Exposure of Human Vascular Endothelial Cells to Sustained Hydrostatic Pressure Stimulates Proliferation
Involvement of the αV Integrins
Abstract—The present study investigated the effects of sustained hydrostatic pressure (SHP; up to 4 cm H2O) on human umbilical vein endothelial cell (HUVEC) proliferation, focal adhesion plaque (FAP) organization, and integrin expression. Exposure of HUVECs to SHP stimulated cell proliferation and a selective increase in the expression of integrin subunit αV. The increase in αV was observed as early as 4 hours after exposure to pressure and preceded detectable increases in the bromodeoxyuridine labeling index. Laser confocal microscopy studies demonstrated colocalization of the αV integrin to FAPs. The individual FAPs in pressure-treated cells demonstrated a reduced area and increased aspect ratio and were localized to both peripheral and more central regions of the cells, in contrast to the predilection for the cell periphery in cells maintained under control pressure conditions. The pressure-induced changes in αV distribution had functional consequences on the cells: adhesivity of the cells to vitronectin was increased, and αV antagonists blocked the pressure-induced proliferative response. Thus, the present study suggests a role for αV integrins in the mechanotransduction of pressure by endothelial cells.
- hydrostatic pressure
- human umbilical vein endothelial cell
- cell proliferation
Anumber of disease processes (such as tumor angiogenesis and glaucoma-induced retinal microangiopathy) and normal physiological events (such as developmental growth, wound healing, and ovulation) are characterized by endothelial cell proliferation. Endothelial cell proliferation is stimulated by a variety of growth factors, such as basic fibroblast growth factor (bFGF)1 and vascular endothelial growth factor (VEGF),2 as well as by mechanical forces such as cyclic stretch (as reviewed in Reference 33 ), fluid shear,4 and sustained hydrostatic pressure [SHP]).5 6
The mechanisms by which endothelial cells distinguish SHP and convert this mechanical force to a biochemical response (ie, cell division) remain unclear. Among the candidate mechanotransducers are integrins, a family of transmembrane glycoproteins, which have been shown to initiate a biochemical cascade of responses leading to activation of cytosolic transcription factors and/or to regulation of gene expression (as reviewed in Reference 77 ). Direct mechanical linkage of cytoskeletal elements to adherent integrins may provide another mechanotransduction pathway.8
Recent research has associated intracellular signals from the vitronectin receptors (VnRs; integrins αVβ3 and αVβ5) with endothelial cell proliferation.2 9 Functional blocking antibodies to αVβ3 integrins selectively inhibit endothelial cell proliferation and angiogenesis in vivo in response to bFGF but not to VEGF.9 Conversely, blocking monoclonal antibodies specific for the αVβ5 VnR selectively inhibit the in vivo proliferative effects of VEGF but not of bFGF.2
The effects of SHP on normal, human endothelial cells are poorly understood; for this reason, the present study used human umbilical vein endothelial cells (HUVECs) as a model system. In addition, the use of human cells permitted investigation of the SHP signal transduction pathways of endothelial cells, because suitable immunological reagents specific for human integrins and other intracellular proteins were available. In the present study, we demonstrated that exposure of human endothelial cells to SHP stimulated cell proliferation and αV integrin expression and altered the distribution of focal adhesion plaques (FAPs).
Materials and Methods
The supplies used in the present study were purchased commercially as follows: HUVECs, EGM-UV medium, and bovine brain extract (BBE) were from Clonetics; FBS was from HyClone; medium 199 (M199) was from Gibco BRL; glass coverslips were from VWR; Thermanox coverslips were from Nunc, Inc; human fibronectin and laminin were from Collaborative Biomedical Products; monoclonal antibodies to vinculin (clone hVIN-1), fluorescein-conjugated streptavidin, 4-(2-aminoethyl)benzenesulfonyl fluoride, Hoechst 33258 dye, salmon sperm DNA, 3,3′-diaminobenzidine tablets (FastDAB), human vitronectin, and bovine gelatin were all from Sigma Chemical Co (St Louis, Mo); BSA was obtained from Bayer; monoclonal antibodies to integrin subunit αV (clone LM142), integrin αVβ3 (clone LM609), and integrin αVβ5 (clone P1F6) and polyclonal antisera to integrin subunits αV, β1, β3, and β5 were from Chemicon International (Temecula, Calif); normal sera (horse, goat, rabbit, and mouse), biotinylated secondary antisera, Vectashield mounting medium, and the Vectastain Elite ABC reagents were from Vector Laboratories; 4′,6-diamidino-2-phenylindole (DAPI) was from Calbiochem; BrdU, FixDenat, and anti-BrdU monoclonal antibody were from Boehringer Mannheim (Indianapolis, Ind); AquaMount was from Lerner Laboratories; protein A–agarose beads and the bicinchoninic acid protein assay kit were from Pierce; and the Enhanced ChemiLuminescence detection kit was from Amersham.
