Bone 58 (2014) 17–25

Contents lists available at ScienceDirect

Bone

j ourna l homepage: www.e lsev ie r .com/ locate /bone

Original Full Length Article

Low-intensity pulsed ultrasound (LIPUS) inhibits LPS-inducedinflammatory responses of osteoblasts throughTLR4–MyD88 dissociation

Juna Nakao, Yasuyuki Fujii, Joji Kusuyama, Kenjiro Bandow, Kyoko Kakimoto,Tomokazu Ohnishi, Tetsuya Matsuguchi ⁎Department of Oral Biochemistry, Field of Developmental Medicine, Kagoshima University Graduate School of Medical and Dental Sciences, 8-35-1 Sakuragaoka, Kagoshima 890-8544, Japan

⁎ Corresponding author. Fax: +81 99 275 6138.E-mail address: tmatsugu@dent.kagoshima-u.ac.jp (T.

8756-3282/$ – see front matter © 2013 Elsevier Inc. All rihttp://dx.doi.org/10.1016/j.bone.2013.09.018

a b s t r a c t

a r t i c l e i n f o

Article history:Received 3 August 2013Revised 12 September 2013Accepted 16 September 2013Available online 30 September 2013

Edited by: Michael Amling

Keywords:ChemokinesLipopolysaccharide (LPS)MechanotransductionOsteoblastsToll-like receptors (TLR)

Previous reports have shown that osteoblasts are mechano-sensitive. Low-intensity pulsed ultrasound (LIPUS)induces osteoblast differentiation and is an established therapy for bone fracture. Here we have examined howLIPUS affects inflammatory responses of osteoblasts to LPS. LPS rapidly induced mRNA expression of severalchemokines including CCL2, CXCL1, and CXCL10 in both mouse osteoblast cell line and calvaria-derived osteo-blasts. Simultaneous treatment by LIPUS significantly inhibited mRNA induction of CXCL1 and CXCL10 by LPS.LPS-induced phosphorylation of ERKs, p38 kinases, MEK1/2, MKK3/6, IKKs, TBK1, and Akt was decreased inLIPUS-treated osteoblasts. Furthermore, LIPUS inhibited the transcriptional activation of NF-κB responsive ele-ment and Interferon-sensitive response element (ISRE) by LPS. In a transient transfection experiment, LIPUS sig-nificantly inhibited TLR4–MyD88 complex formation. Thus LIPUS exerts anti-inflammatory effects on LPS-stimulated osteoblasts by inhibiting TLR4 signal transduction.

© 2013 Elsevier Inc. All rights reserved.

Introduction

Low-intensity pulsed ultrasound (LIPUS) is an established therapyfor bone fracture repair due to its ability to accelerate healing of freshfractures and clinically established non-unions [1]. The fracture-healing activities of LIPUS are associated with its effects on osteoblasts.For example, LIPUS induces the differentiation of rat long bone-derived osteoblastic cells, as evidenced by increased osteocalcin expres-sion [2]. It has also been reported that LIPUS increases calcium accumu-lation, but not type I collagen synthesis, by the mouse osteoblastic cellline, MC3T3-E1 [3]. Thus, osteoblasts in bone tissue are considered asa mechano-sensitive cell type capable of responding to LIPUS.

Osteoblasts are not only simple bone-forming cells but also impor-tant regulators of bone remodeling through the production of polypep-tidic factors including receptor activator of nuclear factor kappa B ligand(RANKL) and macrophage colony-stimulating factor (M-CSF), both of

Matsuguchi).

ghts reserved.

which are potent inducers of osteoclast differentiation and activation[4]. Furthermore, recent lines of evidence have revealed an importantrole for osteoblasts as regulators of local inflammatory responses. Sever-al laboratories including ours reported that osteoblasts express severalToll-like receptor (TLR) family members (TLR2, TLR3, TLR4, TLR6, andTLR9) at relatively high levels recognizing pathogen-associatedmolecu-lar patterns (PAMPs) [5–8]. Consistently, it has previously been report-ed that mouse and human osteoblasts respond to Salmonella infectionby producing a CXC chemokine, CXCL10 (IP-10), through TLR4 signals[6]. Another report showed that human osteoblasts produce a CC che-mokine, CCL2 (MCP-1) in response to LPS stimulation [9]. Thus, in thepresence of bacterial infection, osteoblasts are by themselves capableof inducing bone tissue inflammation by secreting some chemokineswhich recruit inflammatory cells.

As osteoblasts are mechano-sensitive [10], it seems reasonableto presume that mechanical stresses such as LIPUS may affect pro-inflammatory responses of osteoblasts. However, this possibility hasnot previously been elucidated well. Here, we have examined chemo-kine induction profiles of osteoblasts in response to LPS, and analyzedhow they are affected by the simultaneous LIPUS treatment. Our resultshave revealed novel anti-inflammatory effects of LIPUS on osteoblasts,indicating the possibility that LIPUS may be useful in the treatment ofbone infectious diseases.

18 J. Nakao et al. / Bone 58 (2014) 17–25

Materials and methods

Animals and cells

C57BL/6 mice were obtained from Clea Japan (Tokyo, Japan). Prima-ry osteoblasts were isolated from newborn C57BL/6 mouse calvaria aspreviously described [11]. The animals were maintained in accordancewith protocols approved by the Animal Care and Use Committee atKagoshima University, Japan. MC3T3-E1 cells were obtained from RIKENCell Bank (Tsukuba, Japan), and maintained in Eagle's α-minimal essen-tial medium (αMEM) (Sigma-Aldrich, Inc., St. Louis, MO) containing10% fetal bovine serum (FBS), 100 units/ml penicillin, and 100 μg/mlstreptomycin. AD293 cell line (a subline of HEK293) was purchasedfrom Agilent Technologies, Inc. (Santa Clara, CA) and maintained inD-MEM (Sigma-Aldrich) containing 10% FBS, 100 units/ml penicillin,and 100 μg/ml streptomycin.

Antibodies and reagents

Phospho-specific antibodies against JNKs, ERKs, p38 kinases,MEK1/2,MKK3/6, IKKα/β, TBK1, Akt (S473), IRF3, as well as antibodies againstJNKs, ERKs, IκBα were all purchased from Cell Signaling Technology(Danvers, MA). Antibodies against β-actin, p38 kinases, Myc tag, andHA tag were purchased from Santa Cruz Biotechnology (Santa Cruz,CA). The anti-Glyceraldehydes-3-phosphate dehydrogenase (GAPDH)antibody was purchased from Imgenex Corporation (San Diego, CA).The anti-Flag M2 monoclonal antibody and LPS from Escherichia coliO55:B5 were purchased from Sigma.

