RESEARCH ARTICLE Open Access
Ca abolic and p oin lamma o y e ec s o
lep in in chond ocy es a e egula ed by
supp esso o cy okine signaling-3
Anna Koskinen-Kolasa
1
, Ka iina Vuol eenaho
1
, Riku Ko honen
1
, Teemu Moilanen
1,2
and Ee a Moilanen
1*
Abs ac
Backg ound: P e ious s udies p o ide e idence ha adipokine lep in inc eases p oduc ion o ca abolic and
p oin lamma o y ac o s in chond ocy es and se es as a link be ween obesi y and os eoa h i is (OA). Howe e ,
he magni ude o he esponse o lep in ea men a ies g ea ly be ween chond ocy es om di e en dono
pa ien s. In he p esen s udy, we in es iga ed he egula o y ole o supp esso o cy okine signaling-3 (SOCS-3)
in he lep in-induced esponses in OA ca ilage.
Me hods: Ca ilage and syno ial luid samples om 97 pa ien s wi h OA unde going knee eplacemen su ge y
we e collec ed. Ca ilage samples we e cul u ed wi h lep in (10 μg/ml), and he le els o p oin lamma o y and
ca abolic ac o s in syno ial luid and in he ca ilage cul u e media, and SOCS-3 exp ession in he ca ilage we e
measu ed. The ole o SOCS-3 in lep in signaling was u he s udied in H4 mu ine chond ocy es by down egula ing
SOCS-3 wi h siRNA.
Resul s: Lep in-induced exp ession o ma ix me allop o einases MMP-1, MMP-3, MMP-13, in e leukin-6 (IL-6), inducible
ni ic oxide syn hase (iNOS) and cyclooxygenase-2 (COX-2) we e highe in he ca ilage samples wi h low SOCS-3
exp ession. Acco dingly, down egula ion o SOCS-3 by siRNA in H4 chond ocy es led o enhanced lep in-induced
exp ession o MMP-3, MMP-13, IL-6 and iNOS. Syno ial luid lep in was associa ed posi i ely, and ca ilage SOCS-3
nega i ely wi h syno ial luid le els o MMPs in a mul i a ia e model in obese (body mass index (BMI) >30 kg/m
2
)bu
no in non-obese (BMI <30 kg/m
2
)pa ien s.
Conclusions: Ou esul s show, o he i s ime, ha SOCS-3 egula es lep in-induced esponses in ca ilage, and could
hus be a u u e d ug a ge in he ea men o p e en ion o OA, especially in obese pa ien s.
Keywo ds: Lep in, Adipokine, SOCS-3, Os eoa h i is, Chond ocy es, Obesi y
Backg ound
Adipokines a e cy okine-like ho mones p oduced by
adipose issue and o iginally disco e ed o egula e en-
e gy me abolism [1, 2]. Thei ole in in lamma ion and
obesi y- ela ed disease, such as ype 2 diabe es melli us
and ca dio ascula disease, and also in heuma ic dis-
ease has a ac ed inc easing in e es du ing he pas
decade. Lep in was i s cha ac e ized in 1994 [3] and
o da e i is p obably he mos s udied adipokine. The
ci cula ing le els o lep in a e closely associa ed wi h
he amoun o s o ed body a and wi h body mass
index (BMI) [4]. Lep in is, howe e , no only p oduced
by adipose issue, bu also by se e al o he issues, in-
cluding ca ilage and o he join issues [5–7]. In e es -
ingly, syno ial luid lep in le els a e also co ela ed
wi h BMI and lep in exp ession in chond ocy es is
inc eased in obese indi iduals wi h OA [5, 6, 8]. The
exp ession o lep in and i s unc ional ecep o Ob-Rb
is also epo ed o be inc eased in chond ocy es in OA,
in compa ison o heal hy chond ocy es [6].
Obesi y is a majo isk ac o o OA [9]. T adi ionally
obesi y has been hough o explain he isk o de eloping
OA due o inc eased wea -and- ea on weigh -bea ing
join s. Howe e , obesi y is also a isk ac o o hand OA
* Co espondence: [email p o ec ed]
1
The Immunopha macology Resea ch G oup, Uni e si y o Tampe e School
o Medicine and Tampe e Uni e si y Hospi al, Tampe e, Finland
Full lis o au ho in o ma ion is a ailable a he end o he a icle
© 2016 The Au ho (s). Open Access This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 4.0
In e na ional License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and
ep oduc ion in any medium, p o ided you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o
he C ea i e Commons license, and indica e i changes we e made. The C ea i e Commons Public Domain Dedica ion wai e
(h p://c ea i ecommons.o g/publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed.
