Journal of Colloid and Interface Science 230, 84–90 (2000) doi:10.1006/jcis.2000.7080, available online at http://www.idealibrary.com on
The Effect of Fluid Shear Stress on Endothelial Cell Adhesiveness to Polymer Surfaces with Wettability Gradient Jin Ho Lee,∗,1 Sang Jin Lee,∗ Gilson Khang,† and Hai Bang Lee‡ ∗ Department of Polymer Science and Engineering, Hannam University, 133 Ojeong Dong, Daedeog Ku, Taejon 306-791, Korea; †Department of Polymer Science and Technology, Chonbuk National University, 664-14 Dukjin Dong, Dukjin Ku, Chonju 561-756, Korea; and ‡Biomaterials Laboratory, Korea Research Institute of Chemical Technology, P.O. Box 107, Yusong, Taejon 305-606, Korea Received January 12, 2000; accepted July 5, 2000
nonthrombogenic surface. ECs in their resting state maintain a nonthrombogenic surface by various mechanisms, including release of fibrinolytic factors (tissue-type plasminogen activator) and anti-platelet agents (prostacyclin) and by activation of protein C through thrombomodulin (1). In the cardiovascular system, the cells are exposed constantly to hemodynamic forces due to the flow of blood. Endothelialization of the luminal surface reduces thrombogenicity of vascular prostheses, but most of the materials used for vascular prostheses are poor substrates for EC attachment under physiological shear stress, particularly those of small diameter (2, 3). It is recognized that the adhesion and proliferation of different types of cells on polymeric materials depend largely on surface characteristics such as wettability (hydrophilicity/hydrophobicity or surface free energy), chemistry, charge, roughness, and rigidity. A large number of research groups have extensively studied the effect of the surface wettability on the interactions of cultured cells with solid substrates since wettability is one of the most important parameters when biomaterials or implant devices are designed (4). Some groups have studied the interactions of different types of cultured cells with various solid substrates (mainly polymers) with different wettabilities to correlate the relationship between surface wettability and cell or blood compatibility (5–8). Some others have studied the interactions of different types of cells with a range of copolymers with different wettabilities, such as hydroxyethyl methacrylate (HEMA, hydrophilic)/methyl methacrylate (MMA, hydrophobic), HEMA/ethyl methacrylate (EMA, hydrophobic), and HEMA/styrene (hydrophobic) copolymers with different compositions (9–11). Many studies have recently been focused on the preparation of surfaces with properties that are changed gradually along the material length. Such “gradient surfaces” are of particular interest for basic studies of the interactions between biological species and surfaces since the effect of a selected property like wettability can be examined in a single experiment on one surface. Many research groups have prepared wettability gradient surfaces by diffusion of dimethyl dichloro silane through xylene on a flat hydrophilic inorganic substrate and used them to investigate surface hydrophilicity-induced changes of adhered
In this study, the adhesive strength of endothelial cells (ECs) attached on polymer surfaces with different hydrophilicity was investigated using wettability gradient polyethylene (PE) surfaces prepared by corona discharge treatment from a knife-type electrode whose power increases gradually along the sample length. The ECattached wettability gradient surfaces were mounted on parallelplate flow chambers in a flow system prepared for cell adhesiveness test. Three different shear stresses (150, 200, and 250 dyne/cm2 ) were applied to the flow chambers and each shear stress was maintained for 120 min to investigate the effect of shear stress and surface hydrophilicity on the EC adhesion strength. It was observed that the ECs were adhered more onto the positions with moderate hydrophilicity of the wettability gradient surface than onto the more hydrophobic or hydrophilic positions. The maximum adhesion of the cells appeared at around water contact angles of 55◦ . The EC adhesion strength was higher on the hydrophilic positions than on the hydrophobic ones. However, the maximum adhesion strength of the cells also appeared at around water contact angles of 55◦ . More than 90% of the adhered cells remained on that position after applying the shear stress, 250 dyne/cm2 for 2 h, whereas the cells were completely detached on the hydrophobic position (water contact angle, about 86◦ ) within 10 min after applying the same shear stress. It seems that surface hydrophilicity plays a very important role for cell adhesion strength. °C 2000 Academic Press Key Words: wettability gradient; corona discharge; endothelial cells; fluid shear stress; cell adhesion strength.
