The FASEB Journal•Research Communication
Roles of transforming growth factor-␤1and OB-cadherin in porcine cardiac valve
myofibroblast differentiation
Huan Wang,*,†,‡Leslie A.Leinwand,†,‡and Kristi S.Anseth*,‡,§,1
*Department of Chemical and Biological Engineering,†Department of Molecular,Cellular
and Developmental Biology,‡BioFrontiers Institute,and§Howard Hughes Medical Institute,
University of Colorado,Boulder,Colorado,USA
ABSTRACT Calcific aortic stenosis is a common dis-ease,and some of its early causes are the activation and differentiation of residentfibroblasts to myofibroblasts in response to transforming growth factor␤1(TGF-␤1).The aim of this study was to understand how TGF-␤1and its downstream effector,OB-cadherin [cadherin11(CDH11)],regulate porcine myofibroblast phenotypes.Based on whole-genome microarrays,95 and107genes are up-and down-regulated at both the early(8h)and the late(24h)time points of TGF-␤1 treatment.Gene functions related to cell ad
hesion, skeletal system development,and extracellular matrix are up-regulated by TGF-␤1,whereas oxidation-reduc-tion and steroid metabolic process are down-regulated. Notably,one of the cell adhesion molecules,CDH11,is up-regulated byϳ2-fold through both the Smad2/3and the ERK pathways elicited by TGF-␤1.CDH11mediates cell-cell contacts in both valvularfibroblasts and myo-fibroblasts.Knockdown of CDH11by small interfering RNA increases the myofibroblast phenotype,including anϳ2-fold increase in␣-smooth muscle actin(␣-SMA) expression and stressfiber formation.In contrast, increased binding of CDH11through antibody treat-ment inhibits␣-SMA expression.This study presents gene functional changes in response to TGF-␤1at the systems level and supports an inhibitory role of CDH11in myofibroblast differentiation.—Wang,H.,Lein-wand,L.A.and Anseth,K.S.Roles of transforming growth factor-␤1and OB-cadherin in porcine cardiac valve myofibroblast differentiation.FASEB J.28, 4551–4562(2014).
Key Words:aortic valvefibrosis⅐genome-wide transcription⅐␣-smooth muscle actin
Calcific aortic stenosis(CAS)is a predominant cardiac valve disease in Western populations,affecting 2–3%of people older than65yr(1).It is characterized by gradual stiffening and calcific nodule formation in aortic valves.A number of risk factors,including old age,male sex,smoking,hypertension,and high serum lipid levels,have been identified(2).However,recent studie
s suggest that this disease is not purely a result of tissue degeneration,but an active process mediated by the resident cells,named valvular interstitial cells (VICs;ref.2).
VICs are the main cell population in valve leaflets and play important roles in both tissue homeostasis and disease progression.They are largely composed of fibroblasts that can differentiate into pathogenic myo-fibroblasts or osteoblast-like cells(3–5).VICs produce the major extracellular matrix(ECM)components of the valves,including collagen,fibronectin,elastin,and proteoglycans,and secrete ECM-remodeling enzymes, such as matrix metalloproteases(MMPs)and tissue inhibitors of MMPs(6,7).During valve development, VICs actively participate in tissue formation by secreting ECM-related proteins(8,9).On tissue damage,these cells are activated to a myofibroblast phenotype,which is characterized by increased deposition of collagen and increased contractility,mediated by␣-smooth mus-cle actin(␣-SMA)-positive stressfibers(10).Myofibro-blasts are important regulators of tissue repair in mul-
1Correspondence:Howard Hughes Medical Institute,3415 Colorado Ave.,JSCBB A315,596UCB Boulder,CO80303, USA.E-mail:kristi.anseth@colorado.edu
doi:10.1096/fj.14-254623
This article includes supplemental data.Please visit to obtain this information.
