Growth and density-dependent regulation of NO synthase by the actin cytoskeleton in pulmonary artery endothelial cells
Abstract
We previously reported association of eNOS with actin increases eNOS activity. In the present study, regulation of activity of eNOS by actin cytoskeleton during endothelial growth was studied. We found eNOS activity in PAEC increased when cells grew from preconfluence to confluence. eNOS activity was much greater in PAEC in higher density than those in lower density, suggesting increase in eNOS activity during cell growth is caused by increase in cell density. Although eNOS protein contents were also increased when endothelial cells grew from preconfluence to confluence, magnitude of increase in eNOS activity was much higher than increase in eNOS protein content, suggesting posttranslational mechanisms played an important role in regulation of eNOS activity during endothelial growth. Confocal fluorescence microscopy revealed eNOS was colocalized with G-actin in preconfluent cells in perinuclear region, with both G-actin in perinuclear area and cortical F-actin in plasma membrane in confluent cells. There was more β-actin coimmunoprecipitated with eNOS in Triton X-100-soluble fraction in confluent cells in later growth phase and in high density. Decrease in eNOS association with β-actin by silencing β-actin expression using β-actin siRNA causes inhibition of eNOS activity, NO production, and endothelial monolayer wound repair in PAEC. Moreover, PAEC incubation with cytochalasin D and jasplakinolide resulted in increases in eNOS/actin association and in eNOS activity without changes in eNOS protein content. Yeast two-hybrid experiments suggested strong association between eNOS oxygenase domain and β-actin. These results indicate increase in eNOS association with actin is responsible for greater eNOS activity in confluent PAEC.
the endothelial isoform of nitric oxide synthase (eNOS) catalyzes the reaction to produce nitric oxide (NO) from l-arginine. NO plays an important role in the regulation of vascular tone, platelet aggregation, angiogenesis, and smooth muscle cell proliferation (2, 8, 11). It is known that eNOS is tightly regulated by a variety of transcriptional, posttranscriptional, and posttranslational mechanisms. Recent evidence indicates that the actin cytoskeleton regulates eNOS activity (13, 20, 22). The actin cytoskeleton, which is critical for regulation of endothelial cell integrity and motility, consists of filamentous polymerized actin (F-actin), arranged as a string of uniformly oriented globular actin (G-actin) subunits in a tight helix (4). In endothelial cells, there are two groups of F-actin: stress fibers, which are linked to the cell membrane at focal adhesion sites, and cortical F-actin, which is located in a plasma membrane subdomain (5, 25). Disruption of actin filaments by the Rho inhibitor Clostridium botulinum C3 transferase, by cytochalasin D, by swinholide, or decreases in actin stress fiber formation by overexpression of a dominant negative Rho mutant increases eNOS expression (13, 20, 22). We have reported that the actin cytoskeleton is associated with eNOS protein and plays an important role in posttranslational regulation of eNOS activity (20).
During endothelial growth, the actin cytoskeleton undergoes dramatic reorganization to facilitate cell migration and trafficking of intracellular proteins and organelles. For example, in the transition from a resting state to a proliferative state during endothelial monolayer wound repair, endothelial cells lose cortical actin and actin stress fibers became more prominent (14). A dynamic alteration in eNOS activity is also observed during endothelial cell growth. eNOS activity is much greater in confluent quiescent cells than in preconfluent proliferating cells (10, 24). In the present study, we examined the interaction of the actin cytoskeleton and eNOS during growth of pulmonary artery endothelial cells (PAEC). Our results indicate that there is a direct association of actin with eNOS protein and that increased association of actin with eNOS is responsible for increased eNOS activity in confluent quiescent PAEC.
MATERIALS AND METHODS
Reagents and materials.
Human eNOS cDNA was a kind gift of Dr. Philip Marsden (Univ. of Toronto, Toronto, ON, Canada). Human β-actin cDNA was purchased from BD Bioscience Clontech (Palo Alto, CA). Cytochalasin D and jasplakinolide were purchased from Calbiochem (La Jolla, CA). l-[3H]arginine was from Amersham (Piscataway, NJ). Mouse anti-eNOS antibody and monoclonal antibody against eNOS protein phosphorylated on Ser1177 were obtained from Transduction Laboratory (Lexington, KY). Anti-β-actin monoclonal antibody was obtained from Sigma (St. Louis, MO). Monoclonal anti-HA antibody was from Covance (Richmond, CA), and monoclonal anti-c-Myc antibody was from BD Bioscience Clontech. FITC-labeled mouse IgG was from Jackson ImmunoResearch Laboratories (West Grove, PA). Texas red-phalloidin, DNase I conjugated with Texas red, and 4-amino-5-methylamino-2′,7′-difluorescein (DAF-FM) dye were obtained from Molecular Probes (Eugene, OR).
Preparation of PAEC in different growth stages.
PAEC were obtained from the main pulmonary artery of 6- to 7-mo-old pigs from a local slaughterhouse and were cultured as previously reported (18). PAEC in third and sixth passages were used. Cells in different growth stages were prepared in two ways. First, for time-dependent growth, confluent cells were trypsinized and plated at a density of 0.3 × 106 per 100-mm dish. After an initial incubation for 48 h in RPMI 1640 containing 10% FBS, designated as day 1, cells were harvested every 24 h and frozen at −70°C until all samples were collected. At day 3, the medium was discarded and replaced with fresh RPMI 1640 containing 10% FBS. A change in medium at this point did not affect eNOS activity. Cells reached confluence at day 4 (6 days after plating). Confluence and subconfluence were assessed using an inverted phase-contract microscope at ×200 magnification. Determinations were made after examining 10 randomly selected fields per culture dish. Second, for density-dependent growth, confluent cells were trypsinized and plated in 100-mm dishes at densities of 0.3 × 106, 0.6 × 106, 1.0 × 106, and 1.5 × 106 per dish in RPMI 1640 medium containing 10% FBS. Cells were harvested after 48 h.
