of January 5, 2019.This information is current as

Osteoclast Differentiation and FunctionMultifunctional Regulator in Murine Osteoclast Differentiation Factor Acts as a

Tatsuji Nishihara, Naoyuki Takahashi and Tatsuo SudaOkahashi, Kanichiro Kobayashi, Nobuyuki Udagawa, Eijiro Jimi, Shuichi Akiyama, Taro Tsurukai, Nobuo

http://www.jimmunol.org/content/163/1/4341999; 163:434-442; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/163/1/434.full#ref-list-1

, 22 of which you can access for free at: cites 49 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 1999 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on January 5, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on January 5, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

Osteoclast Differentiation Factor Acts as a MultifunctionalRegulator in Murine Osteoclast Differentiation and Function1

Eijiro Jimi,* Shuichi Akiyama,* Taro Tsurukai,* Nobuo Okahashi, † Kanichiro Kobayashi,*Nobuyuki Udagawa,* Tatsuji Nishihara,† Naoyuki Takahashi,* and Tatsuo Suda2*

Osteoclast differentiation factor (ODF), a novel member of the TNF ligand family, is expressed as a membrane-associated proteinby osteoblasts/stromal cells. The soluble form of ODF (sODF) induces the differentiation of osteoclast precursors into osteoclastsin the presence of M-CSF. Here, the effects of sODF on the survival, multinucleation, and pit-forming activity of murine osteoclastswere examined in comparison with those of M-CSF and IL-1. Osteoclast-like cells (OCLs) formed in cocultures of murineosteoblasts and bone marrow cells expressed mRNA of RANK (receptor activator of NF-kB), a receptor of ODF. The survival ofOCLs was enhanced by the addition of each of sODF, M-CSF, and IL-1. sODF, as well as IL-1, activated NF-kB and c-JunN-terminal protein kinase (JNK) in OCLs. Like M-CSF and IL-1, sODF stimulated the survival and multinucleation of prefusionosteoclasts (pOCs) isolated from the coculture. When pOCs were cultured on dentine slices, resorption pits were formed on theslices in the presence of either sODF or IL-1 but not in that of M-CSF. A soluble form of RANK as well as osteoprotegerin/osteoclastogenesis inhibitory factor, a decoy receptor of ODF, blocked OCL formation and prevented the survival, multinucle-ation, and pit-forming activity of pOCs induced by sODF. These results suggest that ODF regulates not only osteoclast differen-tiation but also osteoclast function in mice through the receptor RANK. The Journal of Immunology,1999, 163: 434–442.

Osteoclastic bone resorption consists of multiple stepssuch as the differentiation of osteoclast precursors intomononuclear prefusion osteoclasts (pOC),3 the fusion of

pOCs to form multinucleated osteoclasts, and the activation ofosteoclasts to resorb bone (1–3). Multinucleated cells (MNC) withosteoclast characteristics, including tartrate-resistant acid phospha-tase (TRAP) activity, calcitonin receptors, and pit-forming activityon dentine slices are formed in a coculture of murine hemopoieticcells and osteoblasts/stromal cells in the presence of osteotropicfactors such as 1a,25-dihydroxyvitamin D3, parathyroid hormone,and IL-11 (3). The target cells of those osteotropic factors areosteoblasts/stromal cells but not osteoclast progenitors in the in-duction of osteoclasts in vitro (3). From these results, we proposeda hypothesis that osteoblasts/stromal cells express a common fac-tor, osteoclast differentiation factor (ODF), in response to thoseosteotropic factors. ODF appeared a membrane-associated factor,since cell-to-cell contact between osteoclast precursors and osteo-blasts/stromal cells was prerequisite for osteoclast formation.

Thus, osteoclast precursors of the monocyte-macrophage lineagerecognize ODF through cell-to-cell interaction with osteoblasts/stromal cells, then differentiate into osteoclasts (3).

Osteoblasts/stromal cells also play an essential role in inducingosteoclast function (osteoclast activation) (4). When purified oste-oclast-like cells (OCLs) formed in vitro were cultured on dentineslices, they failed to form resorption pits (4). Osteoblasts/stromalcells, simultaneously added, greatly enhanced the pit-forming ac-tivity of OCLs through a mechanism involving cell-to-cell contactwith OCLs (4). OCLs rapidly died via spontaneously occurringapoptosis in the absence of osteoblasts/stromal cells. The cytokinesIL-1 and M-CSF were shown to prolong the survival of purifiedOCLs (5). The activation of NF-kB was involved in the IL-1-induced survival of OCLs (6). Using pOCs obtained from theechistatin-treated coculture of murine osteoblastic cells and bonemarrow cells, we showed that both IL-1 and M-CSF prolonged thesurvival and induced the multinucleation of pOCs, but only IL-1induced the pit-forming activity of pOCs, even in the absence ofosteoblasts/stromal cells (7). These results indicate that some fac-tors can be replaced with osteoblasts/stromal cells in the inductionof the survival, multinucleation, and activation of osteoclasts.

The identical proteins osteoprotegerin (OPG) and osteoclasto-genesis inhibitory factor (OCIF), which inhibit osteoclast devel-opment in vitro and in vivo, have recently been cloned indepen-dently (8, 9). OPG/OCIF is a member of the TNF receptor family,but it does not have a transmembrane domain, suggesting thatOPG/OCIF functions as a circulating factor. Subsequently, thecDNA encoding the binding molecule of OPG/OCIF was isolatedfrom an expression library of the murine stromal cell line ST2,which supports OCL formation in coculture with hemopoietic cells(10). The binding molecule of OPG/OCIF was a membrane-asso-ciated protein of the TNF ligand family. This molecule satisfied allof the criteria of ODF and was thus renamed ODF. ODF was alsofound to be identical to TNF-related activation-induced cytokine(TRANCE) and receptor activator of NF-kB ligand (RANKL),which were independently cloned from murine T cell hybridomas

*Department of Biochemistry, School of Dentistry, Showa University, Tokyo, Japan;and †Department of Oral Science, National Institute of Infectious Diseases, Tokyo,Japan

Received for publication January 20, 1999. Accepted for publication April 16, 1999.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby markedadvertisementin accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported in part by Grants-in-Aid 09771546 and 08557101 from theMinistry of Education, Science and Culture of Japan.2 Address correspondence and reprint requests to Dr. Tatsuo Suda, Department ofBiochemistry, School of Dentistry, Showa University 1–5-8 Hatanodai, Shinagawa-ku, Tokyo 142, Japan. E-mail address: suda@dent.showa-u.ac.jp3 Abbreviations used in this paper: pOC, prefusion osteoclast-like cell; TRAP, tar-trate-resistant acid phosphatase; ODF, osteoclast differentiation factor; OCL, oste-oclast-like cell; OPG, osteoprotegerin; OCIF, osteoclastogenesis inhibitory factor;TRANCE, TNF-related activation-induced cytokine; RANK, receptor activator ofNF-kB; L, ligand; sODF, soluble form of ODF; JNK, c-Jun N-terminal protein kinase;sRANK, soluble form of RANK; TRAF, TNF receptor-associated factor; DN, dom-inant negative; MNC, multinucleated cell; IL-1ra, IL-1 receptor antagonist; SAPK,stress-activated protein kinase; ERK, extracellular signal-regulated kinase; SEK1,SAPK-ERK kinase 1.

