Monoclonal Antibody Against Vascular Cell Adhesion Molecule-1 Inhibits Neointimal Formation After Periadventitial Carotid Artery Injury in Genetically Hypercholesterolemic Mice

Monoclonal Antibody Against Vascular Cell AdhesionMolecule-1 Inhibits Neointimal Formation After

Periadventitial Carotid Artery Injury in GeneticallyHypercholesterolemic Mice

Sumito Oguchi,* Paul Dimayuga,* Jenny Zhu, Kuang-Yuh Chyu, Juliana Yano, Prediman K. Shah,Jan Nilsson, Bojan Cercek

Abstract—Vascular cell adhesion molecule (VCAM)-1 is induced in smooth muscle cells after arterial injury, in which ithas been implicated in the recruitment of inflammatory cells to the site of injury. To investigate the effect ofhypercholesterolemia on VCAM-1 induction after injury and the role of VCAM-1 in neointimal response to injury, weinjured the carotid artery of wild-type and apolipoprotein E null (KO) mice fed normal and high cholesterol chow. Wedemonstrate a graded response of VCAM-1 induction as well as monocyte/macrophage infiltration by immunohisto-chemistry 3 days after injury that correlated with increasing circulating cholesterol levels. Three weeks after injury, KOmice fed high cholesterol chow (KO HC group) had a significantly greater neointimal formation compared withwild-type and KO mice fed normal chow (P,0.05). Inhibition of VCAM-1 function in the KO HC group by monoclonalantibody treatment significantly reduced monocyte/macrophage infiltration and neointimal formation. There wasreduceda-actin expression in KO HC mice 7 days after injury that was partially inhibited by VCAM-1 antibodytreatment. Cell migration in an in vitro injury model was partially inhibited by monoclonal VCAM-1 antibody treatment.We propose an additional role for VCAM-1 in smooth muscle cell activation and neointimal formation after injury.(Arterioscler Thromb Vasc Biol. 2000;20:1729-1736.)

Key Words: vascular cell adhesion molecule-1n inflammationn apoE knockout micen neointimal formation

Vascular cell adhesion molecule (VCAM)-1 is a memberof the immunoglobulin superfamily known to be ex-

pressed by vascular endothelial cells for the recruitment ofleukocytes during inflammation.1–3 VCAM-1 is also ex-pressed by cytokine-treated vascular smooth muscle cells(SMCs) in vitro and by balloon-injured arteries in vivo, inwhich a concomitant increase in monocyte/macrophage cellsat the site of injury was described.4–6 There is increasingevidence that an atherogenic diet induces inflammatory genespotentially through increased oxidative stress.7,8 Accordingly,hypercholesterolemia may increase VCAM-1 expression andplay an important role in the arterial response to injury.Immune-mediated inflammation has also been implicated inthis response, characterized by the influx of leukocytes,activation of SMCs, and neointimal formation.9,10 We usedthe arterial wall injury model in the genetically mutated

See p 1699hypercholesterolemic apoE knockout mice11 and monoclonalVCAM-1 antibody treatment12 to assess the role of VCAM-1in intimal formation after arterial injury. Our findings suggest

that hypercholesterolemia potentiates the response to injuryby increasing VCAM-1 expression, leading to increasedmacrophage infiltration and subsequent neointimal formation.Neutralization of VCAM-1 function by antibody treatmentinhibited neointimal formation in vivo and SMC migration invitro, suggesting an important role for VCAM-1 in the arterialresponse to injury.

MethodsAnimalsWild-type (WT) mice and apoE knockout (apoE KO) mice with agenetic background of C57BL/6J were purchased from JacksonLaboratory (Bar Harbor, Me). At the age of 25 weeks, the mice wereanesthetized with Avertin (0.016 mL/g of 2.5% solution IP), and theright carotid artery was carefully isolated under a dissection micro-scope. A 3-mm-long tube was applied around the carotid artery, andthe skin incision was closed, as previously described.11 At the timeof euthanasia, the carotid artery was perfused with 0.9% saline for 10minutes and frozen at270°C after embedding in OCT compound(Tissue-Tek, Miles Inc). Serial 4-mm-thick sections of arteries werecollected. Experimental protocols involving these animals wereapproved by the Institutional Animal Care and Use Committee.

Received May 5, 1999; revision accepted November 10, 1999.From the Atherosclerosis Research Center, Burns and Allen Research Institute, Division of Cardiology, Cedars-Sinai Medical Center/UCLA School

of Medicine, Los Angeles, Calif, and the Department of Medicine (J.N.), Lund University, University Hospital MAS, Malmo, Sweden.*S.O. and P.D. contributed equally to the article.Correspondence to Jan Nilsson, MD, PhD, Department of Medicine, Malmo University Hospital, 205 02 Malmo, Sweden. E-mail

[email protected]© 2000 American Heart Association, Inc.

