Oxidized Collagen Stimulates Proliferation of Vascular Smooth Muscle Cells

10
Oxidized Collagen Stimulates Proliferation of Vascular Smooth Muscle Cells L. Bac ˇa ´kova ´, 1 J. Wilhelm, 2 J. Herget, 3 J. Novotna ´, 2 and A. Eckhart 2 1 Institute of Physiology, Academy of Sciences, and 2 Department of Biochemistry and 3 Department of Physiology of 2nd Medical School, Charles University, Prague, Czech Republic Received May 27, 1997, and in revised form July 5, 1997 We hypothesize that the vascular smooth muscle proliferation after lung injury results from oxidative damage to the matrix proteins in the walls of pulmonary blood vessels. The smooth muscle cells (SMC) isolated from rat aorta were cultured on the surface coated with oxidized and nonoxi- dized (control) collagen of type I. Oxidation of collagen was induced by UV irradiation and char- acterized by fluorescence tridimensional spectral arrays and by gel electrophoresis. From day 1 to 6 of the experiment, SMC proliferated more rapidly on the oxidized collagen than on the control surface. At high SMC population densities (day 9 of experiment) the difference disappeared. After 10 min of trypsinization the cells growing on oxidized collagen rounded and detached completely from the growth surface. The control cells on nonoxidized collagen detached only after 30 min of tryp- sinization. We conclude that oxidation of collagen of vascular wall matrix may participate in stimu- lation of SMC proliferation after oxidant tissue injury. Key Words: collagen; oxidative lung injury; pulmonary hypertension; vascular smooth muscle proliferation; cell detachment. © 1997 Academic Press INTRODUCTION Lung vessel injury induces pulmonary vascular remodeling. Resulting increase of pul- monary vascular resistance causes pulmonary hypertension. Structural reconstruction of the walls of peripheral pulmonary arteries in chronic pulmonary hypertension is charac- terized by the deposition of connective tissue fibers and the growth of the smooth muscle cells (Reid, 1986). Recent reports show that hypoxic pulmonary hypertension is associated with oxidant stress of lung tissue (Hoshikawa et al., 1995; Nakanishi et al., 1995). Pulmonary hypertension, including morphologic vascular changes, was successfully pre- vented by administration of antioxidants (Hoshikawa et al., 1995). Reactive oxygen species inhibit the growth of the vascular smooth muscle (Fanburg and Lee, 1996). Exposure to chronic hyperoxia, however, leads to morphologic reconstruction of pulmo- nary vasculature which has many features similar to that seen after temporary hypoxia (Reid, 1986). We hypothesize, therefore, that oxidative damage to extracellular matrix of pulmonary blood vessels is the factor which contributes in initiation of the remodeling of the vascular wall in pulmonary hypertension. The logical target is represented by collagen due to its abundance and close apposition to smooth muscle cells. Type I collagen has an important role in morphogenesis of vessels (Sage and Vernon, 1994). The effects of its oxidation on vascular morphology have not been studied yet. Collagen may be exposed in vivo to ROS originating from various sources (e.g., activated phagocytes or nonenzymic reactions catalyzed by iron) and we can assume that the resulting oxidative damage is nonspecific. Therefore we used UV irradiation as a model source of free radicals. It has an advantage over the other methods that it does not introduce any chemicals interfering with the smooth muscle cell metabolism. The changes induced in the collagen molecule by UV irradiation can be easily detected by measure- ment of its native fluorescence (Fujimori, 1989). For that purpose we used in the present EXPERIMENTAL AND MOLECULAR PATHOLOGY 64, 185–194 (1997) ARTICLE NO. MO972219 185 0014-4800/97 $25.00 Copyright © 1997 by Academic Press All rights of reproduction in any form reserved.

