Resveratrol Suppresses PAI-1 Gene Expression in a Human In Vitro Model of Inflamed Adipose ...

14 Pages·2015·2.15 MB·English

HindawiPublishingCorporation
OxidativeMedicineandCellularLongevity
Volume2013,ArticleID793525,13pages
http://dx.doi.org/10.1155/2013/793525
Research Article
Resveratrol Suppresses PAI-1 Gene Expression in
In Vitro
a Human Model of Inflamed Adipose Tissue
IvanaZagotta,1ElitsaY.Dimova,2Jan-BerndFuncke,1MartinWabitsch,1
ThomasKietzmann,2andPamelaFischer-Posovszky1
1DivisionofPediatricEndocrinologyandDiabetes,DepartmentofPediatricsandAdolescentMedicine,UlmUniversityMedicalCenter,
Eythstraße24,89075Ulm,Germany
2DepartmentofBiochemistryandBiocenterOulu,UniversityofOulu,P.O.B.3000,90014Oulu,Finland
CorrespondenceshouldbeaddressedtoThomasKietzmann;[email protected]
PamelaFischer-Posovszky;[email protected]
Received22March2013;Revised7May2013;Accepted8May2013
AcademicEditor:NilanjanaMaulik
Copyright©2013IvanaZagottaetal.ThisisanopenaccessarticledistributedundertheCreativeCommonsAttributionLicense,
whichpermitsunrestricteduse,distribution,andreproductioninanymedium,providedtheoriginalworkisproperlycited.
Increasedplasminogenactivatorinhibitor-1(PAI-1)levelsareassociatedwithanumberofpathophysiologicalcomplications;among
themisobesity.Resveratrolwasproposedtoimproveobesity-relatedhealthproblems,buttheeffectofresveratrolonPAI-1gene
expressioninobesityisnotcompletelyunderstood.Inthisstudy,weusedSGBSadipocytesandamodelofhumanadiposetissue
inflammationtoexaminetheeffectsofresveratrolontheproductionofPAI-1.TreatmentofSGBSadipocyteswithresveratrol
reducedPAI-1mRNAandproteininatime-andconcentration-dependentmanner.Furtherexperimentsshowedthatobesity-
associated inflammatory conditions lead to the upregulation of PAI-1 gene expression which was antagonized by resveratrol.
AlthoughsignalingviaPI3K,Sirt1,AMPK,ROS,andNrf2appearedtoplayasignificantroleinthemodulationofPAI-1gene
expressionundernoninflammatoryconditions,thosesignalingcomponentswerenotinvolvedinmediatingtheresveratroleffects
onPAI-1productionunderinflammatoryconditions.Instead,wedemonstratethattheresveratroleffectsonPAI-1inductionunder
inflammatoryconditionsweremediatedviainhibitionoftheNF𝜅Bpathway.Together,resveratrolcanactasNF𝜅Binhibitorin
adipocytesandthusthesubsequentlyreducedPAI-1expressionininflamedadiposetissuemightprovideanewinsighttowards
noveltreatmentoptionsofobesity.
1.Introduction [9]. PAI-1 levels can be increased in response to hypoxia
[10, 11], hormones like insulin [12, 13], coagulation factors,
Obesity is becoming an increasing public health problem andcytokines(discussedby[14]).MorerecentlyPAI-1levels
worldwide. The excessive accumulation of adipose tissue havebeenconsideredasoneofthebiomarkersusedtopredict
leads to thedevelopmentofdyslipidemia, impairedglucose obesity-associateddiseases[15].ElevatedPAI-1mRNAlevels
metabolism, hypertension, and proinflammation, processes have been found in adipose tissues from obese ob/ob mice
playing an essential role in the pathogenesis of cardiovas- [16]andalsoinhumanobesitywithhigherexpressionlevels
culardisease, type2diabetes, themetabolicsyndrome,and invisceralcomparedtosubcutaneousadiposetissuedepots
various cancers (reviewed by [1]). Many of those obesity- [17].Thus,highplasmaPAI-1levelsareacommonfindingin
related pathophysiological conditions are associated with obesityinbothmiceandhumans[18–25].Mostimportantly,
increased plasminogen activator inhibitor-1 (PAI-1) levels the obesity-induced PAI-1 increase is reversible by lifestyle
[2–6]. PAI-1 is the primary, fast-acting inhibitor of both intervention.Weightlossduetocalorierestrictiondecreased
tissue-type and urokinase-type plasminogen activators and plasma PAI-1 concentrations in obese individuals [26, 27].
thereforecontrolstheregulationofthefibrinolyticsystemin These data imply that substances that potentially mimic
blood[7,8].Inaddition,PAI-1isanimportantregulatorof calorierestrictionmaybeusedasmodulatorsofPAI-1levels
extracellularmatrixturnover,tissueremodeling,andfibrosis inthetreatmentofobesityandobesity-relateddiseases.
2 OxidativeMedicineandCellularLongevity
Fromanumberofnaturalcompoundsmimickingcalorie Mouseembryonicfibroblasts(MEFs)weremaintainedin
restriction by targeting various metabolic pathways, resver- DMEM supplemented with 10% fetal bovine serum (Invit-
atrol gained special interest. Resveratrol is a polyphenol rogen, Karlsruhe, Germany), 1% nonessential amino acids
producedbyplantsinresponsetoenvironmentalstressand (Invitrogen), and 0.5% antibiotics in an atmosphere of 16%
∘
found in red grape skin, peanuts, a variety of berries, and O2,5%CO2,and97%humidityat37 Cinacellcultureincu-
medicalplants[28].Ithasbeensuggestedtoactasacalorie bator.MouseembryonicfibroblastsSirt1+/+andSirt1−/−were
restrictionmimeticbasedondatafromrodents.Whenmice agenerousgiftfromDr.MichaelMcBurney(OttawaHospital
and/or rats were fed a high-fat diet, resveratrol treatment ResearchInstitute,Canada).WeobtainedAMPK𝛼1,2+/+and
improvedglucosehomeostasis,mitochondrialfunction,lipid AMPK𝛼1,2−/− MEFs [46] from Dr. Benoit Viollet (Institut
parameters, body weight, and survival [29–39]. While the
Cochin,Paris,France).Nrf2wild-typeandNrf2knockdown
resveratrol effects are intensively studied in animal models
MEFswereprovidedbyDr.StephanImmenschuh(Hannover
only few clinical trials were conducted so far to study
MedicalSchool,Germany).
the effects of resveratrol supplementation in the context of
humanobesity[40]andcoronaryarterydisease[41];yetthere
2.2.RNAPreparationandQuantitativeReal-TimePCR. Iso-
existssomecontroversy[42]andtheeffectofresveratrolin
lation of total RNA was performed using the peqGOLD
obesehumanindividualsremainstobefurtherinvestigated.
