Skip to main content
NCBI home page
Clinical and Experimental Immunology logoLink to Clinical and Experimental Immunology
. 2006 Oct;146(1):146–158. doi: 10.1111/j.1365-2249.2006.03181.x

Activated T lymphocytes suppress osteoclastogenesis by diverting early monocyte/macrophage progenitor lineage commitment towards dendritic cell differentiation through down-regulation of receptor activator of nuclear factor-kappaB and c-Fos

D Grčević *, I K Lukić †, N Kovačić †, S Ivčević ‡, V Katavić †, A Marušić †
PMCID: PMC1809724  PMID: 16968409

Abstract

Activated T lymphocytes either stimulate or inhibit osteoclastogenesis from haematopoietic progenitors in different experimental models. To address this controversy, we used several modes of T lymphocyte activation in osteoclast differentiation − mitogen-pulse, anti-CD3/CD28 stimulation and in vivo and in vitro alloactivation. Osteoclast-like cells were generated from non-adherent immature haematopoietic monocyte/macrophage progenitors in murine bone-marrow in the presence of receptor activator of nuclear factor (NF)-κB ligand (RANKL) and monocyte–macrophage colony-stimulating factor (M-CSF). All modes of in vivo and in vitro T lymphocyte activation and both CD4+ and CD8+ subpopulations produced similar inhibitory effects on osteoclastogenesis paralleled by enhanced dendritic cell (DC) differentiation. Osteoclast-inhibitory effect was associated with T lymphocyte activation and not proliferation, and could be replaced by their culture supernatants. The stage of osteoclast differentiation was crucial for the inhibitory action of activated T lymphocytes on osteoclastogenesis, because the suppressive effect was visible only on early osteoclast progenitors but not on committed osteoclasts. Inhibition was associated specifically with increased granulocyte–macrophage colony-stimulating factor (GM-CSF) expression by the mechanism of progenitor commitment toward lineages other than osteoclast because activated T lymphocytes down-regulated RANK, CD115, c-Fos and calcitonin receptor expression, and increased differentiation towards CD11c-positive DC. An activated T lymphocyte inhibitory role in osteoclastogenesis, confirmed in vitro and in vivo, mediated through GM-CSF release, may be used to counteract activated bone resorption mediated by T lymphocyte-derived cytokines in inflammatory and immune disorders. We also demonstrated the importance of alloactivation in osteoclast differentiation and the ability of cyclosporin A to abrogate T lymphocyte inhibition of osteoclastogenesis, thereby confirming the functional link between alloreaction and bone metabolism.

Keywords: animal (mice) models, cell differentiation, cytokine receptors, cytokines, T cells

Introduction

T lymphocytes play an important role in the regulation of bone metabolism, particularly the bone resorption by osteoclasts [1–4]. However, the reports on activated T lymphocyte effects on osteoclastogenesis are still controversial. Activation of T lymphocytes in different experimental models or human diseases has mainly been shown to stimulate osteoclast differentiation and bone resorption by production of cytokines, such as receptor activator of nuclear factor (NF)-κB ligand (RANKL), interleukin (IL)-6, IL-7, IL-17 and tumour necrosis factor (TNF)-α[2,5–9]. On the other hand, T lymphocytes secrete interferon (IFN)-γ, IL-4, IL-10, IL-13 and granulocyte–macrophage colony-stimulating factor (GM-CSF), which mediate the inhibition of osteoclastogenesis from haematopoietic progenitors [2,5,10–14]. Although some of those inhibitory cytokines have been associated with T lymphocyte activation, various in vitro modesof T lymphocyte activation differentially affect osteoclastogenesis [2,8,10,12], and most in vivo models report an enhanced osteoclast formation upon T lymphocyte activation [3,6,15,16].

T lymphocytes have a role not only in osteoclast activation and bone resorption but also in the modulation of multi-potent monocyte/macrophage progenitor differentiation towards either bone or immune lineage, which is determined by ligand binding to cell-surface receptors, particularly receptor activator of NF-κB (RANK) for osteoclasts and Toll-like receptors (TLRs) for mononuclear phagocytes [17]. The interaction of RANK, expressed on osteoclast progenitors, with its ligand RANKL is crucial for osteoclastogenesis in the presence of monocyte–macrophage colony-stimulating factor (M-CSF) [6,7,18–20]. At the level of cell signalling, both RANK and TLRs activate the dimeric transcription factors NF-κB and activator protein (AP)-1. Transcription factor c-Fos, a component of AP-1, plays a positive role in osteoclast but a negative role in macrophage and dendritic cell (DC) differentiation [17,21,22]. GM-CSF controls the bifurcation between osteoclasts and DC from common progenitors by regulating c-Fos expression [22,23].

To study the controversial effect of activated T lymphocytes on osteoclastogenesis we used several activation protocols, including in vivo and in vitro alloactivation. Activation by any method suppressed osteoclastogenesis stimulated by RANKL and M-CSF in murine haematopoietic progenitors. To elucidate the mechanism by which in vitro- and in vivo-activated T lymphocytes suppress differentiation towards the osteoclast lineage from bone-marrow (BM) haematopoietic progenitors we identified the stage at which the suppressive effect was achieved and the expression profile of the treated progenitors. In contrast to some other studies, our findings show clearly that activated T lymphocytes have a net inhibitory effect on osteoclast differentiation by diverting early haematopoietic progenitors towards DC differentiation through down-regulation of RANK and c-Fos.

Materials and methods

Subjects and samples

Twelve-week-old C57BL/6 J (H-2b) female mice were used in all experiments. Allogeneic CBA/J (H-2k) mice were used for allostimulation. Maintenance of animals and all experimental procedures were approved by the Ethics Committee of the Zagreb University School of Medicine.

T lymphocyte preparation

Murine lymph node cells were depleted of non-T lymphocyte populations using anti-CD11b/anti-CD45R monoclonal antibodies (MoAbs) (BD Pharmingen, San Jose, CA, USA) and further separated by anti-CD4/anti-CD8 MoAbs conjugated to magnetic beads (Dynel Biotech, Wirral, UK). Separated T lymphocyte populations had the purity of > 90% as confirmed by flow cytometry.

Immunosuppressant pretreatment consisted of 1 h incubation at 37°C [24] with various concentrations (2·5–40 µg/ml) of cyclosporin A (CsA) (Sigma-Aldrich, St Louis, MO, USA).

