of February 11, 2018.This information is current as

MiceDifferentiation and Prevents Bone Loss in Dynein Light Chain LC8 Inhibits Osteoclast

Han-Sung Kim, Soo Young Lee and Woojin JeongJi Young Huh, Doo Ri Park, Hyojung Lee, Dong-Hyun Seo, Hyeryeon Kim, Seungha Hyeon, Hojin Kim, Yoohee Yang,

ol.1202525http://www.jimmunol.org/content/early/2013/01/04/jimmun

published online 4 January 2013J Immunol 

MaterialSupplementary

5.DC1http://www.jimmunol.org/content/suppl/2013/01/07/jimmunol.120252

        average*  

4 weeks from acceptance to publicationSpeedy Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

?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, Inc. All rights reserved.Copyright © 2013 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 February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on February 11, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

Dynein Light Chain LC8 Inhibits Osteoclast Differentiationand Prevents Bone Loss in Mice

Hyeryeon Kim,* Seungha Hyeon,* Hojin Kim,* Yoohee Yang,* Ji Young Huh,*

Doo Ri Park,* Hyojung Lee,* Dong-Hyun Seo,† Han-Sung Kim,† Soo Young Lee,* and

Woojin Jeong*

NF-kB is one of the key transcription factors activated by receptor activator of NF-kB ligand (RANKL) during osteoclast

differentiation. The 8-kDa dynein L chain (LC8) was previously identified as a novel NF-kB regulator. However, its physiological

role as an NF-kB inhibitor remains elusive. In this study, we showed the inhibitory role of LC8 in RANKL-induced osteoclasto-

genesis and signaling pathways and its protective role in osteolytic animal models. LC8 suppressed RANKL-induced osteoclast

differentiation, actin ring formation, and osteoclastic bone resorption. LC8 inhibited RANKL-induced phosphorylation and

subsequent degradation of IkBa, the expression of c-Fos, and the consequent activation of NFATc1, which is a pivotal determinant

of osteoclastogenesis. LC8 also inhibited RANKL-induced activation of JNK and ERK. LC8-transgenic mice exhibited a mild

osteopetrotic phenotype. Moreover, LC8 inhibited inflammation-induced bone erosion and protected against ovariectomy-induced

bone loss in mice. Thus, our results suggest that LC8 inhibits osteoclast differentiation by regulating NF-kB and MAPK pathways

and provide the molecular basis of a new strategy for treating osteoporosis and other bone diseases. The Journal of Immunology,

2013, 190: 000–000.

Bone undergoes a metabolic remodeling process that de-pends on a balance between the breakdown (resorption)by osteoclasts and the synthesis (formation) of bone by

osteoblasts (1). Most bone diseases such as osteoporosis and rheu-matoid arthritis are caused by excessive osteoclastic activity, whichleads to an imbalance between resorption and formation during boneremodeling that favors resorption (2). Thus, the regulatory mecha-nisms of osteoclast differentiation have significant clinical impli-cations.Receptor activator of NF-kB ligand (RANKL) (3), also known

as TNF-related activation-induced cytokine (4), osteoclast differ-entiation factor (5), and osteoprotegerin ligand (6), is an essentialmolecule for osteoclast differentiation. Specifically, RANKL is ex-pressed in osteoblasts, activated T cells, and stromal cells (7) andstimulates morphologic changes, cell survival, and osteolytic ac-

tivity in osteoclasts, in cooperation with M-CSF (8). The binding ofRANKL to its receptor, RANK, induces the recruitment of TNFR-associated factor (TRAF) family proteins such as TRAF6, whichactivates the NF-kB and MAPK pathways, including JNK, ERK,and p38 (9, 10).TRAF6-activated NF-kB induces the initial expression of

NFATc1, a key determinant of osteoclast differentiation. RANKLalso induces the expression of c-Fos, which is involved in therobust induction of NFATc1 as a component of the AP-1 tran-scription factor family (11). NFATc1 cooperates with AP-1, PU.1transcription factor, and microphthalmia-associated transcriptionfactor to induce expression of several osteoclast-specific genes,such as cathepsin K, tartrate-resistant acid phosphatase (TRAP),calcitonin receptor, and osteoclast-associated receptor (9, 10).Previous studies have shown that c-Fos and NFATc1 play criticalroles in osteoclast differentiation and bone resorption and thatNFATc1 is the downstream target of c-Fos during osteoclasto-genesis: c-Fos–knockout mice were deficient in osteoclasts andexhibited an osteopetrotic phenotype (12), but ectopic expressionof NFATc1 in c-Fos–deficient cells restored the transcription ofosteoclastogenesis-associated genes and resumed osteoclast for-mation and bone resorption (13).The 8-kDa dynein L chain (LC8; also known as DLC8 or DLC1)

is necessary for intracellular motile processes such as mitosis andvesicular transport as an essential component of microtubule-basedmotor dynein (14, 15). In addition, LC8 binds to and regulates thebiological functions of many proteins (16–23). In a previous study,we identified LC8 as a potential substrate of 14-kDa humanthioredoxin-related protein (TRP14), a disulfide reductase thatinhibits TNF-a–induced NF-kB activation by suppressing thephosphorylation of IkBa (24). More recently, we showed that LC8inhibits IkBa phosphorylation by IkB kinase (IKK) via its redox-dependent interaction with IkBa and that TRP14 regulates thisinhibitory activity by facilitating the reduction of LC8 (25, 26).Several genetic studies have shown that the NF-kB plays a cru-

cial role in osteoclast differentiation. NF-kB p50 and p52 double-

*Division of Life and Pharmaceutical Sciences, Department of Life Sciences, andResearch Center for Cellular Homeostasis, Ewha Womans University, Seoul 120-750,Korea; and †Department of Biomedical Engineering, College of Health Sciences,Institute of Medical Engineering, Yonsei University, Wonju 220-710, Korea

Received for publication September 7, 2012. Accepted for publication November 25,2012.

