of February 1, 2019.This information is current as

NFATc1 InductionB Activation andκOsteoclastogenesis via NF-

Pim-1 Regulates RANKL-Induced

and Nacksung KimKabsun Kim, Jung Ha Kim, Bang Ung Youn, Hye Mi Jin

http://www.jimmunol.org/content/185/12/7460doi: 10.4049/jimmunol.1000885November 2010;

2010; 185:7460-7466; Prepublished online 10J Immunol 

Referenceshttp://www.jimmunol.org/content/185/12/7460.full#ref-list-1

, 18 of which you can access for free at: cites 34 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. All rights reserved.1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on February 1, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on February 1, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

Pim-1 Regulates RANKL-Induced Osteoclastogenesis viaNF-kB Activation and NFATc1 Induction

Kabsun Kim, Jung Ha Kim, Bang Ung Youn, Hye Mi Jin, and Nacksung Kim

Pim kinases are emerging as important mediators of cytokine signaling pathways in hematopoietic cells. In this study, we dem-

onstrate that Pim-1 positively regulates RANKL-induced osteoclastogenesis and that Pim-1 expression can be upregulated by

RANKL signaling during osteoclast differentiation. The silencing of Pim-1 by RNA interference or overexpression of a dominant

negative form of Pim-1 (Pim-1 DN) in bone marrow-derived macrophage cells attenuates RANKL-induced osteoclast formation.

Overexpression of Pim-1 DN blocks RANKL-induced activation of TGF-b–activated kinase 1 (TAK1) and NF-kB as well as

expression of NFATc1 during osteoclastogenesis. However, we found that overexpression of TAK1 in the presence of Pim-1 DN

rescues NF-kB activation. Additionally, Pim-1 interacts with RANK as well as TAK1, indicating that Pim-1 is involved in RANKL-

induced NF-kB activation via TAK1. Furthermore, we demonstrate that Pim-1 also regulates NFATc1 transcription activity and

subsequently induces osteoclast-associated receptor expression, an osteoclast-specific gene. Taken together, our results reveal that

Pim-1 positively regulates RANKL-induced osteoclastogenesis. The Journal of Immunology, 2010, 185: 7460–7466.

Osteoclasts are unique cells responsible for resorption ofbone matrix, which is produced by osteoblasts. Althoughbone resorption and formation processes are tightly reg-

ulated in normal bone physiology, certain pathological conditionssuch as postmenopausal osteoporosis and rheumatoid arthritis canoccur when there is excessive bone resorption. Two essential cy-tokines, M-CSF and RANKL, induce osteoclast formation frommonocyte/macrophage lineage precursors.RANKL, a TNF family member, supports osteoclast differen-

tiation, survival, fusion, and activation (1, 2). Upon binding ofRANKL to its cognate receptor RANK, TNFR-associated factors(TRAFs) are recruited to the intracellular domain of RANK. Al-though various TRAFs are known to interact with RANK, onlyTRAF6 seems to play a critical role in RANKL-induced osteoclas-togenesis as well as activation of several intracellular signalingpathways, including stimulation of NF-kB and MAPK pathways,such as JNK and p38 (3, 4). RANKL activates TGF-b–activated ki-nase 1 (TAK1), a MAPK kinase kinase, through a signaling complexcontaining RANK, TAB2, and TRAF6 in RANK-overexpressingcells and RAW264.7 cells (5). TAK1 phosphorylates NF-kB–

inducing kinase to activate the IkB kinase (IKK) abg complex,leading to the activation of the NF-kB pathway (6). RANK also acti-vates AP-1, through induction of its complex c-Fos, and NFATc1 toinduce expression of osteoclast-specific genes, including osteoclast-associated receptor (OSCAR), cathepsin K, calcitonin receptor, andtartrate-resistant acid phosphatase (TRAP). RANKL-inducedNFATc1 expression is dependent on NF-kB and c-Fos during osteo-clastogenesis (7).The Pim family of proto-oncogenes encodes a distinct class of

serine/threonine kinases consisting of Pim-1, Pim-2, and Pim-3.The Pim family kinases have been implicated in the processesof lymphomagenesis, cell growth, and survival (8, 9). These threePim family members share a serine/threonine protein kinase do-main (10). Pim-1 and Pim-2 are widely expressed in hematopoieticcells, whereas Pim-3 expression is highest in kidney, brain, andmammary tissue (11). Little is known about the functional role ofthe Pim kinases in vivo, although a number of observations havepointed to a potential role for Pim-1 in signal transduction. Micedeficient for all three Pim kinases display reduced body size andimpaired responses to hematopoietic growth factors (11). Pim ki-nases have been shown to interact with several cellular substrates,including the proapoptotic protein Bad, NFATc1, Cdc25A, andmembers of the suppressor of cytokine signaling family (12–15).Accumulating evidence demonstrates that Pim-1 controls a di-

verse range of physiological processes, including cell growth, dif-ferentiation, apoptosis, and tumorigenesis. Recently, in vitro assayshave demonstrated that Pim-1 interacts with NFATc1 and phos-phorylates it on several serine residues (14). However, unlike theother known NFATc kinases, Pim-1 enhances NFATc-dependenttransactivation and IL-2 production in Jurkat T cells, without anyeffects on the subcellular localization of NFATc1 (14). AlthoughPim-1 enhances NFATc1 activity directly in a phosphorylation-dependent manner, a role for Pim kinases in RANKL-inducedosteoclastogenesis has not yet been described.In this study, we demonstrate that Pim-1 acts as a modulator of

RANKL-induced osteoclastogenesis via activation of NF-kB andinduction of NFATc1 expression. By overexpressing a dominantnegative form of Pim-1 (Pim-1 DN) in osteoclast precursors, ourdata reveal that Pim-1 DN not only inhibits osteoclast formation,but also attenuates NF-kB activation and NFATc1 and OSCAR

National Research Laboratory for Regulation of Bone Metabolism and Disease,Medical Research Center for Gene Regulation, Research Institute of Medical Scien-ces, Brain Korea 21, Chonnam National University Medical School, Gwangju, SouthKorea

Received for publication March 18, 2010. Accepted for publication October 12,2010.