Transcript of Oxidized Collagen Stimulates Proliferation of Vascular Smooth Muscle Cells

Oxidized Collagen Stimulates Proliferation of Vascular SmoothMuscle Cells

L. Bacakova,1 J. Wilhelm,2 J. Herget,3 J. Novotna´,2 and A. Eckhart2

1Institute of Physiology, Academy of Sciences, and2Department of Biochemistry and3Department ofPhysiology of 2nd Medical School, Charles University, Prague, Czech Republic

Received May 27, 1997, and in revised form July 5, 1997

We hypothesize that the vascular smooth muscle proliferation after lung injury results fromoxidative damage to the matrix proteins in the walls of pulmonary blood vessels. The smooth musclecells (SMC) isolated from rat aorta were cultured on the surface coated with oxidized and nonoxi-dized (control) collagen of type I. Oxidation of collagen was induced by UV irradiation and char-acterized by fluorescence tridimensional spectral arrays and by gel electrophoresis. From day 1 to 6of the experiment, SMC proliferated more rapidly on the oxidized collagen than on the controlsurface. At high SMC population densities (day 9 of experiment) the difference disappeared. After 10min of trypsinization the cells growing on oxidized collagen rounded and detached completely fromthe growth surface. The control cells on nonoxidized collagen detached only after 30 min of tryp-sinization. We conclude that oxidation of collagen of vascular wall matrix may participate in stimu-lation of SMC proliferation after oxidant tissue injury.

Key Words:collagen; oxidative lung injury; pulmonary hypertension; vascular smooth muscleproliferation; cell detachment.© 1997 Academic Press

INTRODUCTION

Lung vessel injury induces pulmonary vascular remodeling. Resulting increase of pul-monary vascular resistance causes pulmonary hypertension. Structural reconstruction ofthe walls of peripheral pulmonary arteries in chronic pulmonary hypertension is charac-terized by the deposition of connective tissue fibers and the growth of the smooth musclecells (Reid, 1986). Recent reports show that hypoxic pulmonary hypertension is associatedwith oxidant stress of lung tissue (Hoshikawaet al., 1995; Nakanishiet al., 1995).Pulmonary hypertension, including morphologic vascular changes, was successfully pre-vented by administration of antioxidants (Hoshikawaet al., 1995). Reactive oxygenspecies inhibit the growth of the vascular smooth muscle (Fanburg and Lee, 1996).Exposure to chronic hyperoxia, however, leads to morphologic reconstruction of pulmo-nary vasculature which has many features similar to that seen after temporary hypoxia(Reid, 1986). We hypothesize, therefore, that oxidative damage to extracellular matrix ofpulmonary blood vessels is the factor which contributes in initiation of the remodeling ofthe vascular wall in pulmonary hypertension. The logical target is represented by collagendue to its abundance and close apposition to smooth muscle cells. Type I collagen has animportant role in morphogenesis of vessels (Sage and Vernon, 1994). The effects of itsoxidation on vascular morphology have not been studied yet.

Collagen may be exposedin vivo to ROS originating from various sources (e.g.,activated phagocytes or nonenzymic reactions catalyzed by iron) and we can assume thatthe resulting oxidative damage is nonspecific. Therefore we used UV irradiation as amodel source of free radicals. It has an advantage over the other methods that it does notintroduce any chemicals interfering with the smooth muscle cell metabolism. The changesinduced in the collagen molecule by UV irradiation can be easily detected by measure-ment of its native fluorescence (Fujimori, 1989). For that purpose we used in the present

EXPERIMENTAL AND MOLECULAR PATHOLOGY 64, 185–194 (1997)ARTICLE NO. MO972219

185

0014-4800/97 $25.00Copyright © 1997 by Academic PressAll rights of reproduction in any form reserved.

study tridimensional spectral arrays. As the wall of peripheral blood vessels containsmainly collagens of type I and III (Mayne, 1986), we have examined the effects ofcollagen I oxidation on the proliferation of vascular smooth muscle cells isolated from rataorta.

METHODS

Collagen Coating of Cultivation Petri Dishes

Collagen I (Sigma) in the concentration 10mg/ml was exposed to UV radiation bymercury lamp (0.5 mW/cm2) for 120 min. Two mililiters of solution of either oxidized ornonoxidized (control) collagen were applied on polystyrene petri dishes (GAMA, Cˇ eskeBudejovice) and dried in a laminar flow box.

