Marie Daniel Lamazière and Jacques Bonnet
Pascale Dufourcq, Huguette Louis, Catherine Moreau, Danièle Daret, Michel R. Boisseau, Jean
Human Atheromatous Plaque
Vitronectin Expression and Interaction With Receptors in Smooth Muscle Cells From
Print ISSN: 1079-5642. Online ISSN: 1524-4636
Copyright © 1998 American Heart Association, Inc. All rights reserved. Greenville Avenue, Dallas, TX 75231
is published by the American Heart Association, 7272
Arteriosclerosis, Thrombosis, and Vascular Biology
doi: 10.1161/01.ATV.18.2.168
1998;18:168-176
Arterioscler Thromb Vasc Biol.
http://atvb.ahajournals.org/content/18/2/168
World 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 online
Arteriosclerosis, Thrombosis, and Vascular Biology
Information about subscribing to Subscriptions:
http://www.lww.com/reprints
Information about reprints can be found online at: Reprints:
document. Answer
Permissions and Rights Question and under Services. Further information about this process is available in the
permission is being requested is located, click Request Permissions in the middle column of the Web page which Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for
can be obtained via RightsLink, a service of the
Arteriosclerosis, Thrombosis, and Vascular Biology
in
Requests for permissions to reproduce figures, tables, or portions of articles originally published Permissions:
Smooth Muscle Cells From Human Atheromatous Plaque
Pascale Dufourcq, Huguette Louis, Catherine Moreau, Danie`le Daret, Michel R. Boisseau,
Jean Marie Daniel Lamazie`re, Jacques Bonnet
Abstract—Vitronectin (VN) is a plasma glycoprotein that promotes cell attachment and induces migration of human smooth muscle cells (SMCs) in culture. VN has been observed to accumulate in human atherosclerotic plaques, although its origin and role in atherosclerosis are not yet established. In the present experiments, synthesis of VN by intimal cells and its colocalization with receptors,avb3 and avb5, were studied by in situ hybridization and immunohistochemistry on 15 human atherosclerotic plaques from carotid arteries obtained after surgery. Strong VN protein and mRNA expression was observed in the intima and in the media. In the intima, VN mRNA expression was colocalized with SMCs, indicating that these cells produce VN, which may account for its accumulation in atherosclerotic plaques. In SMCs in culture, immunoprecipitation after metabolic labeling demonstrated that human SMCs do synthesize vitronectin. Confocal microscopic examination showed that VN colocalized with its receptors,avb3 and avb5, in the atherosclerotic intima. However, the distribution of the VN receptors on SMCs in culture in contact with VN was different. These observations suggest that VN plays various parts in atherogenesis via different SMC membrane receptors. (Arterioscler Thromb Vasc
Biol. 1998;18:168-176.)
Key Words:vitronectinn vitronectin receptors n integrin n smooth muscle cells n atherosclerosis
A
therosclerosis is characterized by the development of an intimal thickening, which contain monocytes, T lympho-cytes, and SMCs with an accumulation of lipid and extracel-lular matrix proteins.1 Medial SMCs during this process areactivated, lose their differentiated phenotype, proliferate, and migrate into the intima. A number of reports have docu-mented the major role of extracellular matrix proteins, partic-ularly glycoproteins, in these pathological processes.2-4 VN is
one of the glycoproteins accumulated in atherosclerotic plaques,5,6although its origin and role in atherosclerosis have
yet to be elucidated.
VN is present in plasma and serum at about 200 to 300 mg/mL,7but in common with other adhesive proteins, it is
localized in the extracellular matrices of various tissues, includ-ing the vascular wall.8The sites of VN synthesis in vivo are
now being identified, with the liver emerging as a major site, although other normal9and pathological tissues may synthesize
VN.10In the atherosclerotic plaques, the origin of VN could
be due to the diffusion of plasmatic VN into the vascular wall, to VN release by activated platelets through damaged endo-thelium, or to the synthesis of cells participating in the atherosclerotic plaques.
VN is a multifunctional protein with a multiple binding domain that is thought to interact with a variety of plasma and cell proteins.8VN has been shown to be involved in adhesion
and migration of SMCs, neural crest cells, and keratino-cytes.10-12 VN has also been shown to induce neurite
out-growth and differentiation13and to promote myocyte
differ-entiation in Drosophilia embryos.14 In addition, VN may
interact with several critical coagulation and fibrinolysis pro-teins. VN binds and stabilizes PAI-1 in its active conforma-tion15,16and binds to u-PA receptor,17suggesting an additional
role for VN in the control of protease activity. It has also been suggested that VN interacts with the complement activation system.8
VN contains an RGD sequence8that intervenes in binding
to integrins, cell surface receptors composed of noncovalently linkeda and b subunits.18VN recognizes the integrins of the
av-dependent system (avb3, avb5, avb1, and avb8 and the platelet receptoraIIbb3).19It has been demonstrated that VN
induces migration of SMCs in vitro, which depends onavb3 expression in these cells.11avb3 expression has recently been
described in human arterial wall in vivo.20Cellular contacts
with extracellular matrix components via various integrins activate signal transduction pathways.21These ultimately lead
to alterations in gene expression, which underpin the changes in phenotype, state of differentiation, and migratory behavior of cells.22Each of these events are thought to play a part in the
development of atherosclerotic plaques1and point to a role for
VN and its receptors in this pathological process.
The present study was designed to find out which cell type produces VN in atherosclerotic plaques and to pinpoint the colocalizations between VN and its receptors,avb3 and avb5. We present evidence that SMCs express and synthesize VN in human carotid plaques. In addition, the distribution of VN was colocalized with its receptors,avb3 and avb5. In vitro studies
Received June 11, 1997; revision accepted September 6, 1997. From INSERM Unite´ 441, Atherocle´rose, Pessac, France.
