help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
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 arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sunyer, T.
Right arrow Articles by Collin-Osdoby, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sunyer, T.
Right arrow Articles by Collin-Osdoby, P.
Endocrinology Vol. 138, No. 5 2148-2162
Copyright © 1997 by The Endocrine Society

Articles

Ca2+ or Phorbol Ester But Not Inflammatory Stimuli Elevate Inducible Nitric Oxide Synthase Messenger Ribonucleic Acid and Nitric Oxide (NO) Release in Avian Osteoclasts: Autocrine NO Mediates Ca2+-Inhibited Bone Resorption1

Teresa Sunyer, Linda Rothe, David Kirsch2, Xinsheng Jiang, Fred Anderson, Philip Osdoby and Patricia Collin-Osdoby

Department of Biology and Division of Bone and Mineral Research, Washington University, St. Louis, Missouri 63130

Address all correspondence and requests for reprints to: Patricia Collin-Osdoby, Department of Biology, Box 1229, Washington Uni-versity, St. Louis, Missouri 63130. E-mail: collin{at}biodec.wustl.edu


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Osteoclast bone resorption is essential for normal calcium homeostasis and is therefore tightly controlled by calciotropic hormones and local modulatory cytokines and factors. Among these is nitric oxide (NO), a multifunctional free radical that potently inhibits osteoclast bone resorption in vitro and in vivo. Previous findings led us to propose that NO might serve as an autocrine, as well as paracrine, regulator of osteoclast function. This premise was investigated using isolated bone-resorptive avian osteoclasts and focusing on the inducible isoform of NO synthase (iNOS) responsible for inflammatory stimulated high-level NO synthesis in other cells. Avian osteoclasts expressed both iNOS messenger RNA (mRNA) and protein. However, inflammatory cytokines that induce iNOS mRNA, protein, and NO in other cells did not do so in avian osteoclasts, consistent with the known role of inflammatory stimuli in promoting osteoclast resorption and localized bone loss. In searching for potential modulators of osteoclast iNOS, protein kinase C activation [by phorbol 12-myristate 13-acetate (PMA)] and intracellular Ca2+ rises (A23187) were each found to elevate osteoclast iNOS mRNA and protein levels, while increasing NO release and reducing osteoclast bone resorption. The iNOS selective inhibitor aminoguanidine suppressed stimulated osteoclast NO production elicited by either signal, but reversed only the resorption inhibition due to raised Ca2+. Thus, whereas additional inhibitory signals are presumably coproduced in osteoclasts treated with PMA, osteoclast iNOS-derived NO may act as an autocrine signal to mediate Ca2+-inhibited bone resorption. These findings document for the first time an iNOS whose mRNA levels are regulated by Ca2+ or PMA, but not inflammatory stimuli, and the autocrine production of NO as a Ca2+ sensing signal to suppress osteoclast bone resorption. The unusual regulation of osteoclast iNOS makes it a potentially attractive target for designing novel therapeutic agents to alleviate excessive bone loss.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
OSTEOCLAST bone resorption is regulated by numerous hormonal and local regulatory factors that act in complex interactive pathways to govern their formation and activity, as well as that of other bone cells (1, 2). Among those local modulators known to have important roles in osteoclast development and function, the inflammatory cytokines interleukin-1 (IL-1), interleukin-6 (IL-6), tumor necrosis factor-{alpha} (TNF{alpha}), interferon-{gamma} (IFN{gamma}), and interleukin-8 (IL-8) have also been implicated in the pathogenesis of various inflammatory disorders typically associated with bone loss (3, 4). Upon inflammatory stimulation, many cells release high levels of another potent intercellular signal molecule, nitric oxide (NO), due to the de novo expression of an inducible isoform of NO synthase (iNOS, type II), and this sustained high level NO release is an important aspect of immune cell-mediated cytotoxicity and tumoricidal activity (5, 6). In addition to the role of NO in inflammation, this highly labile free radical participates as a multifunctional mediator in a wide range of other physiological and pathological processes, some of which involve transient bursts of relatively low levels of NO triggered by signals that elicit calcium/calmodulin enzymatic activation of either of two constitutive NOS (cNOS) isoforms, the endothelial (eNOS, type III) or brain/neuronal (nNOS, type I) isoenzymes (5, 6, 7, 8, 9).

Recently, NO has been found to serve as an important regulator in bone, affecting both osteoclast function and osteoblast proliferation/survival (10, 11, 12, 13, 14). Using exogenous NO donors, we and others have shown that NO potently inhibits, in a dose-dependent fashion, bone resorption in vitro by isolated rat (15) or avian osteoclasts (16) or in bone organ cultures (17, 18, 19), as well as in vivo resorption in rats (20). NO rises in response to inflammatory signals have also been linked with suppressed osteoclastic resorption in complex bone organ cultures (17, 18, 19). Conversely, inhibitors of NO production, including the iNOS-selective inhibitor aminoguanidine (AG) (21, 22), have stimulated bone resorption both in isolated osteoclast (16) or bone organ cultures (17, 19) and in rats in vivo (16, 23, 24), suggesting that basal NO levels may normally temper osteoclastic bone resorption. Other reports involving bone organ cultures have indicated that, whereas modest or high NO levels inhibit, low levels of NO may either not affect or enhance cytokine-mediated bone resorption in these systems (17, 18, 23). Together, these studies suggest that NO released as a paracrine inflammatory signal from osteoblasts, marrow stromal cells, or other cells in the local bone microenvironment can have a dramatic impact on osteoclast resorption and bone remodeling.

Recent studies have indicated that osteoclasts themselves might also serve as a source of NO in bone, signifying that NO may represent a rapid and sensitive autocrine regulator of osteoclast function. Thus, resorbing osteoclasts in avian and rat bone sections, as well as isolated avian osteoclasts engaged in bone resorption in culture, exhibited intense NADPH-diaphorase staining, potentially indicating substantial NOS activity (16, 23, 25). In addition, both iNOS and cNOS isoenzymes have been immunodetected in isolated osteoclasts and bone tissue sections of avian or rat origin (23, 26). Little is known, however, about the possible production of NO or the regulation of NOS isoenzymes in these bone-resorbing cells. Recently, we reported that iNOS messenger RNA (mRNA) and protein, NADPH-diaphorase activity, and NO release were markedly induced by typical iNOS-regulatory inflammatory stimuli (cytokines, bacterial endotoxin), as well as via novel responses to IL-8 or IL-10, in multinucleated giant cells formed in vitro from avian bone marrow mononuclear cell preparations containing osteoclast precursors (27). Inflammatory cytokines have similarly caused substantial increases in NO release from human preosteoclastic FLG 29.1 cells (23). Despite these findings, an inflammatory-mediated autocrine production of NO by osteoclasts seems inconsistent with the stimulatory effects that such inflammatory modulators have been reported to exert on osteoclast development, bone resorption, and localized osteopenia. We therefore undertook the present studies employing isolated highly purified in vivo formed mature avian osteoclasts to investigate whether these cells produced NO by an inflammatory-regulated iNOS-mediated mechanism and whether such NO release influenced osteoclast function. Our results indicate that avian osteoclasts express iNOS (mRNA and protein) and release NO in a noninflammatory regulated fashion, that NO can act as an autocrine signal to suppress osteoclast bone resorption, and that NO may mediate inhibited bone resorption caused by intracellular calcium increases in avian osteoclasts.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Osteoclast isolation, culture, and modulator treatments
Osteoclasts were isolated from the tibiae and humeri of White Leghorn chick hatchlings maintained for 28 days on a low calcium diet to increase osteoclast numbers as previously described (28). All animals were handled in accordance with the institutional Animal Care and Use Committee and standards approved by the NIH Guidelines for the Care and Use of Experimental Animals. Briefly, osteoclasts were released from bones by collagenase and trypsin treatments coupled with physical agitation, and cell suspensions were enriched for osteoclasts by filtration and fractionation on 35% and 6% Percoll (Pharmacia, Piscataway, NJ) gradients. Osteoclasts were resuspended for culture in phenol red-free medium 199 Earle’s salts supplemented with 8.3 mM NaHCO3, 100 mM HEPES (pH 6.8), 5% charcoal-stripped FCS (GIBCO BRL, Gaithersburg, MD), and 2.5% antibiotic/antimycotic (OC medium). Highly purified osteoclast cultures (6- to 10-fold more enriched than 6% Percoll preparations; 70–90% pure for osteoclasts) were obtained by immunoselection of osteoclasts after the 35% Percoll gradient using an osteoclast-specific monoclonal antibody (MAb) 121F coupled to magnetic beads (Dynal Inc., New Hyde Park, NY) as previously described (29). Osteoclasts in OC medium were seeded at 0.1–0.2 x 106 osteoclasts (in 250 µl)/well of a 48-well tissue culture dish containing one circular disc (6 mm in diameter) of devitalized bovine cortical bone, which occupied approximately 30% of the surface area of the well. Some experiments were repeated on ivory and each yielded similar results to those on bone. Cultures were maintained overnight at 37 C in a 95% air, 5% CO2 atmosphere before the medium was replaced with 250 µl fresh OC medium with or without the various modulators, and the cells further incubated for the specified times. The modulators used were: E. coli lipopolysaccharide (LPS) serotype 0111:B4 stored at 4 C as a 5 mg/ml stock in HBSS, AG freshly prepared as a 2 mM stock in culture medium, and PMA stored at -80 C as an 8 x 10-4 M stock in ethanol or dimethylsulfoxide (DMSO), all from Sigma Chemical Co. (St. Louis, MO), Ca2+-ionophore A23187 from Calbiochem (San Diego, CA) stored at -20 C as a 20 mM stock in DMSO, 4-{alpha}-PMA from Biomol (Playmouth Meeting, PA) stored at -20 C as a 1.5 x 10-3 M stock in DMSO, human recombinant cytokines IL-1{alpha}, TNF{alpha}, IL-8, and IL-10 from R & D Systems (Minneapolis, MN), and rat recombinant INF-{gamma} from GIBCO BRL. Cytokines were dissolved in PBS containing 0.1% BSA and stored as concentrated (~10-6 M) aliquots at -80 C.

