Supplemental Materials and Methods
Cell Isolation and Culture
Intact porcine hearts were obtained at a local slaughterhouse (Holifield farms, Porterdale, GA) and immediately placed in cold sterile phosphate buffered saline (PBS, Invitrogen). The aortic valve was excised within 12 hours post-mortem, and the leaflets were trimmed to 1/3 of their length away from their apposition to the base of the aortic root. Leaflets were then rinsed several times in a series of solutions containing gradually reduced amounts of antibiotic-antimycotic (A-A, Invitrogen). The valvular endothelium from both sides of the leaflet was then isolated by incubating the leaflets in a collagenase solution (Worthington, 600 U/mL) for approximately 10 minutes. The surfaces of the leaflets were then gently scraped, and the resulting cell solution collected and spun in a bucket centrifuge at 200 g’s for 5 minutes. Cells from all three leaflets of the valve were collected together. The pelleted cells were then plated onto tissue culture flasks pre-coated with collagen type I (Becton-Dickenson, Rat tail, 50 mg/mL). Endothelial cells were similarly isolated from the porcine ascending aorta, and plated in tissue culture flasks. Cells were cultured with Dulbecco’s Modified Eagle’s Medium (DMEM, Invitrogen) supplemented with 10% fetal bovine serum (FBS, Hyclone), 1% penicillin-streptomycin (Hyclone), and 1% L-Glutamine (Mediatech) at 37°C and 5% CO2. Primary culture media was also briefly supplemented with 50 units/mL Heparin (Sigma) and 0.1% Mycoplasma Removal Agent (MRA, ICN). Medium was changed every 48 hours, and cells were split 1:3 upon confluence.
Cell phenotype comparison. Endothelial phenotype was assessed through indirect immunofluorescent staining and acetylated LDL uptake (a-LDL). Endothelial cell phenotype is generally determined by aLDL uptake, positive von Willebrand factor (vWF) expression, and negative a-smooth muscle actin (a-SMA) expression. Porcine aortic smooth muscle cells (PASMC) served as a control. For a-LDL comparisons, PAECS, PAVECS, and PASMC were incubated with 10 mg/mL a-LDL (Biomedical Technologies #BT-902) in serum-free media for 4 hours. For immunofluorescent staining, PAVECS, PAECS, and PASMC were seeded onto coverslips coated with collagen I and grown to confluence, after which they were rinsed in PBS and fixed in 3.7% paraformaldehyde (Tousimis). Cells were then permeablized with 0.1% Triton X-100 (Sigma), and blocked with 1% neonatal goat serum (Sigma) for one hour. The coverslips were then incubated with either anti-a-smooth muscle actin (Sigma #F3777) or anti-von Willebrand factor (vWF, Sigma #F3520) for one hour. These cells were then incubated with a fluorescently labeled secondary antibody for one hour. The coverslips were then flipped onto glass slides and sealed. The samples were imaged using laser confocal microscopy for qualitative expression of the aforementioned markers.
Shear Flow System. The flow system consists of a polycarbonate block that has been machined to include inlet, outlet, and flow ports. A glass slide containing cultured cells is placed in an aluminum chamber, containing a rubber gasket, and secured with machine screws. This creates a rectangular chamber for fluid flow, the height of which can be changed with the use of polyester spacers. Therefore, a channel with defined geometry is created, and using Navier-Stokes equations of fluid flow with constant density, the wall shear stress (t) can be related to flow rate and channel geometry by the formula:
t = 6Q/mbh2
with Q is the flow rate, m is the fluid viscosity (0.012 poise), and b and h are channel width and height respectively. The flow chamber was connected to a peristaltic pump that controls the flow rate. A medium reservoir and a pulse dampener are also included into the flow circuit, as well as an air filter.
Glass microscope slides were pre-coated with type I collagen for two hours prior to cell seeding. Slides were then seeded with either PAECs or PAVECs, and cultured for 48 hours until confluent. The slides were then placed into the flow chambers, and connected to the flow circuits. The reservoirs were then filled with 125 mL of culture medium. Completed circuits were then placed in an incubator and attached to the pump.
Morphological analysis. Five representative images were taken of confluent monolayers at predetermined locations on the central region of the slides. The slides were aligned such that the image horizontal corresponded to the direction of flow. Image analysis software (LSM Image Browser, Zeiss Corp.) was used to overlay polygons delineating cell areas according to f-actin filaments. The angle between the horizontal (flow direction) and the majority of f-actin filaments was also determined. The software calculated the cell area, perimeter, and orientation angle. Two indices of cell alignment were used: cell shape index (SI) and orientation angle (OA). SI is a non-dimensional parameter that quantifies cell elongation on a scale of 0 to 1, with 0 denoting a straight line, and 1 a perfect circle. It is represented by the following equation:
SI = 4pA/P2
Orientation angle was the deviation of the major actin filament axis from the slide horizontal, which was parallel to the flow direction.
Western Blotting. Additional flow and static culture experiments were conducted to determine the levels of total focal adhesion protein between cell type and condition. After the completion of the experiments, cells were lysed in RIPA buffer containing protease inhibitors (aprotinin, leupeptin, PMSF, 25mg/ml each) and a phosphate analog (sodium orthovanadate, 25 mg/ml) to inhibit phosphatases. Lysate was quantified for total protein using a micro-BCA assay (Pierce), and equal quantities of protein were loaded into 7% polyacrylamide gels. Gels were run at 90V for 90 minutes, transferred to nitrocellulose paper, and blocked overnight in 5% nonfat milk. The samples were then incubated blocking buffer with either anti-b1 integrin (1:500), anti-vinculin (1:1000), or anti-FAK (1:500) antibody for one hour, followed by rinsing in TBS-Tween buffer. Samples were then incubated in biotinylated anti-rabbit or anti-mouse antibodies for one hour, rinsed in TBS-Tween, incubated in alkaline phosphotase anti-biotin for one hour, and rinsed with TBS-Tween. Alkaline phosphatase was then visualized by ECF for 5 minutes, after which the samples were dried and imaged.
