Connective Tissue Research, 50:186–193, 2009Copyright c© Informa Healthcare USA, Inc.ISSN: 0300-8207 print / 1607-8438 onlineDOI: 10.1080/03008200802588451

IL-1� Stimulates the Expression of Prostaglandin ReceptorEP4 in Human Chondrocytes by Increasing Productionof Prostaglandin E2

Yusuke Watanabe, Aki Namba, and Kazuhiro Honda

Nihon University Graduate School of Dentistry, Tokyo, Japan

Yukiko Aida and Hideo Matsumura

Department of Fixed Prosthodontics and Division of Advenced Dental Treatment, Dental ResearchCenter, Nihon University School of Dentistry, Tokyo, Japan

Osamu Shimizu

Department of Oral and Maxillofacial Surgery and Division of Systemic Biology and Oncology, DentalResearch Center, Nihon University School of Dentistry, Tokyo, Japan

Naoto Suzuki

Department of Biochemistry and Division of Functional Morphology, Dental Research Center, NihonUniversity School of Dentistry, Tokyo, Japan

Natsuko Tanabe and Masao Maeno

Department of Oral Health Sciences and Division of Functional Morphology, Dental Research Center,Nihon University School of Dentistry, Tokyo, Japan

Prostaglandin (PG) E2, which exerts its actions via the PG recep-tors EP1–4, is produced from arachidonic acid by cyclooxygenase(COX)-1 and COX-2. The aim of this study was to investigate themechanisms by which interleukin (IL)-1β induces the expressionof PG receptors in cultured human chondrocytes and to explorethe role of PGE2 in this process. The cells were cultured with0, 10, or 100 U/mL IL-1β with or without 1 µM celecoxib, aspecific inhibitor of COX-2, for up to 28 days. Expression of thegenes encoding COX-1, COX-2, and EP1–4 was quantified usingreal-time PCR, and expression of the corresponding proteins wasexamined using immunohistochemical staining. PGE2 productionwas determined using ELISA. IL-1β treatment caused a markeddose- and time-dependent increase in the levels of PGE2, COX-2,and EP4 as compared with the untreated control. It did not affectthe expression of COX-1, and it decreased the expression of EP1and EP2. EP3 expression was not detected in either the absence

Received 24 June 2008; Revised 23 September 2008; Accepted 28October 2008.

Address correspondence to Masao Maeno, PhD, Department ofOral Health Sciences, Nihon University School of Dentistry, 1-8-13, Kanda Surugadai, Chiyoda-ku, Tokyo 101-8310, Japan. E-mail:maeno@dent.nihon-u.ac.jp

or the presence of IL-1β. When celecoxib was also present, IL-1βfailed to stimulate PGE2 production and EP4 expression, but itsstimulatory effect on COX-2 expression and its inhibitory effecton EP1 and EP2 expression were unchanged. IL-1β increases theproduction of PGE2, COX-2, and the PG receptor EP4 in culturedhuman chondrocytes. The increase in EP4 expression appears tobe a result of the increased PGE2 production.

Keywords Chondrocytes, PGE2, IL-1β, PG Receptors, Cyclooxyge-nase

INTRODUCTIONCytokines released at sites of inflammation and infection

can alter the normal processes of cartilage turnover, result-ing in pathologic destruction or formation [1]. We recentlydemonstrated that interleukin (IL)-1β stimulates the turnover ofcartilage matrix primarily by increasing the production of matrixmetalloproteinase-13 by chondrocytes [2]. We also demon-strated that IL-1β promotes the resolution system of cartilagematrix turnover by increasing the production of inflammatorycytokines by chondrocytes and that IL-1β may promote theautocrine action of IL-6 in chondrocytes by increasing IL-6receptor expression [3]. Furthermore, we showed that IL-6

186

Con

nect

Tis

sue

Res

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Frei

e U

nive

rsita

et B

erlin

on

12/0

5/14

For

pers

onal

use

onl

y.

IL-1β PRODUCES EP4 VIA PGE2 187

and soluble IL-6 receptor suppress chondrocyte differentiationand induce the repair of arthrodial cartilage by increasing theexpression of bone morphogenetic protein-7 and its receptors inchondrocytes [4].

In previous studies, prostaglandin (PG) E2 was found in thetemporomandibular joint (TMJ) synovial fluid from patientswith internal derangement of the joint [5–7], and the plasmalevel of PGE2 was related to the general inflammatory activityin these patients [7]. In knee-joint synovial fluid, PGE2 waspresent in higher concentrations in patients with rheumatoidarthritis than in patients with degenerative joint disease, and thelevel of PGE2 was correlated with phospholipase A2 activity[8].