Nitrite assay
NO production was measured based on the levels of nitrite, a stable end-product of NO, in the conditioned medium of cell cultures (briefly centrifuged), using the Greiss reaction in a microplate assay format (27, 30). Results were expressed as µM nitrite accumulated in the conditioned medium for times specified in the figure legends.

RT-PCR
Poly(A) selected RNA was extracted from freshly isolated highly (immunomagnetically) purified osteoclasts using the Quick Prep Micro mRNA purification kit (Pharmacia) and complementary DNA (cDNA) was synthesized using a cDNA cycle kit (Invitrogen, San Diego, CA). PCR was performed using the primers iNOS.F2 and iNOS.R2 as shown in Fig. 1CGo. The 362-bp product was subcloned into pCRII (TA cloning kit, Invitrogen) and sequenced by the dideoxy chain termination reaction method in both strands to completion using the standard techniques described in Sunyer et al. (27). The nucleotide sequence was compared with published cDNA sequences using computation performed at the National Center for Biotechnology Information and the BLAST network service.



View larger version (144K):
[in this window]
[in a new window]
  		Figure 1. Avian osteoclasts express iNOS and cNOS protein isoforms. A-D, Representative phase-contrast (A and B) and brightfield (C and D) photomicrographs of chicken tibial frozen sections reacted with iNOS PAb 587 (A and C) or eNOS PAb 620 (B and D), followed by an enzyme-conjugated secondary antibody and amplified detection system to immunolocalize NOS isoenzymes at sites of blue-green reaction product. Osteoclasts (arrows) associated with bone were specifically immunoreactive in this experiment and in additional trials with both iNOS and eNOS antibodies. Magnifications: A and B, x320; C and D, x1280. E and F, Photomicrographs of isolated avian osteoclasts cultured on bone, many of which are engaged in forming resorption pits (arrows), permeabilized and reacted with iNOS PAb 587 (E) and eNOS PAb 620 (F) antibodies, and subsequently photographed using brightfield reflective light microscopy. Magnification: x260. Similar staining to that shown with PAb 587 (to iNOS C-terminus) was obtained using iNOS PAb 586 (to iNOS N-terminus) for both avian bone sections and isolated osteoclasts seeded onto bone. No such staining was observed either in absence of primary antibody or when incubations were conducted using matched preimmune sera.

 
RNA isolation and ribonuclease protection assay (RPA)
Total RNA was obtained from osteoclast cultures by the method of Chomczynski and Sacchi (31) as previously described (27). Guanidine isothiocyanate denaturing solution (0.4 ml) was added to each well of a 48-well tissue culture dish at harvest time, immediately after removing the culture medium and without rinsing the cells, and the total RNA extracted from each well (2–5 µg) was used for a single RPA assay. RPA was performed using the RPA II kit (Ambion, Austin, Tx) and RNA probes generated from the 362-bp iNOS clone and a 203-bp GAPDH PCR clone as previously described (27). Hybridization was for 18 h at 45 C with 105 cpm iNOS and 104 cpm GAPDH. Protected RNA fragments were separated in a 8% polyacrylamide/8 M urea gel, detected by autoradiography, and quantified in a PhosphorImager (Molecular Dynamics, Sunnyvale, CA) using various exposure times within a linear detection range. Results for iNOS mRNA levels were expressed in arbitrary density units after normalizing for the corresponding GAPDH mRNA levels.

Immunolocalization of NOS
Detection of osteoclast inducible (i) and constitutive (c) NOS isoenzyme expression at the protein level was accomplished using three well-characterized rabbit polyclonal antibodies (PAbs) raised to the N-terminal (PAb 586)(32) or C-terminal (PAb 587)(33) regions of mouse iNOS, or to bovine endothelial eNOS (PAb 620) (34), which were generously provided by Dr. Tom Misko (Monsanto Corp., St. Louis, MO). These PAbs have been shown to specifically react (by Western blot, immunoprecipitation, or immunostaining) with the iNOS (586, 587) or eNOS (620) isoenzymes to which they were raised and to cross-react with homologous mouse, human, rat, or avian NOS isoforms. In a previous study, these PAbs (586, 587) provided a sensitive reflection of the inflammatory-induced iNOS elevation that occurred in avian osteoclast-like multinucleated giant cells (26). Frozen thin sections of bone were prepared from the tibiae of low calcium diet chickens and processed by the method of Sainte-Marie (35), blocked, reacted with the PAbs (1:50 dilution in block), and immunostaining localized using a highly sensitive amplified (biotin/streptavidin) colorimetric ß-galactosidase detection system as described in Sunyer et al. (27). Isolated osteoclasts cultured on bone slices or glass coverslips, with or without modulators, were rinsed three times, fixed, briefly permeabilized by methanol (-20 C, secs.), and immediately processed for immunostaining. Each independent staining trial contained control and treated osteoclasts obtained from the same isolated osteoclast preparation for valid comparisons. More than 30 avian bone sections and over 40 coverslips or bone discs on which isolated osteoclasts were cultured were analyzed. Parallel controls for bone section and isolated cell staining included no primary antibody incubations, corresponding preimmune sera incubations, and staining without prior cell permeabilization, all of which were negative. Sections and isolated osteoclasts on bone slices or coverslips were viewed and photographed with a Leitz Diaplan microscope (Ernst Leitz, Wefzlar, Germany).

Tartrate-resistant acid phosphatase (TRAP) staining and resorption analysis
Osteoclasts cultured on bone or ivory were rinsed, fixed, and stained for TRAP activity as previously described (16). The number of TRAP-positive osteoclasts was determined for a constant number of random fields per bone slice (ranging from 8–25 fields to evaluate a minimum of 80 osteoclasts, but held invariant within an experiment). More than 60 bone slices (>6000 individual osteoclasts) were analyzed to produce the resorption data reported here. Cells were subsequently removed for quantification of the resorption within these exact same fields by computer-linked darkfield reflective light image analysis. The total area of bone resorbed, the total number of pits formed, and the size of each pit excavated was assessed as described in Kasten et al. (16). Results were also normalized to compare the mean area resorbed per osteoclast (area/OC), mean number of pits formed per osteoclast (pits/OC), and the mean pit size (area/pit).

