ava i l ab l e a t www.sc i enced i r ec t . com

www.e l sev i e r . com/ l oca te / sc r

Stem Cell Research (2011) 6, 206214Received 21 August 2010; received in revised form 29 December 2010; accepted 3 January 2011Available online 28 January 2011

Abstract Recent studies suggest that the therapeutic effects of stem cell transplantation following myocardial infarction (MI)are mediated by paracrine factors. One of the main goals in the treatment of ischemic heart disease is to stimulate vascularrepair mechanisms. Here, we sought to explore the therapeutic angiogenic potential of mesenchymal stem cell (MSC)secretions. Human MSC secretions were collected as conditioned medium (MSC-CM) using a clinically compliant protocol. Basedon proteomic and pathway analysis of MSC-CM, an in vitro assay of HUVEC spheroids was performed identifying the angiogenicproperties of MSC-CM. Subsequently, pigs were subjected to surgical left circumflex coronary artery ligation and randomized tointravenous MSC-CM treatment or non-CM (NCM) treatment for 7 days. Three weeks after MI, myocardial capillary density washigher in pigs treated with MSC-CM (645114 vs 98155 capillaries/mm2; P=0.021), which was accompanied by reducedmyocardial infarct size and preserved systolic and diastolic performance. Intravenous MSC-CM treatment after myocardialinfarction increases capillary density and preserves cardiac function, probably by increasing myocardial perfusion. 2010 Elsevier B.V. All rights reserved.a Department of Cardiology, University Medical Center Utrecht, Netherlandsb Institute of Medical Biology, A*STAR, Singaporec Department of Surgery, YLL School of Medicine, NUS, Singapored NUS Graduate School for Integrative Sciences & Engineering, NUS, Singaporee Department of Molecular Cell Biology, Leiden University Medical Center, Netherlandsf Nanyang Technological University, Singaporeg Biotechnology Institute, Singaporeh Department of Cardiology, Academic Medical Center, Amsterdam, Netherlandsi Interuniversity Cardiology Institute of the Netherlands, Utrecht, NetherlandsChaylendra Strijder , Sui Kwan Sze , Andree Choo , Jan J. Piek ,Pieter A. Doevendans a, Gerard Pasterkampa, Dominique P.V. de Kleijn a,i,Human mesenchymal stem cell-conditionedmedium improves cardiac function followingmyocardial infarctionLeo Timmers a, Sai Kiang Limb,c,, Imo E. Hoefera, Fatih Arslana,Ruenn Chai Lai b,d, Angelique A.M. van Oorschot e, Marie Jose Goumanse,

a f g hREGULAR ARTICLE Correspondence to: S.K. Lim, Institute of Medical Biology, A*STAR, Department of Surgery, YLL School of Medicine, NUS, 8A BiomedicalGrove, No. 05505 Immunos, Singapore 136648. Fax: +65 6464 2048. Correspondence to: D.P.V. de Kleijn, Laboratory of Experimental Cardiology, University Medical Center Utrecht, Room G02.523,

Heidelberglaan 100, 3584 CX Utrecht, The Netherlands. Fax: +31 30 2522693.E-mail addresses: saikiang.lim@imb.a-star.edu.sg (S.K. Lim), d.dekleijn@umcutrecht.nl (D.P.V. de Kleijn).

1873-5061/$ - see front matter 2010 Elsevier B.V. All rights reserved.doi:10.1016/j.scr.2011.01.001

207Introduction

Despite significant advances in myocardial revascularization,coronary artery disease and subsequent myocardial infarc-tion (MI) are leading causes of morbidity and mortalityworldwide. One of the main goals in the treatment ofischemic heart disease is the development of effectivestrategies to stimulate vascular repair mechanisms in orderto achieve adequate tissue perfusion.

Although the potential of stem cell therapy for MI is stillheavily debated as a result of studies that failed to showimproved outcome (Moelker et al., 2006; de Silva et al., 2008;Janssens et al., 2006; Lunde et al., 2006; Meyer et al., 2009;Menasche et al., 2008), there are several preclinical andclinical studies that demonstrated the therapeutic effects ofstem cell transplantation (Hashemi et al., 2008; Wolf et al.,2009; Assmus et al., 2006; Schachinger et al., 2006; Straueret al., 2002; Yousef et al., 2009). While robust evidence forcardiomyocyte replacement is still limited, more evidence isaccumulating for vascular repair mechanisms following stemcell transplantation (van Laake et al., 2009; Losordo &Dimmeler, 2004; Erbs et al., 2007). Among the stem cells thathave been tested for cardiac repair, mesenchymal stem cells(MSCs) derived from adult bone marrow have emerged as apromising stem cell type for treating cardiovascular disease(Pittenger & Martin, 2004). Recent studies suggest that atleast some of the therapeutic effects of MSCs aremediated byparacrine factors secreted by the cells (Caplan & Dennis,2006; Gnecchi et al., 2005). These paracrine factors could beexploited to extend the therapeutic possibilities of MSCs forthe treatment of a variety of diseases, including MI.Application of MSC secretions rather than MSCs themselvesin order to restore tissue perfusion could enable us to avoidsome of the limiting factors associatedwith cell therapy, suchas immune incompatibility, tumorigenicity, costs, and wait-ing time for ex vivo expansion.

