Thapsigargin Modulates Osteoclastogenesis Through the Regulation of RANKL-Induced Signaling Pathways and Reactive Oxygen Species Production

Authors

  • Kirk Hm Yip,

    1. Molecular Orthopaedic Laboratory, School of Surgery and Pathology, and Western Australian Institute for Medical Research, Nedlands, Australia
    Search for more papers by this author
  • Ming H Zheng,

    1. Molecular Orthopaedic Laboratory, School of Surgery and Pathology, and Western Australian Institute for Medical Research, Nedlands, Australia
    Search for more papers by this author
  • James H Steer,

    1. Pharmacology Unit, School of Medicine and Pharmacology, University of Western Australia, Nedlands, Australia
    Search for more papers by this author
  • Tindaro M Giardina,

    1. Pharmacology Unit, School of Medicine and Pharmacology, University of Western Australia, Nedlands, Australia
    Search for more papers by this author
  • Renzhi Han,

    1. Molecular Orthopaedic Laboratory, School of Surgery and Pathology, and Western Australian Institute for Medical Research, Nedlands, Australia
    2. Present address: Howard Hughes Medical Institute University of Iowa, College of Medicine, 400 EMRB, Iowa City, Iowa, USA
    Search for more papers by this author
  • Susan Z Lo,

    1. Pharmacology Unit, School of Medicine and Pharmacology, University of Western Australia, Nedlands, Australia
    Search for more papers by this author
  • Anthony J Bakker,

    1. Physiology Unit, School of Biomedical and Chemical Sciences, University of Western Australia, Nedlands, Australia
    Search for more papers by this author
  • A Ian Cassady,

    1. Institute for Molecular Biosciences and School of Molecular and Microbial Sciences, The University of Queensland, Brisbane, Queensland, Australia
    Search for more papers by this author
  • David A Joyce,

    1. Pharmacology Unit, School of Medicine and Pharmacology, University of Western Australia, Nedlands, Australia
    Search for more papers by this author
  • Jiake Xu MD, PhD

    Corresponding author
    1. Molecular Orthopaedic Laboratory, School of Surgery and Pathology, and Western Australian Institute for Medical Research, Nedlands, Australia
    • Molecular Orthopaedic Laboratory, School of Surgery and Pathology, University of Western Australia, QEII Medical Centre, 2nd Floor, M Block, Nedlands, WA 6009, Australia

    Search for more papers by this author

  • The authors have no conflict of interest.

Abstract

The mechanism by which TG modulates osteoclast formation and apoptosis is not clear. In this study, we showed a biphasic effect of TG on osteoclast formation and apoptosis through the regulation of ROS production, caspase-3 activity, cytosolic Ca2+, and RANKL-induced activation of NF-κB and AP-1 activities.

Introduction: Apoptosis and differentiation are among the consequences of changes in intracellular Ca2+ levels. In this study, we investigated the effects of the endoplasmic reticular Ca2+-ATPase inhibitor, thapsigargin (TG), on osteoclast apoptosis and differentiation.

Materials and Methods: Both RAW264.7 cells and primary spleen cells were used to examine the effect of TG on RANKL-induced osteoclastogenesis. To determine the action of TG on signaling pathways, we used reporter gene assays for NF-κB and activator protein-1 (AP-1) activity, Western blotting for phospho-extracellular signal-related kinase (ERK), and fluorescent probes to measure changes in levels of intracellular calcium and reactive oxygen species (ROS). To assess rates of apoptosis, we measured changes in annexin staining, caspase-3 activity, and chromatin and F-actin microfilament structure.

Results: At concentrations that caused a rapid rise in intracellular Ca2+, TG increased caspase-3 activity and promoted apoptosis in osteoclast-like cells (OLCs). Low concentrations of TG, which were insufficient to measurably alter intracellular Ca2+, unexpectedly suppressed caspase-3 activity and enhanced RANKL-induced osteoclastogenesis. At these lower concentrations, TG potentiated ROS production and RANKL-induced NF-κB activity, but suppressed RANKL-induced AP-1 activity and had little effect on ERK phosphorylation.

Conclusion: Our novel findings of a biphasic effect of TG are incompletely explained by our current understanding of TG action, but raise the possibility that low intensity or local changes in subcellular Ca2+ levels may regulate intracellular differentiation signaling. The extent of cross-talk between Ca2+ and RANKL-mediated intracellular signaling pathways might be important in determining whether cells undergo apoptosis or differentiate into OLCs.