Statistical analysis
Data were presented as the mean ± SEM and represent 3–11 independent replicate cell cultures. Differences between treatments were analyzed using single-factor ANOVA. For simultaneous comparisons between multiple treatments, significant differences were determined using the post-ANOVA Bonferroni test. Differences were considered significant for P < 0.05.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Avian osteoclasts express iNOS
The presence of iNOS isoenzymes in avian osteoclasts was investigated in chicken tibial sections as well as in isolated avian osteoclasts cultured on bone slabs by immunohistochemical staining using PAbs raised against murine macrophage iNOS and endothelial eNOS isoforms. Osteoclasts were specifically reactive with PAb 587 (to the C-terminal domain of iNOS, Fig. 1Go, A and C) and PAb 586 (to the N-terminal domain of iNOS, not shown). Many, but not all, resorbing osteoclasts examined in over 30 bone sections were well stained with the anti-iNOS PAbs, and those situated near cartilagenous remodeling zones were strongly reactive. In addition, osteoclasts exhibited specific immunoreactivity with the anti-eNOS PAb 620 (Fig. 1Go, B and D). Other bone cells, such as osteoblasts, and some small cells (but not others) located within the bone marrow were also recognized by the anti-iNOS (PAb 586, PAb 587) and eNOS (PAb 620) antibodies, whereas vascular endothelial cells surrounding the blood vessels of bone notably reacted with the eNOS (PAb 620) but not with the iNOS (PAb 586, 587) antibodies. Similar to osteoclasts in bone sections, isolated avian osteoclasts cultured on pieces of bone were specifically reactive with each of the iNOS antibodies (Fig. 1EGo) and the cNOS antibody (Fig. 1FGo).

To further evaluate whether avian osteoclasts expressed an iNOS, RT-PCR was performed using mRNA obtained from highly purified (immunomagnetically sorted) osteoclasts and chicken-specific iNOS primers (iNOS.F2 and iNOS.R2). A single amplicon of the expected size was obtained (Fig. 2AGo), which was subcloned and sequenced in both strands to completion. The nucleotide sequence of the osteoclast clone was identical to the iNOS transcript expressed in avian bone marrow-derived multinucleated giant cells and 76% homologous to human, rat, and mouse iNOS genes (Fig. 2BGo).



View larger version (24K):
[in this window]
[in a new window]
  		Figure 2. Avian osteoclasts express iNOS mRNA. A, RT-PCR amplicon obtained using highly purified chicken osteoclast mRNA and avian-specific iNOS primers, separated in an agarose gel and stained with ethidium bromide. Lane 1, size markers. Lane 2, PCR reaction product. B, Nucleotide sequence of 362-bp iNOS clone shown in A. Primers used to generate PCR amplicon are underlined.

 
Inflammatory mediators do not elicit osteoclast nitrite production despite modulating osteoclast bone pit resorption
Because inflammatory agents characteristically up-regulate iNOS in various cells, including in vitro formed avian bone marrow-derived multinucleated cells, such modulators were tested for their ability to elicit nitrite release from isolated in vivo formed authentic osteoclasts. Whereas bacterial LPS, IL-8, IL-10, or various combinations of IL-1, TNF{alpha}, and INF{gamma} were formerly shown to markedly raise nitrite release as well as iNOS mRNA and protein levels in avian bone marrow multinucleated giant cells, none of these modulators alone or in various combinations stimulated nitrite production from isolated authentic avian osteoclasts cultured on bone, even when the exact same lot of each agent was used (Table 1Go). To discount the possibility that mononuclear cells remaining in the Percoll fractionated osteoclast preparations were somehow interfering with the detection or induction of osteoclast nitrite release, a subset of these studies were repeated using highly purified (immunomagnetically sorted) osteoclast preparations and these yielded the same results (denoted by superscript c, Table 1Go). Consistent with the inability of inflammatory agents to elicit osteoclast nitrite release, RPA demonstrated that the low basal steady state iNOS mRNA levels associated with osteoclast control cultures were not significantly elevated following their treatment with LPS and various combinations of IL-1{alpha}, INF{gamma}, and/or TNF{alpha}. Thus, for example, the iNOS/GAPDH mRNA levels in osteoclast cultures treated with all four of these agents were 126 ± 33% that of control cultures, in marked contrast to the approximate 1200% rise that occurs in response to the same inflammatory mediators in avian multinucleated giant cells (26, 27). To establish that the lack of inflammatory regulation of nitrite release and iNOS mRNA levels did not stem from an inherent unresponsivity of the osteoclasts to these stimuli, several of these agents were selected for an evaluation of their effectiveness in modulating osteoclast bone resorptive activity. Whether resorption was significantly inhibited by IL-8 (~55%), or stimulated by TNF{alpha} (~2.4-fold) or IL-10 (~1.6-fold), corresponding nitrite levels were not altered in the osteoclast cultures at modulator concentrations proven potently bioactive for eliciting rises in nitrite and iNOS mRNA levels in the related in vitro formed avian bone marrow-derived multinucleated cells (Table 1Go). Therefore, nitrite production and iNOS mRNA levels are not regulated by these inflammatory stimuli in purified avian osteoclast preparations.


View this table:
[in this window]
[in a new window]
  		Table 1. Lack of induction by inflammatory stimuli of nitrite release in avian osteoclasts

 
Protein kinase C (PKC) activation increases osteoclast nitrite release and iNOS mRNA levels
Because the PKC activator PMA has been shown to dampen the inflammatory-mediated induction of iNOS in a number of cell types, but to stimulate iNOS expression in others, we investigated whether PMA might have an effect on nitrite release by isolated avian osteoclasts cultured on bone. Nitrite production was induced by PMA in a dose-dependent manner, reaching near-maximal levels in the conditioned medium of osteoclast cultures by 10-7 M PMA (Fig. 3AGo). Stimulation was not observed with the inactive phorbol ester {alpha}-PMA (10-7 M). Time course experiments documented that the nitrite content of the conditioned medium from osteoclast cultures treated with 10-7 M PMA increased over 24 h of incubation, achieving maximal rates (~0.165 nmol/h) within the first 6.5 h (Fig. 3BGo). The elevated nitrite production in response to PMA stimulation of osteoclasts was fully inhibited by AG, a selective competitive inhibitor of iNOS. AG also inhibited the basal nitrite release from osteoclasts by approximately 60% (Fig. 3CGo). Whether PMA caused an increase in osteoclast nitrite release as a consequence of up-regulating the steady state levels of iNOS mRNA was investigated by RPA using both enriched osteoclast cultures (Percoll fractionated) and highly purified osteoclast cultures (immunomagnetically sorted) with similar results. Thus, PMA treatment (10-7 M, 6.5 h) of osteoclasts led to a significant elevation of iNOS mRNA levels (3.5-fold over control) in parallel with increased nitrite release (2.5-fold over control) into the culture medium (Fig. 3DGo).



View larger version (40K):
[in this window]
[in a new window]
  		Figure 3. PMA stimulation of nitrite production and iNOS mRNA levels in osteoclasts. A, Dose-dependent stimulation of nitrite production by PMA. Osteoclasts cultured on bone or ivory slices were incubated for 24 h with indicated PMA concentrations or 10-7 M {alpha}PMA, after which nitrite levels were determined in conditioned media as a measure of NO release. Data was obtained from three experiments and means ± SEM are presented as a percentage of control. B, Time-dependent stimulation of nitrite production by PMA. Osteoclasts were incubated with 10-7 M PMA for various times (1.5, 4, 6.5, 24 h) and nitrite content was measured in conditioned media (circles). Nitrite levels in untreated cultures were measured at 4 and 24 h (crosses). Data shown is from a representative experiment of isolated osteoclasts cultured on bone that was repeated three times with comparable results. C, Inhibition of control and PMA-stimulated nitrite production by AG. Osteoclasts cultured on bone or ivory were incubated in absence or presence of AG (0.5 mM) and/or PMA (10-7 M), and nitrite levels were measured after 24–48 h. Data was obtained from seven independent cultures and means ± SEM are presented as a percentage of nitrite levels in control cultures. Significant differences from control cultures are denoted by *, P < 0.05 and ****, P < 0.001, and significant differences from PMA treated cultures by ++++, P < 0.001, based on post-ANOVA Bonferroni test for multiple comparisons. D, PMA induction of iNOS mRNA steady state levels. Highly purified osteoclasts cultured on bone were incubated for 6.5 h in absence or presence of PMA (10-7 M) and total RNA was analyzed by RPA. Insert shows protected iNOS band (*) from a representative experiment. Lower bands are protected GAPDH transcripts used to normalize data. Protected mRNA from nine independent experiments were quantified in a PhosphorImager and the results expressed as iNOS/GAPDH mRNA are indicated in comparison to nitrite content of their corresponding culture media. Significant differences from control cultures are denoted by ***, P < 0.005 and **, P < 0.01.

