Signalling pathway involved in nitric oxide synthase type II activation in chondrocytes: synergistic effect of leptin with interleukin-1

Open Access A ailable online h p://a h i is- esea ch.com/con en /7/3/R581 R581 Vol 7 No 3 Resea ch a icle Signalling pa hway in ol ed in ni ic oxide syn hase ype II ac i a ion in chond ocy es: syne gis ic e ec o lep in wi h in e leukin-1 Miguel O e o1, Rocío Lago1, F ancisca Lago2, Juan Jesús Gomez Reino3,4 and O es e Gualillo1 1NEIRID (Neu oEndoc ine In e ac ions in Rheuma ology and In lamma o y Diseases) Labo a o y, San iago Uni e si y Clinical Hospi al, Resea ch Labo a o y 4, San iago de Compos ela, Spain 2Labo a o y o Molecula and Cellula Ca diology, San iago Uni e si y Clinical Hospi al, Resea ch Labo a o y 1, San iago de Compos ela, Spain 3Rheuma ology Di ision, San iago Uni e si y Clinical Hospi al, San iago de Compos ela, Spain 4Depa men o Medicine, School o Medicine, Uni e si y o San iago de Compos ela, San iago de Compos ela, Spain Co esponding au ho : O es e Gualillo, gualill[email p o ec ed] Recei ed: 11 Aug 2004 Re isions eques ed: 16 Sep 2004 Re isions ecei ed: 14 Jan 2005 Accep ed: 3 Feb 2005 Published: 4 Ma 2005 A h i is Resea ch & The apy 2005, 7:R581-R591 (DOI 10.1186/a 1708) This a icle is online a : h p://a h i is- esea ch.com/con en /7/3/R581 © 2005 O e o e al.; licensee BioMed Cen al L d. This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (h p://c ea i ecommons.o g/licenses/by/ 2.0), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. Abs ac The objec i e o he p esen s udy was o in es iga e he e ec o lep in, alone o in combina ion wi h IL-1, on ni ic oxide syn hase (NOS) ype II ac i i y in i o in human p ima y chond ocy es, in he mouse chond ogenic ATDC5 cell line, and in ma u e and hype ophic ATDC5 di e en ia ed chond ocy es. Fo comple eness, we also in es iga ed he signalling pa hway o he pu a i e syne gism be ween lep in and IL-1. Fo his pu pose, ni ic oxide p oduc ion was e alua ed using he G iess colo ime ic eac ion in cul u e medium o cells s imula ed o e 48 hou s wi h lep in (800 nmol/l) and IL-1 (0.025 ng/ml), alone o combined. Speci ic pha macological inhibi o s o NOS ype II (aminoguanidine [1 mmol/l]), janus kinase (JAK)2 ( y phos in AG490 and Tkip), phospha idylinosi ol 3-kinase (PI3K; wo mannin [1, 2.5, 5 and 10 µmol/l] and LY294002 [1, 2.5, 5 and 10 µmol/l]), mi ogen-ac i a ed p o ein kinase kinase (MEK)1 (PD098059 [1, 5, 10, 20 and 30 µmol/l]) and p38 kinase (SB203580 [1, 5, 10, 20 and 30 µmol/l]) we e added 1 hou be o e s imula ion. Ni ic oxide syn hase ype II mRNA exp ession in ATDC5 chond ocy es was in es iga ed by eal- ime PCR and NOS II p o ein exp ession was analyzed by wes e n blo . Ou esul s indica e ha s imula ion o chond ocy es wi h IL-1 esul s in dose-dependen ni ic oxide p oduc ion. In con as , lep in alone was unable o induce ni ic oxide p oduc ion o exp ession o NOS ype II mRNA o i s p o ein. Howe e , co-s imula ion wi h lep in and IL-1 esul ed in a ne inc ease in ni ic oxide concen a ion o e IL-1 challenge ha was elimina ed by p e ea men wi h he NOS II speci ic inhibi o aminoguanidine. P e ea men wi h y phos in AG490 and Tkip (a SOCS-1 mime ic pep ide ha inhibi s JAK2) blocked ni ic oxide p oduc ion induced by lep in/IL-1. Finally, wo mannin, LY294002, PD098059 and SB203580 signi ican ly dec eased ni ic oxide p oduc ion. These indings we e con i med in ma u e and hype ophic ATDC5 chond ocy es, and in human p ima y chond ocy es. This s udy indica es ha lep in plays a p oin lamma o y ole, in syne gy wi h IL-1, by inducing NOS ype II h ough a signalling pa hway ha in ol es JAK2, PI3K, MEK-1 and p38 kinase. In oduc ion Chond ocy es a e he p edominan cells in ma u e ca ilage ha syn hesize and main ain he in eg i y o ca ilage-speci ic ex acellula ma ix. In heuma oid a h i is and os eoa h i is he pheno ype o chond ocy es changes, and apop osis and ex acellula ma ix deg ada ion occu [1-3]. These se e e pe - u ba ions in ca ilage homeos asis may be media ed in pa by ni ic oxide (NO). This gaseous media o is induced by se e al p oin lamma o y cy okines, including IL-1. Lep in, he OB gene p oduc , is a 16 kDa ho mone ha is syn- hesized by adipocy es. Lep in egula es ood in ake and ERK = ex acellula signal- egula