Arterioscler Thromb Vasc Biol.is available at http://www.atvbaha.org

1729 by guest on June 29, 2015http://atvb.ahajournals.org/Downloaded from

Experimental GroupsWT mice were divided into 2 groups. One group was fed normalchow, injured, and euthanized after 3 (n55), 7 (n55), and 21 (n57)days. The second group served as uninjured controls. The apoE-KOmice were maintained on a normal or high cholesterol diet (n534each); half of this group was injured, and the other half served asuninjured controls. The time points were the same as those used inthe WT mice. For neutralization studies, purified rat anti-mouseVCAM-1 monoclonal antibody or rat IgG isotype was administered(1 mg/g body wt, Pharmingen) by tail vein injection to apoE KOmice on cholesterol chow (n58) on the day of injury and every otherday for the duration of the experiment. This antibody has been shownpreviously to inhibit VCAM-1 function in vivo.12 The tissues wereharvested for morphometric analysis. Presence of the antibody inserum was demonstrated by dot-blot analysis of serum from mice 21days after injury. Blots were incubated in anti-rat IgG (DAKO Corp)and detected by enhanced chemiluminescence (ECL, Amersham).

ImmunohistochemistryImmunohistochemical stains were carried out with the followingantisera: biotinylated anti–a-smooth muscle actin (Sigma ChemicalCo), anti-CD4, anti-CD8a, anti–VCAM-1 (Pharmingen), MOMA-2(Serotec), and Mac-1 (Roche). Biotinylated secondary antibody(Pierce) was used with the AEC chromogen detection kit (DAKOCorp). Nonimmune serum or isotype IgG was used as a negativecontrol. Sections from mice 3 days after injury were used forcomputer-assisted image analysis as a semiquantitative assessmentof immunohistochemical stains as previously described.13 Briefly,images were captured and analyzed by use of Optimas 6.1 (OptimasSystem, Bioscan). Color detection was accomplished by sampling,and threshold masking defined the positive area. The same thresholdwas applied to all sections. The area was then standardized againstthe medial area and expressed as percent stained area of the media.For morphometric analysis, serial sections were stained with eosinand hematoxylin, and the intimal and medial areas of 4 to 6 sectionsfrom the middle portion of the injured segment from each animalwere measured by the Optimas System. Results were expressed asmillimeters squared (mean6SD). Representative sections werestained for elastin by using the Accustain Elastic Stain Kit (Sigma)for photography.

Plasma and Tissue CholesterolEDTA plasma from all groups was collected at the time of euthana-sia, and cholesterol levels were measured by using a commerciallyavailable kit (Sigma). For tissue cholesterol levels, a modification ofa previously described protocol was adapted.14 Briefly, aortic tissuewas weighed and homogenized for 1 minute in a mixture of 160mLdistilled water, 200mL chloroform, and 400mL methanol. Afterhomogenization, 200mL chloroform was added, and the mixture wasblended for 30 seconds. Water (200mL) was then added and blendedagain for another 30 seconds. The mixture was centrifuged briefly toseparate the chloroform layer, which was carefully aspirated intoanother tube. The lipid remained in the bottom of the tube after thechloroform was vaporized in a vacuum trap. Lipid was thenresuspended in 50mL of 100% ethanol, and 10mL was used for thecholesterol assay. Results were expressed as micrograms totalcholesterol per milligram tissue.

Cell Culture and InjuryAortic SMCs were cultured by an explant method using 25-week-oldadult C57BL/6J mice as described previously.15 Briefly, mice wereanesthetized, and the aortas were isolated. The adventitia andendothelium were removed with the aid of microscopy. The aorticmedia was cut into 2-mm pieces and placed into 6-well platescontaining DMEM/F-12 with 100 U/mL penicillin, 100mg/mLstreptomycin, 0.25mg/mL amphotericin B (Fungizone), 2 mmol/LL-glutamine (GIBCO-BRL), and 10% FBS (Omega Scientific) incu-bated at 37°C in a humidified atmosphere of 5% CO2/95% airincubator. Cells that had migrated out of the explants were grown in20% FBS to confluence. Verification of cell type was accomplishedby using anti–smooth musclea-actin clone 1A4 (Sigma) immuno-cytochemistry. Cells were subcultured into 75-cm2 plates and grown

to confluence. After incubating in 1% FBS for 48 hours, cell injurywas performed by use of a 4-mm-wide sterile rubber tube gentlypressed for 10 seconds as described previously.16,17 Injured cellswere harvested after 24 and 48 hours. All experiments wereperformed within the first 8 passages.

Western BlotCytosolic protein was extracted by lysing the cell with cold hypo-tonic buffer (10 mmol/L HEPES, 10 mmol/L KCl, 1.5 mmol/LMgCl2, 1 mmol/L dithiothreitol, 1 mmol/L phenylmethylsulfonylfluoride, 1mg/mL leupeptin, 15mg/mL aprotinin, and 0.4% NP-40).After spinning at 9000gfor 30 seconds, the supernatant wascollected as a cytosolic fraction. For detection of VCAM-1 proteinafter injury, equal amounts of the extracted cytosolic protein wereelectrophoresed on 7.5% SDS-PAGE gel and transferred to nitrocel-lulose membrane. The membrane was blocked with 1% milk in PBSwith 0.1% Tween 20 overnight at 4°C. The membrane was subse-quently probed with VCAM-1 antibody (goat polyclonal, 1:1500,Santa Cruz Biotechnology), followed by horseradish peroxidase–conjugated anti-goat antibody. Detection was accomplished by usingthe ECL kit (Amersham). Computer-assisted densitometric analysiswas performed to quantify the detected bands.