Surface-induced thrombosis remains as one of main problems in the development of blood-contacting devices. Generally vascular grafts with a relatively large inner diameter (>5 mm) have been successfully employed for replacement in the human body. However, the use of small diameter grafts is limited, because these grafts rapidly occlude due to the thrombosis. The ideal blood-contacting surface of a prosthesis would be an endothelial cell (EC) lining, because the confluent monolayer of healthy ECs that line natural blood vessels represents the ideal 1
To whom correspondence should be addressed. Fax: 82-42-626-8841.
C 2000 by Academic Press Copyright ° All rights of reproduction in any form reserved.
CELL ADHESIVENESS ON WETTABILITY GRADIENT SURFACES
cells (12–14). Recently, we developed a method for preparing a wettability gradient on polymer surfaces using corona discharge treatment (15). The wettability gradient was produced by treating the polymer sheets with corona from a knife-type electrode whose power was changed gradually along the sample length. The polymer surfaces oxidized gradually with the increasing corona power and the wettability gradient was created on the sample surfaces. The wettability gradient surfaces prepared by the corona discharge treatment were used to investigate the interaction of different types of cells in terms of the surface hydrophilicity/hydrophobicity of polymeric materials (16–18). Most above studies on cell adhesion and proliferation have been carried out under static conditions. In many biomedical applications, particularly in cardiovascular applications, however, the strength of cells to a biomaterials is more determinant for the ultimate patency than adhesion or proliferation. Only a few research works have been carried out to investigate the effect of the strength of adhered cells to substrate wettability under flow conditions (19–22). In this study, the adhesive strength of ECs attached on polymer surfaces with different hydrophilicities was investigated using wettability gradient polyethylene (PE) surfaces prepared by corona discharge treatment. The ECattached wettability gradient surfaces were mounted on parallelplate flow chambers in a flow system and different shear stresses were applied to the flow chambers to correlate the relationship of surface wettability and shear stress with the adhesion strength of ECs. The stable EC adhesion on surfaces under flow conditions is very important for vascular graft application. MATERIALS AND METHODS
Substrate An additive-free low-density PE film (about 50 µm thickness, Hanyang Chemical Co., Korea) was used as the polymeric substrate for the preparation of wettability gradient surfaces. The PE film was cut into 5 × 7-cm pieces, ultrasonically cleaned twice in ethanol for 30 min each, and then dried at room temperature in a clean bench. The pieces were stored in a vacuum oven until use. The cleanliness of the surface was verified by electron spectroscopy for chemical analysis (ESCA). Preparation and Characterization of Wettability Gradient Surfaces The PE film was treated with a radio-frequency (RF) corona discharge apparatus for the preparation of gradient surfaces, in a manner similar to that used in our previous study (15). Briefly, a knife-type electrode was connected to the RF generator and the power increased gradually by a motorized drive (Fig. 1). The cleaned PE film was placed on the sample bed and dry air was purged through the apparatus at a flow rate of 20 L/min. The electrode was 1.5 mm away from the sample surface. At the same time as the sample bed was translated at a constant speed, 1.0 cm/s, the corona from the electrode was discharged onto the sample with gradually increasing power (from 10 to 50 W at
FIG. 1. Schematic diagram showing the corona discharge apparatus for the preparation of wettability gradient surfaces.
