CA2186375A1

CA2186375A1 - Improved blood contact surfaces using endothelium on a subendothelial extracellular matrix

Info

Publication number: CA2186375A1
Application number: CA002186375A
Authority: CA
Inventors: William Carl Bruchman; Paul Christopher Begovac
Original assignee: Individual
Current assignee: WL Gore and Associates Inc
Priority date: 1994-04-29
Filing date: 1995-04-28
Publication date: 1995-11-09
Also published as: JPH09512184A; EP0757562A1; WO1995029713A1; AU2399495A; US5879383A

Abstract

This invention relates to improved blood contact devices such as vascular prostheses ren-dered antithrombotic through the use of recip-ient endothelial cells grown on an appropriate subendothelial matrix. The subendothelial ma-trix layer, which serves as the substratum for growing endothelial cells, may be obtained from either natural donor vessels or from in vitro tis-sue culture sources. This subendothelial matrix is used in situ on the donor vessel, or is grown or applied to a synthetic component, preferably porous expanded polytetrafluoroethylene. Once this subendothelial matrix is prepared, recipi-ent endothelial cells are seeded onto this matrix substratum, which then serves as the immedi-ate blood contact surface. The endothelial cells may be applied as an intra-operative procedure, or grown on the subendothelial matrix substra-tum in vitro until the cells establish a confluent monolayer. A key aspect of this invention is that living, recipient endothelial cells are grown on the appropriate subendothelial matrix substra-tum, thereby providing a substantially nonthrom-botic blood contact surface. Furthermore, the likelihood of recipient immunological response is minimized. This invention results in vascular prostheses that are particularly useful for arterial bypass requiring a diameter of 6 mm or less.

Description

~W095/29713 2186375 p I~ROVED BLOOD CONTACT SURFACES USING ENDOT~LIUM ON A
SUBENDOTHELLAL EXTRACELLULAR MATRIX
B~('KGROUND OF T~F INVENTION
I . Field of the lnvention The present invention relates to improved blood contact surfaces for use in apparatus such as in artificial blood vessels and other implantable appliances and methods for producing the blood 10 contactsurfaces.

2. Description of Related Art Synthetic materials are widely used to replace diseased or damaged portions of the human Lu . . ' system. The use of grafts composed of synthetics such as l,ul~ G t~,." ' ' and expanded pO~; n U~lh,~ . (ePTFE), has provided successful results in the It,UI~ t 15 of large vessels such as the aorta, iliac, or femoral arteries. The application of these same synthetics, however, to small diameter arteries such as in coronary artery bypass or peripheral arterial bypass beyond the popliteal artery often produces thrombosis despite patency-enhancing ~ L ' measures Other applicabons of synthebcs in which clot formation is a frequent problem include the It~ of veins, heart valves, and the artificial heart. Tissue-based 20 prosthetics provide similar l,. . fi .. "~ to synthetics, having acceptable function in some ~, ' . but in others, still having the . , ' of thrombus formation. These thrombotic problems limit the usefulness of both tissue-based and synthetic d~ices, particularly in more demanding ,, ' C~ . 'y, a goal of researchers for many years has been to develop a blood contact surface that provides for the reduction or eliminabon ofthrombosis.
Natural blood contact surfaces, such as those found within blood vessels, possess that prevent blood from clotting during normal passage along the surface. In the case of a mammalian artery, the immediate blood contact surface compnses a layer of l~ur. ~
endothelial cells (ECs). ' ' ~ extemal to the endothelial cell layer is the remainder of the intima: a s.*- - ' ' ' ' matrix layer of basement membrane and underlying ~ U~ bearing . 30 ~ " ' matrix, and the intemal elastic lamina Surrounding the intima layer is the ' ' media structure containing smooth muscle cells (SMCs) and elastm, and ~UI 1~ this media is the adventitia, the most extemal layer comprised of fibroblasts and connective tissue.

