Impaired Splicing of Fibronectin Is Associated With...

Impaired Splicing of Fibronectin Is Associated WithThoracic Aortic Aneurysm Formation in Patients With

Bicuspid Aortic ValveValentina Paloschi, Sanela Kurtovic, Lasse Folkersen, Delphine Gomez, Dick Wågsater, Joy Roy,

Johan Petrini, Maria J. Eriksson, Kenneth Caidahl, Anders Hamsten, Jan Liska, Jean-Baptiste Michel,Anders Franco-Cereceda, Per Eriksson

Objective—Thoracic aortic aneurysm is a common complication in patients with bicuspid aortic valve (BAV).Alternatively spliced extra domain A (EDA) of fibronectin (FN) has an essential role in tissue repair. Here we analyzethe expression of FN spliceforms in dilated and nondilated ascending aorta of tricuspid aortic valve (TAV) andBAV patients.

Methods and Results—The mRNA expression was analyzed in the ascending aorta by Affymetrix Exon arrays in patientswith TAV (n�40) and BAV (n�69). EDA and extra domain B (EDB) expression was increased in dilated aorta fromTAV patients compared with nondilated aorta (P�0.001 and P�0.05, respectively). In contrast, EDA expression wasnot increased in dilated aorta from BAV patients (P�0.25), whereas EDB expression was upregulated (P�0.01). Theexpression of EDA correlated with maximum aortic diameter in TAV (��0.58) but not in BAV (��0.15) patients.Protein analyses of EDA-FN showed concordant results. Transforming growth factor-� treatment influenced the splicingof FN and enhanced the formation of EDA-containing FN in cultured medial cells from TAV patients but not in cellsderived from BAV patients. Gene set enrichment analysis together with multivariate and univariate data analyses ofmRNA expression suggested that differences in the transforming growth factor-� signaling pathway may explain theimpaired EDA inclusion in BAV patients.

Conclusion—Decreased EDA expression may contribute to increased aneurysm susceptibility of BAV patients. (Arterioscler ThrombVasc Biol. 2011;31:691-697.)

Key Words: aneurysms � bicuspid aortic valve � fibronectin � splicing � thoracic aortic aneurysm

Thoracic aortic aneurysms (TAAs) are characterized byloss of smooth muscle cells (SMCs) and degeneration of

extracellular matrix, which together can lead to dilatation andeventually to rupture of the arterial wall. There are multipleetiologies of TAA: monogenic syndromes predisposing indi-viduals to TAA, aneurysm formation associated with bicuspidaortic valve (BAV), and idiopathic causes of TAA. Histolog-ical observations in TAA show similar phenotypes, irrespec-tive of etiology: ie, extracellular matrix breakdown, SMCdisappearance, and areas of mucoid degeneration.1 Thepathogenesis of aneurysm formation in the monogenic syn-dromes has been extensively studied, whereas the molecularmechanisms of the other forms, constituting the majority ofTAAs, remain largely unknown. A common feature of themonogenic syndromes appears to be an impairment of thetransforming growth factor-� (TGF-�) pathway. The Marfansyndrome is caused by mutations in the fibrillin-1 (FBN1)

gene that has been suggested to influence the bioavailabilityof active TGF-�. It was recently reported that Smad2 activa-tion and an increase in stored TGF-�1 are concomitantlyobserved in the media of all types of ascending aorticaneurysms.2

The prevalence of aortic dilatation in patients with BAVwithout significant valve dysfunction has been estimated tobe as high as 50% to 70%.3 BAV is a congenital disorderpresent in 1% to 2% of the population, which makes TAAassociated with BAV very common. BAV is the result ofabnormal aortic cusp formation during valvulogenesis. Theincreased susceptibility to aneurysm formation associatedwith BAV may be independent of hemodynamically signifi-cant valve dysfunction.4 Aortic valve replacement does notprevent future dilatation in BAV patients whereas the aorta ofpatients who have had tricuspid aortic valve (TAV) replace-ment does not dilate.5 This indicates that inherent properties

Received on: June 24, 2010; final version accepted on: November 29, 2010.From the Atherosclerosis Research Unit, Center for Molecular Medicine, Department of Medicine, Karolinska Institutet, Stockholm, Sweden (V.P.,

S.K., L.F., D.W., A.H., P.E.); INSERM, U698, Paris F-75018, France (D.G., J.-B.M.); Vascular Surgery Section, Department of Molecular Medicine andSurgery (J.R.), Clinical Physiology (J.P., M.J.E., K.C.), Cardiothoracic Surgery Unit, Department of Molecular Medicine and Surgery (J.L., A.F.-C.),Karolinska Institutet, Stockholm, Sweden.

Correspondence to Per Eriksson, CMM, L8:03, Karolinska University Hospital, Solna, S-171 76 Stockholm, Sweden. E-mail [email protected]© 2011 American Heart Association, Inc.

Arterioscler Thromb Vasc Biol is available at http://atvb.ahajournals.org DOI: 10.1161/ATVBAHA.110.218461

691

by guest on April 22, 2018

http://atvb.ahajournals.org/D

ownloaded from

by guest on A

pril 22, 2018http://atvb.ahajournals.org/

Dow

nloaded from



ownloaded from

by guest on A


Dow

nloaded from



ownloaded from

by guest on A


Dow

nloaded from



ownloaded from

by guest on A


Dow

nloaded from



ownloaded from

by guest on A


Dow

nloaded from



ownloaded from

http://atvb.ahajournals.org/











of the arterial wall of patients with BAV may lead to anincreased aneurysm susceptibility, whereas aneurysm forma-tion is more dependent on flow conditions in patients withTAV. However, there are also data suggesting a hemodynam-ic origin of BAV aortopathy.6

In addition to fibrillin-1, other extracellular matrix compo-nents with nonstructural functions may be involved in aneu-rysm formation. Fibronectin (FN) is a glycoprotein influenc-ing cell migration, differentiation, and growth and is acommon extracellular matrix component in the vessel wall.The FN protein expression is increased in aneurysmal aorta ofBAV and TAV patients compared with control aortas.7

Several variants of FN exist because of alternative splicing ofa single gene.8 The soluble form of FN present in plasma isdimeric and lacks the alternatively spliced extra domain A(EDA) (exon 33) and extra domain B (EDB) (exon 25). Intissue, multimeric forms containing EDA or EDB are presentin addition to the soluble form. The sequences of EDA andEDB show low homology but have been highly conservedduring evolution, suggesting functional but different roles ofthese domains. The exact functions of EDA and EDB in thevessel wall are not yet known, although many activities havebeen ascribed to these domains. For example, EDA-containing FN (EDA-FN) is associated with an increasednumber of SMCs, more collagen, and less fat in humanatherosclerosis.9 In mice, EDA-FN was shown to have anessential role in pulmonary fibrosis10 and in skin woundhealing.11 Thus, EDA-FN appears to be important for cellmigration and proliferation, mechanisms that are central fornormal functions such as tissue repair and maintenance oftissue integrity. Importantly, both EDA and EDB appear tohave critical roles in vascular morphogenesis duringembryogenesis.12

In the present study, we analyzed the expression of EDAand EDB in dilated and nondilated ascending aorta of TAVand BAV patients. Our data indicate that there is impairmentin the TGF-� mediated splicing mechanisms of the FNtranscript in BAV patients and that the levels of EDA arelower in BAV than in TAV patients. Decreased expression ofthe EDA domain may contribute to the increased susceptibil-ity to TAA formation of BAV patients.