For activation, T lymphocytes were pulsed for 15 min by concanavalin A (Con A) (Sigma-Aldrich), washed and cultured for 24 h. In some experiments T lymphocytes were activated by a 24-h incubation with anti-CD3 (1 µg/ml; Caltag, Burlingame, CA, USA) and anti-CD28 MoAbs (0·5 µg/ml; BD Pharmingen). To inhibit proliferation, T lymphocytes were treated with mitomycin C (Sigma-Aldrich) (20 min at 37°C) after Con A-pulse and 24 h culture. T lymphocyte activation was confirmed by the expression of CD71 and changes in the cell-cycle using flow cytometry [25]. Proliferation was assessed by the colorimetric MTT (Sigma-Aldrich) assay. Activated T lymphocytes were co-cultured with BM cells or cultured alone to obtain conditioned medium needed to treat osteoclastogenic cultures (see ‘Osteoclast-like cell cultures’).

In vivo and in vitro alloactivation

For mixed lymphocyte reaction (MLR), T lymphocytes were co-cultured with allogeneic mitomycin C-treated splenocytes (cell ratio 1 : 1) for 48 h [24]. T lymphocyte activation and proliferation were confirmed as described for mitogen pulse. Conditioned medium of in vitro alloactivated T lymphocytes was used to treat osteoclastogenic cultures.

C57BL/6 J mice were allostimulated in vivo by intraperitoneal (i.p.) injection of allogeneic splenocytes (5 × 107/mouse) followed by foot-pad restimulation with the same cells (107/foot), 10 days after the first injection [25]. Untreated mice and mice treated with syngeneic splenocytes were used as controls. Allostimulated mice provided tibial BM for osteoclast cultures and regionally alloactivated (popliteal) lymph node for T lymphocyte cultures. In vivo alloactivated T lymphocytes were co-cultured with BM cells or additionally cultured for 24 h to obtain conditioned medium needed to treat osteoclastogenic cultures.

Osteoclast-like cell cultures

For osteclast-like (OCL) cell generation, BM were cultured overnight (day − 1) with 5 ng/ml recombinant mouse (rm)M-CSF (R&D Systems, Abingdon, UK) in α-minimum essential medium (MEM)/10% fetal calf serum (FCS) (Sigma-Aldrich) to stimulate monocyte/macrophage lineage [25], followed by harvesting of non-adherent cells as enriched haematopoietic monocyte/macrophage progenitors that are not yet committed towards a certain lineage of differentiation. Non-adherent BM cells (2 × 105/well) were plated (day 0) in 48-well plates with rmRANKL (a gift from Amgen, Thousand Oaks, CA, USA) and rmM-CSF (10 ng/ml for both) to stimulate osteoclast formation (termed ‘osteoclastogenic’ cultures), with medium exchange at day 2·5. Different numbers of T lymphocytes (1·2–5 × 105/well) or different volumes of T lymphocyte-conditioned medium (12·5–100%/well) were used to treat BM cultures at day 0 or 2·5. In some experiments rmGM-CSF (5 or 10 ng/ml; R&D), neutralizing anti-GM-CSF or anti-IFN-γ MoAbs (1 µg/ml for both; BD Pharmingen) were added to osteoclastogenic cultures together with RANKL and M-CSF. At day 5, tartrate-resistant acid phosphatase (TRAP) positive multi-nucleated cells with ≥ three nuclei/cell were considered OCL and counted per well using light microscopy. Differentiated OCL cells highly expressed calcitonin receptor (CtR). There were no TRAP+ OCL cells in cultures without addition of RANKL and M-CSF.

Flow cytometry

Non-adherent BM cell differentiation in osteoclastogenic cultures (after 2·5-day culture) was assessed by: phycoeryhtrin (PE)-anti-CD115 (c-Fms), PE-anti-CD116 (GM-CSF receptor α) (eBioscience, San Diego, CA, USA), fluorescein isothiocyanate (FITC)-anti-CD11b, allophycocyanin (APC)-anti-CD11c (Caltag), and the combination of goat-anti-RANK (R&D Systems) with PE-anti-goat antibodies. For intracellular cytokine staining, T lymphocytes were treated with Brefeldin A (eBioscience), fixed/permeabilized and stained with the combination of rabbit-anti-GM-CSF (eBioscience) and PECy5·5-anti-rabbit antibodies. Apoptotic and dead cells were detected by annexin V/propidium iodide (PI) staining (BD Pharmingen) according to the manufacturer's recommendation. Results, using FACSCalibur (BD Pharmingen), were presented as histograms or dot-plots for 20 000 viable cells/sample. Dead and fragmented cells were excluded from the analysis by their properties on correlated forward/side scatter and PI staining. Applied gates and quadrants were adjusted to non-stained cells and isotype-controls by delineating negative populations to approximately 101 fluorescence intensity (not shown).

Gene expression analysis

RNA was extracted using a commercial kit (TriPure; Roche, Mannheim, Germany) from harvested T lymphocytes (freshly isolated or cultured) or osteoclastogenic cultures (2·5- or 5-day culture). Each sample was obtained from ≥ three animals/group. RNA (1 µg) was converted to cDNA and amplified (20 ng/well) by quantitative polymerase chain reaction (qPCR), using specific amplimer sets designed by PrimerExpress software (Applied Biosystems, Foster City, CA, USA) for mouse RANK, RANKL, osteoprotegerin (OPG), CD115, TNF-related apoptosis-inducing ligand (TRAIL) and β-actin with SYBRGreen chemistry (Applied Biosystems). Expression of c-Fos, GM-CSF, IFN-γ, CD178 (FasL), CtR and CD116 was analysed using commercially available TaqMan assays (Applied Biosystems). qPCR reactions (25 µl/well) were conducted in an ABI Prism 7000 Sequence Detection System (Applied Biosystems) in quadruplicate, as described previously [26]. According to the standard curve, the relative amounts of RNA for target genes were calculated as the ratio of the quantity of target gene normalized to β-actin. RNA quantity for control sample in each experiment was set as 1 and the relative RNA quantities for other samples were calculated in accordance to this value.

Statistics

Experiments were performed at least three times and the representative data were shown (mean ± s.d. of four replicates per sample). Statistical analysis of TRAP+ OCL cell numbers was performed using analysis of variance (anova) and Student–Newman–Keuls post-hoc test (MedCalc, Mariakerke, Belgium). Relative values of RNA quantity were analysed statistically using the comparison of the means t-test with Bonferroni correction for multiple-group comparison (MedCalc). For all experimemts, α-level was set at 0·05.

Results

Activated T lymphocytes inhibit murine osteoclast differentiation

T lymphocytes were activated by a brief mitogen pulse [27] in order to avoid the presence of Con A in the culture supernatants. Activation of T lymphocytes consistently inhibited in vitro osteoclast differentiation in cultures of murine non-adherent BM cells supplemented with RANKL and M-CSF (Fig. 1a). This was true both for co-cultures of activated T lymphocytes and osteoclast progenitors and for cultures where osteoclast progenitors were treated with the supernatants of activated T lymphocyte cultures (Fig. 1b). The inhibitory effect was dose-dependent, i.e. reciprocal to the number of activated T lymphocytes or the volume of activated T lymphocyte conditioned medium. In further experiments, we used 0·25 × 106 T lymphocytes per well or 50% conditioned medium per well, unless stated otherwise.