Koskinen-Kolasa e al. A h i is Resea ch & The apy (2016) 18:215
DOI 10.1186/s13075-016-1112-0
[10], which poin s o a sys emic ac o o ac o s ha
media e he obesi y- ela ed impac on ca ilage. Lep in,
wi h i s s ong posi i e associa ion wi h body a s o es,
i s well in his pic u e; in ac , inc easing e idence
suppo s he ole o lep in as a signi ican ac o in he
pa hogenesis o OA. Lep in has been shown o ha e di ec
p oin lamma o y and ca abolic e ec s on ca ilage in ex-
pe imen al se ings. We and o he s ha e p e iously shown
ha lep in enhances p oduc ion o ca abolic enzymes,
including ma ix me allop o einase 1 (MMP-1), MMP-2,
MMP-3, MMP-9, MMP-13, a disin eg in and me allop o-
einase wi h h ombospondin mo i s 4 (ADAMTS-4) and
ADAMTS-5 and p oin lamma o y media o s, such as
ni ic oxide (NO), in e leukin 6 (IL-6), IL-1β,IL-8and
p os aglandin E
2
(PGE
2
) in chond ocy es, syno iocy es
and in ca ilage [6, 11–19]. These indings sugges ha
lep in is no only a bys ande o ca ilage b eakdown, bu
an ac i e de imen al ac o in he pa hogenesis o OA.
Acco ding o ou expe ience, ca ilage om di e en
dono pa ien s espond o lep in ea men in a qui e e -
sa ile manne : some o he samples p oduce la ge amoun s
o ca abolic/p oin lamma o y media o s like MMPs, IL-6
and NO ollowing lep in ea men , while in some samples
lep in-induced changes in he p oduc ion o hese ac o s
a e e y small. Simila wide a ia ion in he esponse o
lep in is also suppo ed by o he s udies [17]. A s udy by
Pallu e al. showed ha p ima y chond ocy es ecei ed
om obese pa ien s wi h OA espond o smalle amoun s
o lep in o enhance MMP-13 p oduc ion han chon-
d ocy es ob ained om non-obese pa ien s [17], sugges ing
ha obese indi iduals migh be mo e suscep ible o he
ha m ul e ec s o lep in on ca ilage. Howe e , he mecha-
nisms egula ing lep in esponsi eness in chond ocy es
emain unknown.
Supp esso o cy okine signaling 3 (SOCS-3) belongs
o SOCS p o eins, which a e in acellula molecules ha
ha e an impo an unc ion o limi ing excessi e in lam-
ma o y ac i a ion o he inna e and adap i e immune
sys em [20]. In in lamma o y cells SOCS-3 exp ession is
induced by ype I and ype II cy okine ecep o s ia he
JAK-STAT pa hway. SOCS-3 binds o he gp130 subuni
o hose ecep o s and inhibi s he JAK-STAT pa hway,
hus o ming a nega i e eedback loop o limi cy okine
ac ions [21]. In e es ingly, SOCS-3 is also in ol ed in
egula ing lep in esponsi eness in he cen al ne ous
sys em (CNS) [22].
The me abolic unc ion o lep in is o se e as a senso o
body a s o es o he CNS. Ele a ion o blood lep in due
o calo ie in ake, whe he sho - e m o long- e m, in a lean
pe son no mally supp esses ood in ake, whe eas dec eased
lep in le els due o as ing o loss o adipose issue lead o
inc eased ood in ake [23]. In obesi y howe e , ele a ed lep-
in does no lead o he expec ed esponses in weigh con-
ol. This is hough o be due o dis u bed lep in signaling,
also called lep in esis ance. Ele a ed SOCS-3 exp ession in
he CNS is p oposed o be he p ima y mechanism ha
causes lep in esis ance and subsequen ailu e in con-
olling ood in ake in obesi y [22]. Consis en ly, lep in-
de icien mice de elop se e e obesi y [24], whe eas
SOCS-3 condi ional knockou mice a e esis an o
die -induced obesi y [25]. In humans, gene ic lep in de i-
ciency also causes se e e obesi y, hough lep in and lep in-
ecep o - ela ed mu a ions a e ex emely a e [26].
SOCS-3 is also exp essed in ca ilage [27–29], and we
epo ed p e iously ha i s exp ession is lowe in ca i-
lage om obese pa ien s wi h OA han om non-obese
pa ien s [8]. Tha led us o hypo hesize ha SOCS-3
could be a signi ican mechanism behind he a iable
lep in esponsi eness in ca ilage samples om di e en
dono pa ien s. We add essed he hypo hesis by in es i-
ga ing SOCS-3 exp ession and lep in esponsi eness in
ca ilage samples ob ained om 97 pa ien s wi h OA. In
addi ion, he ole o SOCS-3 exp ession in lep in signa-
ling was s udied by down egula ing SOCS-3 by siRNA
in chond ocy e cul u es.