Migration StudiesCells were subcultured into 2-chamber Permanox slides (Laboratory-Tek) and grown to confluence. After synchronizing growth in 1%FBS for 48 hours, cell injury was performed, and the medium wasreplaced with either 1% FBS medium only or 1% FBS with 50mg/mL rat anti-mouse VCAM-1 antibody. Rat isotype IgG at thesame concentration was used as a control. After 48 hours, cells werefixed in acetone and stained with hematoxylin and eosin. Distancebetween the nuclei of migrating cells and the margin of injury wasmeasured by use of the Optimas System. The distance of 50 cellsmigrating from a clear border of injury was averaged and counted as1 experimental value.17

StatisticsNumeric data are expressed as mean6SD. Differences among thegroups were determined by 1-way ANOVA, followed by theTukey-Kramer test for multiple comparisons unless otherwise noted.A value of P,0.05 was considered significant.

ResultsPlasma and Vascular Cholesterol LevelsPlasma cholesterol levels were increased in apoE KO mice onnormal chow and those on high cholesterol chow comparedwith WT mice (5676252 and 13486318 mg/dL versus164647 mg/dL, respectively;P,0.01, ANOVA).

In the KO mice fed normal chow, tissue cholesterol was'50% higher compared with WT mice (3.2460.55 versus2.0460.22mg/mg). Tissue cholesterol was increased sever-alfold (11.262.1mg/mg;P,0.01, ANOVA) in mice fed highcholesterol chow compared with WT and KO mice fednormal chow.

Arterial Response to InjuryNo intima was apparent in uninjured carotid arteries from C57WT and apoE KO mice. Three days after cuff placement, theinjured segments were mostly devoid of endothelial cells (datanot shown), as we have previously described.18Twenty-one daysafter cuff placement, there was neointimal formation observed(Figure 1A and 1D). Injury to the carotid artery of the apoE KOmice fed normal chow (KO N group) resulted in 100% increasedneointimal formation compared with the intimal area in the WTmice after 21 days, which was even greater when the apoE KOmice were fed high cholesterol chow (KO HC, Figure 1Bthrough 1D). Vessel diameter and medial area were not different

1730 Arterioscler Thromb Vasc Biol. July 2000

by guest on June 29, 2015http://atvb.ahajournals.org/Downloaded from

between the WT and KO N mice. The luminal area was slightlyless in KO N mice compared with WT mice, whereas theintima-to-media ratio was significantly increased in KO N mice.Vessel diameter in KO HC mice was similar to that in the othergroups, but the medial area was significantly increased com-pared with KO N and WT mice. The luminal area wassignificantly reduced in the KO HC group compared with the

WT group. the intima-to-media ratio was also significantlyincreased in the KO HC mice (Table).

VCAM-1 Expression and Monocyte/MacrophageInfiltration After InjuryThere was no apparent VCAM-1 staining in the media ofuninjured mice. Injury resulted in increased VCAM-1 expres-

Figure 1. Injury results in intimal formation 21 days after injury in the carotid artery of WT mice (A). The intima is increased in KO Nmice compared with WT mice (B). KO HC mice developed substantial intima compared with WT and KO N mice (C). Arrows indicateinternal elastic lamina. Measurement of the area indicates a 100% increase in intimal area in KO N mice (n57) compared with WT mice(n57). KO HC mice (n54) have significantly more intimal area compared with the other 2 groups (D). *P,0.05 vs WT.

Arterial Response to Injury and Effect of VCAM-1 Antibody Treatment

Group EEL, mm2 Media, mm2 Neointima, mm2 Lumen, mm2 I/M Ratio

WT (n57) 0.095560.0282 0.023460.0045 0.011560.0066 0.06260.018 0.4960.22

KO N (n55) 0.097360.0245 0.025960.0097 0.024560.0101 0.05260.018 0.9660.19*

KO HC (n54) 0.117060.0378 0.040260.0149* 0.045160.0204*† 0.03260.011* 1.1660.49*

KO HC1Ab (n55) 0.085560.0592 0.034860.0155 0.006460.0072‡ 0.09260.018‡ 0.1660.11‡

KO HC1IgG (n55) 0.094660.0392 0.045360.0231 0.027760.0193 0.02260.009§ 0.5860.19‡

Values are mean6SD. EEL indicates external elastic lamina; Ab, antibody.The arterial response 21 days after cuff injury is summarized above. Measurements were made as described in Methods. The effect

of VCAM-1 antibody treatment in KO HC mice (KO HC1Ab) is also summarized. Reduction in neointimal and luminal measurementsas well as the intima-to-media (I/M) ratio is shown.

*P,0.05 vs WT; †P,0.05 vs KO N. Statistical analyses on the effects of the treatments were performed without WT and KO Ngroups. ‡P,0.05 vs KO HC; §P,0.001 vs KO HC1Ab.