Characterization of Oxidized Collagen

The changes in collagens induced by UV irradiation were investigated by fluorescencemeasurements arranged in tridimensional spectral arrays measured at pH 10. Emissionwas measured continuously between 295 and 500 nm for excitation wavelengths adjustedbetween 250 and 290 nm with a step of 5 nm. The spectra were measured on a Perkin–Elmer LS-5 spectrofluorometer and the spectral arrays were created using a softwaredeveloped in our laboratory. The differential spectrum was produced by arithmetic sub-traction of the spectra of oxidized collagen from the control.

The changes in collagen structure were further investigated by means of SDS–PAGE.Electrophoresis was performed on 7.5% polyacrylamide gels using the Mini-Proteansystem (Bio-Rad), according to the method of Laemmli (Laemmli, 1970), and stained bysilver by the method of Swain and Ross (1995).

Isolation and Culture of Cells

Smooth muscle cells (SMC) were obtained from the intima–media complex of thethoracic aorta of four male Wistar–Kyoto rats (age 8 weeks, weight 220–240 g, Instituteof Physiology, Acad. Sci. Cˇ R) by the explantation method (Bacˇakova et al., 1997).Briefly, the aortas were removed under sterile conditions, and after stripping off theadventitia, small pieces of the intima–media complex (cca 0.5–1 mm3) were digested by0.1% collagenase (SEVAC, Prague) in Dulbecco Minimum Essential Medium (MEM) at37°C and pH 7.4 for 1 h. The explants were then seeded in collagen-coated plastic bottles(NUNC, Denmark, cultivation area of 25 cm2, about 200 explants per bottle) in 2 ml ofDulbecco MEM supplemented with 10% fetal calf serum (FCS) and gentamicin (40mg/ml). The cultures were grown at 37°C and 95% humidity in air atmosphere containing5% CO2. At confluence, the cells displayed a characteristic hill-and-valley pattern typicalfor SMC (Chamley-Campbellet al., 1979) and positive staining to vascular smoothmuscle-specific alpha-actin monoclonal antibody (data not shown; Skalliet al.,1986). Forthe growth assays on oxidized collagen, cells at passages 7 to 10 were used.

Evaluation of SMC Growth on Oxidized Collagen

The SMC (passages 7 to 10) were seeded onto polystyrene petri dishes (GAMA, Cˇ eske,Budejovice, Czech Republic, diameter of 5 cm) coated with air-dried oxidized or controlunmodified collagen of type I. Each dish contained 30,000 cells (i.e., 1530 cells per cm2,respectively) and 3 ml of above mentioned Dulbecco MEM supplemented with FCS andgentamicin. One day after inoculation of cells, the culture medium was changed to removenonadherent cells. In days 1, 2, and 3 after seeding, the cell number was determined by

BACAKOVA ET AL.186

phase-contrast microscopy by counting cells using a calibrated eye-piece grid. For eachcollagen sample, 7 dishes were used, and in every dish, 10 randomly selected areas wereevaluated. Described method of cell counting was not suitable on days 6 and 9 after theseeding. The density of the cell population was too high and multilayered areas, hills, werepresent. Therefore the SMC were trypsinized and counted in a Bu¨rker haemocytometer.

The doubling time was calculated in the exponential phase of growth according to thefollowing formula:

DT 4 (t − to)log 2/log Nt − log Nto,

whereto 4 24 h after seeding;t 4 144 h after seeding,Nto4 the number of cells at the

time to; Nt 4 the number of cells at the timet.The mitotic index (i.e., the percentage of cells in mitosis from the whole cell number,

M.I.%) was evaluated by phase-contrast microscopy in cultures at the exponential phaseof growth, i.e., on days 2 and 3 after seeding.

Detachment of SMC from the Growth Support by Trypsin

For assessment of the sensitivity of the SMC growing on oxidized and nonoxidizedcollagens to detachment by trypsin, a modified method described by Leeet al. (1993) wasused. After rinsing in phosphate-buffered saline (PBS) the 6- to 9-day-old cultures wereexposed to 0.1% trypsin (Sigma) in PBS (3 ml per dish, 37°C). The cell morphology wasrecorded by phase-contrast microscopy at 5-min intervals and the time needed for com-plete detachment of cells from the growth support was measured.