Copyright © 1999 by The American Association of Immunologists 0022-1767/99/$02.00

by guest on January 5, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

and murine dendritic cells, respectively (11, 12). Lacey et al. (13)also succeeded in the molecular cloning of a ligand for OPG froman expression library of the murine myelomonocytic cell line 32D.The OPG ligand (OPGL) was identical to ODF (TRANCE/RANKL). The administration of OPGL (ODF) to mice caused re-duced bone volume and extreme hypercalcemia without a signif-icant increase in the number of osteoclasts, suggesting that ODF isinvolved in the activation of osteoclasts as well (13). Fuller at al.(14) also recently reported that TRANCE (ODF) is involved in theosteoclast activation induced by osteoblastic cells treated withparathyroid hormone. A soluble form of TRANCE induced a strik-ing change in the motility and spreading of isolated rat osteoclasts,and inhibited their apoptosis (14). These results suggest that ODFis necessary for the osteoblast-mediated activation of matureosteoclasts.

The utilization of a soluble form of ODF (sODF) has allowed usto elucidate the role of ODF in osteoclast function in more detail.In the present study, we examined the effects of ODF on the sur-vival, multinucleation, and activation of osteoclasts in comparisonwith those of M-CSF and IL-1. sODF, M-CSF, and IL-1 promotedthe survival and multinucleation of pOCs through their respectivereceptors. sODF and IL-1, but not M-CSF, stimulated the pit-form-ing activity of OCLs. sODF, as well as IL-1, activated NF-kB andc-Jun N-terminal protein kinase (JNK) in OCLs. Not only OCIFbut also a soluble form of RANK inhibited all of the events in-duced by sODF. These results indicate that ODF expressed byosteoblasts/stromal cells as a membrane-associated protein is re-sponsible for inducing not only osteoclast differentiation but alsoosteoclast function.

Materials and MethodsAbs and chemicals

A recombinant murine sODF purified by affinity chromatography on anOPG/OCIF-immobilized column and by gel filtration chromatography waskindly provided by Snow Brand Milk Products (Tochigi, Japan). The purityof sODF in this preparation was.95% in SDS polyacrylamide gel elec-trophoresis. Recombinant human IL-1a, recombinant human M-CSF, andmurine IL-1 receptor antagonist (IL-1ra) were obtained from R&D Sys-tems (Minneapolis, MN). Echistatin was purchased from Sigma (St. Louis,MO). Anti-human IkBa rabbit polyclonal Abs were purchased from NewEngland BioLabs (Lake Placid, NY). Anti-human Bcl-2, Bcl-xL, and RelA(p65) rabbit polyclonal Abs were purchased from Santa Cruz Biotechnol-ogy (Santa Cruz, CA). Anti-c-Fms mAb was kindly provided by Dr. S-I.Nishikawa (Kyoto University, Kyoto, Japan). Anti-humanb-actin mAbwas obtained from Boehringer Mannheim Biochemica (Mannheim,Germany).

Coculture system and enrichment of osteoclast-like cells

Osteoblasts obtained from the calvariae of newborn mice and bone marrowcells obtained from the tibiae of male mice were cocultured inaMEM (LifeTechnologies, Grand Island, NY) containing 10% FBS, 1a,25-dihy-droxyvitamin D3 (1028 M) (Wako Pure Chemical, Osaka, Japan) and PGE2

(1026 M) (Sigma) in 100-mm-diameter dishes coated with collagen gels(Nitta Gelatin, Osaka). OCLs were formed within 6 days in culture andwere removed from the dishes by treating with 0.2% collagenase (Wako).The purity of OCLs in this fraction (crude OCL preparation) was about 5%.To further purify the OCLs, the crude OCL preparation was replated onculture dishes. After culture for 8 h, osteoblasts were removed with PBScontaining 0.001% pronase E (Calbiochem, La Jolla, CA) and 0.02%EDTA according to the method described previously (15).

Mononuclear and binuclear pOCs were prepared as described previ-ously (16) with a slight modification. Cocultures of bone marrow cells anda murine calvaria-derived osteoblastic cell line, KS4 (17), were maintainedin 100-mm-diameter dishes for 6 days as described above. The KS4 cellswere removed first from the coculture, using a mixture of collagenase-dispase (Boehringer Mannheim), followed by washing three times with0.1% BSA inaMEM. pOCs were then released from the dish with 30 nMechistatin. More than 90% of the cells in the pOC preparation used in thepresent study were positive for TRAP.

Northern blot analysis

Total RNA was extracted from murine spleen cells, bone marrow cells,primary osteoblasts, purified OCLs, murine myoblastic C2C12 cells, andmurine bone marrow-derived stromal ST2 cells using Trizol solution (LifeTechnologies). The total RNA (10mg) was electrophoresed in 1.0% aga-rose-formaldehyde gels, and the RNA was transferred on nylon membranefilters (Hybond-N, Amersham International, Little Chalfont, U.K.). Themembranes were hybridized for 15 h at 42°C with radioactive cDNAprobes for murine RANK and murine TNF type I and type II receptors,which were cloned by RT-PCR and labeled using a multirandom primeroligonucleotide labeling kit (Takara Shuzo, Osaka, Japan). As an internalcontrol, the membrane was rehybridized with a radioactive cDNA probefor mouse GAPDH. Each membrane was then exposed to an x-ray film.

EMSA and JNK assay

For the EMSA, nuclear extracts were prepared according to the methoddescribed by Dignam et al. (18). The sequence of the NF-kB-binding oli-gonucleotide used as a radioactive DNA probe was 59-AGCTTGGGGACTTTCCGAG-39. The DNA binding reaction was performed at roomtemperature in a volume of 20ml, which contained the binding buffer (10mM Tris/HCl (pH 7.5), 1 mM EDTA, 4% glycerol, 100 mM NaCl, 5 mMDTT, 100 mg/ml BSA), 3mg of poly(dI-dC), 13 105 cpm of a32P-labeledprobe, and 8mg of nuclear proteins. After incubation for 15 min, the sam-ples were electrophoresed on native 5% acrylamide/0.253 TBE gels. Thegels were dried and exposed to an x-ray film. For the determination of JNKactivity, purified OCLs treated with sODF or IL-1 were washed twice withice-cold PBS, then lysed in a lysis buffer (20 mM Tris-HCl (pH 7.5), 150mM NaCl, 1 mM EDTA, 1 mM DGTA, 1% Triton X-100, 2.5 mM sodiumpyrophosphate, 1 mMb-glycerophosphate, 1 mM Na3VO4, 1 mg/ml leu-peptin, 1 mM PMSF). The JNK activity of the cell lysates was determinedusing a stress-activated protein kinase (SAPK)/JNK kinase assay kit (NewEngland BioLabs).

