Caveolin-1 Deficiency Increases Cerebral Ischemic Injurycircres.ahajournals.org/content/circresaha/100/5/721.full.pdf · Caveolin-1 Deficiency Increases Cerebral Ischemic Injury

Caveolin-1 Deficiency Increases Cerebral Ischemic InjuryJean-Francois Jasmin,* Samit Malhotra,* Manjeet Singh Dhallu, Isabelle Mercier,

Daniel M. Rosenbaum, Michael P. Lisanti

Abstract—Caveolins (Cav), the principal structural proteins of the caveolar domains, have been implicated in thepathogenesis of ischemic injury. Indeed, changes in caveolin expression and localization have been reported in renal andmyocardial ischemia. Genetic ablation of the Cav-1 gene in mice was further shown to increase the extent of ischemicinjury in a model of hindlimb ischemia. However, the role of Cav-1 in the pathogenesis of cerebral ischemia remainsunknown. Immunoblot and immunofluorescence analyses of rat brains subjected to middle cerebral artery occlusionrevealed marked increases in endothelial Cav-1 and Cav-2 protein levels. To directly assess the functional role ofcaveolins in the pathogenesis of cerebral ischemic injury, we next investigated the effects of cerebral ischemia incaveolin knockout (KO) mice. Interestingly, Cav-1 KO mice showed a marked increase of cerebral volume of infarction,as compared with wild-type and Cav-2 KO mice. Immunofluorescence analyses showed an increased number ofproliferating endothelial cells in wild-type ischemic brains, as compared with Cav-1 KO ischemic brains. Immunoblotanalyses of wild-type ischemic brains showed an increase in endothelial nitric oxide synthase protein levels. Conversely,the protein levels of endothelial nitric oxide synthase remained unchanged in Cav-1 KO ischemic brains. TUNELanalysis also showed increased apoptotic cell death in Cav-1 KO ischemic brains, as compared with wild-type ischemicbrains. Our findings indicate cerebral ischemia induces a marked increase in endothelial Cav-1 and Cav-2 protein levels.Importantly, genetic ablation of the Cav-1 gene in mice results in increased cerebral volume of infarction.Mechanistically, Cav-1 KO ischemic brains showed impaired angiogenesis and increased apoptotic cell death. (CircRes. 2007;100:721-729.)

Key Words: caveolin � cerebral ischemia � angiogenesis � apoptosis

Caveolae are small vesicular invaginations of the plasmamembrane that have been implicated in endocytosis,

vesicular trafficking, and signal transduction.1–3 Caveolinproteins (Cav) represent the principal structural proteins ofthe caveolar domains.4,5 The caveolin gene family consists ofthree distinct genes, namely Cav-1, -2 and -3.4–7 Cav-1 andCav-2 are often coexpressed and particularly abundant inendothelial cells, smooth muscle cells, fibroblasts and epithe-lial cells.6,8 On the other hand, Cav-3 appears to be muscle-specific and is, thus, solely expressed in cardiac, skeletal andsmooth muscle cells.7,9

Most of the proteins sequestered within caveolar domainspossess caveolin-binding motifs and, consequently, interactwith the caveolin proteins. Interestingly, caveolin proteinsappear as negative regulators of many of these associatedsignaling molecules.10–13 For instance, Cav-1 is well knownfor its inhibitory interaction with the endothelial nitric oxidesynthase (eNOS).12,13 Indeed, the direct interaction of eNOSwith the Cav-1 scaffolding domain (residues 82 to 101) wasshown to inhibit eNOS activity.12,13 The generation of Cav-1

knockout (KO) mice strongly supports the Cav-1-mediatednegative regulation of eNOS.14,15 As a matter of fact, Cav-1KO mice showed decreased vascular tone as well as micro-vascular hyperpermeability secondary to eNOS hyperactiva-tion.14,15 Interestingly, treatment of those mice with nitro-L-arginine methyl ester (L-NAME), a NOS inhibitor,successfully reversed the microvascular hyperpermeabilityphenotype, thus supporting the inhibitory actions of Cav-1 oneNOS activity.15 Therefore, Cav-1 appears as a key regulatorof vascular permeability, vascular tone, as well as angiogen-esis.14–16 Interestingly, although Cav-1 can negatively regu-late NO production, a well known pro-angiogenic factor,Cav-1 KO mice actually show an impaired angiogenic re-sponse to exogenous stimuli.16 Indeed, Cav-1 KO miceshowed reduced vessel infiltration and density in a model ofexogenous tumor cell injection.16 Accordingly, knock-downof the Cav-1 gene expression by antisense oligonucleotideswas shown to dramatically reduce capillary-like tube forma-tion in human umbilical vein endothelial cells (HUVEC).17

Conversely, overexpression of Cav-1 via adenoviral gene

Original received August 2, 2006; revision received January 17, 2007; accepted January 25, 2007.From the Departments of Molecular Pharmacology and Medicine (J.-F.J., M.P.L.), Albert Einstein College of Medicine, Bronx, NY; Department of

Cancer Biology (J.-F.J., I.M., M.P.L.), Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, Pa; Department of Neurology (S.M., M.S.D.,D.M.R.), Albert Einstein College of Medicine, Bronx, NY; Department of Neurology (S.M., D.M.R.), SUNY Downstate Medical Center, Brooklyn, NY.

*Both authors contributed equally to this work.Correspondence to Michael P. Lisanti MD, PhD, Kimmel Cancer Center, Thomas Jefferson University, 233 South 10th Street, Bluemle Building, Room

933B, Philadelphia, Pa, 19107. E-mail [email protected]© 2007 American Heart Association, Inc.