We previously described how human MSC lines can bereproducibly generated from human embryonic stem cell(hESC) lines, which constitute an invariable source ofconsistently uniform batches of MSCs (Lian et al., 2007). Inaddition, we established for the first time the collection ofthese MSC secretions as conditioned medium (MSC-CM) in aclinically compliant manner, circumventing exposure tovirus, mouse cells, or serum (Sze et al., 2007). Recently, wedemonstrated that MSC-CM treatment following ischemia/reperfusion injury reduces myocardial apoptosis and oxida-tive stress (Timmers et al., 2007a). In the current paper, wereveal the angiogenic potential of clinically compatible MSC-CM and demonstrate in a large animal model that it can beused to improve cardiac function following MI.

Results

MSC-CM stimulates angiogenesis in vitro

Ingenuity pathway analysis of the secretory proteome ofMSC-CM revealed that within the physiological systemdevelopment and function group, cardiovascular systemdevelopment and function has the highest overrepresenta-tion (Pb10-12) with 75 different focus proteins. Within thegroup of cardiovascular system development and function,development of blood vessels, angiogenesis, proliferation ofendothelial cells, neovascularization, and vascularizationare the processes with the highest overrepresentation(supplementary data, Table 1).

Having identified the potential involvement of MSC-CM inblood vessel formation, we used an in vitro assay ofangiogenesis to further investigate this potential. In anangiogenesis assay based on formation of capillary sprouts byHUVEC spheroids, MSC-CM induced significantly more capil-lary sprouting compared with NCM (Fig. 1A). The averagelength of the sprouts did not increase following MSC-CMtreatment compared with NCM (Fig. 1B). The total length ofall sprouts, however, significantly increased in spheroidsincubated with MSC-CM when compared with NCM (Fig. 1C).

In addition, a HUVEC/Matrigel angiogenesis assay wasperformed, which revealed a longer average tube length withCM compared with NCM (1.070.046 vs 0.990.041 mm;P=0.003).

MSC-CM treatment after infarction increasescapillary density and decreases collagen density

Capillary density was higher in the border areas followingMSC-CM treatment compared with NCM treatment (Figs. 2Aand B). In the remote area, capillary density did not differbetween MSC-CM treatment and NCM treatment (Fig. 2C).We were unable to reliably quantify capillary density in theinfarct area due to the unorganized structure of the infarctgranulation tissue 3 weeks following MI.

Picrosirius staining revealed that collagen density in theborder area and in the remote myocardium was lower in pigstreated with MSC-CM compared with pigs treated with NCM.In the infarct area, no difference was detected between thetreatment groups (Fig. 3).

MSC-CM preserves cardiac function following MI

One hour after coronary artery ligation, cardiac functiondecreased similarly in both groups. Fractional area shorten-ing decreased from 45.81.9 to 32.32.0% in the NCM groupand from 46.94.6 to 29.73.4% in the MSC-CM group(P=0.670). Also cardiac output (2.800.10 L/min (NCM) vs2.670.22 L/min (MSC-CM); P=0.596), mean aortic pressure(908.3 mm Hg (NCM) vs 845.4 mm Hg (MSC-CM); P=0.544)and dP/dtMAX (113583 mm Hg/s (NCM) vs 109683 mm Hg/s(MSC-CM); P=0.741) were similar in both groups 1 h aftercoronary artery ligation.

In the 3 weeks following MI, the LV internal areasincreased in all pigs. However, the increase in LV internalarea tended to be more pronounced in NCM-treated animals,pointing to reduced remodeling in MSC-CM-treated pigs(Figs. 4A and B). Echocardiographic wall thickness of theinfarct area decreased in pigs treated with NCM, but not inpigs treated with MSC-CM (Fig. 4C). The fractional areashortening was higher following MSC-CM treatment, whichreflects increased systolic cardiac performance (Fig. 4D).Also other systolic functional parameters such as ejectionfraction, dP/dtMAX, stroke volume, and stroke work werehigher in animals treated with MSC-CM. Improved cardiacfunction translated into improved hemodynamic parameters,with the mean arterial pressure and cardiac output being

208 L. Timmers et al.higher in MSC-CM-treated pigs. In addition, diastolic functionimproved, as assessed with dP/dtMIN and diastolic chamberstiffness. All functional parameters are summarized in Table 1.Myocardial infarct size 3 weeks following coronary arteryligation was smaller in MSC-CM-treated pigs compared to NCM-treated pigs (11.62.0% vs 16.61.2% of the LV; P=0.050).

Two pigs died during the surgical procedure due torefractory ventricular fibrillation before treatment withMSC-CM or NCM, and were therefore excluded from thestudy. Two pigs (1 treated with MSC-CM and 1 treated withNCM) died in the third week following MI without signs ofheart failure or cardiac rupture, probably due to ventriculartachycardia/fibrillation.