MethodsExpanded methods are given in the Supplemental Material, availableonline at http://atvb.ahajournals.org.

PatientsA total of 109 patients (80 males, 29 females; 61�12 years old[�SD]) were referred for elective aortic valve surgery or ascendingthoracic aortic surgery and were included in the study (Table 1). ATAV was present in 40 patients, and 69 patients had a bicuspid valve.The ascending aorta was dilated (maximal diameter measured bytransesophageal echocardiography, �45 mm) in 45 BAV and 23TAV patients (50�3 and 54�8 mm�SD, respectively) and nondi-lated (maximal diameter measured by transesophageal echocardiog-raphy, �40 mm) in 24 BAV and 17 TAV patients (36�3 and34�4 mm�SD, respectively). The chosen definition of aorticdilatation with a cutoff at a 4.5-cm diameter is based on recommen-dation of when to perform aortic surgery.13 None of the patients hadsignificant coronary artery disease according to coronary angiogra-phy. Aortic biopsies were taken from the anterior (convex) part ofthe aorta, ie, the site of aortotomy a few centimeters above the aortic

valve. This study was approved by the Ethics Committee at theKarolinska Institutet, and patients were included after written andsigned informed consent was obtained.

Gene ArraysAffymetrix GeneChip Human Exon 1.0 ST arrays and protocolswere used.

Cell CulturePrimary cell cultures were isolated from human aortic tissue bycollagenase and elastase digestion (see Figure 3A and 3B) or by celloutgrowth (see Figure 3C, 3D, and 3E). Normal SMCs were obtainedfrom transplant patients. Passages 3 to 6 were used for experiments.

Principal Component AnalysisPrincipal component (PC) analysis was performed on AffymetrixHuman Exon 1.0 ST Array metaprobe set level data that had beenpreprocessed using robust multichip average (RMA) normalizationfollowed by unit variance scaling and mean centering beforeanalysis.

ResultsInclusion of FN EDA Exon Is Associated WithAneurysm Formation in TAV but NotBAV PatientsThe mRNA expressions of specific FN isoforms were ana-lyzed in dilated and nondilated aortas. Gene expression wasmeasured separately in the medial and adventitial layers ofthe vessel wall. PC analysis was applied to the AffymetrixExon array mRNA data, including 109 patients and 10 888genes. As can be interpreted from the score plot (Supplemen-tal Figure IA), the 2 different tissue phenotypes, nondilatedaorta (black) and dilated aortic samples (red), separate in thePC1 to PC3 plane.

FN mRNA expression in the vessel wall of TAV patientswas initially measured by Affymetrix Exon 1.0 ST arrayanalyses, an array platform that measures the expression of allindividual exons. The mRNA expressions of EDA (exon 33)and EDB (exon 25) were increased in dilated aorta (n�23)compared with nondilated aortic samples (n�17) from TAVpatients (Figure 1A and 1B, P�0.001 and P�0.05, respec-tively). In contrast to patients with TAV, the expression ofFN EDA was not increased in the media of dilated aorta(n�45) compared with nondilated aorta (n�24) of BAVpatients (Figure 1A, P�0.25), whereas EDB expression wasupregulated in dilated aorta from BAV patients (Figure 1B,P�0.01). Because the observed differences could potentially

Table 1. Patient Demographics

BAV TAV

CharacteristicsNondilated(�40 mm)

Dilated(�45 mm)

Nondilated(�40 mm)

Dilated(�45 mm)

n 24 45 17 23

Gender (no. of females) 5 10 5 9

Age, years (mean�SD) 56 � 10 61 � 12 70 � 11 61 � 15

Aortic diameter, (mm�SD) 36 � 3 50 � 3 34 � 4 54 � 8

Aortic insufficiency, n 9 11 5 19

Aortic stenosis, n 14 31 12 1

Normal valve, n 1 3 0 3

692 Arterioscler Thromb Vasc Biol March 2011



ownloaded from


be an effect of hemodynamic changes, we also restricted theanalysis to patients having aortic insufficiency. In agreementwith the total patient material, the EDA expression washigher in dilated aorta of TAV patients (n�19) than BAVpatients (n�11), P�0.01 (Figure 1C). To analyze the FNsplice variants with an independent method, quantitativereal-time polymerase chain reaction (PCR) was used tomeasure the expression of total FN (primers spanning exon41 and 42), EDA-containing FN (EDA-FN; primers spanningexons 32 and 33) and EDB-containing FN (EDB-FN; primersspanning exons 24 and 25). EDA-FN expression was higherin media of dilated aorta of TAV patients compared withdilated aorta of BAV patients (Figure 1D, P�0.05). Further-more, the EDA-FN/total FN ratio was higher in dilated TAVcompared with dilated BAV patients (Figure 1E, P�0.05).Similar results were obtained when analyzing the mRNA

expression using quantitative real-time PCR in media andadventitia of dilated (n�11 and 19, respectively) and nondi-lated (n�6 and 17, respectively) aortas of TAV patients(Supplemental Figure II). Reverse transcription–PCR analy-ses of the FN transcripts showed that EDA-free FN was thepredominant mRNA transcript in RNA samples extractedfrom the aortic wall (Figure 1F). EDA containing FN tran-script could be detected mainly in dilated TAV samples(Figure 1F).

In accordance with the mRNA expression results, Westernblot showed an increased protein expression of EDA-FN indilated aortas of TAV patients compared with nondilatedaortas, whereas EDA-FN was not upregulated in BAVpatients (Figure 1G and 1H).

