The effect of varying fibronectin concentration on the attachment of endothelial cells to polytetrafluoroethylene vascular grafts John S. Budd, F R C S , Karen E. Allen, BSc, Peter R. F. Bell, MD, and R o g e r F. L. James, Phi), Leicester, United Kingdom Endothelial cell seeding onto untreated polytetrafluoroethylene vascular prostheses is inefficient. In an effort to improve cell attachment, numerous investigators have used fibronectin as a coating material to pretreat the luminal surfaces of these prostheses. The concentrations o f fibronectin used have varied enormously, and no one has yet determined the most efficient concentration in terms o f cell attachment and cost. Using endothelial cells labeled with indium 111 oxine we have studied the effect o f varying fibronectin concentration on the attachment of these cells onto polytetrafluoroethylene vas~tlar grafts. Seeding efficiency was significantly better in all groups o f coated grafts, at all times (10, 30, 60, and 120 minutes), compared to uncoated controls (p < 0.01). Overall, fibronectin at a concentration of 20 ~g/ml was found to be the most efficient in terms o f cell attachment and cost since any further increase in concentration was not accompanied by increased cell attachment. We now routinely use fibronectin at a concentration of 20 ttg/ml to coat our grafts before endothelial cell seeding. (J VASe SURG 1990;12:12630.)

Since Herring et al.1 introduced the concept of endothelial cell seeding of vascular grafts in 1978, many researchers throughout the world have continued in their quest to produce a functional, biologic prosthetic graft for use in man. Two main factors appear to influence the success of autologous endothelial cell seeding. The first relates to the method and efficiency of harvesting the endothelial cells and is not the primary concern of this study. The second relates to increasing the seeding efficiency by enhancing endothelial cell attachment onto vascular graft material. Investigators studying ways of improving cell attachment have used various substances to coat the graft materials, and of these fibronectin appears to give consistently good results. 24 Although many people use fibronectin to coat their grafts before seeding, the amount of fibronectin varies considerably. Lindblad et al.s used a concentration of 10

From the Department of Surgery, Universityof Leicester. Supported by a grant from the Trent RegionalHealth Authority. Reprint requests: Mr. J. S. Budd,Department of Surgery, Clinical Sciences Building, Leicester Royal Infirmary, P.O. Box 65, Leicester LE2 7LX, United Kingdom. 24/1/21400

126

~g/ml, Lundgren et al. 6 used 100 ~g/ml, Ramalanjaona et al. 7 used 250 ~g/ml, and Foxall et al.8 used concentrations up to 1000 ~g/ml. The optimal concentration of fibronectin for coating vascular grafts, in particular polytetrafluoroethylene (PTFE), combining good endothelial cell attachment together '~ with minimum cost, has not been determined. To address this question we have studied the attachment ofradiolabeled human umbilical vein endothelial ceils onto the surface of PTFE vascular graft material coated with different concentrations of fibronectin. METHODS Preparation o f graft surfaces

Polytetrafluoroethylene vascular grafts of standard thickness (0.64 mm) and 6 mm internal diameter were used (a gift from W. L. Gore & Associates [U. K.] Ltd., Livingston, West Lothian, Scotland). The grafts were cut into short segments and opened lengthwise to expose the luminal surface. Each segment was then immobilized in a well constructed from a modified Eppendorf tube (Fig. 1). The well allowed 0.5 cm 2 of the luminal surface of the PTFE to be exposed. All subsequent surface treatments and endothelial cell seeding were performed on this immobilized surface.

Volume 12 Nmnber 2 At, ~',.st 1990

Endothelial cell seeding and fibronectin concentration 127

Fibronectin in various concentrations (2, 10, 20, 50, and 100 ~tg/ml made upin Dulbecco's minimum essential medium [DMEM]) was then used to coat the graft surface (fibronectin from Sigma Chemical Co. Ltd., Poole, U. K., and DMEM from Flow Lab~oratories Ltd., Irvine, Scotland, U. K.). One hundred cubic millimeters of the solution to be tested was placed in a well and left for 2 hours at room temperature before being removed. Controls were coated with plain DMEM. Endothelial cell ctfltures Endothelial cells were harvested from human umbilical veins. Umbilical cords were collected within 24 hours of deliver, and all clamp marks and needle holes were removed. The umbilical vein was can_alated at either end and then flushed gently with .DMEM to remove old blood. The vein was then distended with 5 ml 0.1% collagenase solution (CLS I, Worthington Biochemical, Freehold, N. J.; CaCI2 15 retool/L), prewarmed to 37 ° C, and the whole cord was incubated for 15 minutes again at 37 ° C. After it was incubated the vein was flushed through with 20 ml DMEM, and the resulting cell suspension was centrifuged for 7 minutes at 2 0 0 g at 4 ° C. The cell pellet was resuspended in 5 ml complete medium (medium M199, 20% fetal calf serum, 100 units/ml penicillin, 100 txg/ml streptomycin, 20 mmol/L Hepes buffer, 2 mmol/L glutamine, 1 mmol/L pyruvate [all Flow Laboratories, Scotland]), 20 ~,g/ml endothelial cell growth supplement, and 90 >g/ml heparin (both Sigma, England), plated onto a 25 cm 2 tissue culture flask and grown to confluence in an atmosphere of 5% CO2. Identification o f cells Confirmation of cndothelial cell identity was made by appearance at confluence on phase contrast microscopy and also by confirming the presence of yon Willebrand factor in the cells, by use of a rabbit, antihuman yon Willebrand factor (Dakopatts a/s 42, Produktionsvej, Glostrup, Denmark). Labeling o f endothelial cell At confluence the cells were harvested with 0.1% trypsin/0.02% ethylenediamine tetracetic acid (EDTA) at 37 ° C and resuspended in a small volume of serum-free medium. An average of 3.0 + 0.3 × 106 (mean __+SEM) cells were then labeled with 50~xCi indium 111 (mira) oxine at room temperature for 15 minutes. After the cells were labeled they were washed three times with D M E M and finally resus-

Fig. 1. Endothelial cell seeding chamber made from a modified Eppendorf tube. pended in 5 ml DMEM with 10% fetal calf serum in readiness for seeding. Before the cells were seeded a small sample of the labeled cell suspension was taken for cell counting and viability studies by use of trypan blue dye. Attachment o f endothelial cells to graft material One hundred cubic millimeters of the labeled cell suspension was added to individual wells containing the graft segments coated with fibronectin (0 to 1001xg/ml). The wells were incubated in 5% CO2 at 37 ° C for periods ranging from 10 to 120 minutes, and the unattached cells were then removed and saved (10 separate studies were carried out on each coating for each incubation time). The graft surface was then washed, and the washings were added to the previously removed cells. Radioactivity was then measured on the graft segments and in the unattached cells. The degree of cell attachment was expressed as a percentage of the initial cell inoculum and was calculated by means of the following formula: Percentage cell Counts on graft = × 3,00 attachment Counts in washings + on graft

Journal of VASCULAR SURGEv~Y

128 Buddet al.

% attachment lOO

80 Control

60

4o

hol o

10 20

i

30

40

i

i

i

I

50 60 70 80 T i m e (mins)

I

i

I

I

2 ug/ml

)l(

10 IJg/ml

C]

20 u g / m l

)(

50 IJg/ml

0

100 IJg/rnl

i

90 100 110 120

Fig. 2. The percentage attachment of endothelial cells to PTFE pretreated with different concentrations of fibronectin over a time period between 0 to 120 minutes.

Statistical analysis All data were analyzed with both a paired t test and Wilcoxon signed rank test. Results are expressed as mean +_ SEM. RESULTS The average number of endothelial cells seeded onto the graft segments was 1.20 +_ 0.12 × 102 cells per cm 2. The average labeling efficiency w i t h m I n was 50.7% -+ 2.5%. The cell viability after labeling, as determined by staining with trypan blue dye was always greater than 98%. The time trend for endothelial cell attachment to PTFE with the different fibronectin coatings is shown in Fig. 2. Endothelial cell attachment reaches 75% after only 10 minutes with fibronectin concentrations of 20 ~g/rrd and above and then plateaus out at between 80% and 85%. Increasing the fibroneetin concentration above 20~g/rul does not increase cell attachment. In contrast, PTFE with no fibronectin coating only reaches 10% cell attachment even after 120 minutes. Statistically all fibronectin concentrations are significantly better (p < 0.01) than the control at all times. All fibronectin coatings above 2 ~ g / m l are significantly better (p < 0.01) than this concentration at all times. At 10 minutes, fibronectin at 10 ~g/rnl is inferior (p < 0.05) to

concentrations above this, but as time progresses this significance disappears. With 20, 50, and 100 txg/ml there is essentially no overall statistical difference between them, although at sporadic times throughout the experiment, significance between one concentration and another may be reached at the p < 0.05 level. DISCUSSION Fibronectin is a large glycoprotein found in b o ~ blood and other body fluids and in an insoluble form in interstitial connective tissue and in or close to basement membranes. According to immunohistologic data, fibronectin is detected in the subendothelium of the vascular wall. 9'1° It promotes attachment and spreading of many cells, including endothelial cells, on polystyrene and glass, and this is the basis for its use in coating vascular grafts before endothelial cell seeding. 11 The precise mechanism of action of fibronectin as it relates to cell attachment is maknown, but it has been postulated that the interaction between fibronectin and endothelial cells is an active one, occurring between specific receptors on the endothelial cell surface and appropriate binding sites of the fibronectin molecule. 7 Various substances have been used to coat vas-

Volume 12 Number 2 Auo-~st1990

cular grafts before seeding including gelatin, laminin, plasma, collagen, and fibronectin. Most studies suggest that fibronectin is the best of these when used with PTFE grafts, 4,~2 although Thomson et al.13 found that for endothelial cell seeding fibrin preclot -was better, but for mesothelial cell seeding fibronectin was superior. Although fibronectin is used by many people, there have been no previous studies to determine which concentration offibronectin is most efficient in terms of cell attachment and cost. The concentrations used previously have varied 100,fold, s,8 and this study has been carried out to determine the optimum concentration of fibronectin to use for precoating PTFE vascular grafts before endothelial cell seeding. The results indicate that fibronectin concentrations above 20 ~g/ml confer no ben, 5:, in terms of improved endothelial cell attachment, ~and have the disadvantage of higher costs. Although Rarnalanjaona et alp found that higher concentrations of fibronectin bound to PTFE at an increased rate and also that there were more fibronectin molecules bound at the higher concentrations, this seems to make no difference to the number of endothelial cells subsequently bound. We have therefore chosen to use fibronectin at a concentration of 20 p~g/ml for "all our precoating. Recently it has been demonstrated that rain, apart from being a useful marker for endothelial cells, also has an affinity for PTFE. This affinity may be augmented by the presence offibronectin on the graft surface. Although not performed in this study, one "method of assessing this effect would be to take the supernatant from the wells after endothelial cell seeding and measure the radioactivity. Control grafts, coated with appropriate fibronectin concentrations, could then be incubated for the various time periods with medium containing rain with the same activity as the supernatant. This would indicate, in the first instance, whether fibronectin does increase the free rain uptake onto PTFE and, second, whether this is dose dependent. Since pretreatment of PTFE with fibronectin clearly increases initial endothelial cell attachment, 'this would suggest that it will also facilitate growth to confluence as it does in in vitro tissue culture. However, this process takes time, and in the canine model up to 6 weeks is required if immediately harvested endothelial cells are used to seed PTFE vascular grafts. During this period, areas of the graft surface not covered with endothelial cells remain exposed to the blood components and are potential

Endothelial cell seeding and fibronectin concentration

129

points of platelet accumulation. As one might postulate, since fibronectin increases endothelial cell attachment to grafts, it also increases platelet accumulation where it is exposed and therefore increases the thrombogenicity of the graft flow surfaces.~4 This is a distinct disadvantage, and in theory the lower the fibronectin concentration used to coat the graft the better. Even with the lower fibronectin concentrations it is probably still advisable to treat with antiplatelet drugs as suggested by Rama!anjaona. 7 Although fibronectin would appear to be a suitable substance for use in precoating PTFE grafts before endothelial cell seeding further studies on surface preparations and indeed graft materials are necessary to find an optimal seeding environment. To eliminate the adverse effects of exposed areas of graft or coating material seen with immediate seeding methods, it would seem appropriate to try to produce a confluent endothelial cell monolayer on the graft before implantation, as has been postulated by a number of workers. The advantages of using fibronectin as a coating material could then be applied more fully, without the disadvantages of increased ptatelet adherence. We thank W. L. Gore and Associates (U. K.) Ltd. for supplying the graft material and the Radiopharmacy Depamnent, Leicester Royal Infirmary~ for supplying the REFERENCES

1. Herring M, Gardner A, Glover JA. A single-staged technique for seeding vascular grafts with autogenous endothelium. Surgery 1978;84:498-504. 2. Ramalanjaona GR, Kempczinski RF, Ogle JD, Silberstein EB. Fibronectin coating of an experimental PTFE vascular prosthesis. J Surg Res 1986;41:479-83. 3. Seeger JM, Klingman N. Improved endothelial cell seeding with cultured cells and fibronectin coated grafts. J 5urg Res 1985;38:641-7. 4. Budd JS, Bell PRF, James RFL. Attachment of indium-ill labelled endothelial cells to pretreated polytetrafluoroethylene vascadar grafts. Br J Surg 1989;76:1259-61. 5. Lindblad B, Wright SW, Sell RL, Burkel WE, Graham LM, Stanley JC. Alternative techniques of seeding cultured endothelial cells to ePTFE grafts of different diameters, porosities, and surfaces. J Biomed Mater Res 1987;21:1013-22. 6. Lundgren CH, Herring MB, Arnold MP, Glover JL, Bendick PJ. Fluid shear disruption of cultured endothelium: the effect of cell species, fibronectin cross-linking and suppolxiug polymer. Trans Am Soc Artif Interu Organs 1986;32:334-8. 7. Ramalanjaona GR, Kempczinski RF, Rosenman JE, Douvllle E, Silberstein EB. The effect of fibronectin coating on endothelial cell kinetics in polytetrafluoroethylene grafts, l Vase Surg 1986;3:264-272. 8. Foxall TL, Auger KR, Callow AD, Libby P. Adult human

130

9.

10. 11. 12.

Journal of VASCULAR SURGF-RY

Budd et al.

endothelial cell coverage of small diameter Dacron and polytetraBuoroethylene in vitro. J Surg Res 1986;41:158-72. Steuman S, Vaheri A. Distribution of a major connective tissue protein, fibronectin, in normal human tissues. J Exp Med 1978;147:1054-64. Stenman S, von Smitten K, Vaheri A. Fibronectin and atherosclerosis. Acta Med Scand (Suppl). 1980;642:165-79. Grinnel F. Cellular adhesiveness and extracellular substrata. Int Rev Cytol 1978;53:65-144. Hasson JE, Wiebe DH, Sharetkin JB, D'Amore PA, Abbott WM. Use of tritiated thymidine as a marker to compare the

effects of matrix proteins on adult human vascular endothelial cell attachment: implications for seeding of vascular prostheses. Surgery 1986;100:884-91. 13. Thomson GJL, Vohra R, Walker MG. Cell seeding for small diameter ePTFE vascular grafts: a comparison between adult human endothelial and mesothelial cells. Ann Vase Surg 1989;3:140-5. 14. Ihlenfeld IV, Mathis TR, Barber TA, et al. Transient in vivo thrombus deposition onto polymeric biomaterials: role of plasma fibronectin. Trans Am Soc Artif Intern Organs 1978;24:727-35.

B O U N D VOLUMES AVAILABLE TO SUBSCRIBERS Bound volumes of the JOURNAL OF VASCULAR SURGERY for 1990 are available to subscribers only. They may be purchased from the publisher at a cost of $54.00 ($68.00 international) for Vol. 11 (January to June) and Vol. 12 (July to December). Price includes shipping charges. Each bound volume contains a subject and author index, and all advertising is removed. Copies are shipped within 60 days after publication o f the last issue in the volume. The binding is durable buckram with the journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Circulation Fulfillment, M o s b y - Y e a r Book, Inc., 11830 Westline Industrial Drive, St. Louis, M O 63146-3318, USA. In the United States call toll free: (800)325-4177, ext. 7351. In Missouri call collect: (314)872-8370, ext. 7351. Subscriptions must be in force to qualify. Bound volumes are not available in place o f a regular JOURNAL subscription.

The effect of varying fibronectin concentration on the attachment of endothelial cells to polytetrafluoroethylene vascular grafts.

Endothelial cell seeding onto untreated polytetrafluoroethylene vascular prostheses is inefficient. In an effort to improve cell attachment, numerous ...
751KB Sizes 0 Downloads 0 Views