Circulation Research is available at http://circres.ahajournals.org DOI: 10.1161/01.RES.0000260180.42709.29

721

by guest on June 30, 2018http://circres.ahajournals.org/

Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from

http://circres.ahajournals.org/







delivery was shown to accelerate endothelial cell differenti-ation and tubule formation, as well as to increase the numberof capillary-like tubules, using human microvascular endo-thelial cells (HMEC-1) as a model system.18 Importantly,Cav-1 expression was considered essential for the develop-ment of collateral vessels in a mouse model of hindlimbischemia.19 Indeed, Cav-1 KO mice subjected to femoralartery/vein ligation failed to recover a functional vasculature,and in some cases even lost their entire leg.19

Caveolar domains have also been implicated in the com-partmentalization of signaling molecules involved in apopto-sis. For instance, the caspase-3 proenzyme and its activatedcounterpart were both shown to localize within cardiacendothelial caveolae.20 Interestingly, like numerous signalingmolecules, caveolar localization was suggested to maintaincaspase-3 in an inactive state.20 Indeed, disruption of caveo-lae structure was reported to increase staurosporine-inducedcaspase-3 activity.20 Accordingly, ablation of the Cav-1 genein the TRAMP (transgenic adenocarcinoma of mouse pros-tate) mouse model was previously shown to result in in-creased apoptotic cell death.21

Given their implications in angiogenesis and apoptosis,caveolin proteins may well act as key regulators of thepatho-physiological processes of ischemic injury. Accord-ingly, changes in caveolin protein expression and localizationhave previously been reported in ischemic acute renal failureand myocardial ischemia/reperfusion.22,23 Importantly, thegeneration of Cav-1 KO mice strongly supports the functionalrole of caveolin proteins in the pathogenesis of ischemicinjury. Indeed, genetic ablation of the Cav-1 gene was shownto increase the extent of ischemic injury in a model ofhindlimb ischemia.19 However, the role of Cav-1 in thepatho-physiology of cerebral ischemia remains unknown. Inthe present study, we first determined the natural behavior ofcaveolin protein expression in a rat model of middle cerebralartery occlusion (MCAO). Furthermore, to better determinethe implications of caveolin proteins in the pathogenesis ofcerebral ischemia, we subsequently investigated the outcomeof a MCAO in caveolin KO mice.

Materials and MethodsAnimalsThis study was conducted according to the guidelines of the NationalInstitute of Health and the Albert Einstein College of MedicineInstitute for Animal Studies. Male Sprague-Dawley rats weighing200 to 225 g were purchased from Taconic Farms (Hudson, NY).Male Cav-1 KO and Cav-2 KO mice were generated, as previouslydescribed.14,24 All mice used in these studies were in the C57Bl/6genetic background.

MaterialsMouse Cav-1 and Cav-2 monoclonal antibodies (mAbs) were thegenerous gifts of Dr Roberto Campos-Gonzalez (BD Pharmingen,San Diego, Calif). A rabbit polyclonal antibody (pAb) to Cav-1 anda mouse mAb to proliferating cell nuclear antigen (PCNA) werepurchased from Santa Cruz Biotechnology (Santa Cruz, Calif).Rabbit pAbs to glial fibrillary-acidic protein (GFAP) and neurofila-ment heavy chain were respectively purchased from Dako Cytoma-tion (Carpinteria, Calif) and Novus Biologicals (Littleton, Colo). Arabbit pAb to laminin was purchased from abcam (Cambridge,Mass). A mouse mAb to �-actin as well as the nuclear dye Hoechstwere purchased from Sigma-Aldrich (St-Louis, Mo). A mouse mAb

to eNOS, a rabbit pAb to inducible NOS (iNOS), a rabbit pAb toneuronal NOS (nNOS), a rabbit pAb to von Willebrand’s factor(vWF) as well as rabbit and mouse horseradish peroxidase (HRP)-conjugated secondary antibodies were all purchased from BD-Pharmingen. A rabbit pAb to phospho(ser1177)-eNOS was pur-chased from Cell Signaling Technology (Danvers, Mass). Rabbit andmouse fluorescein (FITC)- and rhodamine (TRITC)-conjugated sec-ondary antibodies were purchased from Jackson ImmunoResearch(West Grove, Pa).

Surgical ProceduresTransient MCAO was induced in male Sprague-Dawley rats weigh-ing 200 to 225 g, as previously described.25 Briefly, the rats wereinitially anesthetized by inhalation of 5% halothane through aface-mask in oxygen-enriched air and later maintained at 2.5% of thesame mixture. The left common carotid artery and left externalcarotid artery were exposed through a midline neck incision and theleft external carotid artery was coagulated. A 4–0 monofilamentsuture (Ethicon, Somerville, NJ), whose tip had been rounded byheating near a flame, was inserted into the left external carotid arteryand advanced into the left internal carotid artery past the MCA originuntil the tip reached the proximal anterior cerebral artery, thusoccluding the origin of the MCA. After three hours of MCAO, thefilament was removed and blood flow was restored. Rats were killedby decapitation at either 48 hours, 1-week or 2-weeks postischemia(n�10 for each group). The Sham groups were subjected to the sameprocedure except for the occlusion of the MCA (n�10 for eachgroup).

Figure 1. Immunoblot analysis of Cav-1 and Cav-2 expressionin the brains of Sham and MCAO rats at 48 hours (A), 1-week(B), and 2-weeks postischemia (C) (1 of 6 rats is shown for eachgroup). Immunoblotting against �-actin is shown as an equalloading control. (R) represents the right hemisphere of the brain(contralateral) and (L) represents the left hemisphere of the brain(ischemic).