 
Ca2+-ionophore A23187 increases osteoclast nitrite release and iNOS mRNA levels
Because elevated intracellular Ca2+ ([Ca2+]i) levels have been linked to PKC activation and osteoclast resorption inhibition, the potential effects of the Ca2+-ionophore A23187 on nitrite release and iNOS mRNA expression were investigated in osteoclast cultures on bone. In time course experiments, A23187 (1 µM) stimulated nitrite release into the medium of osteoclast cultures over a 24-h period, with maximal rates (~0.179 nmol/h) within the first 6.5 h (Fig. 4AGo). Like PMA, nitrite induction by A23187 (7-fold over control by 6.5 h) was dependent on an iNOS activity, because cotreatment of osteoclasts with AG and A23187 fully inhibited the rise in nitrite due to A23187 alone (Fig. 4BGo). Because avian osteoclasts also contain cNOS, and Ca2+ is well known to stimulate cNOS-mediated nitrite production, it was possible that the A23187 calcium-dependent rise in osteoclast nitrite production was simply due to cNOS activation. However, RPA analysis demonstrated that A23187 treatment (1 µM, 6.5 h) consistently led to a significant elevation (4-fold over control) in the steady state levels of osteoclast iNOS mRNA expression (Fig. 4CGo). A similar induction of nitrite and iNOS mRNA in response to A23187 was obtained using either highly purified or Percoll-enriched osteoclast preparations. Therefore, a rise in [Ca2+]i induces nitrite release and increased iNOS mRNA levels in avian osteoclasts.



View larger version (25K):
[in this window]
[in a new window]
  		Figure 4. Ca2+-ionophore A23187 stimulation of nitrite production and iNOS mRNA levels in osteoclasts. A, Time-dependent stimulation of nitrite production by A23187. Osteoclasts cultured on bone were treated with 1 µM A23187 for various times (1.5, 4, 6.5, 24 h), and nitrite content was measured in conditioned media (circles). Nitrite levels in untreated cultures were measured at 4 and 24 h (crosses). Data shown is from a representative experiment that was repeated three times with comparable results. B, Inhibition of A23187-stimulated nitrite production by AG. Osteoclasts cultured on bone were incubated for 6.5 h in absence or presence of A23187 (1 µM), with or without AG (0.5 mM), and nitrite levels were measured in conditioned media. Data was obtained from at least eight independent cultures. Significant differences from control cultures are denoted by ****, P < 0.001, and significant differences from A23187-treated cultures by ++++, P < 0.001 as determined by Bonferroni post-ANOVA test for multiple comparisons. C, A23187 induction of iNOS mRNA steady state levels. Osteoclasts cultured on bone were incubated for 6.5 h in absence or presence of A23187 (1 µM), and total RNA was analyzed by RPA. Insert shows protected iNOS band (*) from a representative experiment. Lower bands are protected GAPDH transcripts used to normalize data. iNOS/GAPDH ratios were obtained by PhosphorImager analysis of four independent experiments. Significant differences from control cultures are denoted by ****, P < 0.001.

 
PMA and A23187 elicit similar temporal profiles for rises in osteoclast iNOS mRNA levels
Osteoclast iNOS mRNA steady state levels were raised in time course experiments as early as 1.5 h following the addition of either PMA or the Ca2+-ionophore A23187 (Fig. 5Go, A and B). Although A23187 (1 µM) was more potent than PMA (10-7 M) at every time point measured, both agents elevated iNOS mRNA expression with a similar overall temporal profile: iNOS mRNA attained maximal levels between 4 and 6.5 h of treatment, and these peaks declined to near basal levels by 24 h. Cotreatment of osteoclasts with both PMA and A23187 led to an iNOS mRNA expression profile indistinguishable from that of A23187 alone (Fig. 5BGo), indicating that maximal osteoclast iNOS levels were already achieved in the presence of the Ca2+-ionophore.



View larger version (32K):
[in this window]
[in a new window]
  		Figure 5. PMA and A23187 temporal effects on iNOS mRNA levels in osteoclasts. Osteoclasts cultured on bone were incubated for various times (1.5, 4, 6.5, 24 h) in absence of modulators (open symbols), or in presence of 10-7 M PMA (circles), 1 µM Ca2+-ionophore A23187 (squares), or both agents simultaneously (triangles). iNOS mRNA levels were determined by RPA. A, RPA analysis of a representative experiment. B, Protected bands from two independent cultures were quantified in a PhosphorImager and expressed as a percentage of maximal iNOS mRNA induced by A23187 in 6.5 h.

 
Osteoclasts cultured on bone produce more nitrite and exhibit higher iNOS mRNA levels than those on plastic
Calcium released during the course of normal osteoclastic bone resorption has been shown to transiently elevate osteoclast [Ca2+]i, and based on the foregoing evidence, such Ca2+ rises might therefore influence osteoclast iNOS mRNA levels and nitrite production. To investigate whether resorbing osteoclasts differed from nonresorbing osteoclasts in their levels of iNOS mRNA and nitrite production, osteoclasts were seeded into tissue culture dishes that contained bone slices in some wells, and the cultures were monitored over time for nitrite release into the medium and steady state iNOS mRNA expression. Both parameters were approximately 2.5-fold higher for osteoclast cultures plated on bone than those on plastic (Fig. 6AGo), and differences between bone and plastic were maintained throughout the 24-h culture period (Fig. 6BGo). Bone was not, however, required to obtain a rise in nitrite or iNOS mRNA in osteoclasts by either PMA or A23187, because osteoclasts cultured on plastic were also responsive to PMA and A23187. Thus, the nitrite levels associated with osteoclasts cultured on plastic were increased by PMA (2.3-fold over control), A23187 (3.0-fold over control), or the combined treatment of PMA together with A23187 (3.6-fold over control). iNOS mRNA levels increased following either PMA or A23187 treatment (both 6.5-fold over control) and especially in response to their combined administration (17.5-fold over control) (Fig. 6CGo). Therefore, increased [Ca2+]i or PKC activation induces nitrite formation and raises iNOS mRNA levels in both resorbing and nonresorbing avian osteoclasts. In addition, resorbing osteoclasts attached to bone exhibit greater nitrite production and iNOS mRNA steady state levels than do nonresorbing osteoclasts.