PGs are lipid mediators that are involved in many phys-iological and pathophysiological processes, including thoseassociated with inflammatory airway disease [9]. The PGbiosynthetic pathway begins with the liberation of arachidonicacid from membrane phospholipids via the action of activatedphospholipase A2 or phospholipase C [10]. The liberatedarachidonic acid is converted to PGG2 and PGH2 by PGsynthases, suicide enzymes also known as cyclooxygenases(COXs), of which there are two isotypes; COX-1 and COX-2[11]. PGH2 is the common precursor for all PGs producedthrough terminal enzymes such as the PGE and PGF synthases,which produce PGE2 and PGF2α , respectively [12].

PGE2, one of the principal PGs, exerts its biological effects bybinding to specific cell-surface PG receptors known as EP1, EP2,EP3, and EP4 [13]. Typically, various cells simultaneously ex-press multiple EP receptor proteins, and the relative expressionlevel of these receptors modulates the cellular responsivenessto PGE2. EP1 and EP2 may be present in rat growth-platechondrocytes [14, 15], and EP4 has been identified in bovinearticular chondrocytes [16]. Recently, a full assessment of all EPsubtypes in rabbit growth-plate chondrocytes has been reported[17].

Despite these and other research efforts, little is known aboutthe EP subtypes in human chondrocytes or about the long-termeffects of IL-1β stimulation on EP expression in these cells.Therefore, we assumed cartilage damage and degradation inarthrosis of TMJ and conducted the present study to identifythe mechanisms by which IL-1β induces the expression of PGreceptors in cultured human chondrocytes and also to explorethe putative role of PGE2 in this process. We examined the effectof IL-1β on PGE2 production in human chondrocytes culturedfor up to 28 days and on the expression of COX-1, COX-2, andEP1–4 in these cells. We also examined the effect of celecoxib,a specific inhibitor of COX-2, on the IL-1β response in thesecells at days 21 and 28 of culture.

MATERIALS AND METHODS

Cell CultureChondrocytes derived from normal human femoral cartilage

were obtained from Cell Applications (San Diego, CA, USA).

The cells were maintained in Dulbecco’s modified Eaglemedium/nutrient mixture/F12 (DMEM; Gibco BRL, Rockville,MD, USA) containing 10% (v/v) fetal bovine serum (FBS;HyClone Laboratories, Logan, UT, USA), 1% (v/v) insulin-transferrin-selenium-X supplement (Invitrogen, Carlsbad, CA,USA), and 1% (v/v) penicillin-streptomycin solution (SigmaChemical, St. Louis, MO, USA) at 37◦C in a humidifiedatmosphere consisting of 95% air and 5% CO2.

For treatment with IL-1β (Genzyme/Techne, Minneapolis,MN, USA), the cells were seeded onto 100-mm tissue culturedishes at a density of 5 × 106 cells/cm2. After overnightincubation, the cells were cultured for up to 28 days in DMEMcontaining 10% FBS with 0, 10, or 100 U/mL IL-1β with orwithout 1 µM celecoxib (Astellas Pharma, Tokyo, Japan). Fully1 U of IL-1β corresponds to the unit activity in the colorimetricMTT-assay with CTLL-2 cells [18]. These IL-1β concentrationsare identical to those used in our previous studies [2, 3, 19],and are similar to that reported to occur in inflammatory TMJsynovial fluid [20].

Enzyme-Linked Immunosorbent Assay (ELISA)The cells were placed on 96-well microplates at a density

of 6 × 103 cells/cm2 and incubated in DMEM containing 10%(v/v) FBS with 0, 10, or 100 U/mL IL-1β with or without1 µM celecoxib for up to 28 days. The culture mediumwas then replaced with identical fresh medium, and the cellswere cultured for 24 hr. The amount of PGE2 in the culturemedium was determined using a commercially available ELISAkit (Biomedica Medizinprodukte GmbH & Co. KG, Vienna,Austria) according to the manufacturer’s instructions. Assayswere performed in quadruplicate, and the data were convertedto ng PGE2/mL.

Real-Time PCRCells were incubated in DMEM containing 10% (v/v) FBS

with 0, 10, or 100 U/mL IL-1β with or without 1 µM celecoxibfor up to 28 days. Total RNA was isolated from the cells ondays 1, 7, 14, 21, and 28 using an RNeasy Mini Kit (Qiagen,Valencia, CA, USA). The amounts of mRNA were equalizedusing a human β-actin competitive PCR kit (Takara Bio, Shiga,Japan), and the mRNA was then converted into cDNA usinga RNA PCR kit (GeneAmp; Perkin-Elmer, Branchburg, NJ,USA).

The cDNA mixtures were diluted 1:5 in sterile distilled water,and 2-µL aliquots of the diluted mixtures were subjected toreal-time PCR using Sybr Green I dye (BioWhittaker MolecularApplications, Rockland, ME, USA) in 25-µL reactions contain-ing 1× R-PCR buffer, 1.5 mM dNTP mixture, 1× Sybr Green I,15 mM MgCl2, 0.25 U Ex Taq (R-PCR version; Takara Bio), and0.2 µM of primers (sense and antisense) as shown in Table 1. ThePCR primers were designed using Primer3 software (WhiteheadInstitute for Biomedical Research, Cambridge, MA, USA).PCR was performed on a Smart Cycler II system (Cepheid,

Con

nect

Tis

sue

Res

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Frei

e U

nive

rsita

et B

erlin

on

12/0

5/14

For

pers

onal

use

onl

y.