Fig. 1.

Fig. 1

Open in a new tab

Activated T lymphocytes suppressed murine osteoclast differentiation stimulated by receptor activator of nuclear factor (NF)-κB ligand (RANKL) and monocyte–macrophage colony-stimulating factor (M-CSF). (a) Photomicrographs of osteoclastogenic non-adherent bone-marrow cell cultures with α-minimum essential medium (MEM) (control), unstimulated or activated [concanavalin A (Con A)-pulsed] 0·25 × 106 T lymphocytes per well (lyT) and corresponding 50% T lymphocyte conditioned medium (sup); × 100. Number of Tartrate-Resistant Acid Phosphatase (TRAP)+ osteoclast-like cells (OCLs) in non-adherent bone-marrow cell cultures stimulated by RANKL and M-CSF after addition of (b) different numbers of T lymphocytes/well or different volumes of T lymphocyte conditioned medium/well; (c) 0·25 × 106 T lymphocytes/well or 50% T lymphocyte conditioned medium/well at different time-points of culture (duration of treatment: days 0–2·5, 2·5–5 or 0–5); (d) 0·25 × 106 T lymphocytes/well, 50% T lymphocyte conditioned medium/well, or 0·25 × 106 mitomycin C-treated T lymphocytes/well (lyT + MitC); or (e) 50% conditioned medium/well from unseparated (lyT), CD4+ or CD8+ T lymphocytes. Values, mean ± s.d. (n = 4). *P < 0·01 versus. control culture and respective culture treated with unstimulated T lymphocytes/conditioned medium. T lymphocytes were activated by Con A-pulse, except for (d), where we used Con A-pulse or anti-CD3/CD28 treatment.

Furthermore, we added activated T lymphocytes or conditioned medium at different time-points of osteoclastogenic cultures to test if the stage of osteoclast differentiation was important for the inhibitory effect of activated T lymphocytes. Only immature monocyte/macrophage progenitors (treatment 0–2·5 days) were sensitive to the inhibitory effect of activated T lymphocytes, whereas committed osteoclast progenitors (treatment 2·5–5 days) did not respond to activated T lymphocytes or conditioned medium (Fig. 1c). In further experiments, the treatment by activated T lymphocytes or conditioned medium was performed only for the first 2·5 days of osteoclastogenic culture.

To rule out that the effect of activated T lymphocytes was a consequence of intensive T lymphocyte proliferation and media exhaustion upon co-culture with BM cells, we added mitomycin C-treated activated T lymphocytes to cultures and found a similar inhibitory effect on osteoclastogenesis (Fig. 1d). Inhibition of osteoclastogenesis was not associated specifically with mitogen stimulation, as osteoclastogenesis was blocked in BM cultures treated with the supernatant of T lymphocytes activated by anti-CD28/CD3 antibodies (Fig. 1d). Interestingly, both CD4+ and CD8+ T lymphocyte subpopulations were equally effective in the suppression of osteoclastogenesis, similar to unseparated T lymphocytes (Fig. 1e).

Inhibition of osteoclastogenesis is achieved by changes in cytokine gene expression by T lymphocytes

As assessed by qPCR, activation of T lymphocytes significantly increased the expression of several cytokines known to inhibit osteoclastogenesis such as GM-CSF, IFN-γ and OPG, with the parallel overexpression of the osteoclastogenic factor RANKL (Fig. 2). OPG functions as a decoy receptor for RANKL and competes with RANK/RANKL binding [28]. Because we found the proportional increase in both RANKL and OPG expression upon T lymphocyte activation (Fig. 2), we do not believe that OPG is the important anti-osteoclastogenic factor in our model, particularly because RANKL is added exogenously to the osteoclastogenic cultures and would override the inhibitory effect of OPG.

Fig. 2.

Fig. 2

Open in a new tab

Activated T lymphocytes increased the expression of granulocyte–macrophage colony-stimulating factor (GM-CSF), interferon (IFN)-γ, receptor activator of nuclear factor (NF)-κB ligand (RANKL) and osteoprotegerin (OPG) by reverse transcription–quantitative polymerase chain reaction (RT-qPCR). Relative quantity of RNA represents the ratio of RNA quantity for the respective gene normalized to the quantity of β-actin in freshly isolated T lymphocytes (lyT nat), or in unstimulated and activated [concanavalin A (Con A)-pulsed] T lymphocytes cultured for 24 h. Values, mean ± s.d. (n = 4). *P < 0·01 versus control group and unstimulated T lymphocytes.

Treatment of BM cultures with GM-CSF produced a complete inhibition of osteoclastogenesis despite the presence of RANKL and M-CSF in the culture medium (Fig. 3a). The addition of neutralizing anti-GM-CSF antibodies in cultures treated with activated T lymphocyte-conditioned medium abrogated their inhibitory effect on osteoclastogenesis and restored the number of TRAP+ OCL cells to approximately 90% of the control cultures (Fig. 3b) (ranging from 50 to 90% in the repeated experiments), whereas the addition of anti-IFN-γ antibodies did not produce any changes in the OCL cell number (data not shown). Therefore we concluded that GM-CSF's osteoclast-inhibitory effect prevailed over the IFN-γ in our in vitro conditions. Intracellular cytokine staining confirmed an increased GM-CSF (Fig. 3c) production by activated T lymphocytes. Nevertheless, increased GM-CSF by T lymphocytes was not paralleled by increased CD116 expression in osteoclastogenic cultures (data not shown).

Fig. 3.

Fig. 3

Open in a new tab

Granulocyte–macrophage colony-stimulating factor (GM-CSF) mediates suppression of osteoclastogenesis by activated T lymphocytes. (a) Photomicrographs of osteoclastogenic non-adherent bone-marrow cell cultures with α-minimum essential medium (MEM) (control), activated T lymphocytes conditioned medium (sup) or GM-CSF (5 or 10 ng/ml); × 100. (b) Number of tartrate-resistant acid phosphatase (TRAP)+ osteoclast-like cells (OCLs) in non-adherent bone-marrow cell cultures stimulated by receptor activator of nuclear factor (NF)-κB ligand (RANKL) and monocyte–macrophage colony-stimulating factor (M-CSF) after addition of 50% T lymphocytes conditioned medium with or without 1 µg/ml neutralizing anti-GM-CSF antibodies. Values, mean ± s.d. (n = 4). *P < 0·01 versus control culture, culture treated with unstimulated T lymphocyte conditioned medium and culture treated with activated T lymphocyte conditioned medium plus anti-GM-CSF. (c) Flow cytometric analysis of GM-CSF production by unstimulated T lymphocytes (left) or activated [concanavalin A (Con A)-pulsed] T lymphocytes (right); dot-plots [forward scatter (FSC)/phycoerythrin (PE)Cy5·5-anti-rabbit + rabbit-anti-GM-CSF).