View larger version (28K):
[in this window]
[in a new window]
  		Figure 6. Osteoclast adherence to bone stimulates nitrite production and iNOS mRNA levels. A, Osteoclast cultures were plated in tissue culture wells with or without a devitalized bone slice (occupying ~30% of well surface area). Nitrite production was measured in conditioned media of replicate cultures (n = 11) after 24 h, and mean ± SEM expressed as a percentage of nitrite levels in osteoclast cultures on plastic. iNOS mRNA levels were determined by RPA from four independent osteoclast cultures after 6.5 h of incubation on plastic or bone. B, RPA analyses were performed using at least two osteoclast preparations cultured for various times (1.5, 6.5, 24 h) on plastic or bone slices and iNOS/GAPDH mRNA ratios were determined. C, Osteoclasts were cultured on plastic in absence (control) or presence of PMA (10-7 M), A23187 (1 µM), or both modulators simultaneously (P + A) for 6.5 h, after which nitrite was measured in conditioned media and relative levels of iNOS mRNA were analyzed by RPA. Data shown was obtained from two to four independent osteoclast cultures. iNOS mRNA levels are expressed as fold elevation over control cultures after normalization to GAPDH mRNA. Significant differences from control cultures are denoted by ***, P < 0.005; **, P < 0.01; *, P < 0.05.

 
iNOS protein expression parallels nitrite and iNOS mRNA rises due to PMA, A23187, or bone in osteoclasts
If nitrite production is elevated as a consequence of an up-regulation of osteoclast iNOS mRNA steady state levels in response to PMA, A23187, or culture on bone, then such induction should be reflected in greater iNOS isoenzyme protein levels in osteoclasts. Evidence for this was obtained from immunostaining of untreated and treated osteoclasts with the two specific antibodies directed against the C-terminal (PAb 587) and N-terminal (PAb 586) domains of iNOS. Basal immunoreactivity with iNOS PAbs was detectable in untreated osteoclasts cultured on glass coverslips (Fig. 7AGo) and was notably increased following their exposure to PMA (Fig. 7BGo) or A23187 (not shown), consistent with the increased nitrite release and iNOS mRNA levels resultant from these treatments. In comparison with the basal iNOS PAb immunoreactivity of untreated osteoclasts cultured on glass (Fig. 7AGo) or plastic, stronger staining was generally evidenced by osteoclasts cultured on bone, many of which were associated with clearly defined resorption pits (Fig. 7CGo). This coincides with the higher nitrite production and iNOS mRNA levels demonstrated by osteoclasts adherent to bone. As with osteoclasts cultured on glass, the PAb 587 immunoreactivity of bone-resorbing osteoclasts was enhanced by exposure to PMA and/or A23187 (Fig. 7Go, D and E). By contrast, osteoclasts cultured either on bone or on glass coverslips did not exhibit any detectable alteration in their immunostaining with the iNOS PAbs in response to their treatment with an inflammatory cocktail comprised of IL-1{alpha}, LPS, and IFN{gamma} (not shown). In trials in which cells were subsequently stained for TRAP activity, all of the osteoclasts were TRAP-positive, and the majority were reactive with the iNOS PAbs. Some of the small mononuclear cells present in the cultures were also immunostained by the iNOS PAbs, but none of these were TRAP stained (not shown). Therefore, osteoclast induction by PMA, A23187, or culture on bone leads to increased iNOS mRNA levels, which is manifested in greater iNOS protein levels and nitrite synthesis in both resorbing and nonresorbing osteoclasts.



View larger version (120K):
[in this window]
[in a new window]
  		Figure 7. Immunodetection of iNOS protein expression induced by PMA or A23187 treatment of isolated avian osteoclasts. Representative photomicrographs of isolated avian osteoclasts cultured with or without bone, reacted with iNOS PAb 587 after various treatments, and photographed using brightfield reflective light microscopy. A and B, Osteoclasts (arrowheads) cultured on glass coverslips were incubated for 44 h in absence (A) or presence (B) of 10-7 M PMA. Note spread osteoclasts in (A) and moderate reactivity with iNOS PAb 587, in comparison with greater immunostaining and more contracted appearance of majority of osteoclasts in (B). Magnifications: A and B, x256. C-E, Osteoclasts plated onto bone slices were treated for 22 h in absence (C) or presence of 10-7 M PMA (D), 1 µM A23187 (E), or both modulators simultaneously (insert in E). Whereas a majority of untreated osteoclasts on bone were moderately stained in numerous trials, a few exhibited stronger reactivity with PAb 587; shown is a field containing osteoclasts of both immunophenotypes to illustrate this point (C). Following PMA or A23187 treatment, majority of osteoclasts on bone became strongly reactive with PAb 587 (D and E). Similar results were obtained in at least three additional trials using independent osteoclast preparations; in all cases, stronger iNOS immunoreactivity was detected in cells treated with PMA or A23187 over that of control cultures. To ensure accurate assessments of relative immunostaining, similarly sized osteoclasts in treated and untreated samples were also carefully compared side by side. Analogous findings on bone or glass coverslips were obtained using iNOS PAb 586 across these trials (not shown). PMA-treated cells consistently appeared more contracted than control or A23187 treated cultures. Immunostained osteoclasts were often associated with resorption pits (arrows). Note fewer number of resorption pits formed in PMA or A23187 treated cultures. Magnification: C–E, x260.

 
PMA and A23187 inhibit osteoclast bone pit resorption via mechanisms involving NO
Because PMA, Ca2+ rises, and high NO levels have been shown to inhibit osteoclastic bone resorption, the effects of raising iNOS and nitrite levels via PMA or A23187 induction on bone pit formation were investigated in avian osteoclasts. Quantitative resorption pit analyses revealed that PMA dose dependently inhibited the mean area of bone resorbed per osteoclast, the average number of pits formed per osteoclast, and the mean pit size formed by osteoclasts (Fig. 8Go, A-C), without altering the number of TRAP-positive osteoclasts attached to the bone slices (Table 2Go). Maximal inhibition was achieved with 10-7 M PMA, and no further decline in any of these resorption parameters occurred with 10-6 M PMA. Osteoclasts appeared to have contracted on the bone following PMA administration, consistent with the smaller excavation sites formed. The inactive phorbol ester {alpha}-PMA, which had failed to induce nitrite production in osteoclast cultures, did not affect osteoclast bone pit resorptive activity (not shown). Resorption analyses of osteoclast cultures treated with A23187 revealed that the Ca2+-ionophore significantly inhibited the mean area of bone resorbed per osteoclast (60%) and the number of pits formed per osteoclast (41%), while slightly reducing the average area per pit (26%), to a similar degree as had PMA treatment (61%, 56%, and 22%, respectively) (Table 2Go). In contrast to PMA, A23187 appeared to have generally increased osteoclast cell spread areas. As with PMA, A23187 treatment did not alter the number of TRAP-positive osteoclasts attached to the bone (Table 2Go).



View larger version (33K):
[in this window]
[in a new window]
  		Figure 8. PMA dose-dependent inhibition of osteoclast bone pit resorption. Osteoclast cultures were plated into wells containing ivory slices and treated with indicated PMA concentrations for 30 h. Cells were subsequently fixed, stained for TRAP activity, and number of osteoclasts within eight consecutive random fields were counted (at least 130 osteoclasts per ivory slice for each condition). Resorption pit numbers and excavation areas were analyzed across these same fields after removal of cells, and results were normalized for numbers of osteoclasts or pits. Means ± SEM of areas resorbed (µm2) per osteoclast (A), numbers of pits formed per osteoclast (B), and areas resorbed (µm2) per pit (C) are shown. Comparable results were obtained in at least two additional trials using independent osteoclast cultures plated on devitalized bone slices.

 

View this table:
[in this window]
[in a new window]
  		Table 2. PMA, A23187, and AG regulation of osteoclast bone pit resorption

 
AG, the competitive iNOS inhibitor that earlier was shown to reduce PMA- or A23187-mediated nitrite induction levels, also fully reversed the inhibitory effects of A23187 on osteoclast bone pit resorption, but not those of PMA. Thus, AG treatment resulted in a complete recovery of the A23187-induced inhibition of resorption, because both the area resorbed per osteoclast and the number of pits per osteoclast returned to control levels when AG was added together with A23187 to the osteoclast cultures (Fig. 9AGo). The area per pit, which was inhibited by only 26% in response to A23187 treatment, exhibited a 45% recovery in the presence of AG, although neither of these levels was significantly below that of control. In contrast, only a statistically insignificant trend towards the partial recovery of PMA-induced inhibition of resorption by AG was seen (Fig. 9BGo). AG itself slightly increased the basal area resorbed per osteoclast (19%), the number of pits formed per osteoclast (26%), the mean area per pit (10%), and the number of osteoclasts adherent to bone or ivory (Table 2Go). Therefore, increased levels of iNOS and nitrite production in response to PMA or A23187 stimulation were associated with an inhibition of osteoclast bone pit resorption activity, whereas reversal of induced nitrite levels by AG alleviated only the A23187-dependent inhibition of osteoclast activity.