Plasmids

The coding cDNAs for mouse TLR4, myeloid differentiation factor 88(MyD88), and MD2 were amplified by RT-PCR with specific primersusing total RNA fromC57BL/6mouse peritonealmacrophages as the tem-plate. After the confirmation of the DNA sequences, these cDNAs werecloned into pcDNA4/Myc-His A (Invitrogen, Carlbad, CA), pcDNA3-HA[12], and pEFBOS-Flag [12] vector, producing pcDNA4-mTLR4/Myc,pcDNA3-HA-mMyD88, and pEFBOS-Flag-mMD2, respectively.

Application of LIPUS

Cells were stimulated using a LIPUS-generating system (TeijinPharma Ltd., Tokyo, Japan), which was previously described [13]. TheLIPUS signal consisted of a series of 1.5-MHz, 200 μs burst sine wavesat 1.0 kHz, andwasdelivered at an intensity of 30 mW/cm2. The patternand intensity of the LIPUS signal employed in this studywere essentiallythe same as that used in clinical practice [14] and in animal model ex-periments [15].

Western blot analysis

Total cellular lysateswere prepared in RIPA cell lysis buffer (150 mMNaCl, 5 mMEDTA, 1% Deoxycholate, 0.1% SDS, 1% Triton-X)with the ad-dition of 1 mMNa3VO4 and 1× Proteinase Inhibitor Cocktail (FunakoshiCo., Ltd., Tokyo, Japan). Immunoblotting was performed as previouslydescribed [11].

Real-time PCR analyses

Total RNA was isolated from cells using Isogen II (Nippon Gene Co.,Ltd., Tokyo, Japan) and reverse transcription was performed usingRiverTra Ace (Toyobo, Tokyo, Japan) according to themanufacturers' in-structions. Real-time PCRwas conducted using StepOnePlus™ (AppliedBiosystems, Foster City, CA). Briefly, the cDNA synthesized from 0.05 μgof total RNAwas amplified in a volume of 20 μl with 0.11× SYBRGreen I(CAMBREX, Rockland, ME, USA), 0.2 mM/each of dNTPs, 0.5 μM/each of

a pair of primers, and 0.5 unit Blend Taq® Plus DNA polymerase(Toyobo Co., Ltd., Osaka, Japan) under the following condition: 95 °Cfor 5 min, followed by 55 PCR cycles at 95 °C for 30 s, 60 °C for 20 s,and 72 °C for 40 s. Fluorescent signals were measured in real time,and then each sample was quantified according to the manufacturer'sinstructions. The utilized mouse chemokine primer sequences werelisted in Table 1. The primer sequences for RANKL were 5′-TTGCACACCTCACCATCAAT-3′ (sense) and 5′-TAGGTACGCTTCCCGATGTT-3′(anti-sense). The sequences of the PCR products were checked with aDNA sequencer (Prism 310, Applied Biosystems, Foster City, CA). Inorder to normalize for differences in the amount of total RNA added toeach reaction, GAPDHwas used as the endogenous control. An arbitraryunit was determined by dividing the concentration of each PCR productby the concentration of the GAPDH PCR product.

Luciferase assays

For stable transfection, 8 μg of luciferase reporter plasmid, pNF-κB-Luc or pISRE-Luc (Clontech, Mountain View, CA) was transfected intosub-confluent MC3T3-E1 cells in a 60 mm dish with the addition of0.16 μg of a G418-resistance plasmid, pcDNA3.1(+), using HilyMax™(Dojindo, Kumamoto, Japan) according to the manufacturer's instruc-tions. At 36 h after the transfection, cells were selected with the addi-tion of 1 mg/ml of G418 (Sigma). Twelve independent G418-resistantclones were isolated for each reporter plasmid and screened for consti-tutive luciferase activities. Successful stable transfectants were used forfurther assays. For luciferase assays, MC3T3-E1 cells stably transfectedwith a luciferase reporter plasmid were cultured in 6 well plates tothe sub-confluent state. After the proper stimulation, cells were lysedfor luciferase activity measurements using the ONE-Glo™ luciferaseassay system (Promega, Madison, WI) according to the manufacturer'sinstructions.

Flowcytometric analysis

Cells were stained with PE-conjugated MTS510 (recognizing mouseTLR4/MD2 complex) or its isotype control monoclonal antibody at 4 °Cfor 20 min. The cell surface expression of markers was assessed byEPICS XL-MCL (Beckman Coulter, Inc., Brea, CA).

Transient expression and immunoprecipitation

Subconfluent AD293 cells cultured in 60 mm plates were transientlytransfected with expression plasmids of pcDNA3-HA-mMyd88, pEFBOS-Flag-mMD2, andpcDNA4-mTLR4/Myc (1.3 μg each/plate) usingHyliMax.At 24 h after the transfection, cellswere stimulatedwith 1 μg/ml LPSwithorwithout LIPUS (30 mW/cm2) for 1 h. Cell lysateswere prepared in PLClysis buffer (50 mM HEPES [pH 7.0], 150 mM NaCl, 10% glycerol, 1%Triton X-100, 1.5 mM MgCl2, 1 mM EGTA, 100 mM NaF, 10 mM NaPPi)with the addition of 1 mM Na3VO4 and 1× Proteinase Inhibitor Cocktail(Funakoshi). The cell lysates were incubated with the indicated antibodyfor 2 h at 4 °C, followed by incubation with protein A-Sepharose beads(GE Healthcare Biosciences, Piscataway, NJ) for an additional 1 h. Thebeads were washed three times in ice-cold PLC lysis buffer, suspendedin SDS sample buffer, and heated at 95 °C for 5 min. The eluted proteinswere applied to SDS-polyacrylamide gels, and proteins were detectedby Western blotting.

Results

LPS-induced chemokine gene expression profiles in osteoblasts

Effects of LPS on the chemokine gene expression profiles were ex-amined by quantitative PCR in MC3T3-E1 (a commonly usedmouse os-teoblast cell line) cells and mouse calvaria-derived primary osteoblasts(Table 1). Among the 34 chemokines (including 19 CCL and 13 CXCL

Table 1Chemokine mRNA induction by LPS in osteoblasts.