722 Circulation Research March 16, 2007


Dow

nloaded from


Permanent MCAO was induced in 8 to 10 week-old malewild-type (WT), Cav-1 KO and Cav-2 KO mice, as previouslydescribed (n�16 to 22 for each group).26 Briefly, mice wereanesthetized with inhalation of 3% halothane initially and main-tained at 1.5%. The left MCA was exposed by subtemporal craniot-omy using an operating microscope. Two 11–0 silk sutures werepassed under the artery and the artery was cut in between the 2 sutureknots. Afterward, retracted soft tissue was replaced and wounds weresutured. All mice were killed by cervical dislocation at 72 hourspostischemia.

Infarction Volume MeasurementBrains from WT, Cav-1 KO and Cav-2 KO mice were removed andcut into 2 mm-thick coronal section using a brain matrix (n�10 to 16for each group). The brain sections were then immersed in 2%2,3,5-triphenyltetrazolium chloride (TTC) (Sigma-Aldrich) for 30minutes and then fixed with 4% phosphate buffered formalin. Eachbrain slice was scanned and the infarct area in each image wascalculated using a video image analyzing system (NIH Scion Image,version 1.65) by an observer who was blinded to the study. Infarctvolume corrected for edema was calculated by subtracting thenoninfarcted area of the infarcted hemisphere from the normalcontralateral hemisphere.

Immunoblot AnalysisBrains from Sham and MCAO rats as well as WT and Cav-1 KOmice were cut in half, to separate the left (ischemic) and the right(contralateral nonischemic) hemisphere of the brain (n�6 for eachgroup). The brains were homogenized in RIPA lysis buffer contain-ing protease and phosphatase inhibitors. Proteins were then separatedby SDS-PAGE (6% to 12% acrylamide) and transferred to nitrocel-lulose membranes. The membranes were then placed in blockingsolution for 30 minutes and subsequently washed with 10 mmol/LTris, 150 mmol/L NaCl and 0.05% Tween 20 (1X-TBS-Tween). Themembranes were incubated with a given primary antibody for either1 hour (Cav-1, Cav-2, �-actin) or 3 hours (eNOS, phospho(ser1177)-eNOS, iNOS, nNOS) at room temperature. Afterward, the mem-

Figure 2. Dual-label immunofluorescenceanalysis shows the colocalization ofCav-1 (A, red) and vWF (B, green) in ratischemic hemispheres at 48 hours post-ischemia (C, yellow). However, Cav-1 (Dand G, red) did not colocalize with eitherGFAP (E, green) or neurofilament heavychain antibody (H, green) at 48 hourspostischemia. Panels F and I representthe merged images (yellow) of Cav-1with GFAP and neurofilament heavychain antibody, respectively. All imageswere taken at the same magnification of�40.

Figure 3. Representative TTC staining of coronal brain sectionssliced rostral to caudal (from left to right) shows increased vol-ume of infarction in Cav-1 KO mice, as compared with WT andCav-2 KO mice (A). Quantitation of the volume of infarction isshown in panel B. *P�0.05 vs WT mice; †P�0.05 vs Cav-1 KOmice (n�10 to 16 for each group).

Jasmin et al Caveolin and Cerebral Ischemia 723


Dow

nloaded from


branes were washed and finally incubated with HRP-conjugatedsecondary antibodies. The SuperSignal chemiluminescence substratewas used to detect bound primary antibody. Western blots for Cav-1,Cav-2, eNOS, phospho(ser1177)-eNOS, iNOS and nNOS werequantitated using NIH Image J software (using the mean gray valueof each band).

Immunofluorescence Analysis andApoptosis DetectionBrains from Sham and MCAO rats, as well as WT and Cav-1 KO mice,were immersed in 4% paraformaldehyde for 24 hours and subsequentlyembedded in paraffin (n�4 to 6 for each group). Sections of 10 �mwere cut and stained with Hematoxylin and Eosin (H&E). Paraffin from10 �m-thick sections was removed by immersion in xylene. Thesections were then rehydrated with graded alcohol to water and blockedovernight using HenBLKII (Aves Labs, Tigard, Ore). These sectionswere subsequently incubated with primary antibodies for 3 hours atroom temperature. The primary antibodies were used at the followingdilutions; Cav-1 mAb (1:100), Cav-2 mAb (1:100), vWF pAb (1:50),laminin pAb (1:50), PCNA mAb (1:50), GFAP pAb (1:100) andneurofilament heavy chain pAb (1:100). Hoechst dye was used at aconcentration of 10 �g/mL. Afterward, the sections were washed with1X-PBS and incubated for 1 hour at room temperature with secondaryantibodies. Mouse and rabbit rhodamine (TRITC)- and fluorescein(FITC)-conjugated secondary antibodies were used at a dilution of1:400. The sections were then mounted with ProLong Gold antifadereagent (Molecular Probes, Carlsbad, Calif). Detection of TUNEL-positive cells in WT and Cav-1 KO brains was performed using theTACS 2 TdT In Situ Apoptosis Detection Kit (Trevigen, Gaithersburg,Md), according to the manufacturer’s instructions. All sections wereexamined under a Nikon Te2000-S eclipse microscope (Morrell Instru-ment Company, Melville, NY). Immunofluorescence analyses of PCNAand laminin were performed in the border zone, whereas analysis ofTUNEL-positive cells was performed in both the ischemic core andborder zone.

Statistical AnalysisAll data are expressed as mean � S.E.M and the differences betweengroups were evaluated by either unpaired Student’s t-test or ANOVAfollowed by Tukey’s multiple-group comparisons test, where appro-priate. Statistical significance was assumed at P�0.05.