Statistics

The values are given as means ± SEM. Statistical significance was determined byStudent’st test for unpaired data. ValuesP < 0.05 were considered significant.

RESULTS

Characterization of Collagen I Oxidation

The tridimensional spectral arrays of collagen I are illustrated in Fig. 1. Figure 1Arepresents the control protein, Fig. 1B shows the situation after 120 min of UV irradiation.The position of emission maxima are indicated in attached two-dimensional graphs. Thefluorescence of native collagen is governed by emission of tyrosine which has the majoremission band at pH 10 around 360 nm. The sharp peaks around 300 nm are produced bylight dispersion on the protein molecules. It is apparent that after UV irradiation there isan increase in fluorescence around 425 nm. The changes induced by UV irradiation arerevealed in difference spectra (Fig. 2) in which the control spectra were subtracted fromthe spectra of UV-irradiated sample. The negative value of emission at 350 nm indicatesa destruction of tyrosine after irradiation, and the positive value at 430 nm represents anew fluorophore formed as a sequel of collagen oxidation by UV irradiation. UV irradia-tion induced a fragmentation of collagen molecule as is illustrated in Fig. 3. The observedfragments had molecular weights in the range of 90 to 30 kDa.

Morphology and Adherence of SMC on Oxidized Collagens

In the dishes covered with collagen type I the cell shape, degree of cell spreading, andthe number of initially adhering cells (i.e., cells adhering to the growth support at the end

PROLIFERATION OF VASCULAR SMOOTH MUSCLE 187

of lag phase, day 1) was not affected by oxidation of collagen surface. On both oxidizedand native collagen surfaces the cells were spindle-shaped or polygonal and they adheredto the growth support by areas of similar size.

Detachment of SMC from Collagen by Trypsin

The resistance to trypsin detachment was lower in cells growing on oxidized collagenin comparison with control unmodified samples at all intervals of culture. After 10 min oftrypsinization, the vast majority of cells growing on oxidized collagen rounded, detachedfrom the growth substrate, and floated in the culture medium. At the same time, most cells

FIG. 2. The differential emission spectra between oxidized and native collagen I. The spectra of the nativecollagen were subtracted from the UV-irradiated sample. Thus the negative values indicate lower intensity in theirradiated sample and the positive values represent higher intensity of emission at the specified wavelengthsrelative to the native protein. It is apparent that after UV irradiation the emission of tyrosine at 350 nm decreased,while a new fluorophore was formed emitting around 430 nm.

FIG. 1. Fluorescence spectra of collagen I. A represents control sample, B was measured after UV irradiation.The vertical axis indicates relative fluorescence intensity, horizontal axis the emission wavelengths.

BACAKOVA ET AL.188

cultured on unmodified collagen remained polygonal or spindle-shaped, well spread onthe growth substrate, and it was impossible to release them even by shaking and rinsingoff by needle and syringe (Fig. 4). The control cells were completely detached only after30 min of trypsinization followed by shaking and rinsing.

Growth Kinetics of SMC on Oxidized Collagen

After a 1-day-long lag phase, i.e., between the ends of days 1 and 6 of the experiment,the SMC on both collagens were in the exponential phase of growth. They proliferatedmore rapidly on the oxidized collagen. On this substrate, the doubling time was signifi-cantly shorter (21.4 ± 0.1 h compared to 26.0 ± 1.2 h on control unmodified sample;P <0.02) and the mitotic index significantly higher (3.80 ± 0.44% compared to 2.23 ± 0.37%on unmodified collagen). As a result, the cells proliferating on oxidized collagen reachedhigher population densities on days 2, 3, and 6 after seeding (Fig. 5, left panel).