Analyses of protein expression of EDA-FN using immu-nohistochemistry confirmed the mRNA results. As seen in

Figure 1. Expression of FN spliceformsin vascular tissue from nondilated(�40 mm diameter) and dilated(�45 mm diameter) ascending aorta ofBAV and TAV patients. Shown are EDA(A) and EDB (B) expression in allpatients. C, Analysis of EDA expressionrestricted to dilated TAV and BAVpatients having aortic insufficiency. Geneexpression in A to C was analyzed usingAffymetrix Exon 1.0 ST arrays. Log2-transformed mRNA expression is shown.D and E, EDA-FN expression and ratiobetween EDA-FN and total FN expres-sion in dilated medial aortic vessel wallof BAV and TAV patients analyzed byquantitative real-time PCR. F, Reversetranscription–PCR of exons 32 to 34 ofFN in nondilated TAV. G, Western blot ofprotein lysates of nondilated and dilatedmedial vessel wall from TAV and BAVpatients. H, Summary of Western blotsof EDA-FN protein expression normal-ized for �-actin expression in 3 nondi-lated and 5 dilated TAV and 3 nondilatedand 5 dilated BAV media vessel wallbiopsies.

Paloschi et al Fibronectin and TAA 693



ownloaded from


Figure 2, staining of EDA was mainly detected in the dilatedsamples of TAV patients, where it appeared as a fuzzystaining close to SMCs. Antibodies against total FN gave, inaddition to the fuzzy staining, a more distinct staining close tothe cells. Supplemental Figure III show the correspondingstaining of smooth muscle-� actin.

Of note, the expression of EDA showed a correlation withmaximum aortic diameter in TAV patients (��0.58,P�0.001) but not in BAV patients (��0.08, P�0.53).

TGF-� Influences the Expression of EDAand EDBThe mRNA expression of EDA and EDB was analyzed incultured medial SMCs isolated from normal aorta (n�7), aswell as from the dilated wall of the ascending aorta from TAV(n�6) and BAV (n�6) patients. There was a significantreduction in the expression of FN and its different spliceforms in cells from BAV patients compared with cells fromdilated aorta of TAV patients and control aortas (Figure 3Aand 3B).

Similar to what has been shown previously in fibroblasts,14

incubation of aortic SMCs (American Type Culture Collec-tion PCS-100-012) with 1 and 10 ng/mL of TGF-� for 24hours enhanced the expression of EDA-FN and EDB-FN in adose-dependent manner (Supplemental Figure IVA and IVB).Also, the expression of SM22, a marker of SMC phenotype,showed a similar increase after incubation with TGF-�(Supplemental Figure IVC).

Isolated medial cells from TAV and BAV patients werethereafter induced with 5 to 20 ng/mL of TGF-� for 6 hoursto identify potential differences in splicing of FN. As shownin Figure 3C and 3D, TGF-� treatment enhanced the forma-tion of EDA-containing FN in cultured medial cells fromTAV patients but not in cells derived from BAV patients. Incontrast to RNA isolated directly from tissue (Figure 1F), theEDA-containing FN was the predominant transcript in iso-lated cells held in culture (Figure 3C).

Following previous lines of evidence from other forms ofaortic aneurysm, we tested the hypothesis that TGF-� path-way genes are involved in differences between TAV andBAV patients. The mRNA expression of TGF-�1 was sig-nificantly upregulated in dilated aorta of BAV and TAVpatients (P�0.05 for both). Using the generally applicablegene-set enrichment (GAGE) algorithm as described in theSupplemental Methods section, we examined whether theexpression of genes of the TGF-� pathway on averagediffered to a higher degree than what could be expected for

Figure 2. Immunostaining of EDA-FN and total FN in TAV andBAV patients. A, C, and E, Nondilated aorta (maximal diameter,�40 mm). B, D, and F, Dilated aorta (maximal diameter,�45 mm). G, SM22 staining. H, Negative control.

Figure 3. Expression analyses of FN spliceforms in isolatedhuman cells. A and B, mRNA expression of EDA and EDB incultured SMCs isolated from normal aorta (n�7) and fromdilated wall of the ascending aorta from TAV (n�6) and BAV(n�6) patients. C, Reverse transcription–PCR of exons 32 to 34of FN in medial cells isolated from TAV and BAV patients andtreated with 10 ng/mL of TGF-� for 6 hours. D and E, Quantita-tive real-time PCR analyses of EDA-FN (D) and EDB-FN (E) inmedial cells isolated from TAV and BAV patients and treatedwith 5 to 10 ng/mL of TGF-� for 6 hours. *Significant increase(P�0.05) compared with untreated cells.




ownloaded from


the same number of randomly selected genes. Differences inthe expression of the TGF-� pathway genes between TAVand BAV patients was detected in dilated samples from theintima/media and adventitia (GAGE probability values,0.0248 and 0.0141, respectively), whereas there was nosignificant difference in nondilated samples (GAGE proba-bility values, 0.398 and 0.065, respectively). Because TGF-�genes on average showed more significant differences be-tween BAV and TAV than genes in general, we conclude thatthe TGF-� pathway in particular is worth further investiga-tion. This change of TGF-� pathway genes with BAV andTAV is visualized in Figure 4A. In addition, PC analysis wasapplied to the 26 selected genes of the TGF-� pathway15 toelucidate groupings of the different tissue samples, as well ascontributions of the genes involved in the pathway to thegroupings of the tissues. The first 2 PCs are shown in Figure4B and 4C, where the score plot is composed of 109 tissuesamples belonging to both nondilated (black) and dilated(red) aortas. The dilated aortic tissues cluster together inparticular in the lower right quadrant of the score plot (Figure4B). The expression of the EDA and EDB exons clusterstogether with FN expression in the lower right quadrant of theloading plot (Figure 4C). Genes in the cluster situated in thelower right quadrant of the loading plot (Figure 4C) allcontribute to the grouping of dilated tissue samples in thelower right quadrant of the score plot (Figure 4B). Univariatecorrelations in BAV and TAV patients are shown in Table 2for the expression of TGF-� pathway-related genes and theexpression of EDA and EDB, separately. Genes such asFBN1, TGF-�1, and latent TGF-� binding protein-1(LTBP1) correlated with EDA and EDB expression in BAVand TAV patients. LTBP2, LTBP3, and LTBP4 correlatedsignificantly with EDB but not with EDA expression in bothBAV and TAV patients. The expression of EDA correlatedwith TGF-� receptor-3 (TGF-BR3) in TAV but not in BAVpatients. EDA and EDB correlated with TGF-BR1 in TAVbut not BAV. Taken together, these data suggest that there aredifferent TGF-�-mediated signaling pathways involved in theinclusion of EDA and EDB. Furthermore, the data suggestthat differences in the TGF-� signaling pathway may explainthe impaired EDA inclusion in BAV patients.