Abbreviations:␣-SMA,␣-smooth muscle actin;ANOVA,
analysis of variance;BSA,bovine serum albumin;CAS,calcific
aortic stenosis;CDH2,cadherin2(N-cadherin);CDH11,
cadherin11(OB-cadherin);Col1A1,collagen1A1;CTGF,
connective tissue growth factor;DAVID,Database for Anno-
tation,Visualization and Integrated Discovery;DMSO,di-
methyl sulfoxide;ECM,extracellular matrix;ERK,extracellu-
lar signal-regulated kinase;FBS,fetal bovine serum;FN1,
fibronectin1;GAPDH,glyceraldehyde-3-phosphate dehydro-
genase;GCRMA,GeneChip robust multiarray average;HRP,
horseradish peroxidase;IPA,Ingenuity Pathway Analysis;
MAPK,mitogen-activated protein kinase;MEK1/2,mitogen-
activated protein kinase kinase1/2;MMP,matrix metallopro-
tease;NT-siRNA,nontargeting small interfering RNA;OB-
siRNA,OB-cadherin small interfering RNA;qRT-PCR;
quantitative reverse transcriptase PCR;PBS,phosphate-buff-
ered saline;siRNA,small interfering RNA;TGF-␤1,trans-
forming growth factor␤1;TNF-␣,tumor necrosis factor␣;
TUFT1,tuftelin1;VIC,valvular interstitial cell
4551 0892-6638/14/0028-4551©FASEB
tiple mesenchymal tissues,such as skin,lung,and liver (11–13).However,if the myofibroblast phenotyp
e per-sists,excessive collagen is deposited,leading to tissue fibrosis and stiffening(14,15).Therefore,understand-ing regulation of the myofibroblast phenotypes may shed light on the mechanism of disease progression in various valvular diseases,including CAS.
Many extracellular cues regulate activation of VICs to myofibroblasts,including chemical factors,mechanical stress,and cell-cell interactions.Despite the fact that a number of chemical cues can activate myofibroblasts, transforming growth factor␤1(TGF-␤1)is one of the most common and potent chemokines to induce this phenotype in different tissues(16).It initiates the phosphorylation and translocation of Smad2/3into the nucleus,where they promote transcription of impor-tant myofibroblast genes,including␣-SMA,collagen 1A1(Col1A1),and connective tissue growth factor (CTGF;refs.17–21).TGF-␤1can also activate the noncanonical ,p38mitogen-activated pro-tein kinase(MAPK;ref.22)or phosphatidylinositol 3-kinase(23),to promote myofibroblast differentiation. In addition to chemical factors,mechanical cues,such as dynamic stretching(24,25)or elasticity(26–31),can modulate myofibroblast activation.Although much progress has been made in understanding how chemi-cal and mechanical cues regulate VIC activation,little is known about the role of cell-cell contact in myofibro-blast differentiation.
Cadherins mediate calcium-dependent cell-cell inter-action,and they play critical roles in tissue forma
tion and cell sorting during development(32).In addition to these major functions,cadherins also regulate other cellular functions,including migration,proliferation, and differentiation.During myofibroblast differentia-tion in skin,fibroblast cells had increased expression of OB-cadherin[cadherin11(CDH11)],but decreased expression of N-cadherin[cadherin2(CDH2);ref.33]. Anti-CDH11peptides,but not anti-CDH2peptides,inhib-ited skin myofibroblast contraction(33).In addition,a recent study showed that CDH11-null mice had reduced pulmonaryfibrosis induced by bleomycin(34).Collec-tively,these results support the possibility that CDH11 may regulate the pathogenic myofibroblast phenotype. Because VICs in their native matrix constantly receive many different types of signals,it is important to understand systematically how cells process these sig-nals to give the appropriate phenotypic output.Fur-ther,because TGF-␤1is a pleiotropic factor and has cell type-specific and time-dependent effects,we decided to examine the gene expression program elicited by TGF-␤1over time in porcine aortic VICs.Through microarrays,we found that TGF-␤1induced distinct gene programs in VICs after8and24h of treatment. Intriguingly,genes involved in cell adhesion were up-regulated by TGF-␤1at24h of treatment,with CDH2 and CDH11increased byϳ2-fold.Further,when CDH11was knocked down by small interfering RNA (siRNA),myofibroblast differentiation was increased,as assessed by␣-SMA-positive stressfibers;in contrast,increased CDH11engagement induced by CDH11an-tibody treatment inhibited␣-SMA expression.How-ever,the
effects of CDH11were overridden by high doses of TGF-␤1treatment.This study suggests a unique mechanism as to how CDH11,a downstream target of TGF-␤1,regulates myofibroblast differentia-tion.Whereas the overall effect of TGF-␤1is to promote myofibroblast differentiation,there may be down-stream ,CDH11)of TGF-␤1that act to inhibit or modulate the differentiation process and the signaling pathways.