Measurement of intracellular NO production and NO release in culture medium.
NO production in intact single PAEC was determined using the NO indicator dye DAF-FM as recently described (27). Confluent and preconfluent (60% of confluence as judged by visual analysis) cells were incubated with 10 μmol/l DAF-FM for 10 min at room temperature after which direct visualization of NO production with the fluorescent indicator was performed using a laser scanning confocal microscope with excitation and emission maxima at 495 and 515 nm, respectively. The fluorescence intensities of five cells randomly selected in one observation field were quantitated using LSM 510 (version 3.0 SP3) software for the Carl Zeiss Laser Scanning Microscope for 30 min. The results are expressed as ratio to control for each cell. NO release in the culture medium was measured by determining nitrate formation using an NO quantitation kit (Active Motif, Carlsbad, CA).
Determination of eNOS activity.
eNOS activity was determined by measuring the formation of l-[3H]citrulline from l-[3H]arginine as previously described (21). Cells were scraped and homogenized in a Tris·HCl buffer containing EDTA and EGTA. Homogenates were centrifuged at 100,000 g for 60 min at 4°C, and total membrane pellets were resuspended in buffer containing 2.5 mmol/l CaCl2. The resulting suspensions were used for monitoring l-[3H]citrulline formation as described previously (21). The specific activity of NOS is expressed as l-citrulline pmol·l−1·min−1·mg protein−1 and reflects eNOS activity because our cells do not exhibit inducible NOS activity (26).
Immunofluorescence confocal microscopy.
Confluent and preconfluent (∼60% confluence) PAEC plated on glass coverslips were fixed in 4% paraformaldehyde and incubated with 0.1% Triton X-100 for 10 min and with 5% goat serum for 30 min. eNOS was stained with monoclonal anti-eNOS antibody labeled with FITC goat anti-mouse IgG. F-actin and G-actin were stained with Texas red-conjugated phalloidin and DNase I, respectively. eNOS and actin were detected using a Zeiss LSM 510 laser scanning confocal microscope.
Separation of Triton X-100-insoluble fraction from Triton X-100-soluble fraction.
Separation of proteins in Triton X-100-soluble and -insoluble fractions was done as described previously (20). Briefly, PAEC were washed with ice-cold PBS and then scraped into 1% Triton X-100 buffer. The lysates were incubated at 4°C for 20 min and centrifuged at 100,000 g for 20 min. The supernatants constituted the Triton X-100-soluble fraction and were used for Western blot analysis and immunoprecipitation with eNOS antibody as described in the following sections. The pellets constituted the Triton X-100-insoluble fractions, which were lysed in RIPA buffer at 4°C for 20 min and centrifuged at 14,000 g for 20 min. The supernatants were then used in Western blot analysis.
SDS-PAGE and Western blotting.
Cells were plated as described above for growth- or density-dependent experiments. Alternatively, preconfluent and confluent cells were exposed to appropriate reagents or to vehicle in concentrations and times indicated. Equal amounts of proteins from Triton X-100-insoluble and -soluble fractions were subjected to SDS-PAGE separations. Proteins were then transferred to nitrocellulose membranes and incubated with appropriate primary and secondary antibodies. Specific proteins on the membrane were visualized using ImmunStar-AP chemiluminescent reagents (Bio-Rad, Hercules, CA). The density of the blots was quantitated by Bio-Rad Fluor-S MultiImager.
Coimmunoprecipitation of eNOS and actin.
To study the association of eNOS with β-actin, coimmunoprecipitation of eNOS and actin was performed. Triton X-100-soluble and -insoluble fractions of total lysates were used for immunoprecipitation as described previously (20). Lysates were incubated with monoclonal anti-eNOS antibody or nonimmune IgG at 4°C overnight and then with 30 μl of EZview protein G Sepharose (Sigma) for 2 h at 4°C. Immunoprecipitates were collected by centrifugation and washed, and we eluted proteins from Sepharose beads by boiling the samples in 30 μl of SDS immunoblotting sample buffer. Sepharose beads were pelleted by centrifugation, and supernatants were analyzed by Western blotting as described above.
Transfection of β-actin small interfering RNA.
PAEC were plated at a density of 1.5 × 105 cells/well in six-well plates. Twenty-four hours later, the cells were transfected with 1 μg of β-actin small interfering RNA (siRNA) or a negative control siRNA (Silencer β-actin siRNA kit, Ambion) using Qiagen RNAiFest transfection reagent in RPMI containing 4% FBS according to the manufacturer's protocol. The ratio of siRNA to transfection reagent was 1:3. Three days after transfection, PAEC were lysed. The protein contents of eNOS and β-actin, eNOS activity, and eNOS association with β-actin were measured.
Assay of endothelial monolayer wound repair.
To evaluate the functional consequence of eNOS association with β-actin, we measured endothelial monolayer wound repair in endothelial cells in which eNOS association with β-actin was reduced by β-actin siRNA. The endothelial monolayer wound repair assay was performed as previously described (19). Three days after transfection of β-actin siRNA, a cell-free wound zone was created by scraping the monolayer with a sterile pipette tip. The wound width of monolayers in millimeters was measured under the microscope. Then monolayers were washed and incubated with RPMI 1640 containing 2% FBS, and then wound width was measured again after 16 h. Endothelial monolayer wound repair distance is expressed as the difference between the width of the wound before and after incubation (in millimeters).
Detection of interaction of eNOS and actin by yeast two-hybrid system.