Between days 6 and 9, the proliferation of cells on both collagens was rapidly sloweddown. The increase in cell number from day 6 to 9 was not significant on both collagensamples. Thus, the growth curves began to reach their plateau; i.e., the cells approach theirmaximum population density. The plateau cell population densities (days 6 to 9) rangedfrom 26,500 ± 3,700 cells/cm2 to 33,700 ± 6,100 cells/cm2 for oxidized collagen and from12,100 ± 1,400 to 19,800 ± 2,800 cells/cm2 for control unmodified substrate. At thesedensities, which were about 10-fold higher than the inoculation density of 1,530 cells percm2, the higher growth rate of SMC on oxidized collagen disappeared (Fig. 5, middle andright panel). Similarly, when the cells were seeded at higher inoculation density (15,300cells/cm2), no significant differences in cell population densities on oxidized and controlunmodified collagens were found (data not shown).

FIG. 3. Electrophoretic patterns of native (nat) collagen I and of oxidized protein (ox) after UV irradiation.The low molecular weight fragments are apparent in the UV-irradiated sample below the major collagensubunits.

PROLIFERATION OF VASCULAR SMOOTH MUSCLE 189

FIG. 4. Morphology of SMC growing on oxidized (A, C) and unmodified (B, D) collagen type I after 10 minof trypsinization. (A, B) Only trypsinization; (C, D) trypsinization followed immediately by repeated (five times)shaking and rinsing cells off by needle and syringe containing PBS. Phase contrast, objective magnification, 20×.

BACAKOVA ET AL.190

DISCUSSION

We have shown that isolated vascular smooth muscle cells proliferate more activelywhen they grow on the surface of oxidized collagen than the cells seeded on the unmodi-fied, control collagen surface. In the exponential phase of growth, the doubling time ofSMC on oxidized collagen was shorter and the mitotic index was higher. The higherfrequency of cells in mitosis suggests a shorter cell cycle and/or a higher number ofcycling cells in the population, i.e., a higher growth fraction. The latter two parameterscould be also verified by [3H]thymidine autoradiography, but in this study, the method ofgrowth curves was preferred because it covers a relatively long growth period of severaldays. By this method we revealed that the effect of collagen oxidation on SMC prolif-eration was limited only to relatively low cell population densities, i.e., to no more thanabout 25,000 cells/cm2. At higher cellular densities the proliferative activity in control andexperimental cultures did not differ. Similar results were obtained by Sahata and cowork-ers (Sahataet al., 1990). They showed that differences in the cell population densitybetween vascular SMC cultured on plastic substrates and collagen type I or III decreasedwith the time of duration of culture. The absence of growth differences in higher popu-lation densities may be related to the production of various types of extracellular matrixmolecules by SMC themselves, e.g., collagens, elastin, fibronectin, heparin-like gly-cosaminoglycans (Mayne, 1986; Milliset al., 1985). Secreted material may graduallyseparate the cells from the original experimental surface. Then the proliferative activitywill not differ between the groups. Production of collagen by SMC is higher in the earlyquiescent phase of culture than in the exponential phase of growth (Okadaet al., 1990).Presence of oxidized substances such as linoleic acid hydroperoxide in the culture mediumpotentiates the collagen production by SMC (Nishigakiet al., 1991).

The mechanism of stimulation of the SMC growth by oxidized collagen remains to beelucidated. It was reported that in vascular SMC, the oxidized low density lipids stimulatemitogen-activated protein kinases (Kusuharaet al., 1997), and reactive oxygen species,such as superoxide anion, hydrogen peroxide, and hydroxyl radicals increase autocrineproduction of mitogens in SMC, e.g., insulin-like growth factor (Delafontaine and Ku,1997). Similar effects could be exerted also by oxidized collagen.

FIG. 5. Growth curves of rat aortic SMC seeded at density of 1530 cells/cm2 on oxidized (closed symbols)and unmodified (open symbols) collagens of type I. At lower population densities (days 1 to 3), the cells werecounted in phase contrast using eye-piece grid (left panel,n 4 7). At higher population densities (days 6 to 9),the counting was continued in Burker hemocytometer after trypsinization of cells (middle panel,n 4 3 to 4).Combined figure of both parts of the growth curves is on the right panel.

