International Journal of Cardiology 152 (2011) 337–344

Contents lists available at ScienceDirect

International Journal of Cardiology

j ourna l homepage: www.e lsev ie r.com/ locate / i j ca rd

Imbalance between pro-angiogenic and anti-angiogenic factors in rheumatic andmixomatous mitral valves

Giovanni Mariscalco a,⁎, Roberto Lorusso b, Fausto Sessa c, Vito Domenico Bruno a, Gabriele Piffaretti d,Maciej Banach e, Paolo Cattaneo c, Giuseppe Paolo Cozzi a, Andrea Sala a

a Department of Surgical Sciences, Cardiac Surgery Unit, Varese University Hospital, University of Insubria, Varese, Italyb Cardiac Surgery Unit, Civic Hospital, Brescia, Italyc Department of Pathology and Rehabilitation Cardiology, IRCCS Clinical Institute Multimedica Holding Santa Maria, Castellanza, University of Insubria, Italyd Department of Surgical Sciences, Vascular Surgery Unit, Varese University Hospital, University of Insubria, Varese, Italye Department of Hypertension, Medical University of Lodz, Lodz, Poland

⁎ Corresponding author. Department of Surgical ScieVarese University Hospital, University of Insubria, Via GItaly. Tel.: +39 347 9689055; fax: +39 0332 264394.

E-mail address: giovannimariscalco@yahoo.it (G. Ma

0167-5273/$ – see front matter © 2010 Elsevier Irelanddoi:10.1016/j.ijcard.2010.08.001

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 7 December 2009Received in revised form 2 July 2010Accepted 3 August 2010Available online 15 September 2010

Keywords:Mitral valveAngiogenesisChondromodulin-IVEGF

Background: A balance between angiogenic and anti-angiogenic factors is critical in tissue development,tissue repair and homeostasis. Aberrant angiogenesis has been implicated in several pathologic conditions,including valvular heart disease. The aim of this study was to ascertain the pathogenetic role of angiogenesisin rheumatic and mixomatous mitral valve diseases.Methods: Leaflets from mixomatous (n=20) and rheumatic (n=20) mitral valves removed from surgicalpatients, and normal mitral valve (n=6) obtained at autopsy were collected. Immunohistochemical studieswere performed on sequential valve sections, evaluating CD31, CD34, α smooth muscle actin (α-SMA),vascular endothelial growth factor (VEGF), VEGF receptor-1 (VEGFR1), VEGF receptor-2 (VEGFR-2), andchondromodulin-I (Chm-I).Results: Immunohistochemistry revealed significant differences among groups in CD31 (p=0.001), CD34

(pb0.001), α-SMA (pb0.001), VEGF (pb0.001), VEGFR1 (p=0.007), VEGFR2 (p=0.011), and Chm-I(pb0.001) expressions. Rheumatic valves demonstrated a severe up-regulation and down-regulation in pro-angiogenic and anti-angiogenic factors, respectively, compared with mixomatous and normal mitral valves.On the contrary, mixomatous valves showed a significant up-regulation of anti-angiogenic factors withrespect to rheumatic and normal valves.Conclusions: These findings provide evidence that an imbalance between pro-angiogenic and anti-angiogenicfactors is implicated in mitral valve disease. Pro-angiogenic factors are up-regulated in rheumatic disease,while anti-angiogenic ones in mixomatous mitral valves.

© 2010 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

Angiogenesis is a central feature of several diseases, being governedby an integrated signaling pathway and modulated by numeroussoluble factors and cytokines [1]. In adults, angiogenesis is consideredconstitutively quiescent and a balance between angiogenic and anti-angiogenic factors is critical in tissue development, tissue repair andhomeostasis [1]. Aberrant angiogenesis has been implicated in severalpathologic conditions, including cancer, diabetic retinopathy, athero-sclerosis, various ischemic and inflammatory diseases [2–5].

In addition, deregulatedangiogenesis has been recentlydocumentedin the pathogenesis of heart valve diseases [3,6,7]. An intensevascularization accompanied by an up-regulation of angiogenic factors

nces, Cardiac Surgery Division,uicciardini, 9, I-21100 Varese,

riscalco).

Ltd. All rights reserved.

has been encountered in rheumatic valvular disease and endocarditis[3,6,8]. Despite an intense research in thisfield, ourunderstandingof theangiogenic properties of cardiac valves is still limited. Furthermore, therole of angiogenesis on mixomatous and rheumatic mitral valve lesionsis unclear.