188 Y. WATANABE ET AL.

TABLE 1Real-time PCR primers used

Target Primer sequence (5′-3′)GenBank

Acc.

COX-1 ACCTTGAAGGAGTCAGGCATGAG U63846TGTTCGGTGTCCAGTTCCAATA

COX-2 TTTCTACCAGAAGGGCAGGAT NM 000963TATCACAGGCTTCCATTGACC

EP 1 GGATGTACACCCAAGGGTCCAG NM 000955TCATGGTGGTGTCGTGCATC

EP 2 AGGACTGAACGCATTAGTCTCAGAA NM 000956CTCCTGGCTATCATGACCATCAC

EP 3 GGACTAGCTCTTCGGATAACT NM 000957GCAGTGCTCAACTGATGTCT

EP 4 AACTTGATGGCTGCGAAGACCTAC NM 000958TTCTAATATCTGGGCCTCTGCTGTG

GAPDH GCACCGTCAAGGCTGAGAAC NM 002046ATGGTGGTGAAGACGCCAGT

COX-1, -2 = cyclooxygenase; EP1-4 = prostaglandin receptors;GAPDH = glyceraldehy de-3 phosphate dehydrogenase.

Sunnyvale, CA, USA) using 45 cycles of 95◦C for 5 sec and55◦C for 10 sec, and measurements were made at the end of the60◦C annealing step. Results were analyzed using Smart Cyclersoftware (version 2.0d).

All real-time PCR reactions were performed in quadrupli-cate. Calculated values for gene expression levels were normal-ized to the levels of glyceraldehyde-3-phosphate dehydrogenase(GAPDH) mRNA at the same time point.

ImmunohistochemistryThe cells were placed on BD BioCoat Cellware (BD

Bioscience Discovery Labware, Bedford, MA, USA) at a densityof 6 × 103 cells/cm2 and incubated in DMEM containing10% (v/v) FBS with 0, 10, or 100 U/mL IL-1β with orwithout 1 µM celecoxib for up to 28 days. On days 21and 28, the cells were fixed with 4% (v/v) formalin for30 min and subjected to immunohistochemical staining asdescribed previously [21]. Briefly, the cells were reacted withpolyclonal antibodies specific for EP1–4 (sc-22648, 20675,20676, and 20677, respectively, at 1:500 dilutions; Santa CruzBiotechnology, Santa Cruz, CA, USA) for 30 min at 4◦C.The bound primary antibodies were detected using rhodamineisothiocyanate (RITC)-conjugated anti-goat (for EP1) or anti-rabbit (for EP2–4) IgG. Controls for immunostaining wereprocessed by omitting the primary antibody. Nuclei were stainedwith 4′,6-diamidino-2-phenylindole (DAPI) (BioGenex, SanRamon, CA, USA), and the cells were examined using anepifluorescence microscope (BZ-9000, Keyence, Osaka, Japan)and photographed digitally. In addition, the ratio of EP4-positivecells to total cells was examined on day 28 of culture.

Statistical AnalysisAll experiments were performed in quadruplicate. Each

value shown represents the mean ± standard deviation (S.D.).

Differences were determined using the Bonferroni’s correctionof the Student’s t-test or ANOVA, and differences with p values< 0.05 were considered significant.

RESULTS

Effects of IL-1� and Celecoxib on PGE2 ProductionChondrocyte PGE2 production was determined using ELISA

after human chondrocytes were cultured for 1, 7, 14, 21, or 28days in the absence or presence of 10 or 100 U/mL IL-1β. Withcells cultured in the absence of IL-1β, a minimal level of PGE2

was detected at each time point. In contrast, cells cultured inthe presence of IL-1β produced PGE2 levels markedly higherthan those of the controls at each time point, and the differenceincreased dramatically after 14 days of culture (Figure 1a).

The effect of celecoxib on the IL-1β-induced productionof PGE2 was also examined. When cells were cultured with1 µM celecoxib with or without 100 U/mL IL-1β for 21 or28 days, celecoxib had no effect on PGE2 production in theabsence of IL-1β. But it appeared to block the stimulation ofPGE2 production by IL-1β; the level of PGE2 was similar inIL-1β-treated and non-IL-1β-treated cells (Figure 1b).

COX-1 and COX-2 Gene ExpressionThe expression of the COX-1 and COX-2 genes in human

chondrocytes cultured in the presence of 0, 10, or 100 U/mLIL-1β was determined using real-time PCR. After 21 or 28 daysof culture, COX-1 expression was similar in IL-1β-treated andnontreated cells, but COX-2 expression was significantly higherin IL-1β-treated cells on both days (Figure 2a and b).