HP Total RNA kit (Peqlab, Erlangen, Germany) follow-
Althoughobesityandobesity-associateddiseasesseemto
ing the manufacturer’s instructions. One 𝜇g of total RNA
be positively influenced by resveratrol, not much is known
was used for cDNA synthesis with using SuperScript II
abouttheeffectofresveratrolonPAI-1inobesity.Therefore,
Reverse Transcriptase (Invitrogen, Darmstadt, Germany).
itwastheaimofthisstudytoinvestigatetheeffectsofresver-
Quantitativereal-timePCRwasperformedwithaLightCy-
atrol on the production of PAI-1 in human adipocytes and
cler 2.0 (Roche Diagnostics, Mannheim, Germany) using a
inaninvitromodelofhumanadiposetissueinflammation.
LightCycler FastStart DNA Master PLUS SYBR Green I kit
WefoundthatresveratrolreducesPAI-1levelsinadipocytes
(RocheDiagnostics,Mannheim,Germany).Thequantitative
especially under inflammatory conditions. Thus, our data
real-time PCR results were normalized using hypoxanthine
support the concept that resveratrol can alleviate obesity-
phosphoribosyltransferase (HPRT) as a housekeeping gene.
inducedupregulationofPAI-1inhumanadiposetissue. 󸀠
The following primer sets were used: human PAI-1-F (5 -
ACA AGT TCA ACT ATA CTG AGT TCA CCA CGC
󸀠 󸀠
CC-3 ),humanPAI-1-Rsequence(5 -TGAAACTGTCTG
2.MaterialsandMethods 󸀠 󸀠
AAC ATG TCG GTC ATT CCC-3 ), human HPRT-F (5 -
󸀠
GAGATGGGAGGCCATCACATTGTAGCCCTC-3 ),
2.1. Reagents and Cell Culture. All biochemicals were of 󸀠
and human HPRT-R (5 -CTC CAC CAA TTA CTT TTA
analyticalgradeandwerepurchasedfromcommercialsup- 󸀠
TGTCCCCTGTTGACTGGTC-3 ).Theexperimentsfor
pliers. Resveratrol, sirtinol, and LY204002 were obtained
each data point were carried out in triplicate. The relative
from Sigma (Deisenhofen, Germany). SC-514 was from
quantificationofgeneexpressionwasdeterminedusingthe
Merck Millipore (Darmstadt, Germany). Small molecule
ΔΔCt method [47]. In some experiments conventional RT-
inhibitorsweredilutedinDMSOwhichalonewasalsoused 󸀠
PCRwasperformedusingSp1asareferencegene(PAI-1-F:5-
asvehiclecontrol.Thefollowingconcentrationsofresveratrol 󸀠 󸀠
GTCTGCTGTGCACCATCCCCC-3 ;PAI-1-R:5 -GAA
and inhibitors were used in experiments: resveratrol 10, 50, 󸀠 󸀠
CAGCCTGAAGAAGTGGGGC-3 ,Sp1-F:5 -ACTACC
100𝜇M,sirtinol10𝜇M;LY20400220𝜇M,SC-514100𝜇M. 󸀠 󸀠
AGT GGA TCA TCA GGG-3 ; Sp1-R: 5 -CTG ACA ATG
Simpson-Golabi-Behmel syndrome (SGBS) preadipo- 󸀠
GTGCTGCTTGGA-3 ).
cyteswereculturedaspreviouslydescribed[43].Humanpri-
mary preadipocytes were prepared by collagenase digestion
fromsubcutaneousadiposetissueof3healthywomenusing 2.3.ELISA. SGBSadipocytesweretreatedfor48hwith10%
apreviouslydescribedprotocol[44].Adipogenicdifferentia- MacCM,100𝜇Mresveratrol,and100𝜇MSC-514aloneorin
tion of SGBS and human primary and SGBS preadipocytes combination.TheELISAwasperformedusingthePlatinum
was induced in serum-free DMEM/F12 medium supple- ELISA kit for human PAI-1 (eBioscience, Vienna, Austria).
mentedwith10𝜇g/mLiron-poortransferrin,10nMinsulin, Absorbance was measured on a spectrophotometer using
200pM thyroid hormone, and 0.1𝜇M cortisol and for the 450nmwavelength(ELx800AbsorbanceMicroplateReader,
first fourdays2𝜇Mrosiglitazone,250𝜇Misobutylmethylx- BioTek,BadFriedrichshall,Germany).
anthine, and 25nM dexamethasone. Cells were used for
experimentsonday8ofadipogenicdifferentiation. 2.4. WesternBlotAnalyses. Westernblotanalyseswereper-
THP-1 cells (ATCC, Wesel, Germany) were cultured as formed as previously described [10]. In brief, 24h after
describedearlier[45].Differentiationintomacrophageswas treatmentwithvehicleorresveratrolcellculturemedium(for
induced by 125ng/mL phorbol myristate acetate for 48h. PAI-1)ortotalcelllysateswerecollectedand100𝜇gofprotein
Macrophage-conditioned medium (MacCM) was collected wassubjectedtoSDS-PAGEandblottedontoanitrocellulose
after additional 48h of incubation in serum-free basal membrane. The following primary antibodies were used:
mediumcontaining0.5%BSAandclearedbycentrifugation. PAI-1(polyclonal1:100)(AmericanDiagnostics,Pfungstadt,
MacCM from 5 independently performed productions was Germany), AMPK𝛽1/2 (polyclonal, 1:1000) (Cell Signaling,
pooledandthenusedforexperiments. Hamburg, Germany), Nrf2 (polyclonal, Nrf2 1:200) (Santa
OxidativeMedicineandCellularLongevity 3
Cruz, Heidelberg, Germany), and Sirt1 (polyclonal, 1:1000) 3.Results
(Santa Cruz, Heidelberg, Germany). The secondary anti-
body was anti-rabbit immunoglobulin G (IgG)-horseradish 3.1. Concentration- and Time-Dependent Downregulation of
peroxidase IgG (1:5000) (Biorad, Munich, Germany). The Human PAI-1 mRNA and Protein Levels by Resveratrol in
enhanced chemiluminescence (ECL) system (Amersham, SGBS Adipocytes. To determine how resveratrol modulates
Freiburg, Germany) was used for detection. Blots were PAI-1geneexpressioninSGBSadipocytes,weexaminedPAI-
quantifiedbyusingtheFijiprogram(NCBI). 1 mRNA and protein levels after treatment with increas-
ing concentrations of resveratrol at different time points
(Figure1). Treatment of cells for 12h, 24h, and 48h with
2.5.ROSMeasurement. TodetermineROSproduction,SGBS
differentconcentrationsofresveratrolresultedinareduction
adipocytes were incubated with 2.5𝜇M CM-H2DCFDA ofPAI-1mRNAlevelsinadose-dependentmanner(datanot
(Mol∘ecularProbesEuropeBV,TheNetherlands)for30min shown); 100𝜇M resveratrol reduced PAI-1 mRNA levels by
at37 C.AfterthreewasheswithPBS,cellsweretreatedwith
about40%after12handbyabout60%after48h(Figures1(a)
100𝜇MH2O2or10%MacCMfor15minandanalyzedbyflow and1(b)).Theresveratrol-mediateddecreaseofPAI-1mRNA
cytometry.
