J Neurosurg 77:397-402, 1992
An artificial blood vessel with an endothelial-cell monolayer HIDENORI KOBAYASHI, M.D., MASANORI KABUTO, M.D., HISASHI IDE, M.D.,
KAZUO HOSOTANI, M.D., AND TOSHIHIKO KUBOTA, M.D. Department of Neurosurgery, Fukui Medical School, Matsuoka, Fukui, Japan An artificial blood vessel with an endothelial-cell monolayer was used as an arterial substitute in rats. Endothelial cells were isolated from the aorta of a Wistar rat by the digestion method. The cell identification was established by the cobblestone appearance of a confluent cell monolayer, by an expression of factor VIIIrelated antigen, and by the presence of Weibel-Palade bodies. The luminal surface of the thin-walled polytetrafluoroethylene (PTFE) graft (4 mm in diameter and 10 mm in length) was coated with an endothelialcell monolayer for 7 days in vitro. An interpositional graft was placed using the endothelial cell-coated PTFE prosthesis on the right common carotid artery in seven rats. A total of 10 rats received an interpositional graft with the noncoated PTFE prosthesis as a control. The patency rate at 1 month after implantation was significantly higher in the coated group than in the control group. The vascular prosthesis with an endothelialcell monolayer is a promising technique to inhibit the development of thrombosis. KEY WORDS cell culture 9 endothelial-cell monolayer polytetrafluoroethylene 9 rat ~
ORMAL intact endothelium is the most nonthrombogenic tissue interface known. In neurosurgical bypass operations, relatively small vessels have been used, J1.13,~5but small-caliber artificial vascular grafts have rarely been employed because of their thrombogenicity. Endothelial-cell desquamation or subendothelial tissue exposure leads to rapid activation of platelets and subsequent thrombosis? 4 The absence of an endothelial-cell monolayer coveting the graft's luminal surface is one factor contributing to the relative thrombogenicity of small-diameter prostheses. 2 An artificial vascular prosthesis coated with endothelial cells has been accepted as one of the most promising approaches for ensuring antithrombogenicity. 3's'~9'26 Most of the studies have been based on endothelial-ceU seeding on artificial vascular grafts, with or without a coating layer of adhesive proteins such as collagen and
N
fibronectin
4,8,20,2 i
Expanded polytetrafluoroethylene (PTFE) grafts are currently the most widely used synthetic alternative to autogenous vessels in systemic vascular surgery. Recently, a thin-walled (0.39-ram) PTFE graft has been introduced that is alleged to have the same bursting strength, suture retention ability, and porosity (17 to 30 u) as the standard PTFE graft. :8'3~ Canine grafts seeded with endothelial cells have shown improved small-arJ. Neurosurg. / Volume 77/September, 1992
9 vascular graft
9
tery patencyfl 7 Shindo, et aL, z6 reported that seeded grafts 4 m m in diameter demonstrated an 86% patency rate at 1 month after implantation, in contrast to the significantly lower 14% patency rate of the unseeded control grafts. This article describes endothelial-cell isolation from the rat aorta, endothelial-cell coating onto the PTFE graft in vitro, and an in vivo study. We compared the two grafts with or without endothelial-cell coating for thrombus formation in vivo, following microsurgical interpositional grafting of the common carotid artery (CCA) in rats. Materials and Methods
Preparation of Endothelial Cells A 9-week-old male Wistar rat was anesthetized with sodium pentobarbital (40 mg/kg) injected intraperitoneally, and the aorta was removed using a sterile surgical technique. Endothelial cells were obtained by digestion of the aortic interior with 0.2% crude collagenase in phosphate-buffered saline (PBS) for 15 minutes at 37~ The collagenase solution was drained from the aorta, which was then rinsed with calcium- and magnesium-free (CMF) Hanks' solution. The cells in these solutions were then sedimented by centrifugation at 397
H. Kobayashi, et al. 2000 rpm for 5 minutes, washed once with Dulbecco's modified Eagle's medium (DMEM) containing 20% fetal bovine serum (FBS) and antibiotics, seeded in tissue culture dishes coated with type I collagen in DMEM containing FBS, and incubated at 37~ in a 5% CO2 atmosphere. The cells were maintained in a routine manner and reached confluency in 1 week as determined by visual inspection under an inverted microscope.* The cell monolayer was dissociated by exposure to a solution of 0.25% trypsin and 0.02% ethylenediamine tetra-acetic acid (EDTA) in CMF Hanks' solution, pH 7.4, for 2 to 3 minutes at 37"C with gentle pipetting. After the cells were lifted from the surface, they were transferred to a tube containing approximately 5 ml DMEM with FBS, collected by centrifugation, and washed once with the medium. The cell pellet was then resuspended in the medium, split approximately at a 1:4 ratio, plated in the tissue culture dishes, and incubated as described above. Fifteen-passage endothelial cells were used in this study. Cell identification was established by the cobblestone appearance of confluent-cell monolayers, by expression of factor VIII-related antigen, and by the presence of Weibel-Palade bodies? ~
Artificial Blood Vessels A thin-walled PTFE graft~- with an inner diameter of 4 mm was cut in sections 10 mm in length, and small pieces of lead were attached to both ends as sinkers. Type I collagen~ (0.03% acid soluble, pH 3.0) was diluted with acetic acid (pH 3.0) at a ratio of 1:10 to form a coating solution. This collagen solution (0.15 to 0.2 ml) was applied to the luminal surface of the sterilized grafts placed in 35-mm culture dishes, and the grafts were rotated so that the entire luminal surface was covered completely. The graft coated with type I collagen was dried overnight in a laminar flow hood. Coating With Endothelial Cells The endothelial-cell suspension, collected as previously described, was gently poured onto the luminal surface of the graft in the dish with a pipette. The suspension shed from the lumen onto the dish. The graft was immersed in the culture medium. The cells were incubated at 37"C in a 5% CO2 atmosphere. At 24 hours after seeding, when the culture medium was changed, the graft was rotated 180" and additional endothelial-cell suspension was poured onto the luminal surface of the graft. These cells were seeded at a density of 1 x 106 cells/ml. The culture medium was changed three times, on the 1st, 3rd, and 5th day after
*Inverted microscope, Model CK2, manufactured by Olympus, Tokyo, Japan. t Polytetrafluoroethylene graft was a gift of Japan GoreTex, Tokyo, Japan. Rat type I collagen provided by Sigma Chemical Company, St. Louis, Missouri. 398
seeding. Whenever the culture medium was changed, the graft was rotated 180* again and immersed in the medium. The cells on the culture dish reached confluency in 7 days as determined by visual inspection under an inverted microscope. It was assumed that the cells on the graft would reach confluency when the cells on the culture dish reached confluency. In Vivo Study A total of 17 Wistar rats, weighing 400 to 450 gm each, were used in this study. The rats were divided into an endothelial-ceU coated group (seven rats) and a control group (10 rats). The animals were anesthetized with sodium pentobarbital (40 mg/kg), injected intraperitoneally, and an anterior skin incision was made from shoulder to shoulder. The right CCA was exposed and temporarily occluded with two polyethylene vascular occluders. To relieve spasm, 0.25% bupivacaine was applied to the vessel. A 5-mm segment of the CCA was removed and a 10-mm PTFE graft was interposed using an operating microscope. Two end-to-end anastomoses were performed with 10-0 monofllament nylon suture. The diameter of the PTFE graft was considerably greater than that of the arterial wall. The size discrepancies between the ends were compensated for by taking larger bites set further apart in the wall of the PTFE graft. The vascular occluders were then removed. In the coated group, the internal surface of the PTFE graft was coated with type I collagen and then with an endothelial-cell monolayer. The control group graft was coated with only type I collagen. The animals did not receive any anticoagulant or antithrombotic agents. All animals were sacrificed at 1 month after implantation. The cardiovascular system was perfused with 40 ml of saline followed by 30 ml of 2.5% glutaraldehyde buffered with 0.1 M sodium cacodylate. The perfusion pressure was maintained at 100 cm H20. The PTFE graft was removed for histological examination.
Microscopic Examination The endothelial cells were grown to confluence on a tissue culture chamber/slidew coated with type I collagen. The cell monolayers were washed with PBS, then fixed in 4% paraformaldehyde for 30 minutes at room temperature, followed by treatment with 1% Triton X100 for 5 minutes. The monolayers were treated with rabbit anti-human factor VIII-related antigen antiserum in the peroxidase-antiperoxidase (PAP) technique. II The PTFE grafts were fixed at 4"C in 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer, pH 7.2, for 2 hours. Postfixation was performed for 2 hours in 1% OsO4 in buffer at 4~ and the grafts were then divided into two pieces. Samples for transmission electron mi-
wLat Tek tissue culture chamber and slides manufactured by Miles Scientific Laboratories, Inc., Naperville, Illinois. II Rabbit factor VIII-related antigen antiserum and PAP treatment kit manufactured by DAKO Corporation, Carpinteria, California.
,L Neurosurg. / Volume 77 / September, 1992
Vascular prosthesis with an endothelial-cell monolayer
Fie, 2. Scanning electron micrograph of the luminal surface of the graft coated with type I collagen 7 days after endothelial seeding. A continuous endothelial-cell monolayer is demonstrated, x 515.
FIG. 1. Photomicrograph of 7-day-old cultured endothelial cells grown to confluence. The cells were treated with rabbit anti-human factor VIII-related antigen antiserum using the peroxidase-antiperoxidasetechnique. • 140.
croscopy were rinsed with PBS, dehydrated with graded ethanols through three changes in 100% ethanol, and embedded in Epon 812. Semithin sections were cut with an ultramicrotome and stained with toluidine blue. Ultrathin sections of selected areas were stained with uranyl acetate and lead citrate, and examined with an electron microscope.* For scanning electron microscopy, samples were dehydrated in graded ethanols and amyl acetate after osmium fixation, and were then critical-point dried with C02, cut longitudinally, and coated with gold in an evaporator.t
Statistical Analysis The data concerning patency of the graft versus occlusion were analyzed by the chi-squared test between the control group and the coated group. P values of 0.05 or less were considered statistically significant.
Results
Morphological Changes Or the In Vitro Study Endothelial cells on culture plates were visually identified by their typical cobblestone morphology. Contamination with different cell types was not detected. * Electron microscope, Model H7000, and scanning electron microscope, Model S-450, manufactured by Hitachi, Ltd., Tokyo, Japan. Evaporator, Model 1B-3, manufactured by EIKO, Mito, Japan.
J. Neurosurg. / Volume 77/September, 1992
Figure 1 shows endothelial cells grown to confluence on the tissue culture chamber/slide, which was treated with rabbit anti-human factor VIII-related antigen anti-serum by the PAP technique. Positive staining is seen in the cytoplasm of the endothelial cells. After 1 week, the luminal surfaces of the PTFE prostheses coated with type I collagen were completely covered with endothelial cells. Figure 2 shows a scanning electron micrograph of the luminal surface of a PTFE graft coated with type I collagen at 7 days after endothelial seeding. A continuous endothelial-cell monolayer covered the collagen-coated prosthesis. Light microscopy of the semithin section stained with toluidine blue demonstrated a continuous endothelial-cell monolayer overlying the prosthesis. Figure 3 shows a transmission electron micrograph of the seeded graft which was coated with type I collagen at 7 days after seeding. The luminal surface of the seeded graft revealed ultrastructural characteristics typical of endothelial cells, including a dense nucleus, adhesion-like junctional regions, and extremely attenuated cytoplasmic extensions. Weibel-Palade bodies were also observed in the cytoplasm of the endothelial cells.
Patency Rates in the In Vivo Study Table 1 shows the patency rates at 1 month after implantation, which were 20% in the control group and 86% in the endothelial-cell coated group. The coated group exhibited significantly greater patency than did the control.
Morphological Changes in the In Vivo Study One of seven grafts in the coated group and eight of 10 in the control group were found to have thrombosed and organized. The luminal surfaces of the patent PTFE grafts in both groups were completely covered with endothelial cells. Nylon sutures were also covered with 399
H. K o b a y a s h i , e t a l .
FIG. 3. Transmission electron micrographs of the endothelial cells on the polytetrafluoroethylenegraft (G) coated with type I collagen. An adhesion-likejunction between two cells (left, arrow) and attenuated cytoplasmic extension (right) are observed. L = lumen; x 5400 (left), x 16,300 (right).
TABLE 1 Patency rates in control group and group coated with an endothelial.cell monolayer Group
Total Grafts
PatentGrafts No. Percent
control 10 2 20 coated 7 6 86* *Statisticalsignificance:p < 0.05 versuscontrol(chi-squaredtest). regenerated endothelial cells. Figure 4 shows a scanning electron microscope specimen of the anastomotic site in the coated group. A discrepancy in size is observed between arterial and graft diameters. The luminal surface is completely covered with endothelial cells, which are regularly aligned to the blood stream. Discussion
In Vitro Study Isolation and Identification of Endothelial Cells. The rat aorta was peffused with collagenase instead of trypsin in order to minimize both cell damage j6 and the possibility of digesting the internal elastic lamina of the aorta. The cell identification was established by the cobblestone appearance of confluent cell monolayers, by expression of factor VIII-related antigen, and by the presence of Weibel-Palade bodies in this study? ~ Factor VIII-related antigen can be found in endothelial cells, platelets, and megakaryocytes,9 and is widely considered to be the most reliable marker for endothelial cells. It has been reported that WeibelPalade bodies are difficult to find in bovine, murine, or rat endothelial cells.32 They are not practical markers in some species. It has been shown that angiotensinconverting enzyme, prostacyclin, and endothelin are present in the endothelial cells.7"~:'3~These are also used as endothelial markers. 400
Endothelial-Cell Coating. It has been reported that when endothelial cells are cultured on basement membrane-associated collagen (types IV and V)-coated Petri dishes, the cells organize into tube-like structures} 8 In contrast, when gown on interstitial collagen (types I and III)-coated dishes, the cells proliferate and form a confluent monolayer.'7 The luminal surface of the type I collagen-coated PTFE prosthesis was completely covered with an endothelial-cell monolayer after 7 days in our study. Further investigation will be needed to determine the proper cell density and better culture conditions. In Vivo Study The patency rates at 1 month after implantation were 20% in the control prostheses and 86% in the endothelial cell-coated prostheses. It has been reported that the patency of the small-caliber arterial prostheses may be significantly enhanced with antiplatelet treatment and by lining the luminal surface with autogenous endothelium. 4'5'2~'27 There is considerable evidence that such endothelialization reduces platelet accumulation,TM provides active prostacyclin secretion,24and minimizes the risk of bacterial infection? It has been shown that, with the larger-caliber arterial prostheses used in vascular surgery, the use of autogenous endothelial cells to preclot vascular prostheses just before operation facilitates the development of a complete endothelial lining on the luminal surface of the graft, s Dacron and PTFE vascular graft materials clearly activate the complement system with release of potent and potentially damaging inflammatory mediators} 5 These mediators are capable of stimulating both local platelet aggregation and in situ production of tissue thromboplastin by polymorphonuclear leukocytes.25 The vascular prosthesis coated with an endothelialcell monolayer is promising for antithrombogenicity, especially in small-caliber prosthesis. Ramalanjaona, et J. Neurosurg. / Volume 77/September, 1992
Vascular prosthesis with an endothelial-cell monolayer References
FIG. 4. Scanning electron micrograph of the luminal surface of the prosthesis coated with an endothelial-cell monolayer at 1 month after implantation. The surface was completely covered with endothelial cells. The common carotid artery is at left, the prosthesis at fight. • 730.
al., 2~ have demonstrated that pretreatment of PTFE grafts with an appropriate concentration of fibroneetin significantly increased the initial adherence of harvested endothelial cells, and that there was also a significant reduction in the subsequent loss of these cells from the graft during the first 24 hours after restoration of the flow. In our control group, the patency rate was low for the following reasons: I) a small-caliber (4-ram) prosthesis was coated with type I collagen, which led to rapid activation of platelets and would subsequently cause thrombus formation; 6-14 and 2) as the size discrepancy between the arterial end and the prosthesis end was large, the turbulence at anastomotic sites was great, and the risk of thrombosis was enhanced. Rosenman, eta]., 22 reported that, within 30 minutes after restoration of flow, there is rapid loss of the cells that were harvested from the autogenous vein and seeded onto the prosthesis just before surgery and that, by 24 hours, fewer than 4% of all the cells that were originally adherent remain on the prosthesis. We cannot say how many endothelial cells remained on the prosthesis after restoration of flow in our coated group. Further study will be needed to clarify this point. J. Neurosurg. / Volume 77/September, 1992
1~ Arregui M, Dilley R, Herring M, et al: The effect of bacteremia on arterial prostheses seeded with endothelium. J Am Soc Artif Intern Organs 8:118-121, 1985 2. Berger K, Sauvage LR, Rao AM, et at: Healing of arterial prostheses in man: its incompleteness. Ann Surg 175: 118-127, 1972 3. Gerlach J, Westmeyer K, Sehauwecker HH, et at: Isolation of vascular endothelial cells for investigations on hybrid prostheses. Int J Artif Organs 12:805-810, 1989 4. Graham LM, Burkd WE, Ford JW, el al: Expanded polytetrafluoroethylene vascular prostheses seeded with enzymatically derived and cultured canine endothelial cells. Surgery 91:550-559, 1982 5. Graham LM, Stanley JC, Burkel WE: Improved patency of endothelial-cell-seeded, long, knitted Dacron and PTFE vascular prostheses. J Am Soc Artif Intern Organs 8:65-73, 1985 6. Hattori A, Watanabe T, Izumi T: Scanning electron microscope study on hemostatic reaction. Mural thrombus after the removal of endothelium, with special references to platelet behavior, site of fibrin formation and microhemolysis. Arch Histol Jpn 41:205-227, 1978 7. Hayes LW, Goguen CA, Citing SF, et al: Angiotensinconverting enzyme: accumulation in medium from cultured endothelial cells. Biochem Biophys Res Commun 82:1147-1153, 1978 8. Herring M, Gardner A, GIover J: A single-staged technique for seeding vascular grafts with autogenous endothelium. Surgery 84:498-504, 1978 9. Jaffe EA: Endothelial cells and the biology of factor VIII. N Engl J Med 296:377-383, 1977 10. Jaffe EA: Synthesis of factor VIII antigen by cultured human endothelial cells. Ann NY Acad Sci 240:62-69, 1975 11. Kobayashi H, Hayashi M, Handa Y, et al: EC-IC bypass for adult patients with moyamoya disease. Neurol Res 13: ll3-116, 1991 12. Kobayashi H, Hayashi M, Kobayashi S, et al: Cerebral vasospasm and vasoconstriction caused by endothelin. Neurosurgery 28:673-679, 1991 13. Kobayashi H, Hayashi M, Kawano H, et al: Cerebral blood flow studies using N-isopropyl I- 123 p-iodoamphetamine. Stroke 16:293-296, 1985 14. Kobayashi H, Hayashi M, Kawano H, et al: Effect of flunarizine on artefial thrombosis after endarterectomy in rats. Neurol Med Chir 29:563-567, 1989 15. Kobayashi H, Hayashi M, Kawano H, et al: Evaluation of extracranial-to-intracranial bypass surgery using iodine 123 iodoamphetamine single-photon emission computed tomography. Surg Neuro135:436-440, 1991 16. Lasfarques EY: Collagenase as a cell dispersion agent in tissue culture, in Mandl I (ed): Interdisciplinary Symposium on Collagenase. New York: Gordon & Breach, 1967, pp 83-89 17. Madfi JA, Pratt BM: Endothelial cell-matrix interactions: in vitro models of anglogenesis. J Histochem Cytochem 34:85-91, 1986 18. Madri JA, Williams SK: Capillary endothelial cell cultures: phenotypic modulation by matrix components. J Cell Biol 97:153-165, 1983 19. Matsuda T, Akutsu T, Kira K, et al: Development of hybrid compliant graft: rapid preparative method for reconstruction of a vascular wall. ASAIO Trans 35: 553-555, 1989 20. Pearce WH, Rutherford RB, Whitehill TA, et al: Successful endothelial seeding with omentally derived microvascular endothelial cells. J Vase Surg 5:203-206, 1987 401
H. Kobayashi, et al. 21. Ramalanjaona G, Kempczinski RF, Rosenman JE, et al: The effect of fibronectin coating on endothelial cell kinetics in polytetrafluoroethylene grafts. J Vase Surg 3: 264-272, 1986 22. Rosenman JE, Kempczinski RF, Pearce WH, et al: Kinetics of endothelial cell seeding. J Vasc Surg 2: 778-784, 1985 23. Schmidt SP, Hunter TJ, Falkow LJ, et ah Effects of antiplatelet agents on PGI2 production by endothelialcell-seeded small diameter Dacron grafts. J Am Soc Artif Intern Organs 8:99-103, 1985 24. Sharp WV, Schmidt SP, Donovan DL: Prostaglandin biochemistry of seeded endothelial cells on Dacron prostheses. J Vasc Surg 3:256-263, 1986 25. Shepard AD, Gelfand JA, Callow AD, et ah Complement activation by synthetic vascular prostheses. J Vasc Surg 1:829-838, 1984 26. Shindo S, Takagi A, Whittemore AD: Improved patency of collagen-impregnated grafts after in vitro autogenous endothelial cell seeding, a Vase Surg 6:325-332, 1987 27. Stanley JC, Burkel WE, Ford JW, et al: Enhanced patency of small-diameter, externally supported Dacron iliofemoral grafts seeded with endothelial cells. Surgery 92: 994-1005, 1982
402
28. Takenaka K, Kholoussy AM, Yang Y, et al: The healing of the thin-walled expanded polytetrafluoroethylene vascular graft. Vase Surg 19:383-389, 1985 29. Weibel ER, Palade GE: New cytoplasmic components in arterial endothelia. J Cell Biol 23:101-112, 1964 30. Weksler BB, Marcus AJ, Jaffe EA: Synthesis of prostaglandin Iz (prostacyclin) by cultured human and bovine endothelial cells. Proc Natl Aead Sci USA 74: 3922-3926, 1977 31. Willams GT, Underwood CJ, Charlesworth D: Early resuits of femoropopliteal bypass using a five millimeter "thin wall" polytetraflnoroethylene (Gore-Tex) prosthesis. Ann Vase Sarg 1:208-213, 1986 32. Zetter BR: Culture of capillary endothelial cells, in Jaffe EA (ed): Biology of Endothelial Cells. Boston: Martinus Nijhoff, 1984, pp 14-26
Manuscript received January 24, 1991. Accepted in final form March 4, 1992. Address reprint requests to: Hidenori Kobayashi, M.D., Department of Neurosurgery, Fukui Medical School, Matsuoka, Fukui 910-11, Japan.
J. Neurosurg. / Volume 77/September, 1992
An artificial blood vessel with an endothelial-cell monolayer HIDENORI KOBAYASHI, M.D., MASANORI KABUTO, M.D., HISASHI IDE, M.D.,
KAZUO HOSOTANI, M.D., AND TOSHIHIKO KUBOTA, M.D. Department of Neurosurgery, Fukui Medical School, Matsuoka, Fukui, Japan An artificial blood vessel with an endothelial-cell monolayer was used as an arterial substitute in rats. Endothelial cells were isolated from the aorta of a Wistar rat by the digestion method. The cell identification was established by the cobblestone appearance of a confluent cell monolayer, by an expression of factor VIIIrelated antigen, and by the presence of Weibel-Palade bodies. The luminal surface of the thin-walled polytetrafluoroethylene (PTFE) graft (4 mm in diameter and 10 mm in length) was coated with an endothelialcell monolayer for 7 days in vitro. An interpositional graft was placed using the endothelial cell-coated PTFE prosthesis on the right common carotid artery in seven rats. A total of 10 rats received an interpositional graft with the noncoated PTFE prosthesis as a control. The patency rate at 1 month after implantation was significantly higher in the coated group than in the control group. The vascular prosthesis with an endothelialcell monolayer is a promising technique to inhibit the development of thrombosis. KEY WORDS cell culture 9 endothelial-cell monolayer polytetrafluoroethylene 9 rat ~
ORMAL intact endothelium is the most nonthrombogenic tissue interface known. In neurosurgical bypass operations, relatively small vessels have been used, J1.13,~5but small-caliber artificial vascular grafts have rarely been employed because of their thrombogenicity. Endothelial-cell desquamation or subendothelial tissue exposure leads to rapid activation of platelets and subsequent thrombosis? 4 The absence of an endothelial-cell monolayer coveting the graft's luminal surface is one factor contributing to the relative thrombogenicity of small-diameter prostheses. 2 An artificial vascular prosthesis coated with endothelial cells has been accepted as one of the most promising approaches for ensuring antithrombogenicity. 3's'~9'26 Most of the studies have been based on endothelial-ceU seeding on artificial vascular grafts, with or without a coating layer of adhesive proteins such as collagen and
N
fibronectin
4,8,20,2 i
Expanded polytetrafluoroethylene (PTFE) grafts are currently the most widely used synthetic alternative to autogenous vessels in systemic vascular surgery. Recently, a thin-walled (0.39-ram) PTFE graft has been introduced that is alleged to have the same bursting strength, suture retention ability, and porosity (17 to 30 u) as the standard PTFE graft. :8'3~ Canine grafts seeded with endothelial cells have shown improved small-arJ. Neurosurg. / Volume 77/September, 1992
9 vascular graft
9
tery patencyfl 7 Shindo, et aL, z6 reported that seeded grafts 4 m m in diameter demonstrated an 86% patency rate at 1 month after implantation, in contrast to the significantly lower 14% patency rate of the unseeded control grafts. This article describes endothelial-cell isolation from the rat aorta, endothelial-cell coating onto the PTFE graft in vitro, and an in vivo study. We compared the two grafts with or without endothelial-cell coating for thrombus formation in vivo, following microsurgical interpositional grafting of the common carotid artery (CCA) in rats. Materials and Methods
Preparation of Endothelial Cells A 9-week-old male Wistar rat was anesthetized with sodium pentobarbital (40 mg/kg) injected intraperitoneally, and the aorta was removed using a sterile surgical technique. Endothelial cells were obtained by digestion of the aortic interior with 0.2% crude collagenase in phosphate-buffered saline (PBS) for 15 minutes at 37~ The collagenase solution was drained from the aorta, which was then rinsed with calcium- and magnesium-free (CMF) Hanks' solution. The cells in these solutions were then sedimented by centrifugation at 397
H. Kobayashi, et al. 2000 rpm for 5 minutes, washed once with Dulbecco's modified Eagle's medium (DMEM) containing 20% fetal bovine serum (FBS) and antibiotics, seeded in tissue culture dishes coated with type I collagen in DMEM containing FBS, and incubated at 37~ in a 5% CO2 atmosphere. The cells were maintained in a routine manner and reached confluency in 1 week as determined by visual inspection under an inverted microscope.* The cell monolayer was dissociated by exposure to a solution of 0.25% trypsin and 0.02% ethylenediamine tetra-acetic acid (EDTA) in CMF Hanks' solution, pH 7.4, for 2 to 3 minutes at 37"C with gentle pipetting. After the cells were lifted from the surface, they were transferred to a tube containing approximately 5 ml DMEM with FBS, collected by centrifugation, and washed once with the medium. The cell pellet was then resuspended in the medium, split approximately at a 1:4 ratio, plated in the tissue culture dishes, and incubated as described above. Fifteen-passage endothelial cells were used in this study. Cell identification was established by the cobblestone appearance of confluent-cell monolayers, by expression of factor VIII-related antigen, and by the presence of Weibel-Palade bodies? ~
Artificial Blood Vessels A thin-walled PTFE graft~- with an inner diameter of 4 mm was cut in sections 10 mm in length, and small pieces of lead were attached to both ends as sinkers. Type I collagen~ (0.03% acid soluble, pH 3.0) was diluted with acetic acid (pH 3.0) at a ratio of 1:10 to form a coating solution. This collagen solution (0.15 to 0.2 ml) was applied to the luminal surface of the sterilized grafts placed in 35-mm culture dishes, and the grafts were rotated so that the entire luminal surface was covered completely. The graft coated with type I collagen was dried overnight in a laminar flow hood. Coating With Endothelial Cells The endothelial-cell suspension, collected as previously described, was gently poured onto the luminal surface of the graft in the dish with a pipette. The suspension shed from the lumen onto the dish. The graft was immersed in the culture medium. The cells were incubated at 37"C in a 5% CO2 atmosphere. At 24 hours after seeding, when the culture medium was changed, the graft was rotated 180" and additional endothelial-cell suspension was poured onto the luminal surface of the graft. These cells were seeded at a density of 1 x 106 cells/ml. The culture medium was changed three times, on the 1st, 3rd, and 5th day after
*Inverted microscope, Model CK2, manufactured by Olympus, Tokyo, Japan. t Polytetrafluoroethylene graft was a gift of Japan GoreTex, Tokyo, Japan. Rat type I collagen provided by Sigma Chemical Company, St. Louis, Missouri. 398
seeding. Whenever the culture medium was changed, the graft was rotated 180* again and immersed in the medium. The cells on the culture dish reached confluency in 7 days as determined by visual inspection under an inverted microscope. It was assumed that the cells on the graft would reach confluency when the cells on the culture dish reached confluency. In Vivo Study A total of 17 Wistar rats, weighing 400 to 450 gm each, were used in this study. The rats were divided into an endothelial-ceU coated group (seven rats) and a control group (10 rats). The animals were anesthetized with sodium pentobarbital (40 mg/kg), injected intraperitoneally, and an anterior skin incision was made from shoulder to shoulder. The right CCA was exposed and temporarily occluded with two polyethylene vascular occluders. To relieve spasm, 0.25% bupivacaine was applied to the vessel. A 5-mm segment of the CCA was removed and a 10-mm PTFE graft was interposed using an operating microscope. Two end-to-end anastomoses were performed with 10-0 monofllament nylon suture. The diameter of the PTFE graft was considerably greater than that of the arterial wall. The size discrepancies between the ends were compensated for by taking larger bites set further apart in the wall of the PTFE graft. The vascular occluders were then removed. In the coated group, the internal surface of the PTFE graft was coated with type I collagen and then with an endothelial-cell monolayer. The control group graft was coated with only type I collagen. The animals did not receive any anticoagulant or antithrombotic agents. All animals were sacrificed at 1 month after implantation. The cardiovascular system was perfused with 40 ml of saline followed by 30 ml of 2.5% glutaraldehyde buffered with 0.1 M sodium cacodylate. The perfusion pressure was maintained at 100 cm H20. The PTFE graft was removed for histological examination.
Microscopic Examination The endothelial cells were grown to confluence on a tissue culture chamber/slidew coated with type I collagen. The cell monolayers were washed with PBS, then fixed in 4% paraformaldehyde for 30 minutes at room temperature, followed by treatment with 1% Triton X100 for 5 minutes. The monolayers were treated with rabbit anti-human factor VIII-related antigen antiserum in the peroxidase-antiperoxidase (PAP) technique. II The PTFE grafts were fixed at 4"C in 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer, pH 7.2, for 2 hours. Postfixation was performed for 2 hours in 1% OsO4 in buffer at 4~ and the grafts were then divided into two pieces. Samples for transmission electron mi-
wLat Tek tissue culture chamber and slides manufactured by Miles Scientific Laboratories, Inc., Naperville, Illinois. II Rabbit factor VIII-related antigen antiserum and PAP treatment kit manufactured by DAKO Corporation, Carpinteria, California.
,L Neurosurg. / Volume 77 / September, 1992
Vascular prosthesis with an endothelial-cell monolayer
Fie, 2. Scanning electron micrograph of the luminal surface of the graft coated with type I collagen 7 days after endothelial seeding. A continuous endothelial-cell monolayer is demonstrated, x 515.
FIG. 1. Photomicrograph of 7-day-old cultured endothelial cells grown to confluence. The cells were treated with rabbit anti-human factor VIII-related antigen antiserum using the peroxidase-antiperoxidasetechnique. • 140.
croscopy were rinsed with PBS, dehydrated with graded ethanols through three changes in 100% ethanol, and embedded in Epon 812. Semithin sections were cut with an ultramicrotome and stained with toluidine blue. Ultrathin sections of selected areas were stained with uranyl acetate and lead citrate, and examined with an electron microscope.* For scanning electron microscopy, samples were dehydrated in graded ethanols and amyl acetate after osmium fixation, and were then critical-point dried with C02, cut longitudinally, and coated with gold in an evaporator.t
Statistical Analysis The data concerning patency of the graft versus occlusion were analyzed by the chi-squared test between the control group and the coated group. P values of 0.05 or less were considered statistically significant.
Results
Morphological Changes Or the In Vitro Study Endothelial cells on culture plates were visually identified by their typical cobblestone morphology. Contamination with different cell types was not detected. * Electron microscope, Model H7000, and scanning electron microscope, Model S-450, manufactured by Hitachi, Ltd., Tokyo, Japan. Evaporator, Model 1B-3, manufactured by EIKO, Mito, Japan.
J. Neurosurg. / Volume 77/September, 1992
Figure 1 shows endothelial cells grown to confluence on the tissue culture chamber/slide, which was treated with rabbit anti-human factor VIII-related antigen anti-serum by the PAP technique. Positive staining is seen in the cytoplasm of the endothelial cells. After 1 week, the luminal surfaces of the PTFE prostheses coated with type I collagen were completely covered with endothelial cells. Figure 2 shows a scanning electron micrograph of the luminal surface of a PTFE graft coated with type I collagen at 7 days after endothelial seeding. A continuous endothelial-cell monolayer covered the collagen-coated prosthesis. Light microscopy of the semithin section stained with toluidine blue demonstrated a continuous endothelial-cell monolayer overlying the prosthesis. Figure 3 shows a transmission electron micrograph of the seeded graft which was coated with type I collagen at 7 days after seeding. The luminal surface of the seeded graft revealed ultrastructural characteristics typical of endothelial cells, including a dense nucleus, adhesion-like junctional regions, and extremely attenuated cytoplasmic extensions. Weibel-Palade bodies were also observed in the cytoplasm of the endothelial cells.
Patency Rates in the In Vivo Study Table 1 shows the patency rates at 1 month after implantation, which were 20% in the control group and 86% in the endothelial-cell coated group. The coated group exhibited significantly greater patency than did the control.
Morphological Changes in the In Vivo Study One of seven grafts in the coated group and eight of 10 in the control group were found to have thrombosed and organized. The luminal surfaces of the patent PTFE grafts in both groups were completely covered with endothelial cells. Nylon sutures were also covered with 399
H. K o b a y a s h i , e t a l .
FIG. 3. Transmission electron micrographs of the endothelial cells on the polytetrafluoroethylenegraft (G) coated with type I collagen. An adhesion-likejunction between two cells (left, arrow) and attenuated cytoplasmic extension (right) are observed. L = lumen; x 5400 (left), x 16,300 (right).
TABLE 1 Patency rates in control group and group coated with an endothelial.cell monolayer Group
Total Grafts
PatentGrafts No. Percent
control 10 2 20 coated 7 6 86* *Statisticalsignificance:p < 0.05 versuscontrol(chi-squaredtest). regenerated endothelial cells. Figure 4 shows a scanning electron microscope specimen of the anastomotic site in the coated group. A discrepancy in size is observed between arterial and graft diameters. The luminal surface is completely covered with endothelial cells, which are regularly aligned to the blood stream. Discussion
In Vitro Study Isolation and Identification of Endothelial Cells. The rat aorta was peffused with collagenase instead of trypsin in order to minimize both cell damage j6 and the possibility of digesting the internal elastic lamina of the aorta. The cell identification was established by the cobblestone appearance of confluent cell monolayers, by expression of factor VIII-related antigen, and by the presence of Weibel-Palade bodies in this study? ~ Factor VIII-related antigen can be found in endothelial cells, platelets, and megakaryocytes,9 and is widely considered to be the most reliable marker for endothelial cells. It has been reported that WeibelPalade bodies are difficult to find in bovine, murine, or rat endothelial cells.32 They are not practical markers in some species. It has been shown that angiotensinconverting enzyme, prostacyclin, and endothelin are present in the endothelial cells.7"~:'3~These are also used as endothelial markers. 400
Endothelial-Cell Coating. It has been reported that when endothelial cells are cultured on basement membrane-associated collagen (types IV and V)-coated Petri dishes, the cells organize into tube-like structures} 8 In contrast, when gown on interstitial collagen (types I and III)-coated dishes, the cells proliferate and form a confluent monolayer.'7 The luminal surface of the type I collagen-coated PTFE prosthesis was completely covered with an endothelial-cell monolayer after 7 days in our study. Further investigation will be needed to determine the proper cell density and better culture conditions. In Vivo Study The patency rates at 1 month after implantation were 20% in the control prostheses and 86% in the endothelial cell-coated prostheses. It has been reported that the patency of the small-caliber arterial prostheses may be significantly enhanced with antiplatelet treatment and by lining the luminal surface with autogenous endothelium. 4'5'2~'27 There is considerable evidence that such endothelialization reduces platelet accumulation,TM provides active prostacyclin secretion,24and minimizes the risk of bacterial infection? It has been shown that, with the larger-caliber arterial prostheses used in vascular surgery, the use of autogenous endothelial cells to preclot vascular prostheses just before operation facilitates the development of a complete endothelial lining on the luminal surface of the graft, s Dacron and PTFE vascular graft materials clearly activate the complement system with release of potent and potentially damaging inflammatory mediators} 5 These mediators are capable of stimulating both local platelet aggregation and in situ production of tissue thromboplastin by polymorphonuclear leukocytes.25 The vascular prosthesis coated with an endothelialcell monolayer is promising for antithrombogenicity, especially in small-caliber prosthesis. Ramalanjaona, et J. Neurosurg. / Volume 77/September, 1992
Vascular prosthesis with an endothelial-cell monolayer References
FIG. 4. Scanning electron micrograph of the luminal surface of the prosthesis coated with an endothelial-cell monolayer at 1 month after implantation. The surface was completely covered with endothelial cells. The common carotid artery is at left, the prosthesis at fight. • 730.
al., 2~ have demonstrated that pretreatment of PTFE grafts with an appropriate concentration of fibroneetin significantly increased the initial adherence of harvested endothelial cells, and that there was also a significant reduction in the subsequent loss of these cells from the graft during the first 24 hours after restoration of the flow. In our control group, the patency rate was low for the following reasons: I) a small-caliber (4-ram) prosthesis was coated with type I collagen, which led to rapid activation of platelets and would subsequently cause thrombus formation; 6-14 and 2) as the size discrepancy between the arterial end and the prosthesis end was large, the turbulence at anastomotic sites was great, and the risk of thrombosis was enhanced. Rosenman, eta]., 22 reported that, within 30 minutes after restoration of flow, there is rapid loss of the cells that were harvested from the autogenous vein and seeded onto the prosthesis just before surgery and that, by 24 hours, fewer than 4% of all the cells that were originally adherent remain on the prosthesis. We cannot say how many endothelial cells remained on the prosthesis after restoration of flow in our coated group. Further study will be needed to clarify this point. J. Neurosurg. / Volume 77/September, 1992
1~ Arregui M, Dilley R, Herring M, et al: The effect of bacteremia on arterial prostheses seeded with endothelium. J Am Soc Artif Intern Organs 8:118-121, 1985 2. Berger K, Sauvage LR, Rao AM, et at: Healing of arterial prostheses in man: its incompleteness. Ann Surg 175: 118-127, 1972 3. Gerlach J, Westmeyer K, Sehauwecker HH, et at: Isolation of vascular endothelial cells for investigations on hybrid prostheses. Int J Artif Organs 12:805-810, 1989 4. Graham LM, Burkd WE, Ford JW, el al: Expanded polytetrafluoroethylene vascular prostheses seeded with enzymatically derived and cultured canine endothelial cells. Surgery 91:550-559, 1982 5. Graham LM, Stanley JC, Burkel WE: Improved patency of endothelial-cell-seeded, long, knitted Dacron and PTFE vascular prostheses. J Am Soc Artif Intern Organs 8:65-73, 1985 6. Hattori A, Watanabe T, Izumi T: Scanning electron microscope study on hemostatic reaction. Mural thrombus after the removal of endothelium, with special references to platelet behavior, site of fibrin formation and microhemolysis. Arch Histol Jpn 41:205-227, 1978 7. Hayes LW, Goguen CA, Citing SF, et al: Angiotensinconverting enzyme: accumulation in medium from cultured endothelial cells. Biochem Biophys Res Commun 82:1147-1153, 1978 8. Herring M, Gardner A, GIover J: A single-staged technique for seeding vascular grafts with autogenous endothelium. Surgery 84:498-504, 1978 9. Jaffe EA: Endothelial cells and the biology of factor VIII. N Engl J Med 296:377-383, 1977 10. Jaffe EA: Synthesis of factor VIII antigen by cultured human endothelial cells. Ann NY Acad Sci 240:62-69, 1975 11. Kobayashi H, Hayashi M, Handa Y, et al: EC-IC bypass for adult patients with moyamoya disease. Neurol Res 13: ll3-116, 1991 12. Kobayashi H, Hayashi M, Kobayashi S, et al: Cerebral vasospasm and vasoconstriction caused by endothelin. Neurosurgery 28:673-679, 1991 13. Kobayashi H, Hayashi M, Kawano H, et al: Cerebral blood flow studies using N-isopropyl I- 123 p-iodoamphetamine. Stroke 16:293-296, 1985 14. Kobayashi H, Hayashi M, Kawano H, et al: Effect of flunarizine on artefial thrombosis after endarterectomy in rats. Neurol Med Chir 29:563-567, 1989 15. Kobayashi H, Hayashi M, Kawano H, et al: Evaluation of extracranial-to-intracranial bypass surgery using iodine 123 iodoamphetamine single-photon emission computed tomography. Surg Neuro135:436-440, 1991 16. Lasfarques EY: Collagenase as a cell dispersion agent in tissue culture, in Mandl I (ed): Interdisciplinary Symposium on Collagenase. New York: Gordon & Breach, 1967, pp 83-89 17. Madfi JA, Pratt BM: Endothelial cell-matrix interactions: in vitro models of anglogenesis. J Histochem Cytochem 34:85-91, 1986 18. Madri JA, Williams SK: Capillary endothelial cell cultures: phenotypic modulation by matrix components. J Cell Biol 97:153-165, 1983 19. Matsuda T, Akutsu T, Kira K, et al: Development of hybrid compliant graft: rapid preparative method for reconstruction of a vascular wall. ASAIO Trans 35: 553-555, 1989 20. Pearce WH, Rutherford RB, Whitehill TA, et al: Successful endothelial seeding with omentally derived microvascular endothelial cells. J Vase Surg 5:203-206, 1987 401
H. Kobayashi, et al. 21. Ramalanjaona G, Kempczinski RF, Rosenman JE, et al: The effect of fibronectin coating on endothelial cell kinetics in polytetrafluoroethylene grafts. J Vase Surg 3: 264-272, 1986 22. Rosenman JE, Kempczinski RF, Pearce WH, et al: Kinetics of endothelial cell seeding. J Vasc Surg 2: 778-784, 1985 23. Schmidt SP, Hunter TJ, Falkow LJ, et ah Effects of antiplatelet agents on PGI2 production by endothelialcell-seeded small diameter Dacron grafts. J Am Soc Artif Intern Organs 8:99-103, 1985 24. Sharp WV, Schmidt SP, Donovan DL: Prostaglandin biochemistry of seeded endothelial cells on Dacron prostheses. J Vasc Surg 3:256-263, 1986 25. Shepard AD, Gelfand JA, Callow AD, et ah Complement activation by synthetic vascular prostheses. J Vasc Surg 1:829-838, 1984 26. Shindo S, Takagi A, Whittemore AD: Improved patency of collagen-impregnated grafts after in vitro autogenous endothelial cell seeding, a Vase Surg 6:325-332, 1987 27. Stanley JC, Burkel WE, Ford JW, et al: Enhanced patency of small-diameter, externally supported Dacron iliofemoral grafts seeded with endothelial cells. Surgery 92: 994-1005, 1982
402
28. Takenaka K, Kholoussy AM, Yang Y, et al: The healing of the thin-walled expanded polytetrafluoroethylene vascular graft. Vase Surg 19:383-389, 1985 29. Weibel ER, Palade GE: New cytoplasmic components in arterial endothelia. J Cell Biol 23:101-112, 1964 30. Weksler BB, Marcus AJ, Jaffe EA: Synthesis of prostaglandin Iz (prostacyclin) by cultured human and bovine endothelial cells. Proc Natl Aead Sci USA 74: 3922-3926, 1977 31. Willams GT, Underwood CJ, Charlesworth D: Early resuits of femoropopliteal bypass using a five millimeter "thin wall" polytetraflnoroethylene (Gore-Tex) prosthesis. Ann Vase Sarg 1:208-213, 1986 32. Zetter BR: Culture of capillary endothelial cells, in Jaffe EA (ed): Biology of Endothelial Cells. Boston: Martinus Nijhoff, 1984, pp 14-26
Manuscript received January 24, 1991. Accepted in final form March 4, 1992. Address reprint requests to: Hidenori Kobayashi, M.D., Department of Neurosurgery, Fukui Medical School, Matsuoka, Fukui 910-11, Japan.
J. Neurosurg. / Volume 77/September, 1992