WO 95129713 2 1 8 6 3 7 5 r~
Both the I ' ' ' ' layer and media are generally considered to be ;~ IG in nature in order to mamtain hemostasis when the vascular system is injured.
Intact endothelial linings are considered to be non ~ unless damaged. Because of their blood-Gontacting locahon~ endothelial cells have been thoroughly investigated with respect 5 to their ;' ~ , ,, function. Endothelial cells are known to synthesize or bind a number of substances with . ~,' '' ' ~ or fibrinolytic function including heparan m, dermat~n r 1.~ cofactorlI, i' . ' ' ' 'p~.~l C/protein S, prostacyclin and tissue-type I ' ~ activator r. i =, segments of ' ' ' bearing autologous vessel i . ' ' from one site to another in an individual to bypass diseased vessels 10 exhibit an mcidence of thrombosis ' "~ less than that of synthehcs used in the same application. For these reasons, it has been assumed that endothelial cells are responsible for the non ~ aGtivity of vessels. According to this assumption, synthetic blood contactdevices capable of thrombo-resistance similar to the native vasculature would require a blood contact surface of endothelial cells.
Numerous attempts have been made to provide prosthetic surfaces, and specifically vascular grafts, that include or develop an endothelial cell lining. The u ._. ~.' ' ,, majonty of these attempts have been calried out as an . ~c cell-seeding procedure. T , ~, cell-seeding typically involves harvesting endothelial cells from the recipient during the procedure and ' '.~ seeding these collected cells onto a vascular graft that has been pretreated with a 20 substrate to enhance endothelial cell attachTnent. Substrates frequently applied to the synthetic surface include preclotted blood taken from the patient or; " ' matrix proteins such as fibronectin, collagen, or laminin, either singly or in ' This approach was first reported by M. B. Herring ef al. in "A single staged technique for seeding vascular grafts with autogenous 1( ' ' " ~ç~ 84:498-504 (1978). In this procedure, autologous cells were seeded onto a 25 preclotted DACRON 19 graft. These seeded grafts ~'- ' a decreased thrombus formation compared to control grafts without endothelial cells. In spite of the improved initial adherence of the cells to the synthetic surface afforded through the use of these various substrata, the shear forces resulting from blood flow 1..,.~ ' ' leads to the loss of a substantial fraction of the applied cells. t In a variation of the above, U.S. Patent 4,960,423 to Smith describes the use ofelastin-derived peptides to enhance endothelial cell attachment. Some studies have used SUBSTITUTE SHEET (RULE 26) ~w0 95129713 ~1 8 6 3 ~ 5 r~l,u~
....... 1.;.. ;.. ~ of " ' matrix molecules and cells to provide a substrate for endothelial cell attachment and growth. For example, U S. Patents 4,539,716 and 4,546,500 to Bell describe means by which endothelial cclls are grown on a living smooth muscle cell-collagen lattice. ~n addihon~ U.S. Patent 4,883,755 to Carabasi et al. describes a technical method for seeding 5 endothelial cells onto damaged blood vessel surfaces.
Altemative mearls of growing endothehal cells on vascular grafts have also been reported.
For exarnple, the use of physical force to apply endothelial cells to graft surfaces is described in U.S. Patent 5,037,378 to Muller et al. In another example, U.S. Patent 4,804,382 to Turina and Bittman describe the application of endothelial cells to a semi-pemneable membrane in which the 10 pores are filled with aqueous gels to allow endothelial cell coverage.
Another approach has been the seeding of endothelial cells onto ~ derived sulfaces, including p.,.'.,.u~' , cardiac valve leaf~ets, arnnion, and arteries. These efforts appear to have originated with J. Hoch et al. in "In vrtro ' " ' ' of an aldehyde-stabilized native vessel," J. Sur~. Res. 44:545-554 (1988), where the authors attempted to grow ~ h. l:-..,. on 15 ficin-digested, dialdehyde stabilized, bovine artery. Human venous endothelial cells were found to adhere to and spread on the remnant collagen surface oi' these enzyme-digested grafts, but no implant studies were perfommed. J. Hoch et al. also invesbgated the growth of ' ' ' on human arnnion, on live, ' '~ scraped human artery, and again on ficin-digested, tanned bovine artery. (J. Hoch et al., "Endothelial cell interactions with native surfaces," Ann. V~cr Sur~.
20 2:153-159 (1989)) Although endothelial cell aohesion was observed on these surfaces by 2 hours, the longtemm persistence of ECs on these surfaces was not exaTnined, and, as in the previous study, none of these ~ ' ' ' ' materials were implanted as vascular substitutes. P.A. Schneider et al. showed that endothelial cells could be successfully seeded onto the remaining collagenous surface of baboon vessels from which the intima was removed. (P.A. Schneider et al., "Confiuent 25 durable ' ' ' ' of cl~h~c~ baboon aorta by early attachment of cultured endothelialcells,"J.VA~r Suro. 11:365-372(1990)) S.G.Lalkaetal.useddetergentextractionof canine arteries to produce an ethanol-fixed acellular vascular matrix onto which human umbilical vein endothelial cells were au~ aarull~ seeded in vitro. (S.G. Lalka et al., "Acellular vascular matrix: A natural endothelial cell substrate," Arm Vasc Sur~. 2:108-117 (1989)) In a study similar to that of Hoch et al. in J. Sur~. Res.. supra, L. Bengtsson et al.
au~,~,~arul~ grew human venous ci ' ' ' on the luminal surface of devitalized vessel segments Wo sst2s7l3 ~ 1 8 6 3 7 5 denuded of ~ . l. .'h ! "~ and ~ . (L Bengtsson et al., "Lirling of viable and nonviable allogeneic and xenogeneic car&ovascular tissue with cultured adult human venous ~ ' ' ' ," L
Thn~r Card~ r SUrO. 106 434-443 (1993)) It is emphasized that in none of the foregoing stu&es have attempts to provide endothelial 5 linjngs to biologihal hssues been carried out on tissues with an intact ~. ' ' ' r... =, m none of these attempts have the tissues been tested as chronic vascular substitutes following ~ ' ' ' ' Biological tissues destined for implant use are commonly subjected to a IJlGaGI~.
treatment employing fixative agents such as "' '' ',.', ~( '' ',lt, &aldehyde starches, 10 polyepoxy compounds, or alcohols. Glu~ dlJ~,~.J 1G is the most commonly used crosslinking agent behause it provides reduced ,, ~, excellent tissue l~ltaGl~ and stability in theimplant h.. .l ' The use of _' ' ' ' Jlt as an agent to produce articles denved from biological tissues suitable for . ' is specified in a number of U.S. Patents, including 3,562,820 to Braun, 3,966,40~ to Hancock et al., 3,988,782 to Dar&k, 4,050,893 to Hancock et al., 15 4,098,571 to Miyata et al., 4,323,358 to Lentz et al., and 4,378,224 to Nimni et al. Although g' ' ' ' JJG-treated tissues have generally provided good implant outcome, it has been shown that unreacted, residual _' ' ' ~.IG can result in the inhibibon or death of cells grown in contact with the fixed tissue. Previous i..~ , (see, for exarnple, L. Bengtsson et al. sllpra) have shown that growth of G 1~ ' ' on fixed tissue can be inhibited by residual _' ' ' ' Jle.
The problem of residual ~' ' ' ' J 1G has been addressed by blocking the reacbve site on the aldehyde groups. It is well known that residual or free aldehyde groups can be effectively blocked using compounds containing amino groups such as amino acids or proteins. Grimm et al.
("~`' '''Jlt affects l,;o.,u..., ' '~ of l~u,u~ ' ' heart valves," Surroery 111:74-78 (1992)) describe a postfixahon treatment by which reactive aldehyde groups can be passivated.
25 Endothelial cell-seeded"~,' ' ' ', IG-fixed tissue showed uninhibited cell growth followmg a 48 hour exposure to 8% L-glutamic acid. Nashef et al., in U.S. Patent 4,786,287, specify the use of solutions containing an excess of aldehyde-reactive amines to increase the rate of aldehyde &ffusion from fixed tissue by maintaining a low .~~ of free aldehyde. U.S. Patent 4,553,974 to Dewarljee describes another method to prepare collagenous tissues for 30 G ' ' ' ' using a surfactant-detergent treatment, followed by 1ut~l,1J~,hJdG fixation and anti-calcification treatments. Although: ' ' ' has been previously grown on tissue surfaces, ~nESUEr~E~) ~wo 95129713 2 1 8 6 3 7 5 . ~ J~
the treatments employed, including ficin digestion, detergent treatment, and mechanical removal, each disrupt the ~..1.. 1. .~-- 1:-l layer, resulting in increased i..., ' ) ~ ~ . Additionally, bssues such as pericardium or amnion do not, in the native state, possess the required ~IJV. ' ' ' ' matrix.
Recent efforls to provide a substrate that supports endothelial cell linings on graft surfaces include the use of in vitro cultured i " ' matrix. A. Schneider et al. used comeal endothelial cells to produce i " ' matrix on ePTE~ vascular grafts. (A. Schneider et al., "An improved method of endothelial seedmg on small caliber prosthetic vascular grafts coated with natural; " ' matrix," Clin. Mat. 13:51-55 (1993 After pri~duction of an " ' 10 matrix, these onginal cells were then removed using Triton X-100 and NE~OH~ and the tubes were seeded again with bovine aortic ....I..:h. I; . This approacb showed that ' ' ' could be grown on the ~ " ' matrix lining the ePTE~E grafts, however, no implant studies were performed. Another approach by Y. S. Lee et al. employed fetal human fibroblasts in culture to secrete ,. ~ ' - matrix onto a pvl~ ~ ' tube. The fibroblast cells were then removed by 15 one of several methods and the remaining matrix seeded with human omental endothelial cells.
This method resulted rn a patent graft at 5 weeks after i . ' into a rat aorta. (Y. S. Lee et al., "~niiothelial cell see&g onto the; " ' matrix of fibroblasts for the d~,~, ', of a small diameter pv4 . ' vessel," ASAIO ,T ' 39:M740-M745 (1993)) A similar approach has also been taken by H. Miwa et al., "D~,.. '~, of a ,h"_,~", ' ' '1!~
20 structured hybrid vascular graft ' ' ,, natural arteries," ASAIO 7 ~ 39:M273-77 (1993). In this case, smooth muscle cells are layered over a DACRONi~ graft in an applied artificial matrix of collagen type I and demmatim sulfate sl~ .. A layer of endothelial cells is then grown on the artificial matrix to serve as the blood contact surface.
Since the .. I.. :h. 1:.. is the ~h.u.. l '~, ' source of the ' ' behavior of the 25 nommal vasculature, an endothelial lining has been wi~ely regarded as the means by which improved ' U...Vf ~ of ~d;VV~ implants v~ill be achieved. Actual testing however, has provided mixed outcomes. A number of studies have been conducted in animals, with some reporting clear i~ r-- ._....,..b in patenf;y as a . , of ., ' ' ' ' Several studies, though, report equivocal outcomes with linle measurable , u. (P. Zilla et al., "The ' ' ' A key to the future," J. Card. Surg. 8:32-60 (1993)) P. Ortenwall et al., for example, showed similar patency outcomes between seeded and control grafts in both sheep and dogs. (P.

wo95/29713 21 86375 Ortenwall et al.,"Seeding of ePTFE carotid illb,.r grafrs in sheep and dogs: species dependent results," Sw_erY 103:199-~05 (1988)) Aithough the collective results of ."~,u. ' seeding studies conducted in animals are suggestive of a 1, r.... ~ iIII,UIU ., ' due to the addition of recipient endothelial cells, the data 5 do not indicate a l, r . . ~ A u. .~ to the level expected v~ith the use of the autologous vessel. For example, in a test conducted in sheep, N. L. James et al. report that only one of six (1/6) endothelial cell seeded grafts was patent in comparison to six of six (6/6) autologous artery grafts. (N. L. James et al.,"ln v~vo patency of endothelial cell-lined expanded p~lyt~,t~ "u~ prostheses in an ovine model," Artif, Ore. 16:346-53 (1992)) Similarly, 10 application of the above described c ' ' ' ' methods to synthetic vaScular implants in humans has shown litde ~' ' '~ IIII~IIU._..._... in patency, despite the observation that grafts seeded with recipient cells become; ' ' ' ' i, at least in some cases. (P. Ziiia et al., "The ..... A key to the futwe," J, Card. Suro. 8:32-60 (1993)) Thus, the actual clinical outcome achieved with; ' ' ' ~ ' synthetics has not met the . -~ raised by animal studies.
sr~l\ll\l~RYOFTRF~ ~rrnN
The present invention is directed to an improved blood contact surface suitable for use in artificiai blood vessels and other implantable blood contact appliances.
A key element of the present invention is the ' of living endotheliai cells of vascuiar origin attached to an appropriate ~ ' ' '- ' matrix to forrn a composite. This composite provides an improved patency ~. ~ compared to currently existing vascuiar graft prosthetics, particuiarly in smail diameter a" ' Donor ~ d;uv~ tissue, having had its vaScuiar endotheliai cells removed, can be used to provide an appropriate _' ' ' ' ' matrix for 25 use in the present invention. Altematively, an appropriate _L ' ' ' ' matrix layer can be obtained by removing the _l ' ' ' ' matrix layer from a native vascuiar swface and transferring it to a synthetic component. In yet another method, an appropnate ' ' ' ' ' matrix iayer can be grown in vitro using endotheiiai cells iP, ' with smooth muscle celis.
In any case, the originai: ' ' ' is removed, exposing the s ' ' ' ' ' matrix layer for 30 subsequent seeding v~ith recipient vascular endothelial cells. Autologous vascuiar endothelial cells are preferred for seeding a ' ' ' ' ' matrix layer of the present invention. Geneticaily ~W095/29713 2 1 8 6375 engineered endothelial cells having ,, cell surface factors removed may also be suitable for use u~ the present invenion. In an opional step, the ~ ' ' ' ' matrix layer may be preserved prior to seeding with vascular endothelial cells. The term "preserved ' ' ' ' ' matrix layer," in the instant ~ ;, refers to a ~ "~ matrix layer, that has been 5 treated with a fixative solution, such as ~u~.h~d~, to chemically stabilize the matrix and thereby preserve the ~ ' ~~ ' ' ' ' matrix layer as a surface upon which vascular endothelial cells attach and grow to form a direct blood contact surfæe. Once an appropriate, ~l .. - -- l. ~ - l: l matrix layer is isolated and ophonally preserved, living vascular endothelial cells are seeded onto this matrix and allowed to grow to conrduency to form a direct blood contact surface. A I ' ' 10 matrix layer of the present invention, preserved or I . C~ .,.l, also serves as a direct blood contact surface in the absence of recipient vascular endothelial cells attached to and growing on the l ' ' ' ' matrix layer. Thus, the final blood contact surface compnses a ~ ' ' ' ' matrix substratum preferably with endothelial cells attached to the substratum. In the absence of complete recipient endothelial cell coverage, the ~ l: l matrix itself also serves as a blood 15 contact surface.
Accordingly, the present invention comprises a blood contact surface which compnses a substratum comprising a ~ ' ' ' ' ' matrix layer that supports attachment and growth of applied ~.~d;u~ ' endothelial cells, a layer of applied ~ Luv~l,ul~ endothelial cells attached to the ~ - ~. . l .11. ' -l matrix layer, wherein the wud;u~ endothelial cells serve as a direct 20 blood contact surface and wherein, in the absence of cells attached to the substratum, the substratum serves as a direct blood contact surface.
While the present invention is particularly applicable to blood vessels and similar appliances, it may be applied to many areas where ' ",~ llu.~l..u...l,u~.u blood contact surfaces must be provided. Examples of other appliances that may be produced in accordance with 25 the present invention include, without limitation, flexible sheets, heart valves, artificial hearts, artificial organs such as implantable artificial kidneys, and others.
DESCRIPTION OF TT~F, DRA~IN('~
30 The operahon of the present invention should become apparent from the following descripbon when considered in conjunction with the . , Ill~ drawings, in which:

~snmEsOEr~

WO 95/29713 2 ~ ~ 6 3 7 5 P~I/U_ 5 _J~ ~
Figure I is a perspective sectional view of a mammalian artery.
Figure 2 is a sectional view through tbe longitudinal axis of a mammalian artery wit,'~ t'e endothelial cell layer removed exposing the ' ' ' ' ' layer (14).
Figure 3 is a sectional view of one ....I,,~,I;"....1 of the present invention whereby living 5 endothelial cells (24) have been applied to the surface of the mammalian artery structure (26) of Figure 2.
Figure 4 is a view of a tubular vascular graft e b~ ' (30) of the present invention.
Figure 5 is a cross-sectio~ view of a layer of ondothelial cells (12) applied directly to a layer of smooth muscle cells (20) on a synthetic component (32).
Figure 6 is a cross-sectional view of the structure shown in Figure 5 once a ~ul.. ' ' ' ' matrix layer (14) has been generated between the endothelial cells (12) and the smooth muscle cells (20).
Figure 7 is a cross-sectional view of the structure of Figure 6 in which the endothelial cell layer (12) has been removed exposing the - '.~,--1 ,1'.. 1: 1 matrix layer (14).Figure 8 is a cross-sectional view of one . ~ of the present invention wherein recipient endothelial cells (24) have been applied to the ~ ' ' ' ' matrix layer (14) shown in the structure of Figure 7.
Figure 9 is a view of one i ' ~ of the present invention wherein the structure shown in Figure g is grown on a flat sheet of synthehc base matenal (28).
Figure 10 is a view of a heart valve appliance (30) wherein the structure of Figure 8 is employed.
DETAn .FT) I~ ('R~I~ON OF INVFNnON
By way of L.`~. ' ' , Figure I illustrates the structure of a typical mammalian artery (10). The artery (10) structure consists of an intima layer having an innemmost layer of endot.'lelial cells (12), a ~ 1 l matrix layer (14) consisting of basement membrane and other 1- matrix . ~ , and an intemal elastic lamina layer (16). Extemai to the intima is n media layer (18) composed of smooth muscle cells (SMC) (20), and finally a fibrous comnective 30 bssue or advenhtia layer (22).

SUBSTITUTE SHEET (RULE 26) ~wo ssns7l3 ~ 1 8 ~ 3 7 5 . ~ 5 The present rnvention is directed to improved blood contact surfaces, such as those used in artificial blood vessels and other blood contact appliances. It is recognized that the; ' ' ' provides the primary ,.. ~ ;. character to the surface of the native vasculature. Despite this, implant testing shows that: ' ' ' " ' prostheses exhibit only equivocal ~ Iv.~ ,.lb in 5 patency in humans rn comparison to the results achieved with the non ~ ' ' ' ' l prostheses alone. Since the; ~ is thought to be the sole source of ' ' properties, the absence of patency , J . _...~ in the human is startling.
It has been discovered that an additional element is necessary rn order to realize the properties of the . ' ' ' on prosthetic surfaces. It has been found that the 10 ', ' - function of an; ' ' ' ' ' surfæe is dependent upon the presence of an appropriate biological substrate, specifically a ~u~ ' ' ' ' " ' matrix layer. Although the ,. .l ,. . ,. l, ,ll ,. I l matrix is widely believed to be highly i' I O , in the absence of this layer, an endothelial lirling confers only a marginal 1,, r advantage to the blood contact surface.
In the present invention, three general ' ' ' ~ have been developed to obtain an15 appropriate ~.~l...,.l.. l l; ~l matrix layer on which vascular endothelial cell growth can be achieved.
The first method utilizes donor w.J;~,v~ tissues from which the original G ~ '' '' has been removed, but the rntact ~ '~~' ' ' ' matrix is preserved tn situ. In the second method, the ' ' ' from donor cardiovascular tissues can be removed, and applied to a synthetic component. In the third method, the desired l l ' ' ' layer can be synthesized directly on an 20 appropriate surface usirlg in vitro cell culture techniques. With the third method, both smooth muscle cells and endothelial cells are required to produce the appropriate " ' matrix by co-culture methods. Once a ~ ' ' ' ' matrix surface is prepared by one of these three mearls, recipient endothelial cells can either be grown to confluence prior to . ' or can be seeded onto the surface at the time of implant to form a direct blood contact surface, thereby providing an 25 i ' ' ' ' ' graft with improved patency p~ ~
Irrespective of which of the above sources is employed, the ! l ' ' ' ' matrix of the present invention has several advantages over previously used surfaces. First, the endothelial cells that grow on a I ' ' ' ' matrix surface are tenaciously adhered, reducing concerns for cell loss. As a result, the endothelial cells tend to quickly reach and maintain confluence in comparison 30 to other surfaces. Second, the ' of the ~- ' ' ' ' ' matrix and the endothelial cells is much less prone to thrombosis than an endothelial layer lacking the I ' ' ' ' matrix. Third, ~IIIIESEr(~9,~`