wasfollowedbyadecreaseofPAI-1proteinlevels.Aresver-
atrol concentration of 50𝜇M or 100𝜇M diminished PAI-
2.6.PreparationofNuclearExtractsandElectrophoreticMobil- 1 protein levels in the medium by about 50% after 24h
ity Shift Assay (EMSA). SGBS adipocytes were treated with and by about 75% after 48h (Figure 1(c)). Thus, resveratrol
100𝜇M resveratrol, 100𝜇M SC-514, and 10% MacCM alone reducedPAI-1mRNAandPAI-1proteinlevelsinatime-and
orincombination.TNF𝛼(10ng/mL)wasusedasapositive concentration-dependentmanner.
control.Cellswerecollectedfrom6cmdishesbyscrapingand
∘
centrifugation(10,000gfor5minat4 C).Afterwashingonce
with ice-cold PBS, cell pellets were resuspended in 200𝜇L 3.2.PAI-1GeneExpressionIsUpregulatedinanInVitroModel
low-saltbuffer(10mMHEPES-KOHpH7.9;1.5mMMgCl2; of Inflamed Human Adipose Tissue as well as in Primary
10mMKCl) and incubated for 10min on ice. After addi- Human Adipocytes. Obesity is associated with low-grade
tion of 12.5𝜇L of a 10% Nonidet P-40 solution, samples chronic inflammation [48] and increased circulating PAI-
were mixed vigorously for 30s. Nuclei were collected by 1 levels [6]. Therefore, we mimicked human adipose tissue
centrifugation and resuspended in 25𝜇L high-salt buffer inflammation by using our previously described in vitro
(20mMHEPES-KOHpH7.9;1.5mMMgCl2;420mMNaCl, modelsystem[49]whereweincubatedSGBSadipocyteswith
0.2mM EDTA; 25% glycerol). Both buffers were supple- mediumsupplementedwithincreasingdosesofmacrophage-
mented with a protease-inhibitor cocktail (Sigma), 0.2mM conditioned medium (MacCM) for 48h. As shown in
PMSF, 0.5mM dithiothreitol (DTT), and 1mM sodium- Figure2(a),thepresenceofMacCMincreasedPAI-1mRNA
orthovanadate before use. Nuclei were incubated 15min in SGBS adipocytes; already 5% MacCM induced PAI-1
on ice and vortexed periodically. Nuclear extracts were mRNAbyabout2-fold.Inlinewiththesefindings,treatment
obtainedbycentrifugationat12,500gfor10minat4∘Cand ofprimaryhumanexvivodifferentiatedadipocytesobtained
storedat−80∘C.Proteinconcentrationwasdeterminedwith from healthy donors with MacCM increasing PAI-1 mRNA
the BCA Protein Assay Reagent kit (Pierce, Rockford, IL), by about 1.6-fold (Figure 2(b)). Thus, these data suggested
according to manufacturer’s instructions. Single-stranded that obesity mimicking inflammatory conditions lead to an
oligonucleotides were purchased from Biomers.net (Ulm, upregulationofPAI-1.
Germany):standardNF𝜅Bprobe:sense,5󸀠-AGTTGAGGG
󸀠 󸀠
GAC TTT CCC AGG C-3 ; antisense, 5 -GCC TGG GAA
󸀠 3.3. Resveratrol Reduces Upregulation of PAI-1 Gene Expres-
AGT CCC CTC AAC T-3 . The sense oligonucleotide was
sion in an In Vitro Model of Inflamed Human Adipose
labeled with 𝛾-32P-ATP (Amersham, Freiburg, Germany)
Tissue. To determine the effect of resveratrol on the ele-
using T4-polynucleotide kinase (MBI Fermentas, St. Leon-
vated PAI-1 mRNA and protein levels under inflammatory
Rot, Germany). A 2-fold molar excess of unlabeled anti-
conditions, SGBS adipocytes were cultured in the absence
senseoligonucleotidewasannealed,andthelabeleddouble-
or presence of different concentrations of resveratrol, 10%
stranded oligonucleotide was purified with a spin column
MacCM, or a combination of both for 48h. Treatment of
(MicroBio-SpinP30;Bio-Rad,Munich,Germany).Binding
SGBS adipocytes with increasing doses of resveratrol alone
reactions were performed for 30min on ice in 20𝜇L buffer
resulted in a concentration-dependent reduction of PAI-
(1mMMgCl2, 0.5mMEDTA, 0.5mMDTT, 50mMNaCl, 1 mRNA and protein levels (Figures 3(a), 3(b) and 3(c)).
10mMTris-HCl,pH7.5;4%glycerol)containing5𝜇gnuclear
Incubation of cells with MacCM induced PAI-1 mRNA
extract protein, 1𝜇g poly (dI:dC) (Sigma), and 10,000cpm-
levelsandPAI-1proteinlevelsbyabout3-fold(Figures3(a),
labeledoligonucleotide.