The aim of this studywas to ascertain the pathogenetic importanceof angiogenesis in these two valve diseases, defining the role ofdifferent angiogenic and anti-angiogenic factors and their involve-ment into rheumatic and mixomatous mitral valve lesions.

2. Material and methods

2.1. Study population

Samples comprising 20 rheumatic (9 men; age: 56 to 81 years) and 20mixomatousmitral valves (15 men; age: 50 to 84 years) at the time of surgical valve operation werecollected. Normal mitral valves were obtained from autopsy of 6 patients (4 men; age:44 to 73 years), without an underlying valve disease. Detailed review of clinical chartswas analysed and patients with cancers, inflammatory disorders, connective tissue

338 G. Mariscalco et al. / International Journal of Cardiology 152 (2011) 337–344

disease, endocarditis and hyperparathyroidism were specifically excluded from thestudy. The diagnosis of rheumatic and mixomatous mitral diseases was based onhistory and physical examination of the patients as well as echocardiographic andsurgical findings.

The study was approved by our Institutional Review Board (protocol nr. 322/2007)and was not supported by any external source of funding.

2.2. Histological analysis

Mitral leaflet samples were fixed in buffered formalin (formaldehyde 4% w/v andacetate buffer 0.05 mol/L) for 12 h at room temperature and then routinely dehydrated,embedded in paraffin, and cut into 4-μm serial sections. Sections were stained withhematoxylin and eosin and with Masson trichrome for pathological study focusing onmorphology and connective tissue elements. The sections were examined at differentmagnifications by using an Olympus BX40 light microscope (Olympus, Tokyo, Japan).

Morphometric and immunohistochemical analyses were performed on histologicalsections by two independent pathologists who were blinded to the underlying mitralvalve disease.

2.3. Immunohistochemical staining

Immunohistochemical staining was performed on sections prepared fromformalin-fixed and paraffin-embedded specimens that were dewaxed and rehydratedusing Bio-clear (Bio-Optica, Milan, Italy) and graded alcohols. Endogenous peroxidasewas blocked by dipping sections in 3% aqueous hydrogen peroxide for 10 min at roomtemperature. After antigen retrieval, primary antibody incubation was done at 4 °C for18–20 h and it was followed by the avidin–biotin complex (ABC) procedure. Sectionswere then immersed in 0.03% 3,3′-diaminobenzidine-tetrahydrochloride and counter-stained with Harris' hematoxylin. After treatment with normal rabbit or normal swineserum, the sections were incubated with the corresponding primary antibodies againstCD31 glycoprotein ((JC/70A) 1:20 dilution; Dako, Copenhagen, DK), CD34 ((My10)1:10 dilution; Becton Dickinson, San Jose, CA), chondromodulin ((Chm-I) 1:100dilution; Santa Cruz Biotechnology, Europe), α smooth muscle actin ((α-SMA) (1A4)dilution 1:5000; Sigma, Saint Louis, Missouri, USA), vascular endothelial growth factor((VEGF) dilution 1:400; Santa Cruz Biotechnology), VEGF receptor-1 ((VEGFR-1)dilution 1:100; Santa Cruz Biotechnology), and VEGF receptor-2 ((VEGFR-2) (FLK-1)dilution 1:600; Santa Cruz Biotechnology). CD34 glycoprotein was used to detect thepresence of valve interstitial cells (VIC), while CD31 with CD34 the presence ofendothelial cells, and α-SMA the presence of myofibroblast and smooth muscle cells.Chm-I, VEGF with VEGFR-1 and VEGFR-2, were used for the detection of anti-angiogenic and angiogenic factors.

The antigen retrieval was performed with 0.01 M citrate buffer pH 6 for 10 min in amicrowave oven at 650 W for CD31 and CD34; with 0.01 M citrate buffer pH 6 for20 min in a microwave oven at 650 W for VEGF and VEGFR-1 and VEGFR-2; and withtrypsin for 20 min at 37 °C for Chm-I and α-SMA.

According to Soini and colleagues [3], the immunostainings with the wholeantibodies were evaluated semiquantitatively, as follows:−, no positive immunostain-ing; +, b25% of the cell population showing positivity; ++, 25% to 50% of the cellpopulation showing positivity; or +++, N50% of the cell population showingpositivity.