ed kinase; GAPDH = glyce aldehyde-3-phospha e dehyd ogenase; IFN = in e e on; IL = in e leukin; JAK = janus kinase; MAPK = mi ogen-ac i a ed p o ein kinase; MEK = mi ogen-ac i a ed p o ein kinase kinase; MMP = ma ix me allop o einase; NF-κB = nuclea ac o -κB; NO = ni ic oxide; NOS = ni ic oxide syn hase; PBS = phospha e-bu e ed saline; PI3K = phospha idylinosi ol 3-kinase; RT-PCR = e e se ansc ip ion polyme ase chain eac ion; SOCS = supp esso o cy okine signalling. A h i is Resea ch & The apy Vol 7 No 3 O e o e al. R582 ene gy expendi u e, bu i also modula es neu oend oc ine unc ion [4]. I is in ol ed in immune modula ion in ha i in lu- ences he inna e immune esponse by p omo ing ac i a ion o monocy e/mac ophages, chemo axis and ac i a ion o neu- ophils, and ac i a ion o na u al kille cells [5]. Fu he mo e, lep in in luences adap i e immuni y by inc easing he exp es- sion o adhesion molecules by CD4+ T cells, and p omo ing p oli e a ion and sec e ion o IL-2 by naï e CD4+ T cells [5-7]. Lep in has also been ound o in luence bone g ow h [8] and in lamma ion [9]. High lep in le els a e associa ed wi h obesi y, which is a isk ac o o os eoa h i is [10-12]. In e es ingly, in pa ien s wi h os eoa h i is lep in is p esen in syno ial luid and is exp essed by a icula chond ocy es [13], and no mal human chond o- cy es exp ess he unc ional Ob-Rb lep in ecep o iso o m [14]. I is unlikely ha lep in alone ac s on ca ilage o igge an in lamma o y esponse; a he , i may associa e wi h o he p oin lamma o y cy okines o ampli y in lamma ion and enhance damage o ca ilage. We ecen ly demons a ed a syne gis ic e ec o lep in wi h IFN-γ on ni ic oxide syn hase (NOS) ype II ac i i y in cul u ed chond ocy es ha was medi- a ed by he janus kinase (JAK)2 [15]. In he p esen s udy we in es iga ed whe he lep in syne gizes wi h IL-1, an abundan media o o in lamma ion and ca ilage des uc ion [16,17], o ac i a e NOS ype II in chond ocy es. To gain u he insigh s in o he mechanism o ac ion o his pu a i e syne gism, we also analyzed he ole played by se e al in acellula kinases by using speci ic pha macological inhibi o s. Ma e ials and me hods Reagen s Foe al bo ine se um, issue cul u e media, media supple- men s, mouse and human ecombinan lep in, mouse ecom- binan IL-1, y phos in AG490, wo mannin, LY294002, PD098059 and SB203580 we e pu chased om Sigma (S Louis, MO, USA) unless o he wise speci ied. RT-PCR ea- gen s we e pu chased om In i ogen (Ca lsbad, CA, USA) and S a agene (La Jolla, CA, USA). Tkip (WLVFFVIFYFFR), a supp esso o cy okine signalling (SOCS)-1 mime ic pep ide ha inhibi s JAK2 au ophospho yla ion, was gene ously p o- ided by D Howa d M Johnson (Ins i u e o Food and Ag icul- u al Science, Depa men o Mic obiology and Cell Science, Uni e si y o Flo ida, Gaines ille, FL, USA). Cell cul u e The clonal chond ogenic cell line ATDC5 was chosen o hese s udies because i has been shown o be a use ul in i o model o examining he mul is ep di e en ia ion o chond o- cy es. Undi e en ia ed ATDC5 cells p oli e a e apidly un il hey each con luence, a which poin hey unde go g ow h a es . When ea ed wi h insulin, ans e in and sodium seleni e, con luen ATDC5 cells e-en e a p oli e a i e phase and o m ca ilaginous ma ix nodules (ma u e chond ocy es). As di e en ia ion p og esses, hese cells unde go a la e di e - en ia ion phase, becoming hype ophic, calci ying chond o- cy es ha syn hesize ype X collagen and os eopon in – a ma ke o e minal chond ocy e di e en ia ion [18]. ATDC5 cells we e a kind gi om D Agamemnon E G igo iadis (Depa men o C anio acial De elopmen , King's College, London Guy's Hospi al, London, UK). Unless o he wise spec- i ied, cells we e cul u ed in Dulbecco's modi ied Eagle's medium/Hams' F12 medium supplemen ed wi h 5% oe al bo ine se um, 10 µg/ml human ans e in, 3 × 10-8 mol/l sodium seleni e and an ibio ics (50 U/ml penicillin and 50 µg/ ml s ep omycin). In some expe imen s, conduc ed o demons a e ha lep in/IL- 1 syne gism does no appea o depend on he di e en ia ion s a e o he chond ocy es, chond ogenic ATDC5 cells we e di e en ia ed in o ma u e and hype ophic chond ocy es, as desc ibed by Thomas and cowo ke s [19]. B ie ly, cells we e pla ed a an ini ial densi y o 2 × 104 cells/well in 24-well pla es. Cells we e cul u ed