ResultsIncreased Caveolin Protein Expression in RatsSubjected to MCAOImmunoblot analyses showed marked increases in Cav-1 andCav-2 protein levels in the ischemic hemisphere of MCAOrats at 48 hours, 1-week and 2-weeks postischemia (P�0.05;Figure 1). Dual-label immunofluorescence analysis at 48hours post-ischemia demonstrated the colocalization of Cav-1with the endothelial cell marker, vWF, in rat ischemichemisphere (Figure 2). However, dual-label immunofluores-cence analysis of rat ischemic hemisphere at 48 hourspostischemia did not show colocalization of Cav-1 with eitherGFAP, an astrocyte marker, or neurofilament heavy chainantibody, a neuronal marker (Figure 2). Importantly, immu-nofluorescence analysis of rat brains at 1-week and 2-weekpostischemia gave identical results as the 48 hours postische-mic group (data not shown). Furthermore, dual-label immu-nofluorescence analysis demonstrated identical localizationof Cav-1 and Cav-2 in rat ischemic brains (supplementalFigure I in the online data supplement available athttp://circres.ahajournals.org).

Caveolin-1 Deficiency Increases CerebralInfarction Volume in MiceTo determine the cerebral volume of infarction, 2 mm-thicksections sliced rostral to caudal were stained with 2% 2,3,5-tri-phenyltetrazolium chloride (TTC). Interestingly, as shown inFigure 3A, the cerebral infarct observed in Cav-1 KO miceappears to be larger and to extend further caudally (from left toright). Accordingly, subtraction of the noninfarcted area of theinfarcted hemisphere from the normal contralateral hemispherereveal a marked increase of volume in infarction in Cav-1 KOischemic brains (19.2�3.0 mm3) as compared with WT(10.2�2.6 mm3) and Cav-2 KO (7.0�1.3 mm3) ischemic brains(P�0.05; Figure 3B). Conversely, no significant differences of

Figure 4. Immunofluorescence analysis of the endothelial cell marker, laminin (green), shows an increased number of endothelial cellsin WT ischemic hemispheres (B), as compared with WT contralateral hemispheres (A). Conversely, Cav-1 KO contralateral (C) and is-chemic (D) hemispheres showed a similar number of laminin-positive cells. Quantitation of the total number of endothelial cells isshown in panel E. The quantitation represents the average number of endothelial cells/field of 15 fields per animal (n�6 for eachgroup). *P�0.05 vs WT contralateral hemisphere; †P�0.05 vs WT ischemic hemisphere. Images were taken at the same magnificationof �40.



Dow

nloaded from


volume of infarction were observed between the WT and Cav-2KO ischemic brains (p�ns, Figure 3B).

Caveolin-1 Deficiency Impairs CerebralAngiogenesis in MiceImmunofluorescence analysis using antibodies against lami-nin showed an increased number of endothelial cells in WTischemic hemispheres (16.3�1.6 cells/field), as comparedwith WT contralateral hemispheres (11.7�0.8 cells/field)(P�0.05; Figure 4). Conversely, Cav-1 KO contralateral andischemic hemispheres showed similar numbers of endothelialcells (p�ns; Figure 4). Interestingly, immunofluorescenceanalysis further showed a decreased ratio of PCNA-positivenuclei/total nuclei in Cav-1 KO ischemic brains(13.8�1.2%), as compared with WT ischemic brains(27.0�2.9%) (P�0.05; Figure 5). Importantly, dual-labelimmunofluorescence analyses demonstrated that the percent-age of PCNA-positive cells identified as endothelial cells wasmarkedly decreased in Cav-1 KO ischemic brains(8.4�1.4%), as compared with WT ischemic brains(23.2�3.5%) (P�0.05; Figure 5). However, PCNA-positivecells did not colocalize with either GFAP or neurofilament

heavy chain antibody in both Cav-1 KO and WT ischemicbrains (supplemental Figure II in the online data supplementavailable at http://circres.ahajournals.org). No significant dif-ferences were observed in the immunofluorescence analysisof PCNA, GFAP and neurofilament heavy chain antibodybetween WT and Cav-1 KO contralateral hemispheres (datanot shown).

Interestingly, immunoblot analysis demonstrated that WTischemic brains display increases in both eNOS andphospho(ser1177)-eNOS proteins levels, as compared withCav-1 KO ischemic brains (P�0.05; Figure 6). However,although the phospho(ser1177)-eNOS/total eNOS ratio ap-pear slightly increased in WT ischemic brains, no significantdifferences were observed among all groups (p�ns; Figure6). Furthermore, nNOS and iNOS proteins levels remainedunchanged in the ischemic brains of both WT and Cav-1 KOmice (p�ns; Figure 7).

Caveolin-1 Deficiency Increases CerebralApoptotic Cell Death in MiceCav-1 KO ischemic brains showed an increased number ofTUNEL-positive cells (57.3�3.4 cells/field), as compared

Figure 5. Representative images of thecolocalization of PCNA (red) and thenuclear dye marker Hoechst (blue) showan increased percentage of PCNA-positive nuclei (pink) in WT ischemicbrains (A), as compared with Cav-1 KOischemic brains (B). Representativeimages of the colocalization of PCNA(green), laminin (red), and the nucleardye marker Hoechst (blue) show anincreased percentage of PCNA-positiveendothelial cells (yellow) in WT ischemicbrains (C), as compared with Cav-1 KOischemic brains (D). Quantitations of thepercentage of PCNA-positive nuclei andPCNA-positive vessels are shown inpanel E and F, respectively. The quanti-tations represent the average percentageof PCNA-positive nuclei/field of 15 fieldsper animal as well as the average per-centage of PCNA-positive endothelialcells/field of 100 fields per animal,respectively. *P�0.05 vs WT mice (n�6for each groups). Images were taken atmagnification of �40 (A and B) and�100 (C, and D).



Dow

nloaded from


with WT ischemic brains (28.9�3.1 cells/field) (P�0.05;Figure 8).