PROLIFERATION OF VASCULAR SMOOTH MUSCLE 191

The higher proliferation capacity of SMC growing on oxidized collagen was accom-panied by a lower resistance of the cells to the detachment by trypsin. It suggests a highersensitivity of the oxidized collagen to protease digestion. Oxidation of matrix proteins byreactive oxidant species increases their susceptibility to proteolytic enzymes (Monboiseetal., 1988). The alternative explanation could be different expression ofb1 (i.e., a1b1,a2b1, and a3b1) integrin receptors for collagen. The number and size of focal cell-collagen contacts could be lower on oxidized than unmodified collagen. For example, asdemonstrated by immunocytochemical staining of vinculin, talin, and actin in culturedbovine aortic endothelial cells, the areas of focal contact were smaller on type III than ontype I collagen. Interestingly, at the same time, the endothelial cell shape was not mark-edly influenced by collagen type (Semich and Robenek, 1990). In the present study, wealso found similar morphology in cells on oxidized and unmodified collagens togetherwith different adhesive properties.

The integrin receptors on SMC bind to the RGD and DGEA sequence of collagen(Hynes, 1992). During UV irradiation of collagen, amino acids in these sequences couldbe altered by oxidative modification (e.g., by the introduction of carbonyl groups), whichleads to impaired adherence of cells (Mattanaet al., 1997). Also the adsorption of adhe-sion-mediating proteins (vitronectin and fibronectin) from the serum to the collagenmolecule could differ in oxidized and nonoxidized proteins because of their differentphysicochemical surface properties (Howlettet al., 1994). Changes in the sensitivity ofcells to trypsin detachment was also seen after modification of polyurethane growthsupports by ion implantation (Leeet al., 1993).

Extracellular matrix (ECM) regulates cell migration, growth, and differentiation (Scott-Burden, 1994). The signals between ECM and the surrounding cells can be transmitted tothe cytoskeleton through specific interaction with cell-surface integrin receptors (Bissellet al., 1982). A concept of ECM as an active component of tissues able to transmitinformation to the internal domains of surrounding cells was developed by (Ruoslahti,1988). Collagens of various types are important components of ECM. In the main pul-monary artery, collagens type I and III, the major fibrillar collagens, represent 4% of thetotal protein mass (Bishopet al., 1990). Therefore changes in collagen I or III structuremight represent important signal to the neighboring muscle cells.

Low intensity UV irradiation was shown to oxidize different proteins such as gamma-globulin (Wickenset al., 1983) or lipoproteins (Doussetet al., 1990) by a free radicalmechanism. As UV irradiation of proteins involves several ROS (Vile and Tyrrell, 1995),similar results can be expected after exposure of proteins to ROSin vivo. The proteinoxidation can be observed by measuring the native protein fluorescence as shown also inthe present study. The native fluorescence spectra of a protein represent a sensitive toolfor the detection of subtle changes in its structure. The power of this method is furtherincreased after organizing the spectra into tridimensional spectral arrays which can serveas a ‘‘fingerprint’’ for a given compound. Odetti and coworkers (Odettiet al.,1994) founddramatic changes in tridimensional spectra of collagen after its glycation or after incu-bation with aldehydes produced during free radical-initiated lipid peroxidation. In thepresent study we did not use any chemical mediators of protein oxidation like sugars orfatty acids. Collagen was oxidized directly by free radicals produced during UV irradia-tion. This is probably the basis of a different fluorescence pattern observed in our samplesin comparison with those published by Odetti and coworkers (Odettiet al., 1994). Theresults of fluorescence measurements are confirmed by the electrophoretic patterns indi-cating the breaking of collagen chains into low molecular weight fragments.

It was shown earlier that free-radical-oxidized fatty acids can modulate proliferation of

BACAKOVA ET AL.192

various cell types both positively and negatively. More recently, proliferation of endo-thelial cells was inhibited by free radicals produced during hyperglycemia (Curcio andCeriello, 1992). To our knowledge, however, there was no report dealing with the effectsof oxidized proteins.

We conclude that oxidation of type I collagen promotes the proliferation of vascularSMC. The cells growing on the oxidized substrate are more sensitive to the detachmentby proteolytic enzymes and therefore they can more easily escape from the control ofgrowth mediated by cell–matrix interactions (Cornwell and Morisaki, 1982; Morisakietal., 1984). The described findings may contribute to the explanation of the role of oxi-dative damage of vascular wall matrix in the remodeling of pulmonary vasculature afterlung injury.