HUVECs (passage numbers 2 to 9) were cultured in EGM-UV medium supplemented with 10% FBS and 0.4% BBE. For pressure experiments, HUVECs were trypsinized and seeded at 1/3 confluent density either to borosilicate glass coverslips (for immunocytochemical experiments), Thermanox coverslips (for BrdU labeling experiments), or plastic tissue-culture flasks (for other experiments); all substrates had been precoated with adsorbed human fibronectin (1 μg/mL). Cells were cultured for an additional 16 hours to 3 days before use in experiments. Unless otherwise noted, the cells were cultured in M199 supplemented with 10% FBS and 0.4% BBE for the duration of the experiments. Pressure experiments were conducted in pressure chambers that consisted of sterile, polypropylene cylinders (35-mm diameter; 10-cm height) with a removable, watertight cap on the bottom end. Coverslips (on which HUVECs had been cultured) were secured to the inside of the removable caps before the latter were attached to the pressure chambers. The chambers were then filled with M199 to attain the desired hydrostatic pressure head; the medium was added slowly and gently to minimize shear stress on the cells. Filled chambers were capped loosely with sterile, 35-mm lids and were maintained immobile under standard cell-culture conditions for the duration of the experiments. Previous studies by our group determined the pH, po2, and pco2 characteristics of this model.5 Unless otherwise noted, controls were HUVECs maintained under 0.2 cm H2O pressure, and cells were exposed to 4 cm H2O SHP for various periods. Because a large quantity (25 mL) of medium was required to generate the 4 cm H2O pressure heads in this model, some controls consisted of HUVECs maintained under 0.2 cm H2O pressure in 25 mL of medium in 100-mm Petri dishes; these experiments were conducted at 37°C in a humidified, 5% CO2–95% air atmosphere for periods of 4 hours to 4 days. HUVECs grown on tissue-culture plasticware were either maintained under control conditions or exposed to SHP in M199 at 37°C; these cells were exposed to pressure for various times ranging from 15 minutes to 4 days.
HUVECs on glass coverslips were fixed in 3.7% formalin in PBS for 15 minutes and then permeabilized in 0.1% Triton X-100 in PBS for 5 minutes. Preliminary studies indicated that the monoclonal antibodies to αVβ3 (LM609) and αVβ5 (P1F6) and the polyclonal antisera to β3 and β5 integrins used in the present study did not have adequate sensitivity and/or specificity for their antigens in the formalin-fixed endothelial cells used in the present study; however, the monoclonal antibody to αV (LM142) and the monoclonal antibody to vinculin (hVIN-1) stained single bands on Western blots and recognized antigens in formalin-fixed cells (data not shown). The fixed, permeabilized cells were therefore probed using these antibodies as previously described.10
Microscopy and Image Analysis
The endothelial cells used in the immunocytochemical experiments were examined with a Zeiss LSM 410 laser-scanning confocal microscope (Carl Zeiss) at the University of Connecticut Center for BioImaging Technology, Farmington, Conn. Microscopy was performed with a 63× objective and fluorescein optics (490-nm excitation/520-nm emission). Interference-reflectance microscopy (IRM) was performed with the same microscope and lens. Images were stored digitally as TIFF files and were subsequently analyzed using PhotoShop (Adobe) and ImagePro Plus (Media Cybernetics) software.