Survival and cell fusion assays

The survival rate of OCLs was measured as reported previously (5, 19).After OCLs were purified, some of the cultures were subjected to TRAPstaining. TRAP-positive MNCs containing more than 3 nuclei werecounted as living OCLs. Other cultures were further incubated in the pres-ence or absence of sODF. After incubation for indicated periods, the re-maining OCLs were counted. To examine the effect of sODF on the fusionof osteoclasts, we replated pOCs (15,000 cells/well) on 96-well cultureplates with or without various increasing concentrations of sODF. Afterculture for 18 h, the cells were fixed and stained for TRAP. Some cultureswere also treated with IL-1ra (1mg/ml), anti-c-Fms Ab (10mg/ml), orOCIF (100 ng/ml). The number of TRAP-positive MNCs with more than10 nuclei was counted as pOC-derived OCLs. Actin rings in pOC-derivedOCLs were also visualized by rhodamine-conjugated phalloidin staining,as previously described (15). Results are expressed as the means6 SEMof three cultures.

Immunoblotting analysis and immunofluorescence microscopy

After purified OCL preparations were cultured for various periods in thepresence of ODF, the cells were washed twice with ice-cold PBS and thenlysed in a lysis buffer (20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mMEDTA, 0.5% Triton X-100, 10mg/ml aprotinin, 20mg/ml leupeptin, and 1mM PMSF). The cell lysates (20mg of protein) were resolved by 10%SDS-PAGE and transferred onto polyvinylidene difluoride (PVDF) mem-branes (Millipore, Bedford, MA). After blocking with 2% BSA in TBST,the IkBa, Bcl-2 and Bcl-xL Abs (1/1000 dilution) were added in TBSTcontaining 2% BSA and visualized by an enhanced chemiluminescenceassay using reagents from Amersham and exposed to an x-ray film. Theimmunofluorescence analysis was done as described previously (6). Anti-RelA (p65) Abs (1mg/ml) was used for the immunostaining.

Pit formation assay

pOC preparations (15,000 cells/0.1 ml/well) were seeded on dentine slices(4-mm diameter) which had been placed in 96-well plates. After incubationfor 2 h, dentine slices were transferred to 48-well plates (one slice/well) inthe presence or absence of ODF. Pit formation by pOCs was determinedafter culture for 24 h. For the pit formation assay, cells were removed fromdentine slices, and the resorbed area was stained with Mayer’s hematoxylin(15). The numbers of pits on the slices were counted.

Expression and purification of soluble RANK

A FLAG-tagged soluble form of RANK (sRANK) was generated by clon-ing a PCR product encoding the RANK ectodomain (amino acids 1–213)

435The Journal of Immunology

by guest on January 5, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

into theEcoRI site of an expression vector that carries the promoter regionof human EF1a gene (pEF-BOS) (20). COS 7 cells were transfected withthe expression vectors (16.6mg/100-mm-diameter dish) by cationic lipo-somes (DMRIE-C, Life Technologies) according to the manufacture’s rec-ommendation. The supernatant was harvested 48 h later, passed through a0.45-mm filter, incubated with anti-FLAG M2 affinity gel (Kodak, NewHaven, CT), and eluted with FLAG peptide (250mg/ml, Kodak) as out-lined in the manufacturer’s protocol. The eluant was dialyzed against PBS,

and the protein concentration was determined using a bicinchoninic acid(BCA) protein assay kit (Pierce, Rockford, IL).

ResultsRANK, a novel TNF receptor family member, has been shown tobe a signal-transducing receptor of ODF (12, 21). Total RNA wasobtained from spleen cells, bone marrow cells, purified OCLs, pri-mary osteoblasts, myoblastic C2C12 cells, and bone marrow-de-rived stromal ST2 cells, and the expression of RANK mRNA wasexamined in the RNA preparations (Fig. 1). Few contaminatingosteoblasts (alkaline phosphatase-positive cells) or monocyte-mac-rophages (F4/80 Ag-positive cells) were detected in the purifiedOCL preparations in this study (data not shown). The expression ofother TNF receptor family members, such as TNFRI and TNFRII,was also examined in the same membrane. TNFRI mRNA wassimilarly expressed in most of the cells examined in this study.TNFRII mRNA was highly expressed by OCLs, and the expres-sion level by the other cells was much lower (Fig. 1). OCLs alsohighly expressed RANK mRNA (Fig. 1).

The overexpression of RANK was sufficient to activate NF-kB(12). The ligand-dependent NF-kB activation was also demon-strated by the cotransfection of RANKL with RANK in human 293cells and T cells (12). We then examined whether sODF activatesNF-kB in OCLs. sODF activated NF-kB in OCLs in a dose-de-pendent manner (Fig. 2A). Fig. 2B shows the time course of changein the activation of NF-kB and the levels of IkBa in OCLs afterstimulation with sODF. sODF transiently activated NF-kB in

FIGURE 1. Expression of RANK mRNA by OCLs. Total RNA (10 mg)extracted from spleen cells (lane 1), bone marrow cells (lane 2), purifiedOCLs (lane 3), C2C12 cells (lane 4), ST2 cells (lane 5), and primaryosteoblasts (lane 6) was separated in a 1.0% formaldehyde-agarose gel andhybridized with the cDNA probes for RANK, TNF type I receptor, TNFtype II receptor, and GAPDH as described underMaterials and Methods.

FIGURE 2. sODF activates NF-kB in purified OCLs.A, Purified OCLs were treated with increasing concentrations of sODF for 30 min.B, Time courseof changes in the activation of NF-kB and levels of IkBa in OCLs after stimulation with sODF. Purified OCLs were treated with sODF (100 ng/ml) forthe indicated periods.C, Effects of OCIF on the activation of NF-kB in OCLs treated with sODF. sODF (100 ng/ml) (lanes 2and3) or IL-1 (10 ng/ml)(lanes 4and5) was incubated with OCIF (100 ng/ml) for 1 h at37°C. Purified OCLs were treated with sODF or IL-1 in the absence (lanes 2and4) orpresence (lanes 3and5) of OCIF for 30 min. NF-kB activity in the nuclear extracts was determined by an EMSA, and the amount of IkBa was determinedby immunoblotting, as described underMaterials and Methods. D, Translocation of RelA (p65) into the nuclei of OCLs in response to sODF. Purified OCLswere treated with sODF (100 ng/ml) for 30 min. Cells were then fixed and incubated with Abs against RelA (p65), followed by FITC-conjugated anti-rabbitimmunoglobulins. The subcellular localization of FITC-labeled RelA (P65) was observed by fluorescence microscopy.