Interaction of eNOS and actin protein in yeast two-hybrid system Matchmaker II (BD Bioscience Clontech) was performed according to the manufacturer's protocol. Human eNOS oxygenase domain [amino acids (aa) 1-505] cDNA, eNOS reductase domain (aa 506-1,204) cDNA, and a cDNA encoding the middle part of eNOS (aa 471-874) with the Ca2+ calmodulin-binding domain were cloned into a yeast-expressed vector pGBKT-7 to make plasmids BD-oxy, BD-red, and BD-mid. cDNA encoding the full-length human β-actin protein was cloned into another yeast-expressed vector, pADT-7, to make a plasmid AD-β-actin. Yeast strain AH109 cells were cotransformed with either BD-oxy and AD-β-actin, BD-red and AD-β-actin, BD-mid and AD-β-actin, BD-oxy and empty pADT-7, BD-red and empty pADT-7, and BD-mid and empty pADT-7, or AD-β-actin with empty pGBKT-7, respectively. Cotransformation of plasmid containing murine p53 and GAL-4 DNA-BD cDNA and the one containing simian virus 40 (SV40) large T-antigen and GAL-4 DNA-AD cDNA were used as a positive control because p53 protein and SV40 large T-antigen could be coimmunoprecipitated (9), and interaction of both proteins is extremely strong in a yeast two-hybrid system (15). Cotransformation of plasmid containing human lamin-C and GAL-4 DNA-BD cDNA and the one containing SV40 large T-antigen and GAL-4 DNA-AD cDNA were used as a negative control. Lamin-C has been shown not to interact with T-antigen. Confirmation of protein expression and selection of cotransformants with interacting proteins were performed according to the manufacturer's protocol. The strength of interaction was evaluated by counting the number of colonies on the agar plates with high-stringency (4 dropout) SD/-Ade/-His/-Leu/-Trp and medium-stringency (3 dropout) SD/-His/-Leu/-Trp media and by quantitative analysis of β-galactosidase activity in liquid cultures using O-nitrophenyl β-d-galactopyranoside as a substrate according to the manufacturer's protocol (BD Bioscience Clontech). β-Galactosidase activities were expressed as percentage of positive control.
Statistical analysis.
In each experiment, control and experimental endothelial cells were matched for cell line, age, seeding density, number of passages, and number of days postconfluence to avoid variation in tissue culture factors that can influence the measurement of eNOS activity and eNOS protein analysis. Results are shown as means ± SE in experiments. Student's paired t-test was used to determine the significance of differences between the means of treated and control groups. A P value of <0.05 was taken as significant.
RESULTS
eNOS activity and eNOS protein level increased during endothelial cell growth.
To reveal the alteration of eNOS in endothelial cell growth, we plated PAEC at a low density of 0.3 × 106/100-mm dish. At day 1, 48 h after plating, the endothelial cells grew to 50% confluence. At day 2, day 3, and day 4, confluence stages were 75%, 90%, and 100%, respectively, as judged by visual analysis. Cells at day 5 were designated as postconfluent cells. eNOS activity and protein expression in endothelial cells at different growth stages were measured. eNOS activity was up to eightfold higher in cells at day 4 compared with preconfluent cells at day 1, and eNOS activity continued to increase after cells reached confluence (Fig. 1A). Analysis of the total eNOS protein levels during cell growth revealed an elevation of eNOS protein level of approximately twofold in confluent cells (day 4) compared with cells at day 1 without changes in the protein content of glyceraldehyde-3-phosphate dehydogenase (GAPDH), a housekeeping protein (Fig. 1B). eNOS protein levels did not change after cells reached 90% confluence. Comparison of the increase in eNOS protein level to the increase of eNOS activity (Fig. 1C) showed that the increase in eNOS activity was much higher than the protein level increase, suggesting that a posttranslational regulatory mechanism was primarily, but not exclusively, responsible for the increased eNOS activity in confluent (day 4) and postconfluent (day 5) endothelial cells.
Fig. 1.Increase of endothelial nitric oxide synthase (eNOS) activity and eNOS protein level in pulmonary artery endothelial cells (PAEC) during cell growth. PAEC were plated at low density as described in materials and methods. Cells were harvested every 24 h, and then eNOS activity (A) and protein contents of eNOS and GAPDH (B) were measured. C: changes in eNOS protein level (□, left scale) and in eNOS activity (•, right scale). N = 3. #P < 0.05 vs. day 1, *P < 0.05 vs. day 3.
eNOS activity and eNOS protein level increased as cell density increased.
To determine whether the increase of eNOS activity during cell growth is caused by an increase in cell density, we plated PAEC at cell densities of 0.3 × 106, 0.6 × 106, 1.0 × 106, and 1.5 × 106 per 100-mm dish, respectively. Forty-eight hours after being plated, the cells reached 50%, 75%, 90%, and 100% confluence. As shown in Fig. 2A, eNOS activity was significantly higher in higher-density cell cultures than in lower-density cell cultures. eNOS protein content was also elevated in PAEC at high density without changes in the protein content of GAPDH (Fig. 2B). However, the magnitude of the increase in eNOS activity was much greater than that in eNOS protein level (Fig. 2C). These data suggest that the alterations in eNOS activity and protein content during endothelial cell growth are due to increases in cell density.
Fig. 2.Increase of eNOS activity and protein level in PAEC as cell density increases. PAEC were plated at indicated cell numbers per 100-mm dish and harvested 48 h after being plated. Then, eNOS activity (A) and protein contents of eNOS and GAPDH (B) were measured. C: changes in eNOS protein (□, left scale) and in eNOS activity (•, right scale). N = 3. #P < 0.05 vs. 0.3 × 106/dish, *P < 0.05 vs. 1.0 × 106/dish.
Increase of NO production in confluent vs. preconfluent cells.
To evaluate whether higher eNOS activity in confluent cells is associated with increased NO production, NO production in confluent vs. preconfluent (60% confluence) cells was determined using DAF-FM as a fluorescent substrate by using laser scanning confocal microscopy of single cells. As shown in Fig. 3, NO production was higher in confluent cells than in preconfluent cells (60% confluence). The increase in NO production corresponded to the increase in eNOS activity, suggesting that higher eNOS activity in confluent cells resulted in higher NO production.