This work was supported by Bio & Medical Technology Development Program GrantM10642040002-07N4204-00210, Grant 2011-0018055, Grant 2012M3A9C5048708,and Scientific Research Center Program Grant 2012R1A5A1048236 (all from theNational Research Foundation of Korea), and funds from the Brain Korea 21 ScholarsProgram (to Hyeryeon Kim, S.H., Y.Y., and H.L.).

Address correspondence and reprint requests to Prof. Woojin Jeong, Ewha WomansUniversity, Science Building C, Room 207, 11-1 Daehyun-dong, Seodaemun-gu, Seoul120-750, Korea. E-mail address: jeongw@ewha.ac.kr

The online version of this article contains supplemental material.

Abbreviations used in this article: BMM, bone marrow–derived macrophage; COX-2,cyclooxygenase-2; mCT, micro-computed tomography; HA, hemagglutinin; IKK,IkB kinase; LC8, 8-kDa dynein L chain; MNC, multinucleated cell; OVX, ovariec-tomy; RANKL, receptor activator of NF-kB ligand; ROS, reactive oxygen species;Tg, transgenic; TRAF, TNFR-associated factor; TRAP, tartrate-resistant acid phos-phatase; TRP14, 14-kDa human thioredoxin-related protein.

Copyright� 2013 by The American Association of Immunologists, Inc. 0022-1767/13/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1202525

Published January 4, 2013, doi:10.4049/jimmunol.1202525 by guest on February 11, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

knockout mice (27) and IKKb-deficient mice (28) had severeosteopetrosis caused by the inhibition of osteoclast formation.In addition, dominant-negative IkBa proteins (29) and NF-kBinhibitors (30) were shown to block osteoclastogenesis. More-over, previous studies have shown that LC8 inhibits NF-kB ac-tivation by potent NF-kB stimulators, including TNF-a, LPS,and IL-1b (25, 26). Therefore, we sought to determine whetherLC8 inhibits RANKL-induced NF-kB activation and subsequentosteoclast differentiation and bone resorption using transgenic(Tg) mice with ectopic expression of LC8. An inhibitory roleof LC8 in osteoclast differentiation and bone resorption mightprovide the molecular basis of a new strategy for treating oste-oporosis and arthritic bone diseases.

Materials and MethodsReagents and Abs

Human recombinant RANKL and M-CSF proteins were purchased fromPeproTech and R&D Systems, respectively. A rabbit polyclonal Ab againstb-actin was purchased from Abcam. Rabbit polyclonal Ab against phos-pho-IkBa, phospho-JNK, phospho-ERK, and phospho-p38 was purchasedfrom Cell Signaling Technology. Rabbit polyclonal Abs against IkBa,JNK1, p38, c-Fos, and LC8, a goat polyclonal Ab against cyclooxygenase-2 (COX-2), and mouse mAbs against tubulin, ERK2, and NFATc1 werepurchased from Santa Cruz Biotechnology. LPS (from Escherichia coli0127:B8) was purchased from Sigma-Aldrich.

Construction of LC8 expression vectors

The DNA sequence encoding hemagglutinin (HA)-tagged LC8 was am-plified via PCR using the pCGN-LC8 vector (25, 26) as a template with aforward primer (59-CG GAA TTC ATG GCT TCT AGC TAT CCT TATGAC-39) containing both an EcoRI site (underlined) and the initiationcodon (bold) and a reverse primer (59-GG GGTACC TTA ACC AGATTTGAA CAG AAG AAT G-39) containing both a KpnI site (underlined) andthe stop codon (bold). The PCR product was purified, digested with re-striction enzymes EcoRI and KpnI, and cloned into the corresponding sitesof the pCBA-M vector. The resulting plasmid, pCBA-HA-LC8, was usedto generate Tg mice with overexpression of LC8.

Generation and genotyping of LC8 Tg mice

pCBA-HA-LC8 was digested with the restriction enzymes PvuI, SalI, andPstI, and the resulting 2.6-kb DNA fragment containing the LC8 gene underthe control of a chicken b-actin promoter and a CMVenhancer was purifiedfrom an agarose gel. The linear DNA fragment was injected into thepronuclei of fertilized eggs of C57BL/6 mice. The injected embryos wereimplanted into the uteri of pseudopregnant C57BL/6 mice to generate LC8Tg mice. The genotypes of all of the mice were determined via PCRanalysis using the tail DNA (500 ng) and the following primer pair: 59-CCT ACA GCT CCT GGG CAA CG-39 and 59-GAG CCA GGG CATTGG CCA CA-39. PCR was performed at 94˚C for 30 s, 62˚C for 30 s, and72˚C for 30 s for a total of 32 cycles, and the resulting products wereanalyzed via 1.5% agarose gel electrophoresis to identify the 455-bp DNA.

Preparation of bone marrow–derived macrophages

Bone marrow–derived macrophages (BMMs) were prepared as osteoclastprecursor cells from the femurs and tibiae of 4–8-wk-old C57BL/6 malemice as described below. Bone marrow cells were flushed from the bonemarrow cavity using anti-MEM containing 10% FBS, 100 U/ml penicillin,and 100 mg/ml streptomycin and harvested via centrifugation at 1000 rpmat room temperature. The cell pellets were then resuspended in the samemedium. After incubation at 37˚C for 1 d, nonadherent cells were har-vested and incubated in Gey’s solution for 10 min to remove RBCs. Afterclarification via centrifugation, the cells were cultured in anti-MEM con-taining 10% FBS, 100 U/ml penicillin, 100 mg/ml streptomycin, and 20 ng/ml recombinant human M-CSF. After 3 d, adherent cells were used asosteoclast precursor cells for osteoclastogenesis.

Osteoclast differentiation and TRAP staining

BMMs were loaded onto a 48-well plate and treated with 100 ng/ml RANKLin anti-MEM supplemented with 10% FBS and 20 ng/ml M-CSF for 3–5 d.After osteoclast differentiation, the cells were washed twice with 13 PBS,fixed for 10 min with 4% paraformaldehyde, and stained for TRAP using aleukocyte acid phosphatase cytochemistry kit (Sigma-Aldrich) according to

the manufacturer’s instructions. TRAP-positive multinucleated cells (TRAP+

MNCs) containing three or more nuclei were counted as osteoclasts undera light microscope.