436 REGULATION OF OSTEOCLAST FUNCTION BY ODF

by guest on January 5, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

OCLs, and the maximal activation occurred at 30 min. The deg-radation of IkBa coincided with the activation of NF-kB (Fig. 2B).OCIF, a decoy receptor of ODF, prevented the sODF-inducedNF-kB activation, but it did not affect at all the IL-1-inducedNF-kB activation in the OCLs (Fig. 2C). The immunocytochem-ical analysis revealed that RelA (p65) was translocated from thecytoplasm into almost all the nuclei of the OCLs within 30 minafter the sODF stimulation (Fig. 2D).

JNK is a signal transducer commonly activated by TNF-relatedligands (22). TRANCE (ODF) has been shown to activate JNK inT cells and in bone marrow-derived dendritic cells (11). We ex-amined whether sODF activates JNK in mature OCLs. JNK wasrapidly activated in OCLs after the stimulation with sODF, in adose-dependent manner (Fig. 3,A andB). The activation of JNKwas also induced in OCLs treated with 10 ng/ml of IL-1 (Fig. 3C).OCIF blocked the JNK activation induced by sODF but not thatinduced by IL-1 (Fig. 3,B andC).

We reported that the activation of NF-kB is involved in thesurvival of OCLs promoted by IL-1 (6). Like IL-1, sODF pro-longed the survival of OCLs in a dose-dependent manner (Fig.4A). Addition of OCIF with sODF prevented the sODF-inducedsurvival of OCLs. To the contrary, OCIF could not prevent theIL-1- or M-CSF-induced survival of OCLs. OCIF did not accel-erate spontaneous cell death of OCLs (Fig. 4A). TRANCE (ODF)

has been shown to induce the survival of dendritic cells, which wasmediated by up-regulation of Bcl-xL (23). To determine whethersODF induces Bcl-2 or Bcl-xL, we measured their expression inOCLs treated with sODF. A Western blot analysis showed that theexpression of Bcl-2 and Bcl-xL in OCLs was not affected by sODF(Fig. 4B).

Wesolowski et al. (16) developed a method for obtaining highlypurified (.90%) TRAP-positive mononuclear or binuclear pOCsreleased by the “disintegrin” echistatin from cocultures of murineosteoblastic cells (MB1.8 cells) with murine bone marrow cells.We reported that IL-1 and M-CSF both prolonged the survival andinduced the multinucleation of pOCs in the absence of osteoblasts/stromal cells (7). A small number of TRAP-positive MNCs (pOC-derived OCLs) were formed within 2–5 h in the control cultures(7), but they disappeared by 18 h of the incubation period (Fig.5A). To determine whether the effects of sODF, IL-1, and M-CSFon the survival and multinucleation of pOCs are mediated throughtheir respective receptors, OCIF, IL-1ra, and Ab against c-Fms(M-CSF receptor) were added to pOC cultures (Fig. 5B). IL-1rainhibited only the IL-1-induced multinucleation of pOCs, and anti-c-Fms Ab inhibited only M-CSF-induced multinucleation of pOCs(Fig. 5B). The survival and multinucleation of pOCs induced bysODF were inhibited by adding OCIF but not by adding IL-1ra oranti-c-Fms Ab (Fig. 5B). The formation of ringed structures of

FIGURE 3. sODF activates JNK inpurified OCLs.A, Purified OCLs weretreated with increasing concentrationsof sODF for 30 min.B, Purified OCLswere treated with sODF (100 ng/ml)for the indicated periods in the absence(lanes 1–4) or presence (lanes 5–8) ofOCIF (100 ng/ml).C, Purified OCLswere treated with IL-1 in the presence(lane 4) or absence (lane 3) of OCIFfor 30 min. JNK activities were mea-sured using a SAPK/JNK assay kit asdescribed underMaterials and Methods.

FIGURE 4. sODF prolongs the survival ofOCLs. A, Purified OCLs were incubated for24 h with sODF (100 ng/ml, 1mg/ml), IL-1 (10ng/ml), or M-CSF (100 ng/ml) in the presence(hatched columns) or absence (open columns)of OCIF (100 ng/ml). TRAP-positive OCLscontaining more than three nuclei were thencounted after TRAP staining. Similar resultswere obtained from three independent experi-ments.p, Significantly different from culturestreated with ODF alone;p , 0.01.B, Westernblot analysis of Bcl-2 and Bcl-xL in purifiedOCLs. Purified OCLs were cultured in the ab-sence (lanes 2and4) or presence (lanes 3and5) of sODF (100 ng) for 6 h (lanes 2and3) or18 h (lanes 4and5). The murine leukemia cellline M1 (lane 6) was used as a positive control.

437The Journal of Immunology

by guest on January 5, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

F-actin dots (actin rings) in osteoclasts is closely related to theosteoclast function (24). Actin rings were distributed at the periph-ery of the sODF-induced OCLs with smooth contours (Fig. 5C).This suggests that the smooth contour of pOC-derived OCLs issomehow related to the sODF-induced pit-forming activity.

pOCs were then cultured for 24 h on dentine slices in the pres-ence or absence of sODF (Fig. 6). The number of pOC-derivedOCLs remaining on the slices in the cultures treated with sODFwas much greater than that in the control cultures (data not shown).Resorption pits on dentine slices were observed in the culturetreated with sODF (Fig. 6A). The pit-forming activity of pOCsinduced by sODF was inhibited by adding OCIF (Fig. 6B). M-CSFalso prolonged the survival and induced the multinucleation ofpOCs, but no resorption pits were formed in the cultures treatedwith M-CSF (7).

We then examined the effect of interactions between ODF andRANK on the activation of NF-kB and JNK by adding sRANK topurified OCLs. Addition of sRANK significantly reduced the num-ber of OCLs in a dose-dependent manner (Fig. 7). Proliferation ofstromal cells in bone marrow cultures appeared to not be affectedby the addition of sRANK or OCIF. Addition of sRANK as wellas OCIF reduced the activity of NF-kB and JNK (Fig. 7B). More-over, the sODF-mediated stimulation of the survival, multinucle-ation, and pit-forming activity of pOCs was also abrogated by theaddition of sRANK (Table I).

DiscussionWe previously reported that both IL-1 and M-CSF induced thesurvival and multinucleation of OCLs/pOCs, but only IL-1 stim-ulated their pit-forming activity (7). The present study clearlyshowed that, like IL-1, sODF stimulated not only the survival andmultinucleation of OCLs/pOCs but also their pit-forming activity.Our results are consistent with the findings by Fuller et al. (14),who demonstrated that TRANCE (ODF) is involved in the oste-oclast activation induced by osteoblastic UMR 106 cells treatedwith parathyroid hormone. It was also shown that OPGL (ODF)administered into mice induced osteoclastic bone resorption with-out a significant increase in the number of osteoclasts (13). Theseresults suggest that ODF is a factor essentially involved in not onlyosteoclast differentiation but also osteoclast activation.