Correspondence to Pascale Dufourcq, INSERM Unite´ 441, Atherocle´rose, Avenue du Haut-Le´veˆque, 33600 Pessac, France. © 1998 American Heart Association, Inc.
showed that these receptors bound to VN were not distributed in the same patterns on the surface of SMCs, pointing to a diversity of VN-integrin–triggered mechanisms in these cells.
Methods
Tissue CollectionFifteen specimens of plaque from the internal carotid arteries of patients who underwent surgery for transient ischemic attacks were obtained at surgery. Specimens of radial and mammary arteries obtained during coronary artery surgery and from two aortas obtained during heart transplantation were used as control tissues. All tissues were frozen in isopentane, prechilled in liquid nitrogen, embedded in OCT compound, and stored at270°C.
Antibodies and Adhesive Ligands
mAbs against human VN (VIT-2) were obtained from Sigma Chem-ical Company, and the avb5 integrin (P1F6) was obtained from Gibco-BRL. Anti-av (CLB-706) and anti-b1 (DF5) subunits were purchased from Tebu. Antibodies against human avb3 integrin (LM609) were kindly provided by Dr Cheresh, and anti-vinculin was a gift from Dr M. Glukhova. Human VN and human fibronectin were purchased from Gibco-BRL and Sigma, respectively.
Immunohistochemistry
Seven-micron-thick sections were cut on a cryostat for immunohis-tochemistry and in situ hybridization. They were dried in air, stored at 270°C, and used within 24 hours. Immunohistochemistry was performed by the indirect immunoperoxidase method as previously described.23Serial sections placed on 3-aminopropyl
triethoxysilane-coated slides (Sigma) were fixed in 2% PFA for 10 minutes at room temperature. Endogenous peroxidase activity was blocked by incu-bating the sections in 0.3% hydrogen peroxide. Before staining, cryosections were preincubated for 20 minutes in a blocking solution containing 5% BSA in PBS. Sections were first incubated with primary antibody at appropriate dilutions for 1 hour at room temperature, rinsed with PBS, incubated with biotinylated sheep anti-mouse Ig (Amersham) for 1 hour at room temperature, rinsed in PBS, and covered with streptavidin-horseradish peroxidase complex. After 30 minutes at room temperature, the sections were rinsed in PBS and revealed by incubation with 10% (vol/vol) 3,39-diaminobenzidine tetrahydrochloride dihydrate in stable peroxide substrate buffer (In-terchim). A counterstain of 10% Harris hematoxylin was applied before cover-slipping. The following immunohistochemical controls were performed in each tissue: (1) no primary antibody and (2) preimmune mouse serum as primary monoclonal antibody. The following antibodies were used as cell markers on sections: anti-smooth musclea-actin (1/1000, IA4, Immunotech) to identify SMCs and HAM-56 (1/100, Becton Dickinson) to identify macrophages. An irrelevant isotype-matched immunoglobulin (Sigma) was used as negative control.
To define the colocalization of the VN with its receptors, samples were examined by confocal microscopy. Briefly, the slides were
labeled with anti-integrin mAbs identified by anti-Ig mAb coupled with fluorescein isothiocyanate. After several washes and saturation, the slides were sequentially incubated with anti-VN polyclonal antibody, biotinylated sheep anti-mouse Ig, and finally with strepta-vidin-Texas Red complex (Amersham). The cover slips were mounted in mounting medium for fluorescence H-1000 (Vectashield, Vector Laboratories). Each image was taken at 0.4-mm intervals with a340 oil-immersion objective on a Diaphot THD microscope. Each section was the average of five scans. The typical Z series is composed of optical sections in the x-y optical plane, while the Z series images were projected simultaneously to obtain the final image.
Cloning of VN cDNA and cRNA Preparation for In Situ Hybridization
Total cellular RNA was prepared from confluent human SMC monolayer and was reverse-transcribed. VN-cDNA was amplified by PCR for 35 cycles at 62°C for annealing using primer sets according to the published cDNA sequence for VN24
(CAAGAAGTGCCAGT-GTGA at position 166 and GAAGCTCGAAGATGTCCTCC at position 1019). The 854-bp PCR product obtained was purified, digested into a 650-bp fragment, and subcloned into pBluescript according to standard procedures. The specificity of the PCR product was verified by the DNA sequence of the insert determined using the dideoxy chain-termination method with modified T7 DNA polymer-ase. It corresponded to the published VN sequence.24,25To construct
the riboprobe, pBluescript containing the 650-bp-long human VN cDNA insert was linearized and in vitro-transcribed using T7 and T3 RNA polymerase for sense and antisense, respectively, in the presence of digoxigenin-labeled UTP.
In Situ Hybridization
Seven-micron cryosections were transferred to 3-aminopropyl tri-ethoxysilane-coated slides. Sections were fixed in 4% PFA for 20 minutes, washed, and treated with proteinase K (1mg/mL) in 0.1 M Tris-HCl and 0.05 M EDTA for 20 minutes. The slides were then rinsed in PBS, treated with glycine (2 mg/mL) for 2 minutes, then incubated in triethanolamine buffer, pH 8, for 5 minutes. After several washes, prehybridization buffer was applied for 30 minutes at 37°C. For hybridization, digoxigenin-labeled riboprobes were added, and hybridization was allowed to proceed overnight at 50°C. After hybridization, sections were incubated twice with 50% formamide and 23 SSC at 55°C for 1 hour, washed twice in 23 SSC, treated with RNase A (20 mg/mL) for 30 minutes, and washed twice in 23 SSC, followed by two 13 SSC washes. The slides were immediately processed for immunohistochemical staining with anti-digoxigenin alkaline phosphatase for 90 minutes. After several washes and satura-tion, the slides were incubated in 100 mM Tris-HCl, with 100 mM NaCl, 50 mM MgCl2, and 1 mM levamisole for 5 minutes and
overnight at 37°C with nitroblue tetrazolium chloride and 5-bromo-4-chloro-3-indolyl phosphate solution in the same buffer. After washes, the slides were counterstained with Kernechtrot.