WO 95/29713 2 1 ~ 6 3 7 5 P~
despite the findings of the bulk of the literature to the contraTy, the ,..1,, ~L '' '- ' matrix itself has Ullll,VtlC properties and can serve as a direct blood contact surface. Tbis is of particular value when the present invention is practiced with b.~ cell see&g. Under these condihons, tbe graft can be seeded with recipient endothelial cells with reduced concem for thrombus 5 deposition that could ~,ulll,ulvllfir~c graft function before the cells grow to confiuence. Moreover, that same thrombus could itself interfere with subsequent cell coverage of the surface.
Previous attempts to employ endothelial cells to provide ' u..~ ti~. surfaces have not recogruzed the necessity of the combination of endothelial cells and r ~ 1; 1 matrix, as does the present rnvention. A key element of the present invention, therefore, is that the final endothelial 10 cell layer be grown on the appropriate matrtx. The appropriate matrix, as ~ by the present invention, is either the native ~ ' ' ' ' ' matrix found in the wuV5vvra~.Ulrl system or the matrix produced by an appropnate co-culture of endothelial cells and smooth muscle cells.
Srnce the r~lJb. ~ matrix layer is susceptible to damage by mechanical, enzymatic, or certain chemical processes, it should be prepared accor&gly in order to preserve its l.u..l.lu..lbuti~, 15 properties. One of the crittcal .~ s of the present invention is that the ' ' ' ' ' matrix be maintained rn an intact fomm.
PrPr~ n 'Useof rT s ~ r~
The starting material for graft fabncation is donor l,~ulivvr~l,uLII tissue obtained from a vertebrate specieS, preferably a mammalian species including, but not limited to, bovine, porcine, ovine, equine or primate, including human sources. Avian sources may be u3eful as well. The ~uv~ tissue may be obtained from the heart, arterial vessels, venous vessels, and/or capillaTy beds. For example, arterial vessels may be obtained by freezing a limb intact or by 25 immediate dissection of the ~,u,u.l, ' ' '~, sized vessels. If the limb segment3 are initially frozen, the limbs should be thawed first, then the vessels dissected free and placed rntO 4C salrne buffer at ,ullr3;Jlog;~,rl pH for further processing. Appropriate saline solutions include any number of JI.Jg;.~ balanced salt solution3, including, but not limited to, Dulbecco's ~
saline, Earle's balanced salt solution, Hanks' Balanced Salt Solution (~SS), or phosphate buffered 30 saline, for example. Processing entails dissection of the extraneous adventibal hssue and the ligahon of any branch segments with polymenc suture material. Once these steps have been lû
SUE~STITUTE SHEET (RULE 26) ~wo ss/2s7l3 ~ ~ ~ 6 3 J 5 r ~
carried out, the vessel segments are ready for further processing to remove the endothelial cell layer.
The removal of the endothelial cell layer from the ~ ' '' ' must be carefully perforrned m order to maintain the intact s ' ' " ' ' matrix and its ' . ' properties of 5 the I ' ' ' The use of overly aggressive techniques, such as detergent stripping, are found to damage the ~ ' ' ' ' matrix.
There are a number of relatively simple agents and methods that v~ill remove theendothelial cell layer without adversely affecting the ' ' ' ' ' matrix. The ~
may be treated with a mild chemical stripping solution, such as, ammonium hydroxide (NH40H), 10 for example. One such treatment may consist of incubating the vessel lurnen in an aqueous NH40H solution at a of about 0.01 M to about 0.5 M, for a period of about 30 seconds to about 60 minutes followed by flushing the vessel lumen with a buffer solution. In the preferred ~ ~ ' t the treatment involves treating the vessel with a 0.25 M NH40H solution for about 3 to about 5 minutes. The resulting artery structure is shov~n in Figure 2.
Another suitable tecbrlique for endothelial cell removal is the use of air drying of the luminal surface. One . ' ' air drying treatment rnvolves flushing the vessel lumen with buffer followed by air drying for about 5 to about 10 minutes at a flow rate of about 500 to about 2000 cc/min for a 4 mm intemal diarneter vessel. Preferably, the flow rate is about 1000 cc/mm.
Following air drying, the vessel lumen is flushed with buffer to remove the endothelial cells and 20 rehydratethe~ `- ' ' ' 'matrixpnortofurtherprocessrng.
In another ~ ~ ' of the present invenbon, the endothelial cells are removed by exposing the artery to a series of freeze-thaw cycles, preferably greater than two, followed by a flushing of the artery lumen with a suitable solution such as HBSS (1.3 mM CaCI2, 5 mM KCI, 0.3 mM KH2PO4, 0.5 mM MgCI2, 0.4 mM MgSO4, 138 mM NaCI, 4 mM NaHCO3, 0.3 mM
25 NaHPO4, 5.6 mM glucose).
Once the 5_' ' '' ' ' matrix surface has been prepared by endothelial cell removal as described above, a processing step is employed to prevent or minimize calcification of the vessel tissue following; ~"1 ~ This technique utilizes a solvent solution to extract ~
lipid I r ' One effective treatrnent involves irnmersing the artery in a solution of 30 cbloroforrn and methanol (CHCI3/CH30H) at a volume:volume raho of about 2:1 for a period of about 15 to about 60 minutes, with a preferred treatment period of about 20 to about 40 mmutes.
Il WO 95/29713 2 1 8 6 3 7 5 P ~
The vessel is then removed and rehydrated through immersion in a saline solution. Once re-hydrated, the graft can then be preserved in a manner descnbed below. The chloroform and methanol treatment does not damage the ~ matrix, but serves to limit calcif~cation of the artery once implanted. It is understood that other: ' " treatments may be employed 5 as well.
Following rehydration, the vessel is treated with a fixative in order to preserve the ~.,1,. 1"~}, !'_1 matrix, to reduce ., ~, and to sterilize the vessel segment. Suitable fixative agents include, but are not limited to, alcohols, ~' ' ' ', ', ~ ' ' ' ,.', dialdehyde starches, or polyepoxy compounds, for example. In a preferred form, the vessel is fixed under 10 pressure. This is 9 , ' ' ' by placing the vessel within a tube that will provide an appropnate intemal diameter, filling the vessel with the fixing solution and p.c _ to about 21 to about 35 kPa. Suitable tubes for providing an appropriate diameter include, but are not limited to, an ePIFE
graf~ such as those available under the trademark GORE-TLX~ID Vascular Graft from W. L. Gore & Associates, Inc., Flagstaff, Arizona, for example. The duration of pressure fixation is dependent 15 upon the fixation agent and the l employed. If, for example, _' ' ' ' ', ' is chosen as the fixative agent, suitable solutions would include about 0.1-2.5% in an appropriate buffer. The duration of pressure fixation would then be about I hour to about 72 hours, depending upon the ,,' ' ' ', Jt chosen. Buffers suitable for this use include, but are not limited to, N-2-lydlu~ .,.~w-N'-2-ethane sulfonic acid (HEPES), acetate, 2(N-20 morpholino); ' "` acid ~ES), 3-~N ,' ' ] ~"" '~ - acid (MOPS), tris LJI~O~ '', - '' , phosphate, and others, for example.
In the preferred ~ ~ ' t, the vessel is pressure fixed in 100% ethanol for about 2-72 hours, after which it is rehydrated using a graded ethanol series of decreasing .,... ,.. ~ ... It may be possible to use other agents such as r ~ ',J~ or O' '' 'JJ~i to provide additional 25 fixahon or for use as sterilants. If these agents are employed, however, an additional process step is required to block any residual aldehydes so that subsequent cell growth on the surface is not inhibited. Amino group containing solutions suitable for the blocking step include, for example, 0.1 M glycine, Medium 199, or Dulbecco's Modified Eagle's Medium. Following fixation, or fixation/blocking treatments, the vessels are rinsed and stored in a sterile 4C normal saline solution 30 until ' , ' or further use.

SUBSTI~UTE SHEET (RULE 26) ~ WO95/29713 2 1 8 6 3 7 S
In addition to the ~ l v~ of the present mvention described above, two other c ~ ' of this invenbon employ eit.'er native s '~~'~ ' ' denved from vessel and transferred to a synthebc surface, or an analogue of the ' ' ' ' gro~vn on the synthehc surface via in vifro cell culture. In an optional step, the ~ ~...,.l.,~h. 1;.1 Iayer of either method can 5 be treated with a fixahon agent such as ~'l ' ' ' JJe. In comparison tv blood vessels fixed with.
aldehydes, the decreased mass and diffusional distances of these very thin layers provide a decreased likelihood of residual aldehydes. A blocking step such as that described above for the blood vessel is still ~ -l. J however, to ensure the lack of cell growth inhibition.
The ~ l matrix layer preserved by the above processes contains numervus 10 c " ' matrix , Using , tu~.h.,~fi~,dl assays, it has been determined that this ~ I ' ' ' ' matrix layer includes chondroitin sulfate ~ ,t~,~,gl~. , fibronectin, collagen Type I, collagen Type m, collagen Type IV, and elastin.
Once the - l. I-~1h 1 -1 matrix layer has been prepared, the matrix layer 14 serves as the substratum upon ~vhich recipient endothelial cells 24 are seeded and grown tv produce a structure 15 resemblimg that shown in Figure 3. The endothelial cells may be applied in a number of ways. For example, endothelial cells may be harvested from the graft recipient as an . ~, procedure and i ' '~ applied to the ~ ' ' ' ' matrix substratum. Altematively, endothelial cells may be isolated from a future recipient, grown in vitro on the ' ' ' ' ' matrix substratum umtil confluent, then implanted intv the patient.
In the two ' ' of the present invention that employ derived or in vttro l -.l,. "l~, h l; ,l matrix layers, the ' . ~ properties of the composite are supplied by the biological component, while the requisite mechanical properties are supplied by the synthetic component In one of these ~ :~ ' . the _' ' ' ' is removed from the donor tissue and 25 placed onto a synthetic component. In this case, the synthetic component supplies all of the required mechanical support. In another ~ ' of the proeent invention, the synthetic component consists of an extemal sheath that is placed around the donor vascular tissue, and the support is thus a composite of that provided by the donor hssue and the extemal sheath. While the biological tissue is subject to degradative processes that could reduce its strength below that 30 required for the application, the synthehc portion can resist these same processes where the appropriate polymer and fabrication are chosen.
l3 SUvSTlTUTE SHEET ~RULE ~6) ~09~129713 21 ~375 r~
One skilled in the art will recognize that minimum mechanical integrity ~ .,., exist for each blood appliance application. One minimum requirement is that the appliance resist the forces imposed by blood pressure for the intended durahon of a given application. The S" must thus resist the hoop stress induced by blood pressure v~ithout rupture or 5 aneurysm formation, and must similarly resist the stresses induced at the: without rupture, excessive suture hole elongation, or false aneurysm fommation.
It is preferred to apply a sheath material to the extemal surface of the vessel to provide mechanical strength and to block cellular inflltration from the surrounding tissues. The pore size that is effective in blochng cell passage across the thickness of the sheath is dependent upon the 10 thickness of the wall itself and the tortuosity of the pore connecting the outer surface with the inner surface of the sheath. In the case of regular uniform pores in a thin construction, ranging from about 10 ~m to about 20 llm in thickness, the pores must be selected to be less than about 3 llm in diameter in order to block cell passage across the thickness of the sheath This value is based upon Boyden chamber migration assays where the limit of fibroblast invasion through straight, uniform 15 pores is about 5 llm to about 8 llm and the limit of invasion of leukPcyte invasion through straight, unifomm por3 is about 3 llm to about 5 llm. (A. Albini et al, "Fibroblast chemotaxis," Colla~en R~l~ R3. 5:283-296, (1985); W. Morzycki et al., "Tumour necrosis factor-alpha but not interleukin-l induces l.ul~ .' ' leucocyte migraion through fibroblast layers by a fibrpblast-dependent mecharlism,' ~ . 74:107-113, (1991)) In the case of a thicker, 20 more tPrtUOUs, or laminate structure, the individual por3 can be somewhat larger and still serve as an effective barrier to the passage of cells across the layer.
For an ePT~ or similar Sbrillated sheath, the pore size of the sheath is directly related tP
the fibril length of the sheath material. The fibril length should be chosen to form pores that r3ists cellular access through the sheath, while remaining permeable to ~ In defining the 25 ability of the sheath to resist cell ingrowth, a non-cellular assay method has been developed. Based upon the published valu3 that 3 ~Lm repr3ents the lower limit of cell 1;. ' ' ~ through straight pores, 3 llm ..u.,.u,~,h~,lcs can be used to ' 71~, detemline wbether a given sheath will exclude particles of this diameter. Corlsequently, any sheath material that excludes 3 llm . . ' c~ should effectively prevent cellular movement acrPss the sheath.
~ addihon to the cell exclusion property, the sheath material is pemmeable to bulk flow of l..~.,.u.ll~lL~ . The term " , ' ' " is understood tP include, but not be limited to, SUBSTITUTE SHEET (RULE 26) ~wo 95129713 2 ~ 8 6 3 7 5 molecules having a molecular weight up to ahd including about 2,000,000 MW, for example.
Suitable matenals for the sheath include, but are not limited to, pol~ hdn.. ,. h,!~,~, (PTFE), pul.~. hyL,..~, t.,.c. ' ' ' , ~ol~ u,uJh,..e, fluorinated ethylene propylene (FEP), and of the above, among others, for example. In the preferred ci ~ ' of the present ilnvention, the 5 sheath is IIII~,IU~JUIUU~, constructed of ePTFE, permeable to bulk flow of . _ ' ', , ' ' to host cell ingrowth and of sufficient strength to serve ~ as a blood vessel substitute even in the absence of the tissue it is intended to surroulnd.
The sheath is installed by wrapping a mandrel with mulbple layers of ePTFE film, such as those described in U.S. Patent 3,953,566 and 4,187,390, each issued to Gore, both of which are 10 ' . ' herein by reference, and adhering the film to itself by heating the film and mandrel in an oven at about 380C for about 10 to about 20 minutes. The film tube is removed from the mandrel and the prepared tissue graft installed in the lumen of tne polymeric tube. Other materials, or ' of materials, may be similarly applied ,ho the nssue tube using:
appropriate for the physical properties of the chosen material.
A composite structure is shown in Figure 4 where the sheath (28) has been applied to the external surface of the artery wall (26) that has been covered with a layer of recipient endothelial cells (24) forming a tubular vascular graft (30).
Other methods of " ~ the sheath material include placing the processed artery onto a mandrel and wrapping the graft with multiple layers of an ePTFE and FEP composite film, 20 and heating the film layer very bnefly from about 325C to about 350C ho a&ere the film wrap to itself on the vessel outer surface. The FEP-coated expanded PTFE film is made by a process that includes the steps of: a) contacting a porous PTFE substrate, usually in the form of a membrane or film, with another layer which is preferably a film of FEP or ' ~ of another 1l.... .,u~
polymer; b) heating the composition obtained in step (a) to a temperature above the melting point of 25 the e - polymer; c) stretching the heated composiborl of step (b) while maintaining the ~ , above the melting point of the ~ polymer; and d) cooling the product of step (c).
Another sheath construction method is the fabrication of a tubular form of ePTFEconstructed according to U.S. Patent 3,593,566 to Gore using a uniwal expansion. The processed 30 blood vessel is inserted into the ePTFE tube so constructed. As in the other forms of the sheath, the tubular form must be permeable to the passage of 111~1~ ' ' but exclude the passage of cells.

SUBSTITUTE SHEET (RULE 26) wo 95/29713 2 1 8 6 3 7 5 Construction of the shea* is described in a U.S. Patent application of Bruchman et al., entitled "Cell Excluding Sheath For Vascular Gral~s," USSN 08/235,071, filed Apnl 29, 1994, which is inw~ ' herein by reference.
The ~ ,dbll;L~ L ;` ~ ` of the sheath can be evaluated by testing with markers of 5 known size, such as dextrans and polystyrene ..u~,., . ' c~. For example, dextran, labelled with rduorescein (Sigma Chemical Co., St. Louis, MO), with an average molecular weight of about 2,000,000 MW can be used to test the ability of the sheath to pass ~ ' Cell , of the sheath can be tested using polystyrene .' c. a'~ , Inc., Warrington, PA), with a diameter of about 3 !lm, at a of about 2.5% solids in 10 suspension, for example.
When testing the sheath for ~ , with colored markers, the markers are suspended as an aqueous solution or suspension at su~ficient to provide a distinctly vjsible cwlor. Preferred of the cwmpound of the marker solution is r~ 0.2 mg dextran/ml solution and ~,., '~ 0.02 ml , .' C~/llll suspension, to yield about 4.5 x 15 10~ beads/ml suspension. The evaluation of the sheath ~ occurs at about 23C. The sheath is prepared for p. ' ' ~ testing by rendering it permeable to water, if necessary. For exarnple, sheaths constructed of ePTF~ are wetted with 100% etharlol and then f~ushed with water to remove the alwhol before testing.
To test the ~ of the sheath to l . ' ', the dextran test solution is 20 instilled in the lumen of a sheath and pressurized to about 20.7 kPa using a syringe. The contents of the syringe are forced through the sheath and the liquid that filters through the sheath wall is collected and visually inspected agarnst a white background for evidence of the colored dextran. In preparation for testing the ,~ of a sheath for cells, the nurnber of l....,., . ' ~ in the suspension are detemiined by using an appropnate wunting device, such as a h~,.,.~.ll~., for 25 example. The ~ of the sheath is then tested by forcing the l...,,., .' ~, ~ gsuspension through t*e sheat* material at about 20.7 kPa usrng a syringe. The synnge is refilled with water and the water also forced through the sheath at about 20.7 kPa The liquid that filters through the sheath wall is collected and subjected to c-- ;r,.~,Al;.... at about 300 x g for about 10 minutes. The supematant is decanted and discarded and the pellet of ., ' ~ is ~c~u~ .dcd 30 in a known volume of water. The number of ~ in the ~c~u~ .d~,l pellet are counted ~W0 95129713 2 1 8 6 3 7 S r~ 3~
and compared with the original suspension. The number of .' ~,~ that pass through the sheath are expressed as a percentage of the number introduced into the lumen of the sheath.
An appropriate sheath for this application will pass the about 2,000,000 MW dextran at pressures at or below about 20.7 kPa so that the soluhon that filters through the sheath is visibly - 5 colored when viewed against a white background. Additionally, the sheath will not allow more than about 5% of the 3 ~m J~ a to pass at a pressure of about 20.7 kPa Preparatiorl ~ I Use of Perived ~ thP~ r ~
As previously described, a key element of the present invention is the use of the blood vessel ~ ' matrix layet as a substratum upon which endothelial cells are seeded or grown. The invention is practiced preferably by leaving the preserved ~ '...,.1"1..1.-1 matrix layer in situ on the donor vessel, seeding the matrix surface with endothelial cells, and grafting the donor vessel intact. An altemative ~ ~ ' ' is the removal of the I ' ' ' ' matrix layer ftom the 15 surfæe of the donor ~,~u d;v . ~,ulaJ tissue and the re-applicahon of this ' - ' ' ' ' matrix layer to a synthetic component, followed by the applicahon of endothelial cells to the ~ ' ' ' ' ' matrix.
The first step tn the process of applying a _L ~' ' ' matrix layer to a syntheiccomponent is the removal of the . ' ' ' firom the donor tissue. The ~ ' ' ' ' matrix 20 layer is then ' "~ removed from the donor tissue using, for example, a scalpel blade. Only that layer of ~ -- matrix superficial to the intemal elasic lamina is removed. The intæt intemal elastic lamina is left attæhed to the donor tissue.
A number of methods of attachrnent of the ~ ' ' ' ' ' matrix layer to the synthetic component are possible. One preferred method of attachment of the ~. ,l .. I. h- I -~ matrix layer to 25 the synthetic component is to ' "~ entrap the derived - ~ h 1 -l layer onto the synthetic component. For example, the ~uu. '( ' ' derived from the luminal surface of bovme ærtae is suspended in E~BSS. An e~TF~ vascular graft is wetted m 100% ethanol after mounting on syringe fittings and the solution containing the ~ub. ' ' ' ' matrix suspension forced through the mtershces of the IIII~IU,UUIU ' B using syringe pressure. As the soluhon is forced through 30 the graft, the fragments of the _' ' ' ' ' matrix are left behind entrapped in the interstices and WO 95129713 2 1 ~ 6 ~ 7 5 on the surface of the graft. This applied I ' ' ' ' matrix serves as the substratum for the application of recipient endothelial cells.
Preparation ~ Jse of In V!tro c I " ~ 1- .~1..1; 1 r Another important source of ~..1.. I.,1'.. l;-l matrix for use in the present invenhorl is to create such a matrix in ~itro. The source of - ' ' ' ' ' matrix for use in this c I ' of the present invention is based upon in ntro hssue culture methods. While the precise constituents producing ' . ' _ ~ are not yet fully understood, the following technique may be suitable 10 for producing the matrix.
The present invention uses an anadogue of the - -1.. 1.. ~1.. 1:,1 matrix, synthesized with living vascular cells in vitro, to provide a substratum for seeding recipient ECs on synthehc materiads. The steps of one ~ ~ ' of this process are illustrated in Figures 5 through 7. The generation of the in vi~ro ~ul. ' " ' ' matrix appears to be the natural result of interaction 15 between the endothelial cells and the smooth muscle cells. This I ' ' ' ' matrix is preferably produced by first culturing a biological substrate cell layer on a synthetic base material for a period of time. It is preferred tbat the substrate cell layer is composed of smooth muscle cells (SMCs), and most preferably vascular smooth muscle cells (VSMCs). Once the substrate cell layer is created, it is seeded v~ith endothelial cells (ECs), preferably of vascular origin, to re-establish a 20 w~ J~d vascular cell relahonship shown in Figure 5. This structure consists of SMCs (20) grown on a synthetic base material (32) to which ECs (12) are seeded.
Following a culture penod of sufficient time to allow both the SMCs and ECs to synthesize an in situ I ' ' ' ' matrix (14) as shown in Figure 6, the ECs are specifically removed in such a marlner as to leave the ' ' ' ' ' matrix layer (14) intact overlying the SMCs (20) on a 25 synthetic component consisting of a synthetic base material (32) as shown in Figure 7. This composite graft may then be stabilized by ~,' ' ' ', Jc fixation to minimize c. '!~ and preserve the ,..1.. ---1..11..1 l matrix layer.
The initial step in this process involves the preparation of a synthetic base materiad to support the SMC substrate layer and provide mechanicd integnty. The preferred base materid 30 consists of a synthetic porous, expanded ,~.ul~; n u~ c (e~rF~) graft material, such as those , '1,~ available from W. L. Gore & Associates, Inc., Flagstaff, AZ, under the ~ wo 951297 13 2 1 8 6 3 7 5 A ~
designation GORE-TEX~iD Vascular Graft. The 4 mm intemal diameter vascular grafts used in the following descriphon are , '1~ available product obtained from this source. The 2 5 mrn intemal diameter ePTFE tubing used in following description were constructed from CD 123 fine powder PTFE resin (ICI Americas) as taught in U.S. Patent 3,953,566 to Gore, which is 5 i ~ ~ hereirl by reference. The tubes were expanded by stretching to produce a mean fibril length of about 28 ~m. A fibril length of less than about 60 ,um is preferred for this application.
The finished tubes had an intemal diameter of about 2.5 mm and a wall thickness of about 0.33 mm. Other suitable synthetic base materials may include, but not be limited to, the following:
PTFE, PLI~ JI~, t~.l,''' ' ', pUl~ u~Jl~ PL~I~. I'' , and pul~.' 'yl siloxane, for 10 example.
The fibril length of the porouS expanded PTFE tubes produced as above is defined herein as the average of ten L..~.~lt~ between nodes connected by fibrils in the direction of expansion Ten I~ ts are made in the following manner. First, a I ' ug.a,ull is made of a I ~i,UI ' ' ' . ~ portion of the sarnple surface of adequate _ ~ to show at least 15 five sequential fibrils within the length of the photograph. Tw~ parallel lines are drawn across the length of the I ' ^ ut,.~l. so as to divide the photograph into three equal areas with the lines being drawn in the direction of expansiûn and parallel to the direction of orientation of the fibrils.
Measuring from left to right, five U..l~tb of fibril leng~ are made along the top line in the photograph beginning with the first node to intersect the line near the lefl edge of the photograph 20 and continuing with consecutive nodes intersecting the line. Five more are made along the other line from right to lefl beginning with the first node to intersect the line on the right hand side of the photograph. The ten II.C~ . ' obtained by this method are averaged to obtain the fibril lengdl of the material.
The synthehc base matenal suitable for the present invention was further prepared in the 25 following manner. ~'. , "~, available 4 mrn diameter (~ORE-TEXIID Vascular Grafls and ePTFE~ tubing measuring about 2 5 mm inside diameter were cut to about 7 cm lengths and synnge fittings were tied to both the proximal and distal ends of the grafls Each graft was then mounted in a stainless steel wire holder and a plug inserted into the connector at the distal end of the grafl.
Afler steam StP-ili7~tinn~ the grafls were prepared for cell-seeding by wetting the nommally 30 hJ.llu~h~;~, el~lFk with 100~ ethanol. The ethanol in the graft interstices was djsplaced with about 80-100 ml of Manks' Balanced Salt Solution (E~BSS) (Gibco BRL, Grand Island, NY) using WO g5/29713 2 1 8 6 3 7 5 p~""~ - ~
a syringe attached to the proximal connector. Wetted grafts were stored in HBSS until used for cell-seeding.
Once the synthetic base material tube has been prepared, SMCs are applied to the luminal surface of this supporting structure. The preferred method is the use of posihve pressure to force 5 the media through the graft wall depositing SMCs onto the base matenal luminal surface. Other suitable means for applying the SMCs to the base material may include, but not be limited to: filling the base material tube lumen with a SMC suspension followed by a series of graft rotations to allow the SMCs to settle onto the surface uniformly; using negative pressure to draw the SMCs onto the base material; and using chemotactic agents, for example.
In one experiment, for example, the VSMCs were procured and applied to the synthetic base material in the following manner. Vascular SMCs were isolated by placing about 3-4 cm segments of carotid or femoral arteries obtained from greyhound dogs into a tube containing cold, sterile Medium 199 and 50 llg/ml gentamicin (Gibco BRL). ln a laminar f~ow hood, the artery segment was slit I ,, ' '~ and the endothelial cells were removed by first rubbing the luminal 15 surface with a sterile paper towel followed by scraping with a #10 scalpel blade. Thin strips of arterial media were peeled up wi& forceps and pooled into a puddle of HBSS in a sterile Petri dish.
The strips were then placed into 25 cm2 tissue culture flasks containing about 1.5 ml Smooth Muscle Cell Growth Medium (SMCGM) (43% Dulbecco's Modified Eagle Medium (DMEM);
43% Medium 199; 13% fetal bovine serum; 2 mM glutarnine; 15 units/ml heparin; 23 ~Lg/ml 20 gentamicin; and 12.5 llg/ml endothelial cell growth supplement (Collaborabve Biomedical Products, Bedford, MA)). Culture medium in the flasks was replaced when sigluficant outgrowth of cells from the tissue pieces was observed. Twice weekly, the cells were then fed about 3-5 ml of the SMCGM, depending on the number of cells in the T-25 flask. The cells were generally passaged when about 60-90% confluent, and were usually split 1:4. Smooth muscle cell type was 25 confirmed by ~ g ' critena, positive staining for alpha smooth muscle cell actin, and lack of uptake of acetylated low density lipoprotein which would indicate EC
For graft-seeding purposes, ~ n ' VSMC cultures (passages 3-15, for example) were rinsed briefly with ' ~ free-HBSS (CMF-HBSS) and washed in CMF-HBSS for about 5 minutes. Cells were harvested using trypsin c;h~' ' ~ tetraacetic 30 acid (EDTA) to release cells from the flask, followed by trypsin -~ ' with SMCGM.
Cells were pelleted at about 300 x g for about 5 minutes and the pellet re-suspended in SMCGM