Fig. 3.Increase of nitric oxide (NO) production in confluent vs. preconfluent PAEC. Cells were plated on glass-bottom 35-mm dishes at 5 × 104 or 1.4 × 105 cells/dish, resulting in 60 and 100% confluent cells after a 48-h incubation. Intracellular NO production in a single cell was determined as described in materials and methods at 0, 5, 15, and 30 min. Ratios of the increase in fluorescence intensity at measured time points referenced to zero minute measurements were calculated. N = 3. *P < 0.05 vs. PAEC 60% confluent.
Colocalization of eNOS with actin in preconfluent and confluent endothelial cells.
Intracellular localization of eNOS protein is an important factor regulating eNOS activity (6, 7, 23). We analyzed the localization of eNOS protein in preconfluent and confluent cells by laser confocal microscopy. As shown in Fig. 4A, in preconfluent cells, F-actin existed in the form of stress fibers. eNOS was localized primarily to the perinuclear area where it colocalized with G-actin (Fig. 4B). In contrast, in confluent cells, F-actin existed in the form of cortical actin (Fig. 4C). eNOS protein colocalized with cortical F-actin at the plasma membrane (Fig. 4C). A large amount of eNOS was still colocalized in confluent PAEC with G-actin in the perinuclear area (Fig. 4D).
Fig. 4.Intracellular localization of eNOS and actin in preconfluent and confluent PAEC. Sixty percent confluent PAEC (A and B) and 100% confluent PAEC (C and D) were stained with anti-eNOS antibody coupled with FITC-labeled goat anti-mouse IgG (A1, B1, C1, and D1), phalloidin (A2 and C2), or DNase I conjugated with Texas red (B2 and D2). Cells were examined with a Zeiss LSM 510 laser scanning confocal microscope. A3, B3, C3, and D3 are overlays of A1, B1, C1, D1 and A2, B2, C2, and D2, respectively. Areas of yellow staining represent colocalization of both proteins. The images shown are representative of the results observed in 3 experiments.
Coimmunoprecipitation of eNOS and β-actin in Triton X-100-insoluble and -soluble fractions in endothelial cells in different growth stages and in cells at different densities.
To analyze eNOS protein interaction with β-actin during endothelial cell growth, we studied the distribution of eNOS and β-actin proteins in the Triton X-100-insoluble fraction, which contains only F-actin, and the Triton X-100-soluble fraction, which contains primarily G-actin. As shown in Fig. 5A, eNOS and β-actin levels in the Triton X-100-insoluble fraction increased during cell growth with maximum at 90% confluence and then decreased in confluent cells. eNOS and β-actin protein levels in the Triton X-100-soluble fraction were increased in a growth-dependent fashion and reached maximum when endothelial cells grew to 90% confluence (Fig. 5A). To further study the association of eNOS with actin in the Triton X-100-soluble fraction from PAEC in different growth stages, coimmunoprecipitation of eNOS and β-actin was evaluated. As shown in Fig. 5, B and C, in preconfluent cells (days 1 and 2), there was a very small amount of β-actin protein coprecipitated with eNOS. During cell growth, the amount of β-actin protein coprecipitated with eNOS dramatically increased, although eNOS protein levels remained steady after endothelial cells reached 90% confluence. We further analyzed the distribution of eNOS and actin proteins between the Triton X-100-insoluble and the Triton X-100-soluble fractions as well as eNOS-actin association in PAEC at different cell densities. We found that eNOS and β-actin levels in the Triton X-100-insoluble fraction increased with increasing cell density (Fig. 6A), correlating with those in growth-dependent experiments, indicating increased eNOS association with β-actin (F-actin) in confluent cells. eNOS protein levels increased in the Triton X-100-soluble fraction with increasing cell density with maximum levels observed in cells at 90% of confluence (Fig. 6A). Analysis of β-actin protein coimmunoprecipitated with eNOS from Triton X-100-soluble fractions revealed increased eNOS association with β-actin (G-actin) in confluent cells (Fig. 6, B and C), similar to the increased eNOS/G-actin interaction observed in cells at later growth stages. We found no significant differences in the amounts of heat shock protein 90 (HSP90) and caveolin-1 proteins coprecipitated with eNOS from Triton X-100-soluble fractions from cells at different densities (Fig. 6B). These data indicate that during cell growth or during the increase of cell density, eNOS/β-actin (G- and F-actin) interaction increased, and these increases were correlated with higher eNOS activity in confluent cells.
Fig. 5.Distribution of eNOS protein in Triton X-100-insoluble and -soluble fractions of cell lysates and eNOS/β-actin coimmunoprecipitation during endothelial cell growth. A: Triton X-100-insoluble and -soluble fractions of PAEC plated as described in materials and methods were analyzed by Western blot using monoclonal antibodies against eNOS and β-actin. B: lysates of Triton X-100-soluble fractions of PAEC were subjected to immunoprecipitation using anti-eNOS antibody. eNOS and β-actin protein contents in the pellets were determined by Western blot analysis. C: bar graph summarizing the changes in the ratio of β-actin to eNOS in immunoprecipitation pellets from 3 experiments. *P < 0.05 vs. day 1. Blots shown are representative of the results observed in 3 experiments. IP, immunoprecipitation antibody; IB, Western blot antibody. The relative density was expressed as ratio to day 1.
Fig. 6.Distribution of eNOS protein in Triton X-100-insoluble and -soluble fractions of cell lysates and eNOS/β-actin coimmunoprecipitation during cell density increase. A: Triton X-100-insoluble and -soluble fractions of PAEC plated as described in materials and methods were analyzed by Western blot analysis using monoclonal antibodies against eNOS and β-actin. B: lysates of Triton X-100-soluble fractions of PAEC were subjected to immunoprecipitation using anti-eNOS antibody. eNOS and β-actin protein contents in the pellets were determined by Western blot analysis. C: bar graph summarizing the changes in the ratio of β-actin to eNOS in immunoprecipitation pellets in 3 experiments. *P < 0.05 vs. 0.3 × 106/dish. Blots shown are representative of the results observed in 3 experiments. The relative density was expressed as ratio to group of 0.3 × 106/dish.