in he abo e-men ioned medium supplemen ed wi h 10 µg/ml o human ecombinan insulin (No o No disk A/S, Bags ae d, Denma k). Cul u e was con in- ued o a u he 15 o 21 days, wi h eplacemen o medium e e y o he day. As expec ed, ATDC5 cul u es ea ed wi h insulin unde wen p og essi e di e en ia ion om 0 o 21 days as compa ed wi h un ea ed cul u es. This di e en ia ion was quali a i ely cha ac e ized by inc eased o ma ion o ca ilage nodules and enhanced s aining wi h alcian blue dye, which is indica i e o ca ilage p o eoglycan accumula ion. In o he expe imen s (da a no shown), he di e en ia ion om days 0 o 21 was u he e idenced by sequen ial inc eases in ype II collagen, agg ecan and ype X collagen mRNAs. The ea ly and ma u e chond ocy e ma ke ype II collagen was exp essed in undi e en ia ed ATDC5 cells; he le el began o inc ease a day 3, peaked a days 7–10 and g adually declined a e day 15. The exp ession p o ile o agg ecan mimicked ha o ype II collagen bu wi h a sligh delay o a couple o days. The decline in exp ession o bo h chond ocy e ma ke s coin- cided wi h he onse o la e-s age chond ocy e di e en ia ion. The exp ession o he hype ophic chond ocy e ma ke ype X collagen began a days 12 and 13. The exp ession pa e ns o hese ea ly and la e chond ocy e ma ke s we e consis en wi h p e ious indings in ATDC5 cells ega ding in i o chond o- cy e di e en ia ion. We do no illus a e indings ega ding he di e en ia ion o ATDC5 cells because hey a e ex ensi ely epo ed in li e a u e [19]. Ca ilage ha es and human chond ocy e isola ion Human no mal a icula ca ilage samples we e ob ained om knee join s o pa ien s unde going leg ampu a ions om abo e he knee because o pe iphe al ascula disease. (Pe mission om he local e hical commi ee was g an ed.) None o he pa ien s had a clinical his o y o a h i is o any o he pa hology a ec ing he ca ilage, and he specimens appea ed no mal on mo phological examina ion (no change in colou and no A ailable online h p://a h i is- esea ch.com/con en /7/3/R581 R583 ib illa ion). Fo chond ocy e isola ion, asep ically dissec ed ca ilage was subjec ed o sequen ial diges ion wi h p onase (ca alogue numbe 165921; Roche Molecula Biochemicals, Indianapolis, IN, USA) and collagenase P (ca alogue numbe 1213873; Roche Molecula Biochemicals) a a inal concen- a ion o 1 mg/ml in Dulbecco's modi ied Eagle's medium/F12 plus 10% oe al cal se um and s e ilized by il a ion, in acco dance wi h he manu ac u e 's ins uc ions. In ou hands, his p ocedu e was supe io o enzyma ic isola ion wi h colla- genase alone in e ms o chond ocy e yields and capaci y o a achmen . Ca ilage specimens we e inely diced in phos- pha e-bu e ed saline (PBS), and a e emo ing PBS diced issue was incuba ed o 30 min wi h p onase in a shaking wa e ba h a 37°C. P onase was subsequen ly emo ed om he diges ion lask and he ca ilage pieces we e washed wi h PBS. A e emo al o PBS, diges ion was con inued wi h addi ion o collagenase P; his was done o e 6–8 hou s in a shaking wa e ba h a 37°C. The esul ing cell suspension was il e ed h ough a 40 µm nylon cell s aine (BD Biosciences Eu ope, E embodegem, Belgium) in o de o emo e deb is. Cells we e cen i uged and washed wice wi h PBS, coun ed and pla ed in 24-well issue cul u e pla es o chond ocy e cul- u e. Cells we e se ially passaged o ob ain a su icien numbe o cells and used be ween he i s and second passages. Cell ea men s and ni i e assay ATDC5 cells and human p ima y chond ocy es, wi h a iabili y g ea e han 95% as e alua ed using he ypan blue exclusion me hod, we e cul u ed (as desc ibed abo e) in 24-well pla es. A e 12 hou s o s a a ion in se um- ee medium, cells we e s imula ed o 48 hou s wi h lep in (800 nmol/l), alone o in combina ion wi h IL-1 (0.025 ng/ml). We wished o de e mine whe he inc eased NO p oduc ion was due o NOS ype II ac i a ion and o he in ol emen o JAK2, phospha idylinosi ol 3-kinase (PI3K), mi ogen-ac i a ed p o ein kinase kinase (MEK)1 and p38 kinase. Fo his pu pose, he ollowing spe- ci ic pha macological inhibi o s we e added 1 hou be o e cy okine s imula ion: aminoguanidine (1 mmol/l) o NOS ype II; y phos in AG490 (5 and 10 µmol/l) and Tkip (20 and 50 µmol/l) o JAK2; wo mannin (1, 2.5, 5 and 10 µmol/l) and LY294002 (1, 2.5, 5 and 10 µmol/l) o PI3K; PD098059 (1, 5, 10, 