2.4. Statistical analysis

Histopathologic data were prospectively recorded and tabulated with MicrosoftExcel (Microsoft Corp, Redmond, WA). All comparisons among the three groups wereconducted by one-way ANOVA. Tukey's post-hoc test was used to determinedifferences between groups, where pb0.05 was considered significant. All data arepresented as means±standard error of the mean (SEM). Statistical analysis wascomputed with SPSS, release 16.0 for Windows (SPSS Inc, Chicago, IL).

3. Results

3.1. Histological findings

Figs. 1–3 compare the histological features of normal, mixomatousand rheumatic valves. Calcification and fibrosis processes weremarkedly different amongmitral valve groups as well as angiogenesis.No calcified or fibrotic area without leaflet layer abnormality wasobserved in control valves. Angiogenesis was also absent. Mixomatousvalves showed thick leaflet layers; fibrosis was focally demonstratedwith a few vessel formations, but no calcification was observed(Fig. 2E–F). Rheumatic valves were characterised by a change in thecomposition of all leaflet layers. The normal valvular architecture wasstill maintained only in areas lacking fibrosis and neovascularization.In the majority of rheumatic valves the extracellular matrix wassubstituted by fibrotic tissue with scattered calcification and vessel

formation (Fig. 3D–G). A residual inflammatory cellular infiltrate wasobserved within the rheumatic valve lesions only.

3.2. Immunohistochemistry findings

Immunohistochemistry of mitral valves revealed significantdifferences among groups (Table 1). In particular, angiogenic andanti-angiogenic factor expressions were significantly different amongnormal, mixomatous and rheumatic mitral valves (Table 1). On onehand, VEGF appeared more represented in rheumatic valves com-pared to the normal and mixomatous ones (pb0.001 and p=0.005,respectively; Figs. 1C; 2C; 4A and 5). The same results were obtainedfor VEGFR-1 and VEGFR-2 factors (Figs. 1G–H; 2H; 4C–D). On theother hand, Chm-I expression was significantly higher in mixomatousgroupwith respect to the normal and rheumatic valves (p=0.047 andpb0.001, respectively; Fig. 5).

In detail, normal mitral valves were constituted by a population ofnormal CD34-positive, CD31-negative andα-SMA-negative VICs. Onlya minority of VICs were positive for α-SMA (activated myofibroblast)in these valves (Fig. 1A–B). VEGF was not observed in control valves,and no vessels or CD31-positive endothelial cells were demonstrated(Fig. 1F). Differently, mixomatous valves revealed an increase of VICswith CD34 immunoreactivity (Fig. 1D). A minority of these CD34-positive cells were also α-SMA-immunopositive. In addition, mix-omatous valves expressed the highest Chm-I level and were slightlyimmunopositive for VEGF and its receptors (Figs. 2C and 5). Inrheumatic valves, the majority of VICs showed a transition from CD34positivity to α-SMA positivity as activated myofibroblasts able tomodify the extracellular matrix (Fig. 3D). In addition, residual VICswere highly positive for angiogenic factors, such as VEGF, VEGFR-1and VEGFR-2, while Chm-I was minimally expressed (Fig. 3A–D). Theendothelial CD31 marker was overrepresented in the rheumaticgroup compared with the normal and mixomatous ones (p=0.003and p=0.004, respectively), staining all the endothelial cells liningthe small and the larger vessel channels (Fig. 3G). Finally, α-SMApositivity was observed in the wall of new larger vessel channels ofrheumatic valves (Fig. 3G–H).

4. Discussion

A balance between angiogenic and anti-angiogenic factors iscrucial for tissue development, organ growth, reproductive functionsand tissue repair events [2,9]. Aberrant angiogenesis has beenimplicated in several diseases, particularly in cancer, where itscontribution to tumour growth and metastasis is widely recognized[5]. Pathologic angiogenesis is also a hallmark of atherosclerosis,various ischemic and inflammatory diseases [2,4,5]. In the cartilages,aberrant angiogenesis can contribute to arthritis, and in the retina, it isassociated with vision loss [4,5].

Recent studies suggest a predominant role of angiogenesis inseveral cardiac diseases. Aortic valve stenosis, valve endocarditis andcardiac allograft vasculopathy have been associated to a deregulatedangiogenesis [6,10]. However, despite the intense research in thisfield, our understanding of angiogenic properties of cardiac valves isstill limited, and the role of angiogenesis on mixomatous andrheumatic mitral valve lesions is unclear.