3(b) and 3(c)). The MacCM-dependent induction of PAI-
1 mRNA and protein levels was abolished in the pres-
2.7. Statistics. Data represents mean ± standard error of ence of 100𝜇M resveratrol (Figures 3(a), 3(b) and 3(c)).
means(SEM)of3independentexperimentsunlessotherwise Together, these data suggested that PAI-1 gene expres-
stated. Statistics: statistical significance was evaluated using sion is enhanced under inflammatory conditions and that
one-wayanalysisofvariants(ANOVA)considering𝑃<0.05 this induction is antagonized by the action of resvera-
asstatisticallysignificant. trol.
4 OxidativeMedicineandCellularLongevity
1
els
v
e
A l0.8
N
R
m
1 0.6
p
S
1/
AI-0.4 DMSO Res (100𝜇M)
P
Time (h) 6 12 24 48 6 12 24 48
0.2 PAI-1 mRNA
Sp1 mRNA
6 12 24 48
Time (h)
DMSO
Res (100𝜇M)
(a) (b)
45
40
L)
m 35
g/
n
el ( 30
v
e
n l 25
ei
ot
pr 20
1
AI- 15
P
10
5
0 10 50 100
Resveratrol (𝜇M)
24h
48h
(c)
Figure1:Resveratrol-dependentdownregulationofPAI-1mRNAandproteinlevelsinSGBSadipocytes.SGBSadipocyteswereincubated
inadipogenicmediawithvehiclecontrol(DMSO)or100𝜇Mresveratrol(Res)fortheindicatedtimepoints.(a)PAI-1mRNAlevelswere
measuredbysemiquantitativeRT-PCR.Sp1wasusedasareferencegene.(b)ArepresentativeRT-PCRofPAI-1andSp1mRNAlevelsafter
treatmentwithDMSOor100𝜇MRes.(c)TheaccumulationofPAI-1inthemediawasmeasuredbyELISAaftertreatmentwithincreasing
dosesofResfor24or48h.
3.4. The Effects of Resveratrol on PAI-1 Expression Are Not incubatedwithDMSOasavehiclecontrol,theSirt1inhibitor
Mediated via Sirt1, AMPK, or PI3K. Resveratrol has been sirtinol or PI3K inhibitor LY294002 along with resveratrol,
shown to modulate several key signaling molecules in MacCM,orcombinations,andthePAI-1mRNAlevelswere
adipocytes, including Sirt1 [50, 51], AMPK [52–54], and determined48haftertreatment.
PI3K/Akt [52, 55–57]. To examine whether Sirt1, AMPK, In line with the above mentioned results, resveratrol
and/or PI3K are involved in the resveratrol-dependent decreased PAI-1 mRNA levels in SGBS adipocytes cultured
downregulation of PAI-1 gene expression, we used specific either with or without MacCM. Sirtinol treatment alone
inhibitors of these signaling pathways as well as knockout slightly reduced the basal expression of PAI-1 mRNA in
cells.Concerningtheinhibitorstudies,SGBSadipocyteswere SGBS adipocytes (Figure 4(a)). However, sirtinol had no
OxidativeMedicineandCellularLongevity 5
4 2 ∗
∗
NA level 3 ∗ ∗ mRNA level
R
m 1
PAI-1 2 ∗ ve PAI- 1
ve ati
elati Rel
R 1
0 0
0 1 5 10 20 0 10
MacCM (%) MacCM (%)
(a) (b)
Figure2:PAI-1geneexpressionisupregulatedinaninvitromodelofinflamedhumanadiposetissue.(a)SGBSadipocyteswereincubated
withincreasingdosesofmacrophage-conditionedmedia(MacCM)orvehiclefor48h.PAI-1mRNAlevelswereanalyzedbyqPCRandresults
∗
werenormalizedtoHPRT. significantdifferencecontrolversusMacCM.(b)Primaryhumanexvivodifferentiatedadipocytesisolatedfrom
3patientsweretreatedwith10%macrophage-conditionedmedia(MacCM)orvehiclefor48h.PAI-1mRNAlevelswereanalyzedwithqPCR
∗
andnormalizedtoHPRT. significantdifferencecontrolversusMacCM.
significant effect on the resveratrol-dependent downregula- untreated and MacCM-treated SGBS adipocytes, LY294002
tionofthePAI-1mRNAinSGBSadipocytesincubatedwith treatment did not change the basal PAI-1 mRNA levels
MacCM (Figure 4(a)). To further rule out the role of Sirt1 (Figure 4(f)). Furthermore, incubation with LY294002 did
inresveratrol-dependentregulationofPAI-1expression,we not block the decline of PAI-1 mRNA levels by resveratrol
examined the effect of resveratrol on the PAI-1 protein lev- in both untreated and MacCM-treated SGBS adipocytes,
els in Sirt1-deficient (Sirt1−/−) mouse embryonic fibroblasts implicatingthatthePI3K/Aktpathwayisnotinvolvedinthe
(MEFs).AlthoughthebasalPAI-1proteinlevelswerelowerin resveratrol-modulateddownregulationofPAI-1.
−/−
Sirt1 MEFs,resveratroltreatmentdecreasedthePAI-1lev-
+/+ −/−
elsinbothwild-type(Sirt1 )andtheSirt1 MEFsbyabout
3.5.ROSFormationandtheAntioxidantTranscriptionFactor
50%(Figures4(b)and4(c)).Together,theseresultsindicate
Nrf2 Do Not Contribute to the Effects of Resveratrol on PAI-
thatalthoughSirt1persemightbeinvolvedintheregulation
1GeneExpression. Obesityandinflammationareassociated
ofPAI-1geneexpression,theresveratrol-dependentmodula-
with increased ROS formation [58, 59] and ROS-mediated
tionofPAI-1geneexpressionisindependentofSirt1.
signalinghasbeenreportedtoregulatePAI-1geneexpression
NextweinvestigatedtheroleofAMPKintheresveratrol-
[60,61].Resveratroliswellknownforitsantioxidantpotential
dependent regulation of PAI-1 expression. For this pur-
pose we used wild-type AMPK𝛼1/2+/+ (AMPK𝛽1/2+/+) and andthereforeweaimedtodeterminewhethertheobserved
MacCM-dependentinductionofPAI-1geneexpressionand
AMPK𝛼1/2-deficient (AMPK𝛼1/2−/−) MEFs and measured
hence the effects of resveratrol were dependent on ROS
PAI-1proteinlevelsaftertreatmentwithresveratrolbyWest-
generation.Toaddressthisissue,ROSlevelswereexamined
ern blot. The basal PAI-1 protein levels were significantly
inSGBSadipocytestreatedwithMacCMorforthepurpose
lowerinAMPK𝛼1/2−/−MEFs,butagainresveratroltreatment of a positive control with H2O2. Intracellular ROS levels
resultedinasignificantdecrease(byabout65%)ofthePAI-1 increasedupontreatmentwithH2O2.Bycontrast,nochanges
proteinlevelsinbothwild-typeandtheAMPK𝛼1/2−/−MEFs in ROS generation were detected in MacCM-treated SGBS
(Figures4(d)and4(e)).Together,thesedatashowthateven adipocytes (Figures 5(a) and 5(b)) implying that MacCM-
though AMPK itself might be involved in the regulation dependentPAI-1inductionisindependentofROS.