DiscussionIn the present study, we show that BAV patients have animpaired regulation of EDA-FN expression. Our data suggestthat differences in the TGF-� pathway are responsible for thedifferential EDA expression in BAV and TAV patients. Thelines of evidence linking perturbed expression of the TGF-�pathway to impaired splicing of EDA in BAV patients are asfollows: (1) GAGE (gene set enrichment analysis) demon-strated that the expression of TGF-� pathway genes differsbetween BAV and TAV patients at a higher degree than couldbe expected by chance alone in a sample set of similar size.(2) Connected to this, a volcano plot demonstrated a cleardifference in gene expression of genes belonging to theTGF-� pathway between BAV and TAV patients (Figure4A). (3) Cell culture experiments demonstrated that TGF-�can induce the formation of EDA-containing FN in medialcells isolated from TAV but not from BAV patients (Figure

Figure 4. A, Volcano plot of the TGF-� pathway in BAV andTAV patients. Fold change between the 2 groups is shown onthe x-axis, and the probability value for a t test of differencesbetween samples is shown on the y-axis. B and C, PC analysisof Affymetrix Human Exon 1.0 ST Array metaprobe set level data ofgenes belonging to the TGF-� pathway from 109 biopsies fromdilated and nondilated media of aortic vessel wall. B, The PC scoreplot shows samples from dilated (red dots) and nondilated (blackdots) vessel wall of all patients. C, The loading PC plot shows thecorresponding genes of the TGF-� pathway in all patients. FN1,EDA, and EDB are marked in green. The samples have beenrobust multichip average (RMA) normalized, as well as scaled tounit variance, and mean centered before analysis.




ownloaded from


3C and 3D). (4) PC analysis showed an association betweenthe TGF-�1 expression and EDA expression in the dilatedvessel wall (Figure 4B and 4C). (5) Univariate analysesdemonstrated that there are differences in the correlationbetween gene expression of members of the TGF-� pathwayand EDA expression (Table 2). However, we also observed asignificant correlation between TGF-�1 expression and EDAexpression in samples from both BAV and TAV patients.Although these experiments link impaired expression of theTGF-� pathway to EDA splicing, we have not identified themolecular events leading to the differential EDA expression.TGF-� signaling is extremely complex,15 and only a fewTGF-� signaling pathways are currently known. Further-more, some of the known activating pathways appear to becell or tissue specific, whereas others are active in multiplecell types and tissues.16

Two important features render impaired splicing of FN aninteresting candidate mechanism for both aneurysm suscep-tibility associated with BAV and for BAV itself. First, EDAand EDB containing FN are needed for remodeling processesin response to injury.11 Second, FN plays a major role in thedevelopment of the vasculature and valve formation.12 Aswill be discussed below, the latter may also be attributed tothe presence of EDA and EDB.

Studies in mice have demonstrated the importance of FN invascular morphogenesis. FN-null mice are embryonic lethal.

EDA and EDB double knockout mice showed a similarphenotype, although FN is expressed in similar amounts as inwild-type mice, which suggests that the phenotype of theFN-null mouse is attributed to the absence of EDA andEDB.17 Embryos lacking EDA and EDB die from severecardiovascular defects. Interestingly, the embryos containedsheets of endothelial cells, a phenotype that has been ob-served in other mouse models, such as TGF-�1 mutantmice.18 However, single deletions of either EDA or EDB areviable and fertile (for review, see19). During development,EDA and EDB are spatially and temporally differentiallyexpressed, which suggests distinct functions.20

FN and the TGF-� superfamily have also been implicatedin the development of cardiac valves during embryogenesis.Cardiac valves develop from precursor structures calledcardiac cushions that, after swelling, are invaded by valveprecursor cells formed by endothelial-mesenchymal transdif-ferentiation. It has been demonstrated that TGF-� acts as atriggering signal on endocardial cells to stimulate endotheli-al-mesenchymal transdifferentiation.21 Interestingly, the ini-tiation of endothelial-mesenchymal transdifferentiation is ac-companied by a deposition of FN at the cardiac jelly, amatrix-rich region present between the endothelium andmyocardium of the embryonic heart.22 However, the presenceand exact roles of EDA and EDB containing FN in valveformation remains largely unknown. Of note, FN�/� miceshow defects in endocardial cushion formation.23

It has been suggested that EDA-FN expression is associ-ated with dedifferentiation of SMCs.24 This is in accordancewith our reverse transcription–PCR experiments on SMCs inculture that had shifted to the synthetic/dedifferentiated phe-notype and produced mainly EDA-FN, whereas the prevalentisoform detected in the aortas was EDA free FN. In embryosof EDA and EDB double knockouts, a defect in migration of�SMC-positive cells to dorsal aorta was suggested to be partof the phenotype.17 Furthermore, EDA-FN has been shown tobe necessary for the induction of the TGF-�1-generatedmyofibroblastic phenotype in vitro.25 Differentiation andactivation of myofibroblasts is essential for wound healing,and absence of enhanced expression of EDA in BAV patientsmay be a contributing factor for increased aneurysm suscep-tibility. On the other hand, the presence of myofibroblasts hasbeen suggested to be a contributing factor in aneurysmformation through the expression of matrix-degrading pro-teases associated with myofibroblast migration and prolifer-ation into the dilated aorta.15,26 Defining whether cells in thedilated aorta are dedifferentiated SMCs or myofibroblastsderived from adventitial fibroblasts is not possible. Despitethe appearance of cellular homogeneity in situ, isolation ofmedial cells yields subsets with different morphology,growth, and expression of growth regulatory genes.27

One limitation of the study is that aortic samples fromnondilated vessel wall obtained at aortic valve surgery servedas controls. It is likely that some of the control aortas showsome initial destruction of the medial layer. For obviousreasons, dilated samples from BAV patients should prefera-bly be compared with nondilated aortas of BAV patients;however, these samples are not readily available because thepresumptive donors have nonaffected valves. In the PC

Table 2. Correlations Between the Expressions ofGenes Associated With the TGF-� Pathway and EDA andEDB Expression