Similar to the rheumatic aortic valve stenosis, our study demon-strated an imbalance between angiogenic and anti-angiogenic factorsin rheumatic mitral valves [3,6]. In consonance with Yoshioka and co-workers [11], an intense VEGF expression, neovascularization andcalcification were observed in areas of Chm-I down-regulation. Aplausible mechanism for this mutually exclusive expression of Chm-Iand VEGFmay be explained by an upstream signal, which controls thegenetic switch between angiogenic and anti-angiogenic factors [11].However, the loss of valvular interstitial cells due to the underliningrheumatic inflammatory disease seems to be implicated [11].

Fig. 1. Morphological features of normal mitral valves (magnifications ×100). No vessels were detected both in hematoxylin and eosin stain (A) and CD31 immunostaining (B). On thecontrary, a diffuse chondromodulin-I (C) and CD34 (D) immunoreactivity were observed on valve interstitial cells. Actin (E), VEGF (F), VEGFR-1 (G) and VEGFR-2 (H) are negative.

339G. Mariscalco et al. / International Journal of Cardiology 152 (2011) 337–344

Inflammatory mediators increase VEGF mRNA levels and VEGF itselfreciprocally contributes to the expression of inflammatory cells,promoting their activation and migration [12,13]. Angiogenesis thenpromotes endochondral bone formation [14]. Furthermore, VEGFregulates bone remodelling by attracting endothelial cells and by

stimulating osteoblast differentiation, with the typical observedcalcified valve lesions [15]. In addition, in our study, the intenseangiogenesis encountered in rheumatic valves was substantiatedalso by the up-regulation of VEGFR-1 and VEGFR-2. These tyrosinekinase receptors, throughwhich VEGF acts, are normally implicated in

Fig. 2.Mixomatous mitral valve with mild interstitial edema and increased interstitial cells in hematoxylin and eosin stain (A). An important diffuse CD34 (B) and chondromodulin-I(C) immunoreactivity are evident while a few actin-positive cells (D) were observed (arrow). A small number of vessels with CD31 (E) and CD34 (F) endothelial positive cells weredetected in fibrous area at the edge of the leaflet with few interstitial cells in-between, that are well represented below (arrow). VEGF (G), and VEGFR-1 (H) are detected in a smallnumber of interstitial cells. (A, magnification ×200; B–H, magnification ×100).

340 G. Mariscalco et al. / International Journal of Cardiology 152 (2011) 337–344

Fig. 3. Rheumatic mitral valve showing fibrous tissue (A), small and medium size vessels (B) and area of calcification (C) in hematoxylin and eosin stain. The normal CD34 interstitialcells are replaced by actin-positive myofibroblast cells (D). Vessels of medium size have muscle actin-positive wall (E) lined by endothelial CD34 (F) and CD31 (G) positive cells.Small vessels are only made by endothelial CD31-positive cells (H). (A, C–F, H, magnification ×100; B, G, magnification ×200).

341G. Mariscalco et al. / International Journal of Cardiology 152 (2011) 337–344

different angiogenic processes, stimulating a variety of signallingpathways [16,17].

The present study also confirms that in normal valves angiogenesisis a constitutively quiescent process. VIC CD34-positive and CD31-

negative were detected in normal valves compared with rheumaticsamples. In line with the observations of Chalajour and co-workers[6], the present data raise the possibility that endothelial activationwith the subsequent aberrant angiogenesis is a determinant process

Table 1Comparison among control, rheumatic and mixomatous valves.

Variable Control Rheumatic Mixomatous P*

(n=6) (n=20) (n=20)

CD34 3.00±0.00 1.47±0.15† 2.40±0.14‡ b0.001CD31 0.00±0.00 1.40±0.18† 0.47±0.22‡ 0.001α-SMA 0.00±0.00 2.20±0.17† 0.93±0.20†,‡ b0.001VEGF 0.00±0.00 0.80±0.12† 0.33±0.10‡ b0.001VEGFR-1 1.50±0.00 2.80±0.11† 2.10±0.29‡ 0.007VEGFR-2 1.50±0.00 2.60±0.19† 1.90±0.22‡ 0.011Chm-I 2.00±0.00 1.30±0.14† 2.80±0.17†,‡ b0.001

Data are mean±SEM.*One-way ANOVA between groups.†pb0.05 versus control group; ‡pb0.05 versus rheumatic group (Tukey's test).Abbreviations:α-SMA= α-smooth muscle actin; VEGF = vascular endothelial growth factor; VEGFR-1 = VEGF-receptor-1; VEGFR-2 = VEGF-receptor-2; Chm-I = chondromodulin-I.