Me hods
Ca ilage and cell cul u es
Ca ilage and syno ial luid (SF) samples we e collec ed
om 97 pa ien s wi h OA who we e unde going knee
eplacemen su ge y. All pa ien s ul illed he Ame ican
College o Rheuma ology classi ica ion c i e ia o OA
[30]. Ca ilage samples we e p ocessed o issue cul u e
as p e iously desc ibed [15]. Ca ilage pieces we e incu-
ba ed o 42 hou s wi h o wi hou lep in (10 μg/ml).
The concen a ion o lep in used was chosen based on ou
p e ious s udies and on exis ing li e a u e [15, 17–19].
Recombinan human lep in was pu chased om R&D
Sys ems Eu ope L d, Abindgon, UK. Syno ial luid (SF)
samples om he co esponding pa ien s we e also col-
lec ed a he beginning o he a h oplas y. The SF sam-
ples we e cen i uged a 4000 g a 4 °C and supe na an s
we e collec ed and kep a −70 °C un il assayed.
The immo alized mu ine H4 chond ocy e cell line
[31], de eloped in he Labo a o y o Expe imen al
Rheuma ology, Uni e si y Medical Cen e , Nijmegen, The
Ne he lands, was used in he siRNA expe imen s. The
chond ocy es we e cul u ed a 37 °C in humidi ied 5 % ca -
bon dioxide a mosphe e in Dulbecco’s modi ied Eagle’s
medium (DMEM) wi h L-glu amine and Ham’sF-12
medium (1:1) supplemen ed wi h 5 % e al bo ine se um
(all ob ained om Lonza G oup L d, Basel, Swi ze land).
Immunoassays and ni i e measu emen s
Concen a ions o MMP-1, MMP-3, MMP-13 and IL-6
we e de e mined by immunoassays wi h comme cial e-
agen s acco ding o he p o ocol p o ided by he manu-
ac u e (human o al MMP-1, human o al MMP-3,
Koskinen-Kolasa e al. A h i is Resea ch & The apy (2016) 18:215 Page 2 o 13
human o al MMP-13, mouse o al MMP-3 and mouse IL-
6 ELISA ki s we e om R&D Sys ems; human IL-6 ELISA
ki was om Sanquin, Ams e dam, The Ne he lands;
MMP-1 in SF was de e mined by Mul iplex bead a ay,
Fluo okine® Human MMP Mul i Analy e P o iling Base
Ki , pu chased om R&D sys ems). Ni i e, s able me abo-
li e o ni ic oxide (NO), was measu ed in he cul u e
media by he G iess eac ion [32]. The ca ilage cul u e
media samples we e il e ed h ough Amicon Ul a 10-K
il e s ( om Millipo e, Co k, I eland) a 14,000 g p io o
he G iess analysis in o de o emo e la ge p o eins ha
migh in e e e wi h he G iess analysis.
RNA isola ion and quan i a i e e e se ansc ip ion/
polyme ase chain eac ion
Cul u e medium was emo ed a he indica ed ime
poin s and o al RNA o H4 chond ocy es was ex ac ed
wi h GenElu e™Mammalian To al RNA Minip ep ki
(Sigma-Ald ich, S Louis, MO, USA). To al RNA was
ea ed wi h DNAse (Fe men as UAB, Vilnius, Li huania)
and e e se- ansc ibed o cDNA using TaqMan Re e se
T ansc ip ion eagen s and andom hexame s (Applied
Biosys ems, Fos e Ci y, CA, USA). cDNA ob ained om
he RT eac ion was dilu ed 1:20 wi h RNAse- ee wa e
and subjec ed o quan i a i e PCR using TaqMan Uni e sal
PCR Mas e Mix and he ABI P ism 7000 Sequence de-
ec ion sys em (Applied Biosys ems). P ime s and p obes
o SOCS-3, glyce aldehyde-3-phospha e dehyd ogenase
(GAPDH), iNOS, IL-6 and MMP-13 we e ob ained om
Me abion In e na ional AG (Ma ins ied, Ge many). The
p ime and p obe sequences and concen a ions (Table 1)
we e op imized acco ding o he manu ac u e ’s ins uc-
ions in TaqMan Uni e sal PCR Mas e Mix P o ocol pa
numbe 4304449 e ision C. The exp ession o mouse
MMP-3 mRNA was measu ed using TagMan Gene Ex-
p ession Assay (Mm00440295_m1, Applied Biosys ems).