100 kHz). The sample film (5.0 × 5.0 cm) was treated for 5 s. By this treatment, the sample surface was continuously exposed to the corona with increasing power, resulting in the formation of wettability gradient on the surface. The corona-treated PE surfaces were characterized by the measurement of water contact angle. The water contact angle, an indicator of the wettability of surfaces, was measured by a sessile drop method at room temperature using an optical benchtype contact angle goniometer (Model 100-0, Rame-Hart, Inc.). Drops of purified water, 3 µL, were deposited onto the coronatreated PE surface along the sample length using a microsyringe attached on the goniometer. The amount of peroxides produced on PE surface along the sample length by the corona discharge treatment was determined using radical scavenger, 1,1-diphenyl2-picryl-hydrazyl (DPPH) (23). The corona-treated PE film was cut perpendicular to the gradient into five sections (1.0 cm each). Each section was immersed in 10 mL DPPH ethanol solution (1.0 × 104 mol L−1 ) for 6 h at 70◦ C. The DPPH molecules consumed were quantified from the difference in transmittance at 520 nm between the untreated control and corona-treated section. The adsorption coefficient of DPPH at 520 nm was 9.33 × 10−5 L mol−1 cm−1 . More details in the corona discharge apparatus and the preparation of wettability gradient surfaces were described in a previous paper (15). Cell Adhesion CPAE bovine pulmonary artery ECs (KCLB 10209, Korean Cell Line Bank, Korea) were used to study the effect of shear stress and surface wettability on the EC adhesion strength. The cells routinely cultured in tissue culture polystyrene (PS) flasks (Corning) at 37◦ C under 5% CO2 atmosphere were harvested after the treatment with 0.25% trypsin (Gibco Laboratories). The wettability gradient PE surfaces were mounted in test chambers (dimensions, 7.0 × 5.0 cm) similar to those described by van Wachem et al. (6). The surfaces mounted in the test chambers were equilibrated with prewarmed (37◦ C) Dulbecco’s phosphate-buffered saline (PBS, pH 7.3–7.4; Sigma) free of Ca2+ and Mg2+ for 30 min. After removing the PBS solution from the chambers by pippetting, the cells (4 × 104 /cm2 ) were seeded to the surfaces. The culture medium used was RPMI 1640. The medium was contained 10% fetal bovine serum (FBS), 100 units/mL penicillin, and 100 µg/mL
LEE ET AL.
gentamycin. The medium and FBS were purchased from Gibco Laboratories. The cells were incubated on the wettability gradient surfaces for the adhesion. After 1 day incubation at 37◦ C under 5% CO2 atmosphere, some of the film surfaces were used for the cell adhesiveness test on parallel-plate flow chambers in a flow system (described below). Some others were examined by a scanning electron microscope (SEM; JSM-840A, Jeol Co., Japan). For this, the film surfaces were washed with PBS and the cells attached on the surfaces were fixed with 2.5% glutaraldehyde (Gibco Laboratories) in PBS for 30 min at room temperature. After thorough washing with PBS, the cells on the surfaces were dehydrated in ethanol graded series (50, 60, 70, 80, 90, and 100%) for 10 min each and allowed to dry on a clean bench at room temperature. The cell-attached surfaces were cut into five sections along the gradient and the specimens were gold deposited in vacuum and examined by SEM with a tilt angle of 45◦ . The cell density on the surfaces was estimated by counting the number of attached cells along the gradient. The results for cell adhesion after 1 day were expressed as the average number of cells attached per cm2 of surface. Flow Experiments EC-attached wettability gradient surfaces were placed on parallel-plate flow chambers in a flow system for the cell adhesiveness test (Fig. 2). The flow chamber consisted of a glass-
bottom plate, a polymethyl methacrylate (PMMA) top plate with inlet and outlet silicone rubber tubings (inner diameter, 1.5 mm), and a silicone rubber gasket (thickness, 1 mm) inserted between the two plates. The effective chamber dimensions are 6.0 × 1.8 cm. EC-attached film was placed on the glass-bottom plate. The flow chamber was held together by several clips. The flow chamber was mounted on an inverted microscope stage. Cell culture medium in 37◦ C water bath was pumped into movable cell culture medium reservoir located above the flow chamber to provide a hydrostatic pressure difference, which adjusts the required flow rate. The shear stress τw (dyne/cm2 ) was calculated using the following equation (24): τw = 6Qµ/wh 2 , where Q is the volumetric flow rate (cm3 /s), µ is the viscosity of the medium (0.1 dyne s/cm2 ), w is the chamber width (1.8 cm), and h is the chamber height (0.1 cm). Three different shear stresses (150, 200, and 250 dyne/cm2 ) were applied to the flow chambers by adjusting the level of the movable medium reservoir, which are more severe conditions than those of blood vessels of the human body; physiological levels of venous and arterial shear stresses are 1–5 and 6– 40 dynes/cm2 , respectively (25). Each shear stress was maintained for 120 min to investigate the effect of shear stress on the EC adhesion strength. Under flow conditions the cells attached on the wettability gradient surface along the sample length were observed through a CCD camera attached in the inverted microscope and recorded in a video tape for the image analysis (26). RESULTS AND DISCUSSION
Characterization of Wettability Gradient Surfaces
FIG. 2. Schematic diagram showing the parallel-plate flow chambers in a flow system for the cell adhesiveness test.