20 and 30 µmol/l) o MEK-1; and SB203580 (1, 5, 10, 20 and 30 µmol/l) o p38 kinase. Cy okines and pha maco- logical inhibi o doses we e selec ed on he basis o p io dose– esponse expe imen s (da a no shown) o p e iously published li e a u e [15]. Ni i e accumula ion was measu ed in cul u e medium using he G iess eac ion. B ie ly, 100 µl cell cul u e medium was mixed wi h 100 µl G iess eagen (equal olumes o 1% [weigh / ol] sul anilamide in 5% [ ol/ ol] phospho ic acid and 0.1% [weigh / ol] naph yle hylenediamine-HCl), incuba ed a oom empe a u e o 10 min, and hen he abso bance a 550 nm was measu ed using a mic opla e eade (Ti e ek-Mul i- scan, Labsys em, Helsinki, Finland). F esh cul u e medium was used as blank in all o he expe imen s. The amoun o ni i e in he samples (in mic omola uni s) was calcula ed om a sodium ni i e s anda d cu e eshly p epa ed in cul u e medium. RNA isola ion and eal- ime RT-PCR ATDC5 chond ogenic cells we e seeded in P6 well pla es o each 85–90% con luence. A e 8 hou s o s a a ion in se um- ee medium, cells we e ea ed wi h lep in alone o in combina ion wi h IL-1. In o de o es he in ol emen o JAK2, PI3K, MEK-1 and p38 kinase on NOS ype II mRNA exp es- sion, speci ic inhibi o s ( y phos in AG490 10 µmol/l, wo - mannin and LY294002 10 µmol/l, PD098059 30 µmol/l and SB203580 30 µmol/l) we e added 1 hou be o e cy okine s imula ion. A e 48 hou s o ea men , RNA was isola ed om cell cul u e using he T izol-LS®TM me hod (Gibco-BRL, Li e Technologies, G and Island, NY USA), in acco dance wi h he manu ac u e 's ins uc ions. B ie ly, 5 × 105 cells we e lysed in 1000 µl T izol-LS® eagen , and eco e y o o al RNA a e isop opanol p ecipi a ion was measu ed using a spec o- pho ome e (Beckman DU62, Ame sham Biosciences, Chal- on S . Giles, UK) a 260 nm. Analysis o ni ic oxide syn hase ype II gene exp ession using eal- ime RT-PCR Real- ime RT-PCR analyses we e pe o med in a luo escen empe a u e cycle (MX3000P Real Time PCR Sys em; S a - agene), in acco dance wi h he manu ac u e 's ins uc ions. To al RNA 1 µg was used o each RT eac ion. cDNAs we e syn hesized using 200 uni s o Moloney mu ine leukaemia e e se ansc ip ase (Gibco-BRL) and 6 µl dNTPs mix (10 mmol/l o each dNTP), 6 µl o i s s and bu e (250 mmol/l T is-HCl [pH 8.3], 375 mmol/l KCl, 15 mmol/l MgCl2; Gibco- BRL), 1.5 µl o 50 mmol/l MgCl2, 0.17 µl andom hexame solu ion (3 µg/µl; Gibco-BRL) and 0.25 µl o RNAse Ou TM ( ecombinan ibonuclease inhibi o 40 µg/µl; Gibco-BRL), in a o al olume o 30 µl. Reac ion mix u es we e incuba ed a 37°C o 50 min and a 42°C o 15 min. The RT eac ion was e mina ed by hea ing a 95°C o 5 min and subsequen ly quick chilled on ice. The 50 µl ampli ica ion mix u e (B illian SYBR G een QPC Mas e Mix; S a agene) con ained 2 µl o RT eac ion p oduc s plus 0.75 µl (30 nmol/l) dilu ed e e - ence dye, 150 nmol/l o each p ime and nuclease- ee, PCR g ade wa e o adjus he inal olume o 50 µl. A e a i s enzyme ac i a ion s ep (95°C o 10 min), eac- ions we e cycled 33 imes using he ollowing pa ame e s o NOS ype II de ec ion: dena u a ion a 95°C o 40 s, anneal- ing a 60°C o 1 min and ex ension a 72°C o 1 min. Mouse glyce aldehyde-3-phospha e dehyd ogenase (GAPDH) cDNA (5'-TCCATGACAACTTTGGCATCGTGG-3' o ups eam p ime and 5'-GTTGCTGTTGAAGTCACAGGAGAC-3' o downs eam p ime ; Genebank M32599) was ampli ied unde he same condi ions and was used as a no malize gene. The amoun o PCR p oduc s o med in each cycle was e alua ed A h i is Resea ch & The apy Vol 7 No 3 O e o e al. R584 on he basis o SYBR G een I luo escence. A inal ex ension a 72°C o e 10 min was ollowed by mel ing cu e p o iles as ollows: 95°C o 1 min, amping down o 45°C a a a e o 0.2°C/s, and hea ing slowly (0.5°C/cycle) o 95°C o a o al o 81 cycles (30 s/cycle). Fluo escence was measu ed con in- uously o con i m ampli ica ion o speci ic ansc ip s (da a no shown). The oligonucleo ide p ime s speci ic o mouse NOS ype II we e as ollows: ups eam p ime 5'-CTCACTGGGACAG- CACAGAA-3' and downs eam p ime 5'-TGGT- CAAACTCTTGGGGTTC-3' ( om Genbank U43428). Cycle- o-cycle luo escence emission eadings we e moni- o ed and quan i ied using he second de i a i e maximum me hod om he MX3000P Real Time so wa e package (S a agene). This me hod de e mines he c ossing poin s o indi idual samples