Chemokine Primary OB MC3T3-E1 5′ primer 3′ primer

I-309 CCL1 9.10 48.5 CAGCAAGAGCATGCTTACGG TTGAGGCGCAGCTTTCTCTAMCP-1 CCL2 64.1 81.4 AGCACCAGCCAACTCTCACT CGTTAACTGCATCTGGCTGAMIP-1α CCL3 7.35 n.d. CTCAACATCATGAAGGTCTC GATGAATTGGCGTGGAATCTMIP-1β CCL4 20.8 1.60 CTCTCTCTCCTCTTGCTCGT CCGGGAGGTGTAAGAGAAACRANTES CCL5 13.1 2.03 ATATGGCTCGGACACCACTC TTCTTCGAGTGACAAACACGMCP-3 CCL7 2.19 54.9 CCTGGGAAGCTGTTATCTTCAA TGGAGTTGGGGTTTTCATGTCMCP-2 CCL8 3.32 1.80 GCTGTGGTTTTCCAGACCAA GAAGGTTCAAGGCTGCAGAAMIP-1γ CCL9/10 2.41 5.34 GGGTCTGCCCACTAAGAAGA TCTGTTGCATGTGTGATCTGGEotaxin CCL11 n.d. 5.22 CACGGTCACTTCCTTCACCT GCTTTCAGGGTGCATCTGTTTARC CCL17 2.05 1.20 CAATGTAGGCCGAGAGTGC GGACAGTCAGAAACACGATGGMIP-3β CCL19 39.0 6.10 AGTCTTCTGCCAAGAACAAAGG CTGGGACAGAGCATCAGGAGMIP-3α CCL20 28.6 371 CCAGGCAGAAGCAGCAAG ATCGGCCATCTGTCTTGTGASLC CCL21 n.d. 1.15 TCCGAGGCTATAGGAAGCAA GTTCTGCACCCAGCCTTCMDC CCL22 1.21 1.13 TGATGCAGGTCCCTATGGTG GGCAGGATTTTGAGGTCCAGMPIF-2 CCL24 2.27 0.58 AGGGGTCATCTTCATCACCA GCAAACTTGGTTCTCACTGCTECK CCL25 0.953 1.14 AAGGTGCCTTTGAAGACTGC TGGCGGAAGTAGAATCTCAGAEotaxin-3 CCL26 0.962 0.64 CCTGCTGCCCTAATTTCAGC AAAGAATATCACACCGTCACTGGILC CCL27 0.849 1.02 ATAGACAGCCACTCCCAAGC TCTGGGGATGAACACAGACAMEC CCL28 n.d. n.d. GAGTTCATGCAGCATCCAGA AGGCTCTCATCCACTGCTTCLymphotactin XCL1 1.23 n.d. TGACTTTCCTGGGAGTCTGC AGCCGCTGGGTTTGTAAGTTFractalkine CX3CL1 1.96 n.d. TCATGTGCGACAAGATGACC GTGTCGTCTCCAGGACAATGGRO-α CXCL1 180 1340 CACCCAAACCGAAGTCATAGC ACTTGGGGACACCTTTTAGCAMIP-2 CXCL2 4.97 200 GAAGTCATAGCCACTCTCAAGG TTCCGTTGAGGGACAGCAMIP-2β CXCL3 3.40 544 CAGCCACACTCCAGCCTA CACAACAGCCCCTGTAGCPF-4 CXCL4 n.d. 1.19 AGTCCTGAGCTGCTGCTTCT GATCTCCATCGCTTTCTTCGENA78 CXCL5 1.60 17.9 GCCCTACGGTGGAAGTCATA GTGCATTCCGCTTAGCTTTCNAP-2 CXCL7 0.921 1.19 TGGGCTTCAGACTCAGACCT GGTCCATGCCATCAGATTTTMIG CXCL9 9.12 229 CGGACTTCACTCCAACACAG TAGGGTTCCTCGAACTCCACIP-10 CXCL10 112 145 GGATCCCTCTCGCAAGGA ATCGTGGCAATGATCTCAACAI-TAC CXCL11 n.d. 5.81 GCTCAAGGCTTCCTTATGTTCA TTTGTCGCAGCCGTTACTCSDF-1 CXCL12 1.25 0.78 GCTCTGCATCAGTGACGGTA AGATGCTTGACGTTGGCTCTBLC CXCL13 1.13 2.96 CAGGCCACGGTATTCTGGA CAGGGGGCGTAACTTGAATCBRAK CXCL14 1.17 n.d. ACAATGCCTGGAACGAGAAG TCTCTCAACTGGCCTGGAGTSR-PSOX CXCL16 3.23 1.73 GACCCTTGTCTCTTGCGTTC GGGTGCCAGAAGAAATGGTA

Chemokine mRNA induction ratios of mouse calvaria-derived primary osteoblasts andMC3T3-E1 cells stimulated with LPS (1 μg/ml) for 2 h. Specific primer sequences used in real-timePCR analyses are shown.

19J. Nakao et al. / Bone 58 (2014) 17–25

chemokines) examined, the gene expression of 9 (MC3T3-E1) and 7(primary osteoblasts) chemokines was induced more than 10 timeswith 2 h LPS stimulation. Although the induction profiles were some-what different, they were basically similar between the two cell types(Table 1). Especially, the gene expression of three chemokines, CCL2(MCP-1), CXCL1 (GRO-α), and CXCL10 (IP-10), was markedly induced(more than 50 times) by LPS in both cell types, and was further studiedfor the effects of LIPUS.

Inhibitory effects of LIPUS on LPS-induced chemokine andRANKL expression

In order to study the effects of LIPUS on the chemokine expression ofosteoblasts, MC3T3-E1 cells were stimulated by LPS in the presence orabsence of simultaneous LIPUS treatment. LIPUS was applied at30 mW/cm2 with the same condition used in clinical treatment ofbone fracture (see Materials and methods). Effects of LIPUS on theLPS-induced mRNA expression of three chemokines (CCL2, CXCL1, andCXCL10)were quantitatively analyzed (Fig. 1) by real-time PCR. As a re-sult, simultaneous LIPUS treatment significantly inhibited the LPS-induced mRNA expression of CXCL1 and CXCL10. The LPS-inducedCCL2 mRNA expression was modestly inhibited by LIPUS in a constantmanner, but the degree of inhibition was not statistically significant.As LPS is a potent inducer of RANKL from osteoblasts [5,16], we also ex-amined the effect of LIPUS on RANKL mRNA expression in MC3T3-E1cells. It was found that LIPUS significantly inhibited LPS-inducedmRNA expression of RANKL (Fig. 1). Similar inhibitory effects of LIPUSwere observed at increased intensities (60 and 120 mW/cm2) of LIPUS(data not shown).

Inhibitory effects of LIPUS on the LPS-induced signal transduction

As LPS stimulation induces the activation of various kinases and tran-scription factors in osteoblasts [16,17], we examined the effects of LIPUSon the activities of intracellular signal transducers. LPS is sensed by TLR4,which mediates two distinct (MyD88-dependent and -independent,respectively) intracellular signaling pathways in various cell types. InWestern blot analyses, LPS rapidly induced the transient phosphoryla-tion of three MAP kinases, ERKs, JNKs, p38 kinases, as well as their up-stream kinases, MEK1/2, MKK3/6 in both MC3T3-E1 cells and primaryosteoblasts (Fig. 2A). LPS also rapidly phosphorylates Akt, a downstreamsignaling kinase of phosphoinositide 3-kinase (PI3K), as well as IκB ki-nases (IKKs), upstream activators of a transcription factor, nuclearfactor-κB (NF-κB) in both cell types (Fig. 2A). Furthermore, TANK-binding kinase 1 (TBK1) and its downstream target, Interferon regulato-ry factor 3 (IRF3), both of which are specific signaling components ofMyD88-independent pathway, were phosphorylated by LPS in bothcell types (Fig. 2A). Simultaneous treatment with LIPUS significantlyinhibited the LPS-induced phosphorylation of all these signaling mole-cules except JNK (Fig. 2A). On the other hand, the stand-alone treatmentby LIPUS inducedweak phosphorylation of ERKs and p38, whichwas de-tected by longer exposure of the blot (Fig. 2B), although it did not exertdetectable effects on any other signal molecules examined.

