Overexpression of Vascular Endothelial Growth Factor-B in Mouse Heart Alters Cardiac Lipid...

Kari AlitaloLeif C. Andersson, Eero Mervaala, Ilmo E. Hassinen, Seppo Ylä-Herttuala, Matej Oresic andLeskinen, Riikka Kivelä, Teemu Helkamaa, Mari Merentie, Michael Jeltsch, Karri Paavonen,

Terhi Karpanen, Maija Bry, Hanna M. Ollila, Tuulikki Seppänen-Laakso, Erkki Liimatta, HannaLipid Metabolism and Induces Myocardial Hypertrophy

Overexpression of Vascular Endothelial Growth Factor-B in Mouse Heart Alters Cardiac

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 2008 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/CIRCRESAHA.108.1784592008;103:1018-1026; originally published online August 28, 2008;Circ Res.

http://circres.ahajournals.org/content/103/9/1018World Wide Web at:

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

http://circres.ahajournals.org/content/suppl/2008/08/28/CIRCRESAHA.108.178459.DC1.htmlData Supplement (unedited) at:

http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Researchin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from

http://circres.ahajournals.org/content/103/9/1018

http://circres.ahajournals.org/content/suppl/2008/08/28/CIRCRESAHA.108.178459.DC1.html

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

http://www.lww.com/reprints

http://circres.ahajournals.org//subscriptions/

http://circres.ahajournals.org/

















Overexpression of Vascular Endothelial Growth Factor-B inMouse Heart Alters Cardiac Lipid Metabolism and Induces

Myocardial HypertrophyTerhi Karpanen,* Maija Bry,* Hanna M. Ollila, Tuulikki Seppanen-Laakso, Erkki Liimatta,

Hanna Leskinen, Riikka Kivela, Teemu Helkamaa, Mari Merentie, Michael Jeltsch, Karri Paavonen,Leif C. Andersson, Eero Mervaala, Ilmo E. Hassinen, Seppo Yla-Herttuala, Matej Oresic, Kari Alitalo

Abstract—Vascular endothelial growth factor (VEGF)-B is poorly angiogenic but prominently expressed in metabolicallyhighly active tissues, including the heart. We produced mice expressing a cardiac-specific VEGF-B transgene via the�-myosin heavy chain promoter. Surprisingly, the hearts of the VEGF-B transgenic mice showed concentric cardiachypertrophy without significant changes in heart function. The cardiac hypertrophy was attributable to an increased sizeof the cardiomyocytes. Blood capillary size was increased, whereas the number of blood vessels per cell nucleusremained unchanged. Despite the cardiac hypertrophy, the transgenic mice had lower heart rate and blood pressure thantheir littermates, and they responded similarly to angiotensin II–induced hypertension, confirming that the hypertrophydoes not compromise heart function. Interestingly, the isolated transgenic hearts had less cardiomyocyte damage afterischemia. Significantly increased ceramide and decreased triglyceride levels were found in the transgenic hearts. Thiswas associated with structural changes and eventual lysis of mitochondria, resulting in accumulation of intracellularvacuoles in cardiomyocytes and increased death of the transgenic mice, apparently because of mitochondrial lipotoxicityin the heart. These results suggest that VEGF-B regulates lipid metabolism, an unexpected function for an angiogenicgrowth factor. (Circ Res. 2008;103:1018-1026.)

Key Words: VEGF-B � cardiac hypertrophy � cardiac metabolism � fatty acids � mitochondria

Members of the vascular endothelial growth factor(VEGF) family, currently comprising 5 mammalian

proteins, are major regulators of blood and lymphatic vesseldevelopment and growth. VEGF is essential for vasculogen-esis and angiogenesis, whereas VEGF-C is necessary forlymphangiogenesis. Although not required for embryonicdevelopment, placenta growth factor and VEGF-D arelikely to play more subtle roles in the control of angiogen-esis and lymphangiogenesis, or function under pathologi-cal conditions.1,2

VEGF-B exists as 2 isoforms generated by alternativesplicing. VEGF-B167 has a heparin-binding carboxyl terminus,whereas VEGF-B186 contains a hydrophobic carboxyl termi-nus, is O-glycosylated, and proteolytically processed.3 Bothisoforms bind to VEGF receptor (VEGFR)-1 and neuropilin-1but not to the major mitogenic endothelial cell receptorsVEGFR-2 or VEGFR-3.4,5

VEGF-B has a wide tissue distribution, being most abun-dant in the myocardium, skeletal and vascular smooth mus-cle, as well as in brown adipose tissue.6 Mice lackingVEGF-B are viable and fertile and display only a mildphenotype in the heart. This is manifested as an atrialconduction abnormality characterized by a prolonged PQinterval in 1 strain (C57Bl background),7 or as a smaller heartsize with impaired recovery after myocardial ischemia inanother (129/SvJ or 129/SvJ�C57Bl/6J background).8 Fur-thermore, VEGF-B has also been implicated in pathologicalvascular changes in inflammatory arthritis9 and in protectingthe brain from ischemic injury.10 However, the ability ofVEGF-B to stimulate angiogenesis directly is poor in manytissues. VEGF-B did not stimulate vessel growth whendelivered into muscle or periadventitial tissue via adenoviralvectors.11,12 On the contrary, VEGF-B overexpressed inendothelial cells of transgenic mice was able to potentiate,

Original received October 7, 2007; resubmission received April 29, 2008; revised resubmission received July 14, 2008; accepted August 20, 2008.From the Molecular/Cancer Biology Laboratory and Ludwig Institute for Cancer Research, Biomedicum Helsinki, and Department of Pathology,

Haartman Institute (T.K., M.B., H.O., R.K., M.J., K.P., K.A.) and Institute of Biomedicine (T.H., E.M.), Department of Pharmacology, University ofHelsinki and Helsinki University Central Hospital; VTT Technical Research Centre of Finland (T.S.-L., M.O.), Espoo; Departments of MedicalBiochemistry and Molecular Biology (E.L., I.E.H.) and Pharmacology and Toxicology (H.L.), University of Oulu; A. I. Virtanen Institute (M.M.,S.Y.-H.), Department of Biotechnology and Molecular Medicine, University of Kuopio; and Department of Pathology (L.C.A.), Haartman Institute,University of Helsinki, Finland. Present address for T.K.: Hubrecht Institute, Utrecht, The Netherlands.

*Both authors contributed equally to this work.Correspondence to Kari Alitalo, MD, PhD, Molecular/Cancer Biology Laboratory, Biomedicum Helsinki, PO Box 63, FI-00014 Helsinki, Finland.

E-mail [email protected]© 2008 American Heart Association, Inc.

Circulation Research is available at http://circres.ahajournals.org DOI: 10.1161/CIRCRESAHA.108.178459

1018 by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from


rather than initiate, angiogenesis.13 And unlike VEGF,VEGF-B did not increase vascular permeability.8,13

Here we have studied the effects of VEGF-B in the skinand heart, by overexpression via the keratin-14 and �-myosinheavy chain (�MyHC) promoters, respectively. VEGF-Bpromoted very little angiogenesis. However, mice overex-pressing VEGF-B in the heart gradually developed cardiachypertrophy with increased cardiomyocyte size and, surpris-ingly, had lower blood pressure, lower heart rate and largerblood vessels in the myocardium than their wild-type (WT)littermates. The hypertrophy was associated with changes inmitochondrial morphology and eventual lysis of mitochon-dria. We found marked changes in the lipidomic profiles ofthe transgenic hearts, including increased amounts of cer-amides. These results suggest that VEGF-B is capable ofaltering cardiac lipid metabolism, with overexpression ulti-mately leading to cardiomyopathy.

Materials and MethodsAn expanded Materials and Methods section is available in theonline data supplement at http://circres.ahajournals.org.

Generation of Transgenic MiceTo generate keratin-14–VEGF-B transgenic mice, DNA from thehuman VEGF-B gene was cloned into the keratin-14 expressionvector (provided by Dr Elaine Fuchs, The Rockefeller University,New York, NY). To generate the heart-specific transgene, cDNAencoding human VEGF-B167

14 was ligated into the �MyHC promoterexpression vector (a gift from Dr Jeffrey Robbins, Children’sHospital, Cincinnati, Ohio). Expression cassettes were injected intofertilized mouse oocytes of FVB background. Mice were PCR-ge-notyped using tail DNA. All mouse experiments were approved bythe Provincial State Office of Southern Finland and carried out inaccordance with institutional guidelines.

ImmunohistochemistrySections of paraformaldehyde-fixed, paraffin-embedded mousehearts were stained with rat antibodies against mouse plateletendothelial cell adhesion molecule (Pecam)-1 (BD Pharmingen)using biotinylated anti-rat IgG (Vector Laboratories) and TSA-kit(NEN Life Sciences) for detection. The number and size of Pecam-1–stained vessels were quantified using ImageJ program (NIH).Results are presented as averages�SD.