Cells in situ were first rinsed in 4°C PBS (containing 1 mmol/L CaCl2 and MgCl2) and then in serum-free M199. Proteins and DNA were extracted by incubating the cells in 4°C extraction buffer [10 mmol/L Tris-HCl, 150 mmol/L NaCl, 2 mmol/L CaCl2, 1 mmol/L 4-(2-aminoethyl)benzenesulfonyl fluoride, 40 U/mL aprotinin, 15 μg/mL leupeptin, 0.36 mmol/L 1,10-phenanthroline, 1% (vol/vol) Nonidet NP-40, and 1% (vol/vol) Triton X-100; pH 7.5]. Protein extracts were precleared by using protein A–agarose beads, and the supernatant was reserved; extracts that were not used immediately after extraction were snap-frozen and stored at −80°C. The quantity of protein in each extract sample was determined using a Pierce bicinchoninic acid protein detection kit according to manufacturer’s directions. Total DNA was eluted from the protein extracts by ethanol precipitation, and DNA was quantified by techniques previously described in the literature.11
Proteins used for Western blot experiments were extracted as follows. Cells in situ were rinsed twice in 4°C PBS, removed with plastic cell scrapers, and pelleted by centrifugation. The supernatant was discarded, and 1 mL of immunoprecipitation buffer (0.625 mmol/L HEPES-KOH, pH 7.5; 22.5 mmol/L potassium acetate; 8 μmol/L EDTA; 10 mmol/L DTT; 40 U/mL aprotinin; 15 μg/mL leupeptin; 15 μg/mL pepstatin; 1 mmol/L 4-(2-aminoethyl)benzenesulfonyl fluoride; 1 mmol/L NaF; and 1 mmol/L NaVO4) was added. The cell pellet was sheared in this buffer by aspiration (10 times) through a 23-gauge needle, precleared with protein A–agarose beads at 4°C for 30 minutes, and cleared of debris and beads by centrifugation (15 000g, 5 minutes). Glycerol was added to 30% final concentration, and extracts that were not used immediately were snap-frozen and stored at −80°C.
Aliquots of protein samples were separated by SDS–polyacrylamide gel electrophoresis. Proteins were transferred from the gels to nitrocellulose membranes by electroelution using a GENIE transfer apparatus (Idea Scientific) for 40 minutes. Proteins bound to the membranes were detected by Western blot with polyclonal antisera to integrin subunits αV, β1, β3, and β5, and p65, as well as with monoclonal antibodies to vinculin, and were visualized with an Enhanced ChemiLuminescence detection kit according to manufacturer’s directions. Membranes were stripped of bound antibodies and reprobed several times with different primary antibodies. Stripping was accomplished by soaking the membranes in a solution of 100 mmol/L β-mercaptoethanol, 62.5 mmol/L Tris-HCl, and 2% SDS at 50°C for 30 minutes. Band densities were quantified using PhotoShop software (Adobe).
BrdU and DAPI Labeling
HUVECs that had been seeded to Thermanox coverslips were either maintained under control pressure conditions or exposed to SHP in M199 for periods of 30 minutes to 4 days. In some experiments, the medium contained 100 nmol/L SB223245 (a selective αV integrin antagonist12 ) to determine the requirement of αV integrins for endothelial cell proliferation. BrdU was added to the medium to a final concentration of 10 μmol/L either 24 hours (to assay for inhibition of proliferation using integrin antagonists) or 4 hours (to assay for the time at which proliferation was upregulated in response to pressure) before the end of the experiment. At the end of the experiment, the cells were fixed and permeabilized in FixDenat solution at 37°C for 30 minutes. The fixed, permeabilized cells were rinsed in PBS and subsequently incubated in anti-BrdU monoclonal antibody solution (1:1000 in PBS) in a 4°C, humidified chamber overnight. Cells were rinsed and incubated in biotin-conjugated anti-mouse antisera at 37°C for 2 hours and subsequently incubated in Vector Elite ABC reagent at 37°C for 1 hour. The cells were rinsed 3 times in PBS and incubated in Sigma FastDAB solution that had been prepared according to manufacturer’s instructions immediately before use; cells were incubated in this solution until a visible, brown precipitate had formed (≈2 minutes), at which time the reaction was quenched with a large volume of distilled water.
Cell proliferation was determined after HUVECs were either maintained under control conditions, exposed to SHP, or maintained under control pressure conditions in 25 mL of medium for 24 hours. These cells were subsequently fixed (as described in the Immunocytochemistry section) and incubated in 5 μg/mL DAPI to stain the nuclei, which were then counted.
The BrdU- or DAPI-labeled cells on coverslips were rinsed in PBS, mounted in AquaMount, and examined with a 40× objective either by transmitted-light or UV fluorescence microscopy. Visual fields thus imaged were stored digitally and analyzed with ImagePro Plus software to count the stained nuclei; counts were averaged over 4 to 6 fields per coverslip and 3 to 6 coverslips per data point.
