Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M513225200 on April 12, 2006

J. Biol. Chem., Vol. 281, Issue 23, 15809-15820, June 9, 2006
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
281/23/15809    most recent
M513225200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zheng, H.
Right arrow Articles by Osdoby, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zheng, H.
Right arrow Articles by Osdoby, P.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

RANKL Stimulates Inducible Nitric-oxide Synthase Expression and Nitric Oxide Production in Developing Osteoclasts

AN AUTOCRINE NEGATIVE FEEDBACK MECHANISM TRIGGERED BY RANKL-INDUCED INTERFERON-beta VIA NF-{kappa}B THAT RESTRAINS OSTEOCLASTOGENESIS AND BONE RESORPTION*

Hong Zheng1, Xuefeng Yu, Patricia Collin-Osdoby, and Philip Osdoby2

From the Department of Biology and the Division of Bone and Mineral Metabolism, Washington University, St. Louis, Missouri 63130

Received for publication, December 12, 2005 , and in revised form, March 16, 2006.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Nitric oxide (NO) is a multifunctional signaling molecule and a key vasculoprotective and potential osteoprotective factor. NO regulates normal bone remodeling and pathological bone loss in part through affecting the recruitment, formation, and activity of bone-resorbing osteoclasts. Using murine RAW 264.7 and primary bone marrow cells or osteoclasts formed from them by receptor activator of NF-{kappa}B ligand (RANKL) differentiation, we found that inducible nitric-oxide synthase (iNOS) expression and NO generation were stimulated by interferon (IFN)-{gamma} or lipopolysaccharide, but not by interleukin-1 or tumor necrosis factor-{alpha}. Surprisingly, iNOS expression and NO release were also triggered by RANKL. This response was time- and dose-dependent, required NF-{kappa}B activation and new protein synthesis, and was specifically blocked by the RANKL decoy receptor osteoprotegerin. Preventing RANKL-induced NO (via iNOS-selective inhibition or use of marrow cells from iNOS–/– mice) increased osteoclast formation and bone pit resorption, indicating that such NO normally restrains RANKL-mediated osteoclastogenesis. Additional studies suggested that RANKL-induced NO inhibition of osteoclast formation does not occur via NO activation of a cGMP pathway. Because IFN-beta is also a RANKL-induced autocrine negative feedback inhibitor that limits osteoclastogenesis, we investigated whether IFN-beta is involved in this novel RANKL/iNOS/NO autoregulatory pathway. IFN-beta was induced by RANKL and stimulated iNOS expression and NO release, and a neutralizing antibody to IFN-beta inhibited iNOS/NO elevation in response to RANKL, thereby enhancing osteoclast formation. Thus, RANKL-induced IFN-beta triggers iNOS/NO as an important negative feedback signal during osteoclastogenesis. Specifically targeting this novel autoregulatory pathway may provide new therapeutic approaches to combat various osteolytic bone diseases.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Normal bone remodeling requires a homeostatic balance between the activities of bone-forming osteoblasts and bone-resorbing osteoclasts (OCs).3 Excessive OC bone resorption leads to bone loss in many skeletal pathologies such as rheumatoid arthritis, periodontal disease, postmenopausal osteoporosis, implant osteolysis, and tumor-associated bone loss (1). OCs develop from hematopoietic precursors that fuse and differentiate into multinucleated bone-resorbing OCs in response to the essential tumor necrosis factor (TNF) family-related signal molecule receptor activator of NF-{kappa}B (RANK) ligand (RANKL) in the presence of permissive levels of macrophage colony-stimulating factor (M-CSF) (2, 3). RANKL expressed on the surface of osteoblasts, bone marrow stromal cells, or vascular endothelial cells or secreted by activated T cells directly engages a membrane receptor, RANK, on OC precursors and mature OCs to trigger multiple intracellular signaling cascades that stimulate OC gene expression, development, function, and survival (2, 3). RANKL/RANK interactions are specifically blocked by osteoprotegerin (OPG), a soluble decoy receptor released by osteoblast, stromal, vascular endothelial, and other cells that binds RANKL to inhibit OC formation and bone resorption in vivo and in vitro (2–4). The RANKL/OPG ratio critically determines net effects on OC formation and bone resorption, and increases in this ratio due to various inflammatory or proresorptive stimuli have been shown to significantly contribute to pathological bone loss in multiple skeletal disorders. Interestingly, although clearly essential for promoting OC formation and activity, RANKL was found recently to also trigger an autocrine negative feedback pathway in OC precursors that ultimately limits the extent of osteoclastogenesis concurrently stimulated by RANKL (5, 6). This negative feedback pathway involves RANKL induction of interferon (IFN)-beta in a c-Fos-dependent manner, followed by IFN-beta inhibition of RANKL-induced c-Fos expression necessary for OC formation (5).

OC formation and bone resorption are also inhibited by elevated levels of the multifunctional signal molecule nitric oxide (NO) in vivo and in vitro (7–13). NO is produced from L-arginine in an oxidative reaction catalyzed by NO synthase isoenzymes that are either constitutively expressed and calcium-activated (endothelial and neuronal NO synthase isoforms) or transcriptionally induced (inducible NO synthase (iNOS) isoform) in response to inflammatory stimuli (14). Previously, our group (8, 12, 15) and others (9, 16) have shown that OCs and related OC-like cells (as well as other bone cells) express iNOS and release NO in a regulated manner. NO produced endogenously or supplied by NO donors exerts potent biphasic actions that profoundly affect the recruitment, proliferation, differentiation, activity, and/or survival of OCs and osteoblasts, their precursors, and other cells within bone (8, 11, 17, 18). Whereas low levels of NO may support osteoblast bone formation and OC-mediated bone remodeling (both basal and cytokine-induced) (9, 19–22), high NO levels and NO-generating compounds inhibit OC formation and bone resorption and prevent bone loss, for example, in severe inflammation or estrogen-deficient animals (8, 10–13, 17–24). Conversely, iNOS deficiency or pharmacological inhibition of NO can accelerate OC formation and bone resorption in vivo and in vitro, decrease normal bone mass, exacerbate bone destruction in arthritis or osteoporosis models, and interfere with normal fracture healing (23, 25–28). On the other hand, iNOS-derived NO has been found in some studies to mediate bone loss in ovariectomized mice, interleukin-1 (IL-1)-induced OC resorption, and TNF-dependent OC survival (22, 29, 30). A deeper understanding of NO regulation and actions in bone is needed, especially as NO modulators are currently being evaluated in clinical trials as osteoprotective agents. Here, we report our unexpected discovery that OC precursor cells developing in response to RANKL up-regulate iNOS expression and NO release in a persistent manner. This RANKL-induced iNOS-derived NO functions as a negative feedback signal to limit osteoclastogenesis concurrently stimulated by RANKL. On the basis of such findings, we therefore further investigated whether RANKL-induced IFN-beta might somehow interface with RANKL-induced NO in this novel autocrine inhibitory feedback pathway.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Cell Culture and OC Differentiation—Murine RAW 264.7 cells (3.5 x 105 cells/well in 24-well plates) were cultured in Dulbecco's modified Eagle's medium with 10% fetal bovine serum, passaged, and differentiated into bone pit-resorbing OCs (RAW-OCs) by treatment with soluble recombinant murine RANKL (35 ng/ml, prepared in house, given daily with refeeding) for 4 days, similar to previous reports (31–33). In some cases, large well differentiated RAW-OCs were selectively enriched by fetal bovine serum gradient separation (31, 32). Primary bone marrow mononuclear cells were isolated from the long bones of 6–8-week-old female C57BL/6J wild-type (WT) or iNOS–/– mice (stock no. 002609, The Jackson Laboratory), and the marrow cells from one mouse were placed into a 75-cm2 flask and cultured in phenol red-free {alpha}-minimal essential medium and 10% fetal bovine serum with recombinant human M-CSF (25 ng/ml; R & D Systems, Minneapolis, MN) for 24 h. Nonadherent stroma-depleted cells were replated at 1.2 x 106 cells/well in 24-well plates and differentiated into bone pit-resorbing OCs (MA-OCs) by further treatment with M-CSF (25 ng/ml) and RANKL (50 ng/ml) given every other day with refeeding until day 7 or later when OCs had formed (31–33). Cells were fixed, stained for tartrate-resistant acid phosphatase (TRAP) activity as a marker of OC development, and co-stained with 4',6-diamidino-2-phenylindole dihydrochloride (Molecular Probes, Eugene, OR) to label nuclei, and the number of TRAP+ mononuclear and multinucleated cells (three or more nuclei/cell) and the number of nuclei/TRAP+ cell were counted across a constant number of sequential random fields using an Olympus light fluorescence microscope (33, 34). RANKL preparations were screened for negligible levels of bacterial endotoxin contamination using a commercial LAL kit (Cambrex Corp., Walkersville, MD).