This work was supported in part by the Korea Science and Engineering FoundationNational Research Laboratory program, by the Korean government (Ministry ofEducation, Science and Technology) (Grant R0A-2007-000-20025-0), and by GrantR13-2002-013-03001-0 from the Korea Science and Engineering Foundation throughthe Medical Research Center for Gene Regulation at Chonnam National University.J.H.K. and K.K. were supported by the Brain Korea 21 Project.

Address correspondence and reprint requests to Dr. Nacksung Kim, National Re-search Laboratory for Regulation of Bone Metabolism and Disease, Chonnam Na-tional University Medical School, Hak-Dong 5, Dong-Ku, Gwangju, 501-746, SouthKorea. E-mail address: nacksung@chonnam.ac.kr

Abbreviations used in this paper: BMM, bone marrow-derived macrophage; DN,dominant negative; HA, hemagglutinin; IKK, IkB kinase; IP, immunoprecipitate;MNC, multinuclear cell; OSCAR, osteoclast-associated receptor; siRNA, small in-terfering RNA; TAK1, TGF-b–activated kinase 1; TRAF, TNFR-associated factor;TRAP, tartrate-resistant acid phosphatase; WCL, whole cell lysate; WT, wild-type.

Copyright� 2010 by TheAmericanAssociation of Immunologists, Inc. 0022-1767/10/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1000885

by guest on February 1, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

expression during osteoclastogenesis. Moreover, we show thatPim-1 is involved in RANKL-induced NF-kB activation throughTAK1. Taken together, our results suggest that Pim-1 acts as apositive modulator for RANKL-induced signaling pathways dur-ing osteoclastogenesis.

Materials and MethodsConstructs

Pim-1, Pim-2, Pim-3, and RANK were generated by RT-PCR using RNAfrom bone marrow-derived macrophages (BMMs) and cloned into pMX-IRES-EGFP, pcDNA3.1, or pIRES-hrGFP-2a. The deletion mutant formsof Pim kinase members, that is, Pim-1 DN (aa 81–313), Pim-2 DN (aa 74–311), and Pim-3 DN (aa 84–326), were constructed by deleting theN-terminal region of each Pim gene, which contains a kinase domain (15–17). A point mutant form of Pim-1, Pim-1 K67M (14), was generated bythe QuickChange method of site-directed mutagenesis (Stratagene, LaJolla, CA). CMV-HA-TAK1 wild-type (WT) and CMV-HA-TAK1 K63Wwere provided by Dr. S. Y. Lee (Ewha Womans University, Seoul, Korea)and have been described previously (18).

Osteoclast formation and TRAP staining

Femurs were aseptically removed from 6- to 8-wk-old ICR mice, and bonemarrow cells were flushed out with a sterile 21-gauge syringe. The cellswere cultured in a-MEM (HyClone Laboratories, Waltham, MA) con-taining 10% FBS (HyClone Laboratories) with M-CSF (30 ng/ml) for3 d. Floating cells were removed and adherent cells (BMMs) were used asosteoclast precursors. To generate osteoclasts, BMMs were cultured withM-CSF (30 ng/ml) and RANKL (100 ng/ml) for 3 d. Cultured cells werefixed and stained for TRAP. TRAP+ multinuclear cells (MNCs) containingmore than three nuclei were counted as osteoclasts.

Retroviral gene transduction

To generate retroviral stock, recombinant plasmids and the parental pMXvector were transfected into the packaging cell line Plat-E using FuGENE 6(Roche Applied Sciences, Indianapolis, IN). Plat-E cells were maintained inDMEM supplemented with 10% FBS and 23 105 cells/well were seeded ina 6-well plate. On the following day, the medium was changed to a-MEMcontaining 10% FBS. Viral supernatant was collected from cultured media

24–48 h after transfection. BMMs were incubated with viral supernatantfor 8 h in the presence of 10 mg/ml polybrene (Sigma-Aldrich, St. Louis,MO). After removing the viral supernatant, BMMs were further culturedwith M-CSF (30 ng/ml) and RANKL (100 ng/ml) for 3 d.

Semiquantitative RT-PCR and real-time PCR

Semiquantitative RT-PCR and real-time PCR analyses were performed aspreviously described (19, 20).

In vitro kinase assay

For the TAK1 kinase assay, BMMs were transduced with either a control orPim-1 DN retrovirus and subsequently stimulated for the times indicated inFig. 5C. Whole cell lysates were harvested from cells and immunopreci-pitated with an anti-TAK1 Ab (Santa Cruz Biotechnology, Santa Cruz, CA).Immunoprecipitates were mixed with 30 ml kinase buffer (20 mM PIPES[pH 7.0], 5 mMMnCl2, 7 mM 2-ME, 0.25 mM b-glycerophosphate, 2 mMDTT, 0.1 mM Na3VO4, 0.4 mM spermine, 50 mM ATP, and 5 mCi [g-32P]ATP). GST-IKKb (1.5 mg) (Millipore, Billerica, MA) recombinant pro-tein was added to the kinase buffer as a substrate. After incubation at 30˚Cfor 30 min, reaction mixtures were resolved by SDS-PAGE followed byautoradiography. To detect Pim-1 phosphorylation by RANKL stimulation,BMMs were stimulated with RANKL (100 ng/ml) for the times indicatedin Fig. 4A. A kinase assay was performed using whole cell lysates fromsamples as described above. GST or GST-Pim-1 recombinant purifiedproteins were used as substrates.