The inhibitory effect was achieved by the significant GM-CSF overexpression in two major T lymphocyte subpopulations CD4+ and CD8+ (RNA relative quantity 0·7 ± 0·1 in unstimulated and 3·8 ± 0·7 in activated CD4+; 0·9 ± 0·2 in unstimulated and 3·4 ± 0·5 in activated CD8+; compared with 1·0 ± 0·1 in unstimulated and 4·6 ± 0·3 in activated unseparated T lymphocytes; P < 0·01, t-test with Bonferroni correction for multiple-group comparison), and both subpopulations have been shown previously to produce GM-CSF upon activation [29]. GM-CSF expression was specifically associated with T lymphocytes, as we found increased GM-CSF RNA only in co-cultures of activated T lymphocytes and non-adherent BM cells and not in non-adherent BM cell cultures treated with activated T lymphocyte-conditioned medium (data not shown).

Osteoclast precursors treated with mitogen-activated T lymphocytes or culture supernatants express lower levels of RANK, CD115, c-Fos and calcitonin receptor

We further assessed the expression of RANK and CD115, receptors for essential osteoclastogenic factors RANKL and M-CSF, respectively [30], and found that the RNA for RANK and CD115 were significantly lower in 2·5-day osteoclatogenic cultures treated with activated T lymphocytes or conditioned medium compared with control cultures (Fig. 4). At day 5, the decrease in gene expression was more pronounced for RANK than for CD115, and was stronger in cultures treated with activated T lymphocytes than with conditioned medium. Moreover, expression of c-Fos, the transcription factor important for osteoclast lineage commitment [17], and CtR, the marker of the differentiated osteoclast [21], was suppressed upon treatment with activated T lymphocytes or conditioned medium (Fig. 4). Flow cytometry showed a down-regulation of RANK and CD115 expression on day 2·5 after the addition of activated T lymphocyte-conditioned medium to osteoclastogenic cultures (Fig. 5a).

Fig. 4.

Fig. 4

Open in a new tab

Addition of activated T lymphocytes or conditioned medium to murine osteoclastogenic cultures stimulated by receptor activator of nuclear factor (NF)-κB ligand (RANKL) and monocyte–macrophage colony-stimulating factor (M-CSF) decreased the expression of RANK, CD115, c-Fos and calcitonin receptor (CtR) by reverse transcription–quantitative polymerase chain reaction (RT-qPCR). Osteoclastogenic non-adherent bone-marrow cell cultures, with or without treatment by T lymphocytes (lyT) or conditioned medium (sup) for the first 2·5 days of culture, were harvested for RNA extraction after 2·5 or 5 days of culture. Values, mean ± s.d. (n = 4). *P < 0·01 versus respective time-point of control culture and respective culture treated with unstimulated T lymphocytes/conditioned medium. **P < 0·01 versus only the respective control culture.

Fig. 5.

Fig. 5

Open in a new tab

Addition of activated T lymphocyte conditioned medium to murine osteoclastogenic cultures stimulated by receptor activator of nuclear factor (NF)-κB ligand (RANKL) and monocyte–macrophage colony-stimulating factor (M-CSF) changed the expression of RANK, CD115, CD11c and CD11b by flow cytometry. Non-adherent bone-marrow cells stimulated by RANKL and M-CSF were cultured for 2·5 days in α-minimum essential medium (MEM) (control, left panel) or in 50% activated T lymphocyte conditioned medium [lyT concanavalin A (Con A) sup, right panel]. Cultures of bone-marrow cells treated with unstimulated T lymphocyte conditioned medium (data not shown) were similar to control cultures. (a) Histograms [events/phycoerythrin (PE)-anti-RANK and events/PE-anti-CD115]; or (b) histograms [events/allophycocyanin (APC)-anti-CD11c], arrow indicates the CD11chigh population, and dot plots (PE-anti-RANK/APC-anti-CD11c and fluorescein isothiocyanate (FITC)-anti-CD11b/APC-anti-CD11c).

Increased GM-CS expression mediates transdifferentiation between osteoclast and dendritic lineages

To test if activated T lymphocytes could suppress osteoclast differentiation by GM-CSF-induced differentiation of common monocyte/macrophage progenitors away from the osteoclast towards the immune cell lineages, i.e. DC, we analysed the expression of CD11c, a dendritic lineage marker [22], on non-adherent BM progenitors treated with activated T lymphocyte conditioned medium. Flow cytometry on day 2·5 revealed the presence of CD11chigh population and parallel down-regulation of RANK expression after the treatment with activated T lymphocyte-conditioned medium. Furthermore, we detected a decrease in macrophage CD11b+CD11c– cell population, and an increase in dendritic Cd11b–CD11c+ and Cd11b+CD11c+ cell populations (Fig. 5b).

Because T lymphocytes may produce many pro-apoptotic molecules upon activation [31], we assessed the expression of TRAIL and CD178 in activated T lymphocytes after 24-h culture but detected only a weak, insignificant up-regulation by qPCR (data not shown). In addition, the apoptotic rate in osteoclastogenic cultures did not differ between control cultures and cultures treated with activated T lymphocyte-conditioned medium (19·5% annexin V+/PI– cells in cultures treated with unstimulated versus 18·8% in cultures treated with Con A-pulsed T lymphocyte conditioned medium by flow cytometry). Those results confirmed that progenitor commitment towards dendritic lineage and not the induction of osteoclast progenitor apoptosis was the major mechanism of osteoclast inhibition in our model.