Modulator Treatments—Recombinant murine IFN-{gamma}, TNF-{alpha}, or IL-1{alpha} (all from R & D Systems); Escherichia coli lipopolysaccharide (LPS) or phorbol 12-myristate 13-acetate (PMA) (both from Sigma); or the calcium ionophore A23187 [GenBank] (Calbiochem) was given to RAW cells or RAW-OCs formed by 4 days of RANKL treatment (as described above), and the cells and conditioned medium (CM) were harvested after 24 h for analysis of iNOS mRNA expression and NO release. To inhibit iNOS-derived NO generation during RANKL-induced OC development, RAW or marrow cells were cultured with various doses of the iNOS-selective inhibitor aminoguanidine (AG; Sigma) or L-N 6-(1-iminoethyl)lysine hydrochloride (L-NIL; ALEXIS Biochemicals Corp., San Diego, CA), each of which was freshly dissolved in warm medium just prior to administration on day 1 and given daily thereafter with refeeding until the harvest day noted in the figure legends. Similarly, RANKL-differentiating RAW or marrow cells were incubated with the NF-{kappa}B inhibitor 1-pyrrolidinecarbodithioic acid (PDTC; Sigma), IFN-beta (PBL Biomedical Laboratories, Piscataway, NJ), neutralizing rabbit polyclonal antibody (pAb) to IFN-beta (PBL Biomedical Laboratories), control non-immune rabbit IgG (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), the RANKL inhibitor OPG (OPG-Fc fusion peptide; ALEXIS Biochemicals Corp.), the cGMP inhibitor 1H-[1,2,4]oxadiazolo[4,3,-a]quinoxalin-1-one (ODQ; Sigma), the cGMP-dependent G-protein kinase inhibitor (Rp)-cGMP-S (Sigma), or the cGMP analog 8-pCPT-cGMP (Sigma) for the times and conditions given in the figure legends. Thereafter, NO production, gene expression, OC development, and/or OC bone pit resorption activity was analyzed in harvested cell and CM samples.

Nitrite Assay—CM was harvested from cultured cells, briefly centrifuged, and stored at –20 °C prior to assay. NO production was evaluated based on measuring nitrite as a stable end product of NO using the Griess reagent in a microplate assay (13). Results were normalized for cell protein using the BCA assay (Pierce), and data are expressed as µM nitrite accumulated in CM/mg of cell protein for the times noted in the figure legends.

RNA Isolation and Reverse Transcription (RT)-PCR—Total RNA was isolated from cells using RNA STAT-60 (Tel-Test, Inc., Friendswood, TX) and quantified by spectrophotometry. Semiquantitative RT-PCR amplifications for murine iNOS, TRAP, matrix metalloproteinase-9, cathepsin K, calcitonin receptor, IFN-{alpha}, IFN-beta, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were performed using gene-specific oligonucleotide forward and reverse primers (Table 1), 5–200 ng of total RNA, and Ready-To-Go RT-PCR beads (Amersham Biosciences) as described previously (31). Optimal cycling conditions, linear amplification ranges, lack of genomic DNA contamination, and the sizes and sequences of amplified products were determined (31). PCRs were conducted by initial denaturation at 94 °C for 1 min (TRAP, matrix metalloproteinase-9, cathepsin K, calcitonin receptor), 94 °C for 2 min (IFN-{alpha}, IFN-beta), or 95 °C for 2 min (iNOS, GAPDH), followed by cycling as detailed in Table 1. Products were separated by 1% agarose gel electrophoresis, stained with ethidium bromide, photographed using Gel-Doc (Bio-Rad), and quantified by density determination using Quantity One image analysis software (Bio-Rad). Results were normalized to GAPDH signals determined in parallel for each sample, and data are expressed as a ratio of gene to GAPDH expression. All amplicons were of the expected size (Table 1), and products were directly sequenced using an ABI PRISM cycle sequencing kit (PerkinElmer Life Sciences) to confirm identities by comparison with published sequences using computation performed at NCBI and the BLAST network service.


View this table:
[in this window]
[in a new window]
  		TABLE 1
Primers and conditions used for RT-PCR amplification of murine genes

The forward primer sequences are shown above the reverse primer sequences.

 
Electrophoretic Mobility Shift Assays—Nuclear extracts were prepared from RAW cells stimulated with RANKL (35 ng/ml, 10–40 min) using an NE-PER nuclear and cytoplasmic extraction reagent kit (Pierce) as recommended by the supplier. A double-stranded consensus NF-{kappa}B-binding site oligonucleotide probe (5'-AGTTGAGGGGACTTTCCCAGGC-3') was radiolabeled with [{gamma}-32P]ATP using a gel shift assay system kit (Promega Corp., Madison, WI) and incubated with nuclear extracts (5 µg) as recommended by the manufacturer. Protein-DNA complexes were resolved by electrophoresis on 4% nondenaturing acrylamide gels, and dried gels were exposed to Kodak film (at –80 °C). Signal specificity was confirmed by competition in parallel reactions with a 50-fold excess of unlabeled probe.

Western Blot Analysis—All extraction steps were performed at 4 °C or on ice. RAW cells (107 cells/100-mm dish) were cultured with 35 ng/ml RANKL for various times and washed with ice-cold phosphate-buffered saline, and extracts were prepared by addition of a lysis solution containing 20 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, and a 1:100 dilution of protease inhibitor mixture set III (Calbiochem) and freezing (–80 °C) overnight. Extracts were centrifuged at 18,000 x g for 15 min; protein concentrations were determined using the BCA assay; and equal amounts (50 µg) of protein were loaded per lane on 4–12% BisTris gels (Invitrogen). Following electrophoresis, proteins were transferred using a semidry apparatus to polyvinylidene fluoride membranes (Millipore Corp., Bedford, MA), blocked, probed overnight with rabbit pAb to iNOS (1:5000 dilution; BD Biosciences), reacted for 1 h with alkaline phosphatase-coupled goat anti-rabbit secondary pAb (1:1000 dilution; Santa Cruz Biotechnology, Inc.), and bands were immunodetected by chemiluminescence using CDP-Star (Roche Applied Science) and exposure to Kodak film (34). Band densities on films were quantified using Quantity One software.

Bone Pit Resorption Analysis—RAW cells or primary bone marrow mononuclear cells of WT or iNOS–/– origin were plated in 24-well dishes containing small circular ivory discs (5-mm diameter, 0.4-mm thick) and induced to form OCs via RANKL treatment as described above. Cells on the ivory discs were rinsed, fixed, stained for TRAP activity, and analyzed for TRAP+ cell formation and bone pit resorption as described previously (13, 35). The number of TRAP+ cells was determined by light microscopy in 12 random fields/ivory slice; cells were removed; and resorption pit numbers and areas were quantified in the same fields using a computer-linked dark-field reflective light microscopic image analysis system (13, 35).

Statistical Analysis—Data are presented as the means ± S.E. typically from two to three independent trials, each with two to four replicates per condition. Statistical comparisons between treatments were performed using one-way analysis of variance. For simultaneous comparisons between multiple treatments, significant differences were determined using Bonferroni's post hoc analysis of variance test where appropriate. Differences were considered significant at p < 0.05.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Cytokines and Other Inflammatory Stimuli Differentially Regulate iNOS mRNA Expression and NO Production in Murine RAW Cells and RAW-OCs—Previously, we showed that inflammatory stimuli such as TNF-{alpha}, IL-1, IFN-{gamma}, and LPS significantly increase NO production in avian OC-like cells, but not in authentic isolated avian OCs, whereas protein kinase C activation by PMA or intracellular calcium elevation by the ionophore A23187 [GenBank] stimulates NO production in authentic avian OCs and inhibits their bone pit resorption (12). Here, we investigated whether iNOS mRNA expression and NO release are similarly regulated in a differential manner by such modulators in murine OCs and their precursors. RAW cells or RANKL-differentiated RAW-OCs were treated for 24 h with TNF-{alpha}, IL-1{alpha}, PMA, or A23187. [GenBank] No significant increases in either iNOS mRNA expression or NO release were provoked by any of these agents (Table 2). In contrast, IFN-{gamma} and bacterial LPS each strongly up-regulated iNOS mRNA levels and NO production in both RAW cells and RAW-OCs, with RAW-OCs responding more intensely (by up to 50-fold) than RAW cells (up to 30-fold) to these stimuli (Table 2). Thus, the iNOS/NO system is up-regulated by certain inflammatory signals in murine RAW-OCs and their precursor cells, and this regulation differs in some respects from that seen in avian OCs and their bone marrow precursor cells.