Oguchi et al VCAM-1 Antibody Inhibits Neointimal Formation 1731

by guest on June 29, 2015http://atvb.ahajournals.org/Downloaded from

sion in the media of WT mice, as shown by immunohisto-chemical staining, which peaked after 3 days of cuff place-ment and persisted for 7 days. At 3 days, this increase wasslightly augmented in the KO N group and significantly moreso in the KO HC group at the same time point (Figure 2Athrough 2C). Isotype control antibody stain was negative.Semiquantitative computer analysis indicated that theVCAM-1–stained area was significantly increased in injuredKO HC mice compared with KO N and WT mice (2765%versus 1463% and 1162%, respectively;P,0.01; n53).MOMA-2 staining 3 days after injury showed minimalpresence of monocytes/macrophages in the injured vessels ofWT mice. Injured KO mice stained positively for MOMA-2,which was increased in high cholesterol–fed mice (Figure 2Dthrough 2F). Confirmation of MOMA-2 stains were accom-plished by using Mac-1, which yielded a similar but fainterstain pattern. Isotype control antibody stain was negative.Computer-assisted analysis of the MOMA-2–stained areashowed a trend similar to VCAM-1 expression, with thelowest stained area in the WT mice, increasing in KO N andHC mice (1.361% versus 10.461% and 15.366%, respec-tively; P,0.05). VCAM-1 expression was correlated withmonocyte/macrophage infiltration, as shown by MOMA-2staining at the site of injury (r50.85,P,0.05; Figure 3A).The VCAM-1–stained area 3 days after injury was alsocorrelated with the intimal area 21 days after injury (Figure3B). All groups had minimal stains with CD4 and CD8antibodies.

VCAM-1 Antibody NeutralizationTo determine the function of VCAM-1 in an injury model ofhypercholesterolemia, injured KO mice on a high cholesteroldiet were injected with VCAM-1 antibody. A slot-blotanalysis of serum from treated mice after 21 days indicatedthe presence of rat IgG (Figure 4A). After 3 days, the injuredcarotid arteries were harvested. Localization of the antibody

in the injured arteries was confirmed by use of biotinylatedanti-rat IgG secondary antibody. Staining for rat IgG wasobserved only in the VCAM-1 antibody–treated mice (Figure4B). The 3-day injured arteries were stained for VCAM-1 andMOMA-2. By use of computer quantitative measurements,anti–VCAM-1 treatment did not alter VCAM-1 expression(Figure 4C) but significantly reduced the MOMA-2–stainedarea compared with the area in KO mice fed high cholesterolchow (362% versus 1566%,P,0.05; Figure 4D). Admin-istration of the anti-mouse VCAM-1 antibody significantlyreduced neointimal formation in the KO HC group (Figure5A and 5B). Rat IgG isotype control did not have the sameeffect on neointimal formation after injury (Figure 5C andTable). The tissue cholesterol level was not affected by theinjection of VCAM-1 antibody (11.262.1 versus 12.863.8mg/mg,P5NS).

a-Actin Expression After InjuryTo identify the cells that express VCAM-1 in the media afterinjury, double staining fora-actin and VCAM-1 was per-formed on sections. There was staining for both proteins 3days after injury for WT and KO HC mice (Figure 6A and6B). Seven days after injury, there was a conspicuousreduction in a-actin expression in the media (Figure 6C).Treatment with the anti–VCAM-1 antibody partially inhib-ited thisa-actin reduction (Figure 6D).

VCAM-1 Expression and Migration ofCultured SMCsTo test whether mechanical injury alters VCAM-1 expressionby SMCs, we used a previously described cell injury model.16

Uninjured SMCs showed baseline expression, which wasincreased 48 hours after injury (2.260.9-fold versus beforeinjured,P,0.05; n54) determined by densitometric analysisof Western blots (Figure 7A).

Figure 2. Carotid arteries of WT mice show VCAM-1 expression 3 days after injury (A). ApoE KO mice on normal chow (KO N group)exhibit more stained area (B), which increases in KO HC mice at the same time point (C). Monocyte/macrophage presence is minimalin injured carotid arteries of WT mice after 3 days (D). Monocyte/macrophage infiltration is observed in the KO N mice (E) and increasesin KO HC mice at the same time point (F). Bar in panel C represents top panel; bar in panel F represents bottom panel.

1732 Arterioscler Thromb Vasc Biol. July 2000

by guest on June 29, 2015http://atvb.ahajournals.org/Downloaded from

To investigate the role of VCAM-1 in the migration of SMCs,injured cells were incubated in 50mg/mL anti–VCAM-1 anti-body immediately after injury for 48 hours. The distancemigrated from the injury border was'35% less in antibody-treated compared with untreated and isotype IgG–treated cells(0.10760.012 versus 0.17660.016 and 0.16460.021 mm, re-spectively;P,0.05, ANOVA; Figure 7C).

DiscussionThis report identifies VCAM-1 as a mediator of neointimalformation after carotid injury in mice. Inhibition of VCAM-1function by antibody blockade profoundly reduced neointi-mal formation after injury. VCAM-1 in SMCs has beendemonstrated in atherosclerotic plaques and more recently ininjured rat carotid arteries.4–6,19 Because neointimal forma-tion after injury primarily involves SMC activation, wesought to examine the effect of hypercholesterolemia onVCAM-1 expression in medial SMCs after injury in vivo. Weshow a graded response in VCAM-1 expression after injury,which was associated with different cholesterol profiles. TheapoE KO mice on high cholesterol chow showed the greatestexpression, whereas the least expression was in WT mice.The difference in tissue cholesterol parallels plasma choles-

terol among the groups of mice. The results suggest thathypercholesterolemia increases lipid accumulation in tissueaugmenting VCAM-1 expression after injury; this increasedaccumulation is perhaps due to increased oxidative stress8

and may cause exacerbated intima thickening in response toinjury. This notion is further supported by the correlationbetween VCAM-1 expression and neointimal formation (Fig-ure 3B). Recent work from our laboratory shows the presenceof malondialdehyde-modified lipoprotein in injured arteriesof apoE KO mice,11 which may partially mediate the inflam-matory response by increasing VCAM-1 expression. One canspeculate that VCAM-1 expression might be a predictor ofintimal thickening after injury. This suggestion is furtherstrengthened by a recent report showing increased VCAM-1expression on endothelial cells at sites prone to lesionformation in hypercholesterolemic apoE KO mice comparedwith WT mice.20