Allogeneic stimulation of T lymphocytes in vitro and in vivo produces similar effects on osteoclastogenesis as activation by mitogen or anti-CD28/CD3 antibodies

In addition to mitogen and anti-CD3/CD28 stimulation, which has been used commonly to study the effects on osteoclastogenesis [6,8,12], we also performed T lymphocyte allostimulation in vivo and in vitro. Conditioned medium of T lymphocytes alloactivated in MLR during 48 h produced factor(s) that decreased the number of TRAP+ OCL cells in osteoclastogenic cultures stimulated with RANKL and M-CSF (Fig. 6a,b). After in vivo allostimulation, we confirmed that the addition of in vivo alloactivated T lymphocytes or conditioned medium to osteclastogenic cultures also suppressed osteoclastogenesis (Fig. 6c). In addition, the number of TRAP+ OCL cells differentiated from tibial non-adherent BM cells in osteoclastogenic cultures of in vivo allostimulated mice was significantly lower compared with cultures from control mice (Fig. 6d). The pattern of RANK and CD115 expression was similar among experimental and control groups when allostimulation, anti-CD3/CD28 treatment or Con A-pulse were used for T lymphocyte activation (see Fig. 4). All modes of T lymphocyte activation were associated with an approximate five- to sixfold increase in GM-CSF expression (see Fig. 2).

Fig. 6.

Fig. 6

Open in a new tab

In vitro and in vivo alloactivated T lymphocytes suppressed murine osteoclast differentiation stimulated by receptor activator of nuclear factor (NF)-κB ligand (RANKL) and monocyte–macrophage colony-stimulating factor (M-CSF). (a) Photomicrographs of osteoclastogenic non-adherent bone-marrow cell cultures with α-minimum essential medium (MEM) (control), unstimulated (lyT) or alloactivated (MLR) 50% T lymphocyte conditioned medium (sup); × 100. Number of tartrate-resistant acid phosphatase (TRAP)+ osteoclast-like cells (OCLs) in non-adherent bone-marrow cell cultures stimulated by RANKL and M-CSF after addition of (b) different volumes of unstimulated or in vitro allostimulated T lymphocyte conditioned medium/well; or (c) 0·25 × 106/well or 50% conditioned medium/well of in vivo allostimulated popliteal T lymphocytes, without or with in vitro restimulation for 24 h. (d) In vivo allostimulation was performed by intraperitoneal (i.p.) injection and foot-pad restimulation with allogeneic splenocytes. Left pannel, photomicrographs of osteoclastogenic non-adherent bone-marrow cell cultures from unstimulated animals or from in vivo allostimulated animals, × 100; right panel, number of TRAP+ OCLs in non-adherent bone-marrow cell cultures stimulated by RANKL and M-CSF from unstimulated animals or in vivo allostimulated animals. Values, mean ± s.d. (n = 4). *P < 0·01 versus control culture and respective culture treated with unstimulated T lymphocytes/conditioned medium (b, c) or culture from unstimulated animals (d). **P < 0·01 versus in vivo alloactivated but not in vitro restimulated cells.

CsA abrogates inhibitory effect of T lymphocytes on osteoclastogenesis

We further tested if immunosuppressant CsA, which inhibits T lymphocyte activation, can overcome the suppression of osteoclastogenesis caused by Con A-pulsed T lymphocytes. As T lymphocytes and osteoclasts share the sensitivity to many common cytokines [1,2,5] and suppressory signals, including CsA [32,33], we wanted to separate the effect of CsA on T lymphocytes from the effect on osteoclast progenitors and therefore used CsA only for a brief pretreatment of T lymphocytes and not for actual culture [24]. A brief 1-h preincubation of T lymphocytes with different doses of CsA (2·5, 10 and 40 mg/ml) were performed prior to Con A-pulse and subsequent 24-h culture. CsA pretreatment of T lymphocytes abrogated the suppression of osteoclastogenesis in a dose-dependent manner. The most pronounced reversal of inhibition of osteoclastogenesis was achieved when T lymphocytes were first pretreated with the highest dose of CsA (40 mg/ml), pulsed with Con A, cultured for 24 h and then added as a 50% conditioned medium to osteoclastogenic cultures (Fig. 7a,b). The reversal of osteoclast inhibition was paralleled by changes in gene expression by Con A-pulsed T lymphocytes, evidenced by an insignificant down-regulation of RANKL (P > 0·05) and a significant down-regulation of GM-CSF (P = 0·0025, anova and Student–Newman–Keuls post-hoc test) upon immunosuppressant pretreatment (Fig. 7c).

Fig. 7.

Fig. 7

Open in a new tab

Cyclosporin A (CsA)-treatment prior to concanavalin A (Con A)-pulse abrogates the inhibitory effect of T lymphocytes on murine osteoclast differentiation stimulated by receptor activator of nuclear factor (NF)-κB ligand (RANKL) and monocyte–macrophage colony-stimulating factor (M-CSF). (a) Photomicrographs of osteoclastogenic non-adherent bone-marrow cell cultures with α-minimum essential medium (MEM) (control), unstimulated or activated (Con A-pulse) T lymphocyte (lyT) conditioned medium (sup) with or without CsA pretreatment; × 100. (b) Number of tartrate-resistant acid phosphatase (TRAP)+ osteoclast-like cells (OCLs) in non-adherent bone-marrow cell cultures stimulated by RANKL and M-CSF after the addition of 50% conditioned medium/well of T lymphocytes pretreated with different concentrations of CsA. (c) Reverse transcription–quantitative polymerase chain reaction (RT-qPCR) analysis of RANKL and granulocyte–macrophage colony-stimulating factor (GM-CSF) expression in unstimulated and activated (Con A-pulsed) T lymphocytes, with or without CsA (40 µg/ml) pretreatment, after 24-h culture. Values, mean ± s.d. (n = 4). *P < 0·01 versus respective unstimulated T lymphocytes. **P < 0·01 versus Con A-pulsed T lymphocytes without CsA-pretreatment.

Discussion

Our study demonstrated that activated T lymphocytes suppressed haematopoietic cell differentiation towards the osteoclast lineage by affecting the lineage commitment of early monocyte/macrophage progenitors through the mechanism that involved RANK down-regulation and c-Fos suppression. In addition to mitogen-pulse and anti-CD3/CD28 antibodies, we used in vitro and in vivo T lymphocyte alloactivation and showed for the first time that these methods also produce inhibition of osteoclastogenesis. The stage of osteoclast differentiation was crucial for the inhibitory action of activated T lymphocyte on osteoclastogenesis, as the suppressive effect was visible only on early osteoclast progenitors when activated T lymphocytes were added with RANKL and M-CSF at the beginning of the osteoclastogenic cultures but not on committed osteoclasts. The inhibitory effect was associated specifically with T lymphocyte activation and not proliferation, and did not depend on cell-to-cell contact. The important mediator of osteoclast inhibition and dendritic lineage stimulation in our study was GM-CSF, known to regulate the bifurcation between osteoclast and DC lineages [22]. We found that activated T lymphocytes up-regulated GM-CSF expression and that the osteoclast-inhibitory effect of activated T lymphocytes was abrogated by the addition of anti-GM-CSF antibodies to osteoclastogenic cultures.