MATERIALS AND METHODS
VIC cell culture
Fresh porcine hearts were obtained from Hormel Foods Corp.(Austin,MN,USA)within24h of euthanasia,and aortic valve leaflets were excised.Primary VICs were harvested from porcine aortic valve leaflets based on sequential colla-genase digestion as described previously(35).The isolated cells were cultured in growth medium[medium199,15% fetal bovine serum(FBS),50U/ml penicillin,50␮g/ml streptomycin,and0.5␮g/ml Fungizone]and expanded up to passage3(P3).P3VICs were seeded directly or after trans-fection on plastic plates at different densities in low-serum medium(medium199,1%FBS,50U/ml penicillin,50␮g/ml streptomycin,and0.5␮g/ml Fungizone].Samples were collected after culture in the low-serum medium for2d. For the small molec
ule inhibitor treatment in Fig.3C,cells were seeded at400cells/mm2and treated with SB431542 (S4317;Sigma-Aldrich,St.Louis,MO,USA)at10␮M or CI1040(S1020;Selleck Chemicals,Houston,TX,USA)at2␮M with or without porcine TGF-␤1(5ng/ml;R&D Systems; Minneapolis,MN,USA)for24h.The dimethyl sulfoxide (DMSO)vehicle was treated at0.1%as a control.
Porcine genome microarray
P3VICs were seeded at400cells/mm2in low-serum medium supplemented with1%FBS on plastic plates.All samples were collected at32h after cell seeding,and3experimental conditions were performed:cells without TGF-␤1treatment (control);cells treated with TGF-␤1(5ng/ml)for the last8 h of culture(T8);and cells treated with TGF-␤1(5ng/ml)for the last24h of culture(T24).Total RNA was extracted using an RNeasy Mini Kit(74104;Qiagen,Valencia,CA,USA).All the RNA samples had a260/280ratioՆ2.0and RNA integrity numberՆ9.2verified by an RNA bioanalyzer.Total RNA was amplified and labeled using a Gene Chip3=IVT Express Kit (901228;Affymetrix,Santa Clara,CA,USA)and hybridized to porcine genome microarrays(900624;Affymetrix).Three biological replicates were performed for each condition. Microarray data were analyzed using Spotfire(Tibco,Palo Alto,CA,USA),GeneSpring(Agilent Technologies,Santa Clara,CA,USA),the Database for Annotation,Visualization and Integrated Discovery(DAVID)functional annotation bioinfor
matics analysis(U.S.National Institute of Allergy and Infectious Diseases,Bethesda,MD,USA;david.v/),and Ingenuity Pathway Analysis(IPA;Ingenuity Systems,Inc.,Redwood,CA,USA;www.ingenuity-).The microarray data have been deposited in the U.S. National Center for Biotechnology Information(Bethesda,MD, USA)Gene Expression Omnibus(GEO;bi/ v)archives with accession number GSE48839.
4552Vol.28October2014WANG ET AL.
The FASEB Journal⅐
Real-time quantitative reverse transcriptase PCR(qRT-PCR)
For each condition,total RNA was isolated based on the RNeasy Mini Kit(74104;Qiagen)or according to the manu-facturer’s protocol for TRI Reagent(T9424;Sigma-Aldrich). The RNA concentration and integrity were measured and verified by a NanoDrop1000system(Thermo Scientific, Waltham,MA,USA).cDNA was synthesized from total RNA with SuperScript III reverse transcriptase(18080-051;Life Technologies,Grand Island,NY,USA)and random hexamer primers.Gene expression was determined by SYBR Green-based qRT-PCR using gene-specific prim
er sets(shown in Table1)and a7500real-time PCR machine(Applied Biosys-tems,Foster City,CA,USA).
Immunocytochemistry
VICs werefixed with4%paraformaldehyde(15min at25°C), permeabilized in0.1%Triton X-100,and blocked with3% bovine serum albumin(BSA).Mouse anti-␣-SMA antibody (ab7817;Abcam,Cambridge,MA,USA)was diluted at1:100 in phosphate-buffered saline(PBS)with1%BSA and5% normal goat serum and incubated with the samples overnight at4°C.After washes in PBS with0.05%Tween20(PBST), samples were labeled with goat anti-mouse Alexa Fluor488 secondary antibody(A-11001;Life Technologies).For dou-ble-staining of CDH11with either␣-SMA or F-actin,samples were not permeabilized and were stained with CDH11(clone 15F7,a generous gift from Dr.Michael Brenner,Brigham and Women’s Hospital,Boston,MA,USA)first.Samples were thenfixed with4%paraformaldehyde(10min at25°C), permeabilized,and blocked for staining with␣-SMA or phal-loidin-tetramethylrhodamine B isothiocyanate(P1951;Sig-ma-Aldrich).Samples were mounted and subsequently im-aged on an LSM710laser scanning microscope(Carl Zeiss GmbH,Jena,Germany)with aϫ20objective.Images of differentfluorescence channels were compiled with Zen (Carl Zeiss)or ImageJ(U.S.National Institutes
of Health, Bethesda,MA,USA)software.