Actin-ring staining

BMMs were fixed with 3.7% formaldehyde solution in PBS, permeabilizedwith 0.1% Triton X-100, and incubated with Alexa Fluor 488-phalloidin(Invitrogen) for 20 min. After being washed with PBS, the cells were in-cubated with DAPI (Roche) for 2 min and then photographed under afluorescence microscope.

Bone resorption assay

BMMs were plated onto dentine discs (Immunodiagnostic Systems) andtreated with 20 ng/mlM-CSF and 100 ng/ml RANKL for 7 d. The cells werecompletely removed from the dentine discs via abrasion with a cotton tip,and the dentine discs were stained with hematoxylin. Photographs of re-sorption pits were taken under a light microscope at 340 original mag-nification, and their area was measured via Image-Pro Plus 4.5 software(Media Cybernetics).

TRAP activity assay

BMMswere loaded onto a 12-well plate and lysed in 100 mM sodium acetatebuffer (pH5.2) containing 10mMsodium tartrate, 1%TritonX-100, 10mg/mlaprotinin, and 10 mg/ml leupeptin. After centrifugation, the protein concen-tration of the supernatant was determined using Bradford reagent, and TRAPactivity was assessed via a standard assay that uses para-nitrophenyl phos-phate as a substrate. The 120-ml reaction mixture containing 100 mM sodiumacetate (pH 5.2), 10 mM sodium tartrate, 1 mg/ml para-nitrophenyl phos-phate, and 2 mg total protein was incubated at 37˚C for 2 h, and the reactionwas stopped by adding 80 ml 1 N NaOH. The amount of para-nitrophenolliberated during the reaction was determined by measuring the absorbance at405 nm in a 96-well plate. TRAP activity was expressed as the specific TRAPactivity (A405/min/mg protein).

Calvarial injection and histologic analysis

The LC8 Tg mice were injected via a 28-gauge needle with 15 mg/kg bodyweight of LPS or PBS, both in a 40-ml volume, at a point on the middle ofthe skull in the space between the s.c. tissue and the periosteum. All animalstudy protocols were approved by the Animal Care Committee of the EwhaLaboratory Animal Genomics Center. The mice were killed in a CO2

chamber 7 d after the injection. The calvariae were dissected and fixedwith 4% paraformaldehyde at room temperature overnight. The specimenswere washed with PBS and decalcified with 0.5 M EDTA for 5 d. Then, thespecimens were embedded in low-melting paraffin, cut into 4-mm sections,and stained for TRAP.

Ovariectomy and micro-computed tomography imaging

Ovariectomy (OVX) and micro-computed tomography (mCT) were carriedout as previously described (31). Female 8-wk-old wild-type and LC8 Tgmice underwent either sham operation or OVX. At least four mice in eachgroup were examined. The mice were killed 4 wk after surgery, and thefemurs were evaluated using mCT.

Statistical analysis

Data are presented as the means 6 SD. Unless otherwise stated, all ex-periments were repeated at least two times, and the results from one rep-resentative experiment are reported. Statistical significance was calculatedusing the Student t test, and p values ,0.05 were considered statisticallysignificant.

ResultsLC8 inhibits RANKL-induced osteoclast differentiation

To assess the role of LC8 in RANKL-induced osteoclast differ-entiation, LC8 Tg mice were generated by injecting a linear DNAfragment containing the LC8 gene under the control of a chickenb-actin promoter and a CMVenhancer (Fig. 1A) into the pronucleiof fertilized eggs of C57BL/6 mice. Three lines (18, 79, and 83)of LC8 Tg mice were identified via genotype analysis of theoffspring of surrogate mothers implanted with injected embryos(Fig. 1B). HA-tagged LC8 was expressed in all of the tested tis-sues of the LC8 Tg mice (line 83) (Fig. 1C).

2 LC8 INHIBITS OSTEOCLAST DIFFERENTIATION AND BONE LOSS

by guest on February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

In BMMs of LC8 Tg mice line 83, HA-tagged LC8 wasexpressed at a higher rate than endogenous LC8 (Fig. 2A). WhenRANKL induced the differentiation of the BMMs to osteoclastsin the presence of M-CSF, the formation of TRAP+ MNCs wassignificantly lower in LC8-overexpressing BMMs than in wild-type BMMs (Fig. 2B, 2C). In addition, tests involving anotherLC8 Tg mouse line 18 yielded similar results (data not shown).These results suggest that LC8 inhibits RANKL-induced osteo-clast differentiation of primary osteoclast precursors.

LC8 inhibits RANKL-induced actin-ring formation and boneresorption

Mature osteoclasts have a sealing zone that consists of a ring ofF-actin that is required for bone resorption and have clear margins(32). Therefore, we next determined whether LC8 reduces actin-ring formation and bone resorption. Wild-type BMMs formedactin rings with clear margins in the presence of RANKL, butactin-ring formation was either reduced or absent from LC8-overexpressing BMMs (Fig. 3A). To determine whether the inhi-

bition of actin-ring formation by LC8 affects the resorption abilityof osteoclasts, we induced differentiation of BMMs into osteo-clasts in dentine discs in the presence of RANKL. Consistent withits inhibitory effect on actin-ring formation, LC8 overexpressionmarkedly reduced the formation of resorption pits (Fig. 3B). Theseresults suggest that LC8 inhibits actin-ring formation and boneresorption.