BAV TAV

� P � P

Similar correlations to EDA and EDB, no difference between BAV and TAV

FBN1 vs EDA 0.52 <0.0001 0.72 <0.0001

FBN1 vs EDB 0.63 <0.0001 0.64 <0.0001

TGFB1 vs EDA 0.40 <0.001 0.53 <0.001

TGFB1 vs EDB 0.28 <0.05 0.27 0.08

LTBP1 vs EDA 0.28 0.05 0.53 <0.001

LTBP1 vs EDB 0.70 <0.0001 0.83 <0.0001

Correlations only to EDB, no difference between BAV and TAV

LTBP2 vs EDA 0.11 0.37 0.02 0.90

LTBP2 vs EDB 0.62 <0.0001 0.47 <0.01

LTBP3 vs EDA 0.23 0.06 0.22 0.16

LTBP3 vs EDB 0.45 <0.001 0.55 <0.001

LTBP4 vs EDA 0.23 0.06 0.23 0.14

LTBP4 vs EDB 0.59 <0.0001 0.60 <0.0001

Differences between BAV and TAV

TGFBR1 vs EDA �0.07 0.55 0.38 <0.05

TGFBR1 vs EDB 0.00 0.99 0.48 <0.01

TGFBR2 vs EDA 0.15 0.20 �0.15 0.34

TGFBR2 vs EDB 0.49 <0.0001 0.16 0.30

TGFBR3 vs EDA 0.00 0.99 �0.38 <0.05

TGFBR3 vs EDB 0.18 0.14 �0.13 0.39

Gene expression measured by Affymetrix Exon arrays.Bold text indicates significant correlations.




ownloaded from


analysis based on global gene expression analysis (Supple-mental Figure IA), most of the dilated samples group to-gether. However, some dilated aortic samples are positionedin the area of the nondilated samples, but a distinctiveclustering of all dilated samples with no diverging individualsis not to be expected because of the complexity of the tissuesamples.

In conclusion, our data suggest that BAV patients have animpaired mechanism for the splicing of EDA in response toaneurysm formation. However, the exact function of the EDAdomain in vascular remodeling is unknown, and whether theabsence of EDA expression in BAV patients contributes tothe increased susceptibility to aneurysm formation associatedwith BAV remains to be established. Of particular importanceis determining whether impairment of EDA splicing prevailsduring valve formation and whether this is part of themolecular mechanisms leading to BAV development. Thepresent study implies that dilatation of the ascending aortarelates to an inherent tissue weakness in BAV patients andmay not only be explained by hemodynamic causes, whichfurther substantiates the recommendation of early surgery inaortic dilatation in BAV patients.

Sources of FundingThis study was supported by the Swedish Research Council (12660,14231), the Stockholm County Council, the Swedish Heart-LungFoundation, the European Commission (FAD, Health-F2–2008-200647), and a donation by Fredrik Lundberg.

DisclosuresNone.

References1. Borges LF, Touat Z, Leclercq A, Zen AA, Jondeau G, Franc B, Philippe

M, Meilhac O, Gutierrez PS, Michel JB. Tissue diffusion and retention ofmetalloproteinases in ascending aortic aneurysms and dissections. HumPathol. 2009;40:306–313.

2. Gomez D, Al Haj Zen A, Borges LF, Philippe M, Gutierrez PS, JondeauG, Michel JB, Vranckx R. Syndromic and non-syndromic aneurysms ofthe human ascending aorta share activation of the Smad2 pathway.J Pathol. 2009;218:131–142.

3. Cecconi M, Nistri S, Quarti A, Manfrin M, Colonna PL, Molini E, PernaGP. Aortic dilatation in patients with bicuspid aortic valve. J CardiovascMed (Hagerstown). 2006;7:11–20.

4. Nkomo VT, Enriquez-Sarano M, Ammash NM, Melton LJ III, BaileyKR, Desjardins V, Horn RA, Tajik AJ. Bicuspid aortic valve associatedwith aortic dilatation: a community-based study. Arterioscler ThrombVasc Biol. 2003;23:351–356.

5. Yasuda H, Nakatani S, Stugaard M, Tsujita-Kuroda Y, Bando K,Kobayashi J, Yamagishi M, Kitakaze M, Kitamura S, Miyatake K. Failureto prevent progressive dilation of ascending aorta by aortic valvereplacement in patients with bicuspid aortic valve: comparison withtricuspid aortic valve. Circulation. 2003;108(suppl 1):II291–II294.

6. Hope MD, Hope TA, Meadows AK, Ordovas KG, Urbania TH, AlleyMT, Higgins CB. Bicuspid aortic valve: four-dimensional MR evaluationof ascending aortic systolic flow patterns. Radiology. 2010;255:53–61.

7. Della Corte A, Quarto C, Bancone C, Castaldo C, Di Meglio F, Nur-zynska D, De Santo LS, De Feo M, Scardone M, Montagnani S, CotrufoM. Spatiotemporal patterns of smooth muscle cell changes in ascendingaortic dilatation with bicuspid and tricuspid aortic valve stenosis: focus oncell-matrix signaling. J Thorac Cardiovasc Surg. 2008;135:8–18;18e1–2.

8. Schwarzbauer JE, Tamkun JW, Lemischka IR, Hynes RO. Three differentfibronectin mRNAs arise by alternative splicing within the coding region.Cell. 1983;35:421–431.

9. van Keulen JK, de Kleijn DP, Nijhuis MM, Busser E, Velema E, FijnheerR, van der Graaf Y, Moll FL, de Vries JP, Pasterkamp G. Levels of extradomain A containing fibronectin in human atherosclerotic plaques areassociated with a stable plaque phenotype. Atherosclerosis. 2007;195:e83–e91.

10. Muro AF, Moretti FA, Moore BB, Yan M, Atrasz RG, Wilke CA,Flaherty KR, Martinez FJ, Tsui JL, Sheppard D, Baralle FE, Toews GB,White ES. An essential role for fibronectin extra type III domain A inpulmonary fibrosis. Am J Respir Crit Care Med. 2008;177:638–645.

11. Muro AF, Chauhan AK, Gajovic S, Iaconcig A, Porro F, Stanta G, BaralleFE. Regulated splicing of the fibronectin EDA exon is essential for properskin wound healing and normal lifespan. J Cell Biol. 2003;162:149–160.

12. Astrof S, Hynes RO. Fibronectins in vascular morphogenesis. Angio-genesis. 2009;12:165–175.

13. Bonow RO, Carabello BA, Chatterjee K, de Leon AC Jr, Faxon DP, FreedMD, Gaasch WH, Lytle BW, Nishimura RA, O’Gara PT, O’Rourke RA,Otto CM, Shah PM, Shanewise JS. 2008 focused update incorporated intothe ACC/AHA 2006 guidelines for the management of patients withvalvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (WritingCommittee to revise the 1998 guidelines for the management of patientswith valvular heart disease); endorsed by the Society of CardiovascularAnesthesiologists, Society for Cardiovascular Angiography and Inter-ventions, and Society of Thoracic Surgeons. J Am Coll Cardiol. 2008;52:e1–e142.

14. Borsi L, Castellani P, Risso AM, Leprini A, Zardi L. Transforminggrowth factor-� regulates the splicing pattern of fibronectin messengerRNA precursor. FEBS Lett. 1990;261:175–178.