of PAI-1 gene expression, the resveratrol-dependent down- The NFE2-related factor 2 (Nrf2) is a key transcription
regulation of PAI-1 is mediated by an AMPK-independent factor, involved in the primary cellular defense against the
mechanism. cytotoxic effects of oxidative stress [62]. To further exclude
We further studied whether the PI3K/Akt pathway is the possibility that the effects of MacCM and resveratrol
involved in resveratrol-dependent downregulation of PAI- onPAI-1geneexpressionareindependentofROS,weused
1 and used the PI3K inhibitor LY294002. While resveratrol wild-type and Nrf2 knockdown MEFs. Interestingly, the
treatmentreducedPAI-1mRNAlevelsbyabout50%inboth knockdownofNrf2increasedPAI-1proteinlevelscompared
6 OxidativeMedicineandCellularLongevity
5
∗
vel 4 ∗
e
A l ∗∗
N ∗
R 3
m
AI-1 2 Res (𝜇M) 0 10 30 100
ve P ∗∗ MacCM (10%) − + − + − + − +
Relati 1 ∗ ∗ PAI-1
𝛽-actin
Res (𝜇M) − 10 30 100 − 10 30 100
MacCM (10%) − − − − + + + +
(a) (b)
120 ∗ ∗
L)
m ∗∗
g/ 100 ∗
n
el ( 80
v
e
n l
ei 60 ∗∗
ot
pr
1 40
AI- ∗
P 20
Res (𝜇M) − 10 30 100 − 10 30 100
MacCM (10%) − − − − + + + +
(c)
Figure3:ResveratrolabolishedtheMacCM-dependentPAI-1inductioninSGBSadipocytes.SGBSadipocytesweretreatedwiththeindicated
dosesofresveratrol(Res),10%MacCM,oracombinationofResand10%MacCMfor48h.(a)PAI-1mRNAlevelswereanalyzedbyqPCR
∗ ∗∗
andresultswerenormalizedtoHPRT. SignificantdifferenceuntreatedversusResorMacCM, significantdifferenceMacCMtreatedversus
MacCM+Res.(b)TotalcellproteinlysateswereisolatedandsubjectedtoWesternblotanalysisusinganantibodyagainstPAI-1and𝛽-actin
∗
asaloadingcontrol.(c)AccumulationofPAI-1proteininmediawasmeasuredbyELISA. SignificantdifferenceuntreatedversusResor
∗∗
MacCM, SignificantdifferenceMacCMtreatedversusMacCM+Res.
tothewild-typecells(Figures5(c)and5(d))buttheaddition extracts from cells treated with MacCM and resveratrol or
ofresveratrolcausedadecreaseinPAI-1levelsbyabout80% MacCM and SC-514 (Figure 6(a)). These data demonstrate
inwild-typecellsandbyabout35%inNrf2knockdowncells that resveratrol can lead to a suppression of NF𝜅B DNA-
(Figures 5(c) and 5(d)). Thus, the antioxidant transcription binding activity under inflammatory conditions in SGBS
factorNrf2isnotinvolvedinmediatingtheresveratroleffects adipocytes.Basedontheabovefindings,demonstratingthe
onPAI-1expression. suppressiveeffectofresveratrolonNF𝜅BDNA-binding,we
expectedthatinhibitionofNF𝜅Bbyresveratrolwouldreduce
3.6. The Effects of Resveratrol on PAI-1 Gene Expression in PAI-1 gene expression. Accordingly, SGBS adipocytes were
an In Vitro Model of Inflamed Adipose Tissue Are NF𝜅B treated with MacCM, resveratrol, and SC-514 alone or in
Dependent. ThereductionofPAI-1expressionbyresveratrol combination, and PAI-1 protein levels were measured by
under inflammatory conditions may be partially explained ELISA. In line, resveratrol and SC-514 reduced MacCM-
bytheabilityofresveratroltosuppresstheactivityofNF𝜅B, dependent PAI-1 protein induction (Figure 6(b)), though
a transcription factor critically involved in inflammation. theeffectsofresveratrolweremuchmorepronouncedthan
Therefore,weexaminedtheeffectofresveratrolontheDNA- the effects of the NF𝜅B inhibitor SC-514 alone. These data
binding activity of NF𝜅B in the model of inflamed adipose stronglysuggestthattheeffectsofresveratrolonPAI-1gene
tissue. By performing EMSA, we found that an oligonu- expressioninSGBSadipocytesareNF𝜅Bdependent.