Figure 1
View larger version (40K):
[in this window]
[in a new window]
  		FIGURE 1.
RANKL stimulates iNOS mRNA expression, iNOS protein levels, and NO production in RAW cells via a mechanism fully inhibited by OPG. A and B, RAW cells were cultured with 35 ng/ml RANKL for up to 72 h (without media changes), after which RNA, cell protein, and CM were harvested for analysis of iNOS mRNA expression by RT-PCR (three independent wells; expressed in arbitrary units relative to GAPDH), iNOS protein levels by Western blotting (two independent wells; expressed in relative densitometric units), and NO production by nitrite assay of CM (6–11 independent wells; expressed as µM nitrite released per mg of cell protein) as described under "Experimental Procedures." RANKL increased iNOS mRNA expression, iNOS protein levels, and NO release in a time-dependent manner in RAW cells. C, RAW cells were cultured for 24 h with 5–45 ng/ml RANKL and analyzed (three independent wells) as described above. RANKL dose-dependently increased iNOS mRNA expression and NO release in RAW cells. D, RAW cells were cultured with 35 ng/ml RANKL in the presence or absence of 5–100 ng/ml OPG-Fc for 24 h and analyzed (three independent wells) as described above. OPG-Fc dose-dependently inhibited RANKL-stimulated iNOS mRNA expression and NO release in RAW cells. Data in A–D are presented as the means ± S.E. *, statistical differences from controls without RANKL treatment (p < 0.05); #, differences from RANKL-treated cells (p < 0.05).

 


Figure 2
View larger version (31K):
[in this window]
[in a new window]
  		FIGURE 2.
RANKL stimulates iNOS mRNA expression and NO production throughout the differentiation of either murine RAW cells or primary bone marrow cells into RAW-OCs or MA-OCs, respectively. A, RAW cells were cultured with 35 ng/ml RANKL (administered daily with refeeding) for 1–5 days, at which time RAW-OCs were well formed. RNA, cell protein, and CM were harvested from a group of cells on each day and analyzed for iNOS mRNA expression (three independent wells) and NO production (10 independent wells) as described in the legend to Fig. 1. B, murine bone marrow cells were cultured with 25 ng/ml M-CSF and 50 ng/ml RANKL (administered every other day with refeeding) for 1–7 days, at which time MA-OCs were well formed. RNA, cell protein, and CM were harvested and analyzed as described for A for iNOS mRNA expression (three independent wells) and NO release (11 independent wells). In both murine RAW and marrow cells, iNOS mRNA expression and NO release were significantly elevated at all times during their RANKL differentiation compared with cells cultured without RANKL. Data are presented as the means ± S.E. of RANKL-treated/control ratios for iNOS mRNA and nitrite levels. *, statistical differences from control RAW or marrow cells without RANKL treatment (controls) (p < 0.05). The horizontal dashed lines in A and B indicate control levels.

 


View this table:
[in this window]
[in a new window]
  		TABLE 2
Effects of TNF-{alpha}, IL-1{alpha}, IFN-{gamma}, LPS, PMA, and A23187 on iNOS mRNA expression and NO production by RAW cells or RAW-OCs

RAW cells or RAW-OCs formed by 4 days of culture with RANKL (35 ng/ml) and then enriched by fetal bovine serum gradient separation were cultured for 24 h in the absence or presence of TNF-{alpha} (1 nM), IL-1{alpha} (0.1 nM), IFN-{gamma} (20 units/ml), LPS (1 ng/ml), PMA (100 nM), or A23187 (1 µM). CM and cell RNA and protein were harvested for analysis of iNOS mRNA expression (by RT-PCR) and NO release reflected in medium nitrite levels (by Griess assay) as described under "Experimental Procedures." Expression of iNOS mRNA was determined in two independent wells per condition and normalized to GAPDH. Nitrite levels in CM were calculated in three to four independent wells per condition and are expressed as µM nitrite accumulated per 24 h/mg of cell protein. Results for both iNOS and nitrite are presented as the means ± S.E. of the ratio of treated to untreated (control) RAW cell cultures.

 


Figure 3
View larger version (63K):
[in this window]
[in a new window]
  		FIGURE 3.
Inhibition of RANKL-induced NO production by iNOS-selective inhibitors enables greater RANKL-induced RAW-OC development. RAW cells were cultured with 35 ng/ml RANKL in the presence or absence of the iNOS-selective inhibitor AG (2–500 µM) or L-NIL (1–100 µg/ml) as described under "Experimental Procedures." A, CM was harvested after 24 h for nitrite determination (µM/mg of cell protein). Both AG and L-NIL dose-dependently inhibited RANKL-induced nitrite production. B, RAW cells were cultured in the absence (Control) or presence of 35 ng/ml RANKL with or without 500 µM AG or 100 µg/ml L-NIL, and the cells were refed daily with fresh medium and agents until day 4 or 5, when the cells were fixed, stained for TRAP activity, and examined by light microscopy. Both AG and L-NIL promoted greater RANKL-induced RAW-OC formation than was observed in their absence. The photographs shown are representative of three independent trials, each performed in triplicate, with similar results. Arrowheads point to examples of TRAP+ multinucleated RAW-OCs formed (magnification x100). C, the number of TRAP+ RAW-OCs (containing three or more nuclei/cell) were quantified from the nine independent wells (in 16 consecutive random fields/well) treated in B. Inhibition of RANKL-induced NO production by AG or L-NIL enabled more RAW-OCs to form. D, RAW cells were cultured in the absence or presence of 35 ng/ml RANKL with or without 1–100 µg/ml L-NIL for 24 h, refed according to the same conditions, and cultured for another 24 h, and RNA was harvested for RT-PCR analysis of OC developmental gene expression relative to GAPDH. Results were obtained from three to five independent wells per condition and are expressed as a percentage of the expression level for each gene in RAW-OCs developing with RANKL alone (set at 100% for each marker). L-NIL dose-dependently elevated the expression of multiple OC developmental genes during its enhancement of RANKL-induced RAW-OC formation. Data in A, C, and D are presented as the means ± S.E. *, statistical differences from controls without RANKL treatment (p < 0.05); differences from RANKL-treated cells (p < 0.05). Note that only one set of statistical symbols is displayed for simplicity above the data graphed in D, although such significance pertains to all four genes examined across the RANKL ± L-NIL conditions. CTR, calcitonin receptor; Cath-K, cathepsin K; MMP-9, matrix metalloproteinase-9.

 
RANKL Stimulates iNOS mRNA Levels, iNOS Protein Expression, and NO Production in Murine RAW Cells—In the above studies, we were surprised to observe that iNOS mRNA levels and NO release were higher in RANKL-induced RAW-OCs than in RAW cells. This suggested that RANKL itself might be causing an iNOS/NO response during its induction of OC development. To test this idea, RAW cells were treated with 35 ng/ml RANKL and analyzed for iNOS expression and NO production. Increased expression of iNOS mRNA was apparent by 3 h and reached maximal levels (>15-fold over control levels) by 24 h of RANKL exposure (Fig. 1A). Increased iNOS message was accompanied by a parallel steady rise in iNOS protein over 8–48 h of RANKL treatment (Fig. 1B) and by a time-dependent stimulation of NO release (at 16 h and beyond) (Fig. 1A). Temporally, the rapid induction of iNOS mRNA by RANKL was followed by iNOS protein expression and subsequent NO production. In dose studies, iNOS mRNA expression and NO release in RAW cells were increased in a concentration-dependent fashion by 15–45 ng/ml RANKL (Fig. 1C).

To exclude the possibility that endotoxin contamination of recombinant RANKL preparations was responsible for such iNOS and NO increases, three types of analyses were performed. First, endotoxin levels were measured (using a commercial LAL kit) in the 35 ng/ml RANKL preparations and consistently found to be <0.02 ng/ml (data not shown). Second, LPS concentrations of at least 0.1 ng/ml or more were necessary to induce any detectable (24 h) rises in iNOS mRNA or NO release, and LPS levels of at least 0.5 ng/ml were required to generate iNOS/NO responses comparable with RANKL in our RAW cell system (data not shown). Third and most important, RANKL stimulation of iNOS mRNA expression and NO release in RAW cells was dose-dependently inhibited by OPG-Fc, attaining complete ablation at 100 ng/ml (Fig. 1D). Because OPG acts as a decoy receptor to efficiently block RANKL/RANK interactions and signaling, these findings indicate that iNOS mRNA, iNOS protein, and NO production are stimulated by RANKL in a specific fashion in RAW cells.