Cell Culture
SMCs were isolated from human adult aorta media by enzyme digestion as previously described23and cultured in Ham’s F10 medium
supplemented with 10% fetal calf serum, 5 mM HEPES, 50 U/mL penicillin, and 50 mg/mL streptomycin at 37°C in a 5% CO2/95% air
atmosphere. The experiments were conducted on SMCs at the first passage.
Immunofluorescence Microscopy
Cellular adhesion assays were performed in eight-chamber slides coated with either human purified VN (10mg/mL) or fibronectin (20 mg/mL) in PBS at 4°C overnight. The wells were then washed with PBS, and nonspecific adherence to the coat wells was blocked with 5% BSA in PBS for 2 hours at 37°C.
Primary human SMCs were removed from the culture plates by brief incubation at 37°C with 0.5 mM EDTA, washed in PBS with Ca21 and Mg21, and resuspended in serum-free medium. The cells were placed in coated wells at a density of 50 000 cell/cm2 in Selected Abbreviations and Acronyms
Ig5 immunoglobulin IL5 interleukin mAb5 monoclonal antibody
PAI5 plasminogen activator inhibitor PCR5 polymerase chain reaction
PFA5 paraformaldehyde
RIPA5 radioimmunoprecipitation assay RGD5 Arg-Gly-Asp
SMC5 smooth muscle cell
SSC5 sodium chloride/sodium citrate TGF5 tumor growth factor
VN5 vitronectin
serum-free medium. Synthesis of extracellular matrix proteins was inhibited by cycloheximide (25mg/mL) 2 hours before and during adhesion assays. After 2 hours of adhesion on extracellular matrix proteins, the cells were fixed (2% PFA) and permeabilized with 0.2% Triton X-100. After saturation with 5% BSA in PBS, the cells were immunolabeled with anti-avb3, -avb5, -b1, -av subunit, or -vitro-nectin mAbs. The samples were then incubated for 1 hour with anti-Ig mAb coupled with fluorescein isothiocyanate. For double immunofluorescence, after application and labeling of the first mAbs as above, the samples were then incubated for 1 hour with a focal contact marker, anti-vinculin (1/50) mAb, followed by biotinylated sheep anti-mouse Ig and finally with streptavidin-Texas Red complex for 30 minutes. The coverslips were mounted in mounting medium for fluorescence H-1000 and examined with a Microphot-FXA microscope.
Immunoprecipitation and Immunoblotting
For immunoprecipitation experiments, primary SMCs were preincu-bated for 24 hours in serum-free medium, for 1 hour in methionine-free medium, and metabolically labeled with (35S) methionine and (35S)
cysteine (100mCi/mL) for 18 hours in serum-free medium. The cell supernatants were collected and clarified by centrifugation. The cells were lysed in RIPA buffer (0.05 M Tris-HCl, pH 7.5, 0.15 M NaCl, 1% Triton X-100, 1% deoxycholate, 1 mM EDTA, and proteinase inhibitors) at 4°C. Both culture media and cell lysates were precleared and immunoprecipitated with polyclonal or monoclonal antibodies anti-VN in a rotary shaker overnight, followed by absorption with protein A-Sepharose for 3 hours at 4°C. As control, VN antibodies were preabsorbed on purified VN (10 mg/mL). The Sepharose particles were collected by centrifugation and washed three times in RIPA buffer and twice in PBS and water. The immunocomplexes were eluted from protein A-Sepharose by boiling in Laemmli buffer and analyzed by 7.5% SDS-polyacrylamide gel electrophoresis under reducing conditions. The gel was either dried and exposed for autoradiography or electrophoretically transferred onto nitrocellulose membranes for protein immunoblotting. In this case, the membranes were saturated with 5% BSA and incubated with anti-VN mAb for 1 hour. After several washes, the bound antibody was detected with biotinylated sheep anti-mouse Ig and finally with streptavidin-horse-radish peroxidase complex. The immunoblots were developed with the chemiluminescence detection system according to the manufac-turer’s recommendations (Renaissance, DuPont New England Nu-clear). Relative mass was determined using high molecular mass standards (Biolabs, New England).
Flow Cytometry
Expression of av, b1 subunits, and avb3 and avb5 integrins on human primary SMCs was determined by flow cytometry. Confluent SMCs were harvested by brief incubation in buffer (6 mM glucose, 5.3 mM KCl, 125 mM NaCl, 18 mM HEPES, 0.13 PBS) with 0.5 mM EDTA. After washes in PBS (with Ca21and Mg21), the cells were incubated with mAbs in PBS for 1 hour at 4°C, washed three times in cold PBS, and resuspended in fluorescein isothiocyanate-conjugated goat anti-mouse Ig for 1 hour at 4°C. After three additional washes, the samples were fixed in 1% PFA. Measurements were made in a Coulter type XL flow cytometer. The mean of the fluorescence channels was evaluated by the computer for each sample. The x axis represents the fluorescence intensity, and the y axis represents the cell number; 10 000 events were analyzed for each sample.
Results
VN Expression in Human Carotid Plaques We analyzed VN expression on human atherosclerotic plaques obtained from specimens of internal carotid artery after end-arteriectomy. Fig 1 shows two human plaques at different stages of progression: a fibrotic plaque comprising SMCs and some macrophages (Fig 1, A through D) and a fibroatheroma lesion with macrophage infiltration and a lipid core (Fig 1, E
through G). Serial sections of endarteriectomy fragments were stained for both VN and cell-specific markers for SMCs, smooth musclea-actin (Fig 1, A and E), macrophages (HAM-56) (Fig 1, C and G), and T lymphocytes (anti-CD3 UCHT1, Immunotech) (data not shown). VN labeling was distributed throughout the media of normal and atherosclerotic arteries (Fig 1, B, F, and H). A punctiform VN staining was observed in the intimal thickening of cellular fibrotic plaques (Fig 1, B and F), contrasting with large diffuse masses of VN in the necrotic core of fibroatheroma lesions (Fig 1, F). We found the same labeling with a second antibody, a polyclonal anti-VN. Moreover, preabsorption of mAb anti-VN with purified VN eliminated the pattern observed on carotids arteries (data not shown). It can be seen on Fig 1, B and F, that the intimal VN was mainly expressed ona-actin-positive areas, suggesting an association with intimal and medial SMCs. In the fibrous cap, VN labeling was colocalized with HAM-56 staining (Fig 1, F). These results indicated that VN expression was associated with SMCs and sometimes with macrophages in the intimal thick-Figure 1. Expression of VN in carotid artery plaques.