Reports correlating inflammatory infiltrates with neointi-mal thickening have substantially strengthened the notion thatinflammation plays a significant role in the arterial responseto injury.9,10 Adhesion molecules, such as intercellular adhe-sion molecule (ICAM)-1 and VCAM-1, expressed by endo-thelial and smooth muscle cells, mediate inflammatory cellrecruitment to sites of injury.1,6 Arterial injury has beenshown to increase ICAM-1 expression in rats. Treatment ofrats with anti-inflammatory drugs, such as aspirin, reducedICAM-1 expression and neointimal formation after injury.21

Antibody blockade experiments against ICAM-1 inhibitedintimal formation in balloon-injured rat arteries. The mecha-nism of the effects of ICAM-1 inhibition was not clear;however, monocyte/macrophage accumulation was suggestednot to play a role.22 A recent report describing increasedVCAM-1 expression after rat arterial injury was associatedwith the adhesion of monocytes/macrophages.6 Several ex-perimental models of inflammation have correlated the ex-pression of VCAM-1 with monocyte recruitment.1–3,23There-fore, we tested the hypothesis that VCAM-1 facilitates therecruitment of inflammatory cells after arterial injury. Ourresults show a correlation between VCAM-1 expression andmonocyte/macrophage infiltration after injury. The aug-mented VCAM-1 expression in hypercholesterolemic mice inresponse to injury is associated with increased monocyte/macrophage recruitment and increased neointimal formation.A rabbit cuff-injury model using antibody blockade of leu-kocytes showed that polymorphonuclear influx had littleeffect on SMC migration and intimal thickening.24 It isnotable that mononuclear cells were present in significantquantity only after prolonged exposure of the artery to LDLin vivo. This exposure to LDL also significantly increasedintimal thickening.25 In a cuff model with electrical stimulus–induced injury, mononuclear cell but not polymorphonuclearinflux was shown to influence intimal thickening.10 Althoughthe present study was not designed to determine specific cellinfiltrates, on the basis of previous results outlined above,10,24

it is reasonable to speculate that polymorphonuclear influxdoes not significantly affect intimal thickening in the cuff-injury model. In the present report, the reduction of neointi-mal formation with antibody treatment was associated withreduced inflammatory infiltrates, suggesting further thatmonocyte/macrophage accumulation plays a role in promot-ing neointimal formation in injured arteries of apoE KO mice.

Figure 3. A, Computerized morphometric analysis of stainedarea shows that expression of VCAM-1 is significantly increasedin KO HC compared with WT and KO N mice 3 days after injury.Monocyte/macrophage infiltration is minimal in WT mice but issignificantly increased in KO N and KO HC mice 3 days afterinjury (see Results). There is a correlation between the extent ofVCAM-1 stain and the presence of monocytes/macrophages. B,VCAM-1 expression 3 days after injury is associated with intimalformation 21 days after injury. Intimal areas for 3 mice eachfrom the WT, KO N, and KO HC groups 21 days after injurywere plotted against the VCAM-1 area measurements of a sep-arate group of mice euthanized 3 days after injury, as summa-rized in Methods.

Oguchi et al VCAM-1 Antibody Inhibits Neointimal Formation 1733

by guest on June 29, 2015http://atvb.ahajournals.org/Downloaded from

There was a slight reduction in neointimal formation by IgGtreatment. Although suggestive of a protective effect, theresult was not significantly different compared with the resultin untreated KO HC mice.

The interaction between monocytes/macrophages andSMCs is integrin-ligand–mediated. The integrina4b1

(VLA-4) is constitutively expressed on monocytes and hasbeen identified to bind to VCAM-1.4–6 Antibody blockade ofthe VCAM-1/VLA-4 pathway has produced favorable resultsin animal models of inflammation.2,11 In a different injurymodel, antibody blockade of VLA-4 reduced intimal hyper-plasia in endarterectomized carotid arteries.26 More recently,treatment of hyperlipidemic mice with an antibody againstthe immune mediator CD40L resulted in reduced aorticatherosclerosis, which was attributed to inhibition of inflam-matory cell accumulation. This was shown to occur becauseof the reduction of VCAM-1 expression.27 The observation in

the present study that VCAM-1 antibody treatment blockedthe recruitment of monocytes/macrophages indicates a neu-tralization of VCAM-1 function. Our results concerning therole of VCAM-1 in inflammation concur with their report.