INTRODUCTION

IONIC INTRACELLULAR CALCIUM (Ca2+) signaling regulates proliferation, differentiation, and death in different cell types,(1) as well as osteoclasts, which are bone-resorbing cells of the monocyte-macrophage lineage.(2) Cytosolic Ca2+ mediates cell adhesion, bone resorptive activity, and survival through regulation of RANKL (OPGL/ODF/TRANCE) and the organization of the cytoskeleton of osteoclasts.(3–6) The exact nature of signaling pathways that link changes in cytosolic Ca2+ to osteoclast differentiation and survival has not been fully elucidated.

RANKL is critical for the differentiation of osteoclast precursors and for the survival of osteoclasts.(7–10) RANKL interacts with its receptor RANK and results in the recruitment of the TNF receptor-associated factor (TRAF) adapter proteins and activation of signaling pathways including NF-κB, c-jun N-terminal kinase (JNK), extracellular signal-related kinase (ERK), and nuclear factor of activated T cells (NF-AT).(11) Recent studies have shown that RANKL can also inhibit cell proliferation and induce apoptosis mediated by the JNK pathway.(12) At the same time, RANK signaling also acts through phospholipase C to release Ca2+ from intracellular stores, accelerating nuclear translocation of NF-κB and promoting osteoclast survival.(13) These observations suggest that apoptosis and differentiation are primary consequences of activation of intracellular Ca2+ signaling and RANKL-induced pathways in osteoclasts.

Thapsigargin (TG), a pure compound isolated from Thapsia garganica, is a highly selective inhibitor of the endoplasmic reticular (ER) Ca2+-dependent ATPase. TG is known to deplete intracellular Ca2+ stores and induce a sustained Ca2+ influx in osteoclasts.(4, 5) Because TG promotes apoptosis in a range of cell types, its therapeutic use in the treatment of cancers has been proposed.(14–17) The action of TG on osteoclast formation and RANKL signaling remains unclear. Therefore, in this study, we investigated the effect of TG on osteoclast apoptosis and differentiation through (1) the regulation of caspase-3 activity; (2) its effect on cytosolic Ca2+ concentration; and (3) RANKL-induced activation of NF-κB and activator protein-1 (AP-1), ERK phosphorylation, and reactive oxygen species (ROS) production. Our results show that TG concentrations that rapidly increase the intracellular Ca2+ concentration also induce apoptosis, with an accompanying activation of the caspase-3 pathway. At concentrations of TG that lie below the range generally used to evoke a rapid rise in intracellular Ca2+ concentration, we unexpectedly find that TG decreases caspase-3 activation and increases osteoclastogenesis through RANKL-induced NF-κB activity and ROS production. Thus, selective modulation of Ca2+ and RANKL signaling pathways determine the consequence of osteoclast formation and apoptosis.

MATERIALS AND METHODS

Reagents

RAW264.7 cells were obtained from the American Type Culture Collection (Rockville, MD, USA). TG and diagnostic acid phosphatase kits were purchased from Sigma (Sydney, Australia). The luciferase assay system was obtained from Promega (Sydney, Australia). Annexin V-PE staining reagents was purchased from BD Biosciences. Recombinant GST-rRANKL proteins was expressed and purified as previously described.(18)

In vitro osteoclastogenesis assay

RAW264.7 cells and spleen cells were cultured in α-MEM (Biosciences Pty Ltd.) supplemented with 10% FCS, 2 mM l-glutamine, and 100 U/ml penicillin/streptomycin. For RAW264 cell-derived osteoclast cultures, RAW264.7 cells were seeded in a 6-well plate to a density of 5 × 104 cells/well and cultured for 5–7 days in full α-MEM in the presence of 100 ng/ml of GST-rRANKL. To examine the effect of TG on osteoclastogenesis, RAW264.7 cells were plated in a 96-well plate (2 × 103 cells/well) and cultured in complete medium in the presence of GST-rRANKL, TG, or RANKL plus TG. The growth medium was replaced every 2 days. After 5 days, cultures were fixed for 10 minutes at room temperature with 4% paraformaldehyde in PBS and washed four times with PBS. The fixed cells were stained for TRACP using the Diagnostic Acid Phosphatase kit (Sigma) according to the manufacturer's instructions, and TRACP+ multinucleated cells (>3 nuclei) were scored as osteoclast-like cells (OCLs). In some experiments, OCLs with >15 nuclei were calculated. For primary cell culture, freshly isolated spleen cells from C57/BL mice were seeded in a 96-well plate (1 × 105 cells/well) in the presence of 10 ng/ml of macrophage-colony stimulating factor (M-CSF) for the first 3 days. After that, medium was replaced with fresh medium containing 10 ng/ml of M-CSF and 100 ng/ml of GST-rRANKL every 2–3 days for 10 days. TRACP staining was performed as above.