GM-CSF effect on osteoclastogenesis depends on the experimental model and the duration or timing of GM-CSF administration, and was reported differently in various animal models in vivo and osteoclast cultures in vitro[11–13,22,32,34–36]. In contrast to the study by Wyzga et al. [12], who claimed that the mode of T lymphocyte activation is crucial for the final effect on osteoclast differentiation, we confirmed that the stage of osteoclast precursors is more important by showing that GM-CSF in the presence of RANKL at the beginning of the osteoclastogenic cultures inhibits osteoclast formation through cellular differentiation into DC, whereas it is ineffective after the commitment of RANKL-induced differentiation towards osteoclasts. It has already been shown that GM-CSF treatment of early human osteoclast precursors blocks osteoclastogenesis [34], with a repression of major genes indicative of osteoclast functions [32], but those studies did not investigate activated T lymphocytes as a possible source of GM-CSF in this regulation. Among other effects, GM-CSF inhibits c-Fos in osteoclast progenitors, which disrupts the RANK/c-Fos/nuclear factor of activated T cells (NFAT)c1 transcriptional cascade critical for osteoclastogenesis [17]. IFN-γ also suppresses osteoclastogenesis by disrupting RANKL-induced activation of NF-κB and C-Jun amino terminal kinase (JNK) [1,5]. Nevertheless, increased IFN-γ expression by activated T lymphocytes was not important in our model, as osteoclast inhibition was achieved at the level of decreased RANK expression (upstream of the described IFN-γ action) and could not be neutralized by anti-IFN-γ antibodies. GM-CSF acts at earlier stages of osteoclast differentiation than IFN-γ[1,10,22], supporting our observation of the more important role of GM-CSF than IFN-γ in osteoclast inhibition.

In our experimental in vitro microenvironment, the treatment of early monocyte/macrophage BM progenitors with activated T lymphocytes determined the differentiation towards DC away from the osteoclast lineage, despite the presence of RANKL and M-CSF in culture medium. A number of DC-like adherent clusters were observed in BM cultures treated with all types of activated T lymphocytes or conditioned medium enriched in T lymphocyte derived GM-CSF, similar to those observed after GM-CSF treatment. In addition, we detected a subpopulation of CD11chigh cells characteristic for dendritic lineage, and a shift from macrophage CD11b+CD11c– towards dendritic CD11b–CD11c+ and Cd11b+CD11c+ populations in osteoclastogenic cultures treated with activated T lymphocyte conditioned medium, which has not been found in control cultures. This supports a previous finding that GM-CSF stimulates osteoclast progenitor proliferation but inhibits differentiation, and in combination with RANKL promotes DC formation [22,23,34]. The mechanism underlying the differentiation switch between osteoclast and dendritic lineages certainly involves the regulation of transcription factors essential for osteoclast formation, including c-Fos [37]. c-Fos expression was down-regulated in cultures treated by activated T lymphocytes, showing that osteoclast differentiation is inhibited and DC differentiation stimulated reciprocally through the suppression of c-Fos, supporting our hypothesis that progenitor commitment toward lineages other than osteoclast is the underlying mechanism of osteoclast inhibition in our model.

By analysing RANK and CD115 during in vitro osteoclast differentiation [30], we found that the inhibitory effect of activated T lymphocytes was more pronounced on RANK expression compared to CD115 expression. This suggests that down-regulation of RANK signalling is the major mechanism of activated T lymphocyte action, as RANK expression on haematopoietic progenitors is essential for osteoclast differentiation and activation [19–21]. CD115 gene was down-regulated in cultures treated with activated T lymphocytes at earlier stages of osteoclast differentiation, confirming that CD115 tyrosine kinase provides signals required for survival and proliferation of early osteoclast progenitors [21,30]. Studies using different osteoclast culture systems reported different findings regarding the expression of CD115, RANK and CD116 in the presence of M-CSF, GM-CSF or IL-3 [22,32,36,37]. In contrast to up-regulation of CD115 showing upon GM-CSF treatment of human mononuclear cell cultures [32], we detected a similar CD116 level irrespective of the treatment with activated T lymphocytes, suggesting that the inhibition of osteoclastogenesis is not regulated by CD116 expression.

In addition to T lymphocyte activation in vitro by mitogen-pulse or anti-CD3/CD28 antibodies, we extended the importance of our findings by showing that T lymphocytes alloactivated in vitro and in vivo also exerted a suppressive effect on osteoclastogenesis. These results have clinical significance, as bone homeostasis is often severely disturbed after organ transplantation that involves T lymphocyte alloactivation [38,39]. Moreover, we confirmed that activated T lymphocytes can suppress osteoclastogenesis in BM in vivo by the experiments where we used in vivo alloactivation by i.p. injection and regional foot-pad restimulation with allogeneic cells. When we cultured native non-adherent BM cells from these mice, osteoclastogenesis was suppressed significantly compared with cultures from unstimulated animals. These findings confirmed the complex function of cytokine expression patterns in the interplay between T lymphocytes and osteoclast progenitors in the BM environment, despite a negligible presence of mature T lymphocytes in BM [4,10].

The inhibitory effect of T lymphocyte activation on osteoclastogenesis was abolished by T lymphocyte pretreatment with immunosuppressant CsA. CsA action leads to the inhibition of NFAT and, hence, suppression of the transcription of IL-2 and other cytokine genes, including GM-CSF [24,33,40]. T lymphocyte treatment with CsA prior to the Con A-pulse down-regulated GM-CSF expression, which is consistent with the presence of NFAT response element in the enhancer part of the GM-CSF transcription-control region [13,41]. In addition, we found only an insignificant decrease in RANKL expression after CsA pretreatment, although a previous study reported that CsA potently blocked RANKL in T cell hybridoma A1·1 [7]. Observed changes in the cytokine pattern produced by T lymphocytes upon immunosuppressive treatment may contribute to the post-transplantation osteoporosis in patients receiving long-term immunosuppressive treatment [38,39].

Although it is known that the immune system has powerful effects on bone resorption, and both immune cells resident in BM and immunocompetent cells in inflammatory tissue may interfere with bone metabolism, the exact nature of molecular effects still needs to be discovered before we fully understand the complexity of interactions between T lymphocytes and osteoclasts. As a contribution to this aim, our study clearly demonstrated the inhibitory role of activated T lymphocytes on osteoclastogenesis. The underlying mechanism involved increased production of GM-CSF, which drives the early monocyte/macrophage progenitor lineage commitment towards DC differentiation through the down-regulation of RANK expression and c-Fos transcription. This may be a significant regulatory loop and a possible therapeutic target to counteract the activated bone resorption mediated by T lymphocyte derived pro-osteoclastogenic cytokines in inflammatory and immune-disorders. In our study we also demonstrated the importance of in vivo and in vitro alloactivation in the regulation of osteoclast differentiation, thereby providing the novel evidence of the functional link between alloreaction and bone homeostasis. The ability of CsA to block T lymphocyte activation and hence abrogate the inhibitory effect of T lymphocytes on osteoclastogenesis may be important in clinical states associated with CsA treatment and contribute to the rapid bone loss that occurs after organ transplantation.