Discussion

Here, we demonstrate that human MSC-CM treatmentenhances capillary density, resulting in reduced myocardial

Figure 1 HUVEC spheroid assay of angiogenesis. The number of calength of sprouts (pixels, C) after incubation of HUVEC spheroids withCM and NCM are demonstrated in panels D and E, respectively. * Pbinfarct size and preserved systolic and diastolic function in aporcine model of myocardial infarction.

Recently, it was demonstrated that administration of MSCsecretions improved outcome in a murine model of MI(Gnecchi et al., 2005). Vascular repair mechanisms consti-tute an important means to reduce myocardial injury and toimprove cardiac function following MI. Although the thera-peutic potential of stem cell transplantation is still a subjectof discussion, evidence for vascular repair mechanismsfollowing stem cell transplantation is growing (van Laakeet al., 2009; Losordo & Dimmeler, 2004; Erbs et al., 2007). Ifvascular repair mechanisms are induced by paracrinefactors, application of stem cell secretions may extend therepertoire of stem cell-based therapies with a differentdimension.

An in vitro sprouting assay confirmed Ingenuity analysis ofearlier MSC-CM proteomics data and identified the angio-genic potential of MSC-CM. In order to investigate thetherapeutic potential of MSC-CM to stimulate angiogenesis

pillary sprouts (A), mean length of sprouts (pixels, B), and totalMSC-CM (n=12) or NCM (n=12). Representative pictures of MSC-0.01 vs NCM.

Figure 2 Capillary density. Representative sections of border areafrom pigs treated with NCM (A, n=9) or MSC-CM (B, n=9). Quademonstrated in panel C. Scale bars, 500 m.

Figure 3 Collagen density. Quantification of collaged densityin the infarct, border, and remote areas in pigs treated with NCMand MSC-CM.

209in vivo, we used a porcine model of myocardial infarction.During myocardial ischemia following coronary artery liga-tion, cardiomyocytes in the central infarct perish withinminutes to hours after coronary occlusion. The border zoneof the ischemic area, however, contains injured but viablecells under hypoperfused and hypoxic conditions which mayeither be salvaged or die, depending on the conditions afterischemia. Restoration of myocardial perfusion by means ofangiogenesis restores the metabolic needs of cardiomyocytesin the border area and offers the cells a better chance ofsurvival, which in turn could lead to reduced myocardialinjury. Intracoronary delivery of bone marrow-derivedprogenitor cells has been suggested to promote vascularrepair following MI, potentially via paracrine mechanisms(Erbs et al., 2007). Such effect was also observed in a rodentmodel of hind limb ischemia (Kinnaird et al., 2004). Asbecame evident from our histological analyses, MSC-CMincreased capillary density in the border areas and myocar-dial infarct size was reduced, which was accompanied bypreserved LV dimensions and cardiac function.

Besides stimulation of angiogenesis, it cannot be excludedthat also other mechanisms contributed to reduction of

s after lectin staining to visualize endothelium, 3 weeks after MI,ntification of capillary density in remote and border area is

210 L. Timmers et al.A

C

WT

infa

rct (

cm)

*

LVES

ar

ea (c

m2)myocardial infarct size and preservation of myocardialfunction. With preserved left ventricular dimensions andinfarct thickness, inhibition of post MI cardiac remodeling is apotential contributing mechanism. Collagen density in theborder area and remote myocardium was lower in pigstreated with MSC-CM. Cardiac fibrosis is associated withheart failure following myocardial infarction and maycontribute to progression of systolic and diastolic dysfunc-tion (Burlew & Weber, 2002; Cleutjens et al., 1999). Werecently demonstrated that intravenous and intracoronaryMSC-CM treatment reduces myocardial apoptosis and oxida-tive stress (Timmers et al., 2007a). In that study, cardio-myocyte death was targeted acutely following ischemia andreperfusion injury. In the current study, treatment wasstarted late after coronary artery ligation, and a model ofpermanent coronary artery ligation was used in order toinvestigate the angiogenic potential of MSC-CM and its effecton cardiac function independent of reduction of reperfusioninjury. The fact that cardiac function improved using thisprotocol shows that there may be additional benefit fromprolonged treatment with MSC secretions.