PCR eac ion pa ame e s we e as ollows: incuba ion
a 50 °C o 2 minu es, incuba ion a 95 °C o 10 mi-
nu es, and he ea e 40 cycles o dena u a ion a 95 °C
o 15 s and annealing and ex ension a 60 °C o 1 mi-
nu e. Each expe imen al eac ion was pe o med in du-
plica e. The ela i e mRNA le els o SOCS-3, GAPDH,
iNOS, IL-6 and MMP-13 we e quan i ied using he
s anda d cu e me hod as desc ibed in Applied Biosys-
ems Use Bulle in numbe 2. To calcula e he ela i e
exp ession o MMP-3 mRNA, he 2
(−ΔΔCT)
me hod [33]
was used. Acco ding o he me hod, he cycle h eshold
(C
T
) alues o MMP-3 mRNA exp ession in each sam-
ple we e no malized o he C
T
alues o GAPDH mRNA
in he same sample.
Wes e n blo
P epa a ion o cell lysa es, SDS-polyac ylamide gel elec-
opho esis and wes e n blo analysis we e ca ied ou as
p e iously desc ibed [15]. Mouse monoclonal SOCS-3
an ibody (sc-51699), abbi polyclonal iNOS an ibodies
(sc-651 and sc-650), goa polyclonal cyclooxygenase-2
(COX-2) an ibody (sc-1745) and abbi polyclonal β-ac in
an ibody (sc-1615R), and seconda y ho se adish pe oxidase
(HRP)-conjuga ed goa an i-mouse (sc-2005), goa an i-
abbi (sc-2004) and donkey an i-goa (sc-2020) an ibodies
we e all om San a C uz Bio echnology (San a C uz, CA,
USA). Rabbi polyclonal MMP-13 an ibody (ab39012) was
om Abcam (Camb idge, MA, USA). Lep in-induced
iNOS and COX-2 exp ession was de e mined by unning
he con ol and lep in-induced samples side by side and
he esul is gi en as old o change in he β-ac in-
no malized densi ome y alue o he lep in-induced e sus
he con ol sample.
Down egula ion o SOCS-3 exp ession by siRNA
H4 mu ine chond ocy es we e seeded a 1 × 10
5
cells/well
in 24-well pla es. Cells we e incuba ed o 24 hou s and
ans ec ed wi h SOCS-3 siRNA o wi h non- a ge ing
con ol siRNA. On-Ta ge SMART pool SOCS-3-speci ic
siRNA ( a ge ing sequences o GGCUAGGAGACUCGC
CUUA, GGACCAAGAACCUACGCAU, CUAAUGAAA
CCUCGCAGAU and GAAGGGAGGCAGAUCAACA)
and siGENOME Non-Ta ge ing siRNA we e used a 100
nM o ans ec he cells using Dha maFECT 1. All ans-
ec ion eagen s we e om The mo Scien i ic Dha macon
(La aye e, CO, USA) and ans ec ion was ca ied ou
acco ding o he manu ac u e ’s p o ocol. The expe i-
men s we e s a ed 48 hou s a e he ans ec ion by add-
ing lep in (10 μg/ml) (mouse ecombinan lep in om
R&D sys ems) in esh cul u e medium.
S a is ical analysis
The chi-squa e es , unpai ed es and Mann–Whi ney
es (whe e app op ia e) we e used o analyze di e ences
be ween subg oups o he pa ien s. The Wilcoxon es
was used o calcula e he signi icance o lep in-induced
e ec s in he ca ilage cul u e.
To analyze he di e ences in lep in esponsi eness in
ela ion o SOCS-3 exp ession, he samples on each
wes e n blo gel we e di ided o wo equal sized g oups
(low SOCS-3 o high SOCS-3) acco ding o SOCS-3 ex-
p ession. Median lep in esponses, measu ed as change
in he p oduc ion o MMP-1, MMP-3, MMP-13, IL-6
and NO in he lep in- ea ed e sus con ol sample,
and as old o change in he exp ession o iNOS and
COX-2, we e compa ed be ween he low SOCS-3 and
he high SOCS-3 g oups. Possible in e gel di e ences
in SOCS-3 exp ession we e con olled by analysis o
a iance (ANOVA) in which he lep in esponse a i-
able (e.g., lep in-induced change in p oduc ion o
MMP-1) was se as a dependen a iable, wes e n blo
gel (1 o 8) as a g ouping a iable and SOCS-3
Koskinen-Kolasa e al. A h i is Resea ch & The apy (2016) 18:215 Page 3 o 13
exp ession as a con inuous a iable as a co a ia e. Associ-
a ions we e u he es ed by adjus ing o BMI and age.