Acknowledgments

This work was supported by grants from the Croatian Ministry of Science, Education and Sports (0108342, 0108181 and 0108125). RANKL was a kind gift from Amgen Inc. (Thousand Oaks, CA, USA). We thank Mrs Katerina Zrinski-Petrović for technical assistance.

References

  • 1.Takayanagi H. Mechanistic insight into osteoclast differentiation in osteoimmunology. J Mol Med. 2005;83:170–9. doi: 10.1007/s00109-004-0612-6. [DOI] [PubMed] [Google Scholar]
  • 2.Udagawa N. The mechanism of osteoclast differentiation from macrophages. possible roles of T lymphocytes in osteoclastogenesis. J Bone Miner Metab. 2003;21:337–43. doi: 10.1007/s00774-003-0439-1. [DOI] [PubMed] [Google Scholar]
  • 3.Theill LE, Boyle WJ, Penninger JM. RANK-L and RANK: T cells, bone loss, and mammalian evolution. Annu Rev Immunol. 2002;20:795–823. doi: 10.1146/annurev.immunol.20.100301.064753. [DOI] [PubMed] [Google Scholar]
  • 4.Grcevic D, Lee SK, Marusic A, Lorenzo JA. Depletion of CD4 and CD8 T lymphocytes in mice in vivo enhances 1,25-dihydroxyvitamin D3-stimulated osteoclast-like cell formation in vitro by a mechanism that is dependent on prostaglandin synthesis. J Immunol. 2000;165:4231–8. doi: 10.4049/jimmunol.165.8.4231. [DOI] [PubMed] [Google Scholar]
  • 5.Rho J, Takami M, Choi Y. Osteoimmunology: interactions of the immune and skeletal systems. Mol Cells. 2004;17:1–9. [PubMed] [Google Scholar]
  • 6.Kong YY, Feige U, Sarosi I, et al. Activated T cells regulate bone loss and joint destruction in adjuvant arthritis through osteoprotegerin ligand. Nature. 1999;402:304–9. doi: 10.1038/46303. [DOI] [PubMed] [Google Scholar]
  • 7.Wang R, Zhang L, Zhang X, et al. Regulation of activation-induced receptor activator of NF-kappaB ligand (RANKL) expression in T cells. Eur J Immunol. 2002;32:1090–8. doi: 10.1002/1521-4141(200204)32:4<1090::AID-IMMU1090>3.0.CO;2-P. [DOI] [PubMed] [Google Scholar]
  • 8.Horwood NJ, Kartsogiannis V, Quinn JM, Romas E, Martin TJ, Gillespie MT. Activated T lymphocytes support osteoclast formation in vitro. Biochem Biophys Res Commun. 1999;265:144–50. doi: 10.1006/bbrc.1999.1623. [DOI] [PubMed] [Google Scholar]
  • 9.Pang M, Martinez AF, Jacobs J, Balkan W, Troen BR. RANK ligand and interferon gamma differentially regulate cathepsin gene expression in pre-osteoclastic cells. Bichem Biophys Res Commun. 2005;328:756–63. doi: 10.1016/j.bbrc.2004.12.005. [DOI] [PubMed] [Google Scholar]
  • 10.Walsh MC, Kim N, Kadono Y, et al. Osteoimmunology: interplay between the immune system and bone metabolism. Annu Rev Immunol. 2006;24:2.1–31. doi: 10.1146/annurev.immunol.24.021605.090646. [DOI] [PubMed] [Google Scholar]
  • 11.Horwood NJ, Udagawa N, Elliott J, et al. Interleukin 18 inhibits osteoclast formation via T cell production of granulocyte macrophage colony-stimulating factor. J Clin Invest. 1998;101:595–603. doi: 10.1172/JCI1333. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Wyzga N, Varghese S, Wikel S, Canalis E, Sylvester FA. Effects of activated T cells on osteoclastogenesis depend on how they are activated. Bone. 2004;35:614–20. doi: 10.1016/j.bone.2004.04.022. [DOI] [PubMed] [Google Scholar]
  • 13.Shinoda K, Sugiyama E, Taki H, et al. Resting T cells negatively regulate osteoclast generation from peripheral blood monocytes. Bone. 2003;33:711–20. doi: 10.1016/s8756-3282(03)00230-8. [DOI] [PubMed] [Google Scholar]
  • 14.Choi Y, Woo KM, Ko S-H, et al. Osteoclastogenesis is enhanced by activated B cells but suppressed by activated CD8+ T cells. Eur J Immunol. 2001;31:2179–88. doi: 10.1002/1521-4141(200107)31:7<2179::aid-immu2179>3.0.co;2-x. [DOI] [PubMed] [Google Scholar]
  • 15.Takahashi K, Azuma T, Motohira H, Kinane DF, Kitetsu S. The potential role of interleukin-17 in the immunopathology of periodontal disease. J Clin Periodontol. 2005;32:369–74. doi: 10.1111/j.1600-051X.2005.00676.x. [DOI] [PubMed] [Google Scholar]
  • 16.Teng YT, Mahamed D, Singh B. Gamma interferon positively modulates Actinobacillus actinomycetemcomitans-specific RANKL+ CD4+ Th-cell-mediated alveolar bone destruction in vivo. Infect Immun. 2005;73:3453–61. doi: 10.1128/IAI.73.6.3453-3461.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Matsuo K, Ray N. Osteoclasts, mononuclear phagocytes, and c-Fos: new insight into osteoimmunology. Keio J Med. 2004;53:78–84. doi: 10.2302/kjm.53.78. [DOI] [PubMed] [Google Scholar]
  • 18.Miyamoto T, Suda T. Differentiation and function of osteoclasts. Keio J Med. 2003;52:1–7. doi: 10.2302/kjm.52.1. [DOI] [PubMed] [Google Scholar]
  • 19.Teitelbaum SL. Bone resorption by osteoclasts. Science. 2000;289:1504–8. doi: 10.1126/science.289.5484.1504. [DOI] [PubMed] [Google Scholar]
  • 20.Li J, Sarosi I, Yan XQ, et al. RANK is the intrinsic hematopoietic cell surface receptor that controls osteoclastogenesis and regulation of bone mass and calcium metabolism. Proc Natl Acad Sci USA. 2000;97:1566–71. doi: 10.1073/pnas.97.4.1566. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Boyle WJ, Simonet WS, Lacey DL. Osteoclast differentiation and activation. Nature. 2003;423:337–42. doi: 10.1038/nature01658. [DOI] [PubMed] [Google Scholar]