using an algo i hm ha iden i ies he i s u ning poin o he luo escence cu e. This u ning poin co - esponds o he i s maximum o he second de i a i e cu e and co ela es in e sely wi h he log o he ini ial empla e con- cen a ion. NOS ype II mRNA le els we e no malized wi h espec o mouse GAPDH le el in each sample. Ni ic oxide syn hase ype II wes e n blo analysis ATDC-5 chond ogenic cells we e seeded in P100 pla es un il hey eached 85–90% con luence. A e o e nigh s a a ion in se um- ee medium, cells we e s imula ed o 24 hou s wi h lep in (800 nmol/l), alone o in combina ion wi h IL-1 (0.025 ng/ml). In o de o demons a e he in ol emen o JAK2, PI3K, MEK-1 and p38 kinase, he ollowing speci ic pha macological inhibi o s we e added 1 hou be o e cy okine s imula ion: y - phos in AG490 (5 and 10 µmol/l) and Tkip (20 and 50 µmol/ l) o JAK2; LY294002 (1, 5 and 10 µmol/l) o PI3K; PD098059 (1, 10 and 30 µmol/l) o MEK-1; and SB203580 (1, 10 and 30 µmol/l) o p38 kinase. A e s imula ion, cells we e apidly washed wi h ice cold PBS and sc aped in lysis bu e : 10 mmol/l T is-HCl (pH 7.5), 5 mmol/l EDTA, 150 mmol/l NaCl, 30 mmol/l sodium py ophospha e, 50 mmol/l sodium luo ide, 1 mmol/l sodium o ho anada e (Na3VO4), 10% glyce ol, 0.5% T i on X-100, 1 mmol/l phenylme hylsul- onil luo ide, ap o inin, leupep in and peps a in A (10 mg/ml). Lysed cells we e cen i uged a 13000 g o 15 min. Lysa es om con ol o s imula ed cells we e collec ed and sepa a ed by SDS-PAGE on a 10% polyac ylamide gel. P o eins we e subsequen ly ans e ed o a poly inylidene di luo ide ans e memb ane (Hybond TM-P; Ame sham In e na ional, Li le Chal on , UK) using a ans e semid y blo cell (BioRad Labo- a o ies, He cules, CA, USA). Blo s we e incuba ed wi h he app op ia e an ibody (mouse an i-NOS II an ibody; pu chased om Ups a e Bio ech, Lake Placid, NY, USA). Immunoblo s we e isualized using ECLPlus de ec ion Ki (Ame sham-Pha - macia Bio ech, Ba celona, Spain) using ho se adish pe oxi- dase labelled seconda y an ibody. To con i m equal load in each sample, a e s ipping in glycine bu e a pH 3, mem- b anes we e eblo ed wi h an i-ac in an ibody (San a C uz Bio echnology Inc., San a C uz, CA, USA). The images o au o adiog ams we e cap u ed and analyzed using a Typhoon 9410 digi al a iable mode image (Ame sham Bio ech, Li le Chal on , UK). Da a analysis Da a a e exp essed as mean ± s anda d e o o he mean o a leas h ee independen expe imen s, each wi h a leas h ee o mo e independen obse a ions. S a is ical analysis was pe o med using analysis o a iance ollowed by he S u- den –Newman–Keuls o Bon e oni mul iple compa ison es wi h he Ins a compu e ized package (G aphPad So wa e Inc., San Diego, CA, USA). i < 0.05 was conside ed s a is i- cally signi ican . Resul s Lep in syne gis ic e ec o e IL-1 induced ni i e p oduc ion in chond ocy es A lep in concen a ion o 800 nmol/l was ound o be op imal o co-s imula o y expe imen s. This concen a ion was selec ed based on a b aod se o p e ious dose– esponse expe imen s (da a no shown). Because NOS ype II s imula- ion wi h IL-1 a 0.05 ng/ml was maximal, a dose o 0.025 ng/ ml was selec ed in o de o a oid masking lep in syne gism. As shown in Fig. 1, ATDC5 cells and human p ima y chond o- cy es did no accumula e ni i es when s imula ed wi h lep in alone; howe e , lep in was able o inc ease signi ican ly ni i e accumula ion induced by IL-1 when cells we e co-s imula ed wi h bo h cy okines (Fig 1a,c). This esul was con i med in e ms o p o ein exp ession. Indeed, a clea -cu inc ease in le - els o NOS ype II p o ein was obse ed when cells we e co- s imula ed wi h lep in and IL-1 (Fig. 1b). To con i m whe he NO o ma ion was p oduced ia NOS ype II, ATDC5 cells and human chond ocy es we e incuba ed o 48 hou s wi h bo h cy okines in he p esence o he NOS ype II inhibi o aminoguanidine (1 mmol/l), added 1 hou be o e cy okine adminis a ion. Aminoguanidine comple ely inhibi ed ni i e accumula ion in he cul u e supe na an o human p i- ma y chond ocy es (Fig. 1c) and ATDC5 cells (Fig. 1d). Janus kinase-2 inhibi ion blocks lep in/IL-1 induced ni ic oxide p oduc ion and ni ic oxide syn hase ype II p o ein exp ession We also in es iga ed he ole played by JAK2 in ni i e p oduc- ion e oked by co-s imula ion wi h lep in and IL-1 by using y - phos