁺ DC developed into multinucleated TRAP⁺, CT-R⁺, and cathepsin-k⁺ OC with resorptive activity, whose overall expression profile on days 3 and 5 is as follows: CD11c⁺GMCSF-R^{(low to –)}RANK⁺MHC-II^highCT-R⁺, cathepsin-k⁺, and DEC205^{(low or –)}CD4^–CD8^{(+ or –)}CD11b^{(low to –)}F4/80^{(+ or –)}CD80^{(low to –)}CD86^{(low to –)} integrins _V⁺₃⁺ (Fig. 4A and data not shown). Importantly, OPG addition at the onset of the coculture significantly abolished CT-R expression (<10% by day 5; Fig. 4A), which correlated with the significantly reduced TRAP and resorptive activities detected (Fig. 2, B, E, and G). Furthermore, analysis of CT-R⁺ DDOC revealed that the majority were CD11c⁺RANK⁺MHC-II⁺GM-CSFR^{(low or –)} (Fig. 4B). Note that a relatively high MHC-II expression by DDOC is interesting and distinct from the classical OC phenotype (4). Whether this unique feature is associated with maintaining their APC function to prime CD4⁺ T cells requires further study. On the other hand, GM-CSFR levels were down-regulated regardless of OPG treatment, suggesting that RANKL-RANK signaling may not influence GM-CSFR expression. These data illustrate OC development from activated CD11c⁺ DC and further confirm the critical role of RANKL-RANK/OPG in this process.

View larger version (28K): [in this window] [in a new window]   FIGURE 4. Activated CD11c+ DC develop functional OC phenotype during coculture with CD4+T cells and Aa. CD11c+ DC were collected after 3 and 5 days cocultures with or without OPG (30 ng/ml) and were stained to characterize their cell surface phenotypes. A, Dot plots depict the percentage of DC expressing surface CD11c, GM-CSFR, RANK, MHC-II, CT-R, and intracellular cathepsin-k on days 3 and 5 ± OPG. B, Summary of the cell surface phenotype of CT-R+ DDOC costained for CD11c, GM-CSFR, RANK, and MHC-II. C, SIDIA of CD11c+ DC (day 0) and DDOC (days 3 and 5) fluorescently stained for CT-R and RANK or CT-R and GM-CSFR surface expression levels are shown as mean fluorescence intensity in pixels ± SEM at the single cell level. D, Representative immunofluorescent images (under x400) of single DDOC at days 0, 3, and 5 ± OPG of DC+T+Aa and stained for CT-R (Texas red), GM-CSFR (AMCA), and RANK (AMCA). Representative data out of three independent experiments with similar results are shown here.

M-CSF is required for OC development from CD11c⁺ DC

M-CSF is important for the development of OC precursors as Csf-1^–/–op/op mice manifest severe osteopetrosis (6, 7, 13, 14). Moreover, M-CSF is expressed by activated CD4⁺ T cells (50) and can be detected in our DC+T+Aa culture supernatants (data not shown). It was shown previously that osteoclastogenesis is independent of M-CSF produced by activated T cells in vitro (51). To address whether M-CSF is involved in the process of DDOC development, 10 µg/ml anti-M-CSF Ab was added at the onset of coculture. The results showed that there was no significant change in either TRAP or resorptive activity compared with that of DC+T+Aa (p > 0.5; Fig. 5A), suggesting that M-CSF may not be directly involved in the development of DDOC in the cocultures.