When 1 µM celecoxib was also present in the culture withor without 100 U/mL IL-1β for 21 or 28 days, it appearedto have no effect on COX-2 expression; the levels of COX-2expression were similar in IL-1β-treated and non-IL-1β-treatedcells treated with celecoxib (Figure 2c).

Expression of EP GenesThe expression of genes encoding the PG receptors EP1–4

was determined in chondrocytes cultured in the absence orpresence of IL-1β and celecoxib as described above. EP geneexpression was determined on days 1, 7, 14, 21, and 28 ofculture using real-time PCR. Toward the end of the cultureperiod, EP4 expression was significantly higher in IL-1β-treatedcells than in non-IL-1β-treated controls, and the increase wasIL-1β dose-dependent, whereas the levels of EP1 and EP2 geneexpression were significantly lower in IL-1β-treated cells thanin the controls (Figure 3). No expression of the EP3 receptorgene in either the presence or absence of IL-1β was detected.

Con

nect

Tis

sue

Res

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Frei

e U

nive

rsita

et B

erlin

on

12/0

5/14

For

pers

onal

use

onl

y.

IL-1β PRODUCES EP4 VIA PGE2 189

FIG. 1. Effects of IL-1β and celecoxib on PGE2 production in human chondrocytes. (a) Cells were cultured in the presence of 0, 10, or 100 U/mL IL-1β, andPGE2 production was determined using ELISA on days 1, 7, 14, 21, and 28 of culture. (b) Cells were cultured in the absence and presence of 1 µM celecoxib and100 U/mL IL-1β, and PGE2 production was determined by ELISA on days 21 and 28 of culture. Bars indicate means ± SD for four separate experiments; ∗p <

0.05, ∗∗p < 0.01, IL-1β, and/or celecoxib treatment versus control.

In the absence of IL-1β, EP gene expression was not affectedby the presence of 1 µM celecoxib in cultures of 21 or 28 days.However, EP gene expression was affected by celecoxib when100 U/mL IL-1β was also present; the level of EP4 mRNA incells treated with both IL-1β and celecoxib was similar to that innon-IL-1β-treated controls. However, celecoxib did not affectthe IL-1β-induced decrease in EP1 and EP2 gene expression oneither day (Figure 4).

Expression of EP ProteinsWe next examined the effects of IL-1β and celecoxib

on the expression of the EP1, EP2, and EP4 PG receptorproteins in chondrocytes. After cells were cultured for 21 or28 days in the presence and absence of IL-1β and celecoxibas described above, protein expression was quantified usingimmunohistochemical staining (Figure 5a). In cells culturedwith IL-1β, the level of EP4 protein was higher than in

the controls on both days, whereas the levels of EP1 andEP2 proteins were lower. Thus, the effects of IL-1β on EPprotein expression mirrored its effects on gene expression. Whencelecoxib was present in the culture with IL-1β, it appeared toblock the stimulatory effect of IL-1β on EP4 protein expression;the EP4 protein level in cells treated with both IL-1β andcelecoxib was similar to that in non-IL-1β-treated controls.Celecoxib did not appear to alter the inhibitory effects of IL-1β

on EP1 or EP2 protein expression on either day. In addition,the ratio of EP4-positive cells to total cells on day 28 of culturemirrored the gene expression (Figure 5b).

DISCUSSIONArthrosis of the TMJ results from an imbalance between

anabolic and catabolic processes predominantly controlled bychondrocytes [22, 23]. Cytokines released at sites of inflamma-tion and infection can alter the normal processes of cartilage

FIG. 2. Effects of IL-1β and celecoxib on COX-1 and COX-2 gene expression. (a) and (b) Cells were cultured in the presence of 0, 10, or 100 U/mL IL-1β,and COX-1 and COX-2 mRNA levels were determined using real-time PCR on days 21 and 28 of culture. (c) Cells were cultured in the absence and presence ofcelecoxib (1 µM) and IL-1β (100 U/mL), and COX-2 mRNA level was determined using real-time PCR on days 21 and 28 of culture. Bars indicate means ± SDfor four separate experiments; ∗∗p < 0.01, IL-1β, and/or celecoxib treatment versus control.

Con

nect

Tis

sue

Res

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Frei

e U

nive

rsita

et B

erlin

on

12/0

5/14

For

pers

onal

use

onl

y.

190 Y. WATANABE ET AL.

FIG. 3. Effect of IL-1β on expression of PG receptor genes. Levels of theEP1, EP2, and EP4 mRNAs in cells cultured for the indicated number of daysin the presence of 0, 10, or 100 U/mL IL-1β were determined using real-timePCR. Bars indicate means ± SD for four separate experiments; ∗p < 0.05,∗∗p< 0.01, IL-1β treatment versus control.

turnover, resulting in pathologic destruction or formation [1].Also the levels of inflammatory cytokines are higher than normalin the synovial fluid of TMJ arthrosis patients [24, 25]. Severalstudies [20, 26–28] have shown elevated levels of the cytokineIL-1β in synovial fluid samples obtained from TMJ patientswith osteoarthritis, and elevated concentrations of IL-1β insynovial fluid have been implicated in joint cartilage destruction.In addition, IL-1β in synovial fluid not only plays a central rolein the pathophysiology of cartilage damage and degradation,but also has been associated with pain and hyperalgesia in TMJ[27]. However, little is known about the EP subtypes in humanchondrocytes or about the long-term effects of IL-1β stimulationon EP expression in these cells. In addition, there is no reportthat examines the effect of IL-1β on the production of PGE2

and EP receptors using TMJ-derived chondrocytes. Therefore,we assumed cartilage damage and degradation in arthrosis ofTMJ and conducted the present study.