Immuno-histochemistry on serial sections of carotid arteries. A and E, a-actin, B and F, VN, and HAM-56 (G), and C, and negative control (D). A through D, Atherosclerotic plaque with cellular inti-mal thickening. E through H, Atherosclerotic plaque with extra-cellular matrix accumulation and necrotic core. VN staining (B and F) was observed in the media and in the intima, associated witha-actin (A and E) and HAM-56 staining (G). In advanced plaque, VN was accumulated in the necrotic core (F). As a con-trol, VN expression was studied on specimens of radial artery (H). i indicates intima; m, media; nc, necrotic core; and arrows, internal elastic lamina.
ening. On control radial and mammary arteries and normal aorta, a punctiform staining of VN was observed in normal media (Fig 1, H).
VN mRNA Expression in Human Carotid Plaques
We and others5have detected VN protein in atherosclerotic
plaques, although its cellular origin has yet to be established. We therefore wished to find out whether intimal cells synthe-size VN in situ. After hybridization with an antisense VN probe, strong mRNA labeling was observed in the intima and media of specimens of human atherosclerotic carotid arteries (Fig 2, A). To identify which cells expressed VN mRNA in the plaques, serial sections were processed for immunohistochem-istry with SMCs (Fig 2, C) and macrophage markers (Fig 2, D). In several cases, the plaques were composed of SMCs (Fig 2, C) with occasional macrophages (Fig 2, D). VN mRNA expression colocalized with a-actin staining and, in rare instances, with HAM-56 staining. Most of the VN mRNA labeling in the intima was found in SMCs. However, in the plaque, whole SMCs did not express VN mRNA. As shown in Fig 2, A and C, medial SMCs also expressed VN mRNA. A sense VN riboprobe failed to hybridize to carotid arteries, confirming the specificity of the antisense riboprobe (Fig 2, B). VN Synthesis by SMCs In Vitro
Total RNA of human cultured SMCs was isolated and reverse transcribed, and VN cDNA was amplified by PCR with specific primers. The specificity of the amplified transcripts was confirmed by cloning and sequencing. To find out whether SMCs synthesized and deposited their own VN protein at the cell surface, immunoprecipitation after metabolic labeling was
performed. Primary human SMCs were cultured until conflu-ence on medium supplemented with serum. At confluconflu-ence, SMCs were maintained for 24 hours in serum-free medium, for an additional hour in methionine-free medium, then labeled with 100 mCu/mL of 35S-methionine/35S-cysteine
medium. The supernatants of cultured SMCs and the cell lysates were immunoprecipitated with a polyclonal anti-VN antibody. On autoradiography (Fig 3, A), two bands identified at 65 and 75 kd were detected. The same results were obtained when the immunoprecipitation was performed with mAb anti-VN. Moreover, the precipitation of 75- to 65-kd polypeptides was inhibited by preabsorption of the antibody with purified VN (data not shown). Immunoblotting with mAb anti-VN on the same membrane revealed two bands that migrated at the same positions as purified human VN (Fig 3, B). The majority of VN protein was immunoprecipitated from the cell lysate, while only a small amount was found in SMC supernatant. Low-molecular-weight polypeptides were de-tected, suggesting that VN was proteolytically cleaved from the larger form. On immunofluorescence with mAb anti-VN, a fibrillary pattern of VN deposits was observed on surface membranes of SMCs (Fig 3, C).
Integrin Expression in Human Carotid Plaques Because VN interacts with different membrane integrin receptors, we looked foravb3 and avb5 on human carotid plaques. Both avb3 and avb5 integrins were expressed in intimal thickening and media (Fig 4, A and B, respectively). On serial sections,avb3 colocalized with a-actin staining in the intima (Fig 4, D). Expression of avb5 was less pro-nounced than that ofavb3 in media but stronger in intima of human atherosclerotic plaques (Fig 4, B). Interestingly, avb5 staining colocalized with both a-actin and HAM-56 staining (Fig 4, D and E, respectively). In normal arteries, SMCs expressed bothavb3 and avb5 (data not shown). On serial sections, VN expression (Fig 4, C) was observed in both the fibrotic acellular part of atherosclerotic plaques and in cellular areas. At higher magnification, VN was clearly colocalized with avb3 on the internal side of the cellular rim of the necrotic core (Fig 4, A), contrasting with the more external expression ofavb5 on the rim (Fig 4, B). VN thus appeared to interact with both avb3 and avb5 in human atherosclerotic plaques.
To pinpoint the interactions between VN and its recep-tors in situ, carotid plaques were double immunostained for VN and integrins and examined by confocal microscopy. Colocalization of red-labeled VN with green-labeled recep-tors was indicated by a yellow coloration. It can be seen in Fig 5 that VN was colocalized with both avb3 (Fig 5, A) andavb5 (Fig 5, B) integrins on 0.4-mm-sections of carotid tissue specimens. In contrast, the colocalization between VN and the b1 subunit in the same area was sparse and scattered in comparison with the massive avb3/VN and avb5/VN colocalization (Fig 5, C).