View larger version (26K): [in this window] [in a new window]   FIGURE 5. M-CSF is not required for OC development but is essential for the acquisition of CD11c+ DC potential to act like OC precursors capable of further development into functional OC. WT BM-derived CD11c+ DC (0.5 x 106) were cocultured with CD4+ T cells+Aa (10 µg/ml) in the presence of anti-M-CSF Ab (10 µg/ml) (A). TRAP+ and resorptive pits activities by WT DDOC remained unchanged after the addition of anti-M-CSF Ab. Csf-1–/–CD11c+ DC (105) were cocultured with WT CD4+ T cells or sRANKL (30 ng/ml) and Aa with or without exogenous rmM-CSF (25 ng/ml) for 5 days using WT CD11c+ DC cocultures as positive control. B and C, The addition of M-CSF to Csf-1–/–op/op CD11c+ DC cocultures, or together with RANKL+Aa, significantly restored TRAP (B) and resorptive pit (C) activities by day 5 as measured by total number and surface area of TRAP+ multinucleated cells and resorptive pit in mm2 (n = 3). D, At the clonal level, there was no significant difference of CT-R, GM-CSFR, and RANK expression in Csf-1–/–op/op DDOC cocultures with M-CSF on day 5, compared with WT DDOC (p 0.2).

To confirm the contribution of M-CSF during CD11c⁺ DC’s development into OC, we cocultured WT CD11c⁺ DC with Csf-1^–/–op/op-derived CD4⁺ T cells and Aa. Interestingly, WT CD11c⁺ DC were capable of developing into TRAP⁺ functional OC under these conditions (data not shown). Csf-1^–/–op/op CD11c⁺ DC, on the other hand, did not develop TRAP activity or resorptive pits unless exogenous M-CSF was added in their coculture with T cells or sRANKL and Aa (Fig. 5, B and C). These findings are consistent with the current understanding that RANKL is essential for the differentiation and activation of bona fide OC, whereas M-CSF is required for the development and survival of OC precursors (1, 2, 3, 4, 6, 13, 15).

CD11c⁺ DC-derived OC can induce bone resorption in vivo

To assess whether DDOC can induce bone resorption in vivo, not just in vitro, CFSE-labeled CD11c⁺ DC were cocultured in vitro as described above for 2.5 to 3 days before injection onto the periosteum of NOD/SCID mice calvarias. This protocol allowed the in situ detection of labeled cells postinjection. The results of histological and immunohistochemical analyses revealed that injection of PBS-only, CD4⁺ T cells alone or CD11c⁺ DC alone (Aa stimulated or not) did not induce bone resorption in vivo (Fig. 6, A, B, and F). In contrast, CFSE-labeled CD11c⁺ DC from DC+T+Aa expressed TRAP activity postadoptive transfer (Fig. 6, C, D, E, G, H, and I) and induced morbidly higher bone resorption than the control groups (see the eroded bone surfaces highlighted on the histological sections of Fig. 6, F and G). Moreover, these cells maintained CD11c expression in vivo. This finding suggests that, after proper stimulation, CD11c⁺ DC can develop into activated OC capable of resorbing bone in vivo. However, whether DDOC represent a physiological process of osteoclastogenesis in vivo is unclear and awaits further study.

View larger version (115K): [in this window] [in a new window]   FIGURE 6. DDOC are capable of inducing bone resorption in vivo. A total of 2 x 106 CFSE- and Cy5-labeled WT CD11c+ DC was cocultured with CD4+ T cells+Aa for 2–3 days, then cells were collected and injected onto NOD/SCID mice calvarias. Controls included PBS injection and injection of 2 x 106 viable CD4+ T cells or CD11c+ DC with or without Aa (n = 3 mice/group). All mice were sacrificed 4 days postinjection, and paraffin-embedded tissue sections (4–6 µm) were used for histology and immunohistochemistry. The results of one of three experiments with similar results are shown. A and B, H&E staining of control mice receiving PBS or DC+Aa, respectively (* indicates CFSE+CD11c+ DC injected in vivo; x100–400 magnification). Note: Smooth calvarial bone surface with no erosion or resorption detected. Injection of control CD4+ T cells alone did not yield any significant bone resorption (data not shown). C, H&E staining of mice calvarias exhibiting bone resorption after injection of in vitro-generated DDOC (x100). D, Serial sections of the calvarias postinjection (G, x400) showing cellular infiltration with bone resorption. E, Higher power of the bone resorptive area (D, x600) showing OC-like cells in lacunae on calvarial bone surface. F, CFSE+CD11c+ DC (cocultured with Aa only) residing on "smooth" calvarial bone surface postinjection in vivo (x400). G, CFSE+ DDOC residing in a large "resorbed and eroded" area on calvarial bone surface (the same proximal section of D: x400). H and I, Immunohistochemical staining of TRAP expression on DDOC (under x600) in serial sections of calvarias shown in G. The white arrows linking G to H indicate TRAP+ cells derived from the cells in areas that were CD11c+ (CFSE+) DDOC. I, Four representative areas of the adjacent calvarial tissues harbor additional multinucleated bone-resorbing cells derived from CFSE+CD11c+ pool stained positive for TRAP.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