cleotidewithaNF𝜅Bbindingsitewasabletoformasingle
DNA-protein complex (Figure 6(a)) when incubated with 4.Discussion
nuclear extracts from SGBS adipocytes treated with either
MacCMortheestablishedNF𝜅Bactivator,TNF-𝛼.TheNF𝜅B In this study we investigated the human PAI-1 expression
DNA-binding activity was significantly reduced in nuclear in response to resveratrol in human SGBS adipocytes and
OxidativeMedicineandCellularLongevity 7
4 100 Sirt1+/+ Sirt1−/−
evel ∗ )%
NA l 3 ∗∗∗∗∗ ( lev 80
mR el n 60 ∗
elative PAI-1 12 ∗ ∗∗ ∗∗∗ # ietorp 1-IAP 4200 ∗∗ ∗
R
MacCM (10%) − − − − + + + + Res (100𝜇M) − + − +
Res (100𝜇M) − + − + − + − +
Sirtinol (10𝜇M) − − + + − − + +
(a) (b)
AMPK𝛼1,2+/+ AMPK𝛼1,2−/−
100
+/+ −/−
Sirt1 Sirt1
Res − + − + )% 80
( le
PAI-1 ve
l n 60
ie
Ponceau S torp
1 40 ∗
-IA ∗∗
SIRT1 P
20 ∗
Res (100𝜇M) − + − +
(c) (d)
AMPK𝛼1/2+/+ AMPK𝛼1/2−/− 4
Res − + − + evel ∗ ∗∗
A l 3
PAI-1 N
R
m
Ponceau S AI-1 2 ∗∗∗
P
e
AMPK𝛼1/2 elativ 1 ∗ ∗∗ #
R
MacCM (10%) − − − − + + + +
Res (100𝜇M) − + − + − + − +
LY294002 (20𝜇M) − − + + − − + +
(e) (f)
Figure4:TheeffectsofresveratrolonPAI-1geneexpressioninSGBSadipocytesarenotmediatedviaSIRT1,AMPKandPI3K.(a)Where
indicatedSGBSadipocytesweretreatedwith10𝜇Msirtinol,resveratrol(Res,100𝜇M),andMacCM(10%)for48h.(a)PAI-1mRNAlevelswere
∗ ∗∗
analyzedwithqPCRandresultswerenormalizedtoHPRT. significantdifferenceuntreatedversusRes,sirtinol,orMacCM; significant
∗∗∗
differenceuntreatedversusRes+sirtinolorRes+MacCM; significantdifferenceMacCMversusMacCM+ResorMacCM+sirtinol;
# +/+ −/−
significantdifferenceMacCMversusMacCM+Res+sirtinol.(b)SIRT1 andSIRT1 mouseembryonicfibroblastsweretreatedwith
100𝜇Mresveratrol(Res)orvehiclecontrol(DMSO)for24h.ThePAI-1andSIRT1proteinlevelsweremeasuredbyWesternblot.∗Significant
differenceuntreatedversusRes,∗∗significantdifferencewild-typeversusknockoutcells.(c)RepresentativeWesternblot.(d)AMPK𝛼1/2+/+
andAMPK𝛼1/2−/−mouseembryonicfibroblastsweretreatedwith100𝜇Mresveratrol(Res)orvehiclecontrol(DMSO)for24h.ThePAI-1and
AMPK𝛼1/2proteinlevelsweremeasuredbyWesternblot.∗significantdifferenceuntreatedversusRes,∗∗significantdifferencewildtypeversus
knockoutcells.(e)RepresentativeWesternblot.(f)WhereindicatedSGBSadipocytesweretreatedwith20𝜇MLY294002,100𝜇Mresveratrol
∗
(Res),and10%MacCMfor48h.ThePAI-1mRNAlevelsweremeasuredbyqPCRandresultswerenormalizedtoHPRT. Significantdifference
∗∗ ∗∗∗
untreated versus Res, LY294002, or MacCM; significant difference untreated versus Res + LY294002 or Res + MacCM; significant
#
differenceMacCMversusMacCM+ResorMacCM+LY204002; significantdifferenceMacCMversusMacCM+Res+LY294002.
8 OxidativeMedicineandCellularLongevity
25 50
%) ∗ 40
ve cells (1250 stnuoC30 M1
ositi 20
p
S-10 10
O
R
5 0
100 101 102 103 104
Med H2O2 MacCM (10%) FL1-height
(a) (b)
WT KD-Nrf2
∗∗
WT KD-Nrf2
10
)dlo Res − + − +
f( le 9 PAI-1
ve ∗
l n
ieto 8 Ponceau S
rp
1
-IA 2 Nrf2
P
1
∗
Res (100𝜇M) − + − +
(c) (d)
Figure5:Macrophage-conditionedmediadonotinduceROSformationinhumanadipocytesandtheantioxidanttranscriptionfactorNrf2
doesnotcontributetotheresveratroleffectsonPAI-1geneexpression.(a),(b)SGBSadipocyteswerelabelledwith2.5𝜇MCM-H2DCFDAand
thentreatedwith50𝜇MH2O2and10%MacCMfor15min.ROSproductionwasanalyzedbyflowcytometry.(a)ROS-positiveadipocytes
∗
aftertreatmentwithH2O2 andMacCM; significantdifferenceuntreatedversusH2O2.(b)HistogramsofROS-positivecellpercentagein
+/+
cellsculturedinmediumortreatedwithH2O2for15min.(c)Nrf2 andNrf2knock-downmouseembryonicfibroblastsweretreatedwith
100𝜇Mresveratrol(Res)orcorrespondingvehiclecontrol(DMSO)for24h.ThePAI-1andNrf2proteinlevelsweremeasuredbyWestern
∗ ∗∗
blot. SignificantdifferenceuntreatedversusRes, significantdifferencewildtypeversusknockoutcells.(d)RepresentativeWesternblot.
in a model of inflamed human adipose tissue. Our data chemotacticprotein1,MCP-1),andadipokines(haptoglobin,
demonstrated several new findings with respect to resvera- PAI-1,leptin,visfatin,resistinandVEGF)[63].PlasmaPAI-
trol and human PAI-1 regulation under obesity-mimicking 1 levels are considerably enhanced in obese humans and
conditions.First,itwasfoundthatresveratroldownregulated in patients with insulin resistance, type 2 diabetes, and
PAI-1mRNAandproteinlevelsinatime-andconcentration- cardiovascular diseases [23, 64]. The adipose tissue appears
dependentmannerinhumanSGBSadipocytes.Second,the to be the major source of elevated PAI-1 levels observed in
inhibitoryeffectofresveratrolonPAI-1wasevenstrongeron obesity [65, 66] maybe as a result of its increased capacity
the obesity-associated and inflammation-dependent induc- to produce PAI-1 and/or as an effect of direct stimulation
tion of PAI-1. Third, while resveratrol exerted its effects on of adipocytes by hormones and cytokines upregulated in
inflammatory-dependent PAI-1 gene expression mainly via obesity [67]. Resveratrol is capable of attenuating obesity-
inhibition of NF𝜅B, signaling via Sirt1, AMPK, PI3K, ROS, associated inflammatory responses by inducing changes in
and Nrf2 did not mediate the effect of resveratrol on PAI-1 the secretion profile of adipocytes [68–71]. In particular
production. resveratrolinhibitedTNF𝛼-dependentPAI-1upregulationin
Obesity represents a risk factor for the development of 3T3-L1 adipocytes [68, 70],IL1𝛽-stimulatedPAI-1 secretion
diseases like type 2 diabetes, hypertension, atherosclerosis [69],andPAI-1productioninhumanSGBSadipocytes[71].