NF-κB and AP-1 activation in RAW264.7 cells and bone marrow cells

To examine NF-κB activation, RAW264.7 cells were transiently transfected with luciferase reporter genes.(6, 19) Cells were plated in 24-well plates at a density of 1 × 105 cells/well and treated with RANKL, TG, or RANKL plus TG. At appropriate times, luciferase activity was measured in the cells using the Promega Luciferase Assay System according the manufacturer's instructions (Promega).

The effects of TG on RANKL-induced AP-1 activation were studied using RAW264.7 cells transiently expressing a luciferase reporter construct that contained multiple AP-1 binding sites in its promoter [pAP1(PMA)-TA-Luc; Mercury Pathway Profiling Luciferase System 2; BD Biosciences]. Plasmid DNA was transfected as previously described,(20) and luciferase activity was measured at appropriate times after stimulation.

Confocal microscopy

Morphological changes in the nuclear chromatin of cells undergoing apoptosis were detected by staining with the DNA-binding fluorochrome bis-benzamide (Hoechst 33258; Molecular Probes, Eugene, OR, USA) as previously described.(21) In brief, after treatment with various concentrations of TG, OLCs were washed twice with PBS and fixed with 4% of paraformaldehyde for 20 minutes. The fixed cells were wash twice with PBS and stained with Hoechst 33258 (1 μg/ml) for 3 minutes to visualize the cell nuclei. To examine cytoskeleton structure, OLCs were co-stained for F-actin microfilament with Rhodamine-conjugated Phalloidin (1:200; Molecular Probes) for 2 h at room temperature. The fluorescent images were collected on a BioRad MRC 1000/1024 UV laser scanning confocal microscopy.

Flow cytometry

RAW264.7 cells treated with TG were harvested and incubated with Annexin V-PE according to the manufacturer's instructions (BD Biosciences) and analyzed by flow cytometry (FACSCalibur; Becton Dickinson, San Jose, CA, USA) as previously described.(22) Results are expressed as the percentage of Annexin+ cells in the population (104 cells).

Caspase-3 assay

RAW264.7 cells (2 × 106) were seeded in 24-well plates, cultured overnight, and treated with various doses of TG. The treated cells were released from the plates by trypsinization and vigorous pipetting. The cells were centrifuged at 600g for 5 minutes, washed once in DPBS, and lysed by three freeze thaws in 20 μl of lysis buffer containing 50 mM HEPES, pH 7.4, 100 mM NaCl, 0.1% 3-[(3-choamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS), 0.1 mM EDTA, 1 mM 4-dithio-dl-threitol (DTT), 0.5 mM phenylmethanesulfonyl fluoride (PMSF), 5 μg/ml pepstatin A, and 10 μg/ml leupeptin. The protein content of the lysate was determined by a Bradford assay (BioRad). The activity of caspase-3 in lysates was determined using a kinetic assay in buffer containing 50 mM HEPES, pH 7.4, 100 mM NaCl, 0.1% CHAPS, 1 mM EDTA, and 10% glycerol, by monitoring the cleavage of acetyl-DEVD-AFC in the presence or absence of the caspase-3 inhibitor Ac-DEVD-CHO (1 μM; Promega). The changes in fluorescence of amido-4-trifluoromethylcoumarin (AFC) were measured at 510 nm after excitation at 400 nm in a multifunctional microplate reader (POLARstar OPTIMA; BMG).

Cytosolic Ca2+ measurement

Cells destined for Ca2+ measurement were grown on 10-mm glass coverslips in 35-mm culture Petri dishes. They were washed twice with fresh physiological rodent saline (PRS: 138 mM NaCl, 2.7 mM KCl, 1.8 mM CaCl2, 1.06 mM MgCl2, 12.4 mM HEPES, and 5.6 mM glucose; pH 7.3) and incubated for 45 minutes at room temperature with the fluorescent Ca2+ indicator FFP-18/AM (3 μM; Teflabs, Austin, TX, USA) and 0.0125% pluronic F-127 (wt/vol) in PRS. FFP-18 is a Ca2+ indicator based on fura-2 that remains bound to the intracellular side of the plasma membrane. Like fura-2, it is sensitive to the release of Ca2+ from intracellular stores, but it is also able to detect much smaller local increases in Ca2+ (e.g., at the inner surface of the plasmalemma).(23) After loading, the cells were washed with PRS and placed in the dark at 37°C for 15 minutes to allow time for dye de-esterification. For the Ca2+ measurements, a coverslip was placed in a bath chamber placed on the stage of an inverted epifluorescence microscope (Nikon TE2000). The ratio of fluorescence emission (510 nm) at 340- and 380-nm excitation (F340/F380) was measured with a spectrophotometer (Cairn) and analyzed with proprietary software (Cairn). All Ca2+ measurement experiments were undertaken at ∼35°C.