Cell Cycle Analysis
HUVECs were seeded to fibronectin-coated tissue-culture plasticware at 1/3 confluent density and cultured for 3 days under control pressure conditions. Cells were serum deprived by incubation in M199 containing 1% BSA and no serum for 24 hours. Cells were subsequently incubated in fresh M199 containing either 1% BSA or both 10% serum and 0.4% BBE under control pressure conditions and under 4 cm H2O SHP for 24 hours. HUVECs were subsequently detached from their substrates with 5 mmol/L EDTA in Ca2+/Mg2+-free PBS and then quenched in an equal volume of PBS with Ca2+ and Mg2+. Cells were pelleted by centrifugation and the supernatant was discarded. The cells were resuspended by slow addition of 1 mL 70% ethanol to the pellet under continuous vortexing; these cells were fixed in 70% ethanol at 4°C for 24 hours. Fixed cells were repelleted and incubated in 100 U/mL RNase A (in Ca2+/Mg2+-free PBS) at 37°C for 2 hours, after which time propidium iodide (final concentration 50 μg/mL) was added and the cells incubated for 10 minutes. Fluorescence was analyzed with a FACScan flow cytometer (Becton Dickinson).
Cell Adhesion Experiments
ELISA (96-well, not tissue-culture treated) plates were coated with 1 μg/mL vitronectin, fibronectin, gelatin, or laminin in 0.1 mol/L NaCO3 (pH 9.5) at 37°C for 18 hours and washed with M199 (containing 1% BSA).
HUVECs were either maintained under control conditions or exposed to 4 cm H2O pressure for 4 days. After this time, the cells were detached (as described in the Cell Cycle Analysis section), pelleted by centrifugation, and resuspended in M199 containing 1% BSA; aliquots of these cell suspensions (3×105 cells/mL) were seeded to the protein-coated plates and allowed to adhere for 1 hour under standard cell-culture conditions. At the end of that time, the medium was removed from each plate by flicking, followed by gentle centrifugation (<500 rpm) of the plate, upside-down, for 3 minutes. Cells in each well were counted manually by phase-contrast microscopy and a 10× objective.
Numerical data consisting of multiple comparisons were first analyzed by 1-way ANOVA; if the ANOVA indicated a significant difference between groups, the difference between each control and experimental group was then evaluated with a post hoc Bonferroni’s t test. Numerical data comparing a control versus a single experimental group were evaluated with an unpaired, 2-tailed Student’s t test. Numerical data comparing each control versus a matched experimental group were evaluated with a paired, 1-tailed Student’s t test. For all of these tests, a value of P<0.05 was considered statistically significant.
SHP Stimulated HUVEC Proliferation
Several methods were used to assess the effects of SHP on HUVEC proliferation, including determination of cell numbers, total protein and DNA contents, cell cycle analyses, and BrdU labeling index. Compared with HUVECs maintained under control pressure conditions, exposure of HUVECs to SHP for 4 days resulted in a significant (P<0.05) increase in cell number (from 1.91±0.08×107 to 3.21±0.05×107 cells/175-cm2; mean±SEM, n=4) as well as an increase in total protein and total DNA content (data not shown). Independent experiments also demonstrated that pressure-treated cells showed a significant increase in the number of BrdU-labeled nuclei, observable as early as 24 hours after exposure to elevated pressure (Figure 1A⇓); exposure of HUVECs to large volumes of medium did not affect the number of labeled nuclei (Figure 1B⇓). Cell cycle analysis demonstrated that exposure of HUVECs to SHP for 1 day resulted in an increased percentage of cells in the S/G2/M phases of the cell cycle, with a concomitant decrease in the percentage of cells in the G0/G1 phases (Figure 2⇓). It should be noted that these increases in cell proliferation occurred above a high background proliferation due to the use of culture medium supplemented with serum and BBE.
SHP Altered the Distribution and Concentration of VnRs
The αVβ3 and αVβ5 VnRs play important roles in endothelial cell proliferation2 9 and were investigated in the present study by Western blot, immunocytochemistry, and adhesion assays. The amount of αV, β1, β3, and β5 integrin subunits in HUVEC lysates prepared from cells exposed to SHP for various periods was compared by Western blot analysis (Figure 3A⇓ and 3B⇓). Equal amounts of protein were loaded onto each lane; blots were probed with an antibody to the transcription factor p65; similar band density in each lane demonstrated equal loading of protein across the blots (Figure 3B⇓). Compared with controls, pressure-treated cells exhibited increased total αV, reproducibly detectable after as little as 4 hours of exposure to SHP. Exposure of HUVECs to SHP for 3 days resulted in increased levels of β3 (Figure 3B⇓); this effect, however, was not reproduced consistently in all runs of the experiment. The levels of β5 and β1 were not affected by exposure of HUVECs to SHP at any time examined in the present study (Figure 3A⇓).