DiscussionOur present results demonstrate marked increases in Cav-1and Cav-2 protein levels as well as their specific colocal-ization with endothelial cells in the brains of rats subjectedto transient MCAO. Our results further reveal, for the firsttime, that genetic ablation of the Cav-1 gene increases thecerebral volume of infarction in mice subjected to perma-nent MCAO. Mechanistically, impaired angiogenesis andincreased apoptotic cell death appear to contribute to theincreased cerebral ischemic injury observed in Cav-1 KOmice.

Caveolin proteins have previously been implicated inthe patho-physiology of ischemic injury. Indeed, changesin caveolin protein expression and localization have beenreported in several models of ischemic injury. For in-stance, Cav-1 expression was shown to be markedlyincreased in renal cortical/proximal tubules followingischemic acute renal failure.22 Furthermore, although theirtotal protein expression remained unchanged, a dissocia-tion of Cav-1 and Cav-3 from caveolae to the cytosol wasreported in the hearts of rats subjected to myocardialischemia-reperfusion.23 However, the natural behavior ofcaveolin protein expression in cerebral ischemia-reperfusion remained unclear. Our present results demon-strate marked increases in both Cav-1 and Cav-2 proteinlevels at 48 hours, 1-week, and 2-week postischemia. Thepresent results differ from those of Shen et al (2006) who

have recently reported decreased Cav-1 expression in theischemic core of MCAO rat brains.27 Although unclear,these discrepancies might be due to variations in theexperimental protocols, such as the age of the rats, theduration of ischemia (1 hour versus 3 hours), as well as theimmunoblot analysis of homogenates of the entire ische-mic hemisphere versus homogenates of the ischemic coreonly. Accordingly, Cav-1 expression appears to be differ-entially modulated in the ischemic core and penumbra areaof rat ischemic brains.27 Our immunofluorescence analysesfurther reveal the selective colocalization of both Cav-1and Cav-2 with the endothelial cell marker, vWF, in ratischemic brains. For instance, we and others previouslydemonstrated the selective expression of Cav-1 and Cav-2in endothelial cells of bovine and rat brains.28 These resultssuggest that increased caveolin protein levels might influ-ence the angiogenic processes occurring in cerebral ische-mia/reperfusion. We previously demonstrated that amarked increase in endogenous Cav-1 protein levels pre-

Figure 6. Immunoblot analysis of total eNOS andphospho(ser1177)-eNOS expression levels in the brains of WTand Cav-1 KO mice subjected to MCAO (A) (2 of 6 rats areshown for each group). Quantitation of the phospho(ser1177)-eNOS/eNOS ratio is shown in panel B (n�6 for each group). (R)represents the right hemisphere of the brain (contralateral) and(L) represents the left hemisphere of the brain (ischemic).

Figure 7. Immunoblot analysis of nNOS and iNOS expression inthe brains of WT and Cav-1 KO mice subjected to permanentMCAO (A) (2 of 6 rats are shown for each group). Quantitationof nNOS and iNOS expression is shown in panels B and C,respectively (n�6 for each group). (R) represents the right hemi-sphere of the brain (contralateral) and (L) represents the lefthemisphere of the brain (ischemic).



Dow

nloaded from


cedes the development of capillary-like tubules inHMEC-1 cells cultured on Matrigel.18

Since that in vivo genetic ablation of the caveolin genesis, so far, only available in mice, we subsequently decidedto investigate the outcomes of a MCAO in caveolin KOmice. The use of caveolin KO mice allowed us to clarifythe functional role of caveolin proteins in the pathogenesisof cerebral ischemic injury. Interestingly, our resultsdemonstrate a marked increase of cerebral volume ofinfarction in Cav-1 KO mice, as compared with both WTand Cav-2 KO mice. Accordingly, genetic ablation of theCav-1 gene in mice was previously shown to increase theextent of ischemic injury in a model of hindlimb ische-mia.19 Furthermore, disruption of caveolar domains withmethyl-�-cyclodextrin (M�CD) was previously shown toattenuate the protective effects of ischemic precondition-ing in adult rat cardiomyocytes subjected to simulatedischemia/reperfusion.29 Interestingly, intravenous deliveryof the Cav-1 scaffolding domain peptide was previouslyshown to exert cardio-protective effects in myocardialischemia-reperfusion by increasing endothelium-derivedNO release, as well as by reducing polymorphonuclearneutrophil adherence and infiltration.30 Collectively, thesedata, as well as our present results, suggest that decreasedexpression of Cav-1 might be further detrimental to theischemic injury.

Mechanistically, our present results suggest that im-paired angiogenesis might contribute to the increasedcerebral ischemic injury observed in Cav-1 KO mice.Indeed, our immunofluorescence analyses demonstrate anincreased number of endothelial cells in WT ischemicbrains, as compared with Cav-1 KO ischemic brains. Mostimportantly, dual-label immunofluorescence analysis fur-ther demonstrates a marked increase in the number ofproliferating endothelial cells (PCNA-positive) in WTischemic brains, as compared with Cav-1 KO ischemicbrains. These results are in accordance with those ofSonveaux et al (2004) who previously reported increasedcapillary density, as well as increased numbers of neoves-sels, in WT ischemic hindlimbs, as compared with Cav-1KO ischemic hindlimbs.19 We have also previously re-ported a reduction of vessel density in Cav-1 KO mice,using a model of exogenous tumor cell injection.16 Thecontribution of impaired angiogenesis to the increasedcerebral ischemic injury of Cav-1 KO mice is furthersupported by our findings of reduced volume of infarctionin Cav-2 KO mice. Indeed, whereas Matrigel plugs im-planted in Cav-1 KO mice showed dramatic reduction inboth vessel infiltration and density, those implanted inCav-2 KO mice conversely showed an enhanced angio-genic response.16