ACKNOWLEDGMENTSThe work was supported by a grant from the Charles University (263/95/2) and a grant from GACR (305/

96/0068).

REFERENCESBacakova, L., Mares, V., Lisa, V., Bottone, M.-G., Pellicciari, C., and Kocourek, F. (1997). Sex related

differences in the migration and proliferation of rat aortic smooth muscle cells in short and long term culture.In Vitro Cell. Dev. Biol.33, 410–413.

Bishop, J. E., Guerreiro, D., and Laurent, G. J. (1990). Changes in the composition and metabolism of arterialcollagens during the development of pulmonary hypertension in rabbits.Am. Rev. Respir. Dis.141,450–455.

Bissell, M. J., Hall, H. G., and Parry, G. (1982). How does the extracellular matrix direct gene expression?J.Theor. Biol.99, 31–68.

Chamley-Campbell, J., Campbell, G. R., and Ross, R. (1979). The smooth muscle in culture.Physiol. Rev.5,1–61.

Cornwell, D. G., and Morisaki, N. (1982). Fatty acid paradoxes in the control of cell proliferation: Prostaglan-dins, lipids peroxides and cooxidation reactions, Free Radicals in Biology, W. A. Pryor, Academic Press, NewYork, 95–148.

Curcio, F., and Ceriello, A. (1992). Decreased cultured endothelial cell proliferation in high glucose medium isreversed by antioxidants: New insights on the pathophysiological mechanisms of diabetic vascular compli-cations.In Vitro Cell. Dev. Biol.28A, 787–790.

Delafontaine, P., and Ku, L. (1997). Reactive oxygen species stimulate insulin-like growth factor I synthesis invascular smooth muscle cells.Cardiovasc. Res.33, 216–222.

Dousset, N., Negre-Salvayre, A., Lopez, M., Salvaye, R., and Dousset-Blazy, L. (1990). Ultraviolet treatedlipoproteins as model system for the study of the biological effects of lipid peroxides on cultured cell. I.Chemical modifications of ultraviolet treated low density lipoproteins.Biochim. Biophys. Acta1045,219–223.

Fanburg, B. L., and Lee, S.-L. (1996). ‘‘Modulation of Cell Proliferation by Pro-oxidant Stimuli, Nitric Oxideand Radicals in the Pulmonary Vasculature’’ (E. K. Weir, S. L. Archer, and J. T. Reeves, Eds.), pp. 233–242.Futura, Armonk, NY.

Fujimori, E. (1989). Cross linking and fluorescence changes of collagen by glycation and oxidation.Biochim.Biophys. Acta998,105–110.

Hoshikawa, Y., Ono, S., Tanita, S., Sakuma, T., Noda, M., Tabata, T., Ueda, S., Ashino, Y., and Fujimura, S.(1995). Contribution of oxidative stress to pulmonary hypertension induced by chronic hypoxia.NipponKyobu shokkan Gakki Yas33, 1169–1173.

Howlett, C. R., Evans, M. D. M., Walsh, W. R., Johnson, G., and Steele, J. G. (1994). Mechanism of initialattachment of cells derived from human bone to commonly used prosthetic materials during cell culture.Biomaterials15, 213–222.

Hynes, R. O. (1992). Versatility, modulation and signalling in cell adhesion.Cell 69, 11–25.Kusuhara, M., Chait, A., Cader, A., and Berk, B. C. (1997). Oxidized LDL stimulates mitogen-activated protein

kinases in smooth muscle cells and macrophages.Arterioscler. Thromb. Vasc. Biol.17, 141–148.Laemmli, V. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage.Nature

227,680–685.Lee, J.-L., Kaibara, M., Iwaki, M., Sasahe, H., Suzuki, Y., and Kasukabe, M. (1993). Selective adhesion and

proliferation of cells on ion-implanted polymer materials.Biomaterials14, 968–960.Mattana, J., Margiloff, L., and Singhal, P. C. (1997). Metal-catalyzed oxidation of extracellular matrix proteins

disrupts integrin-mediated adhesion of mesangial cells.Biochem. Biophys. Res. Commun.233,50–55.