RANKL Induces iNOS mRNA Expression and NO Production in Both RAW Cells and Primary Bone Marrow Cells throughout the Course of Their Development into Functional OCs—Extending our analysis beyond the initial stages of RAW-OC development, we found that iNOS mRNA expression and NO release were persistently elevated 4–6-fold by RANKL throughout the differentiation of RAW cells into RAW-OCs over a period of 5 days (Fig. 2A). Even at this point, when RAW-OC formation had peaked, iNOS mRNA levels and NO production did not return to the basal levels associated with undifferentiated RAW cells. To investigate whether RANKL induction of iNOS/NO is uniquely triggered in RAW cells or occurs also in primary isolated OC precursor populations, similar studies were conducted using M-CSF-treated murine bone marrow cells. Analogous to its actions in RAW cells, RANKL stimulated both iNOS mRNA expression (up to 6-fold) and NO production (to nearly 3-fold) throughout a 7-day course of marrow cell differentiation into TRAP+ MA-OCs. Furthermore, both iNOS mRNA and NO release were sustained at elevated levels in the mature RANKL-differentiated (day 7) MA-OCs.

NO Synthase Inhibitors That Suppress NO Release Triggered by RANKL Enable Greater OC Formation and Bone Pit Resorption in RANKL-differentiating RAW Cell Cultures—NO was shown previously to regulate OC formation and function in vitro or in vivo, often by inhibiting these processes (7–13, 17–24). Therefore, it seemed possible that NO induced by RANKL during OC development might be acting to negatively influence the rate or extent of OC formation and differentiation into bone pit-resorbing cells. Consistent with this, RANKL-induced OC formation was enhanced in the presence of AG or L-NIL, two selective competitive inhibitors of iNOS. Specifically, AG and L-NIL dose-dependently inhibited RANKL-induced NO release (Fig. 3A) while significantly increasing the number of TRAP+ RAW-OCs formed (Fig. 3, B and C), mRNA expression levels of a series of characteristic OC differentiation markers (Fig. 3D), and bone pit resorption by RAW-OCs formed under these conditions (Fig. 4, A–F). The latter was a direct consequence of both increased OC numbers and increased OC function. Thus, the greater number of RAW-OCs that formed in the presence of AG or L-NIL led to increases in the overall number of pits formed (Fig. 4C) and the total area of bone resorbed (Fig. 4B). In addition, bone pit resorption was stimulated via the enhanced activation of RAW-OCs, as reflected in significant increases in the mean area of bone resorbed per RAW-OC (Fig. 4E), the number of resorption pit excavations initiated by RAW-OCs (Fig. 4F), and the mean size of individual resorption lacunae (Fig. 4D). Overall, inhibiting RANKL-induced NO synthesis led to greater RANKL-induced RAW-OC formation and bone pit resorption.

Suppression of RANKL-induced iNOS-derived NO (Either by L-NIL Treatment or Using Marrow Cells from iNOS–/– Mice) Leads to Increased OC Formation and Bone Pit Resorption in RANKL-differentiating Murine Marrow Cultures—Two independent strategies to prevent iNOS-derived NO synthesis (either L-NIL treatment of normal marrow cells or the use of iNOS-deficient marrow cells) were employed to determine whether abolishing RANKL-induced NO production in primary murine marrow cell cultures would, as in RAW cells, culminate in greater RANKL-induced OC formation and bone resorption. As expected, only marrow cells from WT (but not iNOS–/–) mice expressed iNOS mRNA and up-regulated this gene in response to RANKL stimulation (Fig. 5A). Moreover, RANKL stimulated NO release only in marrow cells from WT but not iNOS–/– mice (Fig. 5B). NO induction in WT cells was fully inhibited by the presence of 100 µg/ml L-NIL (Fig. 5B). Preventing RANKL-induced NO production (either by L-NIL treatment of WT marrow cells or, alternatively, through the use of iNOS–/– marrow cells) led to significant increases in the number of TRAP+ MA-OCs formed (Fig. 5, C and D), mRNA expression levels of multiple characteristic OC differentiation markers (Fig. 5E), and bone pit resorption activity (both overall and on a per cell basis) of MA-OCs formed under these conditions (Fig. 6, A–F). Notably, pharmacological inhibition of iNOS activity due to L-NIL treatment of WT marrow cells was equally as effective as the complete genetic ablation of iNOS expression with respect to promoting MA-OC formation, differentiation, and function. Thus, RANKL-induced iNOS/NO serves as an important autocrine negative feedback signal that restrains normal OC formation and bone resorption from attaining maximal levels during RANKL-induced osteoclastogenesis in both murine RAW cells and primary bone marrow cells.


Figure 4
View larger version (40K):
[in this window]
[in a new window]
  		FIGURE 4.
Bone pit resorption in response to RANKL is increased by iNOS-selective inhibitors that stimulate RAW-OC formation. RAW cells were plated on ivory discs (3 x 105 cells/well, 24-well dish), treated in the absence or presence of 35 ng/ml RANKL with or without 500 µM AG or 100 µg/ml L-NIL, and refed with fresh medium and agents daily until day 6 or 7. Cells formed on the ivory discs were fixed, stained for TRAP activity, and analyzed for TRAP+ cell numbers and resorption pit formation (after cell removal) as described under "Experimental Procedures."A, representative photomicrographs of RAW-OCs and resorption lacunae generated on ivory discs in response to RANKL with or without AG or L-NIL. Each iNOS inhibitor increased RANKL-induced resorption pit formation in parallel with RAW-OC formation. Similar results were observed in three independent trials, with three to six ivory pieces per condition in each trial. Arrowheads and asterisks highlight examples of RAW-OCs and pit excavations formed, respectively (magnification x200). B–F, quantification of resorption pits formed by RANKL-differentiated RAW-OCs shown in A. Data were obtained from six independent wells (containing 13 ivory pieces used for analysis) per condition. Data also were normalized for the number of TRAP+ cells formed in the exact same surveyed ivory fields (12 fields/ivory chip). Resorption results are expressed as the means ± S.E. of the total area resorbed (B), the total number of pits formed (C), the area resorbed per pit (D), the area resorbed per TRAP+ cell (E), and number of pits formed per TRAP+ cell (F). AG and L-NIL each boosted resorption pit activity both overall and on a per TRAP+ cell basis. Data in B–F are expressedasthemean±S.E. *, statistical differences from controls without RANKL treatment (p < 0.05); #, differences from RANKL-treated cells (p < 0.05).

 


Figure 5
View larger version (48K):
[in this window]
[in a new window]
  		FIGURE 5.
NO suppression by use of iNOS inhibitors or iNOS–/– cells potentiates RANKL-induced MA-OC formation in primary murine marrow cell cultures. Murine bone marrow cells were isolated from WT or iNOS–/– mice and cultured overnight with 25 ng/ml M-CSF, and the nonadherent cells were cultured with M-CSF in the presence or absence of 50 ng/ml RANKL or 100 µg/ml L-NIL as described under "Experimental Procedures" for the times indicated here. A and B, RNA and CM were harvested from WT and iNOS–/– marrow cells treated with RANKL (and M-CSF) in the presence or absence of L-NIL for 24 h and subjected to RT-PCR analysis of iNOS mRNA expression and nitrite release, respectively. Marrow cells from iNOS–/– mice, unlike those from WT mice, did not exhibit basal or RANKL-stimulated iNOS mRNA expression or elevated levels of nitrite in CM in response to RANKL. C–E, marrow cells from WT and iNOS–/– mice were differentiated with M-CSF and RANKL in the presence or absence of L-NIL until day 7, when the cells were either stained for TRAP activity or harvested for RNA. Representative photomicrographs of TRAP-stained culture wells (eight independent wells per condition that yielded similar results) are shown in C. Arrowheads point to examples of MA-OCs formed (magnification x100). The number of TRAP+ MA-OCs (with three or more nuclei/cell) formed was determined in 16 fields/well from eight independent wells per condition (D), and RT-PCR analysis of a series of OC developmental gene markers was performed (E) as described in the legend to Fig. 3. Numbers in E indicate relative gene/GAPDH ratios in densitometric units. RANKL-induced MA-OC formation was significantly greater in marrow cultures either genetically lacking iNOS or unable to produce iNOS-derived NO due to L-NIL inhibition. Data in B and D are expressed as the means ± S.E. *, statistical differences from controls without RANKL treatment (p < 0.05); #, differences from RANKL-treated cells (p < 0.05). CTR, calcitonin receptor; Cath-K, cathepsin K; MMP-9, matrix metalloproteinase-9.