Acetone-fixed cryosections were stained with antibodies againstcollagen IV (CosmoBio), laminin-1 (a gift from Paivi Liesi, Univer-sity of Helsinki, Helsinki, Finland), VEGFR-1 (a gift from Dr BronekPytowski, ImClone Systems), neuropilin-1 or VEGF-B (R&D Systems).AlexaFluor488-conjugated anti-rabbit or AlexaFluor594-conjugatedanti-goat (Molecular Probes) antibodies were used for detection. Cardi-omyocyte sizes were quantified from laminin-1-stained photomicro-graphs using Axiovision program (Zeiss).

EchocardiographyTransthoracic echocardiography was performed using Acuson Ultra-sound System (Sequoia512) and a 15-MHz linear transducer. Micewere anesthetized with fentanyl citrate, fluanisone, and midazolam,and normal body temperature was maintained.

Telemetric AnalysisHeart rate and mean arterial pressure were recorded from left carotidartery using telemetric implants as previously described,15 exceptxylazine was used for anesthesia. Data collection was started on thefifth day when the normal circadian rhythm of the mice had returned.Data were gathered day and night for 2 weeks, every 5 minutes for10 seconds.

Measurements of Mitochondrial Redox State andTolerance of Short-Term Ischemia in Isolated,Perfused HeartsAorta was cannulated immediately after cervical dislocation andperfusion commenced. The heart was enclosed in a thermostatic,light-tight chamber connected to a spectrophotometer–fluorometer.Left ventricular pressure was monitored using a saline-filled cannulaconnected to a Statham P231D pressure transducer and SP1400pressure monitor. Parameters were calculated with custom-designedsoftware. Venous effluent from the heart was collected in 1-minutealiquots and lactate dehydrogenase (LDH) washout measured. Thevalues are presented as means�SE.

Angiotensin II–Induced Pressure OverloadAngiotensin (Ang) II was administered via subcutaneous Alzet-minipumps (Scanbur AB) for 1 week (0.1 mg/kg/h). Echocardiog-raphy was performed using VEVO 779 Ultrasound System(VisualSonics).

Electron MicroscopyLeft ventricular samples were fixed with glutaraldehyde, postosmi-cated, and embedded in epon. Regions of interest were examinedwith JEOL 1400 EX Transmission Electron Microscope.

Lipidomic Analysis of Heart TissueLipid extracts of heart tissue, mixed with a labeled lipid standardmixture, were analyzed on a Waters Q-Tof Premier mass spectrom-eter combined with Acquity Ultra Performance liquid chromatogra-phy. Results are presented as means�SEM.

Results

VEGF-B is Minimally Angiogenic in the SkinWe generated transgenic mice overexpressing VEGF-B in

the epidermis under the keratin-14 promoter (Figure IA andIB in the online data supplement). Dermal vessel density andbranching were only slightly increased (�20%; supplementalFigure IC), thus VEGF-B only minimally promoted angio-genesis in the skin.

VEGF-B Encoded by a Heart-Specific Transgeneis Associated With Pericellular MatrixAs the heart is one of the major sites of VEGF-B expressionand the primary organ affected on deletion of the VEGF-Bgene, we studied the effects of VEGF-B in this tissue. Forthis, we overexpressed the heparin-binding VEGF-B isoform,VEGF-B167, under the �-myosin heavy chain (�MyHC)promoter (Figure 1A). The VEGF-B167 transgene was ex-pressed at 6- and 8-fold compared to the endogenousVEGF-B gene in the myocardium of the 2 transgenic mouselines studied, as analyzed by Northern and Western blotting(supplemental Figure ID and data not shown), and by immu-nohistochemistry (Figure 1B). Much of the extracellularVEGF-B167 protein colocalized with the pericellular matrixcomponent collagen IV (Figure 1B). Intracellular stainingwas also observed in denatured samples (Figure 1B, bottomrow, middle). VEGFR-1 and neuropilin-1 distribution par-tially overlapped with that of collagen IV and laminin (Figure1B and 1C and data not shown), suggesting that extracel-lular VEGF-B was bound to heparan sulfates in the basallaminae of myocardial blood vessels and cardiomyocytes.

Karpanen et al Effects of VEGF-B in the Heart 1019

by guest on August 23, 2014http://circres.ahajournals.org/Downloaded from


The Hearts of the �MyHC-VEGF-B167 MiceDisplay Cardiac Hypertrophy But no SignificantChange in Systolic Function

The hearts of the transgenic mice appeared bigger andmore muscular than the hearts of their WT littermates (Figure2A). In cross-sections the walls of the left ventricle werethicker and the cardiomyocytes larger (Figure 2B and 2C).Hypertrophy of the cardiomyocytes was confirmed by cellarea quantification from sections stained for laminin-1 (Fig-ure 2C and 2D). The heart to body weight ratio wassignificantly higher in the transgenic mice when comparedwith their littermates (5.93�1.36 mg/g for TG males,average�SD, n�16, as compared to 4.95�0.78 mg/g for WTmales, n�16, P�0.02; Figure 2E). An increased heart tobody weight ratio was also observed in female transgenicmice (data not shown).

Echocardiography confirmed the left ventricular hypertro-phy in the �MyHC-VEGF-B167 mice as measured by thethickness of the left ventricular posterior wall (Figure 2F) andinterventricular septum (Figure 2G) in diastole and by thefunctional left ventricular mass (Figure 2H). No change was

observed in diastolic left ventricular diameter (Table). Thecardiac hypertrophy was more severe in male mice andincreased with age (Figure 2F through 2H). However, therewas no change in systolic function as measured by ejectionfraction and fractional shortening (Table).

Telemetric measurement revealed that the transgenic heartrate was significantly lower than the heart rate of WTlittermates (511�52 bpm for TG, n�3, versus 635�20 bpmfor WT mice, n�5, P�0.001; Figure 3A). Both systolic anddiastolic blood pressures were also lower in the transgenicmice (128�3.6 mm Hg versus 143�5.1 mm Hg in systoleand 94�3.7 mm Hg versus 109�5.5 mm Hg in diastole, forTG, n�3, and WT, n�5, mice, respectively, P�0.001; Figure3B and supplemental Figure II). However, �MyHC-VEGF-B167 mice under anesthesia had ventricular extrasystoles andwere more prone to arrhythmias (20/28 of TG versus 5/29 ofWT mice at 12 months of age). We also observed an increasein the death rate of the transgenic mice (Figure 3C).

The Hearts of the �MyHC-VEGF-B167 Mice HaveLarger Blood VesselsImmunohistochemical staining for Pecam-1 revealed a lowerblood vessel density in the hearts of �MyHC-VEGF-B167

mice than in WT littermates (236�33 versus 297�38 vesselsper microscopic field, n�13, P�0.001; Figure 4A through4C). However, when calculated per number of cardiomyocytenuclei in the analyzed area, the number of vessels in trans-genic hearts was comparable to that of WT hearts(1.68�0.19, n�7, versus 1.95�0.24, n�5, P�0.05, Figure4D). The blood vessels in the myocardium of the �MyHC-VEGF-B167 mice were larger than those of their littermates(average vessel area 1223�244 versus 944�130 pixels,P�0.002, Figure 4E). This was associated with an increase inthe number of endothelial cells per vessel cross-section(0.64�0.11, n�6, versus 0.51�0.09, n�4, P�0.004). Thus,the total area covered by blood vessels in the myocardium ofthe transgenic mice was similar to that of controls(277236�44831 versus 274278�37929 pixels per micro-scopic field, P�0.5; Figure 4F). This indicates that althoughVEGF-B167 overexpression did not affect the vessel numberper cardiomyocyte, the vessel area per cardiomyocyte wasincreased.

Mitochondrial Redox StateCentral for energy metabolism and heart function is fatty acidoxidation coupled to ATP production in mitochondria. Themyocardial content of cytochrome c oxidase (cytochromeaa3), a marker of the mitochondrial respiratory chain, wassimilar in transgenic and control hearts. The steady stateoxidation–reduction level of cytochrome aa3 tended to beslightly more reduced in transgenic hearts than in controls(Figure 5A and 5B), which was also the case for thefluorescent flavoproteins. These mainly reflect the activity ofmitochondrial lipoamide dehydrogenase, which is in equilib-rium with the NADH/NAD pool16 and can therefore be usedas a redox indicator for NADH/NAD in the mitochondrialmatrix. When the cytochrome aa3 reduction percentage wasplotted against mechanical work output during electric pacingto varying heart rates, the plots were linear without a

A

BVEGF-B Collagen IV Merge

αMyHC promoter (~6.1 kb) hGH pA (~0.6 kb)VEGF-B167 (~0.6 kb)

TG

VEGFR-1 NRP-1 Laminin-1

WT

C

TG

WT TG WTMerge Merge

TG TG

VEGF-B

ααα

Figure 1. Expression of the �MyHC-VEGF-B167 transgene-encoded protein. A, Schematic structure of the �MyHC-VEGF-B167 transgene. hGH pA indicates human growth hormone poly-adenylation signal. B, Immunohistochemical staining of heartsections from the �MyHC-VEGF-B167 mice (TG) and WT litter-mates with antibodies against VEGF-B (red) and collagen IV(green). The bottom images in the middle and far right representmerged figures of permeabilized samples showing intracellularVEGF-B staining. Blue, DAPI staining of nuclei. C, Immunohisto-chemical staining of heart sections from the �MyHC-VEGF-B167

mice (top row) and littermates (bottom row) with antibodiesagainst VEGFR-1 (red), neuropilin-1 (NRP-1) (red), and laminin-1(green).