Western blot
VICs were scraped into the radioimmunoprecipitation assay buffer(20-188;Upstate,Temecula,CA,USA)with protease inhibitor cocktail(539134;EMD Millipore,Billerica,MA,USA)and phosphatase inhibitors(04906845001;Roche,In-dianapolis,IN,USA).Cell lysates were subsequently rotated at 4°C for30min and centrifuged at12,000rpm for15min.The supernatant was saved for SDS-PAGE.The protein concentra-tions of the samples were measured by a microBCA kit (23235;Thermo Scientific).The same amount of total pro-tein for each sample was loaded into10%TGX mini gels (456–1033;Bio-Rad Laboratories,Hercules,CA,USA)and separated by electrophoresis for1h at200V.The proteins were subsequently transferred onto a nitrocellulose mem-brane using tank transfer at100V for1h.The nitrocellulose membrane was blocked with PBST with5%nonfat milk for1 h at room temperature and then was incubated with primary antibodies overnight at4°C.After washes with PBST,second-ary antibodies conjugated with horseradish peroxidase(HRP) were applied for1h at room temperature.Protein bands were visualized by applying a chemiluminescent substrate of HRP (34075;Thermo Scientific)and exposing the membrane to X-rayfilms.Protein abundance was subseq
uently quantified by ImageJ.The following antibodies were used in this study: glyceraldehyde-3-phosphate dehydrogenase(GAPDH;2118; Cell Signaling Technology,Danvers,MA,USA),␣-SMA (ab7817;Abcam),phospho-Smad2(Ser465/Ser467,3108; Cell Signaling Technology),phospho-extracellular signal-reg-ulated kinase1/2(ERK1/2;Thr202/Tyr204,9106;Cell Sig-naling Technology),ERK1/2(9102;Cell Signaling Technol-ogy),pan-cadherin(4068;Cell Signaling Technology),and CDH11(clone3H10,a generous gift from Dr.Michael Brenner,Brigham and Women’s Hospital).For each condi-tion,Ն3biological replicates were examined by Western blotting.
siRNA transfection
P3VICs were transfected with siRNAs with the U23protocol in Amaxa Nucleofector(Lonza,Basel,Switzerland).In brief, cells were prepared in a single cell suspension at2–3ϫ106 cells/100␮l of the transfection buffer provided by the kit (VPI-1002;Lonza).siRNA against CDH2(sequence CUGU-GUCUGUCACAGUUAUUU)and siRNA against CDH11(se-quence CAGACUUGGACUAUGACUAUU),both of which were customer-designed based on porcine mRNA sequences (Dharmacon RNA Technologies,Lafayette,CO,USA)or nontargeting siRNA(D-001210-02-20;Dharmacon),were used at2␮M in the transfection buffer.After transfection,
TABLE1.Gene primer sequences for qRT-PCR
Gene Forward primer sequence Reverse primer sequence
18S GCCGCTAGAGGTGAAATTCTT CTTTCGCTCTGGTCCGTCTT
Human␣-SMA GCAAACAGGAATACGATGAAGCC AACACATAGGTAACGAGTCAGAGC
Human Col1A1GGGCAAGACAGTGATTGAATACA GGATGGAGGGAGTTTACAGGAA
CTGF CTGGTCCAGACCACAGAGTGG GCAGAAAGCGTTGTCATTGG
FN1GGCATTGATGAAGAACCCTTG GCCTCCACTATGATGTTGTAGGTG
MMP1GGCATCCAGGCCATCTATG CACTTGTGGGGTTTGTGGG
OB-CDH GGGTCCCTGAGCTCCTTAGA CGAGGTCCCCAGTTCTGTAG
TUFT1GGAGAAGATCCACCACTTGGA GTGTCCTTTGACTGGATCACAG
RASAL2ACACGAGCTTTCGGCTTCC GGCTCAGCAAGGATTCATGTG
CHSY1CGCCCAGAAATACCTGCAGA CGACGTGTCTGAACCCTCACTAG
ITGA5CCAAAGGGAACCTCACCTACG ACCTGTTCCCCTGAGAAGTTGTAG
PDK4CCACATTGGCAGCATTGAC ACAGAGCATCTTGGAACACTCAA
MYLIP CTGTGCTGTGAGGGCGAGA CACTCGCGACCTGCAAACG
ADAMTS1GTGATCCCAGTAGAAGCTGCTC CATTGCTCGGCATCATCATG
ACSS3CTGACTTAGGCTGGGTTGTCG CGGAAATAAGCACCAGCATCC
ID2CGCTGACCACCCTAAATACG GAGCGCTTTGCTGTCACTTG
Smad7CCTGTGTGCTGAGCTCTGC GAATCCTTCTTGGTGGGAAG
4553 OB-CADHERIN INHIBITS MYOFIBROBLAST DIFFERENTIATION
cells were recovered in growth medium with15%FBS for15 min in the incubator and seeded at350–400cells/mm2 (accounting for50%survival of cells after the transfection) on plastic plates in low serum medium(1%FBS)for2d.In Fig.5,transfected cells were treated with or without porcine TGF-␤1(5ng/ml)for48h before sample collection.