 
RANKL-stimulated iNOS/NO May Restrain OC Formation via Non-cGMP-mediated Mechanism(s)—NO conveys many of its biological actions through binding to the heme component of guanylate cyclase, thereby stimulating cGMP production and activation of downstream signaling pathways, including protein kinase G activity. To determine whether cGMP signaling is involved in RANKL-induced NO inhibition of osteoclastogenesis, RAW cells were cultured with RANKL (35 ng/ml) in the presence or absence of the cGMP inhibitor ODQ (1, 5, or 10 µM), the G-protein kinase inhibitor (Rp)-cGMP-S (100 µM), or the cGMP analog 8-pCPT-cGMP (100 µM) for up to 4 days. Consistent with their proposed mode of action downstream of NO production, none of these agents affected RANKL-stimulated nitrite release from RAW cells (data not shown). More important, neither ODQ nor (Rp)-cGMP-S inhibition of cGMP production or G-protein kinase activation, respectively, led to any enhancement of RANKL-mediated RAW-OC formation (cell numbers or size), indicating that RANKL-induced NO does not limit OC formation through such pathways (data not shown). Moreover, addition of the cGMP analog 8-pCPT-cGMP did not mimic RANKL-induced NO and further reduce RAW-OC formation in response to RANKL (data not shown). Therefore, RANKL stimulation of iNOS/NO may limit osteoclastogenesis via one or more non-cGMP-dependent mechanisms.

NF-{kappa}B Activation Is Essential for Both RANKL-induced OC Formation and RANKL-induced iNOS/NO Stimulation—NF-{kappa}B is activated by RANKL and is known to play an essential role in OC development and bone resorption. Consistent with this, we found that NF-{kappa}BDNA binding activity assayed in nuclear protein extracts was rapidly increased following RANKL treatment of RAW cells (Fig. 7A). The NF-{kappa}B inhibitor PDTC dose-dependently reduced RANKL induction of iNOS mRNA expression, iNOS protein levels, and NO release, with complete inhibitory actions at 10 µM (Fig. 7B). In addition, as expected from its key role in OC development, inhibition of NF-{kappa}B activation by PDTC also interfered with RANKL-mediated RAW-OC formation (Fig. 7, C and D). Therefore, NF-{kappa}B activation is necessary not only for RANKL to induce OC formation, but also for its concurrent stimulation of the autocrine iNOS/NO negative feedback pathway restraining OC development.


Figure 6
View larger version (49K):
[in this window]
[in a new window]
  		FIGURE 6.
NO suppression by use of iNOS inhibitors or iNOS–/– cells potentiates RANKL-induced MA-OC bone pit resorption in developing murine marrow cell cultures. Marrow cells from WT and iNOS–/– mice were plated on ivory discs (1.2 x 106 cells/well, three ivory chips/well, 24-well dish), cultured with 25 ng/ml M-CSF and 50 ng/ml RANKL in the presence or absence of 100 µg/ml L-NIL, and refed with fresh medium and agents every other day until day 9. Cells formed on the ivory discs were fixed, stained for TRAP activity, and analyzed for TRAP+ cell numbers and resorption pit formation as described in the legend to Fig. 4. A, representative photomicrographs of MA-OCs and resorption lacunae generated from WT and iNOS–/– marrow cells on ivory discs in response to RANKL with or without L-NIL. Suppression of RANKL-induced NO in iNOS–/– cells or WT cells cultured with L-NIL enhanced bone pit resorption by developing MA-OCs. Similar results were observed in four independent wells per condition. Arrowheads and asterisks highlight examples of MAOCs and pit excavations formed, respectively (magnification x100). B–F, quantification of resorption pit formation by RANKL-differentiated MA-OCs from WT or iNOS–/– marrow cultures. Data were obtained from three independent wells (nine ivory pieces) per condition. Results also were normalized for the number of TRAP+ cells formed in the same surveyed ivory fields (12 fields/ivory chip). Resorption results are expressed as the means ± S.E. of the total area resorbed (B), the total number of pits formed (C), the area resorbed per pit (D), the area resorbed per TRAP+ cell (E), and number of pits formed per TRAP+ cell (F). L-NIL boosted resorption pit activity in RANKL-differentiated WT MA-OC cultures to levels that were comparable with the resorption activity of MA-OCs differentiating in iNOS–/– marrow cultures. Data in B–F are expressed as the means ± S.E. *, statistical differences from controls without RANKL treatment (p < 0.05); #, differences from RANKL-treated cells (p < 0.05).

 
RANKL Stimulation of iNOS/NO as an Autocrine Negative Feedback Signal during OC Development Is Triggered by RANKL Induction of IFN-beta through an NF-{kappa}B-dependent Pathway—Because RANKL has been reported to induce IFN-beta as a negative feedback signal that limits RANKL-mediated OC formation in murine OC precursors (5), we investigated whether IFN-beta might interface somehow with iNOS/NO also stimulated by RANKL to convey autocrine negative effects via a common or converging signaling pathway. RANKL-treated RAW cells exhibited a rapid up-regulation of IFN-beta mRNA expression, with peak levels attained by 3 h (Fig. 8A). In contrast, IFN-{alpha} mRNA expression was very low and unaltered (even with more PCR cycles) by RANKL treatment of RAW cells (data not shown). RANKL induction of IFN-beta was inhibited in the presence of 10 µM PDTC (Fig. 8B), but not 10 µM cycloheximide (data not shown). These findings indicate that RANKL directly induces IFN-beta mRNA expression via an NF-{kappa}B-dependent pathway. By contrast, RANKL stimulation of iNOS mRNA levels and NO release (shown above to be inhibited by PDTC) was blocked by concurrent cycloheximide treatment (data not shown), suggesting that some intermediary protein(s) (such as IFN-beta) must be synthesized in order for RANKL to induce iNOS/NO.

Next, the potential for IFN-beta to substitute for RANKL in stimulating iNOS expression and NO release in RAW cells was investigated. Like RANKL, IFN-beta increased iNOS mRNA expression (Fig. 9A) while dose- and time-dependently stimulating NO production in RAW cells, acting either alone or synergistically with RANKL (Fig. 9, B and C). Strikingly, the ability of RANKL to stimulate NO release from RAW cells was reduced to near basal levels by the presence of neutralizing pAb to IFN-beta, but not by an equivalent amount of non-immune IgG (Fig. 9, B and C). Thus, IFN-beta may lie downstream of RANKL and mediate its actions to evoke iNOS-derived NO in developing OCs. Functionally, the ability of neutralizing pAb to IFN-beta to suppress RANKL-stimulated NO release (Fig. 9, B and C) was associated with an ~50% increase in RANKL-induced RAW-OC development (Fig. 9D) to levels resembling those elicited by RANKL in the presence of the iNOS inhibitor L-NIL or AG (Fig. 9D). Conversely, the greater NO production elicited by IFN-beta in combination with RANKL (Fig. 9, B and C) was associated with a pronounced reduction in RANKL-stimulated RAW-OC formation (Fig. 9D), and this inhibition was substantially relieved by inclusion of either L-NIL or AG during RAW cell treatment (Fig. 9D). Overall, interfering with RANKL-generated NO production (by blocking either RANKL-stimulated IFN-beta or NO synthase activity) significantly enhanced RANKL-mediated OC formation, whereas elevating RANKL-generated NO production (by co-incubating with additional IFN-beta) inhibited OC formation.

Proposed Model for NF-{kappa}B-mediated Induction of IFN-beta and iNOS/NO as an Autocrine Negative Feedback Signaling Pathway Elicited by RANKL during OC Development—Fig. 10 summarizes our findings relative to the iNOS/NO negative feedback pathway elicited by RANKL during osteoclastogenesis.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
NO is an important multifunctional signaling molecule in bone that is produced by various cells at basal or stimulated levels and that regulates bone formation, resorption, remodeling, mechanotransduction, and repair in physiological or pathophysiological conditions. Many NO effects in bone are conveyed through targeting OC formation, activity, or survival (18, 36). Here, we report our discovery of a new role for iNOS-derived NO in bone as an autocrine negative feedback inhibitor elicited by RANKL during osteoclastogenesis. A direct mechanistic link was established between this RANKL-induced iNOS/NO and IFN-beta, a previously described RANKL-induced autocrine negative feedback inhibitor of developing OCs. Thus, RANKL induces IFN-beta via an NF-{kappa}B-dependent pathway, thereby evoking iNOS expression and NO production that ultimately limit the extent of osteoclastogenesis and bone resorption elicited by RANKL in murine RAW cell or primary bone marrow cell cultures.