LC8 inhibits RANKL-induced NF-kB activation

To determine whether LC8 inhibits osteoclastogenesis by sup-pressing NF-kB activation, we assessed the effects of LC8 over-expression on RANKL-induced NF-kB activation in BMMs. Inresponse to potent NF-kB activators such as TNF-a, IkBa isphosphorylated at residues Ser32 and Ser36 by IKK and subse-quently degraded by the ubiquitin–proteasome system (33). Pre-

FIGURE 1. Generation of LC8 Tg mice. (A) Schematic representation

of the construct used to generate LC8 Tg mice. The 318-bp HA-tagged

LC8 cDNAs were cloned into EcoRI and KpnI sites downstream of the

chicken b-actin (CBA) promoter. Downstream, the 540-bp of the rabbit

b-globin (RBG) gene includes a polyadenylation signal. The primers used

in PCR analysis for screening Tg mice are indicated by small arrows. (B)

PCR analysis was carried out as described in Materials and Methods to

identify LC8 Tg mice. (C) Immunoblot analysis of tissue lysates prepared

from wild-type (2) and Tg (+, line 83) mice. Each tissue lysate (20 mg)

was subjected to immunoblot analysis using Abs against LC8 and tubulin.

FIGURE 2. Inhibition of RANKL-induced osteoclast differentiation of BMMs by LC8 overexpression. (A) Cell lysates (20 mg) of BMMs obtained from

wild-type (WT) or Tg mice were subjected to immunoblot analysis using Abs against LC8 and b-actin. (B) BMMs were cultured with M-CSF (20 ng/ml)

and RANKL (100 ng/ml) for the indicated times. The cells were fixed, subjected to TRAP staining, and observed under a light microscope. Scale bars, 200

mm. (C) The TRAP+ MNCs generated in (B) were counted. All values represent means 6 SD. n = 3. *p , 0.05, **p , 0.005.

FIGURE 3. Inhibition of RANKL-induced actin ring formation and

bone resorption by LC8 overexpression. (A) After BMMs were incubated

with RANKL (100 ng/ml) for 5 d in the presence of M-CSF (20 ng/ml), the

cells were fixed with 3.7% formaldehyde solution in PBS, permeabilized

with 0.1% Triton X-100, and incubated with Alexa Fluor 488-phalloidin

for 20 min. After being washed with PBS, the cells were incubated with

DAPI for 2 min and then photographed under a confocal microscope. Scale

bars, 50 mm. (B) BMMs on dentine discs were incubated with M-CSF

(20 ng/ml) and RANKL (100 ng/ml) for 7 d, after which the cells were

removed from the dentine discs, and resorption pits were visualized by

staining with hematoxylin. Data represent means 6 SD. n = 3. Scale bars,

200 mm. *p , 0.01.

The Journal of Immunology 3

by guest on February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

vious studies showed that LC8 blocked IkBa phosphorylationby IKK in TNF-a–treated cells (25). Therefore, we examined theeffect of LC8 on IkBa phosphorylation and degradation in BMMsstimulated with RANKL. The level of serine phosphorylation ofIkBa in LC8-overexpressing BMMs was considerably lower thanthat in wild-type BMMs, and this effect was accompanied by adecrease in the rate of IkBa degradation (Fig. 4A, 4B). We alsodetermined the effect of LC8 on the induction of the expression ofCOX-2, one of the NF-kB target genes. Consistently, LC8 over-expression markedly inhibited the induction of COX-2 expressionin RANKL-stimulated BMMs (Fig. 4C). These data indicate thatLC8 inhibits RANKL-induced NF-kB activation by blocking IkBaphosphorylation by IKK as it does in TNF-a signaling.

LC8 suppresses RANKL-induced c-Fos and NFATc1 expression

NF-kB has been known to be activated upstream of c-Fos/NFATc1during RANKL-induced osteoclast differentiation. Thus, we ex-amined the effect of LC8 overexpression on the expression ofc-Fos and NFATc1 in BMMs stimulated with RANKL. RANKL-induced c-Fos expression reached the maximum level after 8 hand was sustained until 24 h in wild-type BMMs, whereas it wasmarkedly decreased in LC8-overexpressing BMMs (Fig. 5A).RANKL robustly induced NFATc1 expression in wild-type BMMs,but this induction was almost abrogated in LC8-overexpressingBMMs (Fig. 5B). Also, a TRAP activity assay revealed that theexpression of TRAP, an NFATc1 target gene, was considerablylower in LC8-overexpressing BMMs than in wild-type BMMs(Fig. 5C).

LC8 inhibits RANKL-induced JNK and ERK activation

RANKL also induces the activation of MAPKs including JNK,p38, and ERK, which are known to play crucial roles in osteo-clastogenesis (34). Therefore, we sought to determine whetherRANKL-induced activation of these MAPKs was affected by LC8overexpression. MAPKs are activated by dual phosphorylation ofthreonine and tyrosine residues (Thr183 and Tyr185 for JNK; Thr180

and Tyr182 for p38; and Thr202 and Tyr204 for ERK) within a Thr-X-Tyr motif. Immunoblot analysis with Abs against the duallyphosphorylated proteins revealed that the level of RANKL-induced JNK and ERK phosphorylation in LC8-overexpressingBMMs was significantly lower than that in wild-type cells (Fig.6). However, the extent of p38 phosphorylation was only mini-mally affected by LC8 overexpression.

LC8 inhibits inflammation- and OVX-induced bone loss

To assess the possibility of using LC8 to treat bone diseases thatresult from excessive osteoclastic activity, we injected LPS intothe calvariae of LC8 Tg mice and examined the effects of LC8 onpathologic osteoclast formation and bone erosion during inflam-mation. The extent of both bone erosion and the formation ofTRAP+ MNCs was significantly lower in LC8 Tg mice than inwild-type mice (Fig. 7A, 7B), which suggests that LC8 inhibitsinflammation-induced bone destruction.In addition, LC8 Tg mice were subjected to an OVX-induced

model of postmenopausal osteoporosis. The bone volume, tra-becular bone number, and bone mineral density were significantlyreduced by OVX in wild-type mice, but such a reduction wasobserved to a much lesser extent in LC8 Tg mice (Fig. 7C, 7D),suggesting that LC8 inhibits pathologic bone destruction.