The survival and multinucleation of pOCs induced by M-CSF,IL-1, and sODF were inhibited by anti-c-Fms (M-CSF receptor)Ab, IL-1ra (antagonist of IL-1 type 1 receptor), and OPG/OCIF (adecoy receptor of ODF), respectively (Fig. 5). OPG/OCIF specif-ically inhibited activation of NF-kB and JNK in OCLs induced bysODF but not that induced by IL-1 (Figs. 2 and 3). These resultssuggest that M-CSF, IL-1, and ODF have similar effects on thesurvival and multinucleation of osteoclasts via respective recep-tors. A soluble form of TNFR has been shown to inhibit the ac-tivity of TNF-a and TNF-b (25–28). Here, sRANK inhibited the

FIGURE 5. sODF induces the multinucleation of pOCs.A, Purified pOCs were incubated with or without sODF (100 ng/ml) for 18 h. The cells werethen stained for TRAP.B, Effects of IL-1ra, anti-c-Fms Ab, and OCIF on the IL-1-, M-CSF-, and sODF-induced multinucleation of pOCs. pOCs werepretreated for 30 min with or without IL-1ra (1mg/ml), anti-c-Fms Ab (10mg/ml), or OCIF (100 ng/ml) and further incubated for 18 h with IL-1 (10 ng/ml)(open columns), M-CSF (100 ng/ml) (hatched columns), or sODF (100 ng/ml) (closed columns). TRAP-positive MNCs with more than 10 nuclei werecounted as pOC-derived OCLs. The results shown are the means6 SEM of three independent experiments.p, Significantly different from the controlcultures;p , 0.001.C, Actin ring formation in pOC-derived OCLs induced by sODF. pOCs were incubated with sODF (100 ng/ml) for 18 h, and thencells were fixed and stained for TRAP. F-actin was detected with rhodamine-conjugated phalloidin. Bar5 100 mm.

438 REGULATION OF OSTEOCLAST FUNCTION BY ODF

by guest on January 5, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

OCL formation in bone marrow cultures treated with sODF to-gether with M-CSF and blocked the sODF-induced activation ofNF-kB and JNK in OCLs (Fig. 7). Nakagawa et al. (21) recentlyreported that polyclonal Abs against the extracellular domains ofRANK induced OCL formation in spleen cell cultures in the pres-ence of M-CSF. This indicates that the clustering of RANK isrequired for the RANK-mediated signal transduction for osteoclas-togenesis. In contrast, the Fab fragment of anti-RANK Abs com-pletely inhibited ODF-mediated osteoclastogenesis (21). These re-sults suggest that RANK is the sole receptor of ODF responsiblefor inducing differentiation and activation of osteoclasts.

We have reported that M-CSF is indispensable for both the pro-liferative phase and the differentiation phase of osteoclast devel-opment (29). ODF and M-CSF cannot be replaced by other cyto-kines in inducing osteoclast differentiation. The present studyshows that M-CSF and IL-1, as well as ODF, prolonged the sur-vival of OCLs and induced their fusion. ODF and IL-1 but notM-CSF induced the pit-forming activity of purified OCLs/pOCs inculture. These results suggest that ODF is the sole factor for in-ducing osteoclast differentiation, but factors other than ODF arealso able to support the survival, fusion, and activation of matureosteoclasts. Fig. 8 summarizes the role of cytokines examined inthis study in the regulation of osteoclast differentiation andfunction.

sODF activated NF-kB in OCLs, which coincided with the deg-radation of IkBa (Fig. 2B). Our immunocytochemical analysis re-vealed that RelA (p65) was translocated from the cytoplasm intoalmost all of the nuclei of OCLs within 30 min after sODF stim-ulation (Fig. 2D). This action of ODF was quite similar to that ofIL-1 previously reported (6, 7). Wong et al. (11) reported that asoluble form of TRANCE (ODF) induced JNK activation in Tcells but not in splenic B cells and bone marrow-derived dendritic

cells. sODF as well as IL-1-activated JNK in OCLs (Fig. 3). Incontrast, M-CSF activated JNK but not NF-kB in OCLs, though itsupported the survival and fusion of OCLs/pOCs (our unpublishedobservation). These results suggest that the survival and fusion ofosteoclasts are not sufficient for inducing osteoclast function.

TRANCE (ODF) has been shown to induce the survival of den-dritic cells through the up-regulation of Bcl-xL (23). It was alsoreported that targeting of both Bcl-xL and SV40 large T Ag to cellsof the osteoclast lineage immortalized osteoclast precursors (30).These results suggest that antiapoptotic proteins such as Bcl-xL

and Bcl-2 are involved in the survival of osteoclasts. However,neither Bcl-2 nor Bcl-xL in OCLs was up-regulated by sODF inour culture condition. IL-1 failed to induce the expression of Bcl-2and Bcl-xL in OCLs (31). This suggests that ODF supports thesurvival of osteoclasts through a mechanism different from theup-regulation of Bcl-2 and Bcl-xL.

The activation of NF-kB has been shown to increase cellularresistance to apoptosis (32–36). Using antisense oligodeoxynucle-otides to NF-kB (RelA/p65 and p50) and proteasome inhibitorsthat inhibit the degradation of IkB, we have shown that the acti-vation of NF-kB is involved in the survival of OCLs promoted byIL-1 (6). The activation of NF-kB appears to be also involved inthe ODF-induced survival of OCLs/pOCs. The precise role of JNKin apoptosis is controversial. Strong activation of JNK was inducedby apoptosis-inducing stresses such as UV and hydrogen peroxide(37–39). Using the knockout mice of the SEK1 gene, which en-codes a direct upstream kinase of JNK, SEK1-induced signalswere shown to play a protective role against various cytotoxicstimuli (40). TNF receptor-associated factor (TRAF) 2 is a signal-transducing protein of the TNF receptor family (39). The activa-tion of JNK was impaired, but the activation of NF-kB was in-duced in thymocytes obtained from dominant negative (DN)

FIGURE 6. sODF induces the pit-forming activity of pOCs. pOCs were plated on dentine slices for 2 h. The dentine slices were then transferred to48-well plates and incubated for 24 h in the absence or presence of sODF (100 ng/ml) with or without OCIF (100 ng/ml).A, Resorption pits formed ondentine slices were stained with Mayer’s hematoxylin.B, The number of resorption pits was determined. The results shown are the means6 SEM of threeindependent experiments.p, Significantly different from the control cultures;p , 0.001. Bar5 100 mm.