PROLIFERATION OF VASCULAR SMOOTH MUSCLE 193

Mayne, R. (1986). Colagenous proteins of blood vessels.Atherosclerosis6, 585–593.Millis, A. J., Hoyle, M., Mann, D. M., and Breenan, M. J. (1985). Incorporation of cellular and plasma fibro-

nectins into smooth muscle cell intracellular matrix in vitro.Proc. Natl. Acad. Sci. USA82, 2746–2750.Monboise, J.-C., Gardes-Albert, M., Randoux, A., Borel, J.-P., and Ferradini, C. (1988). Collagen degradation

by superoxide anion in pulse and gama radiolysis.Biochem. Biophys. Acta965,29–35.Morisaki, N., Lindsey, J. A., Stits, J. M., Zhang, H., and Cornwell, D. G. (1984). Fatty acid metabolism and cell

proliferation. V. Evaluation of pathways for the generation of lipid peroxides.Lipids 19, 381–394.Nakanishi, I., Tajima, F., Nakamura, A., Yagura, S. Y., Ookawara, T., Yamashita, H., Suzuki, K., Taniguchi, N.,

and Ohno, H. (1995). Effects of hypobaric hypoxia on antioxidant enzymes in rats.J. Physiol.489,869–876.Nishigaki, K., Yanagi, H., Sasaguri, Y., Moromatsu, M., Haykawa, T., and Yagi, K. (1991). Synthesis of

collagen by cultured smooth muscle cells from rabbit aorta is increased in the presence of linoleic acidhydroperoxide.J. Clin. Biochem. Nutr.11, 183–190.

Odetti, P., Pronzato, M. A., Noberasco, G., Cosso, L., Traverso, N., Cottalasso, D., and Marinari, U. M. (1994).Relationship between glycation and oxidation related fluorescence in rat collagen during aging.Lab. Invest.70, 61–67.

Okada, Y., Katsuda, S., Matoui, Y., Watanabe, H., and Nakanishi, J. (1990). Collagen synthesis by culturedarterial smooth muscle cells during spontaneous phenotypic modulation.Acta Pathol. Jpn.40, 157–164.

Reid, L. M. (1986). Structure and function in pulmonary hypertension. New perceptions.Chest89, 279–288.Ruoslahti, E. (1988). Fibronectin and its receptors.Annu. Rev. Biochem.57, 375–414.Sage, E. H., and Vernon, R. B. (1994). Regulation of angiogenesis by extracellular matrix: The growth and the

glue.J. Hypertension12(suppl 10),S145–S152.Sahata, N., Kawamura, K., and Takebayashi, S. (1990). Effects of collagen matrix proliferation and differen-

tiation of vascular smooth muscle cells in vitro.Exp. Mol. Pathol.52, 171–191.Scott-Burden, T. (1994). Extracellular matrix: The cellular environment.NIPS9, 110–115.Semich, R., and Robenek, H. (1990). Organization of the cytoskeleton and the focal contacts of bovine aortic

endothelial cells cultured on type I and III collagen.J. Histochem. Cytochem.38, 59–67.Skalli, O., Bloom, W., Ropraz, P., Ayarone, B., and Gabbiani, G. (1986). Cytoskeletal remodeling of rat aortic

smooth muscle cells ‘‘in vitro’’ relationship to culture conditions and analogies to in vitro situations.J.Submicrosc. Cytol.18, 481–493.

Swain, M., and Ross, N. W. (1995). A silver stain control for proteins yielding high resolution and transparentbackground in sodium dodecyl sulfate–polyacrylamide gels.Electrophoresis16, 948–951.

Vile, G. F., and Tyrrell, R. M. (1995). UVA radiation induced oxidative damage to lipids and proteins in vitroand in human skin fibroblasts is dependent on iron and singlet oxygen.Free Rad. Biol. Med.18, 721–730.

Wickens, D. G., Norden, A. G., Lunec, J., and Dormandy, T. L. (1983). Fluorescence changes in human gammaglobulin induced by free radical activity.Biochim. Biophys. Acta742,607–616.

BACAKOVA ET AL.194