As previously suggested, the impaired angiogenesisobserved in Cav-1 KO mice could be ascribed, at least inpart, to the lack of caveolar domains.19 Accordingly,although Cav-1 is well recognized as a natural inhibitor ofeNOS activity, Cav-1 KO ischemic hemispheres did notshow increased eNOS activation. Conversely, WT ische-mic hemispheres show significant increases in both eNOSand phospho(ser1177)-eNOS protein levels. However, al-though the phospho(ser1177)-eNOS/total eNOS ratio ap-pears slightly increased in WT ischemic brains, no signif-icant differences were observed among all groups.Nonetheless, cultured Cav-1 KO aortic endothelial cellswere previously shown to display a marked inhibition ofeNOS phosphorylation on vascular endothelial growthfactor (VEGF) stimulation.19 Therefore, as previouslysuggested,19 the absence of caveolar domains might hinderthe proper compartmentalization of the signaling mole-cules essential to NO synthesis and angiogenic processes.Accordingly, although its total expression remained un-changed, a dissociation of the VEGF receptor-2 (VEGFR2)from low-density to high-density membranes fractions waspreviously reported in cultured endothelial cells derivedfrom Cav-1 KO mice.19 Furthermore, disruption of caveo-lar domains with M�CD was previously shown to inhibitVEGF-induced extracellular signal-regulated kinase(ERK) activation and cell migration in bovine aorticendothelial cells.31 The implication of caveolar domains inangiogenesis is further supported by our findings ofreduced infarction volume in Cav-2 KO mice, as comparedwith Cav-1 KO mice. Indeed, unlike Cav-1 KO mice,Cav-2 KO mice still retain the ability to form caveolaethrough the homo-oligomerization of Cav-1.24

Importantly, although angiogenesis might play a crucialrole in the expansion of an infarct, the implications of

Figure 8. Representative images of TUNEL staining showincreased apoptotic cell death in the Cav-1 KO ischemic brains(B), as compared with the WT ischemic brains (A). Quantitationof TUNEL-positive cells is shown in panel C. The quantitationrepresents the average number of TUNEL-positive cells/field of20 fields per animal (n�6 for each group). *P�0.05 vs WT is-chemic brains. Images were taken at the same magnification of�40.



Dow

nloaded from


apoptosis in such a process cannot be overlooked. Hence,our present results further suggest that increased apoptoticcell death might also contribute to the increased cerebralischemic injury observed in Cav-1 KO mice. As a matter offact, Cav-1 KO ischemic brains show an increased numberof TUNEL-positive cells, as compared with WT ischemicbrains. Accordingly, although controversial, caveolin pro-teins have previously been reported to act as key regulatorsof apoptotic processes. Indeed, Cav-1 expression waspreviously shown to sensitize T24 bladder carcinoma cellsto staurosporine-induced apoptosis.32 Conversely, Cav-1expression was shown to suppress c-Myc-induced apopto-sis in a human epithelial prostate cancer-derived cell line.33

Interestingly, caveolar domains have previously been pro-posed to maintain apoptotic signaling molecules in aninactive state, until reception of the appropriate stimulus.20

Disruption of caveolae with M�CD was shown to increasestaurosporine-induced caspase-3 activity in cardiac endo-thelial cells.20 Importantly, we recently demonstrated thatgenetic ablation of the Cav-1 gene in a transgenic mousemodel of prostate cancer resulted in increased apoptosislevels.21 Therefore, the lack of Cav-1 protein expressionand caveolar domains might result in inappropriate com-partmentalization and regulation of the numerous apoptot-ic signaling molecules and, consequently, result in thehyper-activation of apoptotic signaling pathways, leadingto increased neuronal death in a model of cerebralischemia.

In conclusion, this is the first report of increasedexpression of Cav-1 and Cav-2 as well as their colocaliza-tion with endothelial cells in the brains of rats subjected tocerebral ischemia/reperfusion. This increased expressionof endothelial Cav-1 and Cav-2 suggests essential roles forthe caveolin proteins in postischemic angiogenesis. Inter-estingly, genetic ablation of the Cav-1 gene in micesubjected to MCAO results in increased volume of infarc-tion. Mechanistically, impaired angiogenesis and increasedapoptotic death appear to contribute to the increasedischemic injury observed in Cav-1 KO mice.

Study LimitationsAlthough we quantified the number of capillaries in both WTand Cav-1 KO contralateral and ischemic hemispheres, wedid not determine the impact of collateralization on thecerebral volume of infarction. Therefore, in future studies, itwill be interesting to determine whether genetic ablation ofCav-1 might affect the cerebral collateral circulation in mice.

Sources of FundingThis work was supported by grants from the NIH (to M.P.L.), theAmerican Heart Association (to M.P.L.), and NIH-NEI EY11253 (toD.M.R.). J.F.J. was supported by Fellowship grants from the Fondsde la recherche en sante du Quebec and the Canadian Heart andStroke Foundation. S.M. was supported by the NY State fundedNeuroscience and Psychiatry Fellowship of the Albert EinsteinCollege of Medicine.

DisclosuresNone.

References1. Lisanti MP, Scherer PE, Tang ZL, Sargiacomo M. Caveolae, caveolin and

caveolin-rich membrane domains: a signalling hypothesis. Trends CellBiol. 1994;4:231–235.

2. Okamoto T, Schlegel A, Scherer PE, Lisanti MP. Caveolins, a family ofscaffolding proteins for organizing “preassembled signaling complexes”at the plasma membrane. J Biol Chem. 1998;273:5419–5422.

3. Smart EJ, Graf GA, McNiven MA, Sessa WC, Engelman JA, Scherer PE,Okamoto T, Lisanti MP. Caveolins, liquid-ordered domains, and signaltransduction. Mol Cell Biol. 1999;19:7289–7304.

4. Rothberg KG, Heuser JE, Donzell WC, Ying YS, Glenney JR, AndersonRG. Caveolin, a protein component of caveolae membrane coats. Cell.1992;68:673–682.