FIG. 4. Effect of celecoxib on the expression of PG receptor genes. Levels ofthe EP1, EP2, and EP4 mRNAs in cells cultured for 21 or 28 days in the presenceor absence of celecoxib (1 µM) and IL-1β (100 U/mL) were determined usingreal-time PCR. Bars indicate means ± SD for four separate experiments; ∗p <

0.05, ∗∗p < 0.01, IL-1β, and/or celecoxib treatment versus control.

In the present study, we found that IL-1β stimulatesexpression of the PG receptor EP4 in human chondrocytesby increasing PGE2 production, and also stimulates PGE2

production by increasing the expression of COX-2. In designingthese experiments, we first had to determine an appropriatelength for the culture period. In many previous studies of PGE2

and EP expression in chondrocytes, the culture periods wereshort (24 or 48 hr) [15, 17, 29]. However, TMJ arthritis oftenundergoes a chronic progression. Therefore, we examined thePGE2 production and EP expression levels in these cells over amuch longer culture period of 28 days.

The inflammatory process in the TMJ is complex; it isorchestrated by both infiltrating cells and mesenchymallyderived cells, including synovial fibroblasts and chondrocytes.Proinflammatory cytokines are believed to play important rolesin the initiation and development of joint disease [30–32].

Con

nect

Tis

sue

Res

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Frei

e U

nive

rsita

et B

erlin

on

12/0

5/14

For

pers

onal

use

onl

y.

IL-1β PRODUCES EP4 VIA PGE2 191

FIG. 5. Immunohistochemical staining of PG receptor expression in human chondrocytes. (a) Cells were cultured in the presence or absence of celecoxib(1 µM) and IL-1β (10 or 100 U/mL as indicated) for 21 or 28 days, and the expression of EP1, EP2, and EP4 was examined by immunohistochemical staining.In the presence of IL-1β, the RITC fluorescence (red) signal of EP4 increased in a dose-dependent manner on both days, whereas that of EP1 and EP2 decreased.When celecoxib was also present, the IL-1β-induced EP4 RITC fluorescence signal was reduced to that of the control, but the IL-1β-induced decreases in EP1and EP2 fluorescence remained. Blue in the DAPI fluorescence shows the nucleus of the cell. Bar = 50 µm. (b) The ratio of EP4-positive cells to total cells wasexamined on day 28 of culture. Bars indicate means ± SD for four separate experiments; ∗p < 0.05, ∗∗p < 0.01, IL-1β, and/or celecoxib treatment versus control.

Two cytokines that play prominent roles in the developmentof synovial inflammation and the etiopathology of joint diseaseare tumor necrosis factor-α (TNF-α) and IL-1. IL-1 stimulatessynoviocytes to secrete soluble mediators, including collagenase[33], COX-2 [34], PGE2 [33], and IL-6 [35].

In this study, we found that IL-1β dose-dependently andmarkedly increased the production of PGE2 in the later part of

the culture period. It also significantly increased the expressionof COX-2, which is closely linked to PGE2 production ininflammation, in the later part of the culture period, whereasit did not affect the expression of COX-1, which functions inhomeostasis. In addition, simultaneous culture with celecoxib,which specifically inhibits COX-2, blocked the induction ofPGE2 production by IL-1β but did not block the induction of

Con

nect

Tis

sue

Res

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Frei

e U

nive

rsita

et B

erlin

on

12/0

5/14

For

pers

onal

use

onl

y.

192 Y. WATANABE ET AL.

COX-2 expression by IL-1β. These results suggest that IL-1β

promotes PGE2 production in chondrocytes by increasing COX-2 expression, and that it does not promote COX-2 expressionby increasing PGE2 production. In other words, IL-1β directlystimulates the expression of COX-2.

The effects of PGE2 on specific cells are greatly dependenton its binding to different subtypes of PG receptors. Four humansubtypes of the PG receptors have been identified: EP1, EP2,EP3, or EP4 according to their different signalling pathways[36]. Previous reports have described various effects of IL-1β

on EP expression. In recent studies, IL-1β variously was shownto upregulate the expression of EP2 and EP4 in human synovialfibroblasts [37], to upregulate EP4 and downregulate EP1 inhuman cervical fibroblasts [38], and to upregulate both EP1 andEP4 in rabbit chondrocytes [39].