1020 Circulation Research October 24, 2008



significant difference between the slopes of transgenic andcontrol hearts. Cytochrome aa3 reduction at the intercept(representing 0 work output) was similar for both transgenicand control hearts, indicating no difference in the mitochon-drial coupling efficiency (data not shown).

Ischemia and Pressure Overload ToleranceLDH washout from the transgenic hearts after a 20-minuteperiod of ischemia was significantly less than from WTcontrols (Figure 5C), suggesting less cellular damage onischemia-reperfusion. Despite this indication of myocardialprotection, the relative work output (pressuredevelopment�heart rate) during the first 10 minutes ofreperfusion tended to be lower in the transgenic hearts(64�2% versus 90�18%, for TG, n�3, and WT, n�6,hearts, respectively; Figure 5D). After 20 minutes of reper-fusion this difference disappeared, the work output being82�7% (n�3) in the �MyHC-VEGF-B167 hearts and 85�18(n�3) in the controls (Figure 5D).

Both transgenic and WT mice responded to Ang II–induced pressure overload by decreasing left ventricular

internal diameter and stroke volume (WT untreated34.26�4.47 �L, WT Ang II–treated 28.01�4.48 �L, n�6,P�0.05; TG untreated 39.13�5.17 �L, TG Ang II–treated28.12�4.47 �L, n�6, P�0.01). Furthermore, the Ang II–treated transgenic mice did not show any functional impair-ment when compared to Ang II–treated WT littermates inechocardiographic analysis, further confirming that theVEGF-B induced hypertrophy does not compromise heartfunction.

Mitochondrial Changes in the TransgenicHearts Are Associated With AbnormalLipid AccumulationDespite normal cytochrome concentrations in the transgenichearts, electron microscopic analysis revealed progressivechanges in the structure of cardiomyocyte mitochondria in 2-and 6-month-old transgenic mice (Figure 6A through 6D andsupplemental Figure III), including swelling and alteredmorphology of the cristae (white arrows, Figure 6C), appar-ently leading to mitochondrial lysis (black arrows). These

TG

WT

F

1

**

***

LV

PW

d (

mm

)

12 mo3 mo

C

E

*

0.2

0.4

0.6

% H

eart

wei

gh

t / b

od

y w

eig

ht

TG WT

1

12 mo3 mo

***

***

IVS

d (

mm

)

G

100

200**

***

Fu

nct

ion

al L

V m

ass

(mg

)12 mo3 mo

H

A B D750

500

250

TG WT

Car

dio

myo

cyte

siz

e (µ

m2 )

***

HE Laminin-1

TGWT

Figure 2. The hearts of the �MyHC-VEGF-B167 mice display cardiac hypertrophy but no significant change in systolic function. A, Ven-tral view of the hearts from male �MyHC-VEGF-B167 mice (top) and WT littermates (bottom). B, Hematoxylin/eosin–stained heart sec-tions from a male �MyHC-VEGF-B167 mouse (top) and WT littermate (bottom). Scale bar�1 mm. C, Hematoxylin/eosin–(HE) and lami-nin-1–stained heart sections from a male �MyHC-VEGF-B167 mouse (top) and littermate (bottom). Scale bar�50 �m. D, Size of themyocytes in the �MyHC-VEGF-B167 and WT hearts (n�10 in both groups). E, Heart weight percentage in male �MyHC-VEGF-B167 and WTmice. F through H, Thickness of the left ventricular posterior wall (LVPWd) (F) and of the interventricular septum (IVSd) (G) in diastole andfunctional left ventricular mass (H) measured by echocardiography in 3- and 12-month (mo)-old male �MyHC-VEGF-B167 and WT mice.*P�0.02; **P�0.01; ***P�0.001.




changes accumulated in older transgenic mice, and intracel-lular vacuoles were visible in the transgenic hearts in lightmicroscopy at 1 year of age (arrows in Figure 6E). Mitochon-drial changes suggested altered energy metabolism in thecardiomyocytes, which use mainly fatty acids as an energysource.

Previous studies have indicated that increased myocardialfatty acid uptake or decreased fatty acid utilization leads tocardiac hypertrophy followed by the accumulation of toxiclipid species.17,18 We therefore carried out lipidomic profilingof the hearts. This analysis revealed a significant increase inde novo synthesized ceramide levels in the transgenic hearts,as measured by the ceramide to sphingomyelin ratio (Figure7A and supplemental Figure IV), whereas the triacylglycerolswere decreased, as normalized to total phospholipid concen-trations (Figure 7B) or to tissue weight (data not shown).

Ceramides are known for their toxicity and accumulate inpathological conditions such as diabetes mellitus or adiposetissue inflammation.19,20 Thus, ceramide accumulation wasthe probable reason for the observed mitochondrial damageand increased mortality of the �MyHC-VEGF-B167 transgenicmice.

DiscussionIn the present study, we show that elevated expression ofVEGF-B induces very little angiogenesis in the skin or heart

Table. Echocardiographic Analysis of Three- andTwelve-Month-Old Transgenic Mice and Wild-Type Littermates

Parameter

Three Months Twelve Months

TG WT TG WT

IVSd (mm) 1.1�0.1* 0.9�0.1 1.4�0.1* 1.1�0.1

IVSs (mm) 2.0�0.2* 1.9�0.1 2.2�0.2* 2.0�0.1

LVd (mm) 3.9�0.4 4.0�0.2 3.9�0.3 3.6�0.3

LVs (mm) 1.6�0.3 1.7�0.4 1.8�0.5 1.4�0.6

LVPWd (mm) 1.1�0.1* 0.9�0.1 1.4�0.2* 1.0�0.1

LVPWs (mm) 1.8�0.2* 1.6�0.2 2.0�0.3 1.8�0.2

MWTd (mm) 1.1�0.1* 0.9�0.1 1.4�0.1* 1.1�0.1

MWTs (mm) 1.9�0.2* 1.7�0.1 2.1�0.2* 1.9�0.1

LV FS (%) 60�7 59�8 53�12 63�14

LV EF (%) 93�3 92�4 87�8 93�6

LV mass (mg) 187�41* 137�29 253�40* 148�37

HR (beats/min) 296�40* 416�90 324�28* 377�43

Results are expressed as means�SD. *P�0.05 vs WT (n�8–10 in eachgroup). d indicates diastole; s, systole; IVS, interventricular septum; LV, leftventricle; LVPW, left ventricle posterior wall; MWT, mean wall thickness; FS,fractional shortening; EF, ejection fraction; HR, heart rate.

20

60

100

140

Systolic Diastolic

Blo

od p

ress

ure

(mm

Hg)

***

***

200

400

600

Hea

rt r

ate

(bea

ts/m

in)

***A B

TG WT

TGWT

6 12

wild type

males

females

Sur

viva

l (%

tota

l)

Age (months)

10

2030405060

708090100

C Figure 3. Heart rate, bloodpressure, and survival of thetransgenic and WT mice. Tele-metric measurements of theheart rate (A), as well as sys-tolic and diastolic blood pres-sure (B), of 5-month-oldhomozygous �MyHC-VEGF-B167 male mice and WT con-trols. The values are presentedas averages�SD. ***P�0.001.C, Transgenic male (n�24) andfemale (n�27) mice were fol-lowed for 12 months, and inci-dence of spontaneous deathwas recorded as a function oftime and compared to the pre-viously reported WT FVB/Nsurvival (60% at 24 monthsaccording to Harlan SpragueDawley Inc). Data are present-ed as a Kaplan–Meier survivalcurve.