Figure 7
View larger version (63K):
[in this window]
[in a new window]
  		FIGURE 7.
NF-{kappa}B activation is required for both RANKL-induced OC formation and RANKL-induced iNOS/NO stimulation. A, RAW cells (1 x 107) were treated with or without 35 ng/ml RANKL for various times, and nuclear extracts were prepared for use in electrophoretic mobility shift assay analysis with a 32P-labeled double-stranded probe synthesized to represent a consensus NF-{kappa}B sequence as described under "Experimental Procedures." RANKL time-dependently increased NF-{kappa}B binding activity in RAW cell nuclear extracts. Reaction specificity was confirmed by competition with 50-fold excess unlabeled probe, which reduced NF-{kappa}B binding activity signals in RANKL-treated nuclear extracts to the level in unstimulated control cells (data not shown). B, RAW cells were precultured for 30 min with or without the NF-{kappa}B inhibitor PDTC (0.1–10 µM) and then further cultured with or without PDTC and RANKL (35 ng/ml) for 24 h, after which cells and CM were harvested for analysis of iNOS mRNA expression (three independent wells), iNOS protein levels (two independent wells), and NO release (12 independent wells) as described in the legend to Fig. 1. PDTC dose-dependently inhibited RANKL-induced iNOS mRNA expression, iNOS protein levels, and NO release in RAW cells. C and D, RAW cells were cultured with 35 ng/ml RANKL in the presence or absence of 10 µM PDTC and refed daily with fresh medium and agents until day 4, when the cells were fixed and stained for TRAP activity. The photomicrographs of TRAP-stained cultures shown in C are representative of similar results obtained in eight independent wells per condition. Arrowheads point to examples of RAW-OCs formed (magnification x100). The number of TRAP+ RAW-OCs formed in these cultures (eight wells per condition) is quantified in D. PDTC markedly inhibited RAW-OC differentiation in RANKL-treated RAW cell cultures. Data in B and D are expressed as the means ± S.E. *, statistical differences from controls without RANKL treatment (p < 0.05); #, differences from RANKL-treated cells (p < 0.05).

 
This study was initially undertaken to extend our former work on iNOS/NO regulation in avian OCs and OC-like cells (8, 12, 13, 15). In particular, we wondered if developing murine OCs, like avian OCs, lose the ability for cytokines and acquire the ability for calcium to induce iNOS/NO. However, we found that iNOS/NO in mature bone-resorbing RAW-OCs was not elevated by calcium ionophore or phorbol ester activation of protein kinase C, but was still inducible by IFN-{gamma} or LPS. Neither IL-1 nor TNF-{alpha} raised iNOS/NO in RAW cells or RAW-OCs, whereas these cytokines elevated iNOS/NO in avian precursor cells, but not in mature OCs (12, 15). Together, the findings highlight differences between the avian and murine OC systems, support the known species-dependent nature of iNOS regulation (37), and agree well with reports that the murine macrophage iNOS promoter is highly responsive to IFN-{gamma} and LPS (38) and that neither IL-1 nor TNF-{alpha} stimulates iNOS/NO in RAW cells (39). However, our results partly contrast with a recent study describing iNOS/NO induction by TNF-{alpha} (but not IL-1) in mature murine OCs formed in marrow cell/osteoblast co-cultures in vitro (30).

During this work, we noticed unexpectedly that iNOS expression and NO generation increased following RANKL treatment of RAW cells. This was confirmed in both RAW cell and primary bone marrow cell cultures. Specifically, RANKL (but not M-CSF) stimulated iNOS mRNA and protein expression and NO generation in a time- and dose-dependent and OPG-inhibitable manner throughout murine RAW cell or marrow cell development into TRAP+ multinucleated bone-resorbing OCs. Similarly, two recent reports showed that RANKL increases iNOS mRNA and NO release in murine bone marrow cells and macrophages, although neither study examined if OPG inhibits such responses (41, 42). In contrast, another study reported that mature murine OCs formed by marrow cell/osteoblast co-culture respond to TNF-{alpha} (but not RANKL) with induction of iNOS/NO (30). Possibly, this reflects differences in RANKL sources, manner of OC formation from marrow cells, cell culture, treatment conditions, or other variables.


Figure 8
View larger version (30K):
[in this window]
[in a new window]
  		FIGURE 8.
RANKL rapidly induces IFN-beta mRNA expression in RAW cells via an NF-{kappa}B-mediated pathway. A, RAW cells were treated with or without 35 ng/ml RANKL for various times, after which RNA was harvested from triplicate wells per condition for RT-PCR analysis of IFN-beta and IFN-{alpha} mRNA expression relative to GAPDH. RANKL rapidly induced IFN-beta mRNA expression in RAW cells, whereas IFN-{alpha} was expressed at very low levels (undetectable at 34 cycles) and was not up-regulated by RANKL in RAW cells (data not shown). B, RAW cells were pretreated for 30 min with or without the NF-{kappa}B inhibitor PDTC (10 µM) and further cultured for 3 h with or without PDTC and RANKL (35 ng/ml), and RNA was harvested from triplicate wells per condition for RT-PCR analysis of IFN-beta mRNA expression. PDTC effectively blocked RANKL-induced IFN-beta expression. Data in A are presented as the means ± S.E. *, statistical differences from controls without RANKL treatment (p < 0.05).

 


Figure 9
View larger version (31K):
[in this window]
[in a new window]
  		FIGURE 9.
RANKL stimulation of iNOS/NO as a negative feedback signal in OC development is mediated by RANKL induction of IFN-beta. A, RAW cells were incubated with or without 50 units/ml IFN-beta for 24 h, after which RNA was harvested from duplicate wells per condition for RT-PCR analysis of iNOS mRNA expression relative to GAPDH. IFN-beta mimicked the ability of RANKL to induce iNOS mRNA expression in RAW cells. B, RAW cells were cultured with or without RANKL (35 ng/ml) and/or IFN-beta (2.5–100 units/ml), neutralizing pAb to IFN-beta (2.5–50 units/ml), or control non-immune IgG (250 ng/ml, corresponding to 50 units/ml (250 ng/ml) neutralizing pAb to IFN-beta) for 24 h. CM was harvested from nine independent wells per condition for nitrite analysis. IFN-beta dose-dependently stimulated NO release from RAW cells and synergized with RANKL in eliciting NO, whereas neutralizing pAb to IFN-beta dose-dependently inhibited RANKL-induced NO production in RAW cells. C, shown are the results from the time course analysis of the effects of RANKL (35 ng/ml) with or without IFN-beta (50 units/ml) or pAb to IFN-beta (50 units/ml) on NO release in RAW cells in culture. Results were obtained from eight independent wells per condition. D, RAW cells were cultured in the absence or presence of RANKL (35 ng/ml) with or without IFN-beta (50 units/ml), pAb to IFN-beta (50 units/ml), control IgG (250 ng/ml), L-NIL (100 µg/ml), and/or AG (500 µM) for 4 days, after which they were fixed, stained for TRAP activity, digitally photographed, and analyzed for the number of TRAP+ RAW-OCs formed as described in the legend to Fig. 3. Results were quantified from 16 fields/well (11 independent wells per condition). Data in B–D are presented as the means ± S.E. *, statistical differences from controls without RANKL treatment (p < 0.05); #, differences from RANKL-treated cells (p < 0.05); $, differences from RANKL- and IFN-beta-treated cells (p < 0.05).

 
The biological significance of RANKL-induced iNOS-derived NO was demonstrated by using iNOS-selective inhibitors or marrow precursor cells from iNOS-deficient mice. In all cases, interfering with RANKL-induced iNOS/NO enhanced OC formation and bone pit resorption in RANKL-differentiating RAW cell or marrow cell cultures. This suggests that RANKL-induced iNOS/NO is triggered as an autocrine negative feedback signal to limit osteoclastogenesis promoted by RANKL. Previously, NO was shown to regulate OC recruitment, formation, function, and survival in a complex, potentially biphasic, and context-dependent fashion in which NO can either stimulate (1, 9, 29, 30, 42, 43) or suppress (7–13, 17–24, 43) OC formation and bone resorption. In particular, iNOS deficiency or pharmacological inhibition of NO (including by AG or L-NIL) has increased OC formation and/or bone resorption in murine, rat, and avian cell cultures (9, 10–12, 24); stimulated OC recruitment, development, and function in association with angiogenesis in a chicken egg vascular model (13); decreased bone mass in normal or ovariectomized rats (8, 25); worsened bone destruction in arthritis models (26, 27); and interfered with normal bone fracture healing (28). Consistent with these studies, our present findings strongly support an inhibitory role (in this case, through a novel autocrine mode of action) for NO in restraining RANKL-induced osteoclastogenesis and bone resorption under normal physiological conditions.