15. Jones JA, Spinale FG, Ikonomidis JS. Transforming growth factor-�signaling in thoracic aortic aneurysm development: a paradox in patho-genesis. J Vasc Res. 2009;46:119–137.

16. Forte A, Della Corte A, De Feo M, Cerasuolo F, Cipollaro M. Role ofmyofibroblasts in vascular remodelling: focus on restenosis and aneu-rysm. Cardiovasc Res. 2010;88:395–405.

17. Astrof S, Crowley D, Hynes RO. Multiple cardiovascular defects causedby the absence of alternatively spliced segments of fibronectin. Dev Biol.2007;311:11–24.

18. Dickson MC, Martin JS, Cousins FM, Kulkarni AB, Karlsson S, AkhurstRJ. Defective haematopoiesis and vasculogenesis in transforming growthfactor-�1 knock out mice. Development. 1995;121:1845–1854.

19. White ES, Baralle FE, Muro AF. New insights into form and function offibronectin splice variants. J Pathol. 2008;216:1–14.

20. Ffrench-Constant C, Hynes RO. Alternative splicing of fibronectin istemporally and spatially regulated in the chicken embryo. Development.1989;106:375–388.

21. Brown CB, Boyer AS, Runyan RB, Barnett JV. Antibodies to the Type IITGF-� receptor block cell activation and migration during atrioventric-ular cushion transformation in the heart. Dev Biol. 1996;174:248–257.

22. Mjaatvedt CH, Lepera RC, Markwald RR. Myocardial specificity forinitiating endothelial-mesenchymal cell transition in embryonic chickheart correlates with a particulate distribution of fibronectin. Dev Biol.1987;119:59–67.

23. George EL, Georges-Labouesse EN, Patel-King RS, Rayburn H, HynesRO. Defects in mesoderm, neural tube and vascular development inmouse embryos lacking fibronectin. Development. 1993;119:1079–1091.

24. Glukhova MA, Frid MG, Shekhonin BV, Vasilevskaya TD, Grunwald J,Saginati M, Koteliansky VE. Expression of extra domain A fibronectinsequence in vascular smooth muscle cells is phenotype dependent. J CellBiol. 1989;109:357–366.

25. Serini G, Bochaton-Piallat ML, Ropraz P, Geinoz A, Borsi L, Zardi L,Gabbiani G. The fibronectin domain ED-A is crucial for myofibroblasticphenotype induction by transforming growth factor-�1. J Cell Biol. 1998;142:873–881.

26. Vaughan MB, Howard EW, Tomasek JJ. Transforming growth factor-�1promotes the morphological and functional differentiation of the myofi-broblast. Exp Cell Res. 2000;257:180–189.

27. Zalewski A, Shi Y, Johnson AG. Diverse origin of intimal cells: smoothmuscle cells, myofibroblasts, fibroblasts, and beyond? Circ Res. 2002;91:652–655.




ownloaded from


Jean-Baptiste Michel, Anders Franco-Cereceda and Per ErikssonRoy, Johan Petrini, Maria J. Eriksson, Kenneth Caidahl, Anders Hamsten, Jan Liska,

Valentina Paloschi, Sanela Kurtovic, Lasse Folkersen, Delphine Gomez, Dick Wågsäter, JoyFormation in Patients With Bicuspid Aortic Valve

Impaired Splicing of Fibronectin Is Associated With Thoracic Aortic Aneurysm

Print ISSN: 1079-5642. Online ISSN: 1524-4636 Copyright © 2010 American Heart Association, Inc. All rights reserved.

Greenville Avenue, Dallas, TX 75231is published by the American Heart Association, 7272Arteriosclerosis, Thrombosis, and Vascular Biology

doi: 10.1161/ATVBAHA.110.2184612010;

2011;31:691-697; originally published online December 9,Arterioscler Thromb Vasc Biol.

http://atvb.ahajournals.org/content/31/3/691World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://atvb.ahajournals.org/content/suppl/2010/12/08/ATVBAHA.110.218461.DC1Data Supplement (unedited) at:

http://atvb.ahajournals.org//subscriptions/

at: is onlineArteriosclerosis, Thrombosis, and Vascular Biology Information about subscribing to Subscriptions:

http://www.lww.com/reprints

Information about reprints can be found online at: Reprints:

document. Question and AnswerPermissions and Rightspage under Services. Further information about this process is available in the

which permission is being requested is located, click Request Permissions in the middle column of the WebCopyright Clearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of theArteriosclerosis, Thrombosis, and Vascular Biologyin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:



ownloaded from

http://atvb.ahajournals.org/content/31/3/691

http://atvb.ahajournals.org/content/suppl/2010/12/08/ATVBAHA.110.218461.DC1

http://www.ahajournals.org/site/rights/

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://atvb.ahajournals.org//subscriptions/


Paloschi et al.

Supplementary Materials

Complete Method Section

Patients

A total of 109 patients (80 males, 29 females, 61±12 years old (±SD)) were referred for elective aortic

valve surgery and/or ascending thoracic aortic surgery and included in the study (Table 1). A tricuspid

aortic valve was present in 40 patients and 69 patients had a bicuspid valve. The ascending aorta was

dilated (maximal diameter measured by Transesophageal Echocardiography (TEE) > 45 mm) in 45

BAV and 23 TAV patients (50±3 and 54±8 mm±SD, respectively) and non-dilated (maximal diameter

measured by TEE < 40 mm) in 24 BAV and 17 TAV patients (36±3 and 34±4 mm±SD, respectively).

The chosen definition of aortic dilatation with a cut off at 4.5 cm diameter is based on

recommendation on when to perform aortic surgery1. None of the patients had significant coronary

artery disease according to coronary angiography. Aortic biopsies were taken from the anterior

(convex) part of the aorta i.e. the site of aortotomy a few cm above the aortic valve. This study was

approved by the Ethics Committee at the Karolinska Institutet and patients were included after

informed, written and signed consent.

All patients were operated on through a midline sternotomy using cardio-pulmonary bypass and

cardiac arrest. For isolated aortic valve replacement a biological or mechanical valve prosthesis was

used. Patients with only dilatation of the ascending aorta received a single tubular graft prosthesis

while patients with combined valve and ascending aortic pathology were operated on using a valve

prosthesis combined with a supracoronary graft mechanical composite graft or a biological full root. A

valve sparing procedure was performed in 21 patients. Aortic biopsies were taken from the anterior

(convex) part of the aorta i.e. the site of aortotomy a few cm above the aortic valve.