Western blot analysis

Cells from transfected 293T, transduced BMMs, or osteoclasts were har-vested after washing with ice-cold PBS and then lysed in extraction buffer(50 mM Tris-HCl [pH 8.0], 150 mM NaCl, 1 mM EDTA, 0.5% NonidetP-40, and protease inhibitors). Cell lysates were subjected to SDS-PAGEand Western blotting. Primary Abs used included NFATc1 (Santa CruzBiotechnology), OSCAR (21), c-Fos (Calbiochem, San Diego, CA), IkB,phospho-p38, p38, phospho-JNK, JNK, phospho-ERK, ERK (Cell Sig-naling Technology, Danvers, MA), Pim-1, Pim-2, Pim-3 (Santa Cruz Bio-technology), actin (Sigma-Aldrich), Flag (Sigma-Aldrich), and hemag-glutinin (HA) (Roche Applied Sciences). HRP-conjugated secondary Abs(Amersham Biosciences, Piscataway, NJ) were probed and developed withECL solution (Millipore). Signals were detected and analyzed by anLAS3000 luminescent image analyzer (Fuji, Tokyo, Japan).

FIGURE 1. Expression of Pim family members

during RANKL-mediated osteoclastogenesis. BMMs

were cultured with M-CSF and RANKL for the in-

dicated times. A, Whole cell lysates were harvested

from cultured cells and analyzed by Western blot

using Abs specific for Pim-1, Pim-2, Pim-3, c-Fos,

NFATc1, OSCAR, and actin (control). B, Relative

amounts of each protein are shown as indicated.

The Journal of Immunology 7461

by guest on February 1, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

Transfection and luciferase assay

RAW264.7 cells were plated on 24-well plates at a density of 73 104 cells/well 1 d before transfection. Plasmid DNAwas mixed with Lipofectaminereagent and Lipofectamine Plus reagent in serum-free DMEM and trans-fected into the cells per the manufacturer’s protocol. After 20 h of trans-fection, cells were treated with RANKL (100 ng/ml) for 12–16 h. Cellswere washed twice with PBS and lysed in reporter lysis buffer (Promega,Madison, WI). Luciferase activity was measured with a dual-luciferase re-porter assay system (Promega) according to the manufacturer’s instructions.

Small interfering RNA preparation and transfection

Predesigned mouse Pim-1 small interfering RNAs (siRNAs) (ID nos. 63532and 150114), Pim-2 siRNAs (ID nos. 88062 and 88132), and Pim3 siRNAs(ID nos. 89580 and 169806) were purchased from Ambion (Austin, TX).Among siRNAs, ID no. 150114 (for Pim-1), ID no. 88062 (for Pim-2), andID no. 89580 (for Pim-3) demonstrated a significant inhibitory effect on Pimexpression in BMMs. Therefore, we used GFP siRNA as a control, no.150114 as Pim-1 siRNA, no. 88062 as Pim-2 siRNA, and no. 89580 as Pim-3 siRNA. siRNAs were transfected into BMMs using X-tremeGENE siRNAtransfection reagent (Roche Applied Sciences). After 72 h, cells were usedfor osteoclastogenesis and other analyses as indicated.

ResultsPim-1 expression is upregulated by RANKL during osteoclastdifferentiation

To examine the expression patterns of Pim family members, weperformed Western blot analysis during osteoclast differentiation.Osteoclast differentiation was induced by culturing BMMs in thepresence of M-CSF and RANKL (Fig. 1A). RANKL stimulationincreased the expression of c-Fos and NFATc1, which are key tran-scription factors in osteoclastogenesis. The induction of NFATc1gene expression was followed by the expression of OSCAR, whichis an osteoclast-specific gene (21). Among Pim kinases, Pim-1 andPim-2 expression was readily detectable during osteoclast differ-entiation, whereas expression of Pim-3 was only faintly detected.Pim-1 protein levels peaked at day 2 and subsequently decreased,whereas Pim-2 expression remained unchanged during RANKL-induced osteoclastogenesis (Fig. 1).

Pim-1 positively regulates RANKL-induced osteoclastogenesis

Given our observation that Pim family members are expressed inosteoclast lineage cells, we investigated whether Pim kinases arerequired for efficient osteoclast differentiation. To examine theeffect of Pim family members on osteoclast differentiation, WT orDN forms of each Pim kinase were overexpressed in BMMs.Transduced BMMs were cultured with various concentrations ofRANKL in the presence ofM-CSF and cells were stained for TRAP(Fig. 2A). RANKL treatment of control vector-infected BMMsincreased the number of TRAP+ MNCs. Compared with the controlvector, overexpression of Pim-1 WT in BMMs slightly increasedthe formation of TRAP+ MNCs mediated by RANKL, whereasoverexpression of Pim-1 DN or Pim-1 K67M (a kinase-inactivemutant of Pim-1) in BMMs significantly inhibited RANKL-induced osteoclastogenesis (Fig. 2). However, overexpression ofWT or DN forms of Pim-2 or Pim-3 in BMMs did not affect theformation of TRAP+ MNCs mediated by RANKL when comparedwith overexpression of control vector (Fig. 2).Next, we used siRNA of each Pim kinase to investigate the role

of Pim family members in RANKL-induced osteoclastogenesis.After 48 h of transfection of Pim-1 siRNA, Pim-2 siRNA, Pim-3siRNA, or control GFP siRNA into BMMs, downregulation of Pimkinases by each Pim kinase-specific siRNA was confirmed byWestern blot analysis (Fig. 3A). siRNA-transfected BMMs werecultured for 4 d with M-CSF alone or M-CSF and various con-centrations of RANKL. RANKL treatment of control GFP siRNA-

transfected BMMs increased the number of TRAP+ MNCs ina dose-dependent manner (Fig. 3B, 3C). Compared with the con-trol siRNA, the silencing of Pim-1 in BMMs resulted in a signif-icant decrease in the formation of TRAP+ MNCs mediated byRANKL. Similar to the results described for Pim-2 DN and Pim-3DN above, the siRNAs specific for Pim-2 and Pim-3 did not affectRANKL-mediated osteoclastogenesis (Fig. 3B, 3C). Taken to-gether, these results suggest that Pim-1 may play a role in RANKL-induced osteoclastogenesis.