For this study, a permanent coronary artery ligation wasperformed in anesthetized animals and functional measure-

E

dP/d

t MA

X (m

mHg/s

)

*

Figure 4 Cardiac function. Left ventricular (LV) end-diastolic intmyocardial wall (infarct) thickness (C), fractional area shortening (FAMI and 3 weeks post MI in pigs treated with MSC-CM (black dots, n=B

D

FAS

(%)

*

*

LVES

ar

ea (c

m2)ments were performed with open chest. In addition, we havenot been able to investigate the effect of commoncomorbidities such as hypertension, dyslipidemia, anddiabetes, which might influence the biologic effect of MSCsecretions. Due to the permanent coronary artery ligation,we were not able to assess the area at risk 3 weeks post MI.We were not able to oversee the effects of human proteininjection on the immune system of the porcine host and thepotential influence on study endpoints. Furthermore, themolecular mechanism for the induction of angiogenesis and/or arteriogenesis and an exact determination of theresponsible angiogenic factors are still lacking. Severalclinical trials with potent angiogenic factors failed toimprove outcome in patients (Henry et al., 2003; Simonset al., 2002). It is not clear how paracrine factors in stem celltransplants, or the paracrine factors alone, can induce abetter effect. Most likely, a combination of angiogenicfactors rather than a single protein is responsible. Alsoexosomes could play an important role. We previouslydemonstrated exosomes in MSC-CM to contain and delivercardioprotective compounds preventing cardiomyocytedeath following ischemia and reperfusion injury (Lai et al.,2010). Exosomes constitute a potential delivery tool to

F

*dP/d

t Min

(m

mHg/s

)

ernal area (A), LV end-systolic internal area (B), posterolateralS, D), dPdtMAX (E), and dPdtMIN (F), before MI (baseline), 1 h post9) or NCM (open dots, n=9). * Pb0.05 vs NCM.

3N

1-10213

ESV, ml320

M (. MIthicn; S

211effectively carry angiogenic factors into the intracellularspace of the target tissue, protecting the factors fromproteases. This may explain why relatively low doses canhave such profound angiogenic effects. Whether exosomesand their content are also responsible for the stimulation ofangiogenesis following MI remains a subject for furtherinvestigation.

Here, we demonstrate that human MSC secretions

EF, %SW, mm HgmlStiffness, mm Hg/ml

Summary of cardiac functional parameters of pigs treated with NCconductance catheter-based LV pressure and volumemeasurementspressure; CO, cardiac output; EDP, end-diastolic pressure; WT, wallend-diastolic volume; ESV, end-systolic volume; EF, ejection fractiovs NCM.Table 1 Hemodynamic and Functional Parameters.

Baseline

Parameter NCM CM

HR, bpm 664 736MAP, mm Hg 955 895CO, L/min 3.10.1 3.00.2dP/dt max, mm Hg/s 121377 124156dP/dt min, mm Hg/s -1274122 -113096EDP, mm Hg 8.00.3 7.10.6WT infarct, cm 0.790.04 0.830.05Lvia (ED), cm2 15.80.5 14.21.0Lvia (ES), cm2 8.50.4 7.60.9FAS, % 45.81.9 46.94.6SV, ml 483 422EDV, mlstimulate angiogenesis. These data support the hypothesisthat MSC transplantation induces vascular repair mechanismsvia paracrine pathways. Most importantly, human MSCsecretions harvested as conditioned medium using a clini-cally compliant protocol reduced infarct size and preservedcardiac function in a large animal model of myocardialinfarction. The study therefore identifies MSC-CM as a noveltherapeutic biologic in the improvement of cardiac functionafter myocardial infarction.

Methods

MSC-CM preparation

The protocols for MSC generation and CM preparation havebeen described in previous papers (Lian et al., 2007; Sze et al.,2007). In short, a chemically defined serum-free culturemedium(DMEMwithoutphenol red, supplementedwith insulin,transferrin, and selenium, 5 ng/ml FGF2, 5 ng/ml PDGF AB,glutamine-penicillin-streptomycin, and -mercaptoethanol)was conditioned by MSCs derived from human embryonicstem cells (hESCs) using a clinically compliant protocol. Threepolyclonal, karyotypically stable, and phenotypically MSC-likecultures, that do not express pluripotency-associated markersbut displayed MSC-like surface antigens (CD29+, CD44+,CD49a+/e+, CD105+, CD166+, CD34, CD45) and geneexpression profile, were generated by trypsinization andpropagation of hESCs from either the HuES9 hESC line or theH1 hESC line in feeder- and serum-free selection media (Lianet al., 2007). One of these cultures, HuES9.E1, can be stablyexpanded for at least 80 population doublings. To harvest MSCsecretions, hESC-derived MSC cultures were transferred to achemically defined, serum-free culture medium to conditionthe medium for 3 days before the media containing MSC

weeks post MI Change from baseline (%)

CM CM NCM CM

846 866 2812 1211666 855 * -305 -58 *3.70.6 5.30.5 2321 7820157162 1650124 * 413 409 *88398 -131395 * 279 -1919 *3.31.6 12.11.1 6821 9442.550.02 0.790.05 * -274 -54 *6.01.8 21.81.5 6412 55127.71.6 12.50.9 * 10414 6014 *2.62.2 42.43.3 * -286 -68 *434 623 * -411 5210 *

13416 12999315 7393.74.4 49.13.3 *497333 4595361 *.260.04 0.140.02 *

n=9) and MSC-CM (n=9), determined with echocardiography andindicates myocardial infarction; HR, heart rate; MAP, mean arterialkness; LVia, left ventricular internal area; SV, stroke volume; EDV,W, stroke work; Stiffness, end-diastolic chamber stiffness. * Pb0.05secretions were collected, clarified by centrifugation, con-centrated 25 times using 10-kDa MW cutoff ultrafiltrationmembranes and sterilized by filtration through a 220-nmfilter. The protein concentration after concentration was0.50 mg/ml. As a negative control, the above-noted serum-free culture medium was processed equally (nonconditionedmedium [NCM]).