Co ela ion be ween he ac o s o in e es in SF we e de-
e mined by Pea son’s co ela ion analysis. The associa ions
be ween MMPs o IL-6 and lep in in SF, and SOCS-3 ex-
p ession in ca ilage we e u he analyzed by ANOVA
modeling, by including he a iable o in e es (SF MMP-1,
MMP-3 o IL-6) as a dependen a iable, lep in in SF and
SOCS-3 exp ession in he ca ilage as co a ia es and gel
numbe as a g ouping ac o . The analysis was done in
BMI subg oups (obese, BMI >30 kg/m
2
;non-obese,BMI
<30 kg/m
2
). Na u al loga i hms we e o med o he lep in
esponse alues, SOCS-3 exp ession le els and SF le els o
he measu ed a iables whe e app op ia e in o de o ha e
no mally dis ibu ed a iables o he ANOVA modeling
and o he co ela ion analyses.
Theda awe eanalyzedbyIBMSPSSS a is ics19(IBM
Co po a ion, NY, USA) and G aph-Pad InS a e sion 3.00
so wa e (G aphPad So wa e Inc., San Diego, CA, USA).
The esul s o he siRNA expe imen s a e p esen ed as
means (SEM). The s a is ical signi icance o hese da a was
calcula ed by wo-way ANOVA wi h Bon e oni mul iple
compa isons pos - es using G aph-Pad P ism 5 o Win-
dows e sion 5.04 (G aphPad So wa e Inc.). Di e ences
we e conside ed s a is ically signi ican a p< 0.05.
Resul s
Lep in-induced p oduc ion o p oin lamma o y and
ca abolic ac o s in os eoa h i ic ca ilage in ela ion o
clinical ac o s and SOCS-3 exp ession
Pa ien cha ac e is ics and lep in esponses in he cul u ed
ca ilage ac oss he whole s udy popula ion and in he
obese (BMI >30 kg/m
2
) and non-obese subg oups a e
p esen ed in Table 2. Lep in signi ican ly enhanced he
exp ession o MMP-1, MMP-3, MMP-13, IL-6, iNOS and
COX-2 and NO p oduc ion in OA ca ilage ex i o (Fig. 1).
Howe e , he e was conside able a ia ion in hese e-
sponses be ween he samples om di e en dono pa ien s
(Table 2). The e we e no s a is ically signi ican di e ences
in he lep in esponses be ween obese and non-obese
pa ien s (Table 2), and nei he did he lep in esponses
co ela e wi h age, sex o adiog aphic scaling o OA.
When he pa ien s we e di ided in o subg oups acco -
ding o SOCS-3 exp ession in he ca ilage, lep in-induced
changes in he exp ession/p oduc ion o MMP-1, MMP-3,
MMP-13, IL-6, NO, iNOS and COX-2 in he ca ilage we e
signi ican ly g ea e in he samples wi h low SOCS-3 ex-
p ession han in he samples wi h high SOCS-3 exp ession
(Fig. 2). This sugges s ha he le el o SOCS-3 exp ession
de e mina es he magni ude o lep in-induced in lamma-
o y esponses. The esul s emained s a is ically signi ican
(p< 0.05) o he esponses in he exp ession o MMP-3,
MMP-13, IL-6, NO, iNOS and COX-2, and almos signi i-
can o esponse in he exp ession o MMP-1 (p= 0.10) in
he ANOVA modeling a e con olling o in e gel a i-
a ion, BMI and age.
Syno ial luid le els o MMPs and IL-6 in ela ion o SF
lep in and SOCS-3 exp ession in ca ilage om pa ien s
wi h OA
SF samples we e ob ained om 90 o he 97 pa ien s.