Activation of Pim-1 plays a role in RANKL-induced NF-kBactivation

To investigate the mechanism by which Pim-1 modulates RANKL-induced osteoclastogenesis, we determined whether RANKL couldactivate Pim-1 by an in vitro kinase assay with GST or GST-Pim-1recombinant proteins using whole cell lysates from unstimulatedBMMs and from BMMs cultured in the presence of RANKL. Asshown in Fig. 4A, RANKL stimulation resulted in the phosphory-

FIGURE 2. Overexpression of Pim-1 DN in BMMs inhibits osteoclas-

togenesis. BMMs were transduced with pMX-IRES-EGFP empty vector

(control) or pMX-IRES-EGFP with Pim-1 WT, Pim-1 DN, a kinase-dead

mutant form of Pim-1 (Pim-1 K67M), Pim-2 WT, Pim-2 DN, Pim-3 WT,

or Pim-3 DN, and cultured for 3 d with M-CSF alone or M-CSF and

various concentrations of RANKL as indicated. A, Cultured cells were

fixed and stained for TRAP. B, Numbers of TRAP+ MNCs were counted.

Data represent means 6 SD of triplicate samples. *p , 0.05; **p , 0.01;#p, 0.001 versus control vector. Results are representative of at least three

independent sets of similar experiments.

7462 THE ROLE OF Pim-1 IN OSTEOCLASTOGENESIS

by guest on February 1, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

lation of GST-Pim-1, but not of GST, indicating that activation ofPim-1 can mediate RANKL-induced signaling pathways.Next, we examined the role of Pim-1 in osteoclastogenesis down-

stream of RANK by transducing BMMs with control or Pim-1DN retroviruses followed by RANKL stimulation. Consistent withprevious results (22, 23), RANKL induced degradation of IkB andactivation of ERK, p38 MAPK, and JNK in control vector-infectedBMMs. However, RANKL-induced IkB degradation was blockedby overexpression of Pim-1 DN, while other signaling pathways,including activation of ERK, JNK, and p38 MAPK, were not sig-nificantly affected (Fig. 4B).To examine the role of Pim-1 in RANKL-induced NF-kB ac-

tivation, we used a reporter assay involving transient transfectioninto RAW264.7 cells, which is a monocyte/macrophage cell linecapable of differentiating into multinuclear osteoclasts. An NF-kBluciferase reporter plasmid was cotransfected with Pim-1 WT orPim-1 DN in the presence or absence of RANKL. RANKL stim-ulation induced NF-kB transcriptional activity, and this inductionwas further potentiated by coexpression of Pim-1 WT. In contrast,overexpression of Pim-1 DN inhibited the effects of RANKL onNF-kB transcriptional activity (Fig. 4C). Taken together, these

results suggest that RANKL activates Pim-1, and that activation ofPim-1 plays a role in RANKL-induced NF-kB activation.

Pim-1 regulates RANKL-induced NF-kB activation throughTAK1

TAK1 has been shown to play a role in IL-1– and RANKL-mediated NF-kB activation (5, 6, 24). Given our data indicatingthat Pim-1 modulates RANKL-induced NF-kB activation, we ex-amined whether TAK1 is involved in the RANKL/Pim-1/NF-kBsignaling pathway. Overexpression of Pim-1WT induced RANKL-mediated NF-kB activation (Fig. 5A). However, the induction ofluciferase activity by RANKL/Pim-1 was significantly reduced byoverexpression of TAK1K63W(aDNTAK1), indicating that TAK1might be a downstream target of RANKL/Pim-1 in NF-kB activa-tion. To test this hypothesis, we examined whether overexpressionof TAK1 WT could override the inhibitory effects of Pim-1 DN on

FIGURE 4. Overexpression of Pim-1 DN blocks RANKL-induced NF-

kB activation. A, BMMs were stimulated with RANKL for the indicated

times. Whole cell lysates (WCL) were subjected to in vitro kinase assays

with GST or GST-Pim-1 as a substrate (upper panel). Lysates were im-

munoblotted with anti-IkB and anti-actin Abs (lower panel). B, BMMs

were transduced with pMX-IRES-EGFP (control) or Pim-1 DN retrovirus

and stimulated with RANKL (1 mg/ml) for the indicated times. Whole cell

lysates were subjected to Western blot analysis using Abs specific for IkB,

phospho-p38, p38, phospho-JNK, JNK, phospho-ERK, ERK, and actin. C,

RAW264.7 cells were cotransfected with NF-kB luciferase reporter to-

gether with the indicated plasmids expressing Pim-1 WT (500 ng) or Pim-1

DN (500 ng). After 24 h of transfection, cells were further cultured with

RANKL (100 ng/ml) for 12–16 h and then lysed for luciferase assay.

Luciferase activity was measured using a dual-luciferase reporter assay

system. Data represent means 6 SD of triplicate samples. Results are

representative of at least three independent sets of similar experiments.

**p , 0.01.

FIGURE 3. Downregulation of Pim-1 in BMMs attenuates RANKL-

induced osteoclastogenesis. BMMs were transfected with GFP siRNA

(control) or siRNA specific for Pim-1, Pim-2, or Pim-3. A, Transfected

BMMs were cultured for 2 d with M-CSF. Whole cell lysates were har-

vested from cultured cells and analyzed by Western blot using Abs specific

for Pim-1, Pim-2, Pim-3, and actin (control). B, Transfected BMMs were

cultured for 4 d with M-CSF alone or M-CSF and various concentrations of

RANKL as indicated. Cultured cells were fixed and stained for TRAP. C,

Numbers of TRAP+ MNCs were counted. Data represent means 6 SD of

triplicate samples. **p , 0.01; #p , 0.001 versus GFP siRNA. Results are

representative of at least three independent sets of similar experiments.