FIGURE 4. Inhibition of RANKL-induced NF-kB activation by LC8

overexpression. BMMs were exposed to RANKL (100 ng/ml) in the pres-

ence of M-CSF (20 ng/ml) for the indicated times. (A) Cell lysates were

subjected to immunoblot analysis with Abs against phospho-IkBa, IkBa,

LC8, or b-actin. (B) The chemiluminescence signals for phospho-IkBa and

IkBawere quantified and normalized according to those of b-actin. (C) The

abundance of COX-2, LC8, and b-actin in cell lysates was determined via

immunoblot analysis using their specific Abs.

FIGURE 5. Suppression of RANKL-induced expression of c-Fos and

NFATc-1 by LC8 overexpression. BMMs were incubated with M-CSF (20

ng/ml) and RANKL (100 ng/ml) for the indicated times. The abundance of

c-Fos (A) and NFATc-1 (B) in cell lysates was determined via immunoblot

analysis using their specific Abs. (C) The sp. act. of TRAP was determined

as described in Materials and Methods. All values represent means 6 SD.

n = 3. *p , 0.05, **p , 0.005.

4 LC8 INHIBITS OSTEOCLAST DIFFERENTIATION AND BONE LOSS

by guest on February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

DiscussionIn this study, we showed that LC8 inhibits osteoclast differentia-tion and bone resorption of primary osteoclast precursor cells afterRANKL stimulation and that it prevents inflammation- and OVX-induced bone destruction in mice. During the RANKL-inducedosteoclast differentiation, LC8 inhibited the phosphorylation andsubsequent degradation of IkBa and, consequently, NF-kB acti-vation. Therefore, LC8 likely regulates RANKL-induced NF-kB

activation via the same mechanism that it does in TNF-a signaling(25, 26): it inhibits NF-kB activation by interacting with IkBa andconsequently preventing IkBa phosphorylation by IKK.The NF-kB pathway can contribute to RANKL-induced oste-

oclast differentiation via the expression of NF-kB–dependentgenes such as COX-2, c-Fos, and NFATc1. NF-kB–dependentCOX-2 expression plays a key role in RANKL-induced osteoclastdifferentiation by inducing PGE2 production (35). The RANKL-responsive region of the COX-2 promoter contains an NF-kB site.Blockade of COX-2 by celecoxib inhibits osteoclast differentia-tion of osteoclast precursor cells, and this inhibition can be res-cued by adding exogenous PGE2.NF-kB induces the initial expression of NFATc1, which is a

master regulator of osteoclastogenesis both in vitro and in vivo(9). c-Fos, a major component of transcription factor AP-1, isinvolved in the robust induction of NFATc1 expression (11).Mice lacking c-Fos developed osteopetrosis caused by the inhi-bition of osteoclast formation (36). RANKL did not induce c-Fosor NFATc1 expression in NF-kB p50/p52 double-knockout os-teoclast precursor cells (37), indicating that NF-kB is activatedupstream of c-Fos/NFATc1 during RANKL-mediated osteoclastdifferentiation. Because no consensus NF-kB binding sites arelocated on the c-Fos promoter, NF-kB seems to indirectly acti-vate c-Fos, as previously described in a study of embryonicfibroblasts (38).NFATc1 rescued RANKL-induced osteoclastogenesis in osteo-

clast precursor cells lacking c-Fos (13), indicating that NFATc1is activated downstream of c-Fos. Collectively, RANKL-inducedosteoclastogenesis may be controlled by a hierarchical transcrip-tional regulation in which NF-kB activation is the first event andleads to c-Fos activation followed by robust induction of NFATc1.In the current study, LC8 profoundly suppressed not only the ex-pression of COX-2, an NF-kB target gene, but also the induction ofthe expression of c-Fos and NFATc1 during RANKL-induced os-teoclast differentiation. Therefore, LC8 likely prevents osteoclastformation in the early stages of osteoclast precursor differentiationby inhibiting the activation of the NF-kB pathway, although LC8might directly regulate c-Fos and NFATc1.

FIGURE 6. Inhibition of RANKL-induced MAPK activation by LC8

overexpression. (A) BMMs were incubated with M-CSF (20 ng/ml) and

RANKL (100 ng/ml) for the indicated times, after which cell lysates were

subjected to immunoblot analysis with Abs against phospho-JNK and

JNK1, phospho-ERK and ERK2, or phospho-p38 and p38 (top and bottom

of each pair of images, respectively). (B) The chemiluminescence signals

for phospho-JNK, phospho-ERK, and phospho-p38 were quantified and

normalized according to those of JNK1, ERK2, and p38, respectively.

FIGURE 7. Inhibition of LPS- and OVX-induced bone destruction by LC8 overexpression. (A) LPS (15 mg/kg body weight) or PBS, each in a 40-ml

volume, was injected into the space between the s.c. tissue and the periosteum of the skulls of mice. Seven days after the injection, the mouse calvariae were

fixed, decalcified, embedded in paraffin, and stained with TRAP. Scale bars, 100 mm. (B) Fold differences of bone cavity (left panel) and the number of

TRAP+ osteoclast (right panel) were quantified. Data represent mean 6 SD. n = 3. *p , 0.01, **p , 0.005. (C) Wild-type (WT) and LC8 Tg mice

underwent either sham operation or OVX. Four weeks after surgery, the femurs were examined via mCT, and their three-dimensional reconstructions are

shown. (D) Histograms represent the three-dimensional trabecular structural parameters in the femurs: bone volume fraction (BV/TV), trabecular number

(Tb.N), and bone mineral density (BMD). Data represent mean 6 SD. n = 4. *p , 0.05.