The partial loss of immunodetectable smooth musclea-actin on arterial sections concomitant with the expressionof VCAM-1 seven days after injury suggests an associationbetween phenotypic modulation and VCAM-1 expression.This association is further strengthened by the partial inhibi-tion of the observable decrease ina-actin staining inVCAM-1 antibody–treated mice (Figure 6C and 6D). Theprocess of losinga-actin expression after injury has beendescribed in the rat balloon deendothelialization model.28 Themost conspicuous changes took place at an early time point (5days after injury), which coincides with our 7-day time point.This observation may be another possible mechanism bywhich VCAM-1 may affect SMC activation, leading to

Figure 4. VCAM-1 antibody is detect-able in 21-day injured mouse serum dot-blotted on membrane. Detection wasaccomplished by using biotinylated anti-rat IgG incubated in streptavidin (A).Sections from 3-day injured mice treatedwith VCAM-1 antibody (Ab) show local-ized presence of the Ab as detected bystain for rat IgG (B, bottom panel).Untreated and IgG-treated mice werenegative for rat IgG stain (B, top andmiddle panels). Computer-assisted mor-phometry shows a lack of significanteffect of VCAM-1 Ab treatment ondetectable VCAM-1 stain (C). The treat-ment reduced the presence of mono-cytes/macrophages in the injured arter-ies of KO HC mice (D). *P,0.05 vs KOHC mice (n53).

Figure 5. Low-power magnification of elastic stain shows the extent of intimal formation in the carotid of KO HC mice 21 days afterinjury (A). Treatment with VCAM-1 antibody (1 mg/g body wt, B) inhibits the formation of intima in injured carotid artery of KO HC mice.IgG treatment had no significant effect (C). Bar5100 mm.

1734 Arterioscler Thromb Vasc Biol. July 2000

by guest on June 29, 2015http://atvb.ahajournals.org/Downloaded from

migration and intimal formation. Li et al4 described a similarfinding in atherosclerotic lesions of rabbits, in which SMCspositive for VCAM-1 had reduceda-actin staining. Morerecently, Duplaa et al29 described the inhibitory effects onSMC markers by VCAM-1/VLA-4 blockade in an in vitromodel of SMC differentiation. However, inhibition of inflam-

matory cell infiltration in our model may also lead to reducedlocal release of growth factors by monocytes/macrophagesknown to modulatea-actin expression.

VCAM-1 was shown to stimulate the chemotaxis ofendothelial cells, which was inhibited by the blockade ofVLA-4. Corneal angiogenesis was shown to be mediated byVCAM-1 in the same study.30 Similar results have beenreported on the adhesion, spreading, and subsequent motilityof a human melanoma cell line on VCAM-1.31 Reduction ofthe distance migrated by injured SMCs produced by theantibody against VCAM-1 in the present study suggests theinvolvement of VCAM-1 in cell motility. The mechanism forthe effect of VCAM-1 on migration was not addressed;however, integrin-ligand interaction is likely involved in thisprocess via VLA-4. A recent report using a similar in vitroinjury model inhibited SMC migration by using an antibodyagainst VLA-4.32 That report, in connection with the presentfindings, suggests a novel role for VCAM-1 in SMC migra-tion after injury that may contribute to neointimal formation.The signaling pathway involving VLA-4 in this process iscurrently unknown.

The injury-induced thickening of the neointima occurredwithout changes in the vessel diameter. The observablechange occurred in the medial area of the KO HC group. Themechanism driving this seems to point toward the combina-tion of injury and substantial increase in total circulating andarterial tissue cholesterol. Although the neointima was re-duced, the increase in medial area was not significantlyaffected by antibody treatment of KO HC mice.

Our results indicate that the inhibition of VCAM-1 func-tion after carotid injury in hypercholesterolemic mice pro-foundly reduces neointimal formation. The mechanism maybe the increased monocyte/macrophage infiltration. There isalso evidence that VCAM-1 is involved in phenotypic mod-ulation that may influence neointimal formation after arterialinjury. Furthermore, in vitro migration of SMCs after injuryis partially mediated by VCAM-1. Inhibition of VCAM-1may be an interesting target for future therapeutic interven-tion against vaso-occlusive disease.

Figure 6. Cells expressing VCAM-1three days after injury are SMCs, asshown by double staining for VCAM-1(red) and a-actin (blue) in WT (A) as wellas KO HC (B) mice, although the a-actinstain in the KO mice seems less. At 7days after injury, areas that have intensestain for VCAM-1 lose a-actin immuno-reactivity in the KO HC mice (C). Anti–VCAM-1 antibody treatment partiallyinhibits the loss of a-actin stain in KOHC mice 7 days after injury (D). The barin panel B represents the top panel; thebar in panel D represents the bottompanel.

Figure 7. Western blot analysis of SMCs shows baselineVCAM-1 expression (UI). Twenty-four hours after injury, there isa slight increase in expression, which peaks 48 hours later (A).Tumor necrosis factor (TNF)-a–treated cells (10 ng/mL) wereused as positive control compared with untreated (UT) cells (B).Injury of SMCs induces migration into the injured area (control,n55). VCAM-1 antibody treatment (50 mg/mL) inhibited migra-tion distance by 35% (VCAM-1 Ab, n55). Isotype IgG treatment(IgG, n53) did not affect migration (C).

Oguchi et al VCAM-1 Antibody Inhibits Neointimal Formation 1735

by guest on June 29, 2015http://atvb.ahajournals.org/Downloaded from

AcknowledgmentsThe work described here was supported in part by generouscontributions to the Atherosclerosis Research Center at Cedars-SinaiMedical Center from the Grand Foundation of Los Angeles, HenryWheeler Research Fund, Ornest Family Research Fellowship, andthe United Hostesses Charities. K.-Y.C. is the recipient of a researchfellowship award from the American Heart Association, Greater LosAngeles Affiliate. The authors thank Xiao-Ou Helen Xu for excellenttechnical support.