Integrin Expression In Vitro
Integrin expression on human aortic SMCs in culture was analyzed by flow cytometry (Fig 6). It can be seen in Fig 6, B, that SMCs strongly expressed theav subunit. Analysis of the b Figure 2. Expression of VN mRNA in carotid artery plaques.
Cryosections of human carotid arteries were hybridized to anti-sense (A) and anti-sense (B) VN riboprobe. Hybridization revealed a strong expression of VN mRNA in both intima and media. No staining was observed after hybridization with sense probe. The slides were counterstained with Kernechtrot. On serial sections, immunohistochemical tests were performed witha-actin (C) and HAM-56 (D) mAbs, and sections were counterstained with Harris hematoxylin. VN mRNA colocalized with strong SMC expression and rarely with macrophage expression (original magnification 320).
integrin subunit associated withav showed a higher expression of theavb5 than the avb3 integrin. We did not obtain any direct evidence for avb1 complex expression, although a moderate expression of theb1 subunit was observed, compa-rable to that ofavb3, indicating a relationship between avb1 andav expression.
To determine the distribution of SMC receptors on the membrane surface, cells were placed on VN-coated plates in the presence of cycloheximide to inhibit the synthesis of extracellular matrix proteins. After 2 hours, the cells were
immunostained with a focal contact marker, anti-vinculin mAb (Fig 7, E through G), along with mAb against the av subunit (Fig 7, A), avb3 (Fig 7, B), or avb5 (Fig 7, C) complexes. In the presence of cycloheximide, SMCs were seen to organize focal contacts, as shown by vinculin staining (Fig 7, E through G). As described by Burridge et al,26we observed an
exclusion of the ligand-specific antibody, anti-VN mAb, from focal contacts on cells attached to the VN coat (Fig 7, D). This pointed to an involvement of VN in focal contact organization.
Figure 3. VN synthesis by SMCs in vitro. Immunoprecipitation of SMCs labeled with35S-methionine with polyclonal anti-VN antibody
on total cell lysate (lane 1) and on total supernatant of SMCs (lane 2). The autoradiograph (A) of SDS-polyacrylamide gel electrophore-sis protein separation and the immunoblot (B) exhibit two bands at 65 to 75 kd that migrated at the same position as human purified VN (B, lane 3). Immunofluorescence labeling of cultured SMCs by mAb anti-VN revealed a strong VN staining (C, left). This staining dis-appeared when mAb was preabsorbed on purified VN (C, right).
Furthermore, the two VN receptors were distributed differ-ently on SMC surface membranes in contact with VN.avb3 expression was low and punctiform (Fig 7, B) and not associated with focal contact, evidenced by mAb to vinculin (Fig 7, F). In contrast,avb5 staining was both punctiform and in focal contact (Fig 7, C) and codistributed with vinculin staining (Fig 7, G). Theav subunit pattern revealed a strong focal contact staining (Fig 7, A) as well as a punctate nonfocal staining on some cells. In control experiments with SMCs placed on fibronectin,avb3 expression was in focal contact in association with vinculin staining, whereasavb5 staining was diffuse (data not shown). No labeling was detected in the absence of primary antibody.
Discussion
VN is a multifunctional glycoprotein that appears to play roles in the migration, attachment, and differentiation of cells.11,13
The immunohistochemical and hybridization study presented here provides evidence for the expression of VN in human atherosclerotic arteries. VN protein deposits were observed in fibrous plaques and arterial media. Strong labeling was ob-served deep in the intimal thickening, near the internal elastic
lamina. Moreover, VN seemed to be localized around and sometimes within the necrotic core. In the cellular areas of intimal thickening, most VN protein colocalized with the SMC marker a-actin, suggesting that much of the VN was associated with SMCs in both atherosclerotic plaques and normal media. In addition, some VN protein was colocalized with macrophage HAM-56 staining. Although VN protein was present in normal media, its expression was markedly elevated in the intima of human atherosclerotic plaques. These results are in line with the accumulation of VN observed by numerous authors in human atherosclerotic plaques associated with cell debris, collagen, and elastin and with an activation of complement.6,27In experimental models, an accumulation of
VN expression in arterial wall with no increase in plasma VN has been described in both hypercholesterolemic and mechan-ically injured aortas28as well as in Watanabe heritable
hyper-lipidemic rabbits.29
An accumulation of VN protein within the plaque was demonstrated immunohistochemically, although the source of Figure 4. Integrin expression on carotid artery plaques. Serial
sections of human carotid arteries were immunostained with mAb anti-avb3 (A), avb5 (B), VN (C), a-actin (D), and HAM-56 (E) (original magnification: A through E,35; F through H, 320). i indicates intima; m, media; nc, necrotic core; and arrows, inter-nal elastic lamina.
Figure 5. Confocal analysis. Colocalization of VN (red) and VN
receptors (green),avb3 (A), avb5 (B), and b1 (C) were studied after double immunofluorescence on intimal thickening of human carotid arteries. VN and its receptors were visualized on tissue sections by confocal microscopy. All fluorescence images are the average of five frames per 0.4mm optical section. The colocalization of VN and its receptor was observed in yellow from superimposition of the red and green staining.