Cytochalasin D-induced disruption of the actin cytoskeleton increases, eNOS-actin interaction, and eNOS activity.
To confirm that changes in the actin cytoskeleton structure specifically influence eNOS activity, we treated PAEC with cytochalasin D, a disrupter of the actin cytoskeleton. We found that treatment of preconfluent (90% confluence) PAEC with 1–7 μM cytochalasin D for 1 h leads to disruption of actin stress fibers without pronounced toxic effect (data not shown). In cytochalasin D-treated cells, eNOS activity and the level of Triton X-100-soluble β-actin protein increased in a dose-dependent fashion (Fig. 7, A and B) without alteration in eNOS protein content. Of note, there was more β-actin protein coimmunoprecipitated with eNOS from Triton X-100-soluble fractions in cytochalasin D-treated endothelial cells (Fig. 7C). These data indicate that cytochalasin-induced increases in the amount of actin in Triton X-100-soluble fractions resulted in greater association of β-actin with eNOS in this fraction and was associated with increased eNOS activity.
Fig. 7.Cytochalasin D (cyto-D) induced increases in eNOS activity and eNOS-β-actin interaction. A: eNOS activity in the total membrane fraction of cells treated with cyto-D (1, 5, 7 μM) for 1 h. B: β-actin protein contents in Triton X-100-soluble fractions. C: Triton X-100-soluble fractions of control and treated cells were subject to immunoprecipitation using anti-eNOS antibody. eNOS and β-actin in the precipitates were measured as described in materials and methods. Data shown are representative of the results of 3 experiments. *P < 0.05 vs. control, **P < 0.01 vs. 5 μM. The relative density was expressed as ratio to control.
Jasplakinolide-induced actin reorganization increases eNOS-actin interaction and eNOS activity.
To continue studying the role of the actin reorganization in regulating the eNOS-actin interaction, we stabilized actin polymers with jasplakinolide, which promotes actin polymerization and prevents an increase in G-actin. In confluent PAEC treated with 10, 50, or 100 nM jasplakinolide for 24 h, the level of Triton X-100-insoluble actin increased in a dose-dependent fashion, and the level of Triton X-100-soluble actin decreased insignificantly (Fig. 8B). Unexpectedly, eNOS activity was much higher in jasplakinolide-treated cells (Fig. 8A). To determine whether this higher eNOS activity was due to alteration in the association of eNOS with β-actin, coimmunoprecipitation of eNOS and β-actin in Triton X-100-soluble fraction was performed. We found that there was more β-actin protein coprecipitated with eNOS from Triton X-100-soluble fractions derived from jasplakinolide-treated cells (Fig. 8C), indicating that the elevation of eNOS activity was caused by increased association of β-actin with eNOS.
Fig. 8.Jasplakinolide promotes eNOS-β-actin interaction and increases eNOS activity. A: eNOS activity in total membrane fractions of jasplakinolide-exposed cells. B: level of β-actin protein in Triton X-100-insoluble and Triton X-100-soluble fractions of PAEC exposed to jasplakinolide at indicated concentrations for 24 h was analyzed by Western blot analysis. C: Triton X-100-soluble fractions of cells lysates were subject to immunoprecipitation with anti-eNOS antibody. eNOS and β-actin protein contents were measured as described in materials and methods. Data shown are representative of the results of 3 independent experiments. *P < 0.05 vs. control. The relative density was expressed as ratio to control.
Silencing β-actin expression decreases eNOS/β-actin association, eNOS activity, and endothelial monolayer wound repair.
To determine whether decreasing association of β-actin to eNOS inhibits eNOS activity, the gene expression of β-actin was silenced using siRNA. As shown in Fig. 9, A and B, transfection of PAEC with β-actin siRNA resulted in a significant decrease in β-actin protein content in cell lysates and in precipitates brought down by antibody directed against eNOS, indicating that silencing β-actin expression reduces β-actin association with eNOS. Moreover, PAEC with reduced β-actin association with eNOS manifested a decrease in eNOS activity (Fig. 9C) and a decrease in NO production (Fig. 10A) and in monolayer wound repair (Fig. 10B). The presence of the NO donor NOC-18 prevented the reduction in monolayer wound repair, suggesting that decreased wound repair is not caused by endothelial structural damage but by decreased NO production.
Fig. 9.Silencing β-actin expression decreases eNOS-β-actin association and eNOS activity. PAEC were transfected with scrambled or β-actin small interfering RNA (siRNA). Three days later, protein contents of eNOS and β-actin (A), eNOS-β-actin association (B), and eNOS activity (C) were measured as described in materials and methods. Blots shown are representative of the results of 3 experiments. *P < 0.05 vs. scrambled. The relative density was expressed as ratio to control.
Fig. 10.Silencing β-actin expression decreases NO production and endothelial monolayer wound repair. PAEC were transfected with scrambled or β-actin siRNA. Three days later, NO production (A) and endothelial monolayer wound repair (B) were examined as described in materials and methods. Results are expressed as means ± SE; n = 3 experiments. *P < 0.05 vs. control, **P < 0.05 vs. control with vehicle, #P < 0.05 vs. scrambled with vehicle.
Direct interaction of eNOS and β-actin in yeast two-hybrid system.
To determine whether the eNOS-actin association was a direct interaction between eNOS and β-actin, yeast two-hybrid experiments were performed. We found that β-galactosidase activity was much higher in the cotransformants of eNOS oxygenase domain and β-actin than in those of eNOS reductase domain and β-actin or in those of eNOS middle part and β-actin (Fig. 11, A and B). Consistent with these results, there were more colony numbers in the cotransformants of eNOS oxygenase domain and β-actin than in those of eNOS reductase domain and β-actin or in those of eNOS middle part and β-actin (Fig. 11C). Cotransformants of plasmids BD-oxy, BD-red, and BD-mid with empty pADT-7 and cotransformants of AD-β-actin with empty pGBKT-7 did not produce any β-galactosidase activity and yeast colonies. These results indicate that eNOS can directly interact with the actin protein. The binding site appears to be located in the eNOS oxygenase domain.