SHP Altered αV and Vinculin Distribution
Immunocytochemical techniques were used to determine whether αV and vinculin localization was affected by exposure of HUVECs to SHP. When HUVECs were maintained under control conditions, integrin subunit αV localized predominantly to large, rounded regions around the cell border (Figure 4A⇓). In contrast, after exposure of the cells to SHP for 4 days, αV localized to numerous regions throughout the spread edge of the cells (Figure 4E⇓); these αV-rich regions had an elongated shape (Table⇓). Similarly, regions immunoreactive for vinculin were large, rounded, and localized predominantly to the cell border in HUVECs maintained under control conditions (Figure 4B⇓); after exposure of these cells to pressure, regions immunoreactive for vinculin were located throughout the spread edge of the cells (arrows) and had an elongated shape, as evidenced by the morphology of individual regions (Figure 4F⇓) and confirmed by computer-aided morphology analysis of the aspect ratios of >700 immunoreactive regions (Table⇓). Because vinculin is characteristically associated with FAPs and the distributions of αV and vinculin were similar, IRM was used to determine whether αV colocalized with FAPs. Both under control pressure conditions (Figure 4C⇓ and 4D⇓) and after exposure of the HUVECs to SHP for 4 days (Figure 4G⇓ and 4H⇓), regions of intense αV staining (Figure 4C⇓ and 4G⇓; arrows) colocalized with areas of cell-substrate contact (Figure 4D⇓ and 4H⇓; arrows); the areas of cell-substrate contact were elongated (Figure 4H⇓), identical to the elongated shape of the regions of αV staining in HUVECs that had been exposed to SHP (Figure 4G⇓). In agreement with many published reports, parallel experiments demonstrated that vinculin localized to FAPs (data not shown). Computer-aided morphometric analysis confirmed that, after exposure of HUVECs to SHP, the areas of the individual αV- and vinculin-immunoreactive FAPs were smaller than those in controls (specifically, 23.3±0.5 pixels for αV in cells under pressure versus 44.0±1.3 pixels in control cells; the Table⇓) and were more elongated (2.11±0.03 aspect ratio for αV in cells under pressure versus 1.88±0.02 aspect ratio in control cells; the Table⇓). Cells that had been cultured under control pressure conditions in large volumes of medium (25 mL; similar to that of the hydrostatic pressure heads) had FAP morphology similar to cultures maintained under control pressure conditions in 2 mL medium (data not shown).
Exposure of HUVECs to SHP Selectively Increased Subsequent Cell Adhesion to Vitronectin
The effect of pressure on endothelial cell adhesion to vitronectin-coated surfaces was examined to determine whether the pressure-induced increase in the total pool of αV integrins resulted in a functional change in αV integrin–mediated adhesion. Compared with controls, HUVECs that had been exposed to SHP for 4 days exhibited greater cell adhesion on surfaces that had been precoated with adsorbed vitronectin at concentrations ranging from 0.5 to 10 μg/mL (data not shown). To determine the selectivity of this response, the adhesion of control and pressure-treated HUVECs to fibronectin, laminin, and gelatin was also determined. Pressure-treated cells demonstrated significant increases in adhesion to vitronectin-coated surfaces; in contrast, cell adhesion to laminin-coated, gelatin-coated, and fibronectin-coated surfaces did not change (Figure 5⇓). Furthermore, the adhesion of these cells to the vitronectin-coated surfaces could be completely blocked in the presence of 2.5 μg/mL each LM609 and P1F6 (data not shown). These data are consistent with a pressure-induced, selective increase in functional VnRs.
Integrin αV Inhibitors Blocked Pressure-Induced Cell Proliferation
Cell proliferation was determined by BrdU uptake assays for HUVECs that were either maintained under control conditions or exposed to SHP in the presence and absence of 100 nmol/L SB223245 (a specific αV integrin antagonist12 ) for 4 days. Coincubation of HUVECs with 100 nmol/L SB22324512 blocked the pressure-induced increase in cell proliferation (9.4±0.9 BrdU-labeled nuclei per high-power field for cells under pressure versus 2.1±0.9 for cells under pressure in the presence of SB223245).