  • 22.Miyamoto T, Ohneda O, Arai F, et al. Bifurcation of osteoclasts and dendritic cells from common progenitors. Blood. 2001;98:2544–54. doi: 10.1182/blood.v98.8.2544. [DOI] [PubMed] [Google Scholar]
  • 23.Rivollier A, Mazzorana M, Tebib J, et al. Immature dendritic cell transdifferentiation into osteoclasts: a novel pathway sustained by the rheumatoid arthritis microenvironment. Blood. 2004;104:4029–37. doi: 10.1182/blood-2004-01-0041. [DOI] [PubMed] [Google Scholar]
  • 24.Grcevic D, Batinic D, Ascensao JL, Marusic M. Pre-treatment of transplant bone marrow cells with hydrocortisone and cyclosporin A alleviates graft-versus-host reaction in a murine allogeneic host-donor combination. Bone Marrow Transplant. 1999;23:1145–52. doi: 10.1038/sj.bmt.1701777. [DOI] [PubMed] [Google Scholar]
  • 25.Coligan JE, Kruisbeek AM, Margulies DH, Shevach EM, Strober W. Current protocols in immunology. New York, NY: John Wiley & Sons, Inc.; 1994. [Google Scholar]
  • 26.Katavic V, Lukic IK, Kovacic N, Grcevic D, Lorenzo JA, Marusic A. Increased bone mass is a part of the generalized lymphoproliferative disorder phenotype in the mouse. J Immunol. 2003;170:1540–7. doi: 10.4049/jimmunol.170.3.1540. [DOI] [PubMed] [Google Scholar]
  • 27.Martinez-Lorenzo MJ, Anel A, Alava MA, et al. The human melanoma cell line MelJuSo secretes bioactive FasL and APO2L/TRAIL on the surface of microvesicles. Possible contribution to tumor counterattack. Exp Cell Res. 2004;295:315–29. doi: 10.1016/j.yexcr.2003.12.024. [DOI] [PubMed] [Google Scholar]
  • 28.Yasuda H, Shima N, Nakagawa N, et al. Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci USA. 1998;95:3597–602. doi: 10.1073/pnas.95.7.3597. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Abdalla AO, Kiaii S, Hansson L, et al. Kinetics of cytokine gene expression in human CD4+ and CD8+ T lymphocyte subsets using quantitative real-time PCR. Scand J Immunol. 2003;58:601–6. doi: 10.1111/j.1365-3083.2003.01348.x. [DOI] [PubMed] [Google Scholar]
  • 30.Arai F, Miyamoto T, Ohneda O, et al. Commitment and differentiation of osteoclast precursor cells by the sequential expression of c-Fms and receptor activator of nuclear factor kappaB (RANK) receptors. J Exp Med. 1999;190:1741–54. doi: 10.1084/jem.190.12.1741. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Feng X. Regulatory roles and molecular signaling of TNF family members in osteoclasts. Gene. 2005;350:1–13. doi: 10.1016/j.gene.2005.01.014. [DOI] [PubMed] [Google Scholar]
  • 32.Day CJ, Kim MS, Stephens SR, et al. Gene array identification of osteoclast genes: differential inhibition of osteoclastogenesis by cyclosporin A and granulocyte macrophage colony stimulating factor. J Cell Biochem. 2004;91:303–15. doi: 10.1002/jcb.10780. [DOI] [PubMed] [Google Scholar]
  • 33.Ishida N, Hayashi K, Hoshijima M, et al. Large scale gene expression analysis of osteoclastogenesis in vitro and elucidation of NFAT2 as a key regulator. J Biol Chem. 2002;277:41147–56. doi: 10.1074/jbc.M205063200. [DOI] [PubMed] [Google Scholar]
  • 34.Hodge JM, Kirkland MA, Aitken CJ, et al. Osteoclastic potential of human CFU-GM: biphasic effect of GM-CSF. J Bone Miner Res. 2004;19:190–9. doi: 10.1359/JBMR.0301232. [DOI] [PubMed] [Google Scholar]
  • 35.Kim MS, Day CJ, Morrison NA. MCP-1 is induced by receptor activator of nuclear factor-{kappa}B ligand, promotes human osteoclast fusion, and rescues granulocyte macrophage colony-stimulating factor suppression of osteoclast formation. J Biol Chem. 2005;280:16163–9. doi: 10.1074/jbc.M412713200. [DOI] [PubMed] [Google Scholar]
  • 36.Gorny G, Shaw A, Oursler MJ. IL-6, LIF, and TNF-alpha regulation of GM-CSF inhibition of osteoclastogenesis in vitro. Exp Cell Res. 2004;294:149–58. doi: 10.1016/j.yexcr.2003.11.009. [DOI] [PubMed] [Google Scholar]
  • 37.Grigoriadis AE, Wang ZQ, Cecchini MG, et al. c-Fos: a key regulator of osteoclast-macrophage lineage determination and bone remodeling. Science. 1994;266:443–8. doi: 10.1126/science.7939685. [DOI] [PubMed] [Google Scholar]
  • 38.Lee WY, Baek KH, Rhee EJ, et al. Impact of circulating bone-resorbing cytokines on the subsequent bone loss following bone marrow transplantation. Bone Marrow Transplant. 2004;34:89–94. doi: 10.1038/sj.bmt.1704535. [DOI] [PubMed] [Google Scholar]
  • 39.Kerschan-Schindl K, Mitterbauer M, Fureder W, et al. Bone metabolism in patients more than five years after bone marrow transplantation. Bone Marrow Transplant. 2004;34:491–6. doi: 10.1038/sj.bmt.1704618. [DOI] [PubMed] [Google Scholar]
  • 40.Hirotani H, Tuohy NA, Woo JT, Stern PH, Clipstone NA. The calcineurin/nuclear factor of activated T cells signaling pathway regulates osteoclastogenesis in RAW264.7 cells. J Biol Chem. 2004;279:13984–92. doi: 10.1074/jbc.M213067200. [DOI] [PubMed] [Google Scholar]
  • 41.Shannon MF, Coles LS, Vadas MA, Cockerill PN. Signals for activation of the GM-CSF promoter and enhancer in T cells. Crit Rev Immunol. 1997;17:301–23. doi: 10.1615/critrevimmunol.v17.i3-4.30. [DOI] [PubMed] [Google Scholar]

Articles from Clinical and Experimental Immunology are provided here courtesy of British Society for Immunology

RESOURCES