LPS is known to activate two transcriptional factors, NF-κB and IRF3,through twodistinct (MyD88-dependent and -independent, respective-ly) pathways in various cell types. In order to investigate the effects ofLIPUSon these transcriptional factor activities,MC3T3-E1 cellswere sta-bly transfected with a luciferase reporter plasmid containing NF-κBresponsive element or Interferon-sensitive response element (ISRE).LIPUS significantly inhibited the LPS-induced transcriptional activation

Fig. 1. LIPUS exerts inhibitory effects on LPS-induced chemokine and RANKLmRNA expression in osteoblasts. MC3T3-E1 cells were stimulated with LPS (1 μg/ml) in the absence or pres-ence of simultaneous LIPUS (30 mW/cm2) treatment for 2 h. Untreated MC3T3-E1 cells were used as the control. Total RNA was isolated, reverse-transcribed, and used for quantitativereal-time PCR analyses using the specific primers for CCL2, CXCL1, CXCL10, and RANKL. GAPDHwas used as the endogenous control for normalization. Fold increase represents an exper-imental value divided by the control (untreated) value for each part. Vertical bars indicate mean SD of triplicate samples. *Significant difference from the same day value of the controlgroup by Student's t-test (p b 0.01). Each experiment was repeated at least 3 times with similar results.

20 J. Nakao et al. / Bone 58 (2014) 17–25

of both NF-κB and ISRE promoters (Fig. 3A, B). Additionally, constitutiveactivity of NF-κB, but not ISRE, was modestly but significantly inhibitedby LIPUS treatment alone (Fig. 3A, B).

LIPUS does not affect cell surface expression of TLR4/MD2

We then sought to examine themechanisms how LIPUS inhibits LPSsignals in osteoblast. We and others have previously reported that TLR4is abundantly expressed on osteoblast cell surface and is essential forthe osteoblast responses to LPS [5–8]. TLR4 is known to form a complexwith MD2, and this complex formation is essential for the LPS receptorfunction of TLR4 [18]. The possible influence of LIPUS on the cell surfaceexpression of TLR4/MD2 complex on MC3T3-E1 cells was analyzed byflowcytometry using a monoclonal antibody, which specifically recog-nizes the mouse TLR4/MD2. Stimulation by LIPUS did not affect thecell surface expression level of TLR4/MD2 at least for 6 h (Fig. 4), indi-cating neither TLR4 surface expression nor TLR4/MD2 complex forma-tion was inhibited by LIPUS.

LPS stimulation of cells is known to induce internalization of TLR4,which contributes to the induced refractory period to LPS, termed as“LPS tolerance” [19]. Flowcytometry data showed that LPS-induced in-ternalization of TLR4/MD2 complex similarly occurs in the presence orabsence of LIPUS stimulation (data not shown). Similar flowcytometrydata of TLR4/MD2 surface expression levels were observed in mousecalvaria-derived primary osteoblasts treated by LIPUS (data not shown).

Inhibition of TLR4-MyD88complex formation by LIPUS

The cytoplasmic domain of TLR4 physically associates with MyD88,an intracellular adaptor protein important for the LPS-induced activation

of MAP kinases and NF-κB [20]. For testing the possible influenceof LIPUS on TLR4–MyD88 association, expression plasmids encodingepitope-tagged TLR4, MyD88, and MD2 were transiently transfectedinto AD293 cells (Fig. 5A). In immunoprecipitation analyses, co-precipitation of TLR4 and MyD88 was detected in LPS-stimulatedAD293 cells (Fig. 5B). The level of co-precipitation was significantlydecreased with the simultaneous treatment with LIPUS (Fig. 5B), indi-cating that LIPUS is inhibitory to the complex formation of TLR4 andMyD88.

Discussion

Recent studies have revealed that the immune and skeletal systemshave a variety of regulatorymolecules in common [21]. Chemokines aresmall peptides considered as chemo-attractants for leukocyte subpopu-lationswithin local inflamed sites. Notably, chemokine expression in os-teoblasts is considered to be pathologically significant in some skeletaldiseases. For example, elevated expression of CCL2 is observed in oste-oblasts in a murine model of Staphylococcus aureus osteomyelitis andinfected human bone tissue [22]. Moreover, in patients with osteoar-thritis and rheumatoid arthritis, high expression levels of CCL2, CCL3and CCL5, but not CCL4, was detected in osteoblasts in the lesions [23].In studies of osteoblast cell lines, expression of CCL2 and CXCL10 inMC3T3-E1 cells was induced strongly by TNFα, and weakly by IFNγ orIL-1α, whereas CXCL1 mRNA expression was not induced by any ofthese three cytokines [24]. On the other hand, CXCL1 was reported tobe induced by TGFβ in 10T1/2, another mouse osteoblastic cell line[25]. These reports indicate that osteoblasts are important sources ofkey inflammatory chemokines in bone tissues recruiting various inflam-matory cells under conditions of inflammatory stress.

Fig. 2. Intracellular signal transduction of LIPUS in osteoblasts. (A) Inhibitory effects of LIPUS on LPS-induced intracellular signal transduction in osteoblasts. MC3T3-E1 cells (left)and mouse calvaria-derived primary osteoblast (right) were stimulated with 1 μg/ml LPS for the indicated time in the absence or presence of the simultaneous LIPUS treatment(30 mW/cm2). The fourth lane from the left represents 30 min treatment with LIPUS alone. Cytoplasmic lysates (5 μg protein for each) were analyzed with the indicated antibodies byWestern blotting. (B) Low-grade stimulatory effects of stand-alone LIPUS treatment on ERKs and p38 in MC3T3-E1 cells. MC3T3-E1 cells were stimulated with LIPUS (30 mW/cm2) or1 μg/ml LPS for 30 min. Cytoplasmic lysates (5 μg protein for each) were analyzed with the indicated antibodies byWestern blotting as in (A). Long exposure to the film was performedto visualize the low-intensity signals.