The Journal of Immunology 7463

by guest on February 1, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

activation of NF-kB by RANKL in RAW264.7 cells. As shown inFig. 5B, TAK1 WT rescued activation of NF-kB by RANKL in thepresence of Pim-1 DN.Next, we examinedwhether Pim-1 kinase is involved in RANKL-

induced TAK1 activation via an in vitro kinase assay. BMMs weretransducedwith pMX-IRES-EGFP (control) or Pim-1DN retrovirusand cultured for 2 d with M-CSF and RANKL. Cell lysates frompreosteoclasts were immunoprecipitated with anti-TAK1 Ab, andimmunoprecipitates were subjected to an in vitro kinase assay withGST-IKKb as a substrate. As shown in Fig. 5C and 5D, the in-duction of TAK1 kinase activity by RANKL stimulation was at-tenuated by overexpression of Pim-1 DN. Collectively, theseresults indicate that Pim-1 regulates RANKL-mediated NF-kBactivation through TAK1.

Pim-1 interacts with RANK and TAK1

Given our data indicating that Pim-1 participates in RANKL-induced NF-kB activation, we examined whether Pim-1 interactswith RANK. 293T cells were cotransfected with RANK and Pim-1, and cell lysates were immunoprecipitated with anti-Flag (Fig.6A) or anti-HA (Fig. 6B) Abs. Immunoprecipitated samples weresubjected to SDS-PAGE and Western blotting by anti-Flag or anti-HA Abs. As shown in Fig. 6A and 6B, we confirmed that RANKinteracts with Pim-1. Consistent with previous results (5), RANKalso interacts with TAK1 (Fig. 6C, 6D). When Pim-1 and TAK1were overexpressed in 293T cells, the interaction between Pim-1and TAK1 was observed by immunoprecipitation assay (Fig. 6E).The interaction between TAK1 and both Pim-1 and RANK wasalso observed endogenously in preosteoclast cells (Fig. 6F). Thesedata suggest that the formation of a complex containing RANK,Pim-1, and TAK1 might play a role in RANKL-mediated signal-ing cascades.

Pim-1 regulates RANKL-mediated induction of NFATc1 andOSCAR during osteoclastogenesis

Because Pim-1 modulates RANKL-induced osteoclast differenti-ation, we investigated whether overexpression of Pim-1 DN couldaffect the expression of other genes known to play a role in os-teoclastogenesis. Consistent with previous results (25–27), RANKLinduced the expression of c-Fos, NFATc1, OSCAR, and TRAPduring osteoclast differentiation (Fig. 7A). Compared with samplestransduced with a control vector, overexpression of Pim-1 DN inBMMs inhibited the induction of NFATc1, OSCAR, and TRAPduring RANKL-mediated osteoclastogenesis (Fig. 7A). We con-firmed these results by Western blot (Fig. 7B), demonstrating thatall downstream molecules were affected at both the mRNA andprotein level by overexpression of Pim-1 DN during osteoclasto-geneis, with the exception of c-Fos (Fig. 7A–C).To investigate whether Pim-1 directly regulates NFATc1 trans-

activation, we used an OSCAR luciferase reporter assay. Whenan OSCAR luciferase reporter plasmid was cotransfected withNFATc1, NFATc1 increased OSCAR luciferase activity (Fig. 7D).NFATc1-induced transcriptional activity was further enhanced byPim-1 WT, whereas Pim-1 DN significantly reduced NFATc1transactivation (Fig. 7D). Taken together, our results indicate thatPim-1 is involved in NFATc1 induction as well as NFATc1 tran-scriptional activity during RANKL-mediated osteoclastogenesis.

DiscussionThe Pim kinase family consists of three members: Pim-1, Pim-2,and Pim-3. Because of sequence similarity between these Pimkinases, overlapping functions have been inferred. AlthoughPim kinases are widely expressed, patterns of expression of eachPim kinase are somewhat different: the highest Pim-1 mRNA levels

FIGURE 5. Pim-1 regulates RANKL-stimulated NF-kB transactivation through TAK1. A, RAW264.7 cells were cotransfected with NF-kB luciferase

reporter together with the indicated plasmids expressing Pim-1 WT (500 ng) and increasing concentrations (200 and 500 ng) of a DN form of TAK1 (TAK1

K63W) as indicated. B, RAW264.7 cells were cotransfected with NF-kB luciferase reporter together with the indicated plasmids expressing Pim-1 DN (500

ng) and increasing concentrations (200 and 500 ng) of TAK1 WT as indicated. A and B, After 24 h of transfection, cells were further cultured with RANKL

(100 ng/ml) for 12–16 h and then lysed for luciferase assay. Luciferase activity was measured using a dual-luciferase reporter assay system. Data represent

means 6 SD of triplicate samples. The results are representative of at least three independent sets of similar experiments. C, BMMs were transduced with

pMX-IRES-EGFP (control) or Pim-1 DN retrovirus and stimulated with RANKL (1 mg/ml) for the indicated times. Cell lysates were immunoprecipitated

(IP) with anti-TAK1 Ab, and immunoprecipitates were subjected to in vitro kinase assay with GST-IKKb as a substrate (upper panel) and Western blot

analysis with anti-TAK1 Ab (middle panel). Whole cell extracts were probed with anti-HA (for Pim-1 DN) or anti-actin (control) Abs (lower panel). The

results are representative of two independent sets of similar experiments. D, The relative intensity of phosphorylated IKKb is shown. *p , 0.05; **p ,0.01; #p , 0.001.