Antibody treatment
Untreated plastic plates were incubated with the antibody solution containing human IgG1Fc(110-HG;R&D Systems) or human CDH11Fc(1790-CA;R&D Systems)antibodies at 10␮g/ml overnight at4°C.Before cell seeding,the antibody solution was aspirated off,and the plates were washed once with PBS.Cells were seeded at300–400cells/mm2and collected48h after treatment.
Statistics
A Student’s t test was used to compare data sets with2 conditions and a1-way analysis of variance(ANOVA)with a Newman-Keuls post hoc test was used to compare data sets with Ͼ2conditions.A value of PϽ0.05was considered statistically significant.
RESULTS
TGF-␤1-induced common and distinct gene expression changes globally in VICs after8and24h Gene expression in VICs treated with TGF-␤1over time was measured by porcine genome arrays.Data werefiltered(coefficient of variation,Ͻ50%),nor-malized based on GeneChip robust multiarray aver-age(GCRMA),and analyzed based on hierarchical clustering,as shown in Fig.1A.The
different experi-mental groups spontaneously clustered based on their specific treatment.Two major types of regulation were observed:genes up-regulated or down-regulated by TGF-␤1over time.For the differentially regulated genes after24h of TGF-␤1treatment(fold changeՆ2, PϽ0.05),DAVID functional annotation analysis showed that gene functions related to cell adhesion,skeletal system development,and ECM were significantly up-regulated,whereas those involved in oxidation-reduc-tion and steroid metabolic process were significantly down-regulated(Fig.1B).The lists of genes in each functional category in Fig.1B are summarized in Sup-plemental Tables S1ϪS5.Different numbers of genes were differentially regulated at the2time points of TGF-␤1treatment(Fig.1C).Specifically,127and178 genes were significantly up-regulated by TGF-␤1at8 and24h,respectively,with95overlapping genes.In contrast,154and225genes were significantly down-regulated by TGF-␤1at8and24h,respectively,with 107genes in common.Microarray data were consis-tently validated by qRT-PCR based on5genes up-regulated by TGF-␤1[tuftelin1(TUFT1)RASAL2,fibronectin1(FN1),CHSY1,and ITGA5]and5genes down-regulated by TGF-␤1(PDK4,MYLIP,ADAMTS1, ACSS3,and ID2;Supplemental Fig.S1).TGF-␤1signaling was significantly up-regulated based on IPA
Some known TGF-␤1-responsive genes,including JUNB,TGFB1,KLF10,and SMURF2,were increas
ed in a time-dependent manner based on the microarray data (Fig.2A).In addition,IPA in Fig.2B showed that TGF-␤1signaling was activated with more downstream targets being up-regulated over time,validating the microarray results.