439The Journal of Immunology

by guest on January 5, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

TRAF2-transgenic mice and in embryonic fibroblasts obtainedfrom TRAF2-deficient mice (41, 42). These thymocytes and fibro-blasts were rather apoptotic in the presence of TNF-a. Further-more, thymocytes from IkBa DN and TRAF2 DN double-transgenic mice were more sensitive to TNF-induced apoptosisthan those from normal mice and IkBa DN- or TRAF2 DN-trans-genic mice (43). Therefore, JNK-meditated signals appear to col-laborate with NF-kB in inducing the antiapoptotic action inducedby sODF.

We previously showed that OCLs expressed IL-1 type 1 recep-tors (6). Xu et al. (44) reported that intense signals for IL-1 type Ireceptor mRNA were detected in active osteoclasts in an adjuvantarthritis model in rats, whereas mRNA of IL-1 type II receptor,which serves as a decoy receptor, was expressed preferentially inresting osteoclasts. Bone histological studies of OPG/OCIF knock-out mice revealed that physiological bone resorption was regulatedmainly by ODF and OPG/OCIF (45). IL-1 appears to be involved

in pathological bone resorption, such as that observed in rheuma-toid arthritis and periodontal bone diseases.

Recent studies indicate that the cytoplasmic tail of RANK in-teracts with TRAF1, TRAF2, TRAF3, TRAF5, and TRAF6 (46–49). Mapping of the structural requirements for TRAF/RANK in-teraction revealed that selective TRAF-binding sites clustered intwo distinct domains of the RANK cytoplasmic tail. In particular,TRAF6 interacted with the membrane-proximal domain of the cy-toplasmic tail distinct from binding sites for TRAF1, -2, -3, and -5.When the TRAF6 interaction domain was deleted, RANK-medi-ated NF-kB activation was completely inhibited, and JNK activa-tion was partially inhibited (48). N-terminal truncation of TRAF6(TRAF6 DN) also inhibited RANKL-induced NF-kB activation(48, 49). These results suggest that TRAF6 transduces a signalinvolved in RANK-mediated activation of osteoclast function.

Double knockout mice of p50 (NF-kB1) and p52 (NF-kB2),subunits of NF-kB, showed severe osteopetrosis because of the

FIGURE 7. sRANK prevents both the sODF-induced OCL formation and activation of NF-kB and JNK in OCLs.A, Murine bone marrow cells werecultured with M-CSF (50 ng/ml) for 4 days in the absence or presence of OCIF (50 ng/ml) sRANK (50 ng/ml, 500 ng/ml) or sODF (50 ng/ml) in 96-wellplates. The cells were then fixed and stained for TRAP, and TRAP-positive MNCs containing more than three nuclei were counted as OCLs. The resultsshown are the means6 SEM of three cultures. Significantly different from the culture treated with ODF and M-CSF:p, p , 0.001;pp, p , 0.005.B, Effectsof sRANK on the activation of NF-kB and JNK in OCLs treated with sODF. sODF (100 ng/ml) was preincubated with or without OCIF (50 ng/ml, 100ng/ml) or sRANK (100 ng/ml, 1mg/ml) for 1 h at37°C. Purified OCLs were treated for 30 min with sODF in the absence (lanes 1, 4,and5) or presenceof OCIF (lanes 2, 6,and 7), or sRANK (lanes 3, 8,and 9). NF-kB activity in the nuclear extracts was determined by an EMSA, and JNK assay wasperformed using a SAPK/JNK assay kit as described underMaterials and Methods.

Table I. Effects of sRANK on the survival, multinucleation, and pit-forming activity of osteoclasts promotedby sODFa

Treatment Survival (%)Multinucleation (no. of

pOC-derived OCLs/well)Pit formation (no.

of pits/slice)

Control 20.06 2.8 30.36 5.7 0ODF 71.56 6.4 123.06 2.7 87.56 12.0sODF1 OCIF 29.56 3.5* 41.06 1.4* 0*sODF1 sRANK 35.06 7.1** 62.56 6.4** 26.06 12.7**

a Purified OCLs or pOCs were incubated with sODF (100 ng/ml) in the presence or absence of OCIF (100 ng/ml) or sRANK(1 mg/ml) for 24 h. A survival assay (OCLs), cell fusion assay (pOCs), and pit formation assay (OCLs) were performed asdescribed underMaterials and Methods.Similar results were obtained from two independent experiments. Significantly differentfrom the cultures treated with sODF:p, p , 0.01; pp, p , 0.05.

440 REGULATION OF OSTEOCLAST FUNCTION BY ODF

by guest on January 5, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

impaired osteoclast differentiation (50, 51). The osteopetrotic dis-order was cured by normal bone marrow transplantation. Theseresults indicate that osteoclast progenitors are impaired in the de-ficient mice. sODF activated NF-kB in the target cells includingosteoclasts. This suggests that the ODF-induced activation ofNF-kB in osteoclast progenitors also plays a crucial role in theirdifferentiation into osteoclasts. Further studies are necessary toelucidate the molecular mechanism of the action of ODF in oste-oclast differentiation and function.

AcknowledgmentsWe thank Dr. Takeshi Yanai (Showa University School of Dentistry) forhis technical assistance, and Dr. Hisataka Yasuda (Snow Bland Milk Prod-ucts) for helpful discussion.

References1. Chambers, T. J., P. A. Revell, K. Fuller, and N. A. Athanasou. 1984. Resorption

of bone by isolated rabbit osteoclasts.J. Cell Sci. 66:383.2. Baron, R., L. Neff, D. Louvard, and P. J. Courtoy. 1985. Cell-mediated extra-

cellular acidification and bone resorption: evidence for a low pH in resorbinglacunae and localization of a 100-kD lysosomal membrane protein at the oste-oclast ruffled border.J. Cell Biol. 101:2210.

3. Suda, T., N. Udagawa, and N. Takahashi. 1996. Osteoclast generation. InPrin-ciples of Bone Biology.L. G. Raisz, G. A. Rodan, and J. P. Bilezikian, eds.Academic Press, San Diego, CA, p. 87.

4. Jimi, E., I. Nakamura, H. Amano, Y. Taguchi, T. Tsurukai, M. Tamura,N. Takahashi, and T. Suda. 1996. Osteoclast function is activated by osteoblasticcells through a mechanism involving cell-to-cell contact.Endocrinology137:2187.

5. Jimi, E., T. Shuto, and T. Koga. 1995. Macrophage colony-stimulating factor andinterleukin-1a maintain the survival of osteoclast-like cells.Endocrinology 136:808.

6. Jimi, E., I. Nakamura, T. Ikebe, S. Akiyama, N. Takahashi, and T. Suda. 1998.Activation of NF-kB is involved in the survival of osteoclasts promoted by in-terleukin-1.J. Biol. Chem. 273:8799.