5. Glenney JR, Jr., Soppet D. Sequence and expression of caveolin, a proteincomponent of caveolae plasma membrane domains phosphorylated ontyrosine in Rous sarcoma virus-transformed fibroblasts. Proc Natl AcadSci U S A. 1992;89:10517–10521.

6. Scherer PE, Okamoto T, Chun M, Nishimoto I, Lodish HF, Lisanti MP.Identification, sequence, and expression of caveolin-2 defines a caveolingene family. Proc Natl Acad Sci U S A. 1996;93:131–135.

7. Tang Z, Scherer PE, Okamoto T, Song K, Chu C, Kohtz DS, NishimotoI, Lodish HF, Lisanti MP. Molecular cloning of caveolin-3, a novelmember of the caveolin gene family expressed predominantly in muscle.J Biol Chem. 1996;271:2255–2261.

8. Scherer PE, Lewis RY, Volonte D, Engelman JA, Galbiati F, Couet J,Kohtz DS, van Donselaar E, Peters P, Lisanti MP. Cell-type and tissue-specific expression of caveolin-2. Caveolins 1 and 2 co-localize and forma stable hetero-oligomeric complex in vivo. J Biol Chem. 1997;272:29337–29346.

9. Song KS, Li S, Okamoto T, Quilliam LA, Sargiacomo M, Lisanti MP.Co-purification and direct interaction of Ras with caveolin, an integralmembrane protein of caveolae microdomains. Detergent-free purificationof caveolae microdomains. J Biol Chem. 1996;271:9690–9697.

10. Couet J, Li S, Okamoto T, Ikezu T, Lisanti MP. Identification of peptideand protein ligands for the caveolin-scaffolding domain. Implications forthe interaction of caveolin with caveolae-associated proteins. J BiolChem. 1997;272:6525–6533.

11. Engelman JA, Chu C, Lin A, Jo H, Ikezu T, Okamoto T, Kohtz DS,Lisanti MP. Caveolin-mediated regulation of signaling along the p42/44MAP kinase cascade in vivo. A role for the caveolin-scaffolding domain.FEBS Lett. 1998;428:205–211.

12. Garcia-Cardena G, Oh P, Liu J, Schnitzer JE, Sessa WC. Targeting ofnitric oxide synthase to endothelial cell caveolae via palmitoylation:implications for nitric oxide signaling. Proc Natl Acad Sci U S A. 1996;93:6448–6453.

13. Garcia-Cardena G, Fan R, Stern DF, Liu J, Sessa WC. Endothelial nitricoxide synthase is regulated by tyrosine phosphorylation and interacts withcaveolin-1. J Biol Chem. 1996;271:27237–27240.

14. Razani B, Engelman JA, Wang XB, Schubert W, Zhang XL, Marks CB,Macaluso F, Russell RG, Li M, Pestell RG, Di Vizio D, Hou HJ, KneitzB, Lagaud G, Christ GH, Edelmann W, Lisanti MP. Caveolin-1 null miceare viable but show evidence of hyperproliferative and vascular abnor-malities. J Biol Chem. 2001;276:38121–38138.

15. Schubert W, Frank PG, Woodman SE, Hyogo H, Cohen DE, Chow CW,Lisanti MP. Microvascular hyperpermeability in caveolin-1 (-/-)knock-out mice. J Biol Chem. 2002;277:40091–40098.

16. Woodman SE, Ashton AW, Schubert W, Lee H, Williams TM, MedinaFA, Wyckoff JB, Combs TP, Lisanti MP. Caveolin-1 knockout miceshow an impaired angiogenic response to exogenous stimuli. Am JPathol. 2003;162:2059–2068.

17. Griffoni C, Spisni E, Santi S, Riccio M, Guarnieri T, Tomasi V.Knockdown of caveolin-1 by antisense oligonucleotides impairs angio-genesis in vitro and in vivo. Biochem Biophys Res Commun. 2000;276:756–761.

18. Liu J, Wang XB, Park DS, Lisanti MP. Caveolin-1 expression enhancesendothelial capillary tubule formation. J Biol Chem. 2002;277:10661–10668.

19. Sonveaux P, Martinive P, DeWever J, Batova Z, Daneau G, Pelat M,Ghisdal P, Gregoire V, Dessy C, Balligand JL, Feron O. Caveolin-1expression is critical for vascular endothelial growth factor-induced is-chemic hindlimb collateralization and nitric oxide-mediated angio-genesis. Circ Res. 2004;95:154–161.

20. Oxhorn BC, Buxton IL. Caveolar compartmentation of caspase-3 incardiac endothelial cells. Cell Signal. 2003;15:489–496.



Dow

nloaded from


21. Williams TM, Hassan GS, Li J, Cohen AW, Medina F, Frank PG, PestellRG, Di Vizio D, Loda M, Lisanti MP. Caveolin-1 promotes tumorprogression in an autochthonous mouse model of prostate cancer: geneticablation of Cav-1 delays advanced prostate tumor development in trampmice. J Biol Chem. 2005;280:25134–25145.

22. Mahmoudi M, Willgoss D, Cuttle L, Yang T, Pat B, Winterford C, EndreZ, Johnson DW, Gobe GC. In vivo and in vitro models demonstrate a rolefor caveolin-1 in the pathogenesis of ischaemic acute renal failure.J Pathol. 2003;200:396–405.

23. Ratajczak P, Damy T, Heymes C, Oliviero P, Marotte F, Robidel E,Sercombe R, Boczkowski J, Rappaport L, Samuel JL. Caveolin-1 and -3dissociations from caveolae to cytosol in the heart during aging and aftermyocardial infarction in rat. Cardiovasc Res. 2003;57:358–369.