F

100

200

300

Tot

al P

ecam

-1 p

ositi

ve v

esse

l are

a(1

03 pi

xels

/ m

icro

scop

ic fi

eld)

TG WT

100

200

300

Pec

am-1

pos

itive

ves

sels

/m

icro

scop

ic fi

eld

C

***

TG WT

E

400

800

1200

Ves

sel a

rea

(pix

els)

***

TG WT

A B

0.4

0.8

1.2

1.6

2.0

TG WT

Ves

sels

/ N

ucle

us

D

Figure 4. Altered pattern of myocardial capillaries in the�MyHC-VEGF-B167 hearts. Sections of the hearts from male�MyHC-VEGF-B167 (A) and WT littermate (B) mice were stainedwith antibodies against Pecam-1, and the number of Pecam-1positive vessels was quantified (C). D through F, Number ofPecam-1–positive vessels per cell nucleus (D), average areaof Pecam-1 positive vessels (E), and total area covered byPecam-1 positive vessels (F) in �MyHC-VEGF-B167 and WT lit-termates. The values are presented as averages�SD***P�0.002.




but, instead, hypertrophy of the cardiac muscle withoutsignificant changes in pumping function. Because the trans-genic blood pressure and heart rate were decreased, thehypertrophy was not a consequence of increased workload

but, rather, caused by an intrinsic metabolic effect induced bycardiomyocyte expression of VEGF-B. Although the trans-genic mice had larger myocardial capillaries, their toleranceto cardiac ischemia and reperfusion, as well as to AngII–induced myocardial stress was not compromised. How-ever, evidence was obtained of altered cardiac lipid metabo-lism, ceramide accumulation, gradual mitochondrial damage,and increased death of the transgenic mice. The unexpectedmetabolic and trophic effects of VEGF-B in the myocardium

Wor

k ou

tput

(%

of b

asal

)

20

60

100

D

0-10 20-25

Time period (min)

Cum

ulat

ive

LDH

was

hout

(U/g

wet

wei

ght)

.0.5

1.0

1.5

2.0

***

TG WT

WTTG

TGWT

A

B

8070605040

Time (min)

Fla

vin

oxid

atio

nC

ytoc

hrom

e aa

3 re

duct

ion

1.0

0.6

0.2

-0.2

-0.05

0.05

0.15WTTG

WTTG

30

C Figure 5. Effects of short-termglobal ischemia and reperfu-sion on the hearts of �MyHC-VEGF-B167 mice. Oxidation–reduction state of cytochromec oxidase (cytochrome aa3) (A,values on the y axis presentfluorescence increase 465 to540 nm) and fluorescent fla-voproteins (B, values on the yaxis present fluorescencechange 605 to 530 nm) in�MyHC-VEGF-B167 and controlhearts during 20-minute globalischemia and reperfusion. Theblack horizontal bar depictsthe time period of ischemia,and, in both A and B, upwarddeflection indicates oxidation.Red indicates �MyHC-VEGF-B167 (n�7); blue, controls(n�10). The hatched area rep-resents �SE. C, CumulativeLDH washout from homozy-

gous �MyHC-VEGF-B167 and control mouse hearts during 30 minutes of reperfusion after 20 minutes of global ischemia. The values aremeans�SE from 6 and 10 independent experiments, respectively. ***P�0.005. D, Work output from �MyHC-VEGF-B167 and WT mousehearts during reperfusion after 20 minutes of global ischemia. The values are presented as means�SE from 3 and 6 experiments,respectively.

A B

C D

TG WT

FE

Figure 6. The �MyHC-VEGF-B167 mice display abnormal mito-chondria and vacuoles in cardiomyocytes. A through D, Trans-mission electron micrographs displaying mitochondrial morphol-ogy in 6-month-old �MyHC-VEGF-B167 (left) and WT (right)control hearts. Scale bar�2 �m. C and D, Magnified images ofboxed areas in A and B. E, Vacuoles inside cardiomyocytes in1.5-year-old �MyHC-VEGF-B167 mice, not observed in WT litter-mates (F). Hematoxylin/eosin staining. Scale bar�50 �m.

**

0.1

0.3

0.5

0.7

Ceramides

Cer

amid

e/S

phin

gom

yelin

A

Tria

cylg

lyce

rol/t

otP

L

0.06

0.12

0.18

**

Triacylglycerols

BTGWT

Figure 7. The hearts of the �MyHC-VEGF-B167 mice contain anincreased amount of ceramides but decreased levels of triacyl-glycerols. The concentration of ceramides to sphingomyelin (A)and triacylglycerols to total phospholipid levels (B) in female�MyHC-VEGF-B167 (n�8) and WT (n�9) mice are shown.**P�0.05. Values are presented as means�SEM.




and the absence of major angiogenic effects are striking for amember of the VEGF family.

Several studies, including our present results, indicate thatVEGF-B is not a major angiogenesis-inducing factor. Instead,VEGF-B may have the ability to potentiate neovasculariza-tion in pathological conditions.9 In this regard, VEGF-Bdiffers greatly from placenta growth factor, which also bindsto VEGFR-1 and neuropilin-1 and induces angiogenesis invarious tissues, including the heart.21,22 In the skin, VEGF-Boverexpression only slightly increased dermal capillary den-sity. In the VEGF-B overexpressing hearts, vessel densitywas decreased because of vessel displacement by hypertro-phic cardiomyocytes, whereas vessel diameter was greater sothat the total capillary area per cardiomyocyte was increased.VEGF-B staining decorated cardiomyocytes and the pericel-lular basement membranes, indicating that the transgene wasabundantly expressed, and consistent with the finding thatVEGF-B167 binds to pericellular heparan sulfate.14 The lack ofa significant angiogenic effect is consistent with earlierreports where VEGF-B was expressed in endothelial cells andshown to promote minimal angiogenesis.13

We detected a progressively increasing heart to bodyweight ratio in the transgenic mice, which was associatedwith increased thickness of the cardiac muscle wall. Thereason for the increased cardiac mass appeared to be hyper-trophy of the cardiomyocytes as determined by increased cellsize, and the transgenic hearts lacked areas of fibrosis oredema. In contrast to forms of human hypertrophic cardio-myopathy, parallel sarcomere orientation was preserved inthe transgenic hearts. Increased �-adrenergic stimulation andsystemic hypertension with associated work overload areknown to induce a hypertrophic response in the left ventricle,and angiogenesis is known to be essential for overload-induced adaptive cardiac hypertrophy, whereas inhibition ofangiogenesis leads to heart failure.23,24 The VEGF-B–inducedhypertrophy, however, was not related to these conditions as theheart rate and blood pressure were lower in the transgenic mice, andhypertrophy comprised all parts of the myocardium. Otherwise, theanatomy of the heart and the great vessels was normal. Themyocardial hypertrophy in the VEGF-B–overexpressing heartscontrasts with the loss-of-function phenotype of the VEGF-Bknockout mice, which have smaller hearts,8 and is consis-tent with the recent report that adenoviral VEGF-B treat-ment promotes cardiomyocyte hypertrophy after myocardialinfarction.25

We observed that intracellular vacuoles accumulate in thecardiomyocytes of the transgenic mice. With electron micros-copy these were found to be structures left behind fromdamaged mitochondria, suggesting that mitochondrial energymetabolism is disturbed in the VEGF-B–overexpressinghearts. As cardiomyocytes mainly use lipids as an energysource, it was logical to analyze their lipidomic profiles.Indeed, lipidomic analysis revealed significant changes in thelipid composition of the hearts and suggested a likely reasonfor the cardiac hypertrophy. Previous studies have indicatedthat excess transport of fatty acids can induce cardiac hyper-trophy, which converts to lipotoxicity when the mitochondriacannot use all of the available lipids.17 In particular, ceramideaccumulation is known to induce mitochondrial damage26

and to play a critical role in the pathogenesis of lipotoxiccardiomyopathy.27 The increased lipolysis of triacylglycerolsis the likely reason for their observed depletion in transgenichearts, and would also explain the elevated ceramide levels.Higher availability of fatty acids resulting from lipolysis wouldlead to increased demand for mitochondrial �-oxidation,whereas the excess pool of saturated long-chain fatty acidswould provide substrate for de novo ceramide synthesis.28

Cardiac hypertrophy has also been described followingoverexpression of long-chain acyl-coenzyme A synthetase inthe heart.29 In this case, toxic lipid species accumulated in thehearts. Altered lipid metabolism seems to be the likely causefor the cardiac hypertrophy in the VEGF-B transgenic hearts,and, via increased ceramide accumulation, this would grad-ually lead to cardiomyocyte damage.

Ceramides are also involved in triggering macroautoph-agy.30 The intracellular vacuoles in the transgenic heartssuggest ongoing autophagy, which is a normal process in theheart for removal of damaged cytosolic components.31 Thisprocess is linked both to cardioprotection and lipolysis. Amajor portion of myocardial intracellular triacylglycerol hy-drolysis is lysosomal,32 and lysosomes are involved in auto-phagosome formation.31 Moreover, increasing the autophagiccapacity of cardiomyocytes decreased apoptosis induction onischemia/reperfusion.33 These 2 processes, autophagy and lipol-ysis, both related to lysosomal activities, may link the presentobservations of protection against ischemic damage, a decreasein myocardial triacylglycerols, and an increase in ceramide.There appears to be a fine balance between the proapoptotic andprotective actions of ceramide, because ceramide has beenshown to attenuate hypoxic cell death.34 Elevated levels ofceramide could thus at least partly account for the improved cellsurvival after ischemia in �MyHC-VEGF-B167 hearts.