21J. Nakao et al. / Bone 58 (2014) 17–25

LPS has been identified as an important pathogenic factor in somebone loss diseases, such as inflammatory arthritis [26] and periodontitis[27]. In response to LPS, osteoblasts produce and secrete IL-1, IL-6, pros-taglandin (PG) E2 and RANKL [5,28–30], all of which induce osteoclastmaturation and/or activation. Notably, several chemokines, includingCCL2, CCL3, CCL9, CXCL12, are reported to directly attract osteoclast pre-cursors and increase osteoclast formation [31–34], indicating that someosteoblast-derived chemokines may directly induce osteoclastogenesisand bone resorption. It should also be noted that human osteoclasts

and their progenitors express CXCR1 and respond to its ligand, CXCL8[35,36]. Although the direct effect of CXCL2, a mouse homolog ofhuman CXCL8, on osteoclasts has not been elucidated in mice, it maydirectly promote osteoclastogenesis. In this study, our findings have re-vealed that LPS strongly induces several chemokine mRNAs in osteo-blasts in selective manners (Table 1). Among the 9 chemokines whichare reported or presumed to be involved in osteoclastogenesis (CCL2,3, 4, 5, 9 and CXCL1, 2, 10, 12) as mentioned above, mRNA expressionof 8 chemokines, but not CXCL12, was significantly induced with 2 h

Fig. 3. LIPUS inhibits LPS-induced transcriptional activation of NF-κB responsive element and ISRE in osteoblasts. MC3T3-E1 cells, stably transfectedwith pNF-κB-Luc (A) or pISRE-Luc (B),were stimulated with LIPUS (30 mW/cm2), 1 μg/ml LPS, or their combination for 2 h. Cells were lysed for luciferase activity measurements as described in Materials and methods.

22 J. Nakao et al. / Bone 58 (2014) 17–25

stimulation by LPS (Table 1). Thus, osteoblasts directly respond toGram-negative bacterial infection by secreting chemokines to inducetissue inflammation and bone resorption.

In the present study, it has been revealed that the mechanical stim-ulus applied by LIPUS exerts anti-inflammatory effects on LPS-treatedosteoblasts, inhibiting mRNA induction of CXCL1 and CXCL10 as wellas RANKL, which is a well-known potent inducer of osteoclastogenesis(Fig. 1). Both in vitro and in vivo studies have revealed that CXCL1 is apotent chemo-attractant and activator of neutrophils [37,38]. CXCL1 be-longs to a subclass of CXC chemokines characterized by the presence ofglutamic acid–leucine–arginine motif [39]. Among chemokines in thissubclass, CXCL1, CXCL2, CXCL3, CXCL5, and CXCL7, share a common

Fig. 4. LIPUS does not affect cell surface expression of TLR4/MD2 receptor complex on osteoblastsurface-stainedwith PE-conjugatedMTS510 (recognizingmouse TLR4/MD2 complex) or the isoanalysis. The overlay of MTS510 intensities is shown in the bottom left figure.

receptor, CXCR2 [40], and the CXCR2 knockout mice display a runtedphenotype and fragile alveolar bone, as well as several abnormalitiesof the hematopoietic system, indicating a role for CXCR2 in bone phys-iology [41].

CXCL10, on the other hand, activates CXCR3,which is predominantlyexpressed on activated T cells, natural killer cells, inflammatory dendrit-ic cells,macrophages, and B cells [42]. CXCR3 is shared by two other CXCchemokines, CXCL9 and CXCL11. A previous report showed increasedtissue expression of CXCL10 in rheumatoid arthritis [43]. It has recentlybeen found that CXCL10 significantly increases RANKL expression inCD4+ T cells inducing osteoclastogenesis and bone loss [44]. Thus,osteoblast-derived CXCL1 and CXCL10 are considered to be potent

s.MC3T3-E1 cells were treatedwith LIPUS (30 mW/cm2) for the indicated time. Cells weretype controlmonoclonal antibody. Thefluorescent signalwasmeasuredbyflowcytometric

Fig. 5. LIPUS inhibits TLR4/MyD88 complex formation. AD293 cellswere transiently transfectedwith the expressionplasmids forMyc epitope-taggedmTLR4,HAepitope-taggedmMyD88,and Flag-taggedmMD2. (A) Protein expression ofMyc-taggedmTLR4 and HA-taggedmMyD88was detected byWestern blotting using anti-Myc and anti-HA antibodies at 24 h after thetransfection. The left lane shows untransfectedAD293 cells as a negative control. (B) At 24 h after the transfection, the cellswere stimulatedwith 1 μg/ml LPS for 1 hwith orwithout LIPUStreatment (30 mW/cm2). Immunoprecipitationwas performedwith anti-HA or anti-Myc antibody. The precipitateswere separated by SDS-PAGE followed by detection using the indicat-ed antibodies. (C) Schematic presentation of the inhibitory effects of LIPUS on LPS signaling. BothMyD88-dependent andMyD88-independent (TRIF-mediated) signaling pathways of LPSare inhibited by simultaneous LIPUS treatments. See text for more detailed information.

23J. Nakao et al. / Bone 58 (2014) 17–25

inducers of both bone tissue inflammation and bone resorption duringGram-negative bacterial infection. As osteoblasts are relatively sensitiveto ultrasound compared with other cell types [2,3], and LIPUS signifi-cantly inhibits LPS-induced CXCL1 and CXCL10 expressions in osteo-blasts, the clinical application of LIPUS on bone tissue may effectivelyinhibit osteolytic inflammation induced by Gram-negative bacterialinfection.

Unlike CXCL1 and CXCL10, LPS-induced expression of CCL2 wasonly modestly, and not significantly, decreased by LIPUS (Fig. 1). Thetranscriptional regulatory regions of mouse CCL2, CXCL1, and CXCL10genes have previously been reported [45–47]. The 5′-upstream regulato-ry region of the mouse CXCL1 gene contains two functional NF-κB con-sensus sequences, which cooperatively control LPS responsiveness [46],while the 5′-upstream regulatory region of themouse CXCL10 gene con-tains two functional NF-κB consensus sequences and one functional ISRE,all ofwhich contribute to LPS-induced gene expression [45]. On the otherhand, the LPS responsiveness of the mouse CCL2 gene is essentially con-trolled by an upstream sequence between −169 and −132, to which a

nuclear factor other than NF-κB is bound. Thus, we presume that theweaker response of CCL2 to LIPUS than CXCL1 and CXCL10 is due tothe different gene transcriptional regulatory mechanisms. This hypothe-sis, however, obviously needs further investigation.

Additionally, LIPUS effectively inhibited LPS-induced mRNA expres-sion of RANKL, similar to that of CXCL1 and CXCL10 (Fig. 1). We havepreviously reported that ERK activation is essential for RANKLmRNA in-duction in LPS-stimulated osteoblasts [16]. Thus, this result is consistentwith our intracellular signaling data (Fig. 2A), which demonstrated theeffective inhibition of LPS-induced ERK phosphorylation by LIPUS.

LPS is sensed by Toll like receptor (TLR) 4, and the downstream sig-nal is mediated by adaptor molecules including MyD88, Toll receptor–IL-1 receptor domain containing adapter protein (TIRAP)/MyD88-adapter-like (Mal), Toll receptor–IL-1 receptor domain containingadaptor protein inducing interferon-β (TRIF), and TRIF-related adaptormolecule (TRAM) [48–51] (Fig. 5C).MyD88 forms a complexwith intra-cellular signaling domain of TLR4 and initiates the “MyD88-dependent”signal activating nuclear transport of nuclear NF-κB and a set of MAPKs.