We characterized DDOC phenotype by FACS, and the results were consistent with those of IF at single cell level by SIDIA, suggesting that FACS can be applied to study the phenotypic changes of DDOC. The results from M-CSF blocking, coculturing WT DC with csf-1^–/–op/op-derived T cells and Aa (data not shown), and the M-CSF rescue studies indicate a stage-specific requirement of M-CSF "before" their transition to OC upstream of RANKL-RANK signaling. Exposure of OC precursors to M-CSF has been shown to be essential for the development of their osteoclastogenic potential possibly by up-regulating RANK expression (1, 6). Alternatively, one may suspect the presence of few contaminating precursors among WT DC that are absent in Csf-1^–/–op/op DC cultures. The CD11c⁺ DC subset studied here is highly pure since: 1) it has been shown that CD11c is not expressed on early precursors (49); and 2) the presence of both MQ (CD11b⁺ and F4/80⁺) and Mo (ER-MP12⁺ and ER-MP20⁺) contaminants were excluded (Fig. 3B). Thus, this alternative possibility is unlikely.

In contrast to the proposed model by Miyamoto et al. (28), where "committed" DC derived from common DC/OC progenitors lack the ability to develop into OC, our data demonstrate CD11c⁺ DC development into functional multinucleated TRAP⁺CT-R⁺cathepsin-k⁺MHC-II⁺ OC, whereby they acquire the ability to induce bone resorption not only in vitro but also in vivo (Figs. 2 and 6), suggesting that they posses certain developmental plasticity and may represent an alternative OC differentiation pathway. The molecular mechanisms responsible for such development appear to involve primarily RANKL/RANK signaling in addition to Ag activation and possibly stimulation via TLR. Various effects of LPS on osteoclastogenesis in the presence or absence of OB and stromal cells have been reported, although the precise effects remain unclear (52, 53, 54). By using BM-derived Mo, it has been shown recently that LPS stimulation of OC precursors could result in inhibition or enhancement of osteoclastogenesis depending on their differentiation stage (55). In the present study, Aa-Ags and OMP-1, likely containing traces of LPS, and LPS-free BI protein all yield comparable TRAP and resorptive activities (Fig. 3A). Second, there was no detectable TRAP or resorptive activities in CD11c⁺ DC when cocultured with E. coli-derived LPS in the presence or absence of sRANKL (data not shown). These findings argue against a direct role for LPS in OC development in our system, in agreement with the report by Takami et al. (56), where LPS directly inhibits osteoclastogenesis from OC precursors. Studies are currently underway to examine the role of TLR on osteoclastogenesis using MyD88^–/–CD11c⁺ DC in our coculture system.

The cellular signaling pathways of RANKL/RANK in OC and/or OC precursors involve TNFR-associated factors, p38 MAPK, ERK, and JNK, and downstream transcription factors such as NF-B and c-Fos (57). However, it remains unclear whether the same pathways are also involved in OC development from CD11c⁺ DC. Moreover, multiple Ig-like receptors associated with ITAM-harboring adaptors, FcR common -chain and DNAX-activating protein 12, have been implicated recently in osteoclastogenesis, which further strengthens the link between the immune system and bone remodeling (58). Whether these factors play any role in the development of DDOC remains to be explored. Genetic analyses are being pursued to study the signaling molecules in CD11c⁺ DC responsible for the induction of osteoclastogenesis leading to OC development.