The water contact angles of the corona-treated PE surface gradually decreased (from 96 to 43 degree) along the sample length with increasing corona power (Fig. 3). The decrease in the contact angles (and thus the increase in wettability) along the sample length may be due to the oxygen-based polar functionalities incorporated on the surface by the corona discharge treatment. Figure 4 shows the possible mechanism of surface oxidation which occurs on a polymer surface such as PE by corona discharge treatment. The corona discharge treatment of a polymer surface produces carbon radicals from the hydrocarbon backbone, followed by the formation of unstable hydroperoxides through rapid binding with oxygen in air, and then the decomposition of the hydroperoxides to produce various oxygen-based functionalities (hydroxyl group, ether, ketone, aldehyde, carboxylic acid, carboxylic ester, etc.) by reaction with additional oxygen (27–32). Figure 5 shows the concentration profile of hydroperoxides produced on PE surface along the sample length by the corona discharge treatment. The amount of the peroxides produced on PE surface gradually increased along the sample length with increasing corona power. The oxygen-based functional groups produced on the PE surface also increased
CELL ADHESIVENESS ON WETTABILITY GRADIENT SURFACES
FIG. 3. Changes in water contact angle of corona-treated PE surface along the sample length. Sample numbers, n = 3.
gradually with the increase in corona power, as analyzed by FTIR-ATR and ESCA in a previous study (17). Cell Adhesion on Wettability Gradient Surfaces ECs were cultured on the wettability gradient PE surfaces for 1 day. Figure 6 shows the cell adhesion behavior on the wettability gradient surface. As the surface wettability increased along the sample length, the cell adhesion on the surface increased and then decreased. The cells were adhered more on the positions with moderate hydrophilicity of the wettability gradient surface than more hydrophobic or hydrophilic positions, even though the difference in the number of cells attached was not significant on the hydrophilic positions. The maximum adhesion of the cells appeared at around position 2.5 cm (water contact angle, ∼55 degree; see Fig. 3). The fact that cells are more adherent on the moderately hydrophilic surfaces was also observed by other research groups (6–9) as they cultured EC, HeLa S3 , or
FIG. 4. Possible mechanism of the formation of oxygen-based functionalities on a polymer surface by corona discharge treatment.
FIG. 5. Changes in peroxide concentration on corona-treated PE surface along the sample length (n = 3).
fibroblasts onto various polymer substrates with different surface wettability. This seems closely related to the serum protein adsorption on the surfaces. In our previous study (17), we observed that the serum proteins were also adsorbed more onto the positions with moderate hydrophilicity of the wettability gradient surface. The maximum adsorption of the proteins appeared at around position 2.5 cm, which is the same trend as the cell adhesion behavior (see Fig. 6). Among the serum proteins, some proteins like fibronectin and vitronectin are known as cell adhesive (33–36). The preferential adsorption of theses cell-adhesive proteins from culture medium onto the moderately hydrophilic
FIG. 6. Cell adhesion on wettability gradient PE surface after 1 day culture (number of seeded cells, 4 × 104 /cm2 ) (n = 5).
LEE ET AL.
FIG. 7. The percentage of cells present on the positions of wettability gradient PE surfaces as a function of flowing time for different shear stresses: (A) 150 dyne/cm2 , (B) 200 dyne/cm2 , and (C) 250 dyne/cm2 .