24 J. Nakao et al. / Bone 58 (2014) 17–25

On the other hand, TRIF/TRAM initiates the “MyD88-independent”signal leading to the activation of TBK1, inducing the phosphorylation/activation of IRF3 [52]. Our Western blot data have revealed that LPS-induced phosphorylation of ERKs, p38 kinases, Akt, IKKs, TBK1, andIRF3 is inhibited by the presence of simultaneous LIPUS treatment inboth MC3T3-E1 and primary osteoblasts (Fig. 2A). These findings indi-cated that both MyD88-dependent and -independent TLR4 signalingpathways in osteoblasts are down-regulated by LIPUS. Consistently,transcriptional activation of both NF-κB consensus sequence and ISREby LPS was inhibited by LIPUS in MC3T3-E1 cells (Fig. 3).

Unlike other kinases, phosphorylation of JNKs was not inhibited byLIPUS (Fig. 2A). We currently do not know the molecular mechanismsexplaining this specific unresponsiveness of JNKs. As the inhibition ofLPS signals by LIPUS is partial, one possibility is that the activationthreshold required for JNKs is relatively low and that the decreasedLPS signal intensity in the presence of LIPUS is still sufficient for the nor-mal phosphorylation of JNKs. Although we have no direct experimentalevidence supporting this possibility, it has previously been reportedthat TLR4 physically associates with JNK-interacting protein (JIP) 3, ascaffold protein which facilitates the phosphorylation of JNKs by up-stream kinases [20]. Thus, it may be reasonable to presume that TLR4-associated JIP3 decreases activation threshold of JNKs by LPS.

Our Western blot data also showed that LPS-induced phosphoryla-tion ofMEK1/2 andMKK3/6, upstreamkinases for ERKs and p38 kinasesrespectively, was decreased by the simultaneous LIPUS treatment(Fig. 2A), indicating that the receptor or near-receptor upstream eventsof the MyD88-dependent pathway were negatively affected by LIPUS.Our flowcytometry data, however, showed that the constitutive surfaceexpression level of TLR4/MD2 complex was not affected by LIPUS(Fig. 4). Furthermore, LPS-induced internalization of TLR4, which is aknown mechanism of feedback inhibition of LPS called “LPS tolerance”[19], was not affected by LIPUS.We then performed a transient transfec-tion experiment and found that the co-precipitation of TLR4 andMyD88was significantly inhibited by LIPUS treatment of the cells (Fig. 5). Asthe association of MyD88 and TLR4 is essential for the downstream sig-nals of the MyD88-dependent pathway, it is presumed that LIPUSnegatively affect LPS signals through the inhibition of TLR4/MyD88complex formation. As the MyD88-independent signal represented byTBK1 and IRF3 phosphorylation is also inhibited (Fig. 2A), it seems likelythat the signal transduction by TRIF/TRAM adaptor molecules is alsoinhibited by LIPUS. This possibility, however, obviously needs furtherexamination.

At present we do not know the molecular mechanisms of how theTLR4/MyD88 complex formation is inhibited by LIPUS. Previous studieshave reported that mechanical stresses including LIPUS are specificallysensed by cell surface receptors such as integrins in various cell types[53]. We have previously reported that angiotensin II type1 receptor(AT1) specifically senses mechanical stress by LIPUS in differentiatedosteoblasts [54]. Thus it is possible that there is an interfering crosstalkbetween intracellular signals of LPS and LIPUS. However, this possibilitydoes not seem to be very likely, as the inhibition by LIPUSdirectly affectsTLR4–MyD88 complex formation near the cell surface (Fig. 5), and nei-ther TLR4 nor MyD88 has been identified as an intracellular signalingcomponent of either integrins or AT1.

AsMyD88 is a common signaling component shared by other recep-tors including IL-1 receptor and other TLR members, it is also conceiv-able that LIPUS may signal through one of these receptors competingwith TLR4 for MyD88 binding. However, this possibility seems unlikelyas the stimulation of TLR4/MyD88-transfected AD293 cells with IL-1 didnot inhibit the complex formation of TLR4/MyD88 (data not shown).Additionally, AD293 cells were defective in the protein expression ofTLR2, an important TLR family member, in our flowcytometry analysis(data not shown).

Instead, we would like to propose a hypothesis that the continuousmechanical force on the cell membrane produced by LIPUS physicallyaffects the structural association between TLR4 and MyD88 (Fig. 5C).

Although we do not have any definitive experimental evidence forthis hypothesis, it has previously been reported that the mechanicalstress by ultrasound can induce structural changes in the cellmembraneincluding cell surface receptors [55].

Although the effects of LIPUS on inflammatory skeletal diseaseshave not been fully elucidated, several previous reports seem to be con-sistent with our current findings. For example, in a study using a ratmodel of the surgical muscular incision, application of LIPUS on the in-flammatory region reduced the blood counts of total leukocytes, mono-cytes, segmented neutrophils, and lymphocytes, indicating the anti-inflammatory effects of LIPUS [56]. On the other hand, another reportstudying a rat model of a ruptured Achilles tendon showed that LIPUStreatment increased expressions of anti-inflammatory proteins, E-typeprostanoid receptor (EP) 4 and TGFβ [57].

It should be noted, however, that LIPUS by itself has stimulatoryeffects on some cell types. As a matter of fact, we have previously re-ported that mature osteoblasts, which are more sensitive to LIPUSthan immature osteoblasts, express moderate amounts of chemokinemRNAs in response to LIPUS [54]. In the present study,we examined os-teoblasts (both a cell line and primary culture cells) of immature pheno-type. Although the standalone LIPUS did not induce detectable amountsof chemokine mRNAs (data not shown), it still weakly induced phos-phorylation of ERKs and p38 kinases, indicating that LIPUS by itself isweakly stimulatory to immature osteoblasts (Fig. 2B). Thus the effectsof LIPUS on osteoblasts may vary depending on the existence of extra-cellular stimulatory molecules (such as LPS) and the differentiation sta-tus of the osteoblasts.

Acknowledgments

This work was supported by grants from the Ministry of Education,Culture, Sports, Science and Technology of Japan, and Teijin Pharmawhich supplied the LIPUS device.

References

[1] RomanoCL, RomanoD, LogolusoN. Low-intensity pulsed ultrasound for the treatmentof bone delayed union or nonunion: a review. Ultrasound Med Biol 2009;35:529–36.

[2] Naruse K, Miyauchi A, Itoman M, Mikuni-Takagaki Y. Distinct anabolic response ofosteoblast to low-intensity pulsed ultrasound. J Bone Miner Res 2003;18:360–9.

[3] Saito M, Soshi S, Tanaka T, Fujii K. Intensity-related differences in collagen post-translational modification in MC3T3-E1 osteoblasts after exposure to low- andhigh-intensity pulsed ultrasound. Bone 2004;35:644–55.

[4] Aguila HL, Rowe DW. Skeletal development, bone remodeling, and hematopoiesis.Immunol Rev 2005;208:7–18.