The Journal of Immunology 5

by guest on February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

Given that RANKL stimulates not only the NF-kB pathway butalso the MAPK pathway, including JNK, ERK, and p38 (39), wesought to examine whether LC8 regulates RANKL-inducedMAPK activation. Though a previous report has shown thatcAMP inhibits p38 activation via CREB-induced LC8, whichinterferes with association between MAPK kinase 3/6 and p38(40), LC8 marginally inhibited p38 activation during osteoclastdifferentiation, whereas it significantly inhibited activation of JNKand ERK. In the MAPK pathway, activated JNK and ERK candirectly phosphorylate c-Jun and c-Fos, respectively (41), whichare major components of transcription factor AP-1, suggestingthat LC8 may also contribute to osteoclastogenesis by regulatingAP-1 activity. LC8 has known to bind to a conserved (K/R)XTQTamino acid sequence motif in a wide variety of target proteins(42). However, the failure to find a (K/R)XTQT motif in JNK1/2/3and ERK1/2, and their upstream kinases MAPK kinase 4/7 andMEK1/2, suggest that LC8 may indirectly regulate JNK and ERKactivities.Increasing evidence suggests that reactive oxygen species (ROS)

regulate RANKL-induced osteoclast differentiation. ROS gener-ation transiently increases in RANKL-stimulated osteoclast pre-cursor cells via a signaling cascade involving TRAF6, Rac1, andNADPH oxidase 1 (43). The antioxidants N-acetylcysteine andascorbate, as well as diphenylene iodonium, a NADPH oxidaseinhibitor, inhibited RANKL-induced osteoclastogenesis by sup-pressing ROS generation (43, 44). However, the mechanism bywhich ROS regulate RANKL-induced osteoclast differentiationremains unknown. We recently identified the mechanism under-lying the redox-dependent regulation of TNF-a–induced NF-kBactivation, in which ROS oxidize LC8 to a homodimer linked witha reversible intermolecular disulfide bond that promotes the dis-sociation of LC8 from IkBa and allows IkBa to be phosphory-lated and degraded, thereby resulting in NF-kB activation (25, 26).Therefore, LC8 is likely oxidized by RANKL-induced ROS as itis in TNF-a signaling and is thereby involved in the ROS-depen-dent regulation of osteoclast differentiation. LC8 was previouslyidentified as a substrate of TRP14, a disulfide reductase (24).TRP14 is involved in the redox-dependent regulation of NF-kB byfacilitating the reduction of LC8 (25, 26). Thus, future researchshould focus on elucidating the role of TRP14 in RANKL-inducedosteoclast differentiation.Cytoplasmic dynein is a large multicomponent microtubule-

based motor coupled to the dynactin cofactor complex (45). Mi-crotubule organization, intracellular vesicular trafficking, and theactivity of dynein motor are critical for the activation and functionof osteoclasts (46–48). LIS1 was recently identified as a microtu-bule regulator that interacts with both Plekhm1, a component oflate endosomal trafficking, and the dynein–dynactin motor com-plex in osteoclasts (47). Depletion of LIS1 dramatically inhibitedthe formation and bone resorption function of osteoclasts by in-terfering with dynein function and microtubule organization (47).Exogenous expression of dynamitin, a component of the dynactincomplex, disrupted the dynein–dynactin complex, leading to re-tardation of osteoclast formation (48). Therefore, the role of LC8in osteoclast differentiation as an essential component of the dy-nein motor complex may be worthy to be studied.Because a defect in the development of peripheral lymph nodes

has been reported in mice deficient in RANKL (49) or RANK (50),it was investigated whether the development of lymph nodes isimpaired in LC8 Tg mice. LC8 overexpression had no apparenteffect on the formation of mesenteric lymph nodes (SupplementalFig. 1A). Furthermore, TNF-a–induced apoptosis did not occur inBMM cells of LC8 Tg mice (Supplemental Fig. 1B), indicatingthat the apoptosis is not caused by the inhibition of NF-kB via

LC8. LC8 Tg mice exhibited mild osteopetrotic phenotype andwere resistant to OVX treatment. However, the effect of LC8 onbone phenotype could not be restricted to osteoclasts, because ofits systemic expression. Thus, further study on the role of LC8 inosteoblasts may also be necessary to provide clinical implicationsfor various bone diseases.In conclusion, we showed that LC8 inhibits RANKL-induced

osteoclast differentiation and prevents inflammation- and OVX-induced bone loss. In addition, we elucidated the molecularmechanisms by which LC8 inhibits NF-kB and MAPK activa-tions, leading to attenuation of c-Fos and NFATc1 expression.These findings might lead to the discovery of novel approachesfor preventing and treating various types of inflammatory bonedestruction and postmenopausal osteoporosis.

DisclosuresThe authors have no financial conflicts of interest.

References1. Manolagas, S. C. 2000. Birth and death of bone cells: basic regulatory mecha-

nisms and implications for the pathogenesis and treatment of osteoporosis.Endocr. Rev. 21: 115–137.

2. Rodan, G. A., and T. J. Martin. 2000. Therapeutic approaches to bone diseases.Science 289: 1508–1514.

3. 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, andL. Galibert. 1997. A homologue of the TNF receptor and its ligand enhance T-cell growth and dendritic-cell function. Nature 390: 175–179.

4. Wong, B. R., J. Rho, J. Arron, E. Robinson, J. Orlinick, M. Chao, S. Kalachikov,E. Cayani, F. S. Bartlett, III, W. N. Frankel, et al. 1997. TRANCE is a novelligand of the tumor necrosis factor receptor family that activates c-Jun N-terminal kinase in T cells. J. Biol. Chem. 272: 25190–25194.

5. Yasuda, H., N. Shima, N. Nakagawa, K. Yamaguchi, M. Kinosaki, S. Mochizuki,A. Tomoyasu, K. Yano, M. Goto, A. Murakami, et al. 1998. Osteoclast differ-entiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitoryfactor and is identical to TRANCE/RANKL. Proc. Natl. Acad. Sci. USA 95:3597–3602.

6. 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 93:165–176.

7. Theill, L. E., W. J. Boyle, and J. M. Penninger. 2002. RANK-L and RANK:T cells, bone loss, and mammalian evolution. Annu. Rev. Immunol. 20: 795–823.

8. Ross, F. P., and S. L. Teitelbaum. 2005. alphavbeta3 and macrophage colony-stimulating factor: partners in osteoclast biology. Immunol. Rev. 208: 88–105.