Several pathological conditions, including atherosclerosis,13 tumors,14 15 and glaucoma,16 expose endothelial cells to increased pressure and feature enhanced endothelial proliferation/turnover as an etiological component of the disease. In vivo, endothelial cells are exposed to several mechanical forces, namely, fluid shear, substrate tension, and hydrostatic pressure. Exposure of endothelial cells to these mechanical forces in vitro also affects cell morphology and function.3 5 6 17 In particular, SHP has been shown to affect the proliferation of numerous cell types, including endothelial cells,5 6 chondrocytes,18 fibroblasts,18 19 20 and intimal vascular smooth muscle cells.21 22 Although the direction (ie, stimulatory or inhibitory) of the responses of these different cell types to pressure is likely dependent on cell type, culture conditions, and the pressure regimens to which the cells are exposed,19 these results suggest that molecular signal transduction mechanisms can be activated by SHP and that these mechanisms undoubtedly play a role in physiological and pathological processes in vivo.
The present study used a model of SHP to investigate the effects of this mechanical force independent of shear and tensile stresses. The 4 cm H2O pressure used in the present study falls within the low range of physiological hydrostatic pressures to which the endothelial cells may be exposed in vivo. In the body, the total hydrostatic pressure to which vascular endothelial cells are exposed is not the same as the hydrodynamic pressure value, typically measured as “blood pressure” with a sphygmomanometer. The hydrostatic pressure is the net sum of several pressures, including the vascular hydrostatic pressure (ie, the force containing blood within the vessels) and components of the vascular hydrodynamic pressure, as well as the plasma and tissue oncotic pressures and the tissue hydrostatic pressure (as reviewed in Reference 2323 ). Within a given vascular bed, the hydrostatic pressure to which endothelial cells are exposed is directly correlated with, but generally lower than, arterial blood pressure. For example, the normal abdominal aortic pressure of the rat ranges from 132 to 172 cm H2O; however, the measured tissue interstitial pressure (which is a reasonable estimate of the hydrostatic pressure) in the rat kidney ranges from 4 to 29 cm H2O.24 A change in blood pressure (such as transient changes due to alterations in posture and chronic changes due to disease, eg, hypertension) exposes endothelial cells to changes in hydrostatic pressure. Additionally, the parameters that contribute to variations in blood pressure (namely, posture, hypertension, etc) suggest that endothelial cells at various sites of the vascular tree are exposed to different hydrostatic pressures; for example, in the upright human body, the hydrostatic pressure in the veins of the neck is maintained at atmospheric pressure,25 the veins of the skull at subatmospheric pressure,25 and the large vessels of the feet at pressures that may exceed 100 cm H2O above atmospheric.25
The present study provided evidence that HUVECs respond consistently to a physiological pressure of 4 cm H2O. The findings of this study fall into several major categories: (1) SHP stimulates the proliferation of human endothelial cells, even when these cells have a high baseline proliferation due to the presence of 10% FBS and 0.4% BBE in the culture medium; (2) SHP upregulates integrin αV expression; and (3) SHP results in remodeling of FAPs. The stimulatory effects of pressure on endothelial cell proliferation were confirmed by using various techniques, specifically, determination of total cell number, total DNA and protein contents, ratio of cells entering the proliferative phases of the cell cycle, and BrdU labeling index. These observations are in agreement with previous studies that demonstrated increased proliferation of bovine endothelial cells after exposure to SHP.5 6 The present study, however, is the first to demonstrate that exposure of normal, human endothelial cells to SHP resulted in significantly (P≤0.05) increased cell proliferation. Compared with bovine endothelial cell results,5 6 the magnitude of the response of HUVECs to pressure was less pronounced. It is important to note that, in contrast to bovine endothelial cells, culture of HUVECs requires exogenous growth factors (specifically, BBE, a crude extract rich in bFGF); consequently, the observed increase in HUVEC proliferation under SHP occurred over an artificially high baseline rate of cell proliferation, which (in comparison) caused the pressure-induced increases in proliferation to appear smaller.
Exposure of HUVECs to SHP resulted in a 300% increase in the expression of the integrin subunit αV and in a smaller, more variable increase in the expression of integrin subunit β3. This increase was selective for αV expression because expression of integrin subunits β1 and β5 was not affected (Figure 3⇑). Immunostaining and IRM experiments demonstrated that these biochemical changes in integrins were accompanied by changes in the shape and distribution of FAPs (Figure 4⇑ and the Table⇑). In agreement with these biochemical changes, HUVECs exposed to SHP also demonstrated increased adhesion to vitronectin-coated plasticware; adhesion of these cells to other ligands (such as fibronectin, laminin, and gelatin) was not affected (Figure 5⇑). Furthermore, the pressure-induced increase in cell adhesion to vitronectin could be blocked by the combination of 2.5 μg/mL each of LM609 and P1F6 (but not by 5 μg/mL of either antibody alone) in the culture medium (data not shown); this result provided evidence that the altered expression of VnRs that occurs after exposure of HUVECs to pressure may result in changes in the interactions of these cells with proteins of the extracellular matrix.