[5] Kikuchi T, Matsuguchi T, Tsuboi N, Mitani A, Tanaka S, Matsuoka M, et al. Gene ex-pression of osteoclast differentiation factor is induced by lipopolysaccharide inmouse osteoblasts via Toll-like receptors. J Immunol 2001;166:3574–9.

[6] Gasper NA, Petty CC, Schrum LW, Marriott I, Bost KL. Bacterium-induced CXCL10 se-cretion by osteoblasts can be mediated in part through Toll-like receptor 4. InfectImmun 2002;70:4075–82.

[7] Bandow K, Maeda A, Kakimoto K, Kusuyama J, Shamoto M, Ohnishi T, et al. Molecu-lar mechanisms of the inhibitory effect of lipopolysaccharide (LPS) on osteoblast dif-ferentiation. Biochem Biophys Res Commun 2010;402:755–61.

[8] Zou W, Amcheslavsky A, Bar-Shavit Z. CpG oligodeoxynucleotides modulate theosteoclastogenic activity of osteoblasts via Toll-like receptor 9. J Biol Chem2003;278:16732–40.

[9] Jiang Y, Graves DT. Periodontal pathogens stimulate CC-chemokine production bymononuclear and bone-derived cells. J Periodontol 1999;70:1472–8.

[10] Skerry TM. The response of bone to mechanical loading and disuse: fundamentalprinciples and influences on osteoblast/osteocyte homeostasis. Arch BiochemBiophys 2008;473:117–23.

[11] Matsuguchi T, Chiba N, Bandow K, Kakimoto K, Masuda A, Ohnishi T. JNK activity isessential for Atf4 expression and late-stage osteoblast differentiation. J Bone MinerRes 2009;24:398–410.

[12] Matsuguchi T, Musikacharoen T, Johnson TR, Kraft AS, Yoshikai Y. A novel mitogen-activated protein kinase phosphatase is an important negative regulator oflipopolysaccharide-mediated c-Jun N-terminal kinase activation in mouse macro-phage cell lines. Mol Cell Biol 2001;21:6999–7009.

[13] Iwabuchi S, Ito M, Hata J, Chikanishi T, Azuma Y, Haro H. In vitro evaluation of low-intensity pulsed ultrasound in herniated disc resorption. Biomaterials 2005;26:7104–14.

[14] Cook SD, Ryaby JP, McCabe J, Frey JJ, Heckman JD, Kristiansen TK. Acceleration oftibia and distal radius fracture healing in patients who smoke. Clin Orthop1997:198–207.

25J. Nakao et al. / Bone 58 (2014) 17–25

[15] Wang SJ, Lewallen DG, BolanderME, Chao EY, Ilstrup DM, Greenleaf JF. Low intensityultrasound treatment increases strength in a rat femoral fracture model. J OrthopRes 1994;12:40–7.

[16] Kikuchi T, Yoshikai Y, Miyoshi J, Katsuki M, Musikacharoen T, Mitani A, et al. Cot/Tpl2 is essential for RANKL induction by lipid A in osteoblasts. J Dent Res 2003;82:546–50.

[17] Yang S, Takahashi N, Yamashita T, Sato N, Takahashi M, Mogi M, et al. Muramyl di-peptide enhances osteoclast formation induced by lipopolysaccharide, IL-1 alpha,and TNF-alpha through nucleotide-binding oligomerization domain 2-mediated sig-naling in osteoblasts. J Immunol 2005;175:1956–64.

[18] Kobayashi M, Saitoh S, Tanimura N, Takahashi K, Kawasaki K, Nishijima M, et al.Regulatory roles forMD-2 and TLR4 in ligand-induced receptor clustering. J Immunol2006;176:6211–8.

[19] Nomura F, Akashi S, Sakao Y, Sato S, Kawai T, Matsumoto M, et al. Cutting edge: en-dotoxin tolerance in mouse peritoneal macrophages correlates with down-regulation of surface toll-like receptor 4 expression. J Immunol 2000;164:3476–9.

[20] Matsuguchi T, Masuda A, Sugimoto K, Nagai Y, Yoshikai Y. JNK-interacting protein 3associates with Toll-like receptor 4 and is involved in LPS-mediated JNK activation.EMBO J 2003;22:4455–64.

[21] Takayanagi H, Sato K, Takaoka A, Taniguchi T. Interplay between interferon andother cytokine systems in bone metabolism. Immunol Rev 2005;208:181–93.

[22] Marriott I, Gray DL, Rati DM, Fowler Jr VG, Stryjewski ME, Levin LS, et al. Osteoblastsproduce monocyte chemoattractant protein-1 in a murine model of Staphylococcusaureus osteomyelitis and infected human bone tissue. Bone 2005;37:504–12.

[23] Lisignoli G, Toneguzzi S, Grassi F, Piacentini A, Tschon M, Cristino S, et al. Differentchemokines are expressed in human arthritic bone biopsies: IFN-gamma and IL-6differently modulate IL-8, MCP-1 and rantes production by arthritic osteoblasts.Cytokine 2002;20:231–8.

[24] Ohmori Y, Hamilton TA. Cell type and stimulus specific regulation of chemokinegene expression. Biochem Biophys Res Commun 1994;198:590–6.

[25] Bischoff DS, Zhu JH, Makhijani NS, Yamaguchi DT. KC chemokine expression by TGF-beta in C3H10T1/2 cells induced towards osteoblasts. Biochem Biophys ResCommun 2005;326:364–70.

[26] Yogesha SD, Khapli SM, Srivastava RK, Mangashetti LS, Pote ST, Mishra GC, et al. IL-3inhibits TNF-alpha-induced bone resorption and prevents inflammatory arthritis.J Immunol 2009;182:361–70.

[27] Kirkwood KL, Li F, Rogers JE, Otremba J, Coatney DD, Kreider JM, et al. A p38alphaselective mitogen-activated protein kinase inhibitor prevents periodontal boneloss. J Pharmacol Exp Ther 2007;320:56–63.

[28] Keeting PE, Rifas L, Harris SA, Colvard DS, Spelsberg TC, Peck WA, et al. Evidencefor interleukin-1 beta production by cultured normal human osteoblast-like cells.J Bone Miner Res 1991;6:827–33.

[29] Sakuma Y, Tanaka K, SudaM, Yasoda A, Natsui K, Tanaka I, et al. Crucial involvementof the EP4 subtype of prostaglandin E receptor in osteoclast formation by proinflam-matory cytokines and lipopolysaccharide. J Bone Miner Res 2000;15:218–27.

[30] Ishimi Y, Miyaura C, Jin CH, Akatsu T, Abe E, Nakamura Y, et al. IL-6 is produced byosteoblasts and induces bone resorption. J Immunol 1990;145:3297–303.

[31] Li X, Qin L, Bergenstock M, Bevelock LM, Novack DV, Partridge NC. Parathyroid hor-mone stimulates osteoblastic expression of MCP-1 to recruit and increase the fusionof pre/osteoclasts. J Biol Chem 2007;282:33098–106.