ROS measurement

RAW264.7 cells were cultured overnight in DMEM (Biosciences Pty Ltd.) at an initial density of 50,000 cells/well in clear 96-well plates. Cells were washed in Hanks balanced salt solution (HBSS) and incubated in HBSS containing the stimuli TG, RANKL, or TG plus RANKL and 10 μM dihydroethidium (DHE; Molecular Probes). DHE freely enters cells and is oxidized by superoxide to yield fluorescent ethidium, which binds to DNA, further amplifying the signal. Twenty four hours after stimulation, cells were washed and suspended in 100 μl of HBSS. Fluorescence intensity was measured at 590 nm after excitation at 510 nm using a fluorescent plate reader (POLARstar OPTIMA; BMG). In addition, freshly isolated spleen cells were pretreated with 10 ng/ml M-CSF for 3 days and used for ROS measurements as described above.

Western blotting analysis of active ERK

Proteins were separated by SDS-PAGE and electroblotted onto nitrocellulose membranes (BioRad). Membranes were blocked with 5% (wt/vol) nonfat milk powder in TBST (10 mM Tris, pH 7.5, 150 mM NaCl, 0.1% [vol/vol] Tween 20) and probed with primary antibodies to phosphorylated forms of ERK and β-tubulin (Santa Cruz Biotechnology) in the blocking solution. After washing three times with TBS, membranes were incubated with horseradish peroxidase-conjugated secondary antibodies diluted 1/5000 in 1% (wt/vol) nonfat milk powder in TBST. The membranes were developed using the enhanced chemiluminescence (ECL) system (Amersham Pharmacia Biotech).

Statistical analyses

Results are presented as mean ± SE from three or more experiments. Student's t-test was used to test statistical significance between groups. A p value of <0.05 was considered to be statistically significant.

RESULTS

TG modulates RANKL-induced osteoclastogenesis

To investigate the role of TG-mediated calcium signaling on RANKL-induced osteoclastogenesis, we first examined the effect of TG on osteoclast formation using RAW264.7 cell cultures. RAW264.7 cells were cultured in the presence of 100 ng/ml RANKL in the absence of M-CSF for 5 days as previously described.(6, 18, 24) TG was added into the culture at a final concentration of 10, 2, 1, 0.5, 0.25, 0.125, or 0.0625 nM during the first 2 days (day 1–3), last 2 days (day 3–5), or full course of culture (days 1–5). The treated cells were fixed and stained for TRACP activity, and OLC formation was quantitated. As shown in Fig. 1, a biphasic relationship between OLC numbers and TG concentrations was apparent in each of the three exposure conditions. Exposure to concentrations of TG (days 1–3 or 3–5) between 0.0625 and 0.25 nM increased the numbers of RAW264.7 cells differentiating to OLCs at 5 days. Most notably, addition of TG led to the formation of OLCs with larger sizes and greater multinucleation compared with the untreated controls (Fig. 1). Higher concentrations of TG (0.5–10 nM) caused cell detachment and suppression of OLC formation (Fig. 1; data not shown). Five days of exposure (days 1–5) to TG concentrations between 0.0625 and 0.125 nM also increased OLC formation, whereas 0.25 nM TG caused suppression of osteoclastogenesis (Fig. 1), and higher concentrations caused cell detachment (data not shown).

Figure FIG. 1..

TG modulates RANKL-induced RAW264.7 cell differentiation into OLCs. RAW264.7 cells were cultured in a 96-well plate in the presence of RANKL for 5 days and TG was added for days 1–3, 3–5, or 1–5. (A) After 5 days, the cells were fixed with 4% paraformaldehyde and stained for TRACP activity. (B) High magnification of images from A. Representative images of triplicate wells from one of three experiments are shown. (C) Effect of TG on RANKL-induced osteoclastogenesis. The results represent three experiments. TRACP+ cells with >15 nuclei are scored (*p < 0.05, **p < 0.01, ***p < 0.001 compared with untreated control).