Rivollier et al. (30) recently demonstrated that human circulating blood Mo-derived DC were capable of transdifferentiating into bone resorbing OC in the presence of M-CSF and RANKL in vitro. Our study clearly indicates that interactions of murine BM-derived and splenic CD11c⁺ DC with CD4⁺ T cells in the bone environment can promote DDOC development in response to microbial or protein Ags and RANKL-RANK signaling (Fig. 2). Although in both studies DDOC develop under different conditions and kinetics, they appear to carry similar phenotypes with some differences (see Table II: murine DDOC are CD86^–CD11b^– with dendrites). Nevertheless, based on the results of both studies and findings by Page et al. (59), it is reasonable to speculate that DDOC may develop not only in the presence of M-CSF and RANKL produced by OB, synovial, and stromal cells (60), but also in response to RANKL produced by inflammatory T cells and stimulatory Ags during immune interactions in the bone environment.

	Acknowledgments

 
We thank Annette Marleau at the Department of Microbiology and Immunology, University of Western Ontario (London, Ontario, Canada) for her help in this work.

	Disclosures

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
The University of Rochester has submitted a patent application on CD11c⁺ DC development to form OC functional in the bone environment and has listed Y.-T.A. Teng as the inventor.

	Footnotes

 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

¹ This work was supported by grants from the Ministry of Health of Ontario and the Canadian Institute of Health Research Grant MOP-37960, University of Rochester, and National Institutes of Health Grants DE14473 and DE15786 (to Y.-T.A.T.).

² Address correspondence and reprint requests to Dr. Yen-Tung Andy Teng, Department of Microbiology and Immunology, School of Medicine and Dentistry, University of Rochester, Rochester, NY 14620. E-mail address: andy_teng{at}urmc.rochester.edu

³ Abbreviations used in this paper: RA, rheumatoid arthritis; Aa, Actinobacillus actinomycetemcomitans; AMCA, 7-amino-4-methylcoumarin-3-acetic acid; BI, bovine insulin; BM, bone marrow; CT-R, calcitonin receptor; DC, dendritic cell; DDOC, DC-derived osteoclast; HA, hydroxyapatite; IF, immunofluorescent; m, mouse; MHC-II, MHC class II; Mo/MQ, monocyte/macrophage; RANK, receptor activator of NF-B; RANKL, receptor activator of NF-B ligand; OB, osteoblast; OMP-1, outer membrane protein-1; OC, osteoclast; OPG, osteoprotegerin; PD, periodontal disease (or periodontitis); s, soluble; SIDIA, scanning IF digital images analysis; TRAP, tartrate resistance acid phosphatase; WT, wild type.

Received for publication October 20, 2005. Accepted for publication May 10, 2006.