7464 THE ROLE OF Pim-1 IN OSTEOCLASTOGENESIS

by guest on February 1, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

are found in the thymus and testis (11, 28), whereas Pim-2 mRNAexpression is highest in the brain and thymus (11, 29), and Pim-3mRNA is most abundant in the kidney (11, 30). Additionally, in-duction of Pim kinase expression by a wide range of growth factorsand cytokines differs between the different Pim family members.For example, IL-3 stimulation of bone marrow-derived mast cellsstrongly induces Pim-1 and Pim-2 expression, but only weaklyinduces expression of Pim-3 (11). Therefore, it seems that the spatialand temporal expression of each Pim kinase might be importantfor cell growth and differentiation upon engagement with variouscytokines. Of note, our data demonstrate a previously undescribedpattern of expression during RANKL-induced osteoclastogenesiswith significant upregulation of Pim-1 expression and unchangedPim-2 and Pim-3 expression. Additionally, our results from over-expression of DN forms of each Pim kinase or downregulation ofeach Pim kinase by siRNA demonstrate that only Pim-1 affectsRANKL-induced osteoclastogenesis. Furthermore, unlike Pim-1DN, overexpression of Pim-2 DN or Pim-3 DN in BMMs did notblock RANKL-mediated degradation of IkB and induction ofNFATc1 expression during osteoclastogenesis (data not shown).Thus, our results indicate that Pim-1 acts as a major modulator ofRANKL-mediated osteoclastogenesis.The downstream events of multiple signal-transduction path-

ways used by RANK include activation of transcription factorssuch as NF-kB, induction and activation of NFATc1, and stimu-lation of threeMAPK families: ERK, JNK, and p38 (1, 2). We showthat Pim-1 regulates RANKL-induced NF-kB activation amongRANK-proximal signaling pathways. Similar to our results, Zem-skova et al. (31) reported that Pim-1 mediates docetaxel-inducedactivation of NF-kB transcriptional activity in prostate cancer cells,suggesting that Pim-1 may play a role in NF-kB activation. Of note,although NF-kB activation appears to be the major RANKL-

induced proximal signaling pathway regulated by Pim-1, in someexperimental replicates we did observe a slight effect of Pim-1 onp38 activation as well. Thus, we cannot rule out the possibility thatPim-1 might have an effect on RANKL-induced p38 activation.RANKL activates TAK1 through a signaling complex containing

RANK, TAB2, and TRAF6 (5). It has been reported that TAK1 andMKK6 are required for RANKL-mediated NFATc1 induction andNF-kB transactivation during osteoclast differentiation (32). Ourdata demonstrate that the Pim-1/TAK1 signaling cascade playsa role in RANKL-induced NF-kB activation during osteoclasto-genesis. First, we demonstrated that TAK1 DN blocked RANKL/Pim-1–induced NF-kB transactivation. Second, we showed thatTAK1 WT rescued the inhibitory effect of Pim-1 DN on RANKL-induced NF-kB activation. Third, we found that Pim-1 DN blockedRANKL-induced activation of TAK1, suggesting that TAK1 maybe a downstream target of RANKL/Pim-1 in NF-kB activation.Fourth, we observed the formation of a RANK-Pim-1-TAK1 com-plex. Thus, our data demonstrate that the RANKL/Pim-1/TAK1signaling cascade is potentially important for NF-kB activationduring osteoclastogenesis.

FIGURE 6. Interaction of Pim-1 with RANK and TAK1. 293T cells

were transiently transfected with plasmids expressing Flag-RANK and

HA-Pim-1 (A, B), Flag-RANK and HA-TAK1 (C, D), or HA-TAK1 and

HA-Pim-1 (E) as indicated. After 36 h of transfection, cell lysates were

immunoprecipitated with anti-Flag, anti-HA, or anti-TAK1 Abs, and im-

munoprecipitates were subjected to Western blot analysis with anti-Flag or

anti-HA Abs (upper panel). Whole cell lysates (WCL) were analyzed by

Western blot analysis with anti-Flag or anti-HA Abs (lower panel). F,

Lysates from preosteoclasts were imunoprecipitated (IP) with anti-TAK1

or control IgG Abs. Immunoprecipitated samples (upper panel) or whole

cell lysates (WCL) (lower panel) were subjected to Western blotting for

detection of Pim-1, RANK, and TAK1.FIGURE 7. Pim-1 modulates RANKL-mediated induction of NFATc1

and OSCAR expression during osteoclastogenesis. BMMs were trans-

duced with pMX-IRES-EGFP (control) or Pim-1 DN retrovirus and

cultured with M-CSF and RANKL for the indicated times. A, Total RNA

was collected from each time point. RT-PCR was performed to assess the

expression of indicated genes. B and C, Whole cell lysates were analyzed

by Western blot using Abs specific for OSCAR, NFATc1, TRAP, c-Fos,

and actin. D, 293T cells were cotransfected with OSCAR luciferase

reporter together with the indicated plasmids. Luciferase activity was

measured using a dual-luciferase reporter assay system. Data represent

means 6 SD of triplicate samples. Results are representative of at least

three independent sets of similar experiments. *p , 0.05; **p , 0.01

versus NFATc1 control.