TGF-␤1treatment up-regulated the expression of CDH11in VICs through both the Smad2/3and the ERK signaling pathways
From the microarrays,4types of cadherins(CDH2, CDH5,CDH11,and CDH13)were abundantly expressed in VICs(Fig.3A).Among these,CDH2and CDH11were up-regulatedϳ2-fold after24h of TGF-␤1treatment based on the microarrays(Fig.3A).Up-regulation of CDH11was then validated by both qRT-PCR and West-ern blot.With TGF-␤1treatment for24h,Col1A1, CTGF,FN1,MMP1,and CDH11were all up-regulated3-to7-fold(Fig.3B).At the protein level,TGF-␤1in-creased expression of CDH11relative to the internal control,GAPDH(Fig.3C).However,the effect of TGF-␤1in inducing CDH11expression was blocked by inhibiting either the Smad2/3pathway via SB431542or the ERK pathway via CI1040(Fig.3C).Furthermore, the existing microarray data also showed that CDH11 mRNA was up-regulated with the myofibroblast pheno-type in a number of disease models,such as pulmonary fibrosis and liverfibrosis(Supplemental Fig.S2). CDH11expression was increased with cell density and localized at cell-cell contacts in VICs
To confirm the role of CDH11in mediating VIC-VIC adhesion,we showed that CDH11expression was in-creased with cell density in VICs(Fig.4A).Only when VICs were plated at high densities(Ͼ200cells/mm2), which enable cell-cell contacts,did they have significant CDH11expression.Based on immunostaining images, CDH11was increased with cell density and localized to both cell-cell contacts(arrowheads)and leading edges of the cells(Fig.4B,arrows).
Knocking down CDH11promoted␣-SMA protein expression and stressfiber formation in VICs,which was overridden by TGF-␤1
VICs were transfected with siRNAs designed against porcine OB-cadherin[OB-cadherin siRNA(OB-siRNA)] or nontargeting siRNA(NT-siRNA).OB-siRNA trans-fection caused a significant reduction in CDH11at both the mRNA level(Fig.5A)and the protein level (Fig.5B),compared with NT-siRNA transfection.Mean-while,knocking down CDH11in VICs increased expres-sion of␣-SMA protein,which was overridden by TGF-␤1treatment at5ng/ml(Fig.5B).Further,cells
4554Vol.28October2014WANG ET AL.
The FASEB Journal⅐
treated with OB-siRNA had reduced CDH11staining,but an increased percentage of myofibroblasts (marked by ␣-SMA stress fibers),compared with those treated with NT-siRNA (Fig.5C ).However,TGF-␤1treatment induced similar levels of myofibroblast activation in NT-siRNA and OB-siRNA conditions (Fig.5C ).In addi-tion,CDH11knockdown did not affect the transcrip-tional response to TGF-␤1of some genes (including CTGF ,Col1A1,and Smad7;Supplemental Fig.S4).To rule out the possibility that CDH2may compensate for the function of CDH11,we showed that when CDH11was knocked down,CDH2expression was not induced in VICs (Supplemental Fig.S5A).In addition,when we knocked down both CDH2and CDH11,VICs had increased ␣-SMA protein expression and increased numbers of myofibroblasts with ␣-SMA stress fibers as well (Supplemental Fig.S5).
Increased CDH11engagement induced by antibody treatment inhibited ␣-SMA expression
The effect of increasing CDH11engagement on valvular myofibroblast differentiation was examined by coating plates with CDH11-mimicking antibodies,which has been shown previously to increase CDH11binding (36).When CDH11antibodies were pre-sented to VICs for 48h,␣-SMA mRNA and protein expression was inhibited,compared with the control (Fig.6).
modulate
DISCUSSION
Myofibroblasts are critical during wound healing,serv-ing to deposit more ECM and to contract the wound
A B
C
329583
Up >2 fold with TGF-β1 treatment
47107
118
8 hr
24 hr
Down >2 fold with TGF-β1 treatment
8 hr
24 hr
-90
+24
Figure 1.Global gene expression of porcine VICs in response to TGF-␤1over time.VICs cultured on plastic plates were treated with TGF-␤1at 5ng/ml for 8h (T8)and 24h (T24).Total RNA from 3biological replicates of each condition was collected for microarray analysis.A )Hierarchical clustering of GCRMA-normalized and ANOVA-analyzed (P Ͻ0.05)microarray data showed that the control (Con)condition and 2different durations of TGF-␤1-treated conditions segregated into 3clusters.B )DAVID functional annotation analysis of genes differentially regulated by TGF-␤1at 24h (with fold change Ն2and P Ͻ0.05)showed that genes associated with cell adhesion,skeletal system development,and ECM were up-regulated by TGF-␤1,whereas oxidation-reduction and steroid meta
bolic process were down-regulated.C )Venn diagrams showing the number of gene probes that were differentially regulated by TGF-␤1treatment at different time points with fold change of Ն2and P Ͻ0.05.At both time points,95gene probes were up-regulated,and 107gene probes were down-regulated.
4555
OB-CADHERIN INHIBITS MYOFIBROBLAST DIFFERENTIATION

版权声明：本站内容均来自互联网，仅供演示用，请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198，我们将在24小时内删除。