andmyocardialinfarction.Intriguingly,obesityisalsoasso- These data are very much in line with the results from the
ciatedwithastateofchroniclow-gradeinflammationchar- present study where we have shown that resveratrol not
acterizedbyelevatedplasmaconcentrationsofproinflamma- only downregulated PAI-1 expression (Figure 1) but even
torycytokines(IL-6,IL-1andTNF𝛼),chemokines(monocyte exerted a stronger effect on PAI-1 in a model of inflamed
OxidativeMedicineandCellularLongevity 9
Res (100𝜇M) − − + + − − − 140
MacCM (10%) − + − + − + −
SC-514 (100𝜇M) − − − − + + − 120 ∗
TNF𝛼 − − − − − − +
L)
m100
g/ ∗∗
n
el ( 80
v
e
n l
ei 60
ot ∗∗
AI-1 pr 40 ∗
P ∗
20
Res (100𝜇M) − − + + − −
MacCM (10%) − + − + − +
SC-514 (100𝜇M) − − − − + +
(a) (b)
Figure6:Resveratrol-mediatedsuppressionofNF𝜅BDNAbindingactivitydidnotabrogatePAI-1geneexpression.(a)Electrophoretic
mobilityshiftassayusinga5󸀠-end-labelledconsensusoligonucleotideforNF𝜅BbindingandnuclearextractsfromSGBSadipocytes.SGBS
adipocytes were treated with 10% MacCM, 100𝜇M resveratrol, 100𝜇M SC-514, or combination of them and then incubated for 1h. The
DNA-proteincomplexeswereseparatedbyelectrophoresison5%nativepolyacrilamidegelsandvisualizedbyphosphoimaging.(b)SGBS
adipocytesweretreatedwith10%MacCM,100𝜇Mresveratrol(Res),100𝜇MSC-514,oracombinationofbothResand10%MacCMorSC-514
∗
andMacCMfor48h.AccumulationofPAI-1proteininmediawasmeasuredbyELISA. SignificantdifferenceuntreatedversusMacCM,Res
∗∗
orSC-514; significantdifferenceMacCMtreatedversusMacCM+ResorMacCM+SC-514.
human adipose tissue (Figures 2 and 3). Although all these Resveratrolisknowntoexertpleiotropiceffectsoncells
data indicate that resveratrol can alleviate obesity-induced andSirt1activationisnottheonlyeffectviawhichresveratrol
upregulationofPAI-1inadiposetissue,ithasnotbeenfully exertsitsbeneficialactionsonobesity-associatedpathologi-
elucidatedbywhichmolecularmechanismsresveratrolexerts calconsequences[29,30,78].Therefore,theinhibitoryeffect
itseffectonPAI-1underinflammatoryconditions. of resveratrol on PAI-1 production in obesity may result
Calorie restriction is considered to be one of the most frommodulationofdifferentsignalingpathways.Someofthe
effectivenutritionalinterventionsprotectingagainstobesity, beneficial effects of resveratrol against diet-induced obesity
diabetes,andcardiovasculardisease[72].Theobesity-related and insulin resistance were mediated via AMPK activation
enhancement of PAI-1 levels also appeared to be reversible [29, 35, 36, 38, 78, 79]. In addition, increasing evidence
by calorie restriction diet or calorie restriction mimetics suggeststhatAMPKhasanti-inflammatoryactions[80,81].
[26, 27]. Several signaling pathways have been implicated Therefore, we have tested whether the effects of resveratrol
inmediatingthecalorierestrictioneffect—thesirtuinpath- on PAI-1 expression are mediated via AMPK. Our results
way,theadenosinemonophosphate(AMP)activatedprotein demonstratedthatresveratrol-dependentdownregulationof
kinase(AMPK)pathway,andtheinsulin-likegrowthfactor PAI-1 was still preserved in AMPK-deficient cells (Figures
(IGF-1)/insulin signaling pathway (as discussed by [73]). In 4(d)and4(e))pointingoutthatresveratrolactsonPAI-1in
rodents calorie restriction and calorie restriction mimetics anAMPK-independentmechanism.
seemtoextendthelifespanandarelinkedtosilentmating A number of experimental observations have demon-
type information regulation 2 homolog 1 (Sirt1) activation strated that the PI3K/Akt pathway represents an important
(references in [74]). Resveratrol was identified as a Sirt1 signaling cascade in the initiation of the inflammatory
activator[75]andgainedinterestinanumberofpathological response.Althoughweshowedinanearlierstudythatresver-
settings—amongthemobesity.Inline,theanti-inflammatory atrolinhibitsPI3K-drivenAktphosphorylationinSGBScells
effectsofresveratrolinadipocytesaswellasinhumanadipose [55] the PI3K inhibitor, LY294002, could not abrogate the
tissuewereshowntobemainlydependentonSirt1activation resveratrol-dependentdownregulationofPAI-1(Figure4(f))
[69, 76, 77]. However, in our study, neither inhibition of implicatingthatthePI3K/Aktpathwayisalsonotinvolvedin
Sirt1withsirtinolnordeficiencyofSirt1wasabletoabrogate themodulationofPAI-1expressionbyresveratrol.