The Journal of Immunology 7465

by guest on February 1, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

RANKL/RANK signaling increases expression levels ofNFATc1, akeymodulator inosteoclastogenesis. Takatsunaet al. (33)showed that an NF-kB inhibitor can suppress RANKL-inducedosteoclastogenesis by downregulation of NFATc1. Additionally,chromatin immunoprecipitation experiments have revealed that theNF-kB components p50 and p65 are recruited to the NFATc1promoter by RANKL stimulation (34). Consistent with this,NFATc1 promoter regions contain NF-kB binding sites and NF-kB activates NFATc1, as shown by a luciferase reporter gene assay(7, 34). These data indicate that NFATc1 expression is regulatedby NF-kB in RANKL-induced osteoclastogenesis. We observedthat overexpression of Pim-1 DN attenuates RANKL-induced NF-kB activation as well as induction of NFATc1 expression. There-fore, we propose that Pim-1 regulates NF-kΒ and subsequentlyinduces NFATc1 gene expression during RANKL-mediated osteo-clastogenesis.Phosphorylation by Pim-1 can lead to inactivation of repressors

or activation of transcription factors or regulatory proteins involvedin various signaling pathways (10). It has been reported that Pim-1enhances NFATc-dependent transactivation and IL-2 production inJurkat T cells, and that Pim-1 phosphorylates NFATc1 in vitro onserine residues (14), indicating that NFATc1 is one of the down-stream target genes of Pim-1 kinase. Consistent with previousresults (14), we observed that Pim-1 can directly interact withNFATc1 using a GST pull-down assay (data not shown). Our datafrom reporter assays demonstrate that Pim-1 can directly activateNFATc1, which, in turn, induces expression of downstream targetgenes such asOSCAR.Taken together, our results indicate that Pim-1 activates NF-kB as well as NFATc1 during RANKL-mediatedosteoclastogenesis.In conclusion, this study demonstrates that Pim-1 plays a key

role in RANKL-mediated osteoclastogenesis by regulating NF-kBactivation and induction of NFATc1 expression. Additionally, theRANKL/Pim-1/TAK1/NF-kB signaling cascade may be importantfor induction of NFATc1 gene expression. Thus, our study eluci-dates a role for Pim-1 as a positive regulator of RANKL-mediatedosteoclast differentiation.

AcknowledgmentsWe thank S.Y. Lee for providing CMV-TAK1 WT and CMV-TAK1 K63W,

T. Kitamura for providing the Plat-E cell line, and A. Ko for assistance.

DisclosuresThe authors have no financial conflicts of interest.

References1. Boyle, W. J., W. S. Simonet, and D. L. Lacey. 2003. Osteoclast differentiation

and activation. Nature 423: 337–342.2. Walsh, M. C., N. Kim, Y. Kadono, J. Rho, S. Y. Lee, J. Lorenzo, and Y. Choi.

2006. Osteoimmunology: interplay between the immune system and bone me-tabolism. Annu. Rev. Immunol. 24: 33–63.

3. Kim, N., Y. Kadono, M. Takami, J. Lee, S. H. Lee, F. Okada, J. H. Kim,T. Kobayashi, P. R. Odgren, H. Nakano, et al. 2005. Osteoclast differentiationindependent of the TRANCE-RANK-TRAF6 axis. J. Exp. Med. 202: 589–595.

4. Lee, Z. H., and H. H. Kim. 2003. Signal transduction by receptor activator ofnuclear factor kB in osteoclasts. Biochem. Biophys. Res. Commun. 305: 211–214.

5. Mizukami, J., G. Takaesu, H. Akatsuka, H. Sakurai, J. Ninomiya-Tsuji,K.Matsumoto, and N. Sakurai. 2002. Receptor activator of NF-kB ligand (RANKL)activates TAK1 mitogen-activated protein kinase kinase kinase through a signalingcomplex containing RANK, TAB2, and TRAF6.Mol. Cell. Biol. 22: 992–1000.

6. Ninomiya-Tsuji, J., K. Kishimoto, A. Hiyama, J. Inoue, Z. Cao, andK. Matsumoto. 1999. The kinase TAK1 can activate the NIK-IkB as well as theMAP kinase cascade in the IL-1 signalling pathway. Nature 398: 252–256.

7. Asagiri, M., and H. Takayanagi. 2007. The molecular understanding of osteo-clast differentiation. Bone 40: 251–264.

8. Breuer, M. L., H. T. Cuypers, and A. Berns. 1989. Evidence for the involvementof Pim-2, a new common proviral insertion site, in progression of lymphomas.EMBO J. 8: 743–748.

9. Fox, C. J., P. S. Hammerman, and C. B. Thompson. 2005. The Pim kinases controlrapamycin-resistant T cell survival and activation. J. Exp. Med. 201: 259–266.

10. Bachmann, M., and T. Moroy. 2005. The serine/threonine kinase Pim-1. Int. J.Biochem. Cell Biol. 37: 726–730.

11. Mikkers, H., M. Nawijn, J. Allen, C. Brouwers, E. Verhoeven, J. Jonkers, andA. Berns. 2004. Mice deficient for all PIM kinases display reduced body size andimpaired responses to hematopoietic growth factors. Mol. Cell. Biol. 24: 6104–6115.

12. Chen, X. P., J. A. Losman, S. Cowan, E. Donahue, S. Fay, B. Q. Vuong,M.C.Nawijn,D. Capece, V. L. Cohan, and P.Rothman. 2002. Pim serine/threoninekinases regulate the stability of Socs-1 protein. Proc. Natl. Acad. Sci. USA 99:2175–2180.

13. Mochizuki, T., C. Kitanaka, K. Noguchi, T. Muramatsu, A. Asai, and Y. Kuchino.1999. Physical and functional interactions between Pim-1 kinase and Cdc25Aphosphatase: implications for the Pim-1-mediated activation of the c-Myc sig-naling pathway. J. Biol. Chem. 274: 18659–18666.

14. Rainio, E. M., J. Sandholm, and P. J. Koskinen. 2002. Cutting edge: transcrip-tional activity of NFATc1 is enhanced by the Pim-1 kinase. J. Immunol. 168:1524–1527.

15. Yan, B., M. Zemskova, S. Holder, V. Chin, A. Kraft, P. J. Koskinen, and M. Lilly.2003. The PIM-2 kinase phosphorylates BAD on serine 112 and reverses BAD-induced cell death. J. Biol. Chem. 278: 45358–45367.