These observations were tested using a broader range of experimental conditions, with initial cell densities of between 1 × 103 and 1.6 × 104 per well of a 96-well plate cultured with RANKL (100 ng/ml) and a range of TG concentrations for 5 days. A biphasic effect on osteoclastogenesis was observed at all cell densities. OLC formation was enhanced from 0.125 to 0.25 nM TG but was diminished at 0.5 nM TG (Fig. 2). Exposure to 0.5 nM TG for a period of 5 days caused cell detachment from culture wells consistent with cell death. The greatest enhancement of TRACP+ multinucleated OLCs was observed in RANKL-treated cells grown at a density of 2 × 103 and 4 × 103 cells per well treated with 0.125 nM TG (Fig. 2).

Figure FIG. 2..

Cell density-dependent effect of sublethal doses of TG on RANKL-induced osteoclastogenesis. RAW264.7 cells were seeded at densities ranging from 1000 to 16,000 per well in a 96-well plate and cultured in the presence of RANKL. TG was added at final concentrations of 0, 0.125, 0.25, and 0. 5 nM for days 1–5. (A) After 5 days, the cells were fixed with 4% paraformaldehyde and stained for TRACP activity. (B) High magnification of images from A. Representative images from one of the three experiments are shown. (C) Quantitative analysis was performed using densitometry (**p < 0.01, ***p < 0.001 compared with untreated control).

To further confirm these observations in normal cells, freshly isolated spleen cells were cultured in the presence of 10 ng/ml of M-CSF and 100 ng/ml RANKL as previously described.(18) TG was added into the culture at a final concentration of 10, 2, 1, 0.5, 0.25, and 0.125 nM during the 10-day culture. The treated cells were fixed and stained for TRACP activity, and OLC formation was quantitated. Similar to the effect of TG on RAW cells, a biphasic relationship between OLC numbers and TG concentrations was observed (Fig. 3). Exposure to TG concentrations at 0.125 nM significantly increase OLC formation, whereas 0.5–2 nM TG caused suppression of osteoclastogenesis, and higher concentrations caused cell loss (data not shown). Also notably, addition of TG at 0.125 nM led to the formation of larger OLCs compared with untreated controls. Taken together, these results indicate that TG potentiates RANKL-induced osteoclastogenesis but, at higher concentrations, few OLCs formed, and most cells detached from the dish.

Figure FIG. 3..

TG modulates RANKL-induced osteoclastogenesis from primary cell culture. Freshly isolated spleen cells from C57/BL mice were cultured in a 96-well plate in the presence of M-CSF for the first 3 day and M-CSF and RANKL for an additional 7 days with or without TG. (A) After 10 days, the cells were fixed with 4% paraformaldehyde and stained for TRACP activity. Representative images of triplicate wells from one of three experiments are shown. (B) Quantitation of the effect of TG on RANKL-induced osteoclastogenesis. TRACP+ OCLs are scored (*p < 0.05, ***p < 0.001, NS, no significance compared with control).

TG activates caspase-3, which induces apoptosis of RAW264.7 cell-derived OLCs and pre-osteoclastic RAW264.7 cells

To establish the mechanism of the high TG dosage inhibition in osteoclastogenesis, we examined the morphology of treated cells using confocal microscopy. TG induces nuclear fragmentation (Fig. 4B) and disruption of the F-actin ring of RAW264.7 cell-derived OLCs (Fig. 4D). Under light microscopy, the TG-treated OLCs displayed characteristics of cell shrinkage and cytoplasmic condensation, which are features of apoptosis (Fig. 4F). To confirm the induction of apoptosis, we used immunocytochemical staining of the proapoptotic marker, annexin V, by flow cytometry. These data showed that TG-treated OLC apoptosis was dose-dependent, rising from 1.25% of cells in the absence of TG to 47.2% at 100 nM TG (Fig. 4G).

Figure FIG. 4..

TG triggers apoptosis in OLCs. The cultured RAW264.7 cells derived OLCs were treated with (B, D, and F) 400 nM TG or (A, C, and E) left untreated overnight and fixed with 4% paraformaldehyde in PBS and stained with Hoechst 33258 dye and with F-actin antibody. (A and C) Confocal images showing (A) the nuclei staining and (C) F-actin ring structure in the untreated cells. (B and D) TG induces (B) nuclei fragmentation (arrowheads) and the (D) disruption of F-actin ring formation. (E and F) Light microscopy images showing the effect of TG on RAW264.7 cell-derived OLCs. (F) Note that TG-treated cells exhibited shrinkage and cytoplasmic condensation, morphological changes characteristic of apoptotic death. (G) Flow cytometry analysis showing a dose-dependent effect of TG on rates of apoptosis in osteoclastic precursor RAW264.7 cells. RAW246.7 cells (2 × 105) seeded in 6-well plates were treated with varying concentrations of TG. After overnight incubation, all cells were harvested and stained with Annexin V-PE, and 10,000 cells were subjected to analysis by flow cytometry. Results are representative of three independent experiments. (H and I) The effect of TG on caspase-3 activity. RAW264.7 cells were treated with varying concentrations of TG, and caspase-3 activities were measured using a fluorometric assay. Caspase-3 activity corrected by (H) initial cell number and (I) final amount of protein are presented (***p < 0.001 compared with untreated controls).