Figure 10
View larger version (26K):
[in this window]
[in a new window]
  		FIGURE 10.
Proposed model for the induction of IFN-beta and iNOS/NO as a negative feedback regulatory pathway elicited by RANKL during OC development. RANKL binding to the RANK receptor on OC precursors triggers multiple intracellular signaling pathways that both promote and restrain OC development, thereby sensitively regulating the precise levels of OC formation and differentiation that occur. RANKL activation of NF-{kappa}B, an essential event in OC formation, is also required for induction of the negative feedback regulatory signal IFN-beta, which, in turn, up-regulates iNOS mRNA expression and NO production in murine OC precursor cells. Because the iNOS gene promoter contains binding sites for NF-{kappa}B and because PDTC fully inhibited while anti-IFN-beta pAb incompletely suppressed RANKL-induced iNOS mRNA expression, it is possible that RANKL-stimulated iNOS gene transcription is achieved not only through activated NF-{kappa}B up-regulation of IFN-beta, but also through direct NF-{kappa}B binding to regulatory elements in the iNOS promoter. RANKL-stimulated iNOS-derived NO restrains normal osteoclastogenesis; and interfering with this NO negative feedback signal, by use of iNOS-selective inhibitors (AG, L-NIL) or OC precursor marrow populations from iNOS–/– mice, in all cases markedly increased RANKL-mediated OC formation and bone pit resorption. RANKL-induced OC formation was similarly enhanced by blocking upstream signaling with neutralizing pAb to IFN-beta, highlighting the importance of RANKL-induced IFN-beta in the iNOS/NO negative feedback pathway triggered during osteoclastogenesis. Inhibitory effects on OC formation might result from IFN-beta down-regulation of essential c-Fos protein expression (5, 6) or from other IFN-beta-elicited NO-mediated mechanisms. Although undefined at present in the context of OC biology, such NO-suppressive effects could theoretically include direct NO modification(s) of key signaling molecules such as NF-{kappa}B, c-Fos, and JNK required for osteoclastogenesis (40, 45–47), but do not appear to involve RANKL-induced NO activation of cGMP-mediated signaling.

 
Recently, IFN-beta was identified as a RANKL-stimulated autocrine negative feedback inhibitor that limits osteoclastogenesis in vitro and in vivo (5, 6). Here, we have shown that the RANKL-induced IFN-beta and iNOS/NO autoregulatory pathways are intimately linked, with RANKL-mediated IFN-beta induction being necessary and sufficient, at least in large part, for subsequent iNOS/NO induction in developing OCs. RANKL stimulation of both signals was abolished by the antioxidant NF-{kappa}B inhibitor PDTC, indicating that NF-{kappa}B is essential not only for RANKL-induced OC formation, as is well known (44), but also for triggering this inherent IFN-beta/NO negative feedback pathway. Thus, IFN-beta induction depends on RANKL activation of both c-Fos (5) and NF-{kappa}B, consistent with there being an NF-{kappa}B-binding site within the IFN-beta (but not IFN-{alpha}) gene promoter (5). How RANKL-induced IFN-beta reduces murine OC formation was shown previously to involve a decrease in RANKL-induced c-Fos protein, a transcription factor necessary for OC formation, without any changes in TRAF6, JNK, p38 mitogen-activated protein kinase (MAPK), or NF-{kappa}B protein levels (5). Our new findings further show that IFN-beta inhibits OC formation through induction of iNOS/NO. Whether this NO causes the reduction in c-Fos protein or whether c-Fos declines in an NO-independent manner is a subject for future inquiry. In previous work, NO (supplied by an exogenous NO donor) was reported to diminish OC formation via a cGMP-dependent mechanism in 1,25-dihydroxyvitamin D3-differentiated murine marrow cultures (11). In contrast, no evidence was obtained here for the involvement of a cGMP-dependent mechanism to mediate the inhibitory actions of endogenously stimulated iNOS-derived NO in RANKL-differentiated murine RAW cells. Therefore, exactly how RANKL-induced NO interferes with osteoclastogenesis at a mechanistic level is not yet clear. Possibly, NO targeting of essential transcription factors such as NF-{kappa}B, c-Fos, and JNK via S-nitrosylation or Tyr nitration, as occurs in other cells (40, 45–47), may be involved, and this is a subject of ongoing studies.

In summary, we have shown here for the first time that RANKL induces iNOS expression and NO generation via an NF-{kappa}B- and IFN-beta-mediated mechanism. This NO functions as an autocrine negative feedback signal to limit osteoclastogenesis concurrently stimulated by RANKL. Specifically targeting this autoregulatory pathway may afford new approaches for therapeutic interventions in bone diseases characterized by excessive OC formation and bone resorption.


   FOOTNOTES
 
* This work was supported in part by National Institutes of Health Grant AR42715 (to P. O.). 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. Back

1 Supported by Mineral and Skeletal Metabolism Training Grant 5 T32 AR07033-31 from the National Institutes of Health. Back

2 To whom correspondence should be addressed: Dept. of Biology, Washington University, 1229 McDonnell Hall, P. O. Box 1229, St. Louis, MO 63130. Tel.: 314-935-4044; Fax: 314-935-5134; E-mail: osdoby{at}biology.wustl.edu.

3 The abbreviations used are: OCs, osteoclasts; TNF, tumor necrosis factor; RANK, receptor activator of NF-{kappa}B; RANKL, RANK ligand; M-CSF, macrophage colony-stimulating factor; OPG, osteoprotegerin; IFN, interferon; iNOS, inducible nitric-oxide synthase; IL-1, interleukin-1; RAW-OCs, RAW cell-derived osteoclasts; WT, wild-type; MA-OCs, marrow cell-derived osteoclasts; TRAP, tartrate-resistant acid phosphatase; LPS, lipopolysaccharide; PMA, phorbol 12-myristate 13-acetate; CM, conditioned medium; AG, aminoguanidine; L-NIL, L-N6-(1-iminoethyl)lysine hydrochloride; NF-{kappa}B, nuclear factor-{kappa}B; PDTC, 1-pyrrolidinecarbodithioic acid; pAb, polyclonal antibody; ODQ, 1H-[1,2,4]oxadiazolo[4,3,-a]quinoxalin-1-one; (Rp)-cGMP-S, 8-(4-chlorophenylthio)guanosine 3':5'-monophosphorothioate (Rp isomer triethylammonium salt); 8-pCPT-cGMP, 8-(4-chlorophenylthio)guanosine 3':5'-monophosphate; RT, reverse transcription; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; BisTris, 2-[bis(2-hydroxyethyl)amino]-2-(hydroxymethyl)propane-1,3-diol; JNK, c-Jun N-terminal kinase. Back