In a study of mouse synovial fibroblasts, EP4 was the mostabundant PG receptor isoform, whereas EP1 and EP2 wereat low levels, and the EP3 receptor was not expressed [40].However, since little is known about the expression of EPsubtypes in human chondrocytes or the effects of IL-1β onthis expression, we examined EP subtype expression in thesecells in the absence and presence of IL-1β over 28 daysof culture. We found that IL-1β significantly increased EP4expression in the later part of the culture period, but significantlydecreased EP1 and EP2 expression; no EP3 expression wasdetected in either the absence or presence of IL-1β. Theseresults suggest that by increasing EP4 expression directly orindirectly, IL-1β facilitates the binding of PGE2 to EP4 inchondrocytes. In addition, based on our results and on otherreports, the expression of EP genes and their responses to IL-1β

may differ among animal species or even among different celltypes in the same species.

Stimulation by various cytokines causes cells to releasePGE2, which binds to EP receptor proteins on the same orother cells and enhances various physiologic activities [40,41]. Shoji et al. [41] have shown that lipopolysaccharideenhances the production of nicotine-induced PGE2 by anincrease in COX-2 expression in osteoblasts. White et al.[42] have shown that PGE2 mediates IL-1β-related fibroblastmitogenic effects in acute lung injury through differentialutilization of prostanoid receptors. Therefore, we hypothesizedthat the IL-1β-stimulated increase in EP4 expression and/or theIL-1β-stimulated reduction in EP1 and EP2 expression werecaused not directly by IL-1β but rather by an indirect effect ofthe IL-1β-stimulated increase in PGE2 production.

Thus, we determined the effect of the COX-2 inhibitorcelecoxib on the expression of EP1, EP2, and EP4 in the absenceand presence of IL-1β on days 21 and 28 of culture. In thisexperiment, celecoxib blocked the IL-1β-mediated stimulationof EP4 expression, but it did not affect IL-1β-mediatedreduction in EP1 or EP2 expression. These results suggest thatIL-1β induces EP4 receptor expression in chondrocytes by anautocrine mechanism that increases PGE2 production, whereasit reduces the expression of EP1 and EP2 via direct action.

CONCLUSIONOur results indicate that IL-1β increases the production of

PGE2, COX-2, and EP4 in chondrocytes derived from normalhuman femoral cartilage, and that it may promote the expressionof EP4 by increasing PGE2 production in these cells. In thefuture, we will continue to examine the signalling pathwayslinking EP expression and IL-1β.

ACKNOWLEDGMENTSThis study was supported by a Grant-in-Aid for Scientific

Research (C) and a Grant-in-Aid for Young Scientists (B)from the Japanese Society for the Promotion of Science(19592182 and 19791384, respectively), and by a NihonUniversity Multidisciplinary Research Grant (2006–2007) anda Nihon University Research Grant for Assistants and YoungResearchers (2007). This study was also supported by specialresearch grants for the development of distinctive educationfrom the Promotion and Mutual Aid Corporation for PrivateSchools of Japan and the Sato Fund of Nihon University Schoolof Dentistry.

DECLARATION OF INTERESTThe authors report no conflicts of interest. The authors alone

are responsible for the content and writing of this article.

REFERENCES1. Shinoda, C., and Takaku, S. (2000). Interleukin-1 beta, interleukin-6,

and tissue inhibitor of metalloproteinase-1 in the synovial fluid of thetemporomandibular joint with respect to cartilage destruction. Oral Dis., 6,383–390.

2. Aida, Y., Maeno, M., Suzuki, N., Shiratsuchi, H., Motohashi, M., andMatsumura, H. (2005). The effect of IL-1β on the expression of matrixmetalloproteinases and tissue inhibitor of matrix metalloproteinases inhuman chondrocytes. Life Sci., 77, 3210–3221.

3. Aida, Y., Maeno, M., Suzuki, N., Namba, A., Motohashi, M., Matsumoto,M., Makimura, M., and Matsumura, H. (2006). The effect of IL-1β onthe expression of inflammatory cytokines and their receptors in humanchondrocytes. Life Sci., 79, 764–771.

4. Namba, A., Aida, Y., Suzuki, N., Watanabe, Y., Kawato, T., Motohashi, M.,Maeno, M., Matsumura, H., and Matsumoto, M. (2007). Effects of IL-6and soluble IL-6 receptor on the expression of cartilage matrix proteins inhuman chondrocytes. Connect. Tissue Res., 48, 263–270.

5. Quinn, J.H., and Bazan, N.G. (1990). Identification of prostaglandinE2 and leukotriene B4 in the synovial fluid of painful, dysfunctionaltemporomandibular joints. J. Oral Maxillofac. Surg., 48, 968–971.

6. Murakami, K.I., Shibata, T., and Kubota, E. (1998). Intra-articular levelsof prostaglandin E2, hyaluronic acid, and chondroitin-4 and -6 sulfatesin the temporomandibular joint synovial fluid of patients with internalderangement. J. Oral Maxillofac. Surg., 56, 199–203.