	References

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 

1. Theill, L. E., W. J. Boyle, J. M. Penninger. 2002. RANK-L and RANK: T cell, bone loss and mammalian evolution. Annu. Rev. Immunol. 20: 795-823. [Medline]
2. Lacey, D. L., E. Timms, H. L. Tan, M. J. Kelly, C. R. Dunstan, T. Burgess, R. Elliott, A. Colombero, G. Elliott, S. Scully, et al 1998. Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 93: 165-176. [Medline]
3. Yasuda, H., N. Shima, N. Nakagawa, K. Yamaguchi, M. Kinosaki, S. Mochizuki, A. Tomoyasu, K. Yano, M. Goto, A. Murakami. 1998. Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc. Natl. Acad. Sci. USA 95: 3597-3602. [Abstract/Free Full Text]
4. Teitelbaum, S. L., F. P. Ross. 2003. Genetic regulation of osteoclast development and function. Nat. Rev. Genet. 8: 638-649.
5. Simonet, W. S., D. L. Lacey, C. R. Dunstan, M. Kelley, M. S. Chang, R. Luthy, H. Q. Nguyen, S. Wooden, L. Bennett, T. Boone, et al 1997. Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 89: 309-319. [Medline]
6. Biskobing, D. M., X. Fan, J. Rubin. 1995. Characterization of M-CSF-induced proliferation and subsequent osteoclast formation in murine marrow culture. J. Bone Miner. Res. 10: 1025-1032. [Medline]
7. Tanaka, S., N. Takahashi, N. Udagawa, T. Tamura, T. Akatsu, E. R. Stanley, T. Kurokawa, T. Suda. 1993. Macrophage colony stimulating factor is indispensable for both proliferation and differentiation of osteoclast progenitors. J. Clin. Invest. 91: 257-263. [Medline]
8. Li, J., I. Sarosi, X. Q. Yan, S. Morony, C. Capparelli, H. L. Tan, S. McCabe, R. Elliott, S. Scully, G. Van, et al 2000. RANK is the intrinsic hematopoietic cell surface receptor that controls osteoclastogenesis and regulation of bone mass and calcium metabolism. Proc. Natl. Acad. Sci. USA 97: 1566-1571. [Abstract/Free Full Text]
9. Wong, B. R., R. Josien, S. Y. Lee, B. Sauter, H. L. Li, R. M. Steinman, Y. Choi. 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: 2075-2080. [Abstract/Free Full Text]
10. Anderson, D. M., E. Maraskovsky, W. L. Billingsley, W. C. Dougall, M. Tometsko, E. R. Roux, M. C. Teepe, R. F. DuBose, D. Cosman, L. Galibert. 1997. A homologue of the TNF receptor and its ligand enhance T cell growth and dendritic cell function. Nature 390: 175-179. [Medline]
11. Kong, Y. Y., H. Yoshida, I. Sarosi, H. L. Tan, E. Timms, C. Capparelli, S. Morony, A. J. Oliveira-dos-Santos, G. Van, A. Itie, et al 1999. OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis. Nature 397: 315-323. [Medline]
12. Dougall, W. C., M. Glaccum, K. Charrier, K. Rohrbach, K. Brasel, T. De Smedt, E. Daro, J. Smith, M. E. Tometsko, C. R. Maliszewski, et al 1999. RANK is essential for osteoclast and lymph node development. Genes Dev. 18: 2412-2424.
13. Yoshida, H., S. Hayashi, T. Kunisada, M. Ogawa, S. Nishikawa, H. Okamura, T. Sudo, L. D. Shultz, S. Nishikawa. 1990. The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature 345: 442-444. [Medline]
14. Kodama, H., A. Yamasaki, M. Nose, S. Niida, Y. Ohgame, M. Abe, M. Kumegawa, T. Suda. 1991. Congenital osteoclast deficiency in osteopetrotic (op/op) mice is cured by injections of macrophage-colony stimulating factor. J. Exp. Med. 173: 269-272. [Abstract/Free Full Text]
15. Mizuno, A., N. Amizuka, K. Irie, A. Murakami, N. Fujise, T. Kanno, Y. Sato, N. Nakagawa, H. Yasuda, S. Mochizuki, et al 1998. Severe osteoporosis in mice lacking osteoclastogenesis inhibitory factor/osteoprotegerin. Biochem. Biophys. Res. Commun. 247: 610-615. [Medline]