Figure 6. VN receptor expression on cultured SMCs. Human
SMCs were immunolabeled with either primary antibody (A) or antibody against theav (B), avb3 (C), avb5 (D), b1 (E), or a5 (F) subunits. A fluorescent secondary antibody was applied, and 10 000 cells were analyzed by flow cytometry. The x axes repre-sent fluorescent intensity, and the y axes number of cells. Iden-tical results were obtained in three independent experiments.
the arterial VN was not clear. It may have diffused from plasma, been secreted by activated platelets or been synthesized by cells in the plaque. To determine whether cells in athero-sclerotic lesions synthesize VN and whether intimal SMCs remodel their extracellular matrix via VN, we carried out an immunolabeling and in situ hybridization study of human atheromatous carotid arteries. We found that a large popula-tion of SMCs and a few macrophages expressed VN mRNA. It is of interest that the VN mRNA expression was not restricted to pathological intimal SMCs, because some expres-sion was also observed in medial SMCs. Therefore, normal medial SMCs as well as SMCs in the intima appear to synthesize VN, which would account for its presence in the arterial wall. Some of the VN in atherosclerotic plaques may be derived from increased VN synthesis by SMCs induced by cytokines or growth factors secreted by cells in the plaques. In this respect, in vitro studies have shown that TGF-b, like IL-6, upregulates VN mRNA and protein in Hep G2 cells.30In our
study, we did not find any induction of VN mRNA in cultured human SMCs after treatment with TGF-b (data not shown). However, we cannot rule out a diffusion of VN from plasma into the arterial wall. Indeed, plasmatic VN may be trapped and internalized by endothelial cells.31This suggested
that VN binds to the endothelial surface and may be involved in receptor-ligand endocytosis.
The expression of VN by SMCs was supported by the in vitro findings on cultured human SMCs. VN mRNA was detected in these cells, and we showed by immunoprecipita-tion after metabolic labeling that they synthesized and depos-ited their own VN protein. Primary cultured SMC expressed the two VN forms at 65 and 75 kd, corresponding to the two well-characterized forms of human VN.32Smaller polypeptides
were also detected and were thought to be proteolytic frag-ments. In this respect, it has been reported that VN can be cleaved and degraded by various enzymes. For example, plasmin cleaves VN into 61- to 63-, 56-, 42-, and 35-kd
polypeptides.33The small amounts of VN protein detected in
SMC supernatant suggested that most of the synthesized VN was associated with cells. The immunofluorescence labeling studies showed that it was anchored on the membrane surface in a fibrillary distribution. We therefore concluded that SMCs produce and deposit their own VN.
With respect to cell-matrix interactions, VN interacts with the cell surface via integrins containing the av subunit.34,35
This integrin subunit is associated with multiple b chains, including b1, b3, b5, and b8. Among these heterodimers, avb3, avb5, and avb1 are expressed on SMCs.36,37 We
examined the localization of VN-binding integrins,avb3 and avb5, on human carotid arteries, and we present evidence that both receptors are expressed in human atherosclerotic plaques. The intimal and medial expression ofavb3 in human arteries has been reported by Hoshiga et al.20We present here the first
evidence of avb5 expression in atherosclerotic lesions. On serial sections in situ,avb5 integrin appeared to be expressed by both SMCs and macrophages, which is consistent with the demonstration of avb5 expression by macrophages38 and
SMCs in vitro.4 Interestingly, avb3 and avb5 staining was
observed in VN-positive areas. To obtain more evidence of a colocalization of VN with its receptors, sections of artery were examined by confocal microscopy. VN was observed together withavb3 and avb5 receptors in the fibrous cap of athero-sclerotic lesions. These results were supported by the in vitro findings that human cultured SMCs express both avb3 and avb5 and adhere to VN. However, it can be seen in Figure that not allavb3 and avb5 integrins were associated with VN ligand. It is known thatavb3 and avb5 interact with several ligands, including fibronectin and osteopontin,4,39,40 These
extracellular matrix proteins are all expressed in atherosclerotic plaques.41,42We found that the VN receptorsavb3 and avb5
and the av integrin subunit exhibited different patterns of expression in contact with VN on the surface of SMCs. The avb3 integrin was observed in a punctiform and diffuse Figure 7. Distribution of VN receptors on SMCs. Integrin distribution on human SMCs adherent to VN. Slides were precoated with 10
mg/mL VN, and SMCs were plated and incubated for 2 hours in the presence of cycloheximide. Cells were fixed, permeabilized, and double immunofluorescence labeled with anti-av (A), avb3 (B), avb5 (C), and anti-vinculin (E through G). An exclusion experiment with mAb anti-VN (D) shows focal contact in negative. Arrows point to focal contact organization.
distribution on VN and was not concentrated in focal contacts. This is not in agreement with the findings of Clyman et al,36
but they used a mAb against b3 subunit and ductal SMCs rather than the arterial SMCs used in the present study. In contrast to the distribution ofavb3, avb5 was observed in both punctiform and in focal adhesion sites of the SMC-VN interactions, contrasting with the diffuse distribution on fi-bronectin. So, SMCavb5 can be organized in focal adhesion plaque in contact with VN. Similarly, theav and b1 subunits (data not shown) were distributed in focal contact, suggesting that avb1 was abundant in focal adhesion sites, although we were unable to precisely localize the complexavb1 itself. The participation ofb1 subunit integrin in focal adhesion sites on SMCs has been reported by others.36 avb3 expression on
SMCs has been demonstrated to be involved in SMC migra-tion.11,37,43 The functional roles of the avb5 integrins on
SMCs are as yet unknown. For instance, it has been demon-strated that keratinocytes migrate on VN viaavb5 only after activation by growth factors.44 Many extracellular matrix
proteins have been shown to transmit signals via integrins, giving rise to a variety of intracellular processes, including tyrosine phosphorylation, an increase in intracellular pH, and calcium flux.21,45-47The distributions of VN receptors on the
cell surface as well as in atherosclerotic plaques suggest that various functional changes may be induced by VN in SMCs via different signal transducers.