7. Alstergren, P., and Kopp, S. (2000). Prostaglandin E2 in temporomandibularjoint synovial fluid and its relation to pain and inflammatory disorders. J.Oral Maxillofac. Surg., 58, 180–186.

8. Prete, P.E., and Gurakar-Osborne, A. (1997). The contribution of syn-ovial fluid lipoproteins to the chronic synovitis of rheumatoid arthritis.Prostaglandins, 54, 689–698.

9. Sugimoto, Y., Narumiya, S., and Ichikawa, A. (2000). Distribution andfunction of prostanoid receptors: studies from knockout mice. Prog. LipidRes., 39, 289–314.

Con

nect

Tis

sue

Res

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Frei

e U

nive

rsita

et B

erlin

on

12/0

5/14

For

pers

onal

use

onl

y.

IL-1β PRODUCES EP4 VIA PGE2 193

10. Kudo, I., and Murakami, M. (2002). Phospholipase A2 enzymes.Prostaglandins Other Lipid Mediat., 58–59, 3–58.

11. Simmons, D.L., Botting, R.M., and Hla, T. (2004). Cyclooxygenaseisozymes: The biology of prostaglandin synthesis and inhibition. Phar-macol. Rev., 56, 387–437.

12. Helliwell, R.J., Adams, L.F., and Mitchell, M.D. (2004). Prostaglandinsynthases: Recent developments and a novel hypothesis. ProstaglandinsLeukot. Essent. Fatty Acids, 70, 101–113.

13. Narumiya, S., and Fitzgerald, G.A. (2001). Genetic and pharmacologicalanalysis of prostanoid receptor function. J. Clin. Invest., 108, 25–30.

14. Del Toro, F. Jr., Sylvia, V.L., Schubkegel, S.R., Campos, R., Dean, D.D.,Boyan, B.D., and Schwartz, Z. (2000). Characterization of prostaglandinE2 receptors and their role in 24,25-(OH)2D3-mediated effects on restingzone chondrocytes. J. Cell Physiol., 182, 196–208.

15. Sylvia, V.L., Del Toro, F. Jr., Hardin, R.R., Dean, D.D., Boyan, B.D., andSchwartz, Z. (2001). Characterization of PGE2 receptors (EP) and theirrole as mediators of 1α,25-(OH)2D3 effects on growth zone chondrocytes.J. Steroid Biochem. Mol. Biol., 78, 261–274.

16. Kobayashi, T., and Narumiya, S. (2002). Function of prostanoid receptors:Studies on knockout mice. Prostaglandins Other Lipid Mediat., 68–69,557–573.

17. Brochhausen, C., Neuland, P., Kirkpatrick, C.J., Nusing, R.M., and Klaus,G. (2006). Cyclooxygenases and prostaglandin E2 receptors in growth platechondrocytes in vitro and in situ-prostaglandin E2 dependent proliferationof growth plate chondrocytes. Arthr. Res. Ther., 8, R78.

18. Gillis, S., Ferm, M.M., Ou, W., Smith, K.A. (1978). T cell growth factor:parameters of production and a quantitative microassay for activity. J.Immunol., 120, 2027–2032.

19. Aida, Y., Maeno, M., Ito-Kato, E., Suzuki, N., Shiratsuchi, H., andMatsumura, H. (2004). Effect of IL-1α on the expression of cartilagematrix proteins in human chondrosarcoma cell line OUMS-27. Life Sci.,75, 3173–3184.

20. Alstergren, P., Benavente, C., and Kopp, S. (2003). Interleukin-1β,interleukin-1 receptor antagonist, and interleukin-1 soluble receptor IIin temporomandibular joint synovial fluid from patients with chronicpolyarthritides. J. Oral Maxillofac. Surg., 61, 1171–1178.

21. Zohar, R., Suzuki, N., Suzuki, K., Arora, P., Glogauer, M., Mc. Culloch,C.A., and Sodek, J. (2000). Intracellular osteopontin is an integralcomponent of the CD44-ERM complex involved in cell migration. J. CellPhysiol., 184, 118–130.

22. Poole, A.R. (1999). An introduction to the pathophysiology of osteoarthri-tis. Front. Biosci., 4, d662–d670.

23. Frenkel, S.R., and Di Cesare, P.E. (1999). Degradation and repair of articularcartilage. Front. Biosci., 4, d671–d685.

24. Sandler, N.A., Buckley, M.J., Cillo, J.E., and Braun, T.W. (1998).Correlation of inflammatory cytokines with arthroscopic findings in patientswith temporomandibular joint internal derangements. J. Oral Maxillofac.Surg., 56, 534–543.

25. Alstergren, P., Kopp, S., and Theodorsson, E. (1999). Synovial fluidsampling from the temporomandibular joint: Sample quality criteria andlevels of interleukin-1 beta and serotonin. Acta Odontol. Scand., 57, 16–22.