Hearts isolated from �MyHC-VEGF-B167 mice showedsome protection against cellular damage after ischemia/reperfusion, as evidenced by less LDH release. Nevertheless,there was no significant difference between the �MyHC-VEGF-B167 mice and controls in the mechanical work output25 minutes after the beginning of reperfusion, although,during the first 10 minutes, there seemed to be a lag inattaining the basal work output. The causal connectionbetween the subdued commencement of contractions on reper-fusion and the decrease in LDH release is unclear. However, inisolated hearts, lower pressure during commencement of reper-fusion has been found to protect against injury.35

We monitored the redox state of mitochondria duringvariable workload and substrate availability, but no system-atic effects of the VEGF-B167 overexpression were found(E.L. and I.E.H., unpublished observations, 2004). The re-sponse of the redox state of myocardial cytochrome aa3 toworkload transitions was normal, and the redox level atextrapolated zero workload was similar in the �MyHC-VEGF-B167 and control hearts, indicating no difference inmitochondrial energy coupling or basal metabolic rate. Inter-estingly however, VEGF-B expression is correlated with highmetabolic activity in tissues, suggesting that its expression isassociated with energy metabolism. A recent study has alsoindicated that the metabolic sensor peroxisome proliferator-ac-tivated receptor-� coactivator 1� (PGC-1�) regulates VEGF




expression and angiogenesis in ischemic tissues.36 These resultsshed new light on an important interplay between nutrientdemand, metabolism, and vascular supply.

The changes observed in the VEGF-B167 overexpressingcardiomyocytes may be triggered by receptor activation inendothelial cells, followed by an as yet undetermined signal-ing or metabolite transfer to the cardiomyocytes. The findingthat VEGF-B alters cardiomyocyte mitochondria and lipidmetabolites in the heart in the absence of overt vessel growthis a new finding for an angiogenic factor. Although, in ourcase, the excess of VEGF-B ultimately leads to tissuedamage, the potential of VEGF-B to regulate cardiomyocyteenergy metabolism during shorter treatment periods may turnout to be useful in attempts to salvage myocardial insuffi-ciency where the myocardial energy sources are defective.

AcknowledgmentsWe thank Christian Ehnholm for helpful discussion and RaimoTuuminen and Karl Lemstrom for help with quantitative histopathol-ogy, Ismo Virtanen and Fang Zhao with electron microscopy, JenniHuusko with echocardiography, Berndt Enholm with generation of�MyHC-VEGF-B167 mice, Seppo Kaijalainen with RT-PCR, JaanaRysä for providing cDNA samples, and Sandra Castillo for technicalassistance. We also thank the Molecular Imaging Unit formicroscope support.

Sources of FundingThis study was supported by NIH grant 5-R01-HL075183-02;Academy of Finland Council of Health grants 202852, 204312, and210599; and Novo Nordisk Foundation.

DisclosuresNone.

References1. Shibuya M, Claesson-Welsh L. Signal transduction by VEGF receptors in

regulation of angiogenesis and lymphangiogenesis. Exp Cell Res. 2006;312:549–560.

2. Karpanen T, Alitalo K. Molecular biology and pathology of lymphangio-genesis. Annu Rev Pathol. 2008;3:367–397.

3. Olofsson B, Pajusola K, von Euler G, Chilov D, Alitalo K, Eriksson U.Genomic organization of the mouse and human genes for vascular endo-thelial growth factor B (VEGF-B) and characterization of a second spliceisoform. J Biol Chem. 1996;271:19310–19317.

4. Olofsson B, Korpelainen E, Pepper MS, Mandriota SJ, Aase K, Kumar V,Gunji Y, Jeltsch MM, Shibuya M, Alitalo K, Eriksson U. Vascularendothelial growth factor B (VEGF-B) binds to VEGF receptor-1 andregulates plasminogen activator activity in endothelial cells. Proc NatlAcad Sci U S A. 1998;95:11709–11714.

5. Makinen T, Olofsson B, Karpanen T, Hellman U, Soker S, Klagsbrun M,Eriksson U, Alitalo K. Differential binding of vascular endothelial growthfactor B splice and proteolytic isoforms to neuropilin-1. J Biol Chem.1999;274:21217–21222.

6. Aase K, Lymboussaki A, Kaipainen A, Olofsson B, Alitalo K, ErikssonU. Localization of VEGF-B in the mouse embryo suggests a paracrinerole of the growth factor in the developing vasculature. Dev Dyn. 1999;215:12–25.

7. Aase K, von Euler G, Li X, Ponten A, Thoren P, Cao R, Cao Y, OlofssonB, Gebre-Medhin S, Pekny M, Alitalo K, Betsholtz C, Eriksson U.Vascular endothelial growth factor-B-deficient mice display an atrialconduction defect. Circulation. 2001;104:358–364.

8. Bellomo D, Headrick JP, Silins GU, Paterson CA, Thomas PS, GartsideM, Mould A, Cahill MM, Tonks ID, Grimmond SM, Townson S, WellsC, Little M, Cummings MC, Hayward NK, Kay GF. Mice lacking thevascular endothelial growth factor-B gene (Vegfb) have smaller hearts,dysfunctional coronary vasculature, and impaired recovery from cardiacischemia. Circ Res. 2000;86:e29–e35.

9. Mould AW, Tonks ID, Cahill MM, Pettit AR, Thomas R, Hayward NK,Kay GF. Vegfb gene knockout mice display reduced pathology andsynovial angiogenesis in both antigen-induced and collagen-inducedmodels of arthritis. Arthritis Rheum. 2003;48:2660–2669.

10. Sun Y, Jin K, Childs JT, Xie L, Mao XO, Greenberg DA. Increasedseverity of cerebral ischemic injury in vascular endothelial growth factor-B-deficient mice. J Cereb Blood Flow Metab. 2004;24:1146–1152.

11. Rissanen TT, Markkanen JE, Gruchala M, Heikura T, Puranen A,Kettunen MI, Kholova I, Kauppinen RA, Achen MG, Stacker SA, AlitaloK, Yla-Herttuala S. VEGF-D is the strongest angiogenic and lym-phangiogenic effector among VEGFs delivered into skeletal muscle viaadenoviruses. Circ Res. 2003;92:1098–1106.

12. Bhardwaj S, Roy H, Gruchala M, Viita H, Kholova I, Kokina I, AchenMG, Stacker SA, Hedman M, Alitalo K, Yla-Herttuala S. Angiogenicresponses of vascular endothelial growth factors in periadventitial tissue.Hum Gene Ther. 2003;14:1451–1462.

13. Mould AW, Greco SA, Cahill MM, Tonks ID, Bellomo D, Patterson C,Zournazi A, Nash A, Scotney P, Hayward NK, Kay GF. Transgenicoverexpression of vascular endothelial growth factor-B isoforms by en-dothelial cells potentiates postnatal vessel growth in vivo and in vitro.Circ Res. 2005;97:e60–e70.

14. Olofsson B, Pajusola K, Kaipainen A, von Euler G, Joukov V, Saksela O,Orpana A, Pettersson RF, Alitalo K, Eriksson U. Vascular endothelialgrowth factor B, a novel growth factor for endothelial cells. Proc NatlAcad Sci U S A. 1996;93:2576–2581.

15. Butz GM, Davisson RL. Long-term telemetric measurement of cardio-vascular parameters in awake mice: a physiological genomics tool.Physiol Genomics. 2001;5:89–97.

16. Kunz WS, Kunz W. Contribution of different enzymes to flavoproteinfluorescence of isolated rat liver mitochondria. Biochim Biophys Acta.1985;841:237–246.

17. Chiu HC, Kovacs A, Blanton RM, Han X, Courtois M, Weinheimer CJ,Yamada KA, Brunet S, Xu H, Nerbonne JM, Welch MJ, Fettig NM,Sharp TL, Sambandam N, Olson KM, Ory DS, Schaffer JE. Transgenicexpression of fatty acid transport protein 1 in the heart causes lipotoxiccardiomyopathy. Circ Res. 2005;96:225–233.

18. Sharma S, Adrogue JV, Golfman L, Uray I, Lemm J, Youker K, Noon GP,Frazier OH, Taegtmeyer H. Intramyocardial lipid accumulation in the failinghuman heart resembles the lipotoxic rat heart. FASEB J. 2004;18:1692–1700.