The pathophysiological significance of VN in atheroscle-rotic plaques is still unclear, although several lines of evidence point to a role for VN in the cellular processes involved in atherogenesis. Migration of SMCs into the intima layer is an important contributor to the intimal thickening in atheroscle-rotic lesions,1 and it has been shown that VN mediates
chemotactic and haptotatic activities of SMC in vitro.11
Furthermore, the migration on VN is dependent on the vitronectin receptor, avb3,11 which is strongly expressed in
human atherosclerotic plaques. VN could thus play a role in the recruitment of SMCs from the media to the intima. Peptides that bind toavb3 receptor have been found to reduce neointima formation after balloon injury in rabbit and hamster arteries.48,49Althoughavb3 binds several extracellular matrix
proteins, VN could be involved in the regulation of cellular recruitment. Interestingly, PAI-1, which is highly expressed in human atherosclerotic plaques,50may block SMC migration
on VN by hindering its access to the VN receptor.43Indeed,
PAI-1-binding sites for VN overlap the region containing RGD cell attachment sites for the VN receptor.16 The VN
synthesized and secreted by SMCs may thus modulate its effect on migration as a function of its affinities for the two substrates, PAI-1 or the integrinavb3. Moreover, because VN stabilizes PAI-1 in extracellular matrix proteins,15 thereby protecting
them from plasminogen activator-mediated degradation, it may help stabilize plaques.
In conclusion, we show here that (1) VN protein and mRNA are expressed in human carotid artery plaques; (2) SMCs synthesize VN in the plaque, which may account for its accumulation; (3)avb3 and avb5 integrins are expressed in these plaques and colocalize with VN in the intima; and (4) these integrins were differently distributed on the membrane surface. Comprehension of the true functions of VN and its
interactions with its receptors in atherosclerotic lesions will have to await the results of further studies.
Acknowledgments
We thank Dr Francis Belloc for his excellent technical assistance in the experiments with flow cytometry. This study was supported by the Conseil Regional d’Aquitaine and the Reseau de Recherche Clinique INSERM No. 494014.
References
1. Ross R. Cell biology of atherosclerosis. Annu Rev Physiol. 1995;57: 791-804.
2. Hedin U, Bottger BA, Forsberg E, Johansson S, Thyberg J. Diverse effects of fibronectin and laminin on phenotypic properties of cultured arterial smooth muscle cells. J Cell Biol. 1988;107:307-319.
3. Thyberg J, Hedin U, Sjo¨lund M, Palmberg L, Bottger BA. Regulation of differentiated properties and proliferation of arterial smooth muscle cells.
Arteriosclerosis. 1990;10:966-990.
4. Liaw L, Skinner MP, Raines EW, Ross R, Cheresh D, Schwaertz SM. The adhesive and migratory effects of osteopontin are mediated via distinct cell surface integrins. J Clin Invest. 1995;95:713-724.
5. Niculescu F, Rus HG, Vlaicu R. Immunohistochemical localization of C5b-9, S-protein, C3d and apolipoprotein B in human arterial tissues with atherosclerosis. Atherosclerosis. 1987;65:1-11.
6. Guettier C, Hinglais N, Bruneval P, Kazatchkine M, Bariety J. Immuno-histochemical localisation of S protein/vitronectin in human athero-sclerotic versus atheroathero-sclerotic arteries. Virchows Arch A Pathol Anat
His-topathol. 1989;414:309-313.
7. Preissner K, Wassmuth R, Muller-Berghaus G. Physiochemical characteri-sation of human S protein and its function in the blood coagulation system.
Biochem J. 1985;231:349-355.
8. Preisner KT. Structure and biological role of vitronectin. Annu Rev Cell
Biol. 1991;7:275-310.
9. Seiffert D, Iruelaarispe M, Sage E, Loskutoff D. Distribution of vitronectin mRNA during murine development. Dev Dyn. 1995;203:71-79. 10. Gladson C, Wilcox J, Sanders L, Gillespie G, Cheresh D. Cerebral
micro-environment influences expression of the vitronectin gene in astrocytic tumors. J Cell Sci. 1995;108:947-956.
11. Brown SL, Lundgren CH, Nordt T, Fujii S. Stimulation of migration of human aortic smooth muscle cells by vitronectin: implications for athero-sclerosis. Cardiovasc Res. 1994;28:1815-1820.
12. Delannet M, Martin F, Bossy B, Cheresh D, Reichardt L, Duband J. Specific roles of the alpha V beta 1, alpha V beta 3 and alpha V beta 5 integrins in avian neural crest cell adhesion and migration on vitronectin.
Development. 1994;120:2687-2702.
13. Martinez-Morales J, Marti E, Frade J, Rodrigueztebar A. Developmentally regulated vitronectin influences cell differentiation, neuron survival and process outgrowth in the developing chicken retina. Neuroscience. 1995;68: 245-253.
14. Gullberg D, Fessler LI, Fessler JH. Differentiation, extracellular matrix synthesis, and integrin essembly by Drosophilia embryo cells cultured on vitronectin and laminine substrates. Dev Dyn. 1994;199:116-128. 15. Sigurdardottir O, Wiman B. Identification of a PAI-1 binding site in
vitronectin. Biochim Biophys Acta. 1994;1208:104-110.
16. Seiffert D, Ciambrone G, Wagner N, Binder B, Loskotoff D. The somatomedin B domain of the vitronectin. J Biol Chem. 1994;269: 2659-2666.
17. Wei Y, Waltz D, Rao N, Drummond RJ, Rosenberg S, Chapman HA. Identification of the urokinase receptor as an adhesion receptor for vitro-nectin. J Biol Chem. 1994;269:32380-32388.
18. Ruoslahti E. Integrins. J Clin Invest. 1991;87:1-5.
19. Hynes RO. Integrins: versatility, modulation, and sygnaling in cell adhesion. Cell. 1992;69:11-25.
20. Hoshiga M, Alpers CE, Smith LL, Giachelli CM, Schwartz SM.avb3 integrin expression in normal and atherosclerotic artery. Circ Res. 1995;77: 1129-1135.
21. Juliano RL, Haskill S. Signal transduction from the extracellular matrix. J
Cell Biol. 1993;120:577-585.
22. Adams JC, Watt FM. Regulation of development and differentiation by the extracellular matrix. Development. 1993;117:1183-1198.
23. Dupla`a C, Couffinhal T, Dufourcq P, Llanas B, Moreau C, Bonnet J. The integrin very late antigen-4 is expressed in human smooth muscle cell. Circ
Res. 1997;80:159-169.