To further study the apoptotic pathways induced by TG, RAW264.7 cells treated with TG were evaluated for caspase-3 activity using a fluorometric assay system. The results showed that TG caused a significant increase in caspase-3 activity at concentrations of 10 nM or greater (Figs. 4H and 4I), indicating that TG-induced apoptosis is associated with caspase-3 activation in RAW264.7 cells. Interestingly, concentrations <5 nM TG reduced intrinsic caspase activity in the cells (Figs. 4H and 4I). Taken together, these results show that TG induces apoptosis of RAW264.7 cell-derived OLCs and pre-osteoclastic RAW264.7 cells that involves the activation of caspase-3.

Effect of TG on intracellular Ca2+ elevation

TG-stimulated increases in intracellular Ca2+ are involved in the induction of apoptosis in other cell types,(25) and we have shown that TG induces OLC apoptosis at higher concentrations but modulates osteoclastogenesis at lower concentrations. Using Ca2+ fluorescence microscopy, we found that TG increased intracellular Ca2+ in RAW264.7 cell-derived OLCs and in RAW264.7 cells at concentrations of 100 nM or higher (Fig. 5C). No increase in intracellular Ca2+ was detected at 2 and 10 nM TG (Figs. 5A and 5B). It remains possible that we have not detected very localized increases in cytosolic Ca2+ concentration or exchanges of Ca2+ between organelles in close proximity to each other. Both of these circumstances could be occurring in cells exposed to low concentrations of TG in this study. Nevertheless, these results indicate that the indicative effects of TG on osteoclastogenesis at picomolar concentrations are unlikely to be caused by measurable acute elevation of intracellular Ca2+ concentration.

Figure FIG. 5..

(A-C) High concentrations of TG elevate intracellular Ca2+ in OLCs. RAW264.7 cell-derived OLCs were seeded onto 10-mm coverslips. After loading with FFP-18/AM (3 μM), cells were washed and treated with TG. Changes in F340/F380, the ratio of fluorescent intensity at 510 nm after excitation at 340 and 380 nm, a measure of intracellular Ca2+, were determined using an inverted epifluorescence microscope connected to a spectrophotometer. Traces, representative of six independent experiment, are shown for OLCs treated with (A) 2, (B) 10, and (C) 100 nM TG. (D) TG potentiates RANKL-induced NF-κB-dependent transcription in RAW264.7 cells. RAW264.7 cells, transiently transfected with the 3kB-Luc-SV40 reporter gene, were treated with TG (0, 0.38, 0.75, 1.5, 3, 6.25, 12.5, and 25nM) in the presence or absence of RANKL for 24 h before luciferase activity was measured. Results are expressed as means ± SE of triplicates. (E) TG potentiates RANKL-induced production of ROS. RAW264.7 cells were seeded overnight at a density of 50,000 cells/well in a clear 96-well plate and treated with TG (10 nM), RANKL (100 ng/ml), or TG plus RANKL in the presence of DHE for 24 h. Fluorescence intensities (mean ± SE, n = 5) were measured at 590 nm after excitation at 510 nm using a fluorescent plate reader (*p < 0.05, **p < 0.01, ***p < 0.001).

TG potentiates RANKL-induced activation of NF-κB

Signaling through the NF-κB pathway is critical for osteoclastogenesis and osteoclast progenitor survival,(26–28) which prompted us to ask the question: are the biphasic effects of TG on caspase-3 activity and osteoclast differentiation mediated through altered NF-κB activity? We measured the effects of TG on RANKL-induced NF-κB activity in osteoclast precursors, using an NF-κB-responsive luciferase reporter gene construct. RAW264.7 cells transiently transfected with the NF-κB luciferase reporter construct were pretreated with TG (25, 12.5, 6.25, 3.125, 1.5, 0.75, 0.375 nM) for 4 h followed by addition of RANKL for a further 8 h. The luciferase activity measured after 12 h showed that TG increases the basal level of NF-κB activity and the RANKL-induced activation of NF-κB (Fig. 5D).