19. Kolak M, Westerbacka J, Velagapudi VR, Wagsater D, Yetukuri L,Makkonen J, Rissanen A, Hakkinen AM, Lindell M, Bergholm R,Hamsten A, Eriksson P, Fisher RM, Oresic M, Yki-Jarvinen H. Adiposetissue inflammation and increased ceramide content characterize subjectswith high liver fat content independent of obesity. Diabetes. 2007;56:1960–1968.

20. Medina-Gomez G, Gray SL, Yetukuri L, Shimomura K, Virtue S,Campbell M, Curtis RK, Jimenez-Linan M, Blount M, Yeo GS, Lopez M,Seppanen-Laakso T, Ashcroft FM, Oresic M, Vidal-Puig A. PPARgamma 2 prevents lipotoxicity by controlling adipose tissue expandabilityand peripheral lipid metabolism. PLoS Genet. 2007;3:e64.

21. Luttun A, Tjwa M, Moons L, Wu Y, Angelillo-Scherrer A, Liao F, NagyJA, Hooper A, Priller J, De Klerck B, Compernolle V, Daci E, Bohlen P,Dewerchin M, Herbert JM, Fava R, Matthys P, Carmeliet G, Collen D,Dvorak HF, Hicklin DJ, Carmeliet P. Revascularization of ischemictissues by PlGF treatment, and inhibition of tumor angiogenesis, arthritisand atherosclerosis by anti-Flt1. Nat Med. 2002;8:831–840.

22. Odorisio T, Schietroma C, Zaccaria ML, Cianfarani F, Tiveron C,Tatangelo L, Failla CM, Zambruno G. Mice overexpressing placentagrowth factor exhibit increased vascularization and vessel permeability.J Cell Sci. 2002;115:2559–2567.

23. Sano M, Minamino T, Toko H, Miyauchi H, Orimo M, Qin Y, Akazawa H,Tateno K, Kayama Y, Harada M, Shimizu I, Asahara T, Hamada H, TomitaS, Molkentin JD, Zou Y, Komuro I. p53-induced inhibition of Hif-1 causescardiac dysfunction during pressure overload. Nature. 2007;446:444–448.

24. Shiojima I, Sato K, Izumiya Y, Schiekofer S, Ito M, Liao R, Colucci WS,Walsh K. Disruption of coordinated cardiac hypertrophy and angiogenesiscontributes to the transition to heart failure. J Clin Invest. 2005;115:2108–2118.

25. Tirziu D, Chorianopoulos E, Moodie KL, Palac RT, Zhuang ZW, Tjwa M,Roncal C, Eriksson U, Fu Q, Elfenbein A, Hall AE, Carmeliet P, Moons L,Simons M. Myocardial hypertrophy in the absence of external stimuli isinduced by angiogenesis in mice. J Clin Invest. 2007;117:3188–3197.

26. Scarlatti F, Bauvy C, Ventruti A, Sala G, Cluzeaud F, Vandewalle A,Ghidoni R, Codogno P. Ceramide-mediated macroautophagy involves




inhibition of protein kinase B and up-regulation of beclin 1. J Biol Chem.2004;279:18384–18391.

27. Park TS, Hu Y, Noh HL, Drosatos K, Okajima K, Buchanan J, Tuinei J,Homma S, Jiang XC, Abel ED, Goldberg IJ. Ceramide is a cardiotoxin inlipotoxic cardiomyopathy. J Lipid Res. In press.

28. Sparagna GC, Hickson-Bick DL, Buja LM, McMillin JB. A metabolicrole for mitochondria in palmitate-induced cardiac myocyte apoptosis.Am J Physiol Heart Circ Physiol. 2000;279:H2124–H2132.

29. Chiu HC, Kovacs A, Ford DA, Hsu FF, Garcia R, Herrero P, Saffitz JE,Schaffer JE. A novel mouse model of lipotoxic cardiomyopathy. J ClinInvest. 2001;107:813–822.

30. Zheng W, Kollmeyer J, Symolon H, Momin A, Munter E, Wang E, KellyS, Allegood JC, Liu Y, Peng Q, Ramaraju H, Sullards MC, Cabot M,Merrill AH Jr. Ceramides and other bioactive sphingolipid backbones inhealth and disease: lipidomic analysis, metabolism and roles in membranestructure, dynamics, signaling and autophagy. Biochim Biophys Acta.2006;1758:1864–1884.

31. Gustafsson AB, Gottlieb RA. Recycle or die: the role of autophagy incardioprotection. J Mol Cell Cardiol. 2008;44:654–661.

32. Schoonderwoerd K, Broekhoven-Schokker S, Hulsmann WC, Stam H.Involvement of lysosome-like particles in the metabolism of endogenousmyocardial triglycerides during ischemia/reperfusion. Uptake and degra-dation of triglycerides by lysosomes isolated from rat heart. Basic ResCardiol. 1990;85:153–163.

33. Hamacher-Brady A, Brady NR, Gottlieb RA. Enhancing macroautophagyprotects against ischemia/reperfusion injury in cardiac myocytes. J BiolChem. 2006;281:29776–29787.

34. Lecour S, Van der Merwe E, Opie LH, Sack MN. Ceramide attenuateshypoxic cell death via reactive oxygen species signaling. J CardiovascPharmacol. 2006;47:158–163.

35. Bopassa JC, Vandroux D, Ovize M, Ferrera R. Controlled reperfusionafter hypothermic heart preservation inhibits mitochondrial permeabilitytransition-pore opening and enhances functional recovery. Am J PhysiolHeart Circ Physiol. 2006;291:H2265–H2271.

36. Arany Z, Foo SY, Ma Y, Ruas JL, Bommi-Reddy A, Girnun G, CooperM, Laznik D, Chinsomboon J, Rangwala SM, Baek KH, Rosenzweig A,Spiegelman BM. HIF-independent regulation of VEGF and angiogenesisby the transcriptional coactivator PGC-1alpha. Nature. 2008;451:1008–1012.




ATG WT

VEGF-B

GAPDH

K14-VEGF-B

TG WT

28S

16S

VEGF-B- mVEGF-B

!MyHC-VEGF-B

TG WTD

C

Karpanen et al., Online Figure I

B

TG WT

VE

GF

-B IH

CP

EC

AM

-1 W

M

25

50

Ve

ssels

/hp

f

TG WT

250

500

Bra

nch p

oin

ts/h

pf

TG WT

** **

20

60

100

140mmHg

Control FVB

aMHC-VEGF-B167

2 4 6 8 10 12 14 16

B

20

60

100

140

180

mmHg

2 4 6 8 10 12 14 16

A

day

Karpanen et al., Online Figure II

A B

DC

Karpanen et al., Online Figure III

****

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

18:0 20:0 22:0 24:0 24:1

TG

WT

Ceramide/Sphingomyelin

Karpanen et al., Online Figure IV

CIRCRESAHA/2008/178459/R1 1

Supplemental Figure Legends

Online Figure I. Expression of the transgenes and minimal angiogenic effects of VEGF-B on skin

vasculature. (A) The expression of the K14-VEGF-B transgene was analyzed by Northern blotting of

RNA isolated from the skin of the K14-VEGF-B and littermate control mice with a probe recognizing

VEGF-B mRNA. (B) The proper expression of the K14-VEGF-B transgene was confirmed by

immunohistochemistry of paraffin sections of the skin with antibodies against mouse VEGF-B. The

skin vasculature was analyzed by whole-mount staining of ear skin with antibodies against the

panendothelial marker Pecam-1. (C) Number of Pecam-1 stained vessels and their branching points per

microscopic high-power field (hpf). **, p < 0.05. (D) The expression of the !MyHC-VEGF-B167

transgene was analyzed by RT-PCR (left) and Northern blotting (right) of RNA isolated from the hearts

of !MyHC-VEGF-B167 and littermate control mice.

Online Figure II. Telemetric analysis of blood pressure. Day- and night-time systolic (A) and diastolic

(B) blood pressure of !MyHC-VEGF-B167 and wild type mice as a function of time. Note the

decreasing trend in the blood pressure after the probe insertion procedure (on day 0).

Online Figure III. Analysis of mitochondrial changes in the hearts of !MyHC-VEGF-B167 mice by

electron microscopy. Transmission electron micrographs displaying mitochondrial morphology in the

hearts of six-month-old !MyHC-VEGF-B167 (A) and wildtype (B) mice. (C) Magnification of boxed

area in A, featuring accumulating vacuoles and swollen mitochondria with altered cristae (arrows). (D)

Electron micrograph from a two-month-old !MyHC-VEGF-B167 mouse heart showing progressive

mitochondrial degeneration. Scale bar = 2 µm.


Online Figure IV. Levels of cardiac ceramides with different acyl chain lengths. The ratio of ceramides

to sphingomyelin in the hearts of female !MyHC-VEGF-B167 and wild type mice. **, p < 0.05.