Fig. 11.Direct interaction of eNOS and β-actin in a yeast two-hybrid system. β-galactosidase assay in cotransformants in liquid 3 dropout (3-DO) (A) and 4 dropout (4-DO) (B) medium. Colonies with cotransformants of eNOS oxygenase domain (oxy), eNOS reductase domain (red), and the middle part of eNOS (mid) with β-actin or empty pADT-7 (AD), or β-actin with empty pGBKT-7 (BD) were grown in 3-DO or 4-DO liquid medium, and β-galactosidase was assayed in yeast lysates as described in materials and methods. C: ability of cotransformants to grow on selective media in the absence of several essential amino acids as described in materials and methods. Stringency of growth conditions increased from 2-DO to 4-DO. Growth rate ranged from abundant (+++) to minimal (−). N = 3.
DISCUSSION
In the present study, we have found that a dynamic alteration in eNOS activity is induced during PAEC growth. When PAEC grow from preconfluence to confluence, eNOS activity increases and continues to increase after cells reach confluence. The increase in eNOS activity is accompanied by increased NO production in confluent cells vs. preconfluent cells. The greater eNOS activity in confluent cells compared with preconfluent cells is related to endothelial cell density because higher eNOS activity was observed in endothelial cells at higher density when the cells were split at different densities and harvested at the same time. Confluent endothelial cells do not proliferate and demonstrate a contact inhibition of cell growth (12). Therefore, endothelial cells are an excellent model to study cellular physiological alterations during cell proliferation.
eNOS is regulated by a number of mechanisms at transcriptional, posttranscriptional, and posttranslational levels. Govers et al. (10) reported that eNOS activity is higher in the confluent state than in the preconfluent state in murine microvascular endothelial cells and in bovine aortic endothelial cells. Consistent with the data reported by Whitney et al. (24), our results indicate that the magnitude of the increase in eNOS activity in PAEC is higher than the magnitude of the increase in eNOS protein level. eNOS activity was eightfold greater in confluent PAEC compared with preconfluent cells, whereas eNOS protein contents were just twofold higher in confluent PAEC than in preconfluent cells, indicating that increased eNOS gene expression is partially responsible for greater eNOS activity in confluent endothelial cells. A posttranslational mechanism likely plays an important role in the increased eNOS activity in confluent PAEC.
At the posttranslational level, eNOS is thought to be regulated by phosphorylation, by fatty acylation with myristate and palmitate, by the interaction between the eNOS protein and other proteins, such as caveolin, calmodulin, bradykinin B2 receptor, HSP90, Nostrin, and cytoskeletal proteins. Analysis of the phosphorylation status of eNOS Ser1177 and Thr495, two of the major phosphorylation sites of eNOS, did not reveal any significant differences between preconfluent and confluent cells (data not shown). We also did not find any differences in the interaction of eNOS with caveolin-1 or HSP90 in endothelial cells at different densities. These results suggest that phosphorylation of Ser1177 and Thr495 or changes in the interaction of eNOS with caveolin-1 or HSP90 are unlikely to be responsible for the increase of eNOS activity in confluent endothelial cells.
In the last several years, a large body of evidence has accumulated showing that the actin cytoskeleton associates with eNOS and regulates eNOS activity (10, 13, 20, 28). Venema et al. (23) reported that there is a significant amount of eNOS in the Triton X-100-insoluble portion (the actin cytoskeleton) of aortic endothelial cells. We have reported that eNOS is colocalized with cortical F-actin at the plasma membrane and with G-actin in the perinuclear region (Golgi complex) (20), and incubation of purified eNOS with F-actin and G-actin results in significantly increased eNOS activity (20). In the present study, our results show that in preconfluent PAEC eNOS is colocalized with G-actin in the perinuclear region but not with F-actin, which presents in the form of stress fibers. In the confluent cells, F-actin is in the form of cortical F-actin instead of stress fibers, and eNOS is colocalized with cortical F-actin at the plasma membrane and with G-actin in the perinuclear area. Corresponding to observations in confocal fluorescence microscopy, we found increased eNOS protein contents in Triton X-100-insoluble fractions accompanied by increased β-actin content. In addition, there was more β-actin associated with eNOS in the Triton X-100-soluble fraction in confluent cells in later growth stages and in high-density states, suggesting that increased association of actin with eNOS is an important posttranslational mechanism responsible for greater eNOS activity in confluent PAEC. This conclusion is further supported by the observation that the decrease in eNOS association with β-actin secondary to silencing β-actin expression causes inhibition of eNOS activity, NO production, and endothelial monolayer wound repair in PAEC. Moreover, we have found that increases in the association of β-actin with eNOS by cytochalasin D and jasplakinolide are accompanied by corresponding increases in eNOS activity. Cytochalasin D is able to decrease and jasplakinolide to increase actin polymerization state. However, both agents induce increases in the association of β-actin with eNOS and in eNOS activity. We have previously reported that stabilization of F-actin by phalloidin decreases eNOS activity and association of eNOS with G-actin in PAEC (20). Although two actin polymerization agents, jasplakinolide and phalloidin, have different effects on eNOS activity and eNOS-actin association, eNOS activity is positively correlated to its association with G-actin. Together, these results suggest that it is the eNOS-actin association rather than the actin polymerization state that affects eNOS activity.
Although we cannot exclude the possibility that association of eNOS with the actin cytoskeleton is achieved indirectly through other eNOS-interacting proteins, our data indicate that eNOS has a direct interaction with actin. In vitro incubation of purified eNOS with F-actin and G-actin resulted in significant increases in eNOS activity (20). Moreover, in the present study, the yeast two-hybrid experiments suggest a strong association of the eNOS oxygenase domain with β-actin. The mechanism for enhancement of eNOS activity by actin-eNOS interaction is not known. An increase in eNOS dimerization does not appear to be involved based on preliminary unpublished data from our laboratory. Further experiments are needed to confirm the precise actin binding site within the eNOS oxygenase domain and to identify the mechanism for enhancement of eNOS activity by actin-eNOS interaction.