View larger version (34K):
[in this window]
[in a new window]
  		Figure 9. AG reversal of inhibitory effects of PMA or A23187 on osteoclast bone pit resorption. A and B, Osteoclasts cultured on bone slices were treated for 30 h with A23187 (1 µM) (A) or PMA (10-7 M) (B), in absence (-AG) or presence (+AG) of AG (0.5 mM). Resorption pit analysis was performed as in Fig. 8Go. Each data point was obtained from at least 5 independent osteoclast cultures containing a bone slice, and at least 10 consecutive random fields were quantified for each bone slice. Means ± SEM of areas resorbed per osteoclast (A/Oc), numbers of pits formed per osteoclast (P/Oc), and resorption areas per pit (A/P) were determined for experimental and control osteoclast cultures, and results expressed as a treated to control ratio (1.0 signifying no change from control). Significant differences were determined by post-ANOVA Bonferroni test for AG-treated cultures from those comparably treated but not exposed to AG and are denoted by *, P < 0.05 and **, P < 0.001.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The results of this study demonstrate at a molecular, protein, and physiological level that mature in vivo formed bone-resorptive avian osteoclasts express an iNOS whose mRNA levels are regulated in an unusual manner by raised [Ca2+]i or PKC activation but not by inflammatory stimuli, and further, that the NO released may serve as an autocrine signal to inhibit osteoclast bone resorptive activity. To our knowledge, this is the first reported instance of an iNOS that is not induced by inflammatory cytokines or agents, but is instead responsive at a message level to other stimuli. Because of the important implications of these findings, we tried to eliminate potential alternative explanations for the inability of inflammatory agents to elevate iNOS in osteoclasts by demonstrating in concurrent studies that these same lots and concentrations of inflammatory cytokines or LPS could markedly raise iNOS mRNA and protein levels, together with NO release, in closely related avian bone marrow-derived multinucleated giant cells (27) as well as influence avian osteoclast bone resorption here and in other studies (36, 37). Although we cannot completely exclude the possibility that avian osteoclasts might not possess receptors for one or more of these stimuli, and therefore, the effects of some of these cytokines on osteoclast resorption might have been mediated by other cells in the purified osteoclast cultures, this seems unlikely, because many such cells (e.g. osteoblasts, marrow stromal cells, macrophages) typically respond to these inflammatory cytokines with large rises in iNOS/NO, which should have been detectable in our system. Moreover, although it is not always feasible to evaluate rapid direct actions of particular resorption modulators on osteoclast physiology, we have shown that IL-8 dose dependently inhibits osteoclast bone resorption and causes rapid (within minutes) morphological shape changes in individual avian osteoclasts, consistent with a direct action of IL-8 on osteoclasts (37). Given that IL-8 potently stimulates iNOS/NO in avian giant cells but not in osteoclasts, clear differences exist in how these closely related avian cell types respond to this specific inflammatory mediator. Therefore, we conclude that the above inflammatory stimuli are bioactive and they can act on avian cells to elevate iNOS/NO, and further, that avian osteoclasts can functionally respond to at least some of these potent inflammatory signals in other ways but not with rises in iNOS mRNA, protein, or NO release. It is possible that some unknown combination of inflammatory stimuli, apart from those tested in this study, that are typically inductive for other cells, could elicit an iNOS/NO response in avian osteoclasts.

Although these findings might have suggested that osteoclast iNOS was not subject to regulation at an mRNA level, our subsequent search for potential inductive stimuli led to the discovery of other signals (PMA, Ca2+) that significantly raised osteoclast iNOS mRNA levels, and consequently, iNOS isoenzyme levels and NO production. Because osteoclasts also express a Ca2+-sensitive cNOS, some contribution to the rise in NO release following PMA or Ca2+ stimulation could have been due to activation of this cNOS. However, Ca2+ has not been reported to elevate mRNA levels of either cNOS or iNOS isoenzymes (5, 6, 7, 8, 9), whereas our RPA data showed time- and dose-dependent increases in steady state iNOS mRNA by PMA or Ca2+. Therefore, our findings suggest novel regulatory mechanisms for iNOS mRNA and NO production in osteoclasts. Moreover, the bulk of the NO generated in response to either PMA or Ca2+ probably derived from increased iNOS isoenzyme levels, because the fold elevations in osteoclast iNOS mRNA were sufficient to account for the observed rises in NO production. One could question whether the probe used in the present studies was selective for iNOS, however, such a specificity is supported by the following: 1) the iNOS probe was generated using primers specific to conserved regions of iNOS for human, rat, and mouse that did not match corresponding regions of cNOS isoforms; 2) the sequence of this avian NOS probe was most homologous with corresponding regions of other iNOS isoenzymes and less similar to cNOS isoforms from multiple sources; 3) this probe does not recognize the cNOS of avian vascular endothelial cells or brain tissue (27); 4) this same probe accurately reflected in avian multinucleated giant cell studies the dramatic upregulation of iNOS mRNA in a classical manner by inflammatory stimuli, together with NO production, NADPH-diaphorase staining, and iNOS PAb immunostaining (27); and 5) the results obtained relative to osteoclast iNOS regulation with this probe were independently supported by immunostaining for iNOS protein with two iNOS PAbs (586, 587), thus indicating that Ca2+ ionophore and PMA, but not inflammatory stimuli, elevated osteoclast iNOS mRNA, protein, and NO production. Certain aspects of osteoclast iNOS regulation therefore appear to be novel and have not been reported previously for other iNOS-containing cells. Relative to PMA, this agent frequently interferes with the inflammatory-mediated induction of iNOS in various cells (5, 6, 7, 8, 9), although PKC activation has also been reported not to influence iNOS induction in articular chondrocytes (38) or to promote iNOS induction in macrophages (39) or Swiss 3T3 cells (40). However, unlike bone-resorptive osteoclasts, these other PMA-regulated cells have all exhibited typical iNOS induction responses to inflammatory mediators. Relative to Ca2+, we are not aware of any reports in the literature that demonstrate increased iNOS mRNA levels by Ca2+ ionophore, and so, the Ca2+-mediated elevation of osteoclast iNOS mRNA steady state levels appears to pose the first demonstration of this type of control. It is possible that this unusual regulation by Ca2+ pertains to the unique physiological role that osteoclasts have in bone remodeling and Ca2+ homeostasis.

Contrary to our findings with mature in vivo formed resorptive osteoclasts, inflammatory mediators have been quite effective at increasing iNOS in preosteoclastic human FLG 29.1 multinucleated cells (41) or in osteoclast-like multinucleated giant cells formed in vitro from avian mononuclear bone marrow precursors exposed to developmental promoters (27). In addition, preliminary experiments have indicated that PMA does not raise iNOS/NO in these avian giant cells, whereas it does so in avian osteoclasts. In vitro formed differentiated multinucleated cells, which differ from macrophages in multiple respects, represent the most closely related lineage cell type known to in vivo formed osteoclasts (41, 42). However, despite their similarities to osteoclasts in various molecular, antigenic, and biochemical properties, neither of these partially differentiated related cell types is fully equivalent to mature functional osteoclasts, because they do not excavate resorption pits on bone nor evidence the full complement of phenotypic traits characteristic of osteoclasts (26, 41, 42, 43). Based on RT-PCR cloning and sequence analysis, the iNOS genes of avian osteoclasts and avian bone marrow-derived giant cells are identical over a nearly 400-bp uninterrupted span (27). Thus, it appears that the same iNOS gene may be differentially regulated, whether by transcriptional and/or posttranscriptional mechanisms, between two closely related cell types of hematopoietic origin. If the contrasting findings between avian osteoclasts and giant cells reflects changes in iNOS regulatory circuits that occur during the transition of preosteoclasts to mature bone-resorptive osteoclasts, then iNOS regulation may serve as another defining feature to distinguish functional osteoclasts from other cells belonging to the monocyte/macrophage lineage.