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Heymann, D., Fortun, Y., Redini, F., and Padrines, M. (2005) Drug Discovery Today 10, 242–247[CrossRef][Medline] [Order article via Infotrieve]
  2. Lerner, U. (2004) Crit. Rev. Oral Biol. Med. 15, 64–81[Abstract/Free Full Text]
  3. Hofbauer, L., and Heufelder, A. (2001) J. Mol. Med. 79, 243–253[CrossRef][Medline] [Order article via Infotrieve]
  4. Kostenuik, P., and Shalhoub, V. (2001) Curr. Pharm. Des. 7, 613–635[CrossRef][Medline] [Order article via Infotrieve]
  5. Takayanagi, H., Kim, S., Matsuo, K., Suzuki, H., Suzuki, T., Sato, K., Yokochi, T., Oda, H., Nakamura, K., Ida, N., Wagner, E., and Taniguchi, T. (2002) Nature 416, 744–749[CrossRef][Medline] [Order article via Infotrieve]
  6. Hayashi, T., Kaneda, T., Toyama, Y., Kumegawa, M., and Hakeda, Y. (2002) J. Biol. Chem. 277, 27880–27886[Abstract/Free Full Text]
  7. MacIntyre, I., Zaidi, M., Alam, A., Datta, H., Moonga, B., Lidbury, P., Hecker, M., and Vane, J. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 2936–2940[Abstract/Free Full Text]
  8. Kasten, T., Collin-Osdoby, P., Patel, N., Osdoby, P., Krukowski, M., Misko, T., Settle, S., Currie, M., and Nickols, G. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 3569–3573[Abstract/Free Full Text]
  9. Brandi, M., Hukkanen, M., Umeda, T., Moradi-Bidhendi, N., Bianchi, S., Gross, S., Polak, J., and MacIntyre, I. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 2954–2958[Abstract/Free Full Text]
  10. Ralston, S., and Grabowski, P. (1996) Bone 19, 29–33[Medline] [Order article via Infotrieve]
  11. Holliday, L., Dean, A., Lin, R., Greenwald, J., and Gluck, S. (1997) Am. J. Physiol. 272, F283–F291
  12. Sunyer, T., Rothe, L., Kirsch, D., Jiang, X., Anderson, F., Osdoby, P., and Collin-Osdoby, P. (1997) Endocrinology 138, 2148–2162[Abstract/Free Full Text]
  13. Collin-Osdoby, P., Rothe, L., Bekker, S., Anderson, F., and Osdoby, P. (2000) J. Bone Miner. Res. 15, 474–488[CrossRef][Medline] [Order article via Infotrieve]
  14. Griffith, O., and Stuehr, D. (1995) Annu. Rev. Physiol. 57, 707–736[CrossRef][Medline] [Order article via Infotrieve]
  15. Sunyer, T., Rothe, L., Jiang, X., Osdoby, P., and Collin-Osdoby, P. (1996) J. Cell. Biochem. 60, 469–483[CrossRef][Medline] [Order article via Infotrieve]
  16. Silverton, S., Mesaros, S., Markham, G., and Malinski, T. (1995) Endocrinology 136, 5244–5247[Abstract]
  17. Chae, H., Park, R., Chung H., Kang, J., Kim, M., Choi, D., Bang, B., and Kim, H. (1997) J. Pharm. Pharmacol. 49, 897–902[Medline] [Order article via Infotrieve]
  18. van't Hof, R., and Ralston, S. (2001) Immunology 103, 255–261[CrossRef][Medline] [Order article via Infotrieve]
  19. Ralston, S., Ho, L., Helfrich, M., Grabowski, P., Johnston, P., and Benjamin, N. (1995) J. Bone Miner. Res. 10, 1040–1049[Medline] [Order article via Infotrieve]
  20. Jamal, S., Browner, W., Bauer, D., and Cummings, S. (1998) J. Bone Miner. Res. 13, 1755–1759[CrossRef][Medline] [Order article via Infotrieve]
  21. Chole, R., Tinling, S., Leverentz, E., and McGinn, M. (1998) Acta Otolaryngol. 118, 705–711[CrossRef][Medline] [Order article via Infotrieve]
  22. van't Hof, R., Armour, K., Smith, L., Armour, K., Wei, X., Liew, F., and Ralston, S. (2000) Proc. Natl. Acad. Sci. U. S. A. 14, 7993–7998
  23. Wimalawansa, S. (2000) Calcif. Tissue Int. 66, 56–60[CrossRef][Medline] [Order article via Infotrieve]
  24. van't Hof, R., and Ralston, S. (1997) J. Bone Miner. Res. 12, 1797–1804[CrossRef][Medline] [Order article via Infotrieve]
  25. Tsukahara, H., Miura, M., Tsuchida, S., Hata, I., Hata, K., Yamamoto, K., Ishii, Y., Muramatsu, I., and Sudo, M. (1996) Am. J. Physiol. 270, E840–E845
  26. McCartney-Francis, N., Song, X., Mizel, D., and Wahl, S. (2001) J. Immunol. 166, 2734–2740[Abstract/Free Full Text]
  27. Veihelmann, A., Landes, J., Hofbauer, A., Dorger, M., Refior, H., Messmer, K., and Krombach, F. (2001) Arthritis Rheum. 44, 1420–1427[CrossRef][Medline] [Order article via Infotrieve]
  28. Diwan, A., Wang, M., Jang, D., Zhu, W., and Murrell, G. (2000) J. Bone Miner. Res. 15, 342–351[CrossRef][Medline] [Order article via Infotrieve]
  29. Cuzzocrea, S., Mazzon, E., Dugo, L., Genovese, T., Di Paola, R., Ruggeri, Z., Vegeto, E., Caputi, A., Van de Loo, R., Puzzolo, D., and Maggi, A. (2003) Endocrinology 144, 1098–1107[Abstract/Free Full Text]
  30. Lee, S., Huang, H., Lee, S., Kim, K., Kim, K., Kim, H., Lee, Z., and Kim, H. (2004) Exp. Cell Res. 298, 359–368[CrossRef][Medline] [Order article via Infotrieve]
  31. Yu, X., Huang, Y., Collin-Osdoby, P., and Osdoby, P. (2003) J. Bone Miner. Res. 18, 1404–1418[CrossRef][Medline] [Order article via Infotrieve]
  32. Collin-Osdoby, P., Yu, X., Zheng, H., and Osdoby, P. (2003) in Methods in Molecular Medicine: Bone Research Protocols (Helfrich, M. H., and Ralston, S. H., eds) Vol. 80, p. 153, Humana Press Inc., Totowa, NJ
  33. Yu, X., Huang, Y., Collin-Osdoby, P., and Osdoby, P. (2004) J. Bone Miner. Res. 19, 2065–2077[CrossRef][Medline] [Order article via Infotrieve]
  34. Wright, L., Maloney, W., Yu, X., Kindle, L., Collin-Osdoby, P., and Osdoby, P. (2005) Bone 36, 840–853[Medline] [Order article via Infotrieve]
  35. Collin-Osdoby, P., Rothe, L., Anderson, F., Nelson, M., Maloney, W., and Osdoby, P. (2001) J. Biol. Chem. 276, 20659–20672[Abstract/Free Full Text]
  36. Blair, H., Zaidi, M., and Schlesinger, P. (2005) Biochem. J. 364, 329–341
  37. Pfeilschifter, J., Eberhardt, W., and Huwiler, A. (2003) J. Am. Soc. Nephrol. 14, S237–S240[Abstract/Free Full Text]
  38. Xie, Q., Whisnant R., and Nathan, C. (1993) J. Exp. Med. 177, 1779–1784[Abstract/Free Full Text]
  39. Weisz, A., Oguchi, S., Cicatiello, L., and Esumi, H. (1994) J. Biol. Chem. 269, 8324–8333[Abstract/Free Full Text]
  40. Park, S., Huq, M., Hu, X., and Wei, L. (2005) Mol. Cell. Proteomics 4, 300–309[Abstract/Free Full Text]
  41. Takada, Y., and Aggarwal, B. (2004) Blood 104, 4113–4121[Abstract/Free Full Text]
  42. Gyurko, R., Shoji, H., Battaglino, R., Boustany, G., Gibson, F., Genco, C., Stashenko, P., and Van Dyke, T. (2005) Bone 36, 472–479[Medline] [Order article via Infotrieve]
  43. Armour, K., Armour, K., van't Hof, R., Reid, D., Wei, X., Liew, F., and Ralston, S. (2001) Arthritis Rheum. 44, 2790–2796[CrossRef][Medline] [Order article via Infotrieve]
  44. Wei, S., Teitelbaum, S., Wang, W., and Ross, F. (2001) Endocrinology 142, 1290–1295[Abstract/Free Full Text]
  45. Reynaert, N., Ckless, K., Korn, S., Vos, N., Guala, A., Wouters, E., van der Vliet, A., and Janssen-Heininger, Y. (2004) Proc. Natl. Acad. Sci. U. S. A. 101, 8945–8950[Abstract/Free Full Text]
  46. Marshall, H., and Stamler, J. (2001) Biochemistry 40, 1688–1693[CrossRef][Medline] [Order article via Infotrieve]
  47. Park, H., Huh, S., Kim, M., Lee, S., and Choi, E. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 14382–14387[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Immunol.Home page
J. R. Kiesel, Z. S. Buchwald, and R. Aurora
Cross-Presentation by Osteoclasts Induces FoxP3 in CD8+ T Cells
J. Immunol., May 1, 2009; 182(9): 5477 - 5487.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
O. Uluckan, S. N. Becker, H. Deng, W. Zou, J. L. Prior, D. Piwnica-Worms, W. A. Frazier, and K. N. Weilbaecher
CD47 Regulates Bone Mass and Tumor Metastasis to Bone
Cancer Res., April 1, 2009; 69(7): 3196 - 3204.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
B. Sung, A. Murakami, B. O. Oyajobi, and B. B. Aggarwal
Zerumbone Abolishes RANKL-Induced NF-{kappa}B Activation, Inhibits Osteoclastogenesis, and Suppresses Human Breast Cancer-Induced Bone Loss in Athymic Nude Mice
Cancer Res., February 15, 2009; 69(4): 1477 - 1484.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Pathol.Home page
R. Modarresi, Z. Xiang, M. Yin, and J. Laurence
WNT/{beta}-Catenin Signaling Is Involved in Regulation of Osteoclast Differentiation by Human Immunodeficiency Virus Protease Inhibitor Ritonavir: Relationship to Human Immunodeficiency Virus-Linked Bone Mineral Loss
Am. J. Pathol., January 1, 2009; 174(1): 123 - 135.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
H. Ha, J.-H. Lee, H.-N. Kim, H. B. Kwak, H.-M. Kim, S. E. Lee, J. H. Rhee, H.-H. Kim, and Z. H. Lee
Stimulation by TLR5 Modulates Osteoclast Differentiation through STAT1/IFN-{beta}
J. Immunol., February 1, 2008; 180(3): 1382 - 1389.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
281/23/15809    most recent
M513225200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zheng, H.
Right arrow Articles by Osdoby, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zheng, H.
Right arrow Articles by Osdoby, P.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2006 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement