kinetics of cell adhesion to polymer surfaces

Kinetics of cell adhesion to polymer surfaces

D. R. Absolom,**,**tr*rs C. Thornson,**+ L. A. Hawthorn,*rt W. Zingg,*'f.5 and A. W. N e ~ m a n n * - ~ , ~ **imaging Science Associates, 44 Charles Street West, Suite 801, Toronto, Ontario, Canada M4Y 1R7; *Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada M5G 1 X 8 ; tDepartments of Mechanical Engineering and *Surgery and 'Institute of Biomedical Engineering, University of Toronto, Toronto, Ontario, Canada M5S 1A4

Results of the kinetics of adhesion of granulocytes as well as fresh and glutaraldehyde- fixed erythrocytes, suspended in Hanks Balanced Salt Solution (HBSS; pH 7.2, ionic strength of 0.15) to various polymeric substrates are presented. Cell adhesion increases rapidly initially and reaches a plateau value after approximately 30 minutes. There is no evidence for a lag-time in the onset of cell adhesion, suggesting that electrostatic double-layer forces are negligible under these experimental conditions. For the experiments in which the cells are suspended in HBSS, which has a surface ten-

sion larger than that of the cells, the level of cell adhesion increases with decreasing surface tension of the polymeric substrates. An additional experiment with fresh human granulocytes suspended in HBSS containing 10% dimethyl sulfoxide was also performed. The surface tension of the re- sulting liquid medium is below that of the cells and the pattern of adhesion is reversed, in agreement with the predictions of a thermodynamic model for cell adhesion. The slightly different behavior of siliconized glass as a substrate is discussed in terms of "screening."

INTRODUCTION

The kinetics of cell adhesion are of interest in a number of different areas including thrombus formation, hemostasis, secondary tumor growth and phagocytosis. In our previous work on cell adhesion our main interest was a comparison of the extent of adhesion to various solid substrates.'-' These experiments indicated that the number of cells adhering per unit surface area was dependent on the surface tension of both the suspending aqueous phase and the substrates themselves. For short contact times - typically 30 min or less-it was observed that when the surface tension of the suspending liquid, yLvl is larger than the surface tension of the adhering celIs, ycvl then the extent of cell adhesion decreased in a linear fashion with increasing substrate surface tension, ysv. When yLv < ycv the opposite pattern of cell adhesion was found to exist, i.e., increased adhesion with increasing ysv. These patterns of behavior are exhibited by a wide variety of cell types including erythrocyte^,^,^ and bacteria7 and appear to be a general phenomenon.

We have previously examined, under static test conditions, the kinetics of porcine platelet adhesion to a variety of polymer surfaces for up to 3-hr contact times.6 The results of that work indicated: (a) that platelet adhesion

Address correspondence to: D. R. Absolom at Imaging Science Associates.

Journal of Biomedical Materials Research, Vol. 22, 215-229( 1988) 0 1988 John Wiley & Sons, Inc. CCC 0021-9304/88/030215-15$04.00

21 6 ABSOLOM ET AL.

reaches a saturation level of adhesion, which is not transport controlled but depends on the surface tension, ysv, of the substrate material, i.e., the extent of cell coverage is a function of ysv reflecting the role of van der Waals interactions in cell adhesion; and (b) that for the contact times examined, five to 180 min, that there was no detectable time lag involved in the process of adhesion.

In view of the possible importance of these observations we wished to determine whether they would apply to other cell types as well. We report here on the kinetics of fresh erythrocyte and granulocyte adhesion to a wide range of polymer surfaces. Erythrocytes are known to play a role in the formation of the so-called "red thrombus" under conditions of low shear, whereas granulocytes are being increasingly recognized as being implicated in the thrombotic process both from studies involving complement activation,* lo

as well as cell spreading and endothelialization of polymer surfaces."-'3 In addition we also report here on the kinetics of glutaraldehyde-fixed human erythrocyte adhesion. These cells were examined in order to assess whether the aforementioned observations were indeed a general phenom- ena associated with cell adhesion and also because aldehyde-fixed cells have several properties which make them suitable for studies of this type. These include: 1) fixation prevents the loss of proteins and glycoproteins from the cell surface which might otherwise preadsorb onto carefully prepared and characterized polymer substrates rendering interpretation of the experimental data more difficult; 2) fixation also increases cell rigidity, thereby sta- bilizing the area of contact between the cell and substrate; 3) glutaraldehyde fixation14 produces little change in the overall electrostatic charge on the cell surface'' or in the local distribution of charge-bearing glycoproteins in the membrane lipid bilayer. l6 It should be noted that protein loss and cell rigidity may be important with respect to in vivo cell adhesion. The present experiments with fresh granulocytes and erythrocytes as well as with fixed erythrocytes are thus likely to provide some insight to the relative importance of these variables.

MATERIALS AND METHODS

1. Substrates

Cell adhesion experiments were performed using the substrate materials listed in Table I. Preparation of the substrates was performed as described previously.' Smooth films of these polymers (except for polystyrene and sulphonated polystyrene which are available commercially as thin films) were made in a hydraulic heat press (Wabash, IN) by pressing strips of the polymers between chromic acid cleaned glass slides. All surfaces with the exception of sulphonated polystyrene have no ionizable groups or semiper- manent charges. They are also inert in the sense of not containing chemically reactive groups such as hydroxyl, amine or isocyanate. The surface tension

CELL ADHESION KINETICS 21 7

TABLE I Preparation and Characterization of the Polymer Surfaces Employed

Contact Angle with Water Surface Tension

Po 1 y m e r Source Preparation (“1 (ergs/cm’)

Fluorinated Ethylene Propylene Copolymer (FEP)

(SW Dimeth yldichlorosilane

Polytetrafluoroethylene (PTFE)

Polystyrene (W

Low Density Polyethylene (LDPE)

Acetal (A4

Sulphonated Polystyrene ( S W

Commercial Plastics, Toronto

Eastman- Kodak,

Rochester Commercial

Plastics, Toronto Central

Research Lab, Dow Chemical

Commercial Plastics, Toronto

Commercial Plastics, Toronto Central

Research Lab, Dow Chemical

Heat Press 110 * 3 16.4

Vapor 108 2 2 17.6 Deposition

Heat Press 104 ? 4 19.8

Film 95 * 3 25.6

Heat Press 84 f 5 32.5

Heat Press 64 ? 3 44.6

Film 24 i 3 66.7

(ysv) of the polymers (cf. Table I) was calculated from contact angle data using the equation of state a p p r o a ~ h . ’ ~ Immediately prior to use the samples (with the exception of sulphonated polystyrene) were sonicated in analytical grade ethanol, rinsed in several volumes of deionized distilled water and allowed to dry in a dust-free environment.

2. Liquids

The cells were suspended in calcium- and magnesium-free Hanks’ Bal- anced Salt solution (HBSS), pH 7.2 and an ionic strength of 0.15. The surface tension of HBSS, measured by means of the Wilhelmy Plate method,’’ was found to be 72.8 ergs/cm2 at 25°C. For the granulocyte studies only one additional experiment to determine the effect of liquid surface tension on the kinetics of adhesion was performed. Here the cells were suspended in HBSS containing 10% (vol/vol) dimethyl sulfoxide (DMSO). Following preparation of the HBSS-DMSO mixture the pH of the solution was adjusted to pH 7.2 through the dropwise addition of 0.01 M sodium hydroxide. The surface tension of this solution was 64.6 ergs/cm2 at 25°C.

21 8 ABSOLOM ET AL.

3. Cells a. Granulocytes

Human granulocytes were isolated from whole human blood according to the method of Boyum.” In brief, whole human blood mixed with 0.05% (w/v) Na2EDTA (final concentration) was diluted 1 : 3 with 0.15 M NaC1. In a series of polycarbonate tubes (16 x 125 mm) 6 mL of the diluted blood was then layered on top of 3 mL Lymphoprep (Nyegaard and Co., Oslo, Nor- way) and centrifuged for 10 min at 400g. Following centrifugation the blood cells were separated into two fractions: a white layer at the interface region consisting of mononuclear cells and a bottom fraction containing both erythrocytes and granulocytes. The plasma layer was clear and contained no cells. The latter was removed with about half the volume of Lymphoprep. The remainder of the Lymphoprep is then removed to about 1-2 mm above the erythrocyte layer. The top 4 mm (granulocyte-rich) of this layer was then placed in tubes each containing one ml of the cell free Na2EDTA-plasma. Next, 0.4 mL of dextran T-500 (4.5% (w/v) in saline) (Pharmacia, Piscataway, NJ) was added to each tube. The contents were mixed and transferred to small polycarbonate tubes with an inner diameter of 8.5 mm; the cell column was about 40 mm high. The tubes were then allowed to stand for approximately 1 hr at 4°C during which time the erythrocytes aggregated and were allowed to settle. After this period the plasma layer containing the granulocytes was carefully removed to about 1-2 mm above the aggregated erythrocytes. The granulocyte-rich plasma was centrifuged and the cells washed three times in HBSS to remove plasma proteins. The granulocyte preparation was then examined. Purity of the granulocyte suspension was always greater than 90%. Viability of the isolated granulocytes was evaluated using both the trypan blue dye exclusion2* and fluorescein diacetate hydrolysis2* tests. Viability was always better than 96%. The granulocytes were then suspended in either of the two solutions and the cell concentration adjusted to yield 1 X lo6 cells/mL.

b. Fresh erythrocytes

Fresh erythrocytes were isolated using the isolation procedure described above for granulocytes. Following the initial centrifugation the mononuclear cells were removed together with the remaining Lymphoprep. Thereafter, the top 4 mm of cells (granulocyte-rich fraction) in the erythrocyte column was removed. The remaining cells (>99% erythrocytes) were washed five times in 0.15 M NaCl containing 1 mM of both Na2EDTA and Na2EGTA in order to ensure complete removal of any divalent cations. The cells were then washed twice in calcium- and magnesium-free HBSS. Erythrocyte concentration was adjusted to 1 x lo6 cells/mL.

c. Glutaraldehyde-fixed erythrocytes

Human erythrocytes (isolated as described above) were fixed in fresh 3.3% (v/v) glutaraldehyde-phosphate buffered saline solution (pH 7.1) following

CELL ADHESION KINETICS 21 9

the method of Sabatini et d . 1 4 After fixation, the cells were resuspended and washed five times in 0.15 M NaCl containing 1 mM of both Na2EDTA and Na2EGTA. Thereafter the fixed cells were washed five times in calcium- and magnesium-free HBSS. The cells were suspended in HBSS and used at a final concentration of 1 x lo6 cells/mL.

Microscopic examination of all cell suspensions used in this study revealed that the cells exhibited normal morphology and were not clumped under the experimental conditions employed. All erythrocytes used in this study, i. e., both fresh and fixed, were from a single bleed so as to avoid problems which may arise as a result of extensive heterogeneity of cell surface carbohydrate content of circulating erythrocytes2 which could otherwise influence the adhesive behavior of the cells to the various substrates.

Static adhesion tests to the various surfaces were performed as described previously. ' f 4 t 5 Briefly 1 ml of cell suspension, containing 1 X lo6 cells/mL, in the appropriate liquid, was placed on the surfaces and was retained in wells formed on Teflon molds separated from the polymers by Silastic gaskets. The height of the cell suspension was approximately 1.0 an.

The cells were then incubated at 25°C for various times ranging from 2 min to 120 min. Thereafter the surfaces were placed, without exposure to air, in a large volume of the suspending fluid and rinsed under standardized conditions in order to remove any nonadherent cells. For this purpose we employed a rinsing device described p rev i~us ly .~~ After rinsing, in the case of the fresh erythrocytes only, the cells were futed in 2.5% (v/v) glutaraldehyde-in phosphate-buffered saline, pH 7.2, for 30 min. All slides were then air-dried and in the case of granulocytes stained using the Wright stain. The number of cells adhering to the various surfaces was then determined, after brief immersion in deionized distilled water to dissolve any salt crystals, using an automatic image analysis system (Omnicon 3000, Bausch and Lomb, Rochester, NY). For this purpose a Wild-Leitz Metalloplan microscope was connected to a Bosch camera containing a Chalnicon tube. Computer controlled automation of the microscope stage permitted rapid evaluation of 126 separate fields of view of known area for each well. In view of "wall-effects" in these adhesion tests, illustrated for fixed erythrocytes in Figure 1, only the central section of each well are assessed. The data from each of these individual field measurements were then averaged and expressed as the number of erythrocytes adherent per unit surface area of substrate material. For each experiment at least six wells per type of polymer surface were examined. Each experiment was repeated at least three times.

RESULTS AND DISCUSSION

The contact angles and the calculated surface tensions of the substrates employed are given in Table I.

Results for granulocyte, fresh, and fixed human erythrocytes are given in Figures 2,3, and 4, respectively. It should be noted that the general features

220 ABSOLOM ET AL.

t z

0 1 0 m l Cenier of well

t 1 I

pm from left edge of well 0 1400 2800 4200 5600 7000 8400 9900

Figure 1. The influence of "wall-effects" in determining the extent of glutaraldehyde-fixed erythrocyte adhesion to a FEP surface. The erythrocytes, at a concentration of 1 x lo6 cells/mL in Hanks Balanced Salt Solution (pH 7.2, ionic strength of 0.15), are retained in a Teflon mold. Contact time is 30 min.

c 0

z 0

'"1 Substrates FEP

0 PTFE 0 SIL A PS VLDPE AAcetal

SPS

I I I 1 0 20 40 60 80

Time (min) Figure 2. Kinetics of fresh human granulocyte adhesion to various surfaces. Cell concentration is 1 x lo6 cells/mL; cells suspended in HBSS. Substrates as indicated. The associated error limits are 93% confidence limits. (For graphical reasons errors are shown only for some cases; errors are similar in all cases.)

of the curves for the three types of cells are identical. The key points to consider are:

a) Within experimental accuracy, no lag time in the onset of adhesion was noted;

CELL ADHESION KINETICS 221

Substrates DFEP OSlL A PS V LDPE

- 5 0 0 E E $ 4 0 0 8 SPS

2 2

O 2oa 0

. v)

% 0

5 300

W r

z 100

0 I ' 1 ' 1 ' 1 0 10 20 30 40 50 60 70

Time (min)

Figure 3. Kinetics of fresh human erythrocyte adhesion to various surfaces. Cell concentration is 1 x lo6 cells/mL; cells suspended in HBSS. Substrates as indicated. The associated errors are 95% confidence limits. (For graphical reasons errors are shown only in some cases; errors are similar in all cases.)

2000-

Substrates FEP

OSlL A P S VLDPE .SPS

N

0 10 20 30 40 50 60 70

Time (min)

Figure 4. Kinetics of glutaraldehyde-fixed human erythrocyte adhesion to various surfaces. Cell concentration is 1 x 10' cells/mL; cells suspended in HBSS. Substrates as indicated. The associated errors are 95% confidence limits. (For graphical reasons errors are shown only in some cases; errors are similar in all cases.)

b) Under the present experimental conditions the process of adhesion does not go to completion, i.e., while a large excess of nonadherent cells exists, the maximum surface coverage observed was approximately 10% of the total surface area available;

c) The extent of cell coverage of the various surfaces is a function of substrate surface tension, ysv.

222 ABSOLOM ET AL.

These observations are essentially identical with those noted previously for platelet adhesion.' Thus we conclude that this pattern of behavior is not peculiar to platelets but is a common feature of cell adhesion and probably particle adhesion in general. Furthermore, the extent of cell coverage for any particular cell type is never complete but is constant for any given surface. The finite plateau level of adhesion is different for each substrate suggesting that this level of cell attachment is dependent on substrate surface tension, ysv. That the properties of the substrate materials play an important role in determining the extent of cell-substrate interactions is further illustrated in Figures 5 through 7. Here, the individual plateau levels of cell adhesion for each of the polymer surfaces, obtained from Figures 2,3, and 4, respectively, are replotted as a function of substrate surface tension, ~ S V . It is clear from these plots that the level of cell adhesion decreases with increasing substrate surface tension. In the case of fresh human erythrocytes (Fig. 6) we note for all contact times examined that the level of cell adhesion decreases in a linear manner. However, in the case of granulocytes (Fig. 5) and fixed erythrocytes (Fig. 7) a different pattern is observed. For granulocytes we note a linear decrease for contact times greater than 4 min. At shorter contact times a distinct curvature on the low ysv surfaces is seen. In contrast fixed erythrocytes exhibit a linear decrease in adhesion with increasing substrate surface tension, ysv, for short contact times, i.e., less than 8 min and curvature, again on the low-energy surfaces, at longer contact times. The reasons for the differences between the different types of cells is not yet clear.

These observations are in general agreement with thermodynamic predictions of cell adhesion based on a model which considers only van der Waals

3 7

r 0

2 0

0 lmln a 2 min \ - - .. ... . V4 min 0 8min A 16 mln

32,64 min

0 1 8 \ r 1 , i a - I 1 0 20 30 40 20 6b 70

Substrate Surface Tension (ergs/cm2)

Figure 5. Fresh human granulocyte adhesion as a function of substrate surface tension for various contact items as indicated. The associated errors are 95% confidence limits. (For graphical reasons errors are shown only in some cases; errors are similar in all cases.)

CELL ADHESION KINETICS

1"-

223

500-

400- Contact time! 0 2 rnin

N 0 5 min 0 10 min A 15,30 rnin

E

I - 1 -

Figure 6. Fresh human erythrocyte adhesion as a function of substrate surface tension for various contact times as indicated. The associated errors are 95% confidence limits. (For graphical reasons errors are shown only in some cases; errors are similar in all cases.)

'"1 c*

E E

0 2min Q 4 min 0 8min A 16 min

20,120 min

0 ' 1 I I , I I I 1 1 I I \

:O 20 30 40 50 60 70

Substrate Surface Tension (ergs/cm?

Figure 7. Fixed human erythrocyte adhesion as a function of substrate surface tension for various contact times as indicated. The associated errors are 95% confidence limits. (For graphical reasons errors are shown only in some cases; errors are similar in all cases.)

interaction^."-*^ The agreement between theoretical predictions and experimental observations for the different systems examined strongly suggests that the extent of cell adhesion to polymer surface in the absence of proteins

224 ABSOLOM ET AL.

is primarily governed by van der Waals forces. This conclusion is strength- ened by the absence of any lag time in the adhesion profile which would be indicative of the influence of electrostatic forces in the process of cell attachment to polymer surfaces.

It is clear from Figures 2 through 7 that the extent of cell adhesion is influenced not only by substrate surface properties but also by the surface properties of the adhering cells. We have previously determined the surface tension of the three types of cells employed in the present study using a variety of techniques.26-28 Those surface tensions are summarized in Table 11. From these data it is clear that the aldehyde-fixed erythrocytes are the most hydrophobic cells followed by granulocytes and fresh erythrocytes respectively. When suspended in HBSS the thermodynamic model predicts that the most hydrophobic moiety, i.e., fixed erythrocytes, would adhere to the largest extent. This prediction is indeed confirmed experimentally; for example on an FEP surface with a contact time of 60 min approximately 1650 fixed erythrocytes/mm2 are observed as compared to about only 530 fresh erythrocytes/mm2. It would not be correct to include granulocytes in this comparison since they are considerably larger cells which also exhibit the tendency to spread to a much greater extent on the various surfaces. We would like to point out, however, that previous studies with normal and pathological (more hydrophobic) granulocytes (obtained from patients with juvenile idiopathic periodontal disease) have shown, under experimental conditions comparable to the present study, that the more hydrophobic granulocytes adhered to a greater extent.'

In Figures 2 and 8, we illustrate the effect of the surface tension of the suspending solution, yLv, on the kinetics of granulocyte adhesion. Two experimental conditions were examined. In Figure 2 the granulocytes are suspended in HBSS only so that the surface tension of the granulocytes, ycv, is less than the surface tension ~ L V of HBSS, i.e., yLv > ycV (cf. Table 11). In the second condition (Fig. 8) the cells were suspended in a 10% (v/v) mixture of HBSS-DMSO in which yLv < ycv. First we wish to note that there are no significant differences in the general shape of these curves as compared to Figures 2 through 4. In particular, no lag-time in the adhesion profile is observed. However, the pattern of adhesion to the various surfaces is different. In Figure 2, i.e., when yLv > ycv, it is noted that the plateau Ievel of cell

TABLE I1 Comparison of the Surface Tension of Cells Used.

Corrected to 22 "C assuming dy/dT = 0.1 ergs/cm*/"C.

CeIl Surface Tension (ergsicm')

Method Granulocytes Fresh Erythrocytes Fixed Erythrocytes

Contact Angle 69.1 Adhesion 69.0 Freezing Front 68.3

Sedimentation Volume -

Droplet Sedimentation -

70.1 ~

~

- 70.0

-

64.6 64.2 64.3 64.5


0- 0 20 40 6 0 60

Time lrninl

Figure 8. Kinetics of fresh human granulocyte adhesion to various surfaces. Cell concentration is 1 x lo6 cells/mL; cells suspended in HBSS containing 10% DMSO (v/v). Substrates as indicated. The associated errors are 95% confidence limits. (For graphical reasons errors are shown only in some cases; errors are similar in all cases.)

adhesion is larger on the more hydrophobic surfaces and decreases with increasing substrate surface tension, ysv. However, in Figure 8, when yLv < ycv the extent of granulocyte adhesion was reversed with the extent of cell adhesion being most pronounced on the more hydrophilic surfaces, e.g., sulphonated polystyrene, and least on the more hydrophobic surfaces.

Thus the present study has confirmed that the extent of adhesion, not only of platelets,6 but also of erythrocytes (both fresh and aldehyde-fixed) and granulocytes is not kinetically detem~ined.*~-~l The key observation in this respect is that in the type of experiment described here the maximum cell coverage reached is only of the order of 10% or less of the available substrate surface area. The extent of cell coverage is determined by the relative surface tensions of the adhering cells (ycv), the substrate material (ysv) and the suspending liquid medium (yLv). These parameters determine the net free energy change for the process of cell adhesion for any particular system. A thermodynamic m ~ d e l ' ~ , ~ for cell adhesion to polymer surfaces predicts that the lower, i.e., the more negative the free energy of adhesion, the higher will be the plateau level of cell adhesion. These predictions are in agreement with the present experimental observations. However, this model does not ex- plain why saturation of the polymer surface does not occur.

Since there is a close correspondence between the free energy of adhesion and van der Waals forces, the results imply that cell adhesion to polymer surfaces under physiological conditions of pH and ionic strength, is largely governed by van der Waals interactions between the cells and the polymer substrates. Since there is, within the experimental limitations of our experi-

226 ABSOLOM ET AL.

ments, no evidence for a lag-time in the onset of cell adhesion, we conclude that electrostatic double layer forces do not play a significant role under the present experimental conditions. This conclusion is in agreement with recent results on the pH and ionic strength dependence of the adhesion of aldehyde-fixed erythrocyte^.^ There it was found that double layer interactions become appreciable only at much lower ionic strengths than employed in the present study.

As noted above the level of cell adhesion decreases with increasing substrate surface tension. There is, however, one minor, although significant exception to this general trend. Adhesion to a siliconized glass surface is considerably less than that noted on a hydrophobic FEP surface even though these surfaces both have approximately the same surface tension of about 17 ergs/cm2. This is particularly noticeable for granulocytes in Figure 2 in which in addition to FEP and siliconized glass we have also examined adhesion to another hydrophobic surface - PTFE - which has a surface tension larger than siliconized glass. From Figure 2 it is clear that the extent of adhesion is less on the silane surface than on the more hydrophilic PTFE surface. Generally the level of adhesion on the siliconized surface is similar to that observed on the polystyrene surface which has a surface tension of approximately 26 ergs/cm2. Thus, the siliconized glass behaves as one would expect of a more hydrophilic surface. There are two possible explanations for this pattern of behavior: 1) The hydrocarbon (silane) layer may not be perfect and thus allow some contact between glass and cell. This possibility, however, is not tenable since contact angle measurements performed over various sections of the siliconized surface revealed very little variability. 2) The second possibility, which may be operative even if the silane surface is perfect, is the phenomenon known as "~creening.""~~~ This is due to the fact that the depth of the van der Waals interaction between two phases is of the same order as the separation distance between the two phases. Thus, since the silane layer is only -12 8, thick (as measured by ellipsometry; Model L116B-85Br Gaertner Instruments, Chicago, IL), and since the van der Waals interactions are appreciable over hundreds of Angstroms, the approaching cells will interact not only with the silane layer, but also with the underlying glass substrate. This type of apparent anomally in behavior of siliconized surfaces has been observed not only with cell adhesion but also with protein adsorption35 studies.