Co-stimulatory effect of RANKL and TG on ROS production

Because ROS has been suggested to activate the NF-κB pathway,(29, 30) we set out to study whether TG, RANKL, or a combination of both enhanced ROS production in RAW264.7 cells. RANKL and TG each significantly increased the oxidization of dihydroethedium, a measure of ROS production, and the combination of RANKL and TG acted additively (Fig. 5E). To confirm that TG also increased ROS production from primary cells, freshly isolated spleen cells were pretreated with 10 ng/ml M-CSF for 3 days and treated with RANKL, TG, or a combination of both. TG significantly potentiated ROS production by RANKL (data not shown).

Effects of TG on RANKL-induced activation of AP-1 and ERK

RANKL signals through the RANK/TRAF6/JNK/AP-1 cascade, and this plays a critical role in osteoclastogenesis.(31–35) To examine the effect of TG on RANKL induction of AP-1, RAW264.7 cells were transiently transfected with an AP1-luciferase reporter gene construct and treated with RANKL, TG, or RANKL plus TG. The treated cells were harvested at various time-points to 24 h, and the luciferase activities were measured. As expected, RANKL alone induced AP-1 activation. Treatment with TG alone had little effect on AP-1 activation in RAW264.7, and TG repressed RANKL-induced AP-1 activation at all time-points in a dose-dependent fashion (Fig. 6).

Figure FIG. 6..

TG suppresses RANKL-induced AP-1-dependent transcription in RAW264.7 cells. RAW264.7 cells, transiently expressing an AP-1-luc reporter gene construct, were left untreated (Nil), treated with RANKL (100 ng/ml), TG at (A) 10 or (B) 1 nM, or both RANKL and TG for up to 24 h. Luciferase activities are the mean ± SE of triplicate determinations. p values comparing treatments of RANKL plus TG with TG control: *p < 0.05, **p < 0.01, ***p < 0.001. p values comparing treatments of RANKL with RANKL plus TG:#p < 0.05,##p < 0.01.

To examine the effect of TG on RANKL-induced ERK1/2 activation during osteoclastogenesis, Western blot analysis was performed using a specific antibody against the phosphorylated forms of ERK1/2. The phosphorylation of ERK was increased at both 15 and 30 minutes after RANKL stimulation, whereas TG had no effect on the levels of basal or RANKL-stimulated ERK phosphorylation (Fig. 7). Taken together, these data show that TG suppresses RANKL-induced AP-1 activation but has no effect on ERK phosphorylation.

Figure FIG. 7..

The effect of TG on RANKL-induced ERK phosphorylation in RAW264.7 cells. RAW264.7 cells were treated with RANKL in the presence or absence of TG (10 nM) for 0, 15, and 30 minutes, and proteins from whole cell extracts were separated by SDS-PAGE and transferred to nitrocellulose membranes. The membranes were probed sequentially with antibodies to phosphorylated ERK and β-tubulin. (A) Bands were visualized by ECL and quantitated by densitometry. Results from a representative of triplicate experiments are shown. The levels of (B) phosphorylated pERK1 and (C) pERK2 proteins are shown as the ratios of pERK1 or pERK2 to β-tubulin.

DISCUSSION

Ca2+ signals regulate a host of vital cell functions and are necessary for cell survival. However, the perturbation of intracellular Ca2+ levels in subcellular compartments has been shown to trigger apoptotic cell death.(1) In this study, we showed a biphasic effect of TG on osteoclast formation and apoptosis. At concentrations that rapidly increase intracellular Ca2+ concentration, TG activates caspase-3 and induces apoptosis of OLCs and osteoclast precursor cells. Unexpectedly, we found that concentrations of TG that were insufficient to measurably increase intracellular Ca2+ concentrations have an opposite effect, suppressing caspase-3 activity and enhancing RANKL-induced osteoclastic differentiation. These results reveal a novel cross-talk between Ca2+ and RANKL-mediated intracellular signaling pathways, identifying a potential signaling switch for osteoclast formation and apoptosis.

Previous studies have suggested an indirect effect of TG on osteoclast formation through its action on osteoblasts.(36–38) These authors reported that TG and other substances that elevate intracellular Ca2+, such as ionomycin and cyclopiazonic acid, induce osteoclast formation in mouse bone marrow cells co-cultured with primary osteoblasts by stimulating the gene expression of RANKL and to a lesser extent OPG.(36, 37) In addition, TG and the calcium ionophore A23187 both inhibit OPG promoter expression in rat osteoblast-like UMR106 cells.(38) In our study, the use of the RAW264.7 cell system has an advantage because of the absence of osteoblasts and M-CSF, so that direct effects of TG on RANKL-induced osteoclastogenesis and signaling can be distinguished. Several other groups have used RAW264.7 cells as a convenient model system to study osteoclastogenesis, signaling transduction pathways, and bone resorption.(19, 24, 39, 40) We have also confirmed these results using primary cultured cells.