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

10. Takayanagi, H. 2007. Osteoimmunology: shared mechanisms and crosstalk be-tween the immune and bone systems. Nat. Rev. Immunol. 7: 292–304.

11. Wagner, E. F., and R. Eferl. 2005. Fos/AP-1 proteins in bone and the immunesystem. Immunol. Rev. 208: 126–140.

12. Grigoriadis, A. E., Z. Q. Wang, M. G. Cecchini, W. Hofstetter, R. Felix,H. A. Fleisch, and E. F. Wagner. 1994. c-Fos: a key regulator of osteoclast-macrophage lineage determination and bone remodeling. Science 266: 443–448.

13. Matsuo, K., D. L. Galson, C. Zhao, L. Peng, C. Laplace, K. Z. Wang,M. A. Bachler, H. Amano, H. Aburatani, H. Ishikawa, and E. F. Wagner. 2004.Nuclear factor of activated T-cells (NFAT) rescues osteoclastogenesis in pre-cursors lacking c-Fos. J. Biol. Chem. 279: 26475–26480.

14. Dick, T., K. Ray, H. K. Salz, and W. Chia. 1996. Cytoplasmic dynein (ddlc1)mutations cause morphogenetic defects and apoptotic cell death in Drosophilamelanogaster. Mol. Cell. Biol. 16: 1966–1977.

15. Phillis, R., D. Statton, P. Caruccio, and R. K. Murphey. 1996. Mutations in the 8kDa dynein light chain gene disrupt sensory axon projections in the Drosophilaimaginal CNS. Development 122: 2955–2963.

16. Fan, J., Q. Zhang, H. Tochio, M. Li, and M. Zhang. 2001. Structural basis ofdiverse sequence-dependent target recognition by the 8 kDa dynein light chain.J. Mol. Biol. 306: 97–108.

17. Jaffrey, S. R., and S. H. Snyder. 1996. PIN: an associated protein inhibitor ofneuronal nitric oxide synthase. Science 274: 774–777.

18. Crepieux, P., H. Kwon, N. Leclerc, W. Spencer, S. Richard, R. Lin, and J. Hiscott.1997. I kappaB alpha physically interacts with a cytoskeleton-associated proteinthrough its signal response domain. Mol. Cell. Biol. 17: 7375–7385.

19. Puthalakath, H., D. C. Huang, L. A. O’Reilly, S. M. King, and A. Strasser. 1999.The proapoptotic activity of the Bcl-2 family member Bim is regulated by in-teraction with the dynein motor complex. Mol. Cell 3: 287–296.

20. Puthalakath, H., A. Villunger, L. A. O’Reilly, J. G. Beaumont, L. Coultas,R. E. Cheney, D. C. Huang, and A. Strasser. 2001. Bmf: a proapoptotic BH3-only protein regulated by interaction with the myosin V actin motor complex,activated by anoikis. Science 293: 1829–1832.

6 LC8 INHIBITS OSTEOCLAST DIFFERENTIATION AND BONE LOSS

by guest on February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

21. Lu, J., Q. Sun, X. Chen, H. Wang, Y. Hu, and J. Gu. 2005. Identification ofdynein light chain 2 as an interaction partner of p21-activated kinase 1. Biochem.Biophys. Res. Commun. 331: 153–158.

22. Vadlamudi, R. K., R. Bagheri-Yarmand, Z. Yang, S. Balasenthil, D. Nguyen,A. A. Sahin, P. den Hollander, and R. Kumar. 2004. Dynein light chain 1, a p21-activated kinase 1-interacting substrate, promotes cancerous phenotypes. CancerCell 5: 575–585.

23. Lo, K. W., H. M. Kan, L. N. Chan, W. G. Xu, K. P. Wang, Z. Wu, M. Sheng, andM. Zhang. 2005. The 8-kDa dynein light chain binds to p53-binding protein 1and mediates DNA damage-induced p53 nuclear accumulation. J. Biol. Chem.280: 8172–8179.

24. Jeong, W., T. S. Chang, E. S. Boja, H. M. Fales, and S. G. Rhee. 2004. Roles ofTRP14, a thioredoxin-related protein in tumor necrosis factor-alpha signalingpathways. J. Biol. Chem. 279: 3151–3159.

25. Jung, Y., H. Kim, S. H. Min, S. G. Rhee, and W. Jeong. 2008. Dynein light chainLC8 negatively regulates NF-kappaB through the redox-dependent interactionwith IkappaBalpha. J. Biol. Chem. 283: 23863–23871.

26. Jeong, W., Y. Jung, H. Kim, S. J. Park, and S. G. Rhee. 2009. Thioredoxin-related protein 14, a new member of the thioredoxin family with disulfide re-ductase activity: implication in the redox regulation of TNF-alpha signaling.Free Radic. Biol. Med. 47: 1294–1303.

27. Iotsova, V., J. Caamano, J. Loy, Y. Yang, A. Lewin, and R. Bravo. 1997. Osteo-petrosis in mice lacking NF-kappaB1 and NF-kappaB2. Nat. Med. 3: 1285–1289.

28. Ruocco, M. G., S. Maeda, J. M. Park, T. Lawrence, L. C. Hsu, Y. Cao, G. Schett,E. F. Wagner, and M. Karin. 2005. IkappaB kinase (IKK)beta, but not IKKalpha,is a critical mediator of osteoclast survival and is required for inflammation-induced bone loss. J. Exp. Med. 201: 1677–1687.

29. Abu-Amer, Y., S. F. Dowdy, F. P. Ross, J. C. Clohisy, and S. L. Teitelbaum. 2001.TAT fusion proteins containing tyrosine 42-deleted IkappaBalpha arrest osteo-clastogenesis. J. Biol. Chem. 276: 30499–30503.

30. Alles, N., N. S. Soysa, J. Hayashi, M. Khan, A. Shimoda, H. Shimokawa,O. Ritzeler, K. Akiyoshi, K. Aoki, and K. Ohya. 2010. Suppression of NF-kappaB increases bone formation and ameliorates osteopenia in ovariecto-mized mice. Endocrinology 151: 4626–4634.