16. Lilly, M., J. Sandholm, J. J. Cooper, P. J. Koskinen, and A. Kraft. 1999. ThePIM-1 serine kinase prolongs survival and inhibits apoptosis-related mitochon-drial dysfunction in part through a bcl-2-dependent pathway. Oncogene 18:4022–4031.

17. Deneen, B., S. M. Welford, T. Ho, F. Hernandez, I. Kurland, and C. T. Denny.2003. PIM3 proto-oncogene kinase is a common transcriptional target of di-vergent EWS/ETS oncoproteins. Mol. Cell. Biol. 23: 3897–3908.

18. Lee, N. K., H. K. Choi, D. K. Kim, and S. Y. Lee. 2006. Rac1 GTPase regulatesosteoclast differentiation through TRANCE-induced NF-kB activation. Mol.Cell. Biochem. 281: 55–61.

19. Kim, K., J. H. Kim, J. Lee, H. M. Jin, H. Kook, K. K. Kim, S. Y. Lee, andN. Kim. 2007. MafB negatively regulates RANKL-mediated osteoclast differ-entiation. Blood 109: 3253–3259.

20. Kim, K., J. H. Kim, J. Lee, H. M. Jin, S. H. Lee, D. E. Fisher, H. Kook,K. K. Kim, Y. Choi, and N. Kim. 2005. Nuclear factor of activated T cells c1induces osteoclast-associated receptor gene expression during tumor necrosisfactor-related activation-induced cytokine-mediated osteoclastogenesis. J. Biol.Chem. 280: 35209–35216.

21. Kim, N., M. Takami, J. Rho, R. Josien, and Y. Choi. 2002. A novel member ofthe leukocyte receptor complex regulates osteoclast differentiation. J. Exp. Med.195: 201–209.

22. Kim, J. H., H. M. Jin, K. Kim, I. Song, B. U. Youn, K. Matsuo, and N. Kim.2009. The mechanism of osteoclast differentiation induced by IL-1. J. Immunol.183: 1862–1870.

23. Kim, J. H., K. Kim, H. M. Jin, I. Song, B. U. Youn, J. Lee, and N. Kim. 2009.Silibinin inhibits osteoclast differentiation mediated by TNF family members.Mol. Cells 28: 201–207.

24. Takaesu, G., R. M. Surabhi, K. J. Park, J. Ninomiya-Tsuji, K. Matsumoto, andR. B. Gaynor. 2003. TAK1 is critical for IkB kinase-mediated activation of theNF-kB pathway. J. Mol. Biol. 326: 105–115.

25. Kim, J. H., K. Kim, H. M. Jin, B. U. Youn, I. Song, H. S. Choi, and N. Kim.2008. Upstream stimulatory factors regulate OSCAR gene expression inRANKL-mediated osteoclast differentiation. J. Mol. Biol. 383: 502–511.

26. Kim, K., J. Lee, J. H. Kim, H. M. Jin, B. Zhou, S. Y. Lee, and N. Kim. 2007.Protein inhibitor of activated STAT 3 modulates osteoclastogenesis by down-regulation of NFATc1 and osteoclast-associated receptor. J. Immunol. 178:5588–5594.

27. Lee, J., K. Kim, J. H. Kim, H. M. Jin, H. K. Choi, S. H. Lee, H. Kook, K. K. Kim,Y. Yokota, S. Y. Lee, et al. 2006. Id helix-loop-helix proteins negatively regulateTRANCE-mediated osteoclast differentiation. Blood 107: 2686–2693.

28. Saris, C. J., J. Domen, and A. Berns. 1991. The pim-1 oncogene encodes tworelated protein-serine/threonine kinases by alternative initiation at AUG andCUG. EMBO J. 10: 655–664.

29. Allen, J. D., E. Verhoeven, J. Domen, M. van der Valk, and A. Berns. 1997. Pim-2transgene induces lymphoid tumors, exhibiting potent synergy with c-myc. On-cogene 15: 1133–1141.

30. Feldman, J. D., L. Vician, M. Crispino, G. Tocco, V. L. Marcheselli,N. G. Bazan, M. Baudry, and H. R. Herschman. 1998. KID-1, a protein kinaseinduced by depolarization in brain. J. Biol. Chem. 273: 16535–16543.

31. Zemskova, M., E. Sahakian, S. Bashkirova, and M. Lilly. 2008. The PIM1 kinaseis a critical component of a survival pathway activated by docetaxel and pro-motes survival of docetaxel-treated prostate cancer cells. J. Biol. Chem. 283:20635–20644.

32. Huang, H., J. Ryu, J. Ha, E. J. Chang, H. J. Kim, H. M. Kim, T. Kitamura,Z. H. Lee, and H. H. Kim. 2006. Osteoclast differentiation requires TAK1 andMKK6 for NFATc1 induction and NF-kB transactivation by RANKL. Cell DeathDiffer. 13: 1879–1891.

33. Takatsuna, H., M. Asagiri, T. Kubota, K. Oka, T. Osada, C. Sugiyama, H. Saito,K. Aoki, K. Ohya, H. Takayanagi, and K. Umezawa. 2005. Inhibition ofRANKL-induced osteoclastogenesis by (2)-DHMEQ, a novel NF-kB inhibitor,through downregulation of NFATc1. J. Bone Miner. Res. 20: 653–662.

34. Asagiri, M., K. Sato, T. Usami, S. Ochi, H. Nishina, H. Yoshida, I. Morita,E. F. Wagner, T. W. Mak, E. Serfling, and H. Takayanagi. 2005. Autoamplificationof NFATc1 expression determines its essential role in bone homeostasis. J. Exp.Med. 202: 1261–1269.

7466 THE ROLE OF Pim-1 IN OSTEOCLASTOGENESIS

by guest on February 1, 2019http://w

ww

.jimm

unol.org/D

ownloaded from