surfaces may be a reason for better cell adhesion on the positions with moderate hydrophilicity of the wettability gradient surface. Detachment of Cells from Wettability Gradient Surfaces After 1 day culture, EC-attached wettability gradient PE surfaces were mounted on parallel-plate flow chambers in a flow system. Three different shear stresses (150, 200, and 250 dyne/ cm2 ) were applied to the flow chambers and each shear stress was maintained for 120 min. The shear stresses applied in this study are much more severe conditions than those of blood vessels of the human body; physiological levels of venous and arterial shear stresses are 1–5 and 6–40 dynes/cm2 , respectively, as discussed earlier. We wanted to investigate the effect of shear stress
and surface hydrophilicity on the EC adhesion strength within short time. Figure 7 shows the results of the cell detachment study on the wettability gradient surfaces under different shear stresses. The cells were detached more and sooner on the hydrophobic positions of the wettability gradient surface than on the hydrophilic ones, indicating that the cell adhesion strength is higher on the hydrophilic positions. The maximum adhesion strength of the cells appeared at around water contact angle of 55◦ (position, 2.5 cm), which is the same trend as the cell adhesion behavior (see Fig. 6). More than 90% of the adhered cells remained on that position after applying the shear stress, 250 dyne/cm2 for 2 hr, whereas the cells were completely detached on the hydrophobic position (position, 0.5 cm; water contact angle, about 86◦ ) within
CELL ADHESIVENESS ON WETTABILITY GRADIENT SURFACES
FIG. 8. CCD images of cells adhered on hydrophobic (0.5 cm) and hydrophilic (2.5 cm) positions of wettability gradient PE surfaces as a function of flowing time (shear stress, 250 dyne/cm2 ).
10 min after applying the same shear stress. This phenomenon also seems closely related to the serum protein adsorption on the surfaces, as discussed earlier; the preferentially adsorbed cell-adhesive proteins such as fibronectin and vitronectin from culture medium onto the moderately hydrophilic surface maybe strongly interact with the cells on the surface, providing the increased adhesive strength of the cells during exposure to flow. Other research groups have also observed that cell adhesion strength after exposure to flow increased by preadsorbing fibronectin onto surfaces prior to cell attachment (21, 37– 40). Figure 8 shows a series of sequential images of ECs adhered on hydrophobic (0.5 cm) and hydrophilic (2.5 cm) positions of the wettability gradient surface at shear stress, 250 dyne/cm2 . We can see that almost all cells were detached from the hydrophobic position during the first 1 min after applying the shear stress. It seems that surface hydrophilicity plays a very important role for cell adhesion strength. We demonstrated in this study that the wettability gradient prepared on polymer surfaces by a corona discharge treatment method can be a simple and effective tool for systematically investigating the role of surface hydrophilicity on cell adhesiveness under flow conditions. ACKNOWLEDGMENTS This work was supported by grants from the Korea Science and Engineering Foundation (Grant 95-0502-06-03-3) and the Korea Ministry of Science and Technology (Grant 197-NI-02-05-A-02).
REFERENCES 1. Kottke-Marchant, K., Veenstra, A. A., and Marchant, R. E., J. Biomed. Mater. Res. 30, 209 (1996). 2. Zilla, P., Curr. Opinion Cardiol. 6, 877 (1991). 3. Isher, H. P., and Tattersall, C. W., J. Biomed. Mater. Res. 26, 1383 (1992). 4. Lee, H. B., and Lee, J. H., in “Encyclopedic Handbook of Biomaterials and Bioengineering” (D. L. Wise, D. J. Trantolo, D. E. Altobelli, M. J. Yaszemski, J. D. Gresser, and E. R. Schwartz, Eds.), Vol. 1, p. 371. Dekker, New York, 1995. 5. Schakenraad, J. M., Busscher, H. J., Wildevuur, C. H. R., and Arends, J., J. Biomed. Mater. Res. 20, 773 (1986). 6. Van Wachem, P. B., Beugeling, T., Feijen, J., Bantjes, A., Detmers, J. P., and Van Aken, W. G., Biomaterials 6, 403 (1985). 7. Tamada, Y., and Ikada, Y., J. Colloid Interface Sci. 155, 334 (1993). 8. Tamada, Y., and Ikada, Y., J. Biomed. Mater. Res. 28, 783 (1994). 9. Van Wachem, P. B., Hogt, A. H., Beugeling, T., Feijen, J., Bantjes, A., Detmers, J. P., and Van Aken, W. G., Biomaterials 8, 323 (1987). 10. Horbett, T. A., Schway, M. B., and Ratner, B. D., J. Colloid Interface Sci. 104, 28 (1985). 11. Lydon, M. J., Minett, T. W., and Tighe, B. J., Biomaterials 6, 396 (1985). 12. Elwing, H., Welin, S., Askendahl, A., Nilsson, U., and Lundstrom, I., J. Colloid Interface Sci. 119, 203 (1987). 13. Williams, R. L., Hunt, J. A., and Tengvall, P., J. Biomed. Mater. Res. 29, 1545 (1995). 14. Ruardy, T. G., Schakenraad, J. M., Van der Mei, H. C., and Busscher, H. J., J. Biomed. Mater. Res. 29, 1415 (1995). 15. Lee, J. H., Kim, H. G., Khang, G. S., Lee, H. B., and Jhon, M. S., J. Colloid Interface Sci. 151, 563 (1992). 16. Lee, J. H., and Lee, H. B., J. Biomater. Sci. Polym. Edn. 4, 467 (1993). 17. Lee, J. H., Khang, G., Lee, J. W., and Lee, H. B., J. Colloid Interface Sci. 205, 323 (1998).