Ingenuity pathway analysis

For this analysis we used the proteomic data obtainedpreviously (Sze et al., 2007). Bioinformatic analysis wasperformed using Ingenuity pathway analysis (Ingenuity IPA-2002 version 7.1). The 201 proteins were uploaded inIngenuity with Ingenuity Knowledge Base as a reference setand direct and indirect relationships were included. Allmolecules or relationships were considered for the filtersummary. In the summary, the focus was on Top Bio functionsin which physiological system development and function wasanalyzed in depth.

HUVEC spheroid assay

Spheroids were formed from HUVECs to determine the effectof conditioned medium on angiogenic sprout formation (Korff& Augustin, 1998). Spheroids were formed overnight in 20%

212 L. Timmers et al.methocel in culture medium and subsequently embedded ina collagenmethocel matrix. Sprouting was induced over-night with MSC-CM (n=12) or non-CM (NCM) (n=12) todetermine the effect of the secreted chemotactic proteinson capillary sprouting. After overnight sprouting, spheroidswere fixed in 10% formaldehyde and pictures were taken andthe length and number of sprouts were quantified.

Animals

All experiments were performed in accordance with theGuide for the Care and Use of Laboratory Animals preparedby the Institute of Laboratory Animal Resources andwith priorapproval by the Animal Experimentation Committee of theFaculty of Medicine, Utrecht University, the Netherlands.

Sample size calculation

In a previous study, the mean fractional area shortening(FAS), after LCx ligation was 36.5% (Timmers et al., 2007b).With an alpha of 0.05, power of 0.80, SD of 5.8, anddifference of 20% considered relevant, 10 animals per groupwere needed. To allow for 10% mortality, 11 animals pergroup were enrolled.

Pig study design

Myocardial infarction was induced in 22 Dalland Landracepigs (6070 kg, IDDLO, Lelystad, the Netherlands) by surgicalleft circumflex coronary artery (LCx) ligation. Two pigs diedperioperatively before randomization and were thereforeexcluded from the study. The remaining 20 pigs wererandomized after surgery by using sealed envelopes tointravenous treatment with MSC-CM (2.0 ml MSC-CM, i.e.,1.0 mg protein) or 2.0 ml NCM, initiated 4 h after coronaryartery ligation and the treatment was continued for 7 daystwice daily via a catheter inserted into the jugular vein. Thepigs were sacrificed 3 weeks after MI. To prevent thromboticcomplications and arrhythmias, all pigs were treated withclopidogrel 75 mg/day from 3 days before MI until termina-tion and amiodarone 400 mg/day from 10 days before MIuntil termination.

Myocardial infarction and operational procedure

Pigs were anesthetized as described previously (Timmers etal., 2007b). During the entire operation, electrocardiogram,arterial pressure, and capnogram were continuously moni-tored. After median sternotomy, the left ventricularpressure (LVP) was measured using a pressure tipped Millarcatheter that was inserted through the apex into the leftventricle. A transonic flow probe (Transonic Systems Inc.,Ithaca, NY, USA) was placed around the proximal aorta tomeasure cardiac output. Sutures were then tightened topermanently occlude the proximal LCx. Internal defibrilla-tion with 50 J was used when VF occurred. Prior to and 1 hafter induction of the infarct, echocardiography wasperformed. After stabilization of hemodynamic parametersand heart rhythm, the thorax was closed and the animalswere allowed to recover. Three weeks after induction ofmyocardial infarction, the animals were anesthetized oncemore and the sternum was reopened. Echocardiography andconductance catheter-based pressurevolume (PV) looprecordings were obtained to assess cardiac function andgeometry. After the functional measurements the heart wasexcised for laboratory analysis.

Immunohistochemistry

Three weeks following MI, myocardial biopsies were obtainedfrom the infarct area, border area, and remote area andfixated in 4% formalin for 24 h before being embedded inparaffin. To determine capillary density, a lectin staining,delineating the endothelium, was performed followingantigen retrieval by boiling in 10 mM citrate buffer.Endogenous biotin was blocked using 0.1% avidin solution(DakoCytomation Biotin Blocking system X0590) for 15 min atroom temperature (RT) and subsequently with 0.01% biotinsolution (Dako) for 15 min at RT. After blocking with 3.0% BSAin PBS, the sections were incubated overnight with biotiny-lated BS-1 (Sigma, L3759) in a 1:300 dilution in PBS with 0.1%BSA and subsequently with streptavidin-HRPO (1:1000 in PBS)for 1 h at RT. Finally, the sections were incubated for 30 minin 3-amino-9-ethylcarbazole (AEC) and stained with hema-toxylin before being embedded. Capillary density wasexpressed as the number of capillaries per square millimetertissue. Quantification of collagen density was performedusing picrosirius red staining with circularly polarized lightand digital image microscopy after conversion into gray-value images, as described before (Timmers et al., 2007b).