342 G. Mariscalco et al. / International Journal of Cardiology 152 (2011) 337–344

for rheumatic valve disease. Moreover, the evidence that Chm-I isnormally expressed in control valves, suggests the role of anti-angiogenic factors in the maintenance of the normal valvularmorphology and function. As a matter of fact, normal valves areavascular and oxygen supply is via diffusion from blood stream [18].In addition, oxygen tension is a key regulator of the VEGF geneexpression [19,20].

However, the most relevant and interesting findings of this studywere detected in mixomatous mitral valves. Contrary to the normaland rheumatic valves, the mixomatous ones exhibited an up-regulation of the Chm-I anti-angiogenic factor. To our knowledge,the present study is the first describing an intense Chm-I expressionwith a down-regulation of pro-angiogenic factors, such as VEGF andits receptors, in mixomatous mitral valves (Fig. 2). Interestingly, thisanti-angiogenic up-regulation is an unexpected finding in relationwith the observed molecular processes involved in mixomatousstromal degeneration (MSD). Matrix-degrading enzymes such asmetalloproteinases (MMP)and their tissue inhibitors are involved in the

Fig. 4. Rheumatic mitral valves and pro-angiogenic factors. On one hand, scattered chondrohand, a diffuse and relevant VEGF immunoreactivity was founded in remaining interstitial cel(D, magnification ×100).

pathogenesis, progression and remodeling of MSD [21]. The imbalancebetween MMP and their tissue inhibitors and the increased MMPactivity are theputative precipitatingevents for theMSD.MMPsarepro-angiogenicmodulatorswith an important role in the regulation of tissuecalcifications and extracellular matrix degradation [22]. Furthermore,both MMP and VEGF are critical for bone development, favouringosteoclast and endothelial cell invasion [23]. The observed up-regulation of Chm-I in our study could be explained as a defensivecompensatory process against VEGF andMMP activation because of thevalvular mechanical stress and inflammation encountered in MSD [24].In line with this hypothesis, Kimura and co-workers have recentlyinvestigated the correlation between avascularity and mitral chordaetendineae (CT) rupture [25]. In the outer layer of normal animal andhuman mitral CT, these authors founded that tenomodulin, which is achondromodulin-I-related gene with anti-angiogenic properties, wasabundantly expressed [25]. The same observations were substantiatedin non-ruptured CT of mixomatous mitral valves [25]. On the contrary,the local absence of tenomodulinwith angiogenesis andMMPactivationwere associatedwithCT rupture [25]. Our presentdatadonot consent todeeply investigate this hypothesis and to exclude other possible andmore complex angiogenic MSD mechanisms.

Certainly, angiogenesis is a complexprocess and its imbalancewith asevere valvular compromise is actually debilitating with no effectivetherapy other than the surgical one. The present study possibly providesnew insight into mechanisms by which ongoing and progressiveaberrant angiogenesis may contribute to common clinical pathways. Itremains tobedeterminedwhethermolecular activity and cell activationin the mitral leaflets are causal or whether regurgitation and abnormalmechanical stress induce angiogenic disruptionasa reactivemechanism[27]. This should be possibly clarified in future studies. Nevertheless,these observations on diseased human tissue cannot resolve the natureof the primary stimulus to mitral valvular diseases. It is distinctlypossible that the encountered aberrant angiogenesis could have beentriggered by coexistent patient co-morbidities. As demonstrated withother cardiac diseases (e.g., end-stage coronary disease), co-morbidities

modulin-I cells in a fibrotic tissue were detected (A, magnification ×100). On the otherls (B, magnification ×100) together with VEGFR-1 (C, magnification ×200) and VEGFR-2

1,2 *NS VEGF

**NS VEGFR-1

0,8

1,0

**

2,0

3,0

0,2

0,4

0,6G

rad

ing

Sco

re

1,0

0,0

CD31 Chm-I

0,0

2,0

2,5 **

NS

3,0

4,0 ***

*

0,5

1,0

1,5

Gra

din

g S

core

1,0

2,0

0,0 0,0

NORMAL RHEUMATIC MIXOMATOUS

Fig. 5. The graphs show the semiquantitative analysis of vascular endothelial growth factor (VEGF), VEGF receptor-1 (VEGFR-1); CD31 and chondromodulin-I (Chm-I) amongnormal, rheumatic and mixomatous mitral valves. Bars represent SEM. NS = not significant, *pb0.05 and **pb0.001.