The effect of TG-induced changes in intracellular levels of calcium on osteoclast function and survival has been previously reported. Activation of osteoclast plasma membrane Ca2+ receptors results in the cytosolic release of Ca2+ from intracellular storage organelles. The refilling of such stores depends on a TG-sensitive Ca2+ ATPase, and store depletion induces capacitative Ca2+ influx.(41) Further more, protein tyrosine kinase inhibitors increase cytosolic calcium through activation of a dihydropyridine-insensitive, nonspecific Ca2+ entry pathway that disrupts the formation of actin rings, resulting in suppression of bone resorption activity.(42) More recently Mentaverri et al.(43) reported that, in calcium-free medium, TG decreased both osteoclast survival and bone resorption processes. However, in medium containing 20 mM calcium, TG-treated osteoclasts showed increased survival and capacity to resorb bone, suggesting that increasing extracellular calcium concentration stimulates osteoclast survival when the filling of intracellular stores is prevented. These results are in accordance with our observations on the effect of TG on apoptosis, caspase-3 activation, and intracellular calcium elevation when cells were exposed to high concentrations of TG.

NF-κB activation is central to osteoclastogenesis,(26–28) and suppression of NF-κB activity by parthenolide inhibits osteoclastogenesis and bone resorption.(21) We studied how TG modulates RANKL-induced osteoclastogenesis by measuring its effects on activity in the NF-κB pathway. We found that TG alone has positive effect on the constitutive NF-κB activity in RAW264.7 cells and significantly potentiates the RANKL-induced activation of NF-κB. This is accompanied by the production of ROS, consistent with the notion that cytokine-induced activation of NF-κB has been associated with the production of ROS.(29, 30) These provide mechanistic evidence that TG enhances osteoclastogenesis caused by stimulation of RANKL-induced activation of NF-κB and production of ROS. A recent study has shown that ROS stimulates osteoclastogenesis, whereas the antioxidant, N-acetyl cysteine, prevented osteoclast formation and NF-κB activation.(44) Therefore, the co-stimulatory effect of RANKL and TG on the production of ROS might contribute to the potentiation of NF-κB activation and osteoclastogenesis.

The RANK/TRAF6/JNK/AP-1 signaling pathway is required for efficient osteoclastogenesis.(31–35, 45) Recent studies have shown that inhibition of RANKL-induced JNK activation abolished the RANKL-induced apoptosis,(12) suggesting that the JNK pathway is also involved in apoptosis. We have shown that TG alone has little effect on the activation of AP-1, but suppresses the RANKL-induced AP-1 activity in RAW264.7 cells, revealing another potential link between TG-mediated Ca2+ signaling and osteoclast formation and apoptosis through the JNK/AP-1-mediated pathway.

Given the extent of cross-talk between Ca2+ signaling and other intracellular signaling pathways, other important links might exist to associate TG action with osteoclastogenesis. RANKL has been shown to activate Akt, p38, and ERK kinase pathways and to upregulate gene expression of NF-AT and Jun dimerization protein 2, with evidence that these are also involved in osteoclast differentiation.(11, 40, 46, 47) In addition, modulation of protein kinase C (PKC) activity by TPA has been shown to affect RANKL-induced osteoclastogenesis and the NF-κB signaling pathway.(19) In this study, we have shown that TG has little effect on ERK signaling; however, the effects of TG on other RANKL signaling and transcriptional programs, such as PKC pathways, remain to be determined.

In conclusion, this study showed that TG directly triggers osteoclast apoptosis and modulates osteoclastogenesis in a dose-dependent manner. TG-induced osteoclast apoptosis correlated with activation of the caspase-3 pathway and elevation of intracellular Ca2+ concentration and effects on osteoclastogenesis correlated with effects on RANKL-induced activation of NF-κB and ROS production. The overall effect of TG in osteoclast formation and apoptosis therefore may depend on the amplitude and duration of the TG-mediated Ca2+ signal as well as its cross-talk with RANKL signaling pathways.

Acknowledgements

This work was supported in part by the National Health and Medical Research Council of Australia, the Arthritis Foundation of Western Australia, and a small research grant by the University of Western Australia. KY is a recipient of a Western Australia Medical Research Institute Postgraduate Research Scholarship. Flow cytometry analysis was done in the Lotteries Flow Cytometry Unit, School of Biomedical and Chemical Sciences, The University of Western Australia. Confocal microscope experiments were done in the Biomedical Confocal Microscopy Research Centre at the Pharmacology Unit, School of Medicine and Pharmacology, the University of Western Australia.

Ancillary