Real Time PCR

Media and adventitia tissue samples were homogenized with FastPrep using Lysing Matrix D tubes

(MP Biomedicals, Illkirch, France). Total RNA was isolated using Trizol (Invitrogen, Paisley,

Scotland, UK) and RNeasy Mini kit (Qiagen, Maryland, USA) as a cleanup including treatment with

DNase. RNA quality was analyzed by an Agilent 2100 bioanalyzer (Agilent Technologies Inc., Paolo

Alto, CA, USA) and quantified by a NanoDrop (NanoDrop products, Wilmington, DE, USA). RNA

from each sample (0.5 μg) was reverse transcribed with random primers and Superscript II

(Invitrogen, Carlsbad, CA, USA). Amplification of 2 μL of cDNA was performed with 1× TaqMan

Universal PCR Mastermix (Applied Biosystems, Foster City, CA, USA) on an ABI 7700 Sequence

Detector. Each sample was analyzed in duplicates. PCR amplification was evaluated against a standard

curve. The following Assays on Demand Kits (Applied Biosystems, Foster City, CA, USA) were used:

EDA+, Hs01549958; full FN, Hs00415006; EDA-, Hs01565276; EDB+, Hs 01565270; SM22,

2

Hs00199489. Ribosomal phosphoprotein large P0 (RPLP0, Hs99999902) served as an RNA loading

control. Thermal protocol: 50° C for 2 minutes, 95º C for 10 minutes, 95º C for 15 seconds - 60º C for

1 minute (45 repeats).

RT-PCR

RT-PCR using EDA spanning exons was performed on cDNA isolated from tissues (medial layer of

BAV and TAV patients, with and without dilatation) and smooth muscle cells isolated from biopsies,

cultured for 6 passages and treated with TGFβ. The primers sequences were: Forward 5’-

CACCACTCCCAAAAATGGAC-3’ and Reverse 5’-CTGAGCTGGTCTGCTTGTCA-3’. The

thermal conditions were the following: 95º for 2 minutes, 95º C for 30 seconds - 52° C for 1 minute -

72° C for 1minute (repeated for 20/35 cycles), 72° C for 2 minutes.

Gene arrays

The RNA samples were hybridized and scanned at the Karolinska Institute microarray core facility.

Affymetrix GeneChip® Human Exon 1.0 ST arrays and protocols were used. For probe set and meta

probe set level investigations, i.e. the genome-wide and regional investigations, cel files were pre-

processed and log2 transformed using Robust Multichip Average (RMA) normalization as

implemented in the Affymetrix Power Tools 1.10.2 package apt-probeset-summarize. This

normalization includes a step in which the distribution of gene expression levels on individual arrays

are standardized, essentially normalizing the expression levels to the overall mRNA levels. All

investigations were done on the core set of meta probes provided by Affymetrix. Gene set enrichment

analysis was performed using the GAGE package for R2 comparing TAV and BAV patients. This

algorithm tests if a particular group of genes show more differential expression than the amount that

could be expected in randomly selected group of genes of the same size. The algorithm settings were

group-wise comparison and inclusion of both up and down regulated genes. The gene set investigated

was defined as the meta probe sets for genes belonging to the TGFβ pathway3.

Protein preparation

Tissues were obtained from human biopsies (aortic media) from TAV and BAV patients. Tissues were

freshly dissected, placed in Matrix D tubes (MP Biomedicals, Illkirch, France) and homogenized in

modified RIPA buffer supplemented with 1x protease inhibitor cocktail (Roche Diagnostics) and 1mM

of phenylmethylsulphonyl fluoride (PMSF) to minimize proteolysis. Protein concentration was

analyzed using the Bradford method.

Western Blotting

35-50 µg of the protein lysates were resuspended in running buffer (BioRad) and subjected to sodium

dodecylsulfate–polyacrylamide gel electrophoresis with 4-12% Bis-Tris gels (Invitrogen NuPAGE).

3

Western transfer to polyvinylidene difluoride membranes (Amersham Pharmacia Biotech, Little

Chalfont, UK). Primary monoclonal antibody against EDA+ fibronectin (clone FN-3E2, 1:2000,

SIGMA, St Louis, MO, USA). β-actin (clone AC-15, 1:500, SIGMA, St Louis, MO, USA) was used

as loading control.

Immunohistochemistry

Paraffin embedded sections were used treated with DIVA solution (Biocare Medical, Concord, CA,

USA). Endogenous peroxidase activity was quenched with 3% hydrogen peroxide for 5 min. Primary

antibodies: mouse anti EDA+ fibronectin (Sigma, 1:200), rabbit anti fibronectin (Sigma 1:500), rabbit

anti smooth muscle alpha actin (Sigma 1:3000) or rabbit anti SM22 (Abcam 1:200). Secondary

biotinylated anti-rabbit IgG or anti-mouse IgG (Vector, Peterborough, UK) were used. Avidin-biotin

peroxidase complexes were added, followed by visualization with 3,3′-diaminobenzidine

tetrahydrochloride (Dako, Glostrup, Denmark). All sections were counterstained with Mayer's

haematoxylin (Histolab Products, Göteborg, Sweden).

Cell culture

For figure 3A and B, human aortic tissue preparation consisted of an immediate dissection to separate

medial and adventitial layers followed by enzymatic digestion. Normal SMCs were obtained from

transplant patients. Medial samples were incubated in a collagenase and elastase 0.1% solution (3

hours, 37°C), to obtain SMC cultures4. All cells were routinely cultured in Smooth Muscle Cell

medium (Promocell, Heidelberg, Germany) containing 5 % fetal calf serum (FCS), gentamicin-sulfate

(50 mg/mL), amphotericin B (50 µg/mL), insulin (5mg/mL) and growth factors (human Epidermal and

Fibroblast Growth Factors, respectively 0.5 and 2 µg/mL). Passages 3-6 were used for experiments.

For figures 3C and E, human aortic SMCs were isolated from TAV and BAV aortas biopsies. Media

layer was separated from adventia layer, and kept in culture wells with specific medium (CC-3182,

Clonetics SmBM, LONZA, Basel, Switzerland) for approximately 10 days, time necessary in order to

detect SMCs outgrowing from the tissue. Biopsies were removed and cells held in culture for 2-3

passages before freeze them. Cells were treated at passage 6 with 5-10-20 ng/ml of TGFβ (R&D

Systems, Minneapolis, MN, USA).

For supplementary figure 4, human aortic SMCs purchased from American Type Culture

Collection (PCS-100-012, ATCC, Manassas, VA, USA) were maintained in growth medium

containing supplements (CC-3182, Clonetics SmBM, LONZA, Basel, Switzerland). The cells were

grown to confluence, and treated with 1 and 10 ng/mL, respectively, of TGFβ for 24 hours. Cells were

thawed at passage number 2, splitted once and cultured with stand-by medium (0.5% FCS) 24 hours

before the treatment.