Myocardial infarct size

After excision of the heart, the LV was isolated and cut into 5slices from apex to base. To discriminate infarct tissue fromviable myocardium, the slices were incubated in 1%triphenyltetrazolium chloride (TTC, Sigma-Aldrich Chemi-cals, Zwijndrecht, Netherlands) in 37 C Srensen buffer(13.6 g/L KH2PO4+17.8 g/L Na2H PO42H2O, pH 7.4) for15 min. All slices were scanned from both sides and in eachslide the infarct area was compared to total area using digitalplanimetry software. After correction for the weight of theslices, infarct size was calculated as a percentage of the LV.

Cardiac function

Midpapillary short axis epicardial echocardiography wasperformed before coronary artery ligation, 1 h after and3 weeks after coronary artery ligation (Prosound SSD-5000,5 MHz probe UST-5280-5, Aloka Holding Europe AG, Zug,Switzerland). Wall thickness (WT) of the infarct area and leftventricular internal area (LVia) were measured at end-diastole (ED) and end-systole (ES). Systolic wall thickening(SWT) was calculated as (WT(ES) WT(ED))/WT(ED) * 100 (%)and fractional area shortening (FAS) as (LVia(ED) LVia(ES))/LVia(ED) * 100 (%). Left ventricular (LV) pressurevolumeloops were measured using a conductance catheter asdescribed previously (Timmers et al., 2009). Diastolicchamber stiffness was quantified by means of linearregression of the end-diastolic pressurevolume relationship(Sagawa, 1981).

Transcoronary transplantation of progenitor cells after myocardial

Cleutjens, J.P., Blankesteijn, W.M., Daemen, M.J., Smits, J.F.,

improve remodelling, contractile function, perfusion, or infarct

213size in a swine model of large myocardial infarction. Eur. Heart J.29, 17721782.

Erbs, S., Linke, A., Schachinger, V., Assmus, B., Thiele, H., Diederich,K.W., Hoffmann, C., Dimmeler, S., Tonn, T., Hambrecht, R.,Zeiher, A.M., Schuler, G., 2007. Restoration of microvascularfunction in the infarct-related artery by intracoronary transplan-tation of bone marrow progenitor cells in patients with acutemyocardial infarction: the Doppler Substudy of the Reinfusion ofEnriched Progenitor Cells and Infarct Remodeling in AcuteMyocardial Infarction (REPAIR-AMI) trial. Circulation 116, 366374.

Gnecchi, M., He, H., Liang, O.D., Melo, L.G., Morello, F., Mu, H.,Noiseux, N., Zhang, L., Pratt, R.E., Ingwall, J.S., Dzau, V.J.,2005. Paracrine action accounts for marked protection ofischemic heart by Akt-modified mesenchymal stem cells. Nat.Med. 11, 367368.1999. The infarcted myocardium: simply dead tissue, or a livelytarget for therapeutic interventions. Cardiovasc. Res. 44,232241.

de Silva, R., Raval, A.N., Hadi, M., Gildea, K.M., Bonifacino, A.C.,Yu, Z.X., Yau, Y.Y., Leitman, S.F., Bacharach, S.L., Donahue,R.E., Read, E.J., Lederman, R.J., 2008. Intracoronary infusion ofautologous mononuclear cells from bone marrow or granulocytecolony-stimulating factor-mobilized apheresis product may notinfarction. N. Engl. J. Med. 355, 12221232.Burlew, B.S., Weber, K.T., 2002. Cardiac fibrosis as a cause of

diastolic dysfunction. Herz 27, 9298.Caplan, A.I., Dennis, J.E., 2006. Mesenchymal stem cells as trophic

mediators. J. Cell. Biochem. 98, 10761084.Data analysis

The data were collected and analyzed in a blinded fashion.Data are presented as meanSE. Mortality was comparedusing Fisher's exact test. Capillary sprouting was comparedusing a one-way ANOVA and Bonferroni post hoc tests. A two-way ANOVA with Bonferroni post hoc tests was used tocompare capillary density and collagen density in themyocardial tissue. Cardiac functional parameters that werealso measured before MI (baseline) were compared using atwo-way ANOVA for repeated measures and Bonferroni posthoc tests. For cardiac functional parameters that weremeasured only 3 weeks after MI, Student's t test was used.

Acknowledgments

This work was supported by the Netherlands Heart Founda-tion [Grant 2005T022, L.T.] and an Internationalization grantby the University Medical Center Utrecht, the Netherlands[D.P.V.K.].

Appendix A. Supplementary data

Supplementary data to this article can be found online atdoi:10.1016/j.scr.2011.01.001.