343G. Mariscalco et al. / International Journal of Cardiology 152 (2011) 337–344

like diabetes mellitus, hypertension, and hypercholesterolemia cantrigger and impair the angiogenesis balance. These disease states areassociated with reduced nitric oxide bioavailability and endothelialdysfunction, impaired growth factor signaling, and altered expression ofpro- and anti-angiogenic factors [28]. A better understanding of theseangiogenic influences may allow for the modulation of the response toangiogenic therapy also in patients with mitral disease. Unfortunately,the small number of patients in our study as well as in other studies onthe same topic restricts such ambition. Certainly, further genetic andbiochemical studies are needed to help design future therapeuticstrategies better reflecting the complexity of the underlying biologicprocess of angiogenesis [24–26]. The possibility to treat mixomatousand rheumatic mitral in the early phase of mitral disease is attractive.Specific inhibitor of pro-angiogenic or anti-angiogenic factors could beparticularly effective as preventive agents, slowing valvular disease orthe progressive deterioration of valvular tissues after a successful repairsurgery.

The present study was limited by the small number of the analysedsamples, a difficulty shared with other similar studies [3,6,25]. Investiga-tions of angiogenic mechanisms require a large number of observations,but detailed microscopy of valvular samples limits such ambition andrestrains the statistical power. Certainly, a larger population might beneeded to likely provide a more evident relationship between theangiogenesis imbalance and its involvement into rheumatic andmixomatous mitral valve lesions. In addition, the study would possiblyhave benefit from supporting the obtained data with a molecularapproach. A further limitation is the lack of the assessment of otherpro-angiogenic and anti-angiogenic factors, which could have beenassociatedwith rheumatic andmixomatousmitral diseases. Angiogenesisis a complex process being governed by an integrated signalling circuitry,and its modulation is dependent not only on angiogenic factors, but alsoon inflammatory cytokines and extracellular matrix components [2,9].

In conclusion, our study supports the contention that aberrantangiogenesis is a crucial feature of rheumatic and mixomatous diseasesof the mitral valve. Pro-angiogenic factors are up-regulated in rheumaticdisease, while anti-angiogenic ones in mixomatous mitral valves. Futureclinical studies targeting angiogenesis imbalance in mitral valve diseasesare mandatory to prove these attractive observations. Moreover,understanding the angiogenic properties of mitral valves may providethe basis for therapeutic agents in preventing and controlling rheumaticand mixomatous mitral valve disease progressions.

Acknowledgement

The authors are very grateful to the members of the OperatingRoom for acquisition of the sample valves, with particular thanks toDr. Andrea Musazzi, Dr. Carmelo Dominici, and Dr. Sandro Ferrarese.The authors of this manuscript have certified that they comply withthe Principles of Ethical Publishing in the International Journal ofCardiology [29].

References

[1] Shworak NW. Angiogenic modulators in valve development and disease: doesvalvular disease recapitulate developmental signaling pathways? Curr OpinCardiol 2004;19:140–6.

[2] Carmeliet P, Jain RK. Angiogenesis in cancer and other diseases. Nature 2000;407:249–57.

[3] Soini Y, Salo T, Satta J. Angiogenesis is involved in the pathogenesis ofnonrheumatic aortic valve stenosis. Hum Pathol 2003;34:756–63.

[4] Ambati BK, Nozaki M, Singh N, et al. Corneal avascularity is due to soluble VEGFreceptor-1. Nature 2006;443:993–7.

[5] Folkman J. Angiogenesis in cancer, vascular, rheumatoid and other disease. NatMed 1995;1:27–31.

[6] Chalajour F, Treede H, Ebrahimnejad A, Lauke H, Reichenspurner H, Ergun S.Angiogenic activation of valvular endothelial cells in aortic valve stenosis. Exp CellRes 2004;298:455–64.

344 G. Mariscalco et al. / International Journal of Cardiology 152 (2011) 337–344

[7] Rajamannan NM, Nealis TB, Subramaniam M, et al. Calcified rheumatic valveneoangiogenesis is associated with vascular endothelial growth factor expressionand osteoblast-like bone formation. Circulation 2005;111:3296–301.