~wogs/2s7l3 21 8 6 3 7 5 P~
for cell countmg. After ~ '` v , the cell pellet was rc ~ ,l in SMCGM at a final4~ of about 2.5 - 6.0 x 106 cells per 6-8 ml and trallsferred mto a syringe in preparation for graft seeding. Grafts having an intemal diameter of about 2.5 mm were seeded with about 2.5-3.5 x 106 cells~.7 cm graft in about 6 ml SMCGM, for example, and 4.0 mm internal diameter 5 grafts were seeded with about 4. 0 - 6. 0 x I o6 cells~.7 cm graft in about 8 ml SMCGM, for example.
Cell numbers were quantitated using a l..,.ll~ytuA..~
Graft seeding was performed by attaching the SMC-contau~ing syringe to the proximal connector of the wetted graft and gently forcing the cell suspension into the graft and the media through the base material graft wall. The proximal fitting was then plugged and the seeded vraft 10 placed into a 16 mln culture tube filled with SMCGM with the graft wedged in the culture tube to prevent it from rolling in the tube. The culture tubes were capped securely and placed into an incubator at about 37C on a roller apparatus tummg at about 10-50 revlhr. The medium in the culture tubes was replaced at least twice weekly and grafts were cultured for a minimum of about ten (10) days before further processmg or the addition of endothelial cells.
A layer of endothelial cells (12) attached to a substratum of smooth muscle cells (20) adhered to a synthetic base matenal (32) is shown in Figure 5. This structure may be created in a variety of ways, with the preferred method bemg generating a single cell suspension of endothelial cdls, filling graft lumen with the suspension, and allowing the endothelial cells to attach and spread on the SMC surface. Altemabvely, small patches of endothelial cells may be directly harvested 20 from a donor vessel and seeded into the graft lumen whereby they will attach and proliferate to cover the SMC substratum layer.
One method shown to be effective in isolating endothelial cells (ECs) was by using enzymatic methods to release ECs from arterial or venous vessels obtained from dogs. The vessel lumina were cannulated in a laminar flow hood, rinsed with H~SS, and filled with an endothelial 25 cell harvesting enzyme solution in CMF-HBSS for about 15 minutes at about 37C, for example.
Suitable enzymes include, but are not limited to, collagenase, dispase, and trypsin. Endothelial cells were flushed mto a sterile centrifuge tube and the ECs pelleted at about 300 x g for about 5 minutes. The cells were then plated onto T-25 hssue culture flasks, grown at about 3rc until nearly conf~uent and then passaged. Endothelial cell type was confimmed by ~ cnteria, 30 by positive stainmg for Factor vm, and by uptake of acetylated low density lipoprotein SUBSTITUTE SHEET (RULE 26) WO95/29713 21 86375 P~
For graft-seeding purposes, C..h n endothelial cells (passages 2-10, for example) were rinsed briefly with CMF-HBSS and washed in CMF-HBSS for about 5 minutes. The cells were harvested by using trypsin-EDTA to release cells from the flasks followed by trypsin with complete Endothelial Cell Growth Medium (ECGM, 80% Mediurn 199, 16%