References

Assmus, B., Honold, J., Schachinger, V., Britten, M.B., Fischer-Rasokat, U., Lehmann, R., Teupe, C., Pistorius, K., Martin, H.,Abolmaali, N.D., Tonn, T., Dimmeler, S., Zeiher, A.M., 2006.Hashemi, S.M., Ghods, S., Kolodgie, F.D., Parcham-Azad, K., Keane,M., Hamamdzic, D., Young, R., Rippy, M.K., Virmani, R., Litt, H.,Wilensky, R.L., 2008. A placebo controlled, dose-ranging, safetystudy of allogenic mesenchymal stem cells injected by endo-myocardial delivery after an acute myocardial infarction. Eur.Heart J. 29, 251259.

Henry, T.D., Annex, B.H., McKendall, G.R., Azrin, M.A., Lopez, J.J.,Giordano, F.J., Shah, P.K., Willerson, J.T., Benza, R.L., Berman,D.S., Gibson, C.M., Bajamonde, A., Rundle, A.C., Fine, J.,McCluskey, E.R., 2003. The VIVA trial: vascular endothelialgrowth factor in ischemia for vascular angiogenesis. Circulation107, 13591365.

Janssens, S., Dubois, C., Bogaert, J., Theunissen, K., Deroose, C.,Desmet, W., Kalantzi, M., Herbots, L., Sinnaeve, P., Dens, J.,Maertens, J., Rademakers, F., Dymarkowski, S., Gheysens, O., VanCleemput, J., Bormans, G., Nuyts, J., Belmans, A., Mortelmans, L.,Boogaerts, M., Van de Werf, F., 2006. Autologous bone marrow-derived stem-cell transfer in patients with ST-segment elevationmyocardial infarction: double-blind, randomised controlled trial.Lancet 367, 113121.

Kinnaird, T., Stabile, E., Burnett, M.S., Shou, M., Lee, C.W., Barr,S., Fuchs, S., Epstein, S.E., 2004. Local delivery of marrow-derived stromal cells augments collateral perfusion throughparacrine mechanisms. Circulation 109, 15431549.

Korff, T., Augustin, H.G., 1998. Integration of endothelial cells inmulticellular spheroids prevents apoptosis and induces differen-tiation. J. Cell Biol. 143, 13411352.

Lai, R.C., Arslan, F., Lee, M.M., Sze, N.S., Choo, A., Chen, T.S.,Salto-Tellez, M., Timmers, L., Lee, C.N., El Oakley, R.M.,Pasterkamp, G., de Kleijn, D.P., Lim, S.K., 2010. Exosomesecreted by MSC reduces myocardial ischemia/reperfusioninjury. Stem Cell Res. 4, 214222.

Lian, Q., Lye, E., Suan Yeo, K., KhiaWay Tan, E., Salto-Tellez, M., Liu,T.M., Palanisamy,N., ElOakley, R.M., Lee, E.H., Lim,B., Lim, S.K.,2007. Derivation of clinically compliant MSCs from CD105+, CD24-differentiated human ESCs. Stem Cells 25, 425436.

Losordo, D.W., Dimmeler, S., 2004. Therapeutic angiogenesis andvasculogenesis for ischemic disease, part II, cell-based therapies.Circulation 109, 26922697.

Lunde, K., Solheim, S., Aakhus, S., Arnesen, H., Abdelnoor, M.,Egeland, T., Endresen, K., Ilebekk, A., Mangschau, A., Fjeld, J.G.,Smith, H.J., Taraldsrud, E., Grogaard, H.K., Bjornerheim, R.,Brekke, M., Muller, C., Hopp, E., Ragnarsson, A., Brinchmann, J.E.,Forfang, K., 2006. Intracoronary injection of mononuclear bonemarrow cells in acute myocardial infarction. N. Engl. J. Med. 355,11991209.

Menasche, P., Alfieri, O., Janssens, S., McKenna, W., Reich-enspurner, H., Trinquart, L., Vilquin, J.T., Marolleau, J.P.,Seymour, B., Larghero, J., Lake, S., Chatellier, G., Solomon, S.,Desnos, M., Hagege, A.A., 2008. The Myoblast AutologousGrafting in Ischemic Cardiomyopathy (MAGIC) trial: first ran-domized placebo-controlled study of myoblast transplantation.Circulation 117, 11891200.

Meyer, G.P., Wollert, K.C., Lotz, J., Pirr, J., Rager, U., Lippolt, P.,Hahn, A., Fichtner, S., Schaefer, A., Arseniev, L., Ganser, A.,Drexler, H., 2009. Intracoronary bone marrow cell transfer aftermyocardial infarction: 5-year follow-up from the randomized-controlled BOOST trial. Eur. Heart J. 24, 29782984.

Moelker, A.D., Baks, T., van den Bos, E.J., van Geuns, R.J., deFeyter, P.J., Duncker, D.J., van der Giessen, W.J., 2006.Reduction in infarct size, but no functional improvement afterbone marrow cell administration in a porcine model ofreperfused myocardial infarction. Eur. Heart J. 27, 30573064.