Although several laboratories have reported that eNOS activity is greater in confluent endothelial cells (10, 24), opposite findings have also been reported. Arnal et al. (1) reported that eNOS activity and protein levels were lower in confluent bovine aortic endothelial cells. Zollner et al. (29) showed that specific eNOS activity was upregulated in proliferating bovine atrial endothelial cells and was downregulated in quiescent bovine atrial endothelial cells. Recently, Bayraktutan (3) reported that 50% confluent coronary microvascular endothelial cells possessed approximately threefold increased activity of eNOS and NAD(P)H oxidase compared with 100% confluent cells. Govers et al. (10) have attributed these differences to the different methods with which these experiments were performed. After treatment with trypsin, Arnal's and Zollner's groups split the cells in an equal ratio for all experimental conditions and assayed the cells on consecutive days. We have tried both methods: seeding cells at different densities and harvesting them at the same time or seeding cells at the same low density but collecting cells at several different time points. The data from both methods indicate the same conclusion. In consideration of the fact that Whitney et al. (24) also observed greater eNOS activity in confluent PAEC, we speculate that growth- and density-dependent regulation of eNOS activity might be different for endothelial cells from different vascular beds. There are fundamental differences between pulmonary endothelial cells and aortic or coronary endothelium. Pulmonary endothelium is derived from the splanchnic mesenchyme (17). Aortic and coronary endothelium originate from different embryologic sources. Differences in the actin cytoskeleton may also contribute to the differences observed from cells derived from different vascular beds. For example, Searles et al. (16) recently reported differences in the actin cytoskeleton in bovine aortic endothelial cells compared with our porcine PAEC. In proliferating aortic endothelial cells, F-actin was concentrated near the cell periphery, whereas the actin stress fibers were prominent in proliferating PAEC. In confluent aortic endothelial cells, the distribution of F-actin was diffuse, whereas F-actin was mainly in cortical F-actin in post-confluent PAEC. The aortic endothelium withstands higher blood pressure, so it may need intense and diffuse actin filaments. The pulmonary endothelium is subjected to much lower pressure and thus may need a more prominent cortical actin to facilitate tight intercellular junctions to preserve gas exchange. Another difference between our results and those of Searles et al. in aortic endothelial cells is that Searles et al. reported that 500 nM jasplakinolide caused an increase in eNOS mRNA in aortic endothelial cells. We found that 500 nM jasplakinolide was cytotoxic to porcine PAEC. We found that jasplakinolide (10–100 nM) increased actin-eNOS association and eNOS activity without causing significant changes in eNOS protein content, suggesting that increased association of actin with eNOS protein results in greater eNOS activity. eNOS activity and NO production were not measured in the experiments of Searles et al. (16). The posttranslational modification of eNOS protein by the actin cytoskeleton in preconfluent proliferating vs. confluent quiescent aortic endothelial cells is not known.
In summary, eNOS activity in PAEC increases when cells grow from preconfluence to confluence. The greater eNOS activity in confluent PAEC is caused by the increase in cell density. The mechanism involves enhanced association of eNOS with actin.
GRANTS
This work was supported by the Medical Research Service of the Department of Veterans Affairs; National Heart, Lung, and Blood Institute Grants HL-52136 and HL-67951; Flight Attendant Medical Research Institute Grant 032040; and Florida Department of Health Grant 04TSP-01.
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
We thank Weihong Han and Humberto Herrera for assistance with tissue culture and eNOS activity assay and Dr. Michael Bubb for critical review of the manuscript.
REFERENCES
- 1 Arnal JF, Yamin J, Dockery S, and Harrison DG. Regulation of endothelial nitric oxide synthase mRNA, protein, and activity during cell growth. Am J Physiol Cell Physiol 267: C1381–C1388, 1994.
Link | ISI | Google Scholar - 2 Babaei S and Stewart DJ. Overexpression of endothelial NO synthase induces angiogenesis in a co-culture model. Cardiovasc Res 55: 190–200, 2002.
Crossref | PubMed | ISI | Google Scholar - 3 Bayraktutan U. Nitric oxide synthase and NAD(P)H oxidase modulate coronary endothelial cell growth. J Mol Cell Cardiol 36: 277–286, 2004.
Crossref | PubMed | ISI | Google Scholar - 4 Dos Remedios CG, Chhabra D, Kekic M, Dedova IV, Tsubakihara M, Berry DA, and Nosworthy NJ. Actin binding proteins: regulation of cytoskeletal microfilaments. Physiol Rev 83: 433–473, 2003.
Link | ISI | Google Scholar - 5 Dudek SM and Garcia JG. Cytoskeletal regulation of pulmonary vascular permeability. J Appl Physiol 91: 1487–1500, 2001.
Link | ISI | Google Scholar - 6 Feng Y, Venema VJ, Venema RC, Tsai N, and Caldwell RB. VEGF induces nuclear translocation of Flk-1/KDR, endothelial nitric oxide synthase, and caveolin-1 in vascular endothelial cells. Biochem Biophys Res Commun 256: 192–197, 1999.
Crossref | PubMed | ISI | Google Scholar - 7 Fleming I and Busse R. Signal transduction of eNOS activation. Cardiovasc Res 43: 532–541, 1999.
Crossref | PubMed | ISI | Google Scholar - 8 Goligorsky MS, Li H, Brodsky S, and Chen J. Relationships between caveolae and eNOS: everything in proximity and the proximity of everything. Am J Physiol Renal Physiol 283: F1–F10, 2002.