[8] Yamauchi R, Tanaka M, Kume N, et al. Upregulation of SR-PSOX/CXCL16 andrecruitment of CD8+ T cells in cardiac valves during inflammatory valvular heartdisease. Arterioscler Thromb Vasc Biol 2004;24:282–7.

[9] Pepper MS. Manipulating angiogenesis. From basic science to the bedside.Arterioscler Thromb Vasc Biol 1997;17:605–19.

[10] Raisky O, Nykanen AI, Krebs R, et al. VEGFR-1 and -2 regulate inflammation,myocardial angiogenesis, and arteriosclerosis in chronically rejecting cardiacallografts. Arterioscler Thromb Vasc Biol 2007;27:819–25.

[11] Yoshioka M, Yuasa S, Matsumura K, et al. Chondromodulin-I maintainscardiac valvular function by preventing angiogenesis. Nat Med 2006;12:1151–9.

[12] Ben-Av P, Crofford LJ, Wilder RL, Hla T. Induction of vascular endothelialgrowth factor expression in synovial fibroblasts by prostaglandin E andinterleukin-1: a potential mechanism for inflammatory angiogenesis. FEBSLett 1995;372:83–7.

[13] Clauss M, Gerlach M, Gerlach H, et al. Vascular permeability factor: a tumor-derived polypeptide that induces endothelial cell and monocyte procoagulantactivity, and promotes monocyte migration. J Exp Med 1990;172:1535–45.

[14] Mohler III ER, Gannon F, Reynolds C, et al. Bone formation and inflammation incardiac valves. Circulation 2001;103:1522–8.

[15] O'Brien KD, Kuusisto J, Reichenbach DD, et al. Osteopontin is expressed in humanaortic valvular lesions. Circulation 1995;92:2163–8.

[16] Rahimi N. VEGFR-1 and VEGFR-2: two non-identical twins with a uniquephysiognomy. Front Biosci 2006;11:818–29.

[17] Yancopoulos GD, Davis S, Gale NW, Rudge JS, Wiegand SJ, Holash J. Vascular-specific growth factors and blood vessel formation. Nature 2000;407:242–8.

[18] Hammon Jr JW, O'Sullivan MJ, Oury J, Fosburg RG. Allograft cardiac valves. A viewthrough the scanning electron microscope. J Thorac Cardiovasc Surg 1974;68:352–60.

[19] Ferrara N. Molecular and biological properties of vascular endothelial growthfactor. J Mol Med 1999;77:527–43.

[20] Ferrara N, Davis-Smyth T. The biology of vascular endothelial growth factor.Endocr Rev 1997;18:4–25.

[21] Togashi M, Tamura K, Nitta T, Ishizaki M, Sugisaki Y, Fukuda Y. Role of matrixmetalloproteinases and their tissue inhibitor of metalloproteinases in myxoma-tous change of cardiac floppy valves. Pathol Int 2007;57:251–9.

[22] Liu AC, Joag VR, Gotlieb AI. The emerging role of valve interstitial cell phenotypesin regulating heart valve pathobiology. Am J Pathol 2007;171:1407–18.

[23] Engsig MT, Chen QJ, Vu TH, et al. Matrix metalloproteinase 9 and vascularendothelial growth factor are essential for osteoclast recruitment into developinglong bones. J Cell Biol 2000;151:879–89.

[24] Rabkin E, AikawaM, Stone JR, Fukumoto Y, Libby P, Schoen FJ. Activated interstitialmyofibroblasts express catabolic enzymes and mediate matrix remodeling inmyxomatous heart valves. Circulation 2001;104:2525–32.

[25] Kimura N, Shukunami C, Hakuno D, et al. Local tenomodulin absence,angiogenesis, and matrix metalloproteinase activation are associated with therupture of the chordae tendineae cordis. Circulation 2008;118:1737–47.

[26] Heissig B, Hattori K, Friedrich M, Rafii S, Werb Z. Angiogenesis: vascularremodeling of the extracellular matrix involves metalloproteinases. Curr OpinHematol 2003;10:136–41.

[27] Kalluri R, Zeisberg E. Controlling angiogenesis in heart valves. Nat Med 2006;12:1118–9.

[28] Boodhwani M, Nakai Y, Mieno S, et al. Hypercholesterolemia impairs themyocardial angiogenic response in a swine model of chronic ischemia: role ofendostatin and oxidative stress. Ann Thorac Surg 2006;81:634–41.

[29] Coats AJ. Ethical authorship and publishing. Int J Cardiol 2009;131:149–50.