Pittenger, M.F., Martin, B.J., 2004. Mesenchymal stem cells andtheir potential as cardiac therapeutics. Circ. Res. 95, 920.

Sagawa, K., 1981. The end-systolic pressure-volume relation of theventricle: definition, modifications and clinical use. Circulation63, 12231227.

Schachinger, V., Erbs, S., Elsasser, A., Haberbosch, W., Ham-brecht, R., Holschermann, H., Yu, J., Corti, R., Mathey, D.G.,Hamm, C.W., Suselbeck, T., Assmus, B., Tonn, T., Dimmeler,S., Zeiher, A.M., 2006. Intracoronary bone marrow-derivedprogenitor cells in acute myocardial infarction. N. Engl. J. Med.355, 12101221.

Simons, M., Annex, B.H., Laham, R.J., Kleiman, N., Henry, T.,Dauerman, H., Udelson, J.E., Gervino, E.V., Pike, M., White-house, M.J., Moon, T., Chronos, N.A., 2002. Pharmacologicaltreatment of coronary artery disease with recombinant fibroblastgrowth factor-2: double-blind, randomized, controlled clinicaltrial. Circulation 105, 788793.

Strauer, B.E., Brehm, M., Zeus, T., Kostering, M., Hernandez, A.,Sorg, R.V., Kogler, G., Wernet, P., 2002. Repair of infarctedmyocardium by autologous intracoronary mononuclear bonemarrow cell transplantation in humans. Circulation 106,19131918.

Sze, S.K., de Kleijn, D.P., Lai, R.C., Khia Way, E., 2007. Tan, H.Zhao, K.S. Yeo, T.Y. Low, Q. Lian, C.N. Lee, W. Mitchell, R.M. ElOakley, S.K. Lim, Elucidating the secretion proteome of humanembryonic stem cell-derived mesenchymal stem cells. Mol. Cell.Proteomics 6, 16801689.

Timmers, L., Lim, S.K., Arslan, F., Armstrong, J.S., Hoefer, I.E.,Doevendans, P.A., Piek, J.J., El Oakley, R.M., Choo, A., Lee, C.N.,Pasterkamp, G., de Kleijn, D.P., 2007a. Reduction of myocardialinfarct size by humanmesenchymal stemcell conditionedmedium.Stem Cell Res. 1, 129137.

Timmers, L., Sluijter, J.P., Verlaan, C.W., Steendijk, P., Cramer,M.J.,Emons, M., Strijder, C., Grundeman, P.F., Sze, S.K., Hua, L., Piek,J.J., Borst, C., Pasterkamp, G., de Kleijn, D.P., 2007b. Cycloox-ygenase-2 inhibition increases mortality, enhances left ventricularremodeling, and impairs systolic function after myocardialinfarction in the pig. Circulation 115, 326332.

Timmers, L., Henriques, J.P., de Kleijn, D.P., Devries, J.H.,Kemperman, H., Steendijk, P., Verlaan, C.W., Kerver, M.,Piek, J.J., Doevendans, P.A., Pasterkamp, G., Hoefer, I.E.,2009. Exenatide reduces infarct size and improves cardiacfunction in a porcine model of ischemia and reperfusion injury.J. Am. Coll. Cardiol. 53, 501510.

van Laake, L.W., Passier, R., denOuden, K., Schreurs, C.,Monshouwer-Kloots, J.,Ward-vanOostwaard, D., van Echteld, C.J., Doevendans,P.A., Mummery, C.L., 2009. Improvement of mouse cardiacfunction by hESC-derived cardiomyocytes correlates with vascular-ity but not graft size. Stem Cell Res. 3, 106112.

Wolf, D., Reinhard, A., Seckinger, A., Gross, L., Katus, H.A., Hansen,A., 2009. Regenerative capacity of intravenous autologous,allogeneic and human mesenchymal stem cells in the infarctedpig myocardium-complicated by myocardial tumor formation.Scand. Cardiovasc. J. 43, 3945.

Yousef, M., Schannwell, C.M., Kostering, M., Zeus, T., Brehm, M.,Strauer, B.E., 2009. The BALANCE Study: clinical benefit andlong-term outcome after intracoronary autologous bone marrowcell transplantation in patients with acute myocardial infarction.J. Am. Coll. Cardiol. 53, 22622269.

214 L. Timmers et al.

Human mesenchymal stem cell-conditioned medium improves cardiac function following myocardial infarctionIntroductionResultsMSC-CM stimulates angiogenesis in vitroMSC-CM treatment after infarction increases capillary density and decreases collagen densityMSC-CM preserves cardiac function following MI

DiscussionMethodsMSC-CM preparationIngenuity pathway analysisHUVEC spheroid assayAnimalsSample size calculationPig study designMyocardial infarction and operational procedureImmunohistochemistryMyocardial infarct sizeCardiac functionData analysis

AcknowledgmentsSupplementary dataReferences