26. Kubota, E., Imamura, H., Kubota, T., Shibata, T., and Murakami, K.(1997). Interleukin 1β and stromelysin (MMP3) activity of synovial fluidas possible markers of osteoarthritis in the temporomandibular joint. J. OralMaxillofac. Surg., 55, 20–27.

27. Alstergren, P., Ernberg, M., Kvarnstrom, M., and Kopp, S. (1998).Interleukin-1β in synovial fluid from the arthritic temporomandibular jointand its relation to pain, mobility, and anterior open bite. J. Oral Maxillofac.Surg., 56, 1059–1065.

28. Takahashi, T., Kondoh, T., Fukuda, M., Yamazaki, Y., Toyosaki, T., andSuzuki, R. (1998). Proinflammatory cytokines detectable in synovial fluidsfrom patients with temporomandibular disorders. Oral Surg. Oral Med.Oral Pathol. Oral Radiol. Endod., 85, 135–141.

29. de Brum-Fernandes, A.J., Morisset, S., Bkaily, G., and Patry, C. (1996).Characterization of the PGE2 receptor subtype in bovine chondrocytes inculture. Br. J. Pharmacol., 118, 1597–1604.

30. Feldmann, M., Brennan, F.M., and Maini, R.N. (1996). Role of cytokinesin rheumatoid arthritis. Annu. Rev. Immunol., 14, 397–440.

31. Westacott, C.I., and Sharif, M. (1996). Cytokines in osteoarthritis:Mediators or markers of joint destruction? Semin. Arthr. Rheum., 25,254–272.

32. Isomaki, P., and Punnonen, J. (1997). Pro- and anti-inflammatory cytokinesin rheumatoid arthritis. Ann. Med., 29, 499–507.

33. Dayer, J.M., De Rochemonteix, B., Burrus, B., Demczuk, S., and Dinarello,C.A. (1986). Human recombinant interleukin 1 stimulates collagenase andprostaglandin E2 production by human synovial cells. J. Clin. Invest., 77,645–648.

34. Rauk, P.N., and Chiao, J.P. (2000). Interleukin-1 stimulates humanuterine prostaglandin production through induction of cyclooxygenase-2expression. Am. J. Reprod. Immunol., 43, 152–159.

35. Guerne, P.A., Zuraw, B.L., Vaughan, J.H., Carson, D.A., and Lotz, M.(1989). Synovium as a source of interleukin 6 in vitro. Contribution to localand systemic manifestations of arthritis. J. Clin. Invest., 83, 585–592.

36. Narumiya, S., Sugimoto, Y., and Ushikubi, F. (1999). Prostanoid receptors:Structures, properties, and functions. Physiol. Rev., 79, 1193–1226.

37. Inoue, H., Takamori, M., Shimoyama, Y., Ishibashi, H., Yamamoto, S., andKoshihara, Y. (2002). Regulation by PGE2 of the production of interleukin-6, macrophage colony stimulating factor, and vascular endothelial growthfactor in human synovial fibroblasts. Br. J. Pharmacol., 136, 287–295.

38. Schmitz, T., Leroy, M.J., Dallot, E., Breuiller-Fouche, M., Ferre, F., andCabrol, D. (2003). Interleukin-1β induces glycosaminoglycan synthesis viathe prostaglandin E2 pathway in cultured human cervical fibroblasts. Mol.Hum. Reprod., 9, 1–8.

39. Alvarez-Soria, M.A., Largo, R., Sanchez-Pernaute, O., Calvo, E., Egido,J., and Herrero-Beaumont, G. (2007). Prostaglandin E2 receptors EP1 andEP4 are up-regulated in rabbit chondrocytes by IL-1β, but not by TNF-α.Rheumatol. Int., 27, 911–917.

40. Wei, X., Zhang, X., Zuscik, M., Drissi, M.H., Schwarz, E.M., and O’Keefe,R.J. (2005). Fibroblasts express RANKL and support osteoclastogenesisin a COX-2-dependent manner after stimulation with titanium particles. J.Bone Miner. Res., 20, 1136–1148.

41. Shoji, M., Tanabe, N., Mitsui, N., Suzuki, N., Takeichi, O., Katono,T., Morozumi, A., and Maeno, M. (2007). Lipopolysaccharide enhancesthe production of nicotine-induced prostaglandin E2 by an increase incyclooxygenase-2 expression in osteoblasts. Acta Biochim. Biophys. Sin.,39, 163–172.

42. White, K.E., Ding, Q., Moore, B.B., Peters-Golden, M., Ware, L.B.,Matthay, M.A., and Olman, M.A. (2008). Prostaglandin E2 mediatesIL-1β-related fibroblast mitogenic effects in acute lung injury throughdifferential utilization of prostanoid receptors. J. Immunol., 180, 637–646.

Con

nect

Tis

sue

Res

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Frei

e U

nive

rsita

et B

erlin

on

12/0

5/14

For

pers

onal

use

onl

y.