LEE ET AL.
18. Lee, J. H., Lee, S. J., Khang, G., and Lee, H. B., J. Biomater. Sci. Polym. Edn. 10, 283 (1999). 19. Van Kooten, T. G., Schakenraad, J. M., Van der Mei, H. C., and Busscher, H. J., Biomaterials 13, 897 (1992). 20. Horbett, T. A., Waldburger, J. J., Ratner, B. D., and Hoffman, A. S., J. Biomed. Mater. Res. 22, 383 (1988). 21. Burmeister, J. S., Vrany, J. D., Reichert, W. M., and Truskey, G. A., J. Biomed. Mater. Res. 30, 13 (1996). 22. Ruardy, T. G., Moorlag, H. E., Schakenraad, J. M., Van der Mei, H. C., and Busscher, H. J., J. Colloid Interface Sci. 188, 209 (1997). 23. Iwasaki, Y., Sawada, S., Nakabayashi, N., Khang, G., Lee, H. B., and Ishihara, K, Biomaterials 20, 2185 (1999). 24. Bird, R. B., Stewart, W. E., and Lightfoot, E. N., “Transport Phenomena,” p. 47. Wiley, New York, 1960. 25. Zarines, C. K., Giddens, D. P., Bharadvaj, B. K., Sottiurai, V. S., Mabon, R. F., and Glagov, S., Circ. Res. 53, 502 (1983). 26. Van Kooten, T. G., Schakenraad, J. M., Van der Mei, H. C., and Busscher, H. J., J. Biomed. Mater. Res. 26, 725 (1992). 27. Iwata, H., Kishida, A., Suzuki, M., Hata, Y., and Ikada, Y., J. Polym. Sci. Polym. Chem. 26, 3309 (1988). 28. Suzuki, M., Kishida, A., Iwata, H., and Ikada, Y., Macromolecules 19, 1804 (1986).
29. Owens, D. W., J. Appl. Polym. Sci. 19, 265 (1975). 30. Blythe, A. R., Briggs, D., Kendall, C. R., Rance, D. G., and Zichy, V. J. I., Polymer 19, 1273 (1978). 31. Catoire, B., Bouriot, P., Demuth, O., Baszkin, A., and Chevrier, M., Polymer 25, 766 (1984). 32. Carley, J. F., and Kitze, P. T., Polym. Eng. Sci. 20, 330 (1980). 33. Steele, J. G., Dalton, B. A., Johnson, G., and Underwood, P. A., Biomaterials 16, 1057 (1995). 34. Underwood, P. A., and Bennett, F. A., J. Cell Sci. 93, 641 (1989). 35. Van Wachem, P. B., Mallens, B. W. L., Dekker, A., Beugeling, T., Feijen, J., Bantjes, A., Detmers, J. P., and Van Aken, W. G., J. Biomed. Mater. Res. 21, 1317 (1987). 36. Grinnel, F., and Feld, M. K., J. Biol. Chem. 257, 4888 (1982). 37. Truskey, G. A., and Pirone, J. S., J. Biomed. Mater. Res. 24, 1333 (1990). 38. Iuliano, D. J., Saavedra, S. S., and Truskey, G. A., J. Biomed. Mater. Res. 27, 1103 (1993). 39. Kesler, K. A., Herring, M. B., Arnold, M. P., Glover, J. L., Park, H. M., Helmus, M. N., and Bendick, P. J., J. Vasc. Surg. 3, 58 (1986). 40. Prendiville, E. J., Coleman, J. E., Callow, A. D., Gould, K. E., Laliberte-Verdon, S., Ramberg, K., and Connolly, R. J., Eur. J. Vasc. Surg. 5, 311 (1991).