7. Jimi, E., I. Nakamura, L. T. Duong, T. Ikebe, N. Takahashi, G. A. Rodan, andT. Suda. 1999. Interleukin 1 induces multinucleation and bone-resorbing activityof osteoclasts in the absence of osteoblasts/stromal cells.Exp. Cell Res. 247:84.

8. Simonet, W. S., D. L. Lacey, C. R. Dunstan, M. Kelley, M. S. Chang, R. Luthy,H. Q. Nguyen, S. Wooden, L. Bennett, T. Boone, et al. 1997. Osteoprotegerin: anovel secreted protein involved in the regulation of bone density.Cell 89:309.

9. Tsuda, E., M. Goto, S. Mochizuki, K. Yano, F. Kobayashi, T. Morinaga, andK. Higashio. 1997. Isolation of a novel cytokine from human fibroblasts that specif-ically inhibits osteoclastogenesis.Biochem. Biophys. Res. Commun. 234:137.

10. Yasuda, H., N. Shima, N. Nakagawa, K. Yamaguchi, M. Kinosaki, S. Mochizuki,A. Tomoyasu, K. Yano, M. Goto, A. Murakami, et al. 1997. Osteoclast differ-entiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory fac-tor and is identical to TRANCE/RANKL.Proc. Natl. Acad. Sci. USA 95:3597.

11. Wong, B. R., J. Rho, J. Arron, E. Robinson, J. Orlinick, M. Chao, S. Kalachikov,E. Cayani, F. S. Bartlett III, W. N. Frankel, S. Y. Lee, and Y. Choi. 1997.TRANCE is a novel ligand of the tumor necrosis factor receptor family thatactivates c-Jun N-terminal kinase in T cells.J. Biol. Chem. 272:25190.

12. Anderson, D. M., E. Maraskovsky, W. L. Billingsley, W. C. Dougall,M. E. Tometsko, E. R. Roux, M. C. Teepe, R. F. DuBose, D. Cosman, and

L. Galibert. 1997. A homologue of the TNF receptor and its ligand enhance T-cellgrowth and dendritic-cell function.Nature 390:175.

13. Lacey, D. L., E. Timms, H.-L. Tan, M. J. Kelley, C. R. Dunstan, T. Burgess,R. Elliott, A. Colombero, G. Elliott, S. Scully, et al. 1998. Osteoprotegerin ligandis a cytokine that regulates osteoclast differentiation and activation.Cell 93:165.

14. Fuller, K., B. Wong, S. Fox, Y. Choi, and T. J. Chambers. 1998. TRANCE isnecessary and sufficient for osteoblast-mediated activation of bone resorption inosteoclasts.J. Exp. Med. 188:997.

15. Suda, T., E. Jimi, I. Nakamura, and N. Takahashi. 1997. Role of 1a,25-dihy-droxyvitamin D3 in osteoclast differentiation and function.Methods Enzymol.282:223.

16. Wesolowski, G., L. T. Duong, P. T. Lakkakorpi, R. M. Nagy, K. Tezuka,H. Tanaka, G. A. Rodan, and S. B. Rodan. 1995. Isolation and characterizationof highly enriched, prefusion mouse osteoclastic cells.Exp. Cell Res. 219:679 .

17. Yamashita, T., K. Asano, N. Takahashi, T. Akatsu, N. Udagawa, T. Sasaki,T. J. Martin, and T. Suda. 1990. Cloning of an osteoblastic cell line involved inthe formation of osteoclast-like cells.J. Cell Physiol. 145:587.

18. Dignam, J. D., R. M. Lebovitz, and R. G. Roeder. 1983. Accurate transcriptioninitiation by RNA polymerase II in a soluble extract from isolated mammaliannuclei.Nucleic Acids Res. 11:1475.

19. Okahashi, N., I. Nakamura, E. Jimi, M. Koide, T. Suda, and T. Nishihara. 1997.Specific inhibitors of vacuolar H1-ATPase trigger apoptotic cell death of oste-oclasts.J. Bone Miner. Res. 12:1116.

20. Mizushima, S., and S. Nagata. 1990. pEF-BOS, a powerful mammalian expres-sion vector.Nucleic Acids Res. 18:5322.

21. Nakagawa, N., M. Kinosaki, K. Yamaguch, N. Shima, H. Yasuda, K. Yano,T. Morinaga, and K. Higashio. 1998. RANK is the essential signaling receptor forosteoclast differentiation factor in osteoclastogenesis.Biochem. Biophys. Res.Commun. 253:395.

22. Baker, S. J., and E. P. Reddy. 1996. Transducers of life and death: TNF receptorsuperfamily and associated proteins.Oncogene 12:1.

23. Wong, B. R., R. Josien, S. Y. Lee, B. Sauter, H. L. Li, R. M. Steinman, andY. Choi. 1997. TRANCE (tumor necrosis factor [TNF]-related activation-in-duced cytokine), a new TNF family member predominantly expressed in T cells,is a dendritic cell-specific survival factor.J. Exp. Med. 186:2075.

24. Nakamura, I., E. Jimi, L. T. Duong, T. Sasaki, N. Takahashi, G. A. Rodan, andT. Suda. 1998. Tyrosine phosphorylation of p130Casis involved in actin organi-zation in osteoclasts.J. Biol. Chem. 273:11144.

25. Nophar, Y., O. Kemper, C. Brakebusch, H. Englemann, R. Zwang, D. Aderlka,H. Holtmann, and D. Wallach. 1990. Soluble forms of tumor necrosis factorreceptors (TNF-Rs): the cDNA for the type I TNF-R, cloned using amino acidsequence data of its soluble form, encodes both the cell surface and a solubleform of the receptor.EMBO J. 9:3269.

26. Smith, C., L. Lauffer, T. Davis, M. Widner, S. Dower, C. Jacobs, K. Mohler, andR. G. Goodwin. 1991.Cytokines 3:448.

27. Ashkenazi, A., S. A. Marsters, D. J. Capon, S. M. Chamow, I. S. Figari,D. Pennica, D. V. Goeddel, M. A. Palladino, and D. H. Smith. 1991. Protectionagainst endotoxic shock by a tumor necrosis factor receptor immunoadhesin.Proc. Natl. Acad. Sci. USA 88:10535.

28. Peppel, K., D. Crawford, and B. Beutler. 1991. A tumor necrosis factor (TNF)receptor-IgG heavy chain chimeric protein as a bivalent antagonist of TNF ac-tivity. J. Exp. Med. 174:1483.

29. Tanaka, S., N. Takahashi, N. Udagawa, T. Tamura, T. Akatsu, E. R. Stanley,T. Kurokawa, and T. Suda. 1993. Macrophage colony-stimulating factor is in-dispensable for both proliferation and differentiation of osteoclast progenitors.J. Clin. Invest. 91:257.