References1. Henseleit U, Steinbrink K, Sunderkotter C, Goebeler M, Roth J, Sor C.

Expression of murine VCAM-1 in vitro and in different models ofinflammation in vivo: correlation with immigration of monocytes.ExpDermatol. 1994;3:249–256.

2. Foster CA. VCAM-1/a4-integrin adhesion pathway: therapeutic target forallergic inflammatory disorders.J Allergy Clin Immunol. 1996;98:S270–S277.

3. Meerscaert J, Furie M. The adhesion molecules used by monocytes formigration across endothelium include CD11a/CD18, CD11b/CD18, andVLA-4 on monocytes and ICAM-1, VCAM-1, and other ligands onendothelium.J Immunol. 1995;154:4099–4112.

4. Li H, Cybulsky M, Gimbrone MA Jr, Libby P. Inducible expression ofvascular cell adhesion molecule-1 by vascular smooth muscle cellsinvitro and within rabbit atheroma.Am J Pathol. 1993;143:1551–1559.

5. Couffinal T, Duplaa C, Moreau C, Lamaziere JMD, Bonnet J. Regulationof vascular cell adhesion molecule-1 and intercellular adhesionmolecule-1 in human vascular smooth muscle cells.Circ Res. 1994;74:225–234.

6. Landry DB, Couper L, Bryant SR, Linder V. Activation of the NF-kB andIkB system in smooth muscle cells after rat arterial injury: induction ofvascular cell adhesion molecule-1 and monocyte chemoattractantprotein-1.Am J Pathol. 1997;151:1085–1095.

7. Liao F, Andalibi A, deBeer FC, Fogelman AM, Lusis AJ. Genetic controlof inflammatory gene induction and NF-kB-like transcription factor acti-vation in response to an atherogenic diet in mice.J Clin Invest. 1993;91:2572–2579.

8. Liao F, Andalibi A, Qiao JH, Allayee H, Fogelman AM, LusisAJ. Genetic evidence for a common pathway mediating oxidative stress,inflammatory gene induction, and aortic fatty streak formation in mice.J Clin Invest. 1994;94:877–884.

9. Kornowski R, Hong MK, Tio FO, Bramwell O, Wu H, Leon MB. In-stentrestenosis: contributions of inflammatory responses and arterial injury toneointimal hyperplasia.J Am Coll Cardiol. 1998;31:224–230.

10. Kling D, Fingerle J, Harlan JM, Lobb RR, Lang F. Mononuclear leu-kocytes invade rabbit arterial intima during thickening formation viaCD18- and VLA-4–dependent mechanisms and stimulate smooth musclemigration.Circ Res. 1995;77:1121–1128.

11. Dimayuga P, Zhu J, Oguchi S, Chyu K-Y, Xu X-OH, Yano J, Shah PK,Nilsson J, Cercek B. Reconstituted HDL containing human apoli-poprotein A-1 reduces VCAM-1 expression and neointima formationfollowing periadventitial cuff induced carotid injury in apoE null mice.Biochem Biophys Res Commun. 1999;264:465–468.

12. Hakugawa J, Bae SJ, Tanaka Y, Katayama I. The inhibitory effect ofanti-adhesion molecule antibodies on eosinophil infiltration in cutaneouslate phase response in Balb/c mice sensitized with ovalbumin (OVA).JDermatol. 1997;24:73–79.

13. Aikawa M, Rabkin E, Voglic SJ, Shing H, Nagai R, Schoen FJ, Libby P.Lipid lowering promotes accumulation of mature smooth muscle cellsexpressing smooth muscle myosin heavy chain isoforms in rabbit ather-oma.Circ Res. 1998;83:1015–1026.

14. van Ree JH, van den Broek WJAA, Dahlmans VEH, Wieringa B, FrantsRR, Havekes LM, Hofker MH. Variability in cholesterol content in serumand aortic tissue in apolipoprotein E-deficient mice is comparable ininbred (129/Sv) and outbred (mixed 129/Sv and C57BL/6) mice.Athero-sclerosis. 1995;118:165–167.

15. Zhu Z, Zhang SH, Wagner C, Kurtz A, Maeda N, Coffman T,Arendshorst WJ. Angiotensin AT1B receptor mediates calcium signaling

in vascular smooth muscle cells of AT1A receptor–deficient mice.Hypertension. 1998;31:1171–1177.

16. Calara F, Ameli S, Hultgårdh-Nilsson A, Cercek B, Kupfer J, Hedin U,Forrester J, Shah PK, Nilsson J. Autocrine induction of DNA synthesis bymechanical injury of cultured smooth muscle cells: potential role of FGFand PDGF.Arterioscler Thromb Vasc Biol. 1996;16:187–193.

17. Jovinge S, Hultgårdh-Nilsson A, Regnstrom J, Nilsson J. Tumor necrosisfactor-a activates smooth muscle cell migration in culture and isexpressed in the balloon-injured rat aorta.Arterioscler Thromb Vasc Biol.1997;17:490–497.