CONCLUSIONS

The present experiments show for several cell types, under conditions of high ionic strength and a solution pH of 7.2, that there is no discernible time lag in the kinetics of adhesion to a wide range of polymer surfaces; this suggests that electrostatic double-layer forces, under these conditions, play no or only a minor role. For any particular cell type the extent of adhesion reveals a characteristic level of saturation, i.e., degree of cell coverage, which is dependent on the surface tension of the substrate materials. In the pres-

227 CELL ADHESION KINETICS

ence of a large excess of cells the maximum cell coverage is of the order of 10% of the available substrate surface area indicating that cell adhesion is not determined by kinetic factors. Since the patterns of adhesion can be pre- dicted in large part by means of a thermodynamic model for the free energy of adhesion, it appears that van der Waals forces (under physiological conditions) are the most important factor in determining the extent of cell adhesion to polymer surfaces.

Supported in part by research grants from the Medical Research Council of Canada (MT-5462, MT-8024, MA-9114), the Natural Sciences and Engineering Research Council of Canada (A-8278, UO-408) and the Ontario Heart Foundation (4-12, AN-402). One of us (D. R. A,) acknowledges support through receipt of an Ontario Heart Foundation Senior Fellowship.

References 1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

A. W. Neumann, D. R. Absolom, W. Zingg, and C. J. van Oss. "Surface thermodynamics of cell adhesion," Cell Biophys., 1, 79-92 (1979). D.R. Absolom, C. J. van Oss, R. J. Genco, D. W. Francis, and A.W. Neumann, "Surface thermodynamics of normal and pathological human granulocytes," Cell Biopkys., 2, 113-126 (1980). W. Zingg, D. R. Absolom, 0. S. Hum, A. B. Strong, and A. W. Neu- mann, "Platelet adhesion to smooth and rough surfaces," Can. J. Surg.,

D. R. Absolom, W. Zingg, C. Thomson, Z . Policova, C. J. van Oss, and A. W. Neumann, "Erythrocyte adhesion to polymer surfaces," J. Colloid Interface Sci., 104, 51-59 (1985). D.R. Absolom, C. Thomson, G. Kruzyk, W. Zingg, and A. W. Neu- mann, "Erythrocyte adhesion to polymer surfaces. Effect of pH and ionic strength," Colloids and Surfaces, 21, 447-456 (1986). D. R. Absolom, F. V. Lamberti, Z. Policova, W. Zingg, C. J. van Oss, and A. W. Neumann, "Surface thermodynamics of bacterial adhesion," J. Appl. Environ. Microbiol., 46, 90-97 (1983). A. W. Neumann, 0. S. Hum, D. W. Francis, W. Zingg, and C. J. van Oss, "Kinetic and thermodynamic aspects of platelet adhesion from suspension to various substrates," J. Biomed. Mater. Res., 14,499-509 (1980). C. A. Ward, "Activation of complement at plasma-air or serum-air interface of rabbits," J. Appl. Pkysiol., 60, 1651-1658 (1986). G. A. Herzlinger and R. D. Cumming, "Role of complement activation in cell adhesion to polymer blood contact surfaces," Trans. A m . SOC. Artif. Intern. Organs, 26, 165-171 (1980). G. A. Herzlinger, D. H. Bing, R. Stein, and R. Cumming, "Quantitative measurement of Cs activation at polymer surfaces," Blood, 57, 764-770 (1981). D. R. Absolom, L. A. Hawthorn, and G. C. Chang, "Endothelialization of polymer surfaces," J. Biom. Mater. Res., in press. R. Fasol, P. Zilla, P. Groscurth, E. Wolner, and R. Moser, "Experimental in vitro cultivation of human endothelial cells on artificial surfaces," Trans. Am. SOC. Artif. Intern. Organs, 31, 276-283 (1985). P. B. van Wachem, T. Beugeling, J. Feijen, A. Bantjes, J. P. Detmers, and W. G. van Aken, "Interaction of cultured endothelial cells with polymer surfaces of different wettabilities," Biomateriuls, 6, 403-408 (1986). D. D. Sabatini, K. Bensch, and R. J. Barnett, "Cytochemistry and elec- tron microscopy- preservation of cellular ultrastructure and enzymatic activity by aldehyde fixation," J. Cell B i d , 17, 19-58 (1963).

25, 16-19 (1982).