6 fetal bovine serum, 2 mM glutamine, 15 units/ml heparin, 25 ~g/ml gentamicin, 12.5 llg/ml Endothelial Cell GrowLh Supplement (Collaborative Biomedical Products, Bedford, MA)). The cells were pelleeed at about 300 x g for about 5 minutes, and the pellet ro ~ ~rPn~Pd rn ECGM at a final of about 1.1 - 1.3 x 106 cells/ml, for example. The cell suspension was transferred rnto a syringe for seeding.
A previously seeded SMC-graft, described above, was prepared for endothelial cell seeding by removing both end plugs and briefly rinsing the graft lumen with HBSS. The syringe containing the endothelial cell suspension was then attached to the proximal connector of the SMC-graft and the graft lumen filled with the cell suspension without forcing fluid through the graft wall. The syringe fittings were plugged and the grafts placed rnto 16 mm culture tubes filled with ECGM.
15 Grafts were wedged into the tubes so they could not rotate ', ' '~ of tbe culture tube during incubabon.
Once the composite structure of endothelial cells (12) and smooth muscle cells (20) on a synthetic base material (32) was created, it resembled the structure shown in Figure 5. The composite structure was placed under the following conditions that have proven successful.
20 Culture tubes were incubated at about 37C in a roller apparatus turning at about 10-50 r' 1u. The medium in the culture tubes was replaced at least twice weekly with ECGM
for the first week to establish ECs and then switched to SMCGM for the remainder of the culture period, a minimum of about 10 days total culture bme.
Following culturing, a layer of I ' ' ' ' matrix (14) will form between the endothelial 25 cells (12) and the smooth muscle cells (20) m the manner shown in Figure 6. Generally the - ' - - ' ' ' ' matrix is less than about one llm in tbickness.
Once a suitable layer of ' ' ' ' ' matrix is created, the endothelial cell layer is removed tbrough one of the following processes. Preferably, the endotbelial cells are stripped by rmsing the graft about three times with BSS, and treating with a 0.025 M ammonium hydroxide 30 solution for about 4-4.5 minutes to remove the endothelial cells, and rinsed again about tbree times in HBSS. Other suitable treatments may include 0.01-0.5 M NEI4OH for about 30 seconds to ~wossl2s7l3 21~3637~ r~ J~-~
about 60 minutes, for example. Other possibly effective methods of removing the endothelial cells may include air drying, or treatment ~vit.'. other stripping solutions, for example, chloroform, methanol, ammonium hydroxide, or sodium chloride, either alone or in, ' The ~,hlu~urv~ treatment may also be employed to reduce potenbal calcification of O' ' ' ~de fixed cellular material as well. Other treatments known to those skilled in the art may also be suitable. Once the endothelial cell layer is rernoved, the structure resembles that sho~-vn m Figure 7, Scanning electron microscopy inspection of ~ a~ ll;v~ samples treated in the above manner conflrmed near total loss of the endothelial cells and the l of the ~ '.. l" h 1: .1 10 matrix layer. It is understood that for purposes ofthe present irlvention, removal of endothelial cells from the ~ l matrix layer in amounts greater than about 80% is considered to render the ' matrix layer ' "~ free of donor endothelial cells.
The ' ' ' ' ' matrix analogue exposed after removal of the endothelial cells contains numerous ~ " ' matrix , Usrng ~tv.,l.. ,.. ,~l assays, it has been 15 determined that the matrix includes chondroitin sulfate ~ tuv51~. , collagen L collagen m, cûllagen IV, elastin, and fibronectin all of which are present both on the exposed blood contact surface as weU as within portions of the graft wall. In addition, laminin is an abundant component of the luminal a l ---1 ~ l matrix blood contact surface and is present to a lesser degree within the graft wall Following EC removal, the grafts are optionally treated v~ith a fixative. The purpose of this step is to preserve the ! ' ' '' '' ' matrix layer, to reduce ,, - ~, and to sterilize the graft. 'rhis f xation can be ~ by placing the graft into a fixing solubon, such as, for example, 0.1-2.5% ,,' - ' ' ' Jl-., in a suitable buffer for about l-72 hours depending upon the r~ of .,' ' ' ' JV~ used. Suitable buffers may include, but are not limited to, N-2-2~ hJ~UA~ ' J!p r N'-2-ethane sulfonic acid (HEPES), acetate, 2-(N I -;" ~r '- acid ~S), 3-~rN-morpholino] l , r ' acid (MOPS), tris llydlv ' ~, phosphate, and others.
In the preferred .~ - ' ' t, the graft is fixed in greater than about 0.5% 0~ - _' ' bj de in 20 mM HEPES buffer for a minimum of about two hours. The fixed grafts are rinsed at least three 30 times in sterile normal saline and washed for a minimum of about 24 hours in fresh stenle normal SUBST~TUTE SHEET (RULE 26j WO9~29713 Zl 86375 ~ C.
saline and then stored at about 4C in firesh stenle normal saline. Other suitable fixatives, such as dl" for example, may be used in addition to ,,' ' ' J 11; to assure sterility.At this stage, remaining aldehyde reactive sites are blocked with an amino group containing solution. Suitable reagents for this procedure include, but are not limited to, 0.1 M glycine, 5 Medium ]99, Dulbecco's Modified Eagle Medium, and other ~),L..~ :Jlo&.~l culture media, for example. The graft is incubated in one of these solutions for a minimum of about 12 hours at about 37C. Once the blocking step has been completed, the &raft is placed into ECGM and incubated for several hours to equilibrate the vessel segment to the endothelial cell culture medium. Single cell suspensions of endothelial cells are then placed into the vessel lumen, both ends of the vessel 10 plu&ged, and the graft placed into rotation culture to allow the endothelial cells to forrn a conf~uent monolayer on the surface as shown in Figure 8. A structure resembling that shown in Figure 8 is created with recipient endothelial cells (24) resting on an in vi~ro ~ ~1,....I.,~h. 1 ~l matrix (14). The endothelial cells may be arterial or venous in ongin and it is preferred that these cells be derived from the recipient to prevent immune responses.
It is also possible to produce the s ' 7 '~ ~' I matrix layer throu&h a variety of other methods. One suitable method, for example, involves using mixed culture seeding in which both ECs and SMCs are combined in rabos of about 1:10 to 1:1 (EC:SMC) and both cell types are seeded onto the synthetic base material ' _ ~,. Once placed mto culture, the ECs will form a confluent monolayer on the luminal surface thereby re-establishing the normal EC and SMC
20 l, ' ', After extended co-culture, the - L_ ' '' '' ~ matrix will be produced between the cell layers. The - . .l ,. .. l. ,l' .. ! ~l matrix layer may then be exposed as outlined above and processed accordingly.
While VSMCs are the preferred use as the substrate cell layer in the production of the s ` ~ ~ ' ' ' ' matrix layer of the present invenhon, it is reco&nized that it may be possible to use 25 other cell types to provide a similar function. Among the other potential cell types are smooth muscle cells from the digestive system or urinary tract, as well as fibroblasts, among others, for example.
Additionally, while vascular endothelial cells are the preferred ~ , it may also be possible to use other cell types to produce the ' ' ' ' ' matrix layer in conjunction with the 30 VSMCs, SMCs, fibroblasts, or other similar cell tvpes. These cell types may include, but are not ~W095129713 21 8 6 3 7 5 r~
limited to, ~ UV~ endotheiiai cells, comeai; ' ' ' , glomeruiar epithelium, and mesotheliai cells amon~ others, for example.
While the present invention is particuiarly applicable to blood vessels and similar appliances, it may be applied to many areas where ~ lUliflilUll~lJUt;L blood contact 5 surfaces must be provided. Examples of ûther appliances which may be produced in accordance with the present invention include, without limitation, flexible sheets as shown in Figure 9, heart vaives as shown in Figure lû, artificiai hearts, artificiai or~ans such as implantable artificiai i~idneys, and others.
Without intending to iimit the scope of the present invention, the foiiowing examples 10 illustrate how the present invention can be made and used.
EXAMp! .F.~
EXAMPLE I (Preserved S ' ~ ' " ' Seeded with Endotheliai Cells) Pairs of canine carotid arteries were coiiected into a sol~tion of HBSS containing 15%
DMSO and flash frozen in liquid nitrogen. The artery segments were thawed, the lumen flushed v~lith HBSS or nomui saiine, and the adventiha was removed by dissection. Both arteries were then everted and placed onto a 2.5 mm mandrel. The inventive artery, having an intact ' ' ' ' without endotheiiai cells, was then removed fro2n the mandrel and retumed to its naturai position. The control artery had the ' ' ' ' removed by rubbing the endotheliai surface of the a~tery with a cotton swab. The artery was then rir sed in HBSS and retumed to its naturai position for further processing. Both inventive and control arteries were subjected to further processing in parailel.
The arteries were each sleeved with a GORE-TEX~ Vascuiar Gra~ of a diameter sufficient to produce a fixed graft ûf about 2.5 mm in diameter (usuaily 3 .0 mm). The sleeved arteries were blotted dry and immersed in lû0% medicai grade ethanol. The arteries were pressure fixed in 100% ethanol for 2 hours at 23C under about 27.6 i~Pa pressure After fixation, the artery segments were sterilized by ovemight incubation in 70% ethanol. Following ' , the artery 30 segments were rehydrated in a graded ethanol series into HBSS. The artery segments were then placed into Medium 199 for about 4-5 hours at 37C, followed by an overnight incubation in fresh SUBSTITUTE SHEET (RULE 26) WO 95/29713 21 8 6 3 7 5 r~
Medium 199. The segments were finally incubated 4-5 hours in ECGM at 3rc. Autologous venous endothelial cells were harvested by ~ ..,...uu., and .~u~,u~ .,J in ECGM at 1.1 x 1o6 cells/ml. The invenhve and control arteries were filled with the cell suspension, the ends ûf each rlrtery were plugged, and the artenes were incubated in culture tubes contauling ECGM at 37C.
5 The day following seeding, the artery end plugs were removed and grafts cultured until the endothelial cells reæhed conduency on the artery luminal surface (generally greater than 7-10 days). The . ' ' ' of an endothelial cell monolayer was confirmed by staining with acetylated lûw density lipoprotein and scanning electron microscûpy.
One day prior to . ' ' " , the grafts were placed into Endothelial Serum Free Medium 10 (Gibco BRL, Grand Island, NY). Thir~y minutes prior to . ' , the grafts were rinsed thoroughly in sterile 37C BSS and stored in warm HBSS until implanted. These ~ ' ' ' ' inventive and control grafts were implanted as test pairs into the brachial arteries of an adult greyhound dog. Each graft of this pair was implanted into one brachial artery so that a direct comparison of the gral`ts within the dog was achieved. The vessel grafs, 2.5 cm long with an 15 intemal diameter of 2.5 mm, were a good diameter match to the native brachial artery ofthe adult greyhound recipient. Standard end-to-end surgical technique was used fûr . l and no or anhplatelet agents were ' cd to the dogs at any time. The course of the implans was followed using duplex ultrasound: At 20 days ~ J1 the brachial implans were surgically exposed and the patency tested by transecting the artery distal to 20 the implant location. The inventive graft having autologous endothelial cells grown on a preserved ' matrix was widely patent and the control graft having autologous endothelial cells grown on a vessel without a preserved s ' ' ' ' was occluded. Thus, the use of a preserved ~( ' ' as a substratum for endothelial cells provided improved patency.
EXAMPLE 2 (In Vitro c ~ ' ' ' ' Matrix Seeded with Endothelial Cells) A vascular smooth muscle Cdl~ cell matrix invenhve graft fabncated as described in the detailed description above was prepared resulting in a base graft having a fixed, g~ d ~ ' ' ' ' matrix. The inventive graft was placed into Medium l 99 for a 30 minimum of 12 hours at 37~C to block unreacted aldehyde groups followed by incubation in Endothelial Cell Growth Medium (ECGM). Autologous venous endothelial cells were then seeded ~wo 95/29713 2 1 8 6 3 7 ~
onto the graft ~,lb. 1 ' ' ' matrix surface by filling the graft lumen with a suspension of endothelial cells at a of I . I - 1.3 x I o6 cells/ml of ECGM. The graft was then wedged into a culture tube and placed into a 37C incubator on a roller apparatus for at least 10 days to allow the endothelial cells to establish a monolayer and become firmly attached.
5 Endothelial cell coverage was confimmed by observing uptake of acetylated low density lipoprotein and scanning electron microscopy.
These ~ ' " ' ' ' matrix invenhve grafts were implanted as test paurs usung control 30 ~Im ePTFE grafts into the brachial artenes of 2 adult greyhounds. Each pair was implanted into the brachial arteries of one dog so that a direct comparison of the vessels within each dog was 10 achieved. The vessel grafts, 2.5 cm long with an intemal dia~neter of 2.5 mm, were a good diameter match to the nabve brachial artery of the adult greyhound recipients. Standard end-to-end surgical technique was used for , ' and no " _ ' or antiplatelet agents were ' ' ' tJ to the dogs at any time. The course of the implants was followed using contrast and duplex ultrasound: In both dogs, the inventive and control grafts 15 were both patent at 74 and 27 days, respectively.
EXAMPLE 3 (Mixed SMC-EC In Vitro S ' .. - l. ,11,. 1; -I Matrix Seeded with ECs) Individual cultures of vascular SMC and EC were harvested for graft fabrication by 20 removing growth medium, rinsing the cells with CMF-HBSS, followed by washing the cells in CMF-HBSS for 3-~ minutes. The CMF-~SS was then removed and depending on flask siz4 1.5-3.0 ml of trypsin-EDTA was added to release cells from the flask. Complete SMCGM was added to flasks to inactivate the trypsin, the cells were triturated, and pelleted by, r _ '' at 300 x g for 5 minutes. The supematant was discarded, the cells were ~t .~ ...d.,J in SMCGM and 25 cell counts were carried out, followed by a second pelleting of cells. The final individual SMC and EC cell pellets were r~ I in SMCGM at a of I x I o6 cells/ml, and then cell types mixed in EC:SMC ratios of 1:9, 1:4, and 1:2.
Several porous ePTFE tubes of 2.5 mm intemal diameter, having about 30 ~m fibnl lengths, were cut to 7 cm and mounted onto syringe connector fittings. A pcl~ , plug was 30 fitted onto the distal end of each tube. The ePTFE tubes were wetted with 100% ethanol and the alcohol ' . ~ displaced with HBSS by E~ ,, the tubes, thereby forcing E~3SS