The observation that a mechanical force (ie, hydrostatic pressure) changes human endothelial cell morphology, FAP composition, and FAP organization is consistent with results of studies that exposed endothelial cells to cyclic strain and shear; however, the details of the biochemical responses of endothelial cells to each of these mechanical stimuli are unique. For example, exposure of bovine endothelial cells to SHP results in changes in cell shape (from polygonal to elongated) over a time course of 4 days.5 In contrast, exposure of bovine and human endothelial cells to cyclic strain and laminar shear stress results in similar polygonal to elongated changes in cell shapes as early as 4 hours after the onset of stimulation.26 27 Furthermore, exposure of HUVECs to cyclic strain does not affect the level (either total or surface) of α2, α5, β1, and β3 integrins27 ; under cyclic strain conditions, however, integrin subunits β1, α5 (cells seeded on collagen), and α2 (cells seeded on fibronectin) are localized to long streaks parallel to the long axes of the cells.27 Bovine aortic endothelial cells exposed to shear stress upregulate α5β1 but downregulate αVβ3 expression and redistribute αVβ3 to the “upstream” end of the cells.26 In contrast, the results of the present study demonstrate that exposure of human endothelial cells to SHP results both in the upregulation and redistribution of specific integrins (namely, αV heterodimers) and in a unique reorganization of FAP morphology and distribution. These differences in integrin distribution and expression in endothelial cells exposed to different external mechanical forces are consistent with the “tensegrity” model proposed by Ingber.28 The tensegrity model hypothesizes that an external mechanical force acting on a cell is transmitted by tensile actin filaments to proteins (specifically, integrins) that anchor the cell to the extracellular matrix28 ; thus, different mechanical forces that apply distinct stresses to endothelial cells could induce specific reorganization of integrins to a pattern that is optimally suited to resist those stresses.
Through tensegrity and/or other mechanisms, integrins and other proteins present in the FAPs (eg, focal adhesion kinase and paxillin) mediate both outside-in and inside-out signaling pathways (as reviewed in References 7 and 297 29 through 31); activation of these pathways leads to altered gene expression, cell migration, and cell proliferation. The αV integrins in particular have been reported to modulate proliferation of endothelial cells in vivo.2 9 32 Blocking monoclonal antibodies to αVβ3 (LM609) and to αVβ5 (P1F6) inhibited angiogenesis in several in vivo models.2 9 In the present investigation, time-course studies demonstrated that increases in total immunoreactive αV could be detected after 4 hours of HUVEC exposure to 4 cm H2O SHP, yet a significant increase in cell proliferation (determined by BrdU labeling index) was not detected until 24 hours, suggesting that alteration of αV preceded the proliferative response. Use of functional blocking antibodies in the present in vitro study, however, suggests that LM609 and P1F6 (at 5 μg/mL) did not affect pressure-induced proliferation (data not shown). It is important to note that a mixture of 2.5 μg/mL each of LM609 and P1F6 antibodies significantly reduced adhesion of HUVECs to vitronectin-coated tissue-culture plasticware (data not shown), suggesting that the concentrations of antibodies used to block integrin-mediated proliferation were sufficient to inhibit attachment of the VnRs to their ligands but may not have effected detachment of adherent VnRs from their ligands. Higher concentrations of these antibodies were not tested due to the very high cost of such experiments. However, 100 nmol/L SB223245 did prevent pressure-induced increases in HUVEC proliferation; SB223245 has been reported in the literature to specifically inhibit interactions of αVβ3 integrins with their ligands.12 It should be noted that the observations with SB223245 must take into account the potential “toxic” effects of αVβ3 inhibitors: detachment of cells from substrates leads to death in anchorage-dependent cells; furthermore blocking of αV integrins in proliferating cells can lead to apoptosis in the absence of cell detachment.9 Although no evidence of extensive cell death was observed in the present study, a reduction in the number of proliferating cells in the presence of αV integrin inhibitors may be due to cell death rather than exclusively to an inhibition of integrin-mediated signaling leading to pressure-induced cell proliferation. Thus, at present, the precise role that αV heterodimers may play in the pressure-induced proliferative response of endothelial cells cannot be unequivocally proven.