24. Jenne D, Stanley KK. Nucleotide sequence and organization of the human S-protein gene: repeating peptide motifs in the pexin family and a model for their evolution. Biochemistry. 1987;26:6737-6742.
25. Suzuki S, Oldberg A, Hayman EG, Pierschbacher MD, Ruoslahti E. Complete amino acid sequence of human vitronectin deduced from cDNA. EMBO J. 1985;4:2519-2524.
26. Burridge K, Fath K, Kelly TN, Nuckolls G, Turner C. Focal adhesions: transmembrane junctions between the extracellular matrix and the cytoskeleton. Annu Rev Cell Biol. 1988;4:487-525.
27. Niculescu F, Rus HG, Porutiu D, Ghiurca V, Vlaicu R. Immunoelectron-microscopic localisation of S-protein/vitronectin in human atherosclerotic wall. Atherosclerosis. 1989;78:197-203.
28. Sawa H, Sobel BE, Fujii S. Potentiation by hypercholesterolemia of the induction of aortic intramural synthesis of plasminogen activator inhibitor type-1 by endothelial injury. Circ Res. 1993;73:671-680.
29. Sato R, Komine Y, Imada T, Takano T. Monoclonal antibody EMR 1a/212D recognizing site of deposition of extracellular lipid in atheroscle-rosis: isolation and characterisation of a cDNA clone for the antigen. J Biol
Chem. 1990;265:21232-21236.
30. Koli K, Lohi J, Hautanen A, Keski-Oja J. Enhancement of vitronectin expression in human HepG2 hepatoma cells by transforming growth factor-b1. Eur J Biochem. 1991;199:337-345.
31. Deboer HC, Preissner KT, Bouma BN, Degroot PG. Internalization of vitronectin-thrombin-antithrombin complex by endothelial cells leads to deposition of the complex into the subendothelial matrix. J Biol Chem. 1995;270:30733-30740.
32. Barnes DW, Silnutzer J, See C, Shaffer M. Characterization of human serum spreading factor with monoclonal antibody. Proc Natl Acad Sci
U S A. 1983;80:1362-1366.
33. Kost C, Benner K, Stockmann A, Linder D, Preissner KT. Limited plasmin proteolysis of vitronectin. Eur J Biochem. 1996;236:682-688.
34. Felding-Habermann B, Mueller BM, Romerdahl CA, Cheresh DA. Involvement of integrin av gene expression in human melanoma tumori-genicity. J Clin Invest. 1992;89:2018-2022.
35. Leavesley DI, Ferguson GD, Wayner EA, Cheresh DA. Requirement of the integrin b3 subunit for carcinoma cell spreading or migration on vitronectin and fibrinogen. J Cell Biol. 1992;117:1101-1107.
36. Clyman RI, Mauray F, Kramer RH.b1 and b3 integrins have different roles in the adhesion and migration of vascular smooth muscle cells on extracellular matrix. Exp Cell Res. 1992;200:272-284.
37. Jones JI, Prevette T, Gockerman A, Clemmons DR. Ligand occupancy of the avb3 integrin is necessary for smooth muscle cells to migrate in response to insulin-like growth factor I. Proc Natl Acad Sci U S A. 1996; 93:2482-2487.
38. De Nichilo MO, Yamada KM. Integrin avb5-dependent serine phosphor-ylation of paxillin in cultured human macrophages adherent to vitronectin.
J Biol Chem. 1996;271:11016-11022.
39. Cheresh D. Human endothelial cells synthesized and expressed an Arg-Gly-Asp directed adhesion receptor involved in attachment to fibrinogen and von Willebrand factor. Proc Natl Acad Sci U S A. 1987;84:6471-6475. 40. Smith J, Vestal D, Irwin S, Burke T, Cheresh D. Purification and functional
characterisation of integrinavb5. J Biol Chem. 1990;265:11008-11013. 41. Clausell N, Delima V, Molossi S, et al. Expression of tumour necrosis factor
alpha and accumulation of fibronectin in coronary artery restenotic lesions retrieved by atherectomy. Br Heart J. 1995;73:534-539.
42. Giachelli C, Bae N, Almeida M, Denhardt D, Alpers C, Schwartz S. Osteopontin is elevated during neointima formation in rat arteries and is a novel component of human atherosclerotic plaques. J Clin Invest. 1993;92:1686-1696.
43. Stefansson S, Lawrence DA. The serpin PAI-1 inhibits cell migration by blocking integrinavb3 binding to vitronectin. Nature. 1996;383:441-443. 44. Yebra M, Filardo E, Bayna E, Kawahara E, Becker J, Cheresh D. Induction of carcinoma cell migration on vitronectin by NF-kappa B-dependent gene expression. Mol Biol Cell. 1995;6:841-850.
45. Rosales C, Obrien V, Kornberg L, Juliano R. Signal transduction by cell adhesion receptors. Biochim Biophys Acta. 1995;1242:77-98.
46. Schwartz MA. Integrins as signal transducing receptors. In: Tanaka Y, ed.
Integrins. Boca Raton, Fla: CRC Press, 1994:33– 47.
47. Guan J, Shalloway D. Regulation of focal adhesion-associated protein tyrosine kinase by both cellular adhesion and oncogenic transformation.
Nature. 1992;358:690-692.
48. Choi ETM, Engel LP, Callow ADMP, et al. Inhibition of neointimal hyperplasia by blockingavb3 integrin with a small peptide antagonist GpenGRGDSPCA. J Vasc Surg. 1994;19:125-134.
49. Matsumo H, Stassen JM, Vermylen J, Deckmyn H. Inhibition of integrin function by a cyclic RGD-containing peptide prevents neointima for-mation. Circulation. 1994;90:2203-2206.
50. Lupu F, Heim DA, Bachmann F, Hurni M, Kakkar VV, Kruithof EKO. Plasminogen activator expression in human atherosclerotic lesions.