Supplemental Results

VEGF-B is minimally angiogenic in transgenic mouse skin. Previous studies have shown that when

overexpressed under the K14 promoter in transgenic mouse skin, VEGF and PlGF induce an

angiogenic phenotype and VEGF-C and VEGF-D a lymphangiogenic phenotype1-4. We generated

transgenic mice overexpressing VEGF-B under the K14 promoter and confirmed transgene expression

by Northern blotting and immunohistochemistry (Supplemental Figure 1, A-B). The mice developed

normally, appeared healthy, were fertile, and had a normal life span. The skin of the ears, snout and

paws had a normal pinkish blush, and hair growth over the body was normal, without any signs of

spontaneous skin lesions or abnormalities, for up to 1.5 years of age. VEGF-B overexpression only

slightly increased vessel density (42 ± 5 vessels/mm2 in K14-VEGF-B mice versus 35 ± 4 vessels/mm2

in WT mice, n = 10, p < 0.05) and vessel branching (392 ± 74 branchpoints/mm2 in K14-VEGF-B mice

versus 331 ± 54 branchpoints/mm2 in WT mice, n = 7, p < 0.05) in the skin (Supplemental Figure 1C).

These vessels had a normal morphology, size, patterning and mural cell coating as analyzed by

immunostaining for smooth muscle alpha-actin (SMA; data not shown). Thus, VEGF-B only

minimally promoted angiogenesis when overexpressed in the skin.


Supplemental Materials and Methods

Generation of the transgenic mice. To generate the K14-VEGF-B transgenic mice, DNA from the

human VEGF-B gene corresponding to nucleotides 745-5059 of Genbank accession number AF468110

was cloned into the K14 expression vector (kindly provided by Dr. Elaine Fuchs5), and one non-

initiating upstream ATG was mutated into a GTG. The expression cassette was excised from the vector

backbone and injected into fertilized mouse oocytes of FVB background. The mice were PCR-

genotyped using tail DNA with the primers 5’-TCTCCCAGCCTGATGCCCCT-3’ and 5’-

GGACTTGGTGCTGCCCAGTG-3’.

To generate a heart specific transgene, the recessed 3'-ends of the EcoRI fragment from a human

VEGF-B167/pCRII vector6 were filled in with the Klenow fragment of DNA polymerase I and ligated to

the Sal I-opened and Klenow filled-in !MyHC promoter expression vector7 (a kind gift from Dr.

Jeffrey Robbins). The expression cassette was excised from the vector backbone with BamHI and

injected into fertilized mouse oocytes of FVB background. The mice were genotyped by PCR of tail

DNA with the primer pair 5’-TCTCCCAGCCTGATGCCCCT-3’ and 5’-

GCCATGTGTCACCTTCGCAG-3’.

The expression of the transgene was confirmed by subjecting total RNA to Northern blotting and

hybridization with a 32P-labeled VEGF-B cDNA probe (nucleotides 1-382, Genbank Accession No

U48800), by RT-PCR using the primer pair 5’-TCTCCCAGCCTGATGCCCCT-3’ and 5’-

CTAAGCCCCGCCCTTGGC-3’, and by immunohistochemistry and Western blotting. The phenotypes

were analyzed from two !MyHC-VEGF-B founder lines expressing the transgene. All experiments


involving mice were approved by the Provincial State Office of Southern Finland and carried out in

accordance with institutional guidelines.

Immunohistochemistry. For quantification of blood vessels in the heart, 7 µm sections of 4%

paraformaldehyde (PFA) fixed, paraffin-embedded mouse tissues were deparaffinized, rehydrated, and

pretreated with trypsin (0.25 mg/ml trypsin in 9 mM CaCl2, 50 mM Tris pH 7.8) for 30 minutes at

37°C, and the endogenous peroxidase was inactivated with 3% H2O2 in methanol. After blocking, the

slides were incubated with purified rat anti-mouse Pecam-1 (MEC13.3, BD Pharmingen) at a

concentration of 0.625 µg/ml overnight at +4°C, washed, and incubated with biotinylated rabbit anti-rat

IgG at a concentration of 1.7 µg/ml (BA-4001, Vector Laboratories) for 30 min at room temperature.

The slides were stained using the TSA-kit (NEN Life Sciences/PerkinElmer Life and Analytical

Sciences) according to the manufacturer’s instructions. The number and size of the Pecam-1 stained

vessels were quantified from four photomicrographs per mouse photographed with an Olympus AX70

microscope and DP50 Camera (Olympus) using the ImageJ program (National Institutes of Health).

The number of Pecam-1 positive endothelial cell nuclei per blood vessel cross-section was quantified

using Pecam-1/hematoxylin staining (four photomicrographs per mouse). Results are expressed as

average ± SD.

8 µm cryosections of frozen tissues were fixed in acetone, blocked, and incubated with rabbit anti-

mouse collagen IV (Cosmo Bio), rabbit anti-laminin-1 antiserum (a kind gift from Päivi Liesi), anti-rat

VEGFR-1 (5B12, ImClone Systems Incorporated), goat anti-rat neuropilin-1 (AF566, R&D Systems)

or goat anti-human VEGF-B (AF751, R&D Systems) primary antibodies for 2 hours. Intracellular

VEGF-B staining was observed when using microwave pretreatment of slides in DakoCytomation Low


pH Target Retrieval Solution. AlexaFluor488-conjugated anti-rabbit and AlexaFluor594-conjugated

anti-goat (Molecular Probes) antibodies were used for detection. Zeiss Axioplan2 epifluorescence

microscope (Zeiss) was used for imaging. Cardiomyocyte sizes in transgenic and control hearts were

quantified from four correspondingly located laminin-1 stained high-power field photomicrographs per

heart using the Axiovision program (Zeiss).

Mouse ears were dissected for whole-mount immunostaining, fixed in 4% PFA, and incubated with

antibodies against Pecam-1, followed by detection with peroxidase-labeled IgGs (Dako). For

quantification of branch points and vessel densities, 24 optical fields (400x) per mouse, sampled

randomly in the center and periphery of the ear, were photographed with a digital camera. For

quantification of vessel densities, prints of the images were divided by horizontal lines into five equal

sectors and all crossing points of the vessels with the horizontal lines were counted. Vessel branching

was quantified from the same prints by counting the branch points. Values are presented as average ±

SD.

Echocardiography. Transthoracic echocardiography was performed using an Acuson Ultrasound

System (SequoiaTM 512) and a 15-MHz linear transducer (15L8) (Acuson) or a VEVO 779 Ultrasound

System (VisualSonics). Mice were anesthetized with fentanyl citrate 8 µg/10 g, fluanisone 250 µg/10 g

and midazolam 125 µg/10 g. Normal body temperature was maintained during the examination with a

warming pad and lamp.

Using two-dimensional imaging, a short axis view of the left ventricle at the level of the papillary

muscles and two-dimensionally guided M-mode recordings through the anterior and posterior walls of


the left ventricle were obtained. Left ventricular end-systolic (LVDs) and end-diastolic (LVDd)

dimensions as well as thickness of the interventricular septum (IVS) and left ventricular posterior wall

(LVPW) in diastole (d) and systole (s) were measured from the M-mode tracings. Left ventricular

shortening fraction (LVFS) was calculated from the M-mode left ventricular dimensions using the

following equation: LVFS (%) = [(LVDd-LVDs) / LVDd] x 100. Ejection fraction (EF) was also

calculated from the M-mode dimensions using the equation: EF (%) = [(LVDd)3 – (LVDs)3 / LVDd3] x

100. Functional left ventricular mass was calculated using the equation: LVmass = 1.055 x [(IVSd +

LVd + LVPWd) 3 – LVDd3]. All the measurements were made from three subsequent cycles and

calculated as the mean of these three measurements. Results are expressed as mean ± SD.

Function of the mitral, tricuspidal, aortic and pulmonary valves was evaluated by using color flow

mapping and pulsed Doppler.

Telemetric analysis. During the telemetric measurements, the mice were housed one per cage in a

thermostatically controlled environment at 23±2ºC and relative humidity of 50-70%. Mice were

allowed free access to chow and drinking water, available ad libitum. The room was artificially

illuminated from 7am to 7pm. The heart rate and mean arterial pressure were recorded from the left

carotid artery using TA11PA-C20 telemetric implants as previously described, except that xylazine

instead of promazine was used for anaesthesia8. Mice were allowed a recovery period after surgery and

data collection was started on the fifth day when the normal daily circadian rhythm of the mice had

returned. The data were sampled continuously day and night for two weeks, every 5 min for 10 s. The

presented values, average ± SD, are from a 24-hour period during the fifth day after the start of the

measurements. No significant changes were noted in the telemetric variables, 24 h mean daily values of

the blood pressure, or heart rate after the fifth day.