Reorganization of adhesion plaques in response to SHP is probably just one of the steps in the mechanotransduction/mechanoresponse to this ubiquitous mechanical force. It is clear that signaling events that occur downstream of integrin activation must also be triggered by hydrostatic pressure. Additionally, a role for either soluble or intracellular growth factors cannot be ruled out. The results of Acevedo et al5 provided evidence that exposure of bovine pulmonary artery endothelial cells to SHP resulted in release of bFGF from the cells and in the accumulation of this growth factor, and perhaps others (for example, VEGF), in the supernatant medium. These conditioned media from pressure-treated cells stimulated subsequent proliferation of cells under control pressure conditions, providing evidence that the released factor(s) mediated the cell proliferation response to SHP. Thus, the proliferative response of HUVECs to SHP could also be due to either increased synthesis and/or release of soluble growth factor(s) or to changes in growth factor signaling. It should be noted that, in the present study, it was necessary for the survival of the HUVECs in culture that the experiments in the present study be carried out in the presence of high concentrations of exogenous growth factors (BBE), a situation that complicated attempts to examine growth factor signaling in the current model.
In summary, the results of the present study demonstrated, for the first time, that a modest (specifically, 4 cm H2O) level of SHP over a period of 1 to 4 days stimulates cell proliferation, changes in integrin expression, and changes in organization of FAPs in human endothelial cells. The present study also suggests that VnRs play a role (to date, not specifically determined) in the pressure-induced increase in endothelial cell proliferation. It is unclear at this time, however, whether changes in αV expression, distribution, and function mediate or occur in response to the observed changes in cell proliferation rate. Further research on the mechanism(s) responsible for regulation of αV synthesis and on the role of various integrin-associated signaling molecules is needed to elucidate pertinent details of the intracellular signaling mechanism(s) that governs pressure-induced increases in endothelial cell proliferation.
The authors wish to thank the Whitaker Foundation for a graduate fellowship to Eric A. Schwartz.
Correspondences to Dr Mary E. Gerritsen, Senior Scientist, Cardiovascular Research, Genentech, Inc, 1 DNA Way, South San Francisco, CA 94080-4990.
- Received September 10, 1998.
- Accepted November 24, 1998.
- © 1999 American Heart Association, Inc.
Friedlander M, Brooks PC, Shaffer RW, Kincaid CM, Varner JA, Cheresh DA. Definition of two angiogenic pathways by distinct αv integrins. Science. 1995;270:1500–1502.
Clark EA, Brugge JS. Integrins and signal transduction: the road taken. Science. 1995;268:233–239.
Wang N, Butler JP, Ingber DE. Mechanotransduction across the cell surface and through the cytoskeleton. Science. 1993;260:1124–1127.
Brooks PC, Clark RA, Cheresh DA. Requirement of vascular integrin αvβ3 for angiogenesis. Science. 1994;264:569–571.
Read MA, Neish AS, Gerritsen ME, Collins T. Postinduction transcriptional repression of E-selectin and vascular cell adhesion molecule 1. J Immunol. 1996;157:3472–3479.
Aiello LP, Avery RL, Arrigg PG, Keyt BA, Jampel HD, Shah ST, Pasquale LR, Thieme H, Iwamoto MA, Park JE, Nguyen HV, Aiello LM, Ferrara N, King GL. Vascular endothelial growth factor in ocular fluid of patients with diabetic retinopathy and other retinal diseases. N Engl J Med. 1994;331:1480–1487.
Stromberg DD, Wiederhielm CA. Intravascular and tissue space oncotic and hydrostatic pressures. In: Kaley G, Altura B, eds. Microcirculation. Baltimore, Md: University Park Press; 1977:187–196.
Guyton AC. Textbook of Medical Physiology. Philadelphia. Pa: Saunders; 1986:1057.
Yano Y, Geibel J, Sumpio BE. Cyclic strain induced reorganization of integrin α5β1 and α2β1 in human umbilical vein endothelial cells. J Cell Biochem. 1997;65:505–513.
Ingber DE. Cellular tensegrity: defining new rules of biological design that govern the cytoskeleton. J Cell Sci. 1993;104:613–627.
Leavesley DI, Schwartz MA, Rosenfeld M, Cheresh DA. Integrin β1- and β3-mediated endothelial cell migration is triggered through distinct signaling mechanisms. J Cell Biol. 1993;121:163–170.