Both PKC activation and elevation of cytosolic Ca2+ inhibited avian osteoclast bone resorption in this study, consistent in temporal and concentration profiles with their reported actions in causing cell contraction and resorption inhibition for osteoclasts from various species (44, 45, 46, 47). AG, a competitive selective inhibitor of iNOS (21, 22), completely suppressed the stimulated osteoclast NO release elicited by PMA or the Ca2+ ionophore A23187. The return to basal NO levels by AG was not, however, sufficient to reverse inhibited osteoclast bone resorption caused by PMA, suggesting that additional PMA-generated inhibitory signals may be coproduced that independently inhibit osteoclast function (although NO may also contribute to such inhibition). By contrast, the return to basal NO levels by AG was associated with the full recovery of resorptive activity in A23187-treated osteoclasts, indicating that NO may serve as a Ca2+-elicited autocrine signal to limit osteoclast bone resorption. Previously, AG was shown to inhibit the inflammatory-mediated rise in NO release by avian giant cells (27), to dose dependently stimulate in vitro avian osteoclast bone resorption (16), and to cause a significant loss in the bone mineral densities of spinal and femoral sites, while decreasing urinary NO2/NO3 excretion, in normal or ovariectomized rats administered AG in vivo, effects that were reversed by concomitant L-arginine supplementation (16, 24). Therefore, the current findings complement and extend these former studies while suggesting that the mechanism by which an [Ca2+]i rise inhibits osteoclast function involves an elevation in osteoclast iNOS mRNA and protein with an ensuing autocrine production of the potent antiresorptive inhibitor NO.

The present findings provide one possible explanation for the apparent paradox over how inflammatory cytokines known to enhance osteoclast development and activity could ultimately stimulate osteoclast bone resorption if such signals simultaneously induced iNOS/NO in osteoclasts. Thus, inflammatory cytokines that affect osteoclast bone resorption did not stimulate osteoclasts to produce NO, and so, did not evoke release of a conflicting autocrine inhibitory signal that could oppose their actions in potentiating osteoclast bone resorption. Although these findings obtained with isolated osteoclasts are consistent with the overall established linkage between elevated levels of inflammatory cytokines and localized bone loss associated with such disorders as rheumatoid arthritis, periodontal disease, or trauma (3, 4), additional complexities must also be considered, because other cells within the bone microenvironment can respond to inflammatory stimuli with an elevation of iNOS/NO (17, 18, 19, 48, 49, 50). In acute inflammatory conditions, NO has been detected at up to 1–5 µM concentrations (51), levels shown in this and additional studies using 100 µM sodium nitroprusside (SNP) as an exogenous NO donor (that yields 5 µM nitrite in our osteoclast cultures) to be in the range sufficient for inhibiting osteoclast resorption. However, given these NO levels, why does bone loss typically occur in inflammatory diseases? The answer presumably lies in our understanding that NO does not function alone, but rather, as part of a complicated interdependent network of locally acting bone signals that govern overall bone remodeling. Recent studies have begun to decipher this hierarchy of interactions. Thus, inflammatory cytokine-induced NO rises have suppressed osteoclast bone resorption in organ cultures of neonatal mouse calvariae (18, 19) or long bones (17), whereas NOS inhibitors conversely reduced stimulated NO levels and alleviated inhibited bone resorption, consistent with reports that NO inhibited bone resorption by isolated rat or avian osteoclasts (15, 16). The degree of NO production and its local concentrations relative to other signals elicited during inflammation may critically determine whether bone resorption is augmented or dampened. Hence, in murine calvarial (18, 19) or long bone (17) organ cultures, NO concentrations below approximately 10 µM have not caused the inhibitory responses evoked by approximately 30 µM or greater, and may have been associated with cytokine-induced bone resorption, an effect possibly due to elevated coproduction of other resorption-promoting modulators (e.g. PGE2) (19). Furthermore, differential NO levels may underlie the selective inhibition of bone resorption by IFN{gamma}, because its combination with proresorptive cytokines IL-1 and TNF (which together elicit moderate NO rises) boosts NO levels into a range inhibitory for osteoclast resorption in murine calvarial cultures (17). These findings agree well with reports that high IL-1 and TNF, but not IFN{gamma}, levels are found in inflammatory disorders exhibiting localized bone loss, whereas in disorders characterized by exceedingly high NO levels (e.g. septicemia), bone loss does not occur (19). Other studies have also indicated potential biphasic effects of elevated NO levels on osteoclast formation and resorption, and contrasting effects of various NOS inhibitors on basal bone resorption (2, 10, 11, 14, 23, 26, 27). Inflammatory cytokines like IL-1 and TNF can also elicit release of degradative enzymes (e.g. collagenase) and depress osteoblast proliferation and bone formation, contributing to inflammatory-mediated osteopenia. Proresorptive effects of inflammatory cytokines may be offset by their concomitant induction of mediators that can scavenge NO and promote osteoclast resorption (e.g. superoxide anions and other reactive oxygen species) (12, 13, 14, 51, 52), inhibit NO production (IL-4, IL-8, IL-10, transforming growth factor-ß) (5, 6, 7, 8, 9, 10, 11, 12, 13, 14), alter the redox status of the cell (7, 10, 51, 53), or counteract NO effects (e.g. PGE2 suppression of iNOS and stimulation of osteoclast resorption) (10, 12, 13, 18, 19). Clearly, further complementary isolated cell and organ culture studies will be needed to describe the complicated autocrine and paracrine signal pathways that may control the production and physiological consequences of NO as part of a larger integrated regulatory network governing normal and pathological bone remodeling processes.

The unusual Ca2+-sensitive regulation of an iNOS in osteoclasts is consistent with the central role these cells have in maintaining calcium homeostasis. During bone resorption, local and ambient circulating Ca2+ levels rise via osteoclast-mediated release from the matrix, signaling osteoclasts to reduce bone resorptive activity. We propose that this mechanism involves an elevation of iNOS and autocrine release of the antiresorptive inhibitor NO. High extracellular Ca2+ concentrations ([Ca2+]e) have been shown to rapidly increase free [Ca2+]i levels in osteoclasts, probably via activation of a membrane ryanodine receptor-like sensor (54), leading to cell retraction and inhibited bone resorption. Excessive resorption has therefore been proposed to increase [Ca+2]i levels which, in turn, convey a feedback signal to restrain osteoclast bone resorption. Both elevation of [Ca+2]i and resorption of bone were shown here to raise NO release by osteoclasts. Elevated [Ca2+]e levels have also activated PKC isoforms in rat osteoclasts or human osteoclastoma cells (55), and so, it is possible that Ca2+ mediates PKC increases of iNOS in avian osteoclasts. PKC activation has also been proposed to mediate NO actions on rat osteoclasts (23). Other physiological signals besides [Ca2+]e can trigger a rise in [Ca2+]i in osteoclasts, and therefore may generate NO as a second-messenger signal to inhibit resorption. An important example is the calciotropic hormone calcitonin, which raises intracellular cAMP and [Ca2+]i (46, 47) and activates PKC in osteoclasts (56), signals that might increase iNOS/NO in osteoclasts and perhaps contribute to the resorption inhibitory actions of calcitonin. Other stimuli that could potentially elevate osteoclast iNOS via raising [Ca2+]i include IL-4 (57), ATP (58), platelet-activating factor (59), caffeine (60), ipriflavone (61), and mechanical perturbation (62). Furthermore, because engagement of osteoclast integrin {alpha}vß3 to RGD-containing components of bone (e.g. osteopontin, bone sialoprotein) may provoke a rise in [Ca2+]i (63), this may be at least partly responsible for the increased iNOS levels, and NO production we observed in this study in osteoclasts cultured on bone in comparison to those on plastic. In addition, Brandi et al. (23) reported that NOS inhibitors suppressed rat osteoclast contraction and substrate detachment elicited by elevated [Ca2+]e, and they attributed this to a putative stimulation of cNOS activity. Based on our findings, these NOS inhibitors also may have functioned to reduce NO synthesis following a Ca2+-mediated elevation of osteoclast iNOS. Thus, osteoclast release of the potent bone resorptive inhibitor NO is likely to be a key element in the Ca2+-mediated regulation of osteoclast activity.