30. Hentunen, T. A., S. V. Reddy, B. F. Boyce, R. Devlin 9, H. R. Park, H. Chung,K. S. Selander, M. Dallas, N. Kurihara, D. L. Galson, S. R. Goldring, B. A. Koop,J. J. Windle, and G. D. Roodman. 1998. Immortalization of osteoclast precursors

FIGURE 8. The differentiation pathway of osteoclast progenitors into functionally active osteoclasts and the cytokines required for each step of thepathway.

441The Journal of Immunology

by guest on January 5, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

by targeting Bcl-XL and Simian virus 40 large T antigen to the osteoclast lineagein transgenic mice.J. Clin. Invest. 102:88.

31. Okahashi, N., M. Koide, E. Jimi, T. Suda, and T. Nishihara. 1998. Caspases(interleukin-1b-converting enzyme family proteases) are involved in the regula-tion of the survival of osteoclasts.Bone 23:33.

32. Beg, A. A., and D. Baltimore. 1996. An essential role for NF-kB in preventingTNFa-induced cell death.Science 274:782.

33. Wang, C. Y., M. W. Mayo, and A. S. Baldwin, Jr. 1996. TNF- and cancertherapy-induced apoptosis: potentiation by inhibition of NF-kB. Science 274:784.

34. Van Antwerp, D. J., S. J. Martin, T. Kafri, D. R. Green, and I. M. Verma. 1996.Suppression of TNFa-induced apoptosis by NF-kB. Science 274:787.

35. Liu, Z. G., H. Hsu, D. V. Goeddel, and M. Karin. 1996. Dissection of TNFreceptor 1 effector functions: JNK activation is not linked to apoptosis whileNF-kB activation prevents cell death.Cell 87:565.

36. Ozaki, K., H. Takeda, H. Iwahashi, S. Kitano, and S. Hanazawa. 1997. NF-kBinhibitors stimulate apoptosis of rabbit mature osteoclasts and inhibit bone re-sorption by these cells.FEBS Lett. 410:297.

37. Zanke, B. W., K. Boudreau, E. Rubie, E. Winnett, L. A. Tibbles, L. Zon,J. Kyriakis, F. F. Liu, and J. R. Woodgett. 1996. The stress-activated proteinkinase pathway mediates cell death following injury induced bycis-platinum, UVirradiation or heat.Curr. Biol. 6:606.

38. Dickens, M., J. S. Rogers, J. Cavanagh, A. Raitano, Z. Xia, J. R. Halpern,M. E. Greenberg, C. L. Sawyers, and R. J. Davis. 1997. A cytoplasmic inhibitorof the JNK signal transduction pathway.Science 277:693.

39. Arch, R. H., R. W. Gedrich, and C. B. Thompson. 1998. Tumor necrosis factorreceptor-associated factors (TRAFs): a family of adapter proteins that regulateslife and death.Genes Dev. 12:2821.

40. Nishina, H., K. D. Fischer, L. Radvanyi, A. Shahinian, R. Hakem, E. A. Rubie,A. Bernstein, T. W. Mak, J. R. Woodgett, and J. M. Penninger. 1997. Stress-signalling kinase Sek1 protects thymocytes from apoptosis mediated by CD95and CD3.Nature 385:350.

41. Lee, S. Y., A. Reichlin, A. Santana, K. A. Sokol, M. C. Nussenzweig, andY. Choi. 1997. TRAF2 is essential for JNK but not NF-kB activation and reg-ulates lymphocyte proliferation and survival.Immunity 7:703.

42. Yeh, W.- C., A. Shahinian, D. Speiser, J. Kraunus, F. Billia, A. Wakeh am,J. L. de la Pompa, D. Ferrick, B. Hum, N. Iscove, P. Ohashi, M. Rothe,

D. V. Goeddel, and T. W. Mak. 1997. Early lethality, functional NF-kB activa-tion, and increased sensitivity to TNF-induced cell death in TRAF2-deficientmice. Immunity 7:715.

43. Lee, S. Y., D. R. Kaufman, A. L. Mora, A. Santana, M. Boothby, and Y. Choi.1998. Stimulus-dependent synergism of the antiapoptotic tumor necrosis factorreceptor-associated factor 2 (TRAF2) and nuclear factorkB pathways.J. Exp.Med. 188:1381.

44. Xu, L. X., T. Kukita, Y. Nakano, H. Yu, T. Hotokebuchi, T. Kuratani, T. Iijima,and T. Koga. 1996. Osteoclasts in normal and adjuvant arthritis bone tissuesexpress the mRNA for both type I and II interleukin-1 receptors.Lab. Invest.75:677.

45. Bucay, N., I. Sarosi, C. R. Dunstan, S. Morony, J. Tarpley, C. Capparelli,S. Scully, H. L. Tan, W. Xu, D. L. Lacey, W. J. Boyle, and W. S. Simonet. 1998.Osteoprotegerin-deficient mice develop early onset osteoporosis and arterial cal-cification.Genes Dev. 12:1260.

46. Darnay, B. G., V. Haridas, J. Ni, P. A. Moore, and B. B. Aggarwal. 1998. Char-acterization of the intracellular domain of receptor activator of NF-kB (RANK):interaction with tumor necrosis factor receptor-associated factors and activationof NF-kB and c-Jun N-terminal kinase.J. Biol. Chem. 273:20551.

47. Wong, B. R., R. Josien, S. Y. Lee, M. Vologodskaia, R. M. Steinman, andY. Choi. 1998. The TRAF family of signal transducers mediates NF-kB activa-tion by the TRANCE receptor.J. Biol. Chem. 273:28355.

48. Galibert, L., M. E. Tometsko, D. M. Anderson, D. Cosman, and W. C. Dougall.1998. The involvement of multiple tumor necrosis factor receptor (TNF)-asso-ciated factors in signaling mechanisms of receptor activation of NF-kB, a mem-ber of the TNFR superfamily.J. Biol. Chem. 273:34120.

49. Darnay, B. G., J. Ni, P. A. Moore, and B. B. Aggarwal. 1999. Activation ofNF-kB by RANK requires tumor necrosis factor receptor-associated factor(TRAF) 6 and NF-kB-inducing kinase: identification of a novel TRAF6 interac-tion motif. J. Biol. Chem. 274:7724.

50. Iotsova, V., J. Caamano, J. Loy, Y. Yang, A. Lewin, and R. Bravo. 1997. Os-teopetrosis in mice lacking NF-kB1 and NF-kB2. Nat. Med. 3:1285.

51. Franzoso, G., L. Carlson, L. Xing, L. Poljak, E. W. Shores, K. D. Brown,A. Leonardi, T. Tran, B. F. Boyce, and U. Siebenlist. 1997. Requirement forNF-kB in osteoclast and B-cell development.Genes Dev. 11:3482.

442 REGULATION OF OSTEOCLAST FUNCTION BY ODF

by guest on January 5, 2019http://w

ww

.jimm

unol.org/D

ownloaded from