the resveratrol effects on PAI-1 (Figures 4(a), 4(b) and Inflammationiswellknowntoexistincombinationwith
4(c)) implicating that Sirt1 activation is not necessary to oxidativestresswhichinturnisapotentmodulatorofPAI-
mediate the action of resveratrol on PAI-1 synthesis under 1 gene expression in different systems [60, 82] as well as
inflammatoryconditions. in this study. In this context, an important transcription
10 OxidativeMedicineandCellularLongevity
factor mediating responses to oxidative stress is Nrf-2 [83]. Acknowledgments
Resveratrolsupplementationhasbeenshownsignificantlyto
increaseNrf2activityinhumansafterameal[84].However, TheauthorsthankDr.MichaelMcBurney(OttawaHospital
the conditions of our inflammatory model did not induce ResearchInstitute,Canada)forthegenerousgiftoftheSirt1-
ROSgeneration(Figures5(a)and5(b)).Inlinewiththat,the wild-type and Sirt1-deficient mouse embryonic fibroblasts
knockdownofNrf2didnotimpairtheresveratroleffecton (MEFs). They gratefully acknowledge Dr. Benoit Viollet
PAI-1secretion(Figures5(c)and5(d)). (InstitutCochin,Paris,France)andDr.StephanImmenschuh
AnincreaseinplasmaPAI-1levelsobservedinobesitycan (Hannover medical School, Germany) for providing them
alsobetheresultofacytokine-dependentinductionofPAI- with AMPK𝛼1,2+/+, AMPK𝛼1,2−/−, and Nrf2 wild-type and
1transcriptionwheretheproinflammatorycytokinessuchas Nrf2knock-downMEFs,respectively.Theauthorsalsothank
IL-1, IL-6, and TNF𝛼 play the major role [85–87]. Interest- AlexandraKillianforexcellenttechnicalassistance.PFPwas
ingly, no STAT3 binding element participating in the IL-6 fundedbyaMargaretevonWrangellscholarshipfinancedby
response could be mapped in the PAI-1 promoter whereas theBaden-WuerttembergMinistryofScience,Researchand
theso-calledNF𝜅B-likesiteswithinthePAI-1promoterand Arts; the European Social Fund; and Ulm University. Ivana
a TNF𝛼-responsive enhancer located 15kb upstream of the Zagotta is funded by the International Graduate School in
transcriptionstartsitewereshowntoparticipateinresponse MolecularMedicineUlm.Thisstudywasinpartsupported
toIL-1andTNF𝛼(referencesin[14]). bytheResearchTrainingGroupGRK1041“MolecularDia-
Nuclear factor (NF)𝜅B is a transcription factor with a betologyandEndocrinologyinMedicine”toPamelaFischer-
central role in the induction of a chronic inflammatory Posovszky and Martin Wabitsch and the foundation “Das
stateassociatedwithobesity,developmentoftype2diabetes, zuckerkranke Kind” to Pamela Fischer-Posovszky. Work in
cardiovascular risk, and insulin resistance [88]. Previous theThomasKietzmannlabissupportedbygrantsfromthe
reports established resveratrol as an inhibitor of NF𝜅B [41, Biocenter Oulu, Academy of Finland, and Sigrid Juselius
89]andresveratroltreatmentofTNF𝛼-stimulatedadipocytes Foundation.
reduced the expression of proinflammatory cytokines [88].
Therefore,ourresultsshowingthattheresveratroleffectson References
PAI-1geneexpressionwereNF𝜅B-dependent(Figure6)are
inlinewiththosefindings. [1] S. D. De Ferranti and S. K. Osganian, “Epidemiology of
Interestingly a number of in vivo and in vitro studies paediatric metabolic syndrome and type 2 diabetes mellitus,”
showed an inhibitory role of the resveratrol target Sirt1 on DiabetesandVascularDiseaseResearch,vol.4,no.4,pp.285–
NF𝜅B signaling [76, 77, 90, 91]. Similarly, AMPK signal- 296,2007.
ing has been shown to inhibit the inflammatory responses [2] A. de Hamsten, G. Walldius, G. Dahlen et al., “Plasminogen
inducedbyNF𝜅BviaseveraldownstreamtargetsofAMPK activatorinhibitorinplasma:riskfactorforrecurrentmyocar-
dialinfarction,”TheLancet,vol.2,no.8549,pp.3–9,1987.
(referencesin[92]).Moreover,severalpreviousfindingshave
[3] I. Juhan-Vague, C. Roul, M. C. Alessi, J. P. Ardissone, M.
demonstrated that the PI3K/Akt pathway has a crucial role
in the activation of the NF𝜅B pathway [93, 94]. Based on Heim,andP.Vague,“Increasedplasminogenactivatorinhibitor
activityinnoninsulindependentdiabeticpatients:relationship
thesestudiesandtheroleofresveratrolasaSirtandAMPK
withplasmainsulin,”ThrombosisandHaemostasis,vol.61,no.
activator, PI3K inhibitor as well as ROS scavenger, and we
3,pp.370–373,1989.
were expecting that Sirt, AMPK, PI3K, or ROS would be
[4] D. J. Schneider, T. K. Nordt, and B. E. Sobel, “Attenuated
involvedintheresveratroleffects.Surprisingly,noneofthese fibrinolysis and accelerated atherogenesis in type II diabetic
upstream NF𝜅B modulators contributed to the effects of patients,”Diabetes,vol.42,no.1,pp.1–7,1993.
resveratrol;however,inlinewithpreviousstudies[77,88,95– [5] I. Juhan-Vague and M. C. Alessi, “PAI-1, obesity, insulin
100] our findings show that resveratrol can act as an NF𝜅B resistance and risk of cardiovascular events,” Thrombosis and
inhibitor,mostlikelyviasofarnotacharacterizedpathway. Haemostasis,vol.78,no.1,pp.656–660,1997.
[6] C.DellasandD.J.Loskutoff,“HistoricalanalysisofPAI-1from
itsdiscoverytoitspotentialroleincellmotilityanddisease,”
5.Conclusions
Journal of Thrombosis and Haemostasis, vol. 93, pp. 631–640,
2005.
Together, our study showing that resveratrol mediates an
[7] H. R. Lijnen, L. Nelles, B. Van Hoef, E. Demarsin, and D.
inhibitoryeffectonPAI-1maybeusefultofurtherestablish Collen,“Characterizationofachimericplasminogenactivator
PAI-1asamarkerforobesity-associatedinflammatorycon- consistingofaminoacids1to274oftissue-typeplasminogen
ditions.In addition,we add at least onenovelaspect to the activatorandaminoacids138to411ofsingle-chainurokinase-
pleiotropyoftheresveratrolactionbyshowingthatitcanact typeplasminogenactivator,”JournalofBiologicalChemistry,vol.
asanNF𝜅BinhibitorwithoutinvolvingSirt1,AMPK,PI3Kor 263,no.35,pp.19083–19091,1988.
ROS. [8] H.P.KohlerandP.J.Grant,“Plasminogen-activatorinhibitor
type1andcoronaryarterydisease,”TheNewEnglandJournalof
Medicine,vol.342,no.24,pp.1792–1801,2000.
Authors’Contribution [9] L. R. Lund, B. Georg, L. S. Nielsen, M. Mayer, K. Dano,
andP.A.Andreasen,“Plasminogenactivatorinhibitortype1:
IvanaZagottaandElitsaY.Dimovacontributedequallytothis cell-specificanddifferentiation-inducedexpressionandregu-
work. lation in human cell lines, as determined by enzyme-linked