Obese pa ien s had signi ican ly highe SF lep in han
non-obese pa ien s, while SF MMP-1 and MMP-3 did
no signi ican ly di e be ween obese and non-obese
Table 1 P ime and p obe sequences o quan i a i e RT-PCR
Gene Oligunucleo ide Sequence Conc. (nM)
Fo wa d p ime GCATGGCCTTCCGTGTTC 300
Mouse GAPDH Re e se p ime GATGTCATCATACTTGGCAGGTTT 300
P obe TCGTGGATCTGACGTGCCGCC 150
Fo wa d p ime GCGGGCACCTTTCTTATCC 300
Mouse SOCS-3 Re e se p ime AAGCTGCCCCCCTCACA 300
P obe CTCGGACCAGCGCCACTTCTTCA 150
Fo wa d p ime CCTGGTACGGGCATTGCT 300
Mouse iNOS Re e se p ime GCTCATGCGGCCTCCTT 300
P obe CAGCAGCGGCTCCATGACTCCC 150
Fo wa d p ime TCGGAGGCTTAATTACACATGTTC 900
Mouse IL-6 Re e se p ime CAAGTGCATCATCGTTGTTCATAC 300
P obe CAGAATTGCCATTGCACAACTCTTTTCTCA 200
Fo wa d p ime TTGTGTTTGCAGAGCACTACTTGA 900
Mouse MMP-13 Re e se p ime AACTGTGGAGGTCACTGTAGACTTCTT 900
P obe CATCCTGCGACTCTTGCGGGAATC 250
SOCS-3 supp esso o cy okine signaling-3, iNOS inducible ni ic oxide syn hase, IL-6 in e leukin-6, MMP-13 ma ix me allop o einase-13, Conc. concen a ion
Koskinen-Kolasa e al. A h i is Resea ch & The apy (2016) 18:215 Page 4 o 13
pa ien s (Table 2). Lep in co ela ed posi i ely wi h MMP-
1 and wi h MMP-3 in SF om obese bu no om non-
obese pa ien s (Fig. 3). In ANOVA modeling, lep in con-
cen a ions in SF and SOCS-3 exp ession in ca ilage sig-
ni ican ly explained le els o SF MMP-1 and MMP-3 in
he obese bu no in he non-obese g oup (Table 3) poin -
ing o obesi y- ela ed associa ion o lep in and SOCS-3 in
OA pa hophysiology. In addi ion, SF IL-6 le els we e ex-
plained by SOCS-3 in he obese bu no in he non-obese
g oup, while lep in did no signi ican ly explain SF IL-6
le els in ei he o he BMI subg oups (Table 3).
SOCS-3 modula es lep in esponses in chond ocy es
In o de o in es iga e u he he ole o SOCS-3 in he
egula ion o lep in-induced esponses in chond ocy es,
we used siRNA o down egula e SOCS-3 in he H4 chon-
d ocy e cell line. H4 chond ocy es exp essed SOCS-3
mRNA a ela i ely high le els and i was educed by ap-
p oxima ely 80 % in he SOCS-3-siRNA- ea ed cells
when compa ed o he cells ans ec ed wi h con ol
siRNA. Lep in had a clea e ec on inducing MMP-3,
MMP-13, IL-6 and iNOS exp ession in he SOCS-3-
de icien cells, whe eas in he con ol siRNA- ea ed cells
lep in did no ha e any s a is ically signi ican e ec on
he p oduc ion o hese ac o s (Fig. 4), con i ming ha
SOCS-3 nega i ely egula es lep in-induced p oin lamma-
o y esponses in chond ocy es.
Discussion
Lep in has been shown o ha e de imen al e ec s on
ca ilage me abolism in se e al s udies [6, 11–19]. How-
e e , conside able a ia ion in lep in esponsi eness
be ween ca ilage/chond ocy es om di e en pa ien s
has been obse ed. Ou p esen esul s indica e ha a
signi ican mechanism behind he di e en ial lep in
esponsi eness could be SOCS-3.
SOCS-3 is a known nega i e egula o o in lamma o y
signals [20]. I s ole in con olling he e ec s o lep in in
chond ocy es has no been p e iously in es iga ed, bu i
has been epo ed o egula e he esponses o lep in in
he CNS [22]. In he p esen s udy we show, o he i s
ime, ha SOCS-3 egula es he p oin lamma o y and
ca abolic e ec s o lep in in chond ocy es. This was
demons a ed as g ea e lep in esponsi eness in ca il-