Supplemental Results
Phenotype marker expression. Figure I details the phenotype comparison of the PAECs and PAVECs. PAECs and PAVECs express each marker similarly, but opposite of the negative control PASMC. Both of these endothelial cell types were positive for aLDL uptake and vWF, but negative for a-SMA. PASMCs were negative for aLDL and vWF, but positive for a-SMA. These results show that the valvular endothelial cells express similar markers and aLDL uptake to vascular endothelial cells, this in contrast to the smooth muscle cells.
Additional cell morphology data. Figures II and III are additional data supporting statements mentioned in the printed text. Refer to the figure legends for greater detail. In figure III, the shape index of PAVECS and PAECS are significantly different in static culture. This is because of the stellate pattern these cells present owing to their filamentous extensions. The shape index is a robust measurement of elongation in polarized cells, but less accurate for more stellate cells. This pattern changes somewhat over time in culture, presumably because the cells are spreading further without any influence of flow. The shape index of PAECS does not change because of their lack of extensions and a generally polygonal shape.
Confirmation of alignment tendency. As an additional demonstration of the perpendicular alignment tendency of the valvular endothelial cells, a glass coverslip seeded to confluence with PAVECs was fixed to a microscope slide. This combination was then exposed to 20 dynes/cm2 for 24 hours, then rotated 90 degrees, and again subjected to flow for 24 hours. The PAVECs aligned perpendicular to flow after the first 24 hours, and then re-aligned perpendicular to flow after the second 24 hours (Figure V, panels A and B), demonstrating their dynamic responses to fluid flow and a preference for a perpendicular alignment. The alignment after the second 24 hours is not as dramatic as after the first, probably due to the fact that the cells had to progress from an oppositely aligned state, instead of a more random orientation distribution.
Cell alignment dependent on actin polymerization. Experiments were conducted to determine the mechanism for valvular endothelial alignment to flow. Valvular endothelial cells were incubated with Cytochalasin-D (CD, Sigma #30385, 0.1 mm) for one hour prior to exposure to flow, and identical concentrations of either chemical were included in the flow medium as well. Alignment was then determined as previously described after 48 hours, with static cultures serving as controls. The aforementioned concentrations were determined through 48 hour static culture titration assays. The alignment change of the valvular endothelial cells was found to be correlated with the reorganization of actin filaments. Disruption of actin filaments with CD inhibited the alignment changes observed previously (Figure VI, panels A and B). Fragmented actin filaments prohibited an accurate determination of alignment parameters, but no tendency of cell alignment was observed.
No differences in total levels of focal adhesion proteins with cell type of flow condition. Figure VII shows the results of the Western blots. It is evident that there are no significant differences in any of the focal adhesion complexes studied. This confirms that the arrangement of focal adhesion proteins, and not the total amount of adhesion proteins, is important for the morphological differences see between these two cell types.
Supplemental Figure Legends
Figure I: Phenotype comparison of PAECs (A,D,G), PAVECs (B,E,H), and PASMCs (C,F,I). Cells analyzed for acetylated LDL uptake (A-C), alpha-smooth muscle actin expression (D-F), and von Willlebrand Factor expression (G-I). Scale bar = 50 mm.
Figure II: Cell alignment regressions. Static culture control vs.20 dynes/cm2 for either 24 hour or 48 hour flow condition. Flow regressions significantly different from control conditions P<0.05. PAECS 48 hour flow regression significantly different from 24 hour flow condition P<0.05. PAVECS 48 hour flow condition not significantly different from 24 hour flow condition.
Figure III: Cell Orientation changes with experimental condition (Panel A). Cell shape index changes with experimental condition (Panel B). Bars indicate standard deviation. (*) Indicates significance P<0.05, (NS) indicates no significance. Visually obvious significant differences not denoted for clarity.
Figure IV: Confocal images of cell monolayers stained for a-SMA (Green), f-actin (red) and cell nuclei (blue). Top row is PAECS, while the bottom row is PAVECS. Scale bar = 50 mm. Data indicates that there is no a-SMA expression by either cell type under flow, suggesting no transdifferentiation of these cells to a more smooth muscle-like phenotype.
Figure V: Re-orientation of valvular endothelial cells under flow. PAVECS were seeded on a coverslips coated with collagen I were adhered to a glass slide using vacuum grease. The ensemble was then placed in the flow system for 24 hours (A). The coverslip was then rotated 90 degrees, and subjected to flow for an additional 24 hours (B). PAVECs re-oriented perpendicular to flow, demonstrating their preference for this arrangement. Scale bar = 50 mm.
Figure VI: PAVECs with disrupted actin polymerization through Cytochalasin D failed to align under flow. Confocal images in 48 hour static (A) and 20 dynes/cm2 shear stress for 48 hours (B). Cells stained for f-actin (red) and cell nuclei (blue). Scale bar = 50 mm.
Figure VII: Western blots of b1 integrin, vinculin, and focal adhesion kinase. + or – indicates whether the samples were treated with 20 dynes/cm2 steady laminar shear stress for 48 hours (+) or culture statically (-) for 48 hours.