31. Lee, S. K., J. F. Kalinowski, C. Jacquin, D. J. Adams, G. Gronowicz, andJ. A. Lorenzo. 2006. Interleukin-7 influences osteoclast function in vivo but is nota critical factor in ovariectomy-induced bone loss. J. Bone Miner. Res. 21: 695–702.

32. Teitelbaum, S. L. 2000. Bone resorption by osteoclasts. Science 289: 1504–1508.33. Hayden, M. S., and S. Ghosh. 2004. Signaling to NF-kappaB. Genes Dev. 18:

2195–2224.34. Lee, Z. H., and H. H. Kim. 2003. Signal transduction by receptor activator of nuclear

factor kappa B in osteoclasts. Biochem. Biophys. Res. Commun. 305: 211–214.35. Han, S. Y., N. K. Lee, K. H. Kim, I. W. Jang, M. Yim, J. H. Kim, W. J. Lee, and

S. Y. Lee. 2005. Transcriptional induction of cyclooxygenase-2 in osteoclast pre-cursors is involved in RANKL-induced osteoclastogenesis. Blood 106: 1240–1245.

36. Wang, Z. Q., C. Ovitt, A. E. Grigoriadis, U. Mohle-Steinlein, U. Ruther, andE. F. Wagner. 1992. Bone and haematopoietic defects in mice lacking c-fos.Nature 360: 741–745.

37. Yamashita, T., Z. Yao, F. Li, Q. Zhang, I. R. Badell, E. M. Schwarz, S. Takeshita,E. F. Wagner, M. Noda, K. Matsuo, et al. 2007. NF-kappaB p50 and p52 regulatereceptor activator of NF-kappaB ligand (RANKL) and tumor necrosis factor-induced osteoclast precursor differentiation by activating c-Fos and NFATc1. J.Biol. Chem. 282: 18245–18253.

38. Fujioka, S., J. Niu, C. Schmidt, G. M. Sclabas, B. Peng, T. Uwagawa, Z. Li,D. B. Evans, J. L. Abbruzzese, and P. J. Chiao. 2004. NF-kappaB and AP-1connection: mechanism of NF-kappaB-dependent regulation of AP-1 activity.Mol. Cell. Biol. 24: 7806–7819.

39. Boyle, W. J., W. S. Simonet, and D. L. Lacey. 2003. Osteoclast differentiationand activation. Nature 423: 337–342.

40. Zhang, J., T. N. Bui, J. Xiang, and A. Lin. 2006. Cyclic AMP inhibits p38 ac-tivation via CREB-induced dynein light chain. Mol. Cell. Biol. 26: 1223–1234.

41. Miyazaki, T., H. Katagiri, Y. Kanegae, H. Takayanagi, Y. Sawada, A. Yamamoto,M. P. Pando, T. Asano, I. M. Verma, H. Oda, et al. 2000. Reciprocal role of ERKand NF-kappaB pathways in survival and activation of osteoclasts. J. Cell Biol.148: 333–342.

42. Lo, K. W., S. Naisbitt, J. S. Fan, M. Sheng, and M. Zhang. 2001. The 8-kDadynein light chain binds to its targets via a conserved (K/R)XTQT motif. J. Biol.Chem. 276: 14059–14066.

43. Lee, N. K., Y. G. Choi, J. Y. Baik, S. Y. Han, D. W. Jeong, Y. S. Bae, N. Kim, andS. Y. Lee. 2005. A crucial role for reactive oxygen species in RANKL-inducedosteoclast differentiation. Blood 106: 852–859.

44. Lean, J. M., J. T. Davies, K. Fuller, C. J. Jagger, B. Kirstein, G. A. Partington,Z. L. Urry, and T. J. Chambers. 2003. A crucial role for thiol antioxidants inestrogen-deficiency bone loss. J. Clin. Invest. 112: 915–923.

45. Vale, R. D. 2003. The molecular motor toolbox for intracellular transport. Cell112: 467–480.

46. Van Wesenbeeck, L., P. R. Odgren, F. P. Coxon, A. Frattini, P. Moens, B. Perdu,C. A. MacKay, E. Van Hul, J. P. Timmermans, F. Vanhoenacker, et al. 2007.Involvement of PLEKHM1 in osteoclastic vesicular transport and osteopetrosisin incisors absent rats and humans. J. Clin. Invest. 117: 919–930.

47. Ye, S., T. W. Fowler, N. J. Pavlos, P. Y. Ng, K. Liang, Y. Feng, M. Zheng,R. Kurten, S. C. Manolagas, and H. Zhao. 2011. LIS1 regulates osteoclast for-mation and function through its interactions with dynein/dynactin and Plekhm1.PLoS ONE 6: e27285.

48. Ng, P. Y., T. S. Cheng, H. Zhao, S. Ye, E. S. Ang, E. C. Khor, H. T. Feng, J. Xu,M. H. Zheng, and N. J. Pavlos. 2012. Disruption of the dynein-dynactin complexunveils motor specific functions in osteoclast formation and bone resorption. J.Bone Miner. Res. DOI: 10.1002/jbmr.1725.

49. Kong, Y. Y., H. Yoshida, I. Sarosi, H. L. Tan, E. Timms, C. Capparelli,S. Morony, A. J. Oliveira-dos-Santos, G. Van, A. Itie, et al. 1999. OPGL is a keyregulator of osteoclastogenesis, lymphocyte development and lymph-node or-ganogenesis. Nature 397: 315–323.

50. Dougall, W. C., M. Glaccum, K. Charrier, K. Rohrbach, K. Brasel, T. De Smedt,E. Daro, J. Smith, M. E. Tometsko, C. R. Maliszewski, et al. 1999. RANK isessential for osteoclast and lymph node development. Genes Dev. 13: 2412–2424.

The Journal of Immunology 7

by guest on February 11, 2018http://w

ww

.jimm

unol.org/D

ownloaded from