4

Principal component analysis

Principal component analysis (PCA)5 was performed on Affymetrix Human Exon 1.0 ST Array meta

probe set level data that had been pre-processed using RMA normalization. The data was filtered with

respect to signal levels of Y chromosome genes on array analyses of female samples. This filtering

resulted in inclusion of 10888 genes and 109 tissue samples belonging to dilated (n=68) and non-

dilated aortas (n=41). Prior to PCA, the data was scaled to unit variance and mean centered. PCA was

performed using the SIMCA P+12.0.1 x64 (Umetrics, Umeå, Sweden) software. The R2 and Q2 model

paremeters for the first three PCs describe 48.2% and 42.7% respectively of the variation seen in the

data. PCA was also performed on a sub selection of 26 genes belonging to TGFβ pathway together

with EDA and EDB exons. The tolerance ellipse based on Hotelling's T2 at a significance level 0.05 is

calculated and shown in the score plot of the TGFβ pathway. The R2 and Q2 model paremeters for the

first two PCs describe 47.3% and 44.6% respectively of the variation seen in the data.

Univariate statistics

Statistical significance was determined using the Mann-Whitney test and the Spearman rank

correlation test. A p-value <0.05 was considered statistically significant.

References

1. Bonow RO, Carabello BA, Chatterjee K, de Leon AC, Jr., Faxon DP, Freed MD, Gaasch WH,

Lytle BW, Nishimura RA, O'Gara PT, O'Rourke RA, Otto CM, Shah PM, Shanewise JS. 2008

focused update incorporated into the ACC/AHA 2006 guidelines for the management of

patients with valvular heart disease: a report of the American College of Cardiology/American

Heart Association Task Force on Practice Guidelines (Writing Committee to revise the 1998

guidelines for the management of patients with valvular heart disease). Endorsed by the

Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and

Interventions, and Society of Thoracic Surgeons. J Am Coll Cardiol. 2008;52:e1-142.

2. Luo W, Friedman MS, Shedden K, Hankenson KD, Woolf PJ. GAGE: generally applicable

gene set enrichment for pathway analysis. BMC Bioinformatics. 2009;10:161.

3. Jones JA, Spinale FG, Ikonomidis JS. Transforming growth factor-beta signaling in thoracic

aortic aneurysm development: a paradox in pathogenesis. J Vasc Res. 2009;46:119-37.

4. Battle T, Arnal JF, Michel JB. Hyperproliferation of aortic smooth muscle cells and

fibroblasts from young SHR rats is not shared by endothelial cells. Clin Exp Pharmacol

Physiol. 1994;21:981-9.

5. Krzanowski WJ. Principles of Multivariate Analysis. New York: Oxford University Press;

2000.

Paloschi et al. Supplementary Figures Supplementary Figure legends Supplementary Figure I. Principal component analysis (PCA) of Affymetrix Human Exon 1.0 ST

Array meta probe set level data containing a total of 10888 genes expressed in aortic samples from

109 patients. The PCA is presented in three-dimensional orthogonal plane projections. (A) PC score

plot showing dilated (red dots) and non-dilated (black dots) aortic samples. (B) Loading PC plot

showing the corresponding genes (black dots) on meta probe set level. The samples have been RMA

normalized as well as scaled to unit variance and mean centered prior to the analysis.

Supplementary Figure II. Increased mRNA expression of EDA-containing FN in dilated aortic tissue

(maximal aortic diameter >45mm) of TAV patients. FN mRNA was analyzed by quantitative real-time

PCR. A-D, media layer of the vessel wall; E-H, adventitia. (A, E) Total FN; (B, F) EDA-containing

FN (EDA-FN); (C, G) EDB-containing FN (EDB-FN); (D, H) ratio between EDA-FN and total FN.

Supplementary Figure III. Immunostaining of EDA containing FN in BAV and TAV patients. (A)

EDA-FN in non-dilated (<40mm) aorta from BAV, (B) EDA-FN in dilated (>45mm) aorta from BAV,

(C) EDA-FN in dilated (>45mm) aorta from TAV, (E-G) the corresponding staining of smooth

muscle alpha actin (D,H) negative controls.

Supplementary Figure IV. TGFβ-induced expression of EDA-FN and EDB-FN. Aortic SMCs were

treated with 1 and 10 ng/ml, respectively, of TGFβ for 24h. EDA-FN (A), EDB-FN (B) and SM22 (C)

mRNA expression was analyzed by quantitative real-time PCR.

A B

Supplementary Fig I

A EMedia Adventitia

7

8

Tot FN P=0.06

1600

0

Tot FN P=0.06

0

1

2

3

4

5

6

Tot

FN

/RP

LP

0

TAV<40 mm

TAV>45 mm

0

400

800

1200

Tot

FN

/RP

LP0

TAV<40 mm

TAV>45 mm

B F

ED

A-F

N/R

PL

P0

EDA-FN p<0.01

50

100

150

200

250

300

350

ED

A-F

N/R

PL

P0

EDA-FN p<0.001

1

2

3

4

C G

TAV<40 mm

0TAV

>45 mm

1500

2000

N/R

PL

P0

EDB-FN P=0.19

TAV<40 mm

TAV>45 mm

0

N/R

PLP

0

EDB-FN P=0.14

4

5

6

7

8

D H

0

500

1000

ED

A-F

N

TAV<40 mm

TAV>45 mm

EDA-FN/Tot FNP=0.056

ED

B-F

N

TAV<40 mm

TAV>45 mm

0

1

2

3

4

EDA-FN/Tot FNP=0.11

0

0.1

0.2

0.3

Ra

tio

TAV TAV

Ra

tio

TAV TAV0

1

2

3

TAV<40 mm

TAV>45 mm

TAV<40 mm

TAV>45 mm

Supplementary Fig II

EDA-FN SMC α-actin

A BAV <40mm E BAV <40mm

B BAV >45mm F BAV >45mm

C TAV >45mm G TAV >45mm

D control H control

Supplementary Fig III

0.6

0.8

1.0

1.2

1.4

A-F

N/R

PLP

0

EDA-FNA

0

0.2

0.4

C 1ng TGFβ

2.0

2.5

ED

A0

EDB-FNB

10ng TGFβ

0

0.5

1.0

1.5

ED

B-F

N/R

PLP

SM22C

C 1ng TGFβ 10ng TGFβ

0

0.2

0.4

0.6

0.8

1.0

1.2

SM

22/R

PLP

0

C 1ng TGFβ 10ng TGFβC 1ng TGFβ 10ng TGFβ

Supplementary Fig IV

Impaired Splicing of Fibronectin Is Associated With...

Documents

Transcript of Impaired Splicing of Fibronectin Is Associated With...