Link | ISI | Google Scholar - 9 Gonin S, Diaz-Latoud C, Richard MJ, Ursini MV, Imbo A, Manero F, and Arrigo AP. p53/T-antigen complex disruption in T-antigen transformed NIH3T3 fibroblasts exposed to oxidative stress: correlation with the appearance of a Fas/APO-1/CD95 dependent, caspase independent, necrotic pathway. Oncogene 18: 8011–8023, 1999.
Crossref | PubMed | ISI | Google Scholar - 10 Govers R, Bevers L, de Bree P, and Rabelink TJ. Endothelial nitric oxide synthase activity is linked to its presence at cell-cell contacts. Biochem J 361: 193–201, 2002.
PubMed | ISI | Google Scholar - 11 Huang PL. Endothelial nitric oxide synthase and endothelial dysfunction. Curr Hypertens Rep 5: 473–480, 2003.
Crossref | PubMed | ISI | Google Scholar - 12 Johnson AR. Human pulmonary endothelial cells in culture. Activities of cells from arteries and cells from veins. J Clin Invest 65: 841–850, 1980.
Crossref | PubMed | ISI | Google Scholar - 13 Laufs U, Endres M, Stagliano N, Amin-Hanjani S, Chui DS, Yang SX, Simoncini T, Yamada M, Rabkin E, Allen PG, Huang PL, Bohm M, Schoen FJ, Moskowitz MA, and Liao JK. Neuroprotection mediated by changes in the endothelial actin cytoskeleton. J Clin Invest 106: 15–24, 2000.
Crossref | PubMed | ISI | Google Scholar - 14 Lee TY, Rosenthal A, and Gotlieb AI. Transition of aortic endothelial cells from resting to migrating cells is associated with three sequential patterns of microfilament organization. J Vasc Res 33: 13–24, 1996.
Crossref | PubMed | ISI | Google Scholar - 15 Li B and Fields S. Identification of mutations in p53 that affect its binding to SV40 large T antigen by using the yeast two-hybrid system. FASEB J 7: 957–963, 1993.
Crossref | PubMed | ISI | Google Scholar - 16 Searles CD, Ide L, Davis ME, Cai H, and Weber M. Actin cytoskeleton organization and posttranscriptional regulation of endothelial nitric oxide synthase during cell growth. Circ Res 95: 488–495, 2004.
Crossref | PubMed | ISI | Google Scholar - 17 Stenmark KR and Mecham RP. Cellular and molecular mechanisms of pulmonary vascular remodeling. Annu Rev Physiol 59: 89–144, 1997.
Crossref | PubMed | ISI | Google Scholar - 18 Su Y and Block ER. Role of calpain in hypoxic inhibition of nitric oxide synthase activity in pulmonary endothelial cells. Am J Physiol Lung Cell Mol Physiol 278: L1204–L1212, 2000.
Link | ISI | Google Scholar - 19 Su Y, Cao W, Han Z, and Block ER. Cigarette smoke extract inhibits angiogenesis of pulmonary artery endothelial cells: the role of calpain. Am J Physiol Lung Cell Mol Physiol 287: L794–L800, 2004.
Link | ISI | Google Scholar - 20 Su Y, Edwards-Bennett S, Bubb MR, and Block ER. Regulation of endothelial nitric oxide synthase by the actin cytoskeleton. Am J Physiol Cell Physiol 284: C1542–C1549, 2003.
PubMed | ISI | Google Scholar - 21 Su Y, Han W, Giraldo C, De Li Y, and Block ER. Effect of cigarette smoke extract on nitric oxide synthase in pulmonary artery endothelial cells. Am J Respir Cell Mol Biol 19: 819–825, 1998.
Crossref | PubMed | ISI | Google Scholar - 22 Takemoto M, Sun J, Hiroki J, Shimokawa H, and Liao JK. Rho-kinase mediates hypoxia-induced downregulation of endothelial nitric oxide synthase. Circulation 106: 57–62, 2002.
Crossref | PubMed | ISI | Google Scholar - 23 Venema VJ, Marrero MB, and Venema RC. Bradykinin-stimulated protein tyrosine phosphorylation promotes endothelial nitric oxide synthase translocation to the cytoskeleton. Biochem Biophys Res Commun 226: 703–710, 1996.
Crossref | PubMed | ISI | Google Scholar - 24 Whitney JA, German Z, Sherman TS, Yuhanna IS, and Shaul PW. Cell growth modulates nitric oxide synthase expression in fetal pulmonary artery endothelial cells. Am J Physiol Lung Cell Mol Physiol 278: L131–L138, 2000.
Link | ISI | Google Scholar - 25 Wong MK and Gotlieb AI. Endothelial cell monolayer integrity. I. Characterization of dense peripheral band of microfilaments. Arteriosclerosis 6: 212–219, 1986.
Crossref | PubMed | Google Scholar - 26 Zhang J, Patel JM, Li YD, and Block ER. Proinflammatory cytokines downregulate gene expression and activity of constitutive nitric oxide synthase in porcine pulmonary artery endothelial cells. Res Commun Mol Pathol Pharmacol 96: 71–87, 1997.
PubMed | ISI | Google Scholar - 27 Zharikov SI, Krotova KY, Belayev L, and Block ER. Pertussis toxin activates l-arginine uptake in pulmonary endothelial cells through downregulation of PKC-α activity. Am J Physiol Lung Cell Mol Physiol 286: L974–L983, 2004.
Link | ISI | Google Scholar - 28 Zharikov SI, Sigova AA, Chen S, Bubb MR, and Block ER. Cytoskeletal regulation of the l-arginine/NO pathway in pulmonary artery endothelial cells. Am J Physiol Lung Cell Mol Physiol 280: L465–L473, 2001.
PubMed | ISI | Google Scholar - 29 Zollner S, Aberle S, Harvey SE, Polokoff MA, and Rubanyi GM. Changes of endothelial nitric oxide synthase level and activity during endothelial cell proliferation. Endothelium 7: 169–184, 2000.
Crossref | PubMed | Google Scholar