18. Chyu K-Y, Dimayuga P, Zhu J, Nilsson J, Kaul S, Shah PK, Cercek B.Decreased neointimal thickening after arterial wall injury in induciblenitric oxide synthase knockout mice.Circ Res. 1999;85:1192–1198.

19. O’Brien KD, McDonald T, Chait A, Allen MD, Alpers CE. Neovascularexpression of E-selectin, intercellular adhesion molecule-1, and vascularcell adhesion molecule-1 in human atherosclerosis and their relation tointimal leukocyte content.Circulation. 1996;93:672–682.

20. Nakashima Y, Raines EW, Plump AS, Breslow JL, Ross R. Upregulationof VCAM-1 and ICAM-1 at atherosclerosis-prone sites on the endothe-lium in the apoE-deficient mouse.Arterioscler Thromb Vasc Biol. 1998;18:842–851.

21. Cercek B, Yamashita M, Dimayuga P, Zhu J, Fishbein M, Kaul S, ShahPK, Nilsson J, Regnstrom J. Nuclear factor-k B activity and arterialresponse to balloon injury.Atherosclerosis. 1997;131:59–66.

22. Yasukawa H, Imaizumi T, Matsuoka H, Nakashima A, Morimatsu M.Inhibition of intimal hyperplasia after balloon injury by antibodies tointercellular adhesion molecule-1 and lymphocyte function-associatedantigen-1.Circulation. 1997;95:1515–1522.

23. Harrison AA, Stocker CJ, Chapman PT, Tsang YT, Huehns TY, GundelRH, Peters AM, Davies KA, Geroge AJ, Robinson MK, Haskard DO.Expression of vascular cell adhesion molecule-1 by vascular endothelialcells in immune and nonimmune inflammatory reactions in the skin.J Immunol. 1997;159:4546–4554.

24. Van Put DJM, Van Osselaer N, De Meyer GRY, Andries LJ, Kockx MM,De Clerck LS, Bult H. The role of polymorphonuclear leukocytes incollar-induced intimal thickening in the rabbit carotid artery.ArteriosclerThromb Vasc Biol. 1998;18:915–921.

25. Matthys KE, Van Hove CE, Kock MM, Andries LJ, Van Osselaer N,Herman AG, Bult H. Local application of LDL promotes intimalthickening in the collared carotid artery of the rabbit.Arterioscler ThrombVasc Biol. 1997;17:2423–2429.

26. Lumsden AB, Chen C, Hughes JD, Kelly AB, Hanson SR, Harker LA.Anti-VLA-4 antibody reduces intimal hyperplasia in the endarterec-tomized carotid artery in nonhuman primates.J Vasc Surg. 1997;26:87–93.

27. Mach F, Schonbeck U, Sukhova GK, Atkinson E, Libby P. Reduction ofatherosclerosis in mice by inhibition of CD40 signalling.Nature. 1998;394:200–203.

28. Kocher O, Gabbiani F, Gabbiani G, Reidy MA, Cokay MS, Peters H,Huttner I. Phenotypic features of smooth muscle cells during the evo-lution of experimental carotid artery intimal thickening: biochemical andmorphologic studies.Lab Invest. 1991;64:459–470.

29. Duplaa C, Couffinhal T, Dufourcq P, Llanas B, Moreauo C, Bonnet J.The integrin very late antigen-4 is expressed in human smooth musclecell: involvement ofa4 and vascular cell adhesion molecule-1 duringsmooth muscle cell differentiation.Circ Res. 1997;80:159–169.

30. Kock AE, Halloran MM, Haskell CJ, Shah MR, Polverini PJ. Angio-genesis mediated by soluble forms of E-selectin and vascular celladhesion molecule-1.Nature. 1995;376:517–519.

31. Mould AP, Askari JA, Craig SE, Garratt AN, Clements J, HumphriesMJ. Integrina4b1-mediated melanoma cell adhesion and migration onvascular cell adhesion molecule-1 (VCAM-1) and the alternativelyspliced IIICS region of fibronectin.J Biol Chem. 1994;269:27224–27230.

32. Slepian MJ, Fritz A, Dehdashti B. Smooth muscle cella4b1 integrin-matrix interactions participate in migration and neointimal thickeningfollowing injury. Circulation. 1998;98(suppl I):I-227. Abstract.

1736 Arterioscler Thromb Vasc Biol. July 2000

by guest on June 29, 2015http://atvb.ahajournals.org/Downloaded from

Shah, Jan Nilsson and Bojan CercekSumito Oguchi, Paul Dimayuga, Jenny Zhu, Kuang-Yuh Chyu, Juliana Yano, Prediman K.

Hypercholesterolemic MiceFormation After Periadventitial Carotid Artery Injury in Genetically

Monoclonal Antibody Against Vascular Cell Adhesion Molecule-1 Inhibits Neointimal

Print ISSN: 1079-5642. Online ISSN: 1524-4636 Copyright © 2000 American Heart Association, Inc. All rights reserved.

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

doi: 10.1161/01.ATV.20.7.17292000;20:1729-1736Arterioscler Thromb Vasc Biol. 

http://atvb.ahajournals.org/content/20/7/1729World Wide Web at:

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

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

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

  http://www.lww.com/reprints

Information about reprints can be found online at: Reprints: 

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

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

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

by guest on June 29, 2015http://atvb.ahajournals.org/Downloaded from