in AG490. This JAK2 inhibi o , added 1 hou be o e cy okine co-s imula ion, comple ely blocked ni i e p oduc ion (Fig. 2a). This esul was con i med in e ms o p o ein exp es- sion, because cell p e ea men wi h y phos in AG490 signi - ican ly dec eased NOS II p o ein exp ession in lep in/IL-1 co- s imula ed cells (Fig. 2d). In iguingly, y phos in AG490 was also able o inhibi ni i e accumula ion induced by IL-1 alone, sugges ing ha lep in syne gizes wi h undamen al pa hways A ailable online h p://a h i is- esea ch.com/con en /7/3/R581 R585 in IL-1 esponses. To gain u he insigh s in o he in ol emen o JAK2, Tkip (a 12-me SOCS-1 mime ic pep ide ha binds o he au ophospho yla ion si e o JAK2) was added o ATDC5 cells 1 hou be o e hey we e s imula ed wi h lep in o IL-1, o bo h cy okines. Tkip a 50 µmol/l was able o blun comple ely lep in/IL-1 induced ni i e accumula ion and NOS II p o ein exp ession (Fig. 2b,e). A lipophilic i ele an pep ide, MuIFN- γ95–125 (AKFEVNNPQVQRQAFNELIRVVHQLLPESSL), was used as con ol. In iguingly, Tkip was also able o inhibi , in a dose– esponse manne , ni i e accumula ion and NOS II p o- ein exp ession in ATDC5 cells s imula ed wi h IL-1 alone (Fig. 2c,e). E ec o he speci ic signalling pa hways inhibi o s LY294002, PD098059 and SB203580 on lep in/IL-1 co- s imula ion In o de o de ine he signalling pa hway in ol ed in he syne - gis ic induc ion o NOS ype II media ed by co-s imula ion wi h lep in and IL-1 in cul u ed ATDC5 cells, we e alua ed he e ec s o speci ic pha macological inhibi o s on o he kinases, speci ically PI3K, MEK-1 and p38 kinase. We i s in es iga ed he e ec o a speci ic inhibi o o PI3K, namely LY294002 (1, 2.5, 5 and 10 µmol/l) on lep in/IL-1 induced NO p oduc ion. The addi ion o LY294002 1 hou be o e cy okine co-s imula ion esul ed in signi ican and dose- dependen dec eases in NO p oduc ion and NOS ype II p o- ein exp ession (Fig. 3a,a1). In o de o es whe he MEK-1 ( he mi ogen-ac i a ed p o ein kinase [MAPK] kinase in ol ed in ex acellula signal- egula ed kinase [ERK]-1 and ERK-2 phospho yla ion/ac i a ion) pa ici- pa es in NOS ype II induc ion ia lep in/IL-1 co-s imula ion, we used he speci ic MEK-1 inhibi o PD98059. When his inhibi o was added 1 hou be o e cy okine co-s imula ion, sig- ni ican dose-dependen dec eases in NO p oduc ion and NOS II p o ein exp ession we e obse ed (Fig. 3b,b1). Figu e 1 Lep in syne gizes wi h IL-1 in inducing ni ic oxide syn hase (NOS) ype IILep in syne gizes wi h IL-1 in inducing ni ic oxide syn hase (NOS) ype II. Syne gis ic e ec o lep in (OB) on ni i e (NO2-) accumula ion and NOS ype II p o ein exp ession induced by IL-1. S imula ions we e conduc ed in se um- ee condi ions (a,b) in ATDC5 chond ogenic cells and (c) in human p ima y chond ocy es. NO2- accumula ion is selec i ely inhibi ed by aminoguanidine (AG) bo h in (d) ATDC5 cells and in (panel c) human p i- ma y chond ocy es. Values a e exp essed as mean ± s anda d e o o he mean. WB, wes e n blo . A h i is Resea ch & The apy Vol 7 No 3 O e o e al. R586 Finally, because i has been shown ha p38 kinase is in ol ed in apop o ic p ocesses induced by NO in chond ocy es, we es ed whe he his MAPK is also in ol ed in NOS ype II syn- e gis ic ac i a ion s imula ed by lep in/IL-1. Fo his pu pose, we used he speci ic p38 kinase inhibi o SB203580. Addi ion o his inhibi o 1 hou be o e lep in/IL-1 co-s imula ion caused signi ican and dose-dependen dec eases in NO p oduc ion and NOS II p o ein exp ession (Fig. 3c,c1 [lowe panel]). Lep in syne gism does no depend on chond ocy e di e en ia ion s a e In o de o de e mine whe he lep in/IL-1 syne gism and i s sig- nalling pa hway depend on he di e en ia ion s a e o chond o- cy es, we conduc ed simila expe imen s in ma u e and hype ophic chond ocy es. We di e en ia ed ATDC5 cells (see Ma e ials and me hods, abo e) in o ma u e and hype - ophic chond ocy es, and es ed co-s imula ion and ea - men s wi h all speci ic inhibi o s. Ni i e accumula ion, e alua ed in 15-day (ma u e) and in 21-day (hype ophic) di - e en ia ed ATDC5 cells a 24 and 48 hou s a e ea men , was simila o ha obse ed in he ATDC5 chond ogenic undi e en ia ed cell line (Fig. 4a–d). No e ha in o de o e al- ua e he in ol emen o PI3K, in some expe imen s we also used wo mannin a 10 µmol/l (a