228 ABSOLOM ET AL.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

I? Pinto da Silva, ”Translational mobility of the membrane intercalated particles of human erythrocyte ghosts,” J. Cell Biol., 53, 777-787 (1972). P. S. Vassar, J. M. Hards, D. E. Brooks, B. Hargenberger, and G. V. F. Seamen, ”Physicochemical effects of aldehydes on human erythrocyte,” 1. Cell Biol., 53, 809-814 (1972). A. W. Neumann, R. J. Good, C. 1. Hope, and M. Sejpal. “An equation- of-state approach to determine surface tensions of low-energy solids from contact angle measurements,” 1. Colloid Interface Sci., 49, 291-297 (1974). A. W. Neumann, R. J. Good, P. Ehrlich, P. K. Basu, and G. J. Johnston, ”The temperature dependence of the surface tension of solutions of atactic polystyrene,” J. Mucromol. Sci. Phys., B7, 525-531 (1973). A. Boyum, “A one-stage procedure for isolation of granulocytes and lymphocytes from human blood- general sedimentation properties of white blood cells in a 1 g gravity field,” Scund. J. Clin. Lab. Invest., 21,

D. R. Absolom, C. J. van Oss, and A. W. Neumann, ”Elution of human granulocytes from nylon fibres by means of repulsive van der Waals forces,” Trunsftlsion, 21, 663-674 (1982). B. Rotman and B. W. Papermaster, ”Membrane properties of living mammalian cells studied by enzymatic hydrolysis of fluorogenic esters,” Proc. Nutl. Acud. Sci. (U.S.A.), 55, 134-141 (1966). A. Baxter and J.G. Beeley, ”Surface carbohydrates of aged erythrocytes,“ Biochem. Biophys. Res. Comm., 83, 466-471 (1978). 0. S. Hum, A. W. Neumann, and W. Zingg, “Platelet interaction with smooth solid substrates determined by an open static method,” Thromb. Res., 7, 461-470 (1975). D. W. Francis, ”Interfacial tensions and van der Waals interactions of small particles at solid-liquid interfaces,” 1’h.D. Dissertation, University of Toronto, 1983. R.P. Smith, D.R. Absolom, J .K. Spelt, and A. W Neumann, “Ap- proaches to determine the surface tension of small particles: Equa- tion of state considerations,” l. Colloid Interface Sci., 110, 521-532 (1986). D. R. Absolom, W. Zingg, and A. W. Neumann, “Measurement of surface tensions of cells and biopolymers with application to the evaluation of biocompatibility,“ in Comprehensive Biofechnology, Vol. I l l , Chapter 22, C. C. Cooney and A. E. Humphrey (eds.), Pergamon Press, New York, 1985, pp. 433-446. A.W. Neumann, D.R. Absolom, D. W. Francis, S.N. Omenyi, J .K. Spelt, C. Thomson, W, Zingg, and C . J. van Oss. ”Measurement of surface tensions of blood cells and other particles,” Ann. N.Y. Acud. Sci.,

D. R. Absolom, ”Measurement of the surface properties of phagocytes,” in Methods in Enzymology, S . P. Calowick and N. 0. Kaplan (eds.) Aca- demic Press, New York, N.Y. 1986, pp. 16-95. E. Ruckenstein, A. Marmur, and S. R. Rakower, ”Sedimentation and adhesion of platelets onto a horizontal glass surface,” Thrombos. Hue- mostas. (Stutfgurt), 36, 334-342 (1976). E. Ruckenstein and R. Srinivasan, ”The origin of platelet deposition-is it kinetic or thermodynamic?” J. Biomed. Mat. Res., 16, 169-172 (1982). A. W. Neumann, D. W. Francis, W. Zingg, C. J. van Oss, and D. R. Abso- lom, ”Comments on two recent publications: (I) The origin of platelet deposition and (2) Guest editorial,” 1. Biomed. Muter. Res., 17, 375-381 (1983). D. Langbein, “Van der Waals attraction in multilayer structures,” 1. Adhesion, 3, 213-235 (1972).

(Suppl. 97), 51-98 (1968).

416, 276-298 (1983).

H. Krupp, ”Particle adhesion theory and experiment,” Adv. Colloid Inter- face Sci., 1, 111-239 (1967).


34.

35.

J. Visser, “The concepts of negative Hamaker coefficient history and present status,” Adv. Colloid Interface Sci., 15, 157-169 (1981). D. R. Absolom, W. Zingg, and A. W. Neumann, ”Protein adsorption to polymer particles: Role of surface properties,” I . Biorned. Muter. Res., 21, 161-171 (1987).

Received November 26, 1986 Accepted October 15, 1987

kinetics of cell adhesion to polymer surfaces

Documents