age explan s wi h low SOCS-3 exp ession in compa ison
o lowe lep in esponsi eness in he explan s wi h high
SOCS-3 exp ession. The causali y o his associa ion was
illus a ed by down egula ion o SOCS-3 by siRNA in
he chond ocy e cell line, which led o inc eased lep in-
induced exp ession o p oin lamma o y and ca abolic
genes. In addi ion, SF lep in le els we e shown o be
posi i ely associa ed, and ca ilage SOCS-3 exp ession
nega i ely associa ed wi h SF MMP le els in obese, bu
no in non-obese pa ien s wi h OA. This poin s o dys-
egula ion o he lep in-SOCS-3 axis, especially in obese
Table 2 Pa ien cha ac e is ics and lep in esponses in ca ilage cul u es in he whole s udy popula ion and compa ed ac oss body
mass index subg oups
All Non-obese, BMI <30 kg/m
2
Obese, BMI >30 kg/m
2
n=97 n=49 n=48 P
Gende ( emale/male)
a
60/37 26/23 34/14 0.072
Body mass index (kg/m
2
)
b
30.9 (6.1) 26.2 (2.4) 35.7 (4.6) <0.001
Age (yea s)
b
69.8 (10.0) 72.8 (9.7) 66.8 (9.4) 0.003
Syno ial luid lep in (ng/ml)
c, d
12.8 (17.8) 7.6 (11.0) 21.5 (26.7) <0.001
Syno ial luid IL-6 (pg/ml)
c, d
118.9 (196.0) 126.8 (204.3) 114.0 (280.2) 0.784
Syno ial luid MMP-1 (ng/ml)
c, d
14.4 (25.7) 10.4 (16.6) 18.1 (27.3) 0.325
Syno ial luid MMP-3 (ng/ml)
c, d
649.5 (929.6) 591.6 (571.0) 764.9 (1159.0) 0.106
Lep in esponse in ca ilage
MMP-1 (change pg/mg ca ilage)
c
145.8 (247.1) 123.2 (253.6) 150.0 (258.3) 0.773
MMP-3 (change ng/mg ca ilage)
c
5.2 (8.8) 6.0 (8.6) 4.9 (10.2) 0.920
MMP-13 (change pg/mg ca ilage)
c
5.8 (13.6) 6.3 (11.8) 5.4 (15.9) 0.983
IL-6 (change pg/mg ca ilage)
c
123.2 (310.2) 114.6 (295.1) 130.9 (312.0) 0.740
NO (change pmol/mg ca ilage)
c
44.5 (133.4) 31.0 (125.8) 52.2 (140.0) 0.359
iNOS ( old o inc ease)
c, e
11.7 (160.6) 5.4 (143.1) 15.2 (209.6) 0.501
COX-2 ( old o inc ease)
c, e
6.9 (18.4) 7.1 (15.0) 6.4 (22.6) 0.748
a
Values a e numbe s o emale/male subjec s; p alue was calcula ed o compa ison be ween non-obese and obese subjec s using he chi-squa e es .
b
Values a e
mean (SD); p alues we e calcula ed o compa ison be ween non-obese and obese subjec s using he unpai ed es .
c
Values a e median (IQR); p alues
we e calcula ed o compa ison be ween non-obese and obese subjec s using he Mann–Whi ney es .
d
Syno ial luid sample was ob ained om 90 pa ien s.
e
Numbe s o pa ien s (non-obese/obese) in he analysis we e 26/31 o inducible ni ic oxide syn hase (iNOS) and 25/29 o cyclooxygenase-2 (COX-2). MMP ma ix
me allop o einase, IL in e leukin, NO ni ic oxide
Koskinen-Kolasa e al. A h i is Resea ch & The apy (2016) 18:215 Page 5 o 13

indi iduals, and o a possible obesi y- ela ed pa hogenic
mechanism in OA.
In he p esen s udy we obse ed a posi i e associa ion
be ween lep in le els and ma ix me allop o einases in
SF ha was only p esen in he obese pa ien s wi h OA.
Howe e , obesi y did no explain he di e en ial lep in
esponsi eness in he ca ilage cul u e expe imen s, un-
like in he s udy by Pallu e al. whe e g ea e lep in
Fig. 1 E ec o lep in on he p oduc ion o ma ix me allop o einase-1 (MMP-1)(a), MMP-3 (b), MMP-13 (c), in e leukin-6 (IL-6)(d), ni ic oxide (NO) (e)
and on he exp ession o inducible ni ic oxide syn hase (iNOS)( ) and cyclooxygenase-2 (COX-2)(g) in ca ilage om pa ien s wi h os eoa h i is (OA).