classical bu no e y speci ic PI3K inhibi o ), yielding esul s simila o hose ob ained wi h LY294002. Finally, a simila pa e n was obse ed in human cul u ed p i- ma y chond ocy es. In hese cells, lep in induced a s ong inc ease in ni i e accumula ion o e ha induced by IL-1, and Figu e 2 Janus kinase (JAK)2 inhibi ion blocks lep in/IL-1-induced ni ic oxide (NO) p oduc ion and ni ic oxide syn hase (NOS) ype II p o ein exp essionJanus kinase (JAK)2 inhibi ion blocks lep in/IL-1-induced ni ic oxide (NO) p oduc ion and ni ic oxide syn hase (NOS) ype II p o ein exp ession. E ec o y phos in AG490 and Tkip on NO p oduc ion and NOS II p o ein exp ession. The e ec o y phos in AG490 was e alua ed in e ms o (a) ni i e accumula ion in ATDC5 cells s imula ed wi h lep in and IL-1, and in e ms o (d) NOS II p o ein exp ession. The e ec o Tkip was e alua ed by ni i e accumula ion in (b) lep in/IL-1 ATDC5 co-s imula ed cells and in (c) IL-1 s imula ed cells (panel c). (e) E ec o Tkip on NOS ype II p o ein exp ession in lep in/IL-1 co-s imula ed cells. A ailable online h p://a h i is- esea ch.com/con en /7/3/R581 R587 he syne gis ic esponse was signi ican ly inhibi ed by y phos in AG490, wo mannin, LY294002, PD98059 and SB203580 (Fig. 5). E ec o lep in/IL-1 co-s imula ion on ni ic oxide syn hase ype II RNA exp ession We inally s udied NOS II mRNA exp ession in o de o de e - mine whe he NO inc ease/inhibi ion was due o modula ion o NOS ype II mRNA exp ession. As shown in Fig. 6, NOS ype II mRNA, e alua ed using eal- ime PCR, was s ongly exp essed when cells we e co-s imula ed wi h lep in plus IL-1, and his exp ession was signi ican ly educed by y phos in AG490, wo mannin, LY294002, PD098059 and SB203580. Discussion In he p esen s udy we in es iga ed he e ec o lep in on NO p oduc ion s imula ed by IL-1. We ound ha lep in had a syn- e gis ic e ec in he ATDC5 mu ine chond ogenic cell line, in di e en ia ed ma u e and hype ophic ATDC5 chond ocy es, and in human p ima y chond ocy es. Lep in has been classi ied as a cy okine-like ho mone, because o i s s uc u e and he homology o i s ecep o s wi h membe s o he class I cy okine ecep o supe amily. A p oin- lamma o y ole o lep in has p e iously been p oposed. Se - e al da a show ha lep in le els a e inc eased by p oin lamma o y cy okine adminis a ion and in animal models o acu e in lamma ion [9]. In addi ion, lep in egula es no only humo al bu also cellula immune esponses in an igen- induced a h i is models [20]. Ne e heless, he e a e only ew Figu e 3 In ol emen o phospha idylinosi ol 3-kinase (PI3K), mi ogen-ac i a ed p o ein kinase kinase (MEK)-1 and p38-kinase in lep in/IL-1-induced ni ic oxide syn hase (NOS)In ol emen o phospha idylinosi ol 3-kinase (PI3K), mi ogen-ac i a ed p o ein kinase kinase (MEK)-1 and p38-kinase in lep in/IL-1-induced ni ic oxide syn hase (NOS). Dose-dependen e ec o (a,a1) LY294002, (b,b1) PD098059 and (c,c1) SB203580 on ni i e (NO2-) p oduc ion and NOS ype II p o ein exp ession in s imula ed and uns imula ed ATDC5 cells. S imula ions we e conduc ed in se um- ee condi ions. Each inhibi o was added 1 hou be o e cy okine co-s imula ion. Values a e exp essed as mean ± s anda d e o o he mean. OB, lep in; WB, wes e n blo . A h i is Resea ch & The apy Vol 7 No 3 O e o e al. R588 epo s o a di ec ac ion o lep in a he cellula le el in ca i- lage [14,15]. NO con ols a a ie y o ca ilage unc ions, including loss o chond ocy e pheno ype, chond ocy e apop osis, and ex acel- lula ma ix deg ada ion [2,3]. NOS ype II is mainly exp essed by immune cells in esponse o a wide ange o p oin lamma- o y cy okines [21,22]. In i o, human a icula ca ilage is able o p oduce la ge amoun s o NO [23], which can be enhanced by p oin lamma o y cy okines. In addi ion, NO p oduc ion can be signi ican ly inc eased by he p esence o lep in, as shown in ou p e ious wo k [15] and in he p esen s udy. He e, we show ha he IL-1 induced p oduc ion o NO by ATDC5 mu ine chond ocy es and by human chond ocy es is signi ican ly enhanced by lep in. I is no ewo hy ha , apa om blood, se e al sou ces o lep in and IL-1 ha e been iden- i ied in o a ound he join s in pa hological condi ions. IL-1 is p oduced by in lamed syno ium and pe ia icula a pad [24]. In e es ingly, mul ipo en s omal cells om he in apa