In summary, we have shown in this study using avian osteoclasts the first case of an iNOS whose mRNA levels are regulated by Ca2+ or PMA, but not by inflammatory stimuli. Together with our previous demonstration that avian osteoclast bone resorptive activity is directly correlated with levels of NO (16), we propose that NO synthesized by osteoclasts provides a rapid, sensitive autocrine signal to down-regulate osteoclast bone resorptive function, and that NO has a prominent role as a second-messenger of osteoclast Ca2+ sensing. NO therefore represents an important bidirectional bone signal communication molecule whose production by osteoclasts could contribute in a paracrine manner to the modulation of multiple cell types within bone (osteoblasts, stromal, endothelial), as well as in an autocrine fashion to limit osteoclast bone resorptive activity. Discovery of this iNOS regulation pathway in osteoclasts potentially poses a novel target for the development of selective therapeutic agents to alleviate excessive bone loss.


   Footnotes
 
1 This work was supported by NIH Grants AR-32927, AR-42715, AR-32087, and DE-11060. Portions of this work were presented in preliminary form at the American Society for Bone and Mineral Research meetings held in Kansas City, MO (1994), Baltimore, Maryland (1995), and Seattle, Washington (1996). Back

2 Supported by a Howard Hughes Medical Investigation Undergraduate Fellowship. Back

Received September 23, 1996.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Gowen M 1992 Cytokines and Bone Metabolism. CRC Press, Boca Raton, FL
  2. Riancho JA, Mundy GR 1995 The role of cytokines and growth factors as mediators of the effects of systemic hormones at the bone local level. Crit Revs Eukar Gene Express 5:193–217
  3. Arend WP, Dayer J-M 1990 Cytokines and cytokine inhibitors or antagonists in rheumatoid arthritis. Arthritis Rheum 33:305–315[Medline]
  4. Williams R, Jeffcoat M, Kaplan M, Goldhaber P, Johnston H, Wechter W 1992 Flurbiprophen: a potent inhibitor of alveolar bone resorption in beagles. Science 227:640–642
  5. Griffith OW, Steuhr DJ 1995 Nitric oxide synthases: properties and catalytic mechanisms. Ann Rev Physiol 57:707–736[CrossRef][Medline]
  6. Knowles RG, Moncada S 1994 Nitric oxide synthases in mammals. Biochem J 298:249–258
  7. Gross SS, Wolin MS 1995 Nitric oxide: pathophysiological mechanisms. Ann Rev Physiol 57:737–769[CrossRef][Medline]
  8. Bredt DS, Snyder SH 1994 Nitric oxide: a physiologic messenger molecule. Ann Rev Biochem 63:175–195[CrossRef][Medline]
  9. Lowenstein CJ, Dinerman JL, Snyder SH 1994 Nitric oxide: a physiologic messenger. Ann Inter Med 120:227–237[Abstract/Free Full Text]
  10. Collin-Osdoby P, Osdoby P 1996 The role of nitric oxide in regulating bone remodeling. In: Davidovitch Z (ed) The Biological Mechanisms of Tooth Eruption, Resorption and Replacement by Implants. Harvard Society for the Advancement of Orthodontics, Boston MA, pp 69–82
  11. Evans C, Stefamovic-Racic M, Lancaster J 1995 Nitric oxide and its role in orthopedic disease. Clin Orthop Rel Res 312:275–294
  12. Collin-Osdoby P, Nickols GA, Osdoby P 1995 Bone cell function, regulation, and communication: a role for nitric oxide. J Cell Biochem 57:399–408[CrossRef][Medline]
  13. Hall TJ, Chambers TJ 1996 Molecular aspects of osteoclast function. Inflamm Res 45:1–9[CrossRef][Medline]
  14. Evans DM, Ralston SH 1996 Nitric oxide and bone. J Bone Miner Res 11:300–305[Medline]
  15. MacIntyre I, Zaidi M, Alam ASMT, Datta HK, Moonga BS, Lidbury PS, Hecker M, Vane JR 1991 Osteoclastic inhibition: an action of nitric oxide not mediated by cGMP. Proc Natl Acad Sci USA 88:2936–2940[Abstract/Free Full Text]
  16. Kasten T, Collin-Osdoby P, Patel N, Osdoby P, Krukowski M, Misko T, Settle S, Currie M, Nickols A 1994 Potentiation of osteoclast bone resorptive activity by inhibition of nitric oxide synthase. Proc Natl Acad Sci USA 91:3569–3573[Abstract/Free Full Text]
  17. Lowik C, Nibbering P, van de Ruit M, Papapoulos S 1994 Inducible production of nitric oxide in osteoblast-like cells and in fetal mouse bone explants is associated with suppression of osteoclastic bone resorption. J Clin Invest 93:1465–1472
  18. Ralston SH, Ho L-P, Helfrich MH, Grabowski PS, Johnston PW, Benjamin N 1995 Nitric oxide: a cytokine-induced regulator of bone resorption. J Bone Miner Res 10:1040–1049[Medline]
  19. Ralston SH, Grabowski PS 1996 Mechanisms of cytokine induced bone resorption: role of nitric oxide, cyclic guanosine monophosphate, and prostaglandins. Bone 19:29–33[Medline]
  20. Wimalawansa SJ, De Marco G, Gangula P, Yallampalli C 1996 Nitric oxide donor alleviates ovariectomy-induced bone loss. Bone 18:301–304[Medline]
  21. Misko TP, Moore WM, Kasten TP, Nickols GA, Corbett JA, Tilton RG, McDaniel ML, Williamson JR, Currie MG 1993 Selective inhibition of the inducible nitric oxide synthase by aminoguanidine. Eur J Pharmacol 233:119–125[CrossRef][Medline]
  22. Griffiths MJD, Messent M, MacAllister RJ, Evans TJ 1993 Aminoguanidine selectively inhibits inducible nitric oxide synthase. Br J Pharmacol 110:963–968[Medline]
  23. Brandi ML, Hukkanen M, Umeda T, Moradi-Bidhendi N, Bianchi S, Gross S, Polak JM, MacIntyre I 1995 Bidirectional regulation of osteoclast function by nitric oxide synthase isoforms. Proc Natl Acad Sci USA 92:2954–2958[Abstract/Free Full Text]
  24. Tsukahara H, Miura M, Tsuchida S, Hata I, Hata K, Yamamoto K, Ishii Y, Muramatsu I, Sudo M 1996 Effect of nitric oxide synthase inhibitors on bone metabolism in growing rats. Am J Physiol 270:E840–E845
  25. Schmidt HH, Gerard DG, Nakane M, Pollock JS, Miller MF, Murad F 1992 Mapping of neural nitric oxide synthase in the rat suggests frequent co-localization with NADPH diaphorase but not with soluble guanylyl cyclase, and novel paraneural functions for nitrinergic signal transduction. J Histochem Cytochem 40:1439–1456[Abstract]
  26. Sunyer T, Jiang X-S, Greenman R, Osdoby P, Nickols GA, Currie M, Rodi C, Misko T, Collin-Osdoby P 1994 Expression of inducible nitric oxide synthase and PCR cloning from avian osteoclasts and osteoclast-like cells. J Bone Miner Res [Suppl 1] 9:C114
  27. Sunyer T, Rothe L, Jiang X-S, Osdoby P, Collin-Osdoby P 1996 Proinflammatory agents, IL-8 and IL-10, upregulate inducible nitric oxide synthase expression and nitric oxide production in avian osteoclast-like cells. J Cell Biochem 60:469–483[CrossRef][Medline]
  28. Oursler MJ, Collin-Osdoby P, Anderson F, Li L, Webber D, Osdoby P 1991 Isolation of avian osteoclasts: improved techniques to preferentially purify viable cells. J Bone Miner Res 6:375–385[Medline]
  29. Collin-Osdoby P, Oursler MJ, Webber D, Osdoby P 1991 Osteoclast-specific monoclonal antibodies coupled to magnetic beads provide a rapid and efficient method of purifying avian osteoclasts. J Bone Miner Res 6:1353–1365[Medline]
  30. Imai Y, Kolb H, Burkhardt V 1993 Nitric oxide production from macrophages is regulated by arachidonic acid metabolites. Biochem Biophys Res Commun 197:105–109[CrossRef][Medline]
  31. Chomczynski P, Sacchi N 1987 Single-step method of RNA isolation by acid guanidium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156–159[Medline]
  32. Hewett SJ,