16. Kong, Y. Y., U. Feige, I. Sarosi, B. Bolon, A. Tafuri, S. Morony, C. Capparelli, J. Li, R. Elliott, S. McCabe, et al 1999. Activated T cells regulate bone loss and joint destruction in adjuvant arthritis through osteoprotegerin ligand. Nature 402: 304-309. [Medline]
17. Teng, Y. A., H. Nguyen, X. Gao, Y. Y. Kong, R. M. Gorczynski, B. Singh, R. P. Ellen, J. M. Penninger. 2000. Functional human T cell immunity and osteoprotegerin-ligand (OPG-L) control alveolar bone destruction in periodontal infection. J. Clin. Invest. 106: R59-R67.
18. Hofbauer, L. C., M. Schoppet. 2004. Clinical implication of the osteoprotegerin /RANKL/RANK system for bone and vascular diseases. J. Am. Med. Assoc. 28: 490-495.
19. Borwn, J. M., J. Zhang, E. T. Keller. 2004. OPG, RANKL and RANK in cancer metastasis: expression and regulation. Cancer Treat. Res. 118: 149-172. [Medline]
20. Mahamed, D. A., A. Marleau, M. Alnaeeli, B. Singh, X. Zhang, J. M. Penninger, A. Y.-T. Teng. 2005. G^– anaerobes-reactive CD4⁺ T cells trigger RANKL-mediated enhanced alveolar bone loss in diabetic NOD mice. Diabetes 54: 1477-1486. [Abstract/Free Full Text]
21. Banchereau, J., F. Briere, C. Caux, J. Davoust, S. Lebecque, Y. J. Liu, B. Pulendran, K. J. Palucka. 2000. Immunobiology of dendritic cells. Annu. Rev. Immunol. 18: 767-811. [Medline]
22. Ardavin, C., S. Amigorena, C. Reis e Sousa. 2004. Dendritic cell: immunobiology and cancer immunotherapy. Immunity 1: 17-23. [Medline]
23. Ardavin, C.. 2003. Origin, precursors and differentiation of mouse dendritic cells. Nat. Rev. Immunol. 3: 582-590. [Medline]
24. Leon, B., G. Martinez del Hoyo, V. Parrillas, H. H. Vargas, P. Sanchez-Mateos, N. Longo, M. Lopez-Bravo, C. Ardavin. 2004. Dendritic cell differentiation potential of mouse monocytes: monocytes represent immediate precursors of CD8^– and CD8⁺ splenic dendritic cells. Blood 103: 2668-2676. [Abstract/Free Full Text]
25. Traver, D., K. Akashi, M. Manz, M. Merad, T. Miyamoto, E. G. Engleman, I. L. Weissman. 2000. Development of CD8a-positive dendritic cells from a common myeloid progenitor. Science 290: 2152-2154. [Abstract/Free Full Text]
26. Pullendran, B., K. Palucka, J. Banchereau. 2001. Sensing pathogens and tuning immune responses. Science 293: 253-256. [Abstract/Free Full Text]
27. Shortman, K.. 2000. Burnet oration: dendritic cells: multiple subtypes, multiple origins, multiple functions. Immunol. Cell Biol. 78: 161-165. [Medline]
28. Miyamoto, T., O. Ohneda, F. Arai, K. Iwamoto, S. Okada, K. Takagi, D. M. Anderson, T. Suda. 2001. Bifurcation of osteoclasts and dendritic cells from common progenitors. Blood 98: 2544-2554. [Abstract/Free Full Text]
29. Servet-Delprat, C., S. Arnaud, P. Jurdic, S. Nataf, M. F. Grasset, C. Soulas, C. Domenget, O. Destaing, A. Rivollier, M. Perret, et al 2002. Flt3⁺ macrophage precursors commit sequentially to osteoclasts, dendritic cells and microglia. BMC Immunol. 3: 15[Medline]
30. Rivollier, A., M. Mazzorana, J. Tebib, M. Piperno, T. Aitsiselmi, C. Rabourdin-Combe, P. Jurdic, C. Servet-Delprat. 2004. Immature dendritic cell transdifferentiation into osteoclasts: a novel pathway sustained by rheumatoid arthritis microenvironment. Blood 104: 4029-4037. [Abstract/Free Full Text]
31. Da Costa, C., N. E. Annels, C. M. J. M. Faaji, R. G. Forsyth, P. C. W. Hogendoorn, R. M. Egeler. 2005. Presence of osteoclast-like multinucleated giant cells in the bone and nonostotic lesions of Langerhans cell histiocytosis. J. Exp. Med. 201: 687-693. [Abstract/Free Full Text]