[32] Scheven BA, Milne JS, Hunter I, Robins SP. Macrophage-inflammatory protein-1alpha regulates preosteoclast differentiation in vitro. Biochem Biophys ResCommun 1999;254:773–8.

[33] Yang M, Mailhot G, MacKay CA, Mason-Savas A, Aubin J, Odgren PR. Chemokine andchemokine receptor expression during colony stimulating factor-1-induced osteoclastdifferentiation in the toothless osteopetrotic rat: a key role for CCL9 (MIP-1gamma) inosteoclastogenesis in vivo and in vitro. Blood 2006;107:2262–70.

[34] Wright LM, Maloney W, Yu X, Kindle L, Collin-Osdoby P, Osdoby P. Stromal cell-derived factor-1 binding to its chemokine receptor CXCR4 on precursor cells pro-motes the chemotactic recruitment, development and survival of human osteoclasts.Bone 2005;36:840–53.

[35] Bendre MS, Montague DC, Peery T, Akel NS, Gaddy D, Suva LJ. Interleukin-8 stimula-tion of osteoclastogenesis and bone resorption is a mechanism for the increasedosteolysis of metastatic bone disease. Bone 2003;33:28–37.

[36] Bendre MS, Margulies AG, Walser B, Akel NS, Bhattacharrya S, Skinner RA, et al.Tumor-derived interleukin-8 stimulates osteolysis independent of the receptor acti-vator of nuclear factor-kappaB ligand pathway. Cancer Res 2005;65:11001–9.

[37] Cacalano G, Lee J, Kikly K, Ryan AM, Pitts-Meek S, Hultgren B, et al. Neutrophil and Bcell expansion in mice that lack the murine IL-8 receptor homolog. Science1994;265:682–4.

[38] Bozic CR, Kolakowski Jr LF, Gerard NP, Garcia-Rodriguez C, von Uexkull-GuldenbandC, Conklyn MJ, et al. Expression and biologic characterization of the murine chemo-kine KC. J Immunol 1995;154:6048–57.

[39] Baggiolini M, Dewald B, Moser B. Interleukin-8 and related chemotactic cytokines—CXC and CC chemokines. Adv Immunol 1994;55:97–179.

[40] Bizzarri C, Beccari AR, Bertini R, Cavicchia MR, Giorgini S, Allegretti M. ELR + CXCchemokines and their receptors (CXC chemokine receptor 1 and CXC chemokine re-ceptor 2) as new therapeutic targets. Pharmacol Ther 2006;112:139–49.

[41] Bozic CR, Gerard NP, von Uexkull-Guldenband C, Kolakowski Jr LF, Conklyn MJ,Breslow R, et al. Themurine interleukin 8 type B receptor homologue and its ligands.Expression and biological characterization. J Biol Chem 1994;269:29355–8.

[42] Liu M, Guo S, Stiles JK. The emerging role of CXCL10 in cancer (review). Oncol Lett2011;2:583–9.

[43] Hanaoka R, Kasama T, Muramatsu M, Yajima N, Shiozawa F, Miwa Y, et al. A novelmechanism for the regulation of IFN-gamma inducible protein-10 expression inrheumatoid arthritis. Arthritis Res Ther 2003;5:R74–81.

[44] Lee EY, Seo M, Juhnn YS, Kim JY, Hong YJ, Lee YJ, et al. Potential role and mechanismof IFN-gamma inducible protein-10 on receptor activator of nuclear factor kappa-Bligand (RANKL) expression in rheumatoid arthritis. Arthritis Res Ther 2011;13:R104.

[45] Ohmori Y, Hamilton TA. Cooperative interaction between interferon (IFN) stimulusresponse element and kappa B sequence motifs controls IFN gamma- andlipopolysaccharide-stimulated transcription from the murine IP-10 promoter. J BiolChem 1993;268:6677–88.

[46] Ohmori Y, Fukumoto S, Hamilton TA. Two structurally distinct kappa B sequencemotifs cooperatively control LPS-induced KC gene transcription in mouse macro-phages. J Immunol 1995;155:3593–600.

[47] Ueno M, Sonoda Y, Funakoshi M, Mukaida N, Nose K, Kasahara T. Differential induc-tion of JE/MCP-1 in subclones from a murine macrophage cell line, RAW 264.7: roleof kappaB-3 binding protein. Cytokine 2000;12:207–19.

[48] Rock FL, Hardiman G, Timans JC, Kastelein RA, Bazan JF. A family of human receptorsstructurally related to Drosophila Toll. Proc Natl Acad Sci U S A 1998;95:588–93.

[49] Fitzgerald KA, Palsson-McDermott EM, Bowie AG, Jefferies CA, Mansell AS, Brady G,et al. Mal (MyD88-adapter-like) is required for Toll-like receptor-4 signal transduc-tion. Nature 2001;413:78–83.

[50] Yamamoto M, Sato S, Mori K, Hoshino K, Takeuchi O, Takeda K, et al. Cutting edge: anovel Toll/IL-1 receptor domain-containing adapter that preferentially activates theIFN-beta promoter in the Toll-like receptor signaling. J Immunol 2002;169:6668–72.

[51] Oshiumi H, Matsumoto M, Funami K, Akazawa T, Seya T. TICAM-1, an adaptor mol-ecule that participates in Toll-like receptor 3-mediated interferon-beta induction.Nat Immunol 2003;4:161–7.

[52] O'Neill LA. When signaling pathways collide: positive and negative regulation oftoll-like receptor signal transduction. Immunity 2008;29:12–20.

[53] Watabe H, Furuhama T, Tani-Ishii N, Mikuni-Takagaki Y. Mechanotransduction acti-vates alpha(5)beta(1) integrin and PI3K/Akt signaling pathways inmandibular oste-oblasts. Exp Cell Res 2011;317:2642–9.

[54] Bandow K, Nishikawa Y, Ohnishi T, Kakimoto K, Soejima K, Iwabuchi S, et al. Low-intensity pulsed ultrasound (LIPUS) induces RANKL, MCP-1, and MIP-1beta expres-sion in osteoblasts through the angiotensin II type 1 receptor. J Cell Physiol2007;211:392–8.

[55] Brayman AA, Coppage ML, Vaidya S, Miller MW. Transient poration and cell surfacereceptor removal from human lymphocytes in vitro by 1 MHz ultrasound. Ultra-sound Med Biol 1999;25:999–1008.

[56] Signori LU, Costa ST, Neto AF, Pizzolotto RM, Beck C, Sbruzzi G, et al. Haematologicaleffect of pulsed ultrasound in acute muscular inflammation in rats. Physiotherapy2011;97:163–9.

[57] Kosaka T, Masaoka T, Yamamoto K. Possible molecular mechanism of promotion ofrepair of acute Achilles tendon rupture by low intensity-pulsed ultrasound treat-ment in a rat model. West Indian Med J 2011;60:263–8.

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具