Ca ilage samples om 97 pa ien s wi h OA we e cul u ed wi h and wi hou lep in (10 μg/ml) o 42 hou s. Concen a ions o MMP-1, MMP-3, MMP-13
and IL-6 we e measu ed by ELISA; NO p oduc ion was de e mined as i s me aboli e ni i e by he G iess eac ion and iNOS and COX-2 p o eins by
wes e n blo ing. The ci cles ep esen he medians. The whiske s ep esen 95 % con idence in e al o he median. S a is ical signi icance was calcula ed
using he Wilcoxon es ; ***p< 0.001
Koskinen-Kolasa e al. A h i is Resea ch & The apy (2016) 18:215 Page 6 o 13
Fig. 2 Lep in-induced p oduc ion/exp ession o ma ix me allop o einase-1 (MMP-1)(a), MMP-3 (b), MMP-13 (c), in e leukin-6 (IL-6)(d), ni ic oxide
(NO)(e), inducible ni ic oxide syn hase (iNOS)( ) and cyclooxygenase-2 (COX-2)(g) in ca ilage om pa ien s wi h os eoa h i is (OA) in subg oups
s a i ied by supp esso o cy okine signaling-3 (SOCS-3) exp ession in he non- ea ed ca ilage. Human os eoa h i ic ca ilage was cul u ed wi h
lep in (10 μg/ml) o 42 hou s. Concen a ions o MMP-1, MMP-3, MMP-13 and IL-6 we e measu ed by ELISA, NO was de e mined as i s me aboli e
ni i e by he G iess eac ion and iNOS and COX-2 p o eins we e analyzed by wes e n blo ing. The ci cles ep esen he median change in he
lep in-induced e ec s. The whiske s ep esen he 95 % con idence in e al o he median. Numbe s o pa ien s om whom he ca ilage samples
we e collec ed a e indica ed. S a is ical signi icance was calcula ed using he Mann–Whi ney es ; *p< 0.05, **p< 0.01
Koskinen-Kolasa e al. A h i is Resea ch & The apy (2016) 18:215 Page 7 o 13
Fig. 3 Co ela ion be ween lep in and ma ix me allop o einase-1 (MMP-1) and MMP-3 in non-obese (a) and obese (b) pa ien s wi h os eoa h i is.
Lep in and MMPs we e measu ed in syno ial luid (SF) by immunoassay. Na u al loga i hms (LN) we e o med o he SF le els o lep in and MMPs
in o de o ha e no mally dis ibu ed a iables o he Pea son co ela ion analysis. Co ela ion coe icien s ( ) and p alues a e indica ed. Samples
we e collec ed om 90 pa ien s (non-obese, BMI <30 kg/m
2
,n= 44; obese, BMI >30 kg/m
2
,n= 46)
Table 3 Associa ions be ween in e leukin-6 (IL-6), ma ix me allop o einase-1 (MMP-1), MMP-3 and lep in in syno ial luid and
supp esso o cy okine signaling-3 (SOCS-3) exp ession in ca ilage om non-obese and obese pa ien s wi h os eoa h i is
Non-obese, BMI <30 kg/m
2
Obese, BMI >30 kg/m
2
Dependen a iable Co a ia es R
2
adjus ed PR
2
adjus ed P
LN (SF MMP-1) 0.15 0.30
LN SOCS-3 0.818 0.007
LN (SF lep in) 0.884 0.023
LN (SF MMP-3) 0.03 0.27
LN SOCS-3 0.608 0.004
LN (SF lep in) 0.733 0.015
LN (SF IL-6) −0.05 0.20
LN SOCS-3 0.945 0.003
LN (SF lep in) 0.808 0.466
P alues a e calcula ed o co a ia es in analysis o a iance modeling. The model is con olled o in e gel a ia ion in SOCS-3 exp ession le els. Analysis was
pe o med in body mass index (BMI) subg oups. Na u al loga i hms (LN) we e o med whe e app op ia e. SF syno ial luid
Koskinen-Kolasa e al. A h i is Resea ch & The apy (2016) 18:215 Page 8 o 13
Fig. 4 The e ec o silencing o supp esso o cy okine signaling-3 (SOCS-3) by siRNA on lep in-induced exp ession o ma ix me allop o einase-3
(MMP-3)(a,e), MMP-13 (b, ), in e leukin-6 (IL-6)(c,g) and inducible ni ic oxide syn hase (iNOS)(d,h) in H4 mu ine chond ocy es. The cells we e
ans ec ed wi h SOCS-3 siRNA o non- a ge ing siRNA (siNEG) and ea ed wi h lep in (10 μg/ml) o 4 (c,d), 8 (a,b,h)o 24(e-g) hou s. mRNA
exp ession (a-d) was de e mined by quan i a i e RT-PCR, he le els o MMP-3 (e) and IL-6 (g) in he cul u e media supe na an s by ELISA, and
MMP-13 ( ) and iNOS (h) exp ession in he chond ocy e lysa es by wes e n blo ing. Resul s a e exp essed as means ± SEM; n=6(a-eand g) and
n=3( ,h). MMP-3 p o ein le el in siNEG and in non- ea ed SOCS-3 siRNA samples was below he de ec ion limi and is se as hal o he lowes
s anda d. Rep esen a i e bands o he wes e n blo s a e shown. S a is ical analysis was ca ied ou by wo-way analysis o a iance wi h Bon e oni
mul iple compa isons pos hoc es ; *p< 0.05, **p< 0.01, ***p< 0.001. n.s. no signi ican
Koskinen-Kolasa e al. A h i is Resea ch & The apy (2016) 18:215 Page 9 o 13