24. Razani B, Wang XB, Engelman JA, Battista M, Lagaud G, Zhang XL,Kneitz B, Hou H, Jr., Christ GJ, Edelmann W, Lisanti MP. Caveolin-2-deficient mice show evidence of severe pulmonary dysfunction withoutdisruption of caveolae. Mol Cell Biol. 2002;22:2329–2344.

25. Giorgi FS, Malhotra S, Hasson H, Veliskova J, Rosenbaum DM, MosheSL. Effects of status epilepticus early in life on susceptibility to ischemicinjury in adulthood. Epilepsia. 2005;46:490–498.

26. Wiessner C, Allegrini PR, Ekatodramis D, Jewell UR, Stallmach T,Gassmann M. Increased cerebral infarct volumes in polyglobulic miceoverexpressing erythropoietin. J Cereb Blood Flow Metab. 2001;21:857–864.

27. Shen J, Ma S, Chan P, Lee W, Fung PC, Cheung RT, Tong Y, Liu KJ.Nitric oxide down-regulates caveolin-1 expression in rat brains during

focal cerebral ischemia and reperfusion injury. J Neurochem. 2006;96:1078–1089.

28. Ikezu T, Ueda H, Trapp BD, Nishiyama K, Sha JF, Volonte D, GalbiatiF, Byrd AL, Bassell G, Serizawa H, Lane WS, Lisanti MP, Okamoto T.Affinity-purification and characterization of caveolins from the brain:differential expression of caveolin-1, -2, and -3 in brain endothelial andastroglial cell types. Brain Res. 1998;804:177–192.

29. Patel HH, Head BP, Petersen HN, Niesman IR, Huang D, Gross GJ, InselPA, Roth DM. Protection of adult rat cardiac myocytes from ischemic celldeath: role of caveolar microdomains and delta opioid receptors. Am JPhysiol Heart Circ Physiol. 2006;291:H344–H350.

30. Young LH, Ikeda Y, Lefer AM. Caveolin-1 peptide exerts cardiopro-tective effects in myocardial ischemia-reperfusion via nitric oxidemechanism. Am J Physiol Heart Circ Physiol. 2001;280:H2489 –H2495.

31. Labrecque L, Royal I, Surprenant DS, Patterson C, Gingras D, BeliveauR. Regulation of vascular endothelial growth factor receptor-2 activity bycaveolin-1 and plasma membrane cholesterol. Mol Biol Cell. 2003;14:334–347.

32. Liu J, Lee P, Galbiati F, Kitsis RN, Lisanti MP. Caveolin-1 expressionsensitizes fibroblastic and epithelial cells to apoptotic stimulation. Am JPhysiol Cell Physiol. 2001;280:C823–C835.

33. Timme TL, Goltsov A, Tahir S, Li L, Wang J, Ren C, Johnston RN,Thompson TC. Caveolin-1 is regulated by c-myc and suppresses c-myc-induced apoptosis. Oncogene. 2000;19:3256–3265.



Dow

nloaded from


Rosenbaum and Michael P. LisantiJean-François Jasmin, Samit Malhotra, Manjeet Singh Dhallu, Isabelle Mercier, Daniel M.

Caveolin-1 Deficiency Increases Cerebral Ischemic Injury

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 2007 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.0000260180.42709.292007;100:721-729; originally published online February 9, 2007;Circ Res.

http://circres.ahajournals.org/content/100/5/721World 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/2007/02/10/01.RES.0000260180.42709.29.DC1Data 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:


Dow

nloaded from

http://circres.ahajournals.org/content/100/5/721

http://circres.ahajournals.org/content/suppl/2007/02/10/01.RES.0000260180.42709.29.DC1

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

http://www.lww.com/reprints

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


Online Supplement Jasmin et al.

Online Figure I. Co-localization of Cav-1 (A, green), Cav-2 (B, red) and the nuclear

dye marker Hoechst (C, blue) in rat ischemic hemispheres. Panel (D) represent the

merged image of A, B, and C (yellow). Images were taken at the same magnification of

10X.

Online Figure II. Dual-label immunofluorescence analysis of PCNA (red) with either

GFAP (A, green) or neurofilament heavy chain antibody (B, green) failed to show co-

localization of PCNA with either astrocytes or neurons in Wild-Type ischemic brains.

Images were taken at the same magnification of 40X.

Online Figure III. Immunoblot analysis of Cav-1 expression in the brains of Wild-

Type mice subjected to permanent middle cerebral artery occlusion for 72 hrs.

Immunoblotting against β-actin is shown as an equal loading control. (R) represents the

right hemisphere of the brain (contralateral) and (L) represents the left hemisphere of the

brain (ischemic).

Online Figure IV. Quantitation of the nuclear dye marker Hoechst shows increased

nuclei counts in WT ischemic hemispheres as compared to WT contralateral

hemispheres. Conversely, Cav-1 KO contralateral and ischemic hemispheres showed

similar nuclei counts. *p<0.05 vs Wild-Type contralateral hemisphere (n=6 for each

group).

Online Figure I .Jasmin J-F et al.

A B

C D

Online Figure IIJasmin J-F et al.

A

B

Online Figure IIIJasmin J-F et al.

β-Actin

Cav-1

R R R RL L L L

WT Cav-1 KO

Supplemental Data 4.Jasmin J-F et al.

Nuclei Counts

0

20

40

60

80

100

120

Num

ber o

f Nuc

lei/

Fiel

d

ContralateralIschemic

Wild-Type Cav-1 KO

*

Circulation Research Manuscript #CIRCRESAHA/2006/138115 - R2 Original Contribution

michelle

Text Box

Online Figure IV

Documents

Caveolin-1 Deficiency Increases Cerebral Ischemic Injurycircres.ahajournals.org/content/circresaha/100/5/721.full.pdf · Caveolin-1 Deficiency Increases Cerebral Ischemic Injury