32. Mohamadzadeh, M., S. Olson, W. V. Kalina, G. Ruthel, D. L. Demmin, K. L. Warfield, S. Bavari, T. R. Klaenhammer. 2005. Lactobacilli activate human dendritic cells that skew T cells towards T helper 1 polarization. Proc. Natl. Acad. Sci. USA 102: 2880-2885. [Abstract/Free Full Text]
33. Hafsi, N., P. Voland, S. Schwendy, R. Rad, W. Reindl, M. Gerhard, C. Prinz. 2004. Human dendritic cells respond to Helicobacter pylori, promoting NK cell and Th1 effector responses in vitro. J. Immunol. 173: 1249-1257. [Abstract/Free Full Text]
34. Thomas, R., K. P. MacDonald, A. R. Pettit, L. L. Cavanagh, J. Padmanabha, S. Zehntner. 1999. Dendritic cells and the pathogenesis of rheumatoid arthritis. J. Leukocyte Biol. 66: 286-292. [Abstract]
35. Highton, J., A. Kean, P. A. Hessian, J. Thomson, J. Rietveld, D. N. Hart. 2000. Cells expressing dendritic cell markers are present in the rheumatoid nodule. J. Rheumatol. 27: 339-346. [Medline]
36. Cirrincione, C., N. Pimpinelli, L. Orlando, P. Romagnoli. 2002. Lamina propria dendritic cells express activation markers and contact lymphocytes in chronic periodontitis. J. Periodontol. 73: 45-52. [Medline]
37. Teng, Y.-T.. 2003. The role of acquired immunity and periodontal disease progression. Crit. Rev. Oral Biol. Med. 14: 237-252. [Abstract/Free Full Text]
38. Santiago-Schwarz, F., P. Anand, S. Liu, S. E. Carsons. 2001. Dendritic cells (DCs) in rheumatoid arthritis (RA): progenitor cells and soluble factors contained in RA synovial fluid yield a subset of myeloid DCs that preferentially activate Th1 inflammatory-type responses. J. Immunol. 67: 1758-1768.
39. Schuler, G., M. B. Lutz, A. Bender, B. Thurner, C. Röder, J. Young, N. Romani. 1999. A guide to the isolation and propagation of dendritic cells. M. T. Lotze, and A. W. Thomson, eds. Dendritic Cells: Biology and Clinical Applications 27 515-533. Academic, San Diego.
40. Inaba, K., M. Inaba, N. Romani, H. Aya, M. Deguchi, S. Ikehara, S. Muramatsu, R. M. Steinman. 1992. Generation of large numbers of dendritic cells from mouse bone marrow cultures supplemented with granulocyte/macrophage colony-stimulating factor. J. Exp. Med. 176: 1693-1702. [Abstract/Free Full Text]
41. Teng, Y.-T., R. M. Gorczynski, S. Iwasaki, D. B. Williams, N. Hozumi. 1995. Evidence for Th2 cell-mediated suppression of antibody response in transgenic beef insulin-tolerant mice. Eur. J. Immunol. 25: 2522-2527. [Medline]
42. Leenen, P. J. M., M. Melis, W. A. T. Slieker, W. V. Ewijk. 1990. Murine macrophage precursor characterization II. Monoclonal antibodies against macrophage precursor antigens. Eur. J. Immunol. 20: 27-34. [Medline]
43. Horwood, N. J., V. Kartsogiannis, J. M. W. Quinn, E. Romas, T. J. Martin, M. T. Gillespie. 1999. Activated T cells support osteoclast formation in vitro. Biochem. Biophys. Res Commun. 256: 144-150.
44. Pan, Y.-J., G. Loo. 2000. Effect of copper deficiency on oxidative DNA damage in Jurkat T lymphocytes. Free Radic. Biol. Med. 28: 824-830. [Medline]
45. Teng, Y.-T., H. Nguyen, A. Hassanloo, R. P. Ellen, N. Hozumi, R. M. Gorczynski. 1999. Periodontal immune responses of human lymphocytes in Actinobacillus actinomycetemcomitans-inoculated NOD/SCID mice engrafted with peripheral blood leukocytes of periodontitis patients. J. Periodontal Res. 1: 54-61.
46. Klerck, B. D., L. Geboes, S. Hatse, H. Kelchtermans, Y. Meyvis, V. Kurt, B. Gary, B. Alfons, D