ella a p oduce lep in [25]. In addi ion, os eoa h i ic human chond o- cy es p oduce lep in, and lep in adminis a ion in a s induces o e -exp ession o his ho mone by a icula chond ocy es [13]. Thus, in pa ien s wi h in lamma o y syno i is o os eoa - h i is, he e is a unique mic oen i onmen in he ca ilage cha - ac e ized by ele a ed le els o bo h lep in and IL-1, due no only o local p oduc ion bu also o sys emic inc ease [10,13,26]. I is concei able ha in his scena io lep in plays a signi ican p oin lamma o y ole, as sugges ed by he indings p esen ed he e. O u he in e es is ou p e ious epo [15] o he co-s imula o y e ec o lep in and IFN-γ a he chond o- cy e le el. Figu e 4 Lep in syne gism does no depend upon chond ocy e di e a ion s a eLep in syne gism does no depend upon chond ocy e di e a ion s a e. E ec o di e en inhibi o s on ni i e (NO2-) accumula ion in 15-day di e en- ia ed ATDC5 cells s imula ed o no wi h lep in, alone o in combina ion wi h IL-1, du ing (a) 24 and (b) 48 hou s. The e ec o inhibi o s was also e alua ed in 21-day di e en ia ed ATDC5 cells, a e (c) 24 o (d) 48 hou s o s imula ion wi h lep in and IL-1 (alone o in combina ion). Values a e exp essed as mean ± s anda d e o o he mean. OB, lep in. A ailable online h p://a h i is- esea ch.com/con en /7/3/R581 R589 We p e iously es ablished ha he ea ly e en in lep in/IFN-γ syne gis ic NOS ype II ac i a ion was he in ol emen o JAK2 [15]; he p esen esul s con i m ha JAK2 ac i a ion is also an ea ly s ep in lep in/IL-1 induced NOS ype II co-s imula ion. The ac ha y phos in AG490 blocks he lep in/IL-1 esponse implies ha lep in syne gizes wi h c i ical pa hways in IL-1 esponse. I was su p ising ha y phos in AG490 also blocked he esponse o IL-1 alone, because JAK2 is no known o be equi ed o IL-1 ecep o ansduc ion, and so one would expec he e ec o y phos in AG490 o be pa ial. Howe e , ou esul s a e in ag eemen wi h hose epo ed by o he in es iga o s [27,28]. We also used Tkip in ou expe imen s; Tkip is a 12-me SOCS-1 mime ic lipophilic pep ide (WLVFFVIFYFFR) ha inhibi s JAK2 au ophospho yla ion [29]. In e es ingly, he beha iou o his pep ide was simila o ha o y phos in AG490 in e ms o NOS II inhibi ion. I is concei able ha his pep ide, because o i s SOCS-1 mime ic p ope ies, could inhibi IL-1/Toll-like ecep o unc ion in chond ocy es. SOCS- 1 is a nega i e egula o o lipopolysaccha ide-induced mac o- phage ac i a ion [30,31] and has been shown o bind o IL-1 ecep o associa ed kinase [32]. This dis up s he cascade ha leads o nuclea ac o -κB (NF-κB) signalling and causes NOS inhibi ion. O no e, i has been demons a ed ha y - phos in AG490 inhibi s IL-1 induced NF-κB ac i a ion in con- cen a ions ha also inhibi NOS II mRNA and p o ein syn hesis. These indings sugges ha JAK2 is equi ed o NF-κB ac i a ion, which in u n media es IL-1 induced NOS II exp ession in chond ocy es [28]. To gain u he insigh s in o he mechanism by which lep in, oge he wi h IL-1, p omo es NO p oduc ion, we e alua ed he oles played by downs eam signalling cascades using spe- ci ic pha macological inhibi o s. Fi s , we analyzed he in ol e- men o PI3K. The ole played by his kinase in he ac i a ion o NOS ype II is qui e con o e sial and emains he subjec o deba e. A numbe o s udies suppo he iew ha PI3K ac i - i y down- egula es NOS ype II, bu he e a e se e al ca ea s Figu e 5 Lep in ac s syne gis ically wi h IL-1 in human p ima y chond ocy esLep in ac s syne gis ically wi h IL-1 in human p ima y chond ocy es. Ni i e (NO2-) accumula ion in lep in (OB)/IL-1 co-s imula ed human p i- ma y chond ocy es. S imula ions we e conduc ed in se um- ee condi- ions in he p esence o absence o y phos in AG490, wo mannin, LY294002, PD98059 and SB203580 inhibi o s. Values a e exp essed as mean ± s anda d e o o he mean. Figu e 6 E ec o lep in/IL-1 co-s imula ion on ni ic oxide syn hase (NOS) ype II mRNA exp essionE ec o lep in/IL-1 co-s imula ion on ni ic oxide syn hase (NOS) ype II mRNA exp ession. Real- ime RT-PCR analysis o he exp ession o he inducible NOS ype II mRNA in lep in (OB)/IL-1 co-s imula ed ATDC5 cells. S imula ions (24 hou s) we e conduc ed in se um- ee condi ions. Speci ic inhibi o s we e added 1 hou be o e cy okine co-s imula ion. Values a e exp essed as mean ± s anda d e o o he mean.