Measurements of mitochondrial redox state and tolerance of short-term ischemia in the isolated,

perfused hearts. The mice were sacrificed by decapitation after cervical dislocation. The aorta was

immediately cannulated and perfusion commenced in situ with ice-cold perfusion fluid. The heart was

dissected out and perfusion continued with Krebs-Henseleit buffer consisting of 118.5 mmol/L NaCl,

4.7 mmol/L KCl, 2.5 mmol/L CaCl2, 0.25 mmol/L Ca-EDTA, 1.2 mmol/L MgSO4, 1.2 mM KH2PO4, 25

mmol/L NaHCO3 and 10 mmol/L glucose (pH 7.4, 37°C) with a pressure of 100 cm H2O (9.81 kPa) and

the medium gassed with O2/CO2 (19:1).

The heart was enclosed in a thermostatic, light-tight chamber equipped with fiber optics for

simultaneous epicardial readout of reflectance spectrum changes at 605 and 630 nm and fluorescence at

520 nm with excitation at 460 nm by means of a three-channel spectrophotometer-fluorometer9. The

photometric data were collected via a data acquisition card (PCI-6024E, National Instruments) and

stored on a PC at two-second intervals. Left ventricular pressure was monitored through the ventricular

wall by inserting a saline-filled Teflon cannula connected to a Statham P231D pressure transducer

linked to a Statham SP1400 pressure monitor. The pressure wave signal was led to a Lab-PC+ data

acquisition card (National Instruments) and heart function parameters (heart rate, peak systolic

pressure, diastolic pressure and peak pressure development) were calculated on-line with custom

designed software and stored on a PC at 4 s intervals. Coronary flow was measured during the

experiments by using a drop counter with an analog output. Oxygen consumption was calculated from

the arteriovenous concentration difference multiplied by coronary flow. Venous effluent from the heart

was collected in 1 min aliquots and lactate dehydrogenase washout measured as described10. The values

are presented as mean ± SE.


Angiotensin II-induced pressure overload. AngII was administered to mice via subcutaneous osmotic

Alzet minipumps (Scanbur AB) for 1 week (0,1 mg/kg/h). Mice were sacrificed one week later.

Echocardiographic analysis of cardiac dimensions was carried out before the treatment and before

sacrifice using a Vevo 779 high-resolution in vivo imaging system (VisualSonics).

Electron microscopy. Tissue samples from the left ventricle were fixed with 2% glutaraldehyde,

postosmicated and embedded in epon. Semithin sections were stained with toluidine blue, and on the

basis of initial analysis under light microscope, regions of interest were selected for thin 100 nm

sectioning and analysis using a JEOL 1400 EX Transmission Electron Microscope equipped with

Morada CCD Camera (Olympus SIS).

Lipidomic analysis of heart tissue. Hearts were perfused with PBS upon excision, dissected, and snap-

frozen. Corresponding pieces (5-9 mg) were mixed with an internal standard mixture containing 1-

heptadecanoyl-sn-glycero-3-phosphocholine, N-(heptadecanoyl)-sphing-4-enine, 1,2-diheptadecanoyl-

sn-glycero-3-phosphocholine, 1,2-diheptadecanoyl-sn-glycero-3-phosphoethanolamine and 1,2,3-

triheptadecanoyl-sn-glycerol at a concentration level of 0.5-1 µg/sample and 200 µl

chloroform:methanol (2:1). The tissues were homogenized with grinding balls in a Mixer MILL at 25

Hz for 2 min and 50 µl of 0.9% NaCl was added. The samples were vortexed for 2 min and after 30

min standing, centrifuged at 10 000 rpm for 3 min. The labeled lipid standard mixture was added into

the separated lipid extracts (1 µg/sample) before ultra performance liquid chromatography-mass

spectrometric analysis.

Lipid extracts were analyzed on a Waters Q-Tof Premier mass spectrometer (Waters, Inc., Milford,

MA) combined with an Acquity Ultra Performance LCTM (UPLC). The column (at 50°C) was an


Acquity UPLCTM

BEH C18 10 x 50 mm with 1.7 µm particles. The solvent system included A.

ultrapure water (1% 1M NH4Ac, 0.1% HCOOH) and B. LC/MS grade acetonitrile/isopropanol (5:2, 1%

1M NH4Ac, 0.1% HCOOH). The gradient started from 65% A / 35% B, reached 100% B in 6 min and

remained there for the next 7 min. There was a 5 min re-equilibration step before the next run. The

flow rate was 0.200 ml/min and the injected amount 1.0 µl (Acquity Sample Organizer, Waters, Inc.,

Milford, MA). Reserpine was used as the lock spray reference compound. The lipid profiling was

carried out using ESI+ mode and the data was collected at mass range of m/z 300-1200 with scan

duration of 0.2 sec. The data was processed by using MZmine software v. 0.60 11 and the lipid

identification was based on an internal spectral library12. Results are presented as mean ± SEM.

Statistical analysis of data. Two-group comparison was made with the Student’s two-tailed t-test by

using the method of summary measures13 when appropriate.

CIRCRESAHA/2008/178459/R1 10

References

1. Detmar M, Brown LF, Schon MP, Elicker BM, Velasco P, Richard L, Fukumura D, Monsky W,

Claffey KP, Jain RK. Increased microvascular density and enhanced leukocyte rolling and

adhesion in the skin of VEGF transgenic mice. J Invest Dermatol. 1998;111(1):1-6.

2. Odorisio T, Schietroma C, Zaccaria ML, Cianfarani F, Tiveron C, Tatangelo L, Failla CM,

Zambruno G. Mice overexpressing placenta growth factor exhibit increased vascularization and

vessel permeability. J Cell Sci. 2002;115(Pt 12):2559-2567.

3. Jeltsch M, Kaipainen A, Joukov V, Meng X, Lakso M, Rauvala H, Swartz M, Fukumura D, Jain

RK, Alitalo K. Hyperplasia of lymphatic vessels in VEGF-C transgenic mice. Science.

1997;276(5317):1423-1425.

4. Veikkola T, Jussila L, Makinen T, Karpanen T, Jeltsch M, Petrova TV, Kubo H, Thurston G,

McDonald DM, Achen MG, Stacker SA, Alitalo K. Signalling via vascular endothelial growth

factor receptor-3 is sufficient for lymphangiogenesis in transgenic mice. EMBO J.

2001;20:1223-1231.

5. Vassar R, Rosenberg M, Ross S, Tyner A, Fuchs E. Tissue-specific and differentiation-specific

expression of a human K14 keratin gene in transgenic mice. Proc. Natl. Acad. Sci. USA.

1989;86:1563-1567.

6. Olofsson B, Pajusola K, Kaipainen A, von Euler G, Joukov V, Saksela O, Orpana A, Pettersson

RF, Alitalo K, Eriksson U. Vascular endothelial growth factor B, a novel growth factor for

endothelial cells. Proc. Natl Acad. Sci. USA. 1996;93:2576-2581.

7. Jeltsch M. Functional analysis of VEGF-B and VEGF-C. Master's Thesis. University of

Helsinki, Finland; 1997. (http://ethesis.helsinki.fi/julkaisut/mat/bioti/pg/jeltsch/)

CIRCRESAHA/2008/178459/R1 11

8. Butz GM, Davisson RL. Long-term telemetric measurement of cardiovascular parameters in

awake mice: a physiological genomics tool. Physiol Genomics. 2001;5(2):89-97.

9. Hassinen IE. Reflectance spectrophotometric and surface fluorometric methods for measuring

the redox state of nicotinamide nucleotides and flavins in intact tissues. Methods Enzymol.

1986;123:311-320.

10. Bergmeyer H, Bernt E. Lactat-Dehydrogenase. UV-Test mit Pyruvat und NADH. Methoden der

enzymatischen Analyse. Bergmeyer HU (Ed.) Verlag Chemie, Weinberg, Germany. 1970:533-

538.

11. Katajamaa M, Oresic M. Processing methods for differential analysis of LC/MS profile data.

BMC Bioinformatics. 2005;6:179.

12. Yetukuri L, Katajamaa M, Medina-Gomez G, Seppanen-Laakso T, Vidal-Puig A, Oresic M.

Bioinformatics strategies for lipidomics analysis: characterization of obesity related hepatic

steatosis. BMC Syst Biol. 2007;1:12.

13. Matthews JN, Altman DG, Campbell MJ, Royston P. Analysis of serial measurements in

medical research. Bmj. 1990;300(6719):230-235.

Overexpression of Vascular Endothelial Growth Factor-B in Mouse Heart Alters Cardiac Lipid...

Documents

Transcript of Overexpression of Vascular Endothelial Growth Factor-B in Mouse Heart Alters Cardiac Lipid...