~095ng713 21 86375 F~~
through the porous tube walls. The wetted tubes were then stored in HBSS until seedirlg with the vascular cells.
CellseedingofeachePTFEtubewascarriedoutbyplacing2.5-3.5xlO6cellsofthe mixed SMC-EC suspensions into a total volume of 6-8 ml SMCGM in a syr~nge. The 2.5 mm 5 intemal diameter graft was seeded with 2.5 to 3 .5 x I o6 cells/graft. For each graft, a syringe containing the cell suspension was attached to the proximal sy~inge connector arld the slur~y was gently injected into the graft lumen. After placing cells into each graft, a second pol~.v~"~k,..
plug was attached to the opGn connector to seal the cells in the graft lumen.
Each seeded graft was wedged into a 16 mm culture tube filled with SMCGM to ensure 10 complete rotation ofthe graft during incubation. The culture tubes were capped securely and placed in a roller apparatus tuming at about ~ 0 rev/hr. in a 37C irlcubator. Seeded grafts were cultured for 7- 10 days with fresh medium feedings every 2-4 days. During this culture period, the endothelial cells were segregated from the SMCs forming a cor 'luent EC monolayer on top of the SMCs. This was verified by both scarlning electron microscopy analysis and by stairling with 15 acetylated low density lipoprotein to visualize ECs on the graft luminal surface.
The B were then rinsed three hmes with ~BSS, and the lumen treated with 0.025 M
NH4OH for 4.5 minutes to remove the ECs. The grafts were rinsed with HBSS, fixed with 0.25%
O'a~ d~ for 24 hours at 23C, followed by extensive washing in sterile normal saline.
The grafts were then incubated in Medium 199 ovemight at 37C, followed by incubation 20 in ECGM for 4-6 hours to block and neutralize free and unreacted O'l ' ' ', i~ groups. The luminal space of each graft was filled with a suspension of ECs at a seeding density of I . I x I o6 cell per ml. The grafts were wedged into culture tubes and incubated on a roller apparatus at 10 RPM at 37'C for 7-14 days to establish the endothelial cell monolayer. ~nnfirn~otmn of EC
coverage was carried out by scanning electron microscopy analysis and staining with acetylated low 25 density lipoprotein to identify the presence of endothelial cells EXAMPLE4 (Derived ~ ' ' ' ' SeededwithEndothelial Cells) A ~,UIIh..~ J available 4 mm GORE-TEXiD Vascular Graft was modified through the 30 addition of 5111~ 1' derived from a donor cow aortae. The veSsel was obtained at slaughter and frozen at -20C until use. After opening with a longitudinal incision, the intimal surface was ~wo 9~/29713 2 1 8 6 3 7 5 P~
gently rubbed with a paper towel to ensure removal of endotbelial cells remaining after the freezing-thawing cycle. Then, using a scalpel blade, tbe inhmal surface of the donor vessel was carefully scraped, taking care to not penetrate the intemal elastic lamina. The ~"l .. l.. 1l .. 1; -I matrix derived in this manner from ~ 5 cm of cow aorta was used for a single 4 mm x 5 cm 5 rnventive graf~ The scrapings from the cow aorta were suspended in about 50 ml HBSS and subjected to I ~, with a Verti-shear mixer for about 60 seconds. The I -~- ~ ' was allowed to sit for about 10 minutes to allow the larger clumps to settle out and the remainder ofthe I ,~ decanted. The suspension was placed in a syringe and the inventive graft mounted on barbed fittings. The 4 mm GORE-TEXa9 Vascular Graft was wetted 10 with 100% ethanol and the etharlol displaced with EIBSS. An aliquot containing the scrapings was tben forced tbrough the pores of the graft, depositing the ' ' ' ' ' ~ . l. r- ~Il- ~I ~ matrix on the surface of the graft. The grat't was sterilized in 70% ethanol ovemight and rehydrated in a graded ethanol series irlto HBSS. The graft was then incubated in Medium 199 ovemight followed by ECGM to equilibrate the graft to the culture medium. Endothelial cells were harvested and seeded 15 onto the graft ' ' ' ' ' surface by filling the graft lumen with a suspension of endothelial cells at a of I . I - 1.3 x I o6 cells/ml of ECGM. The graft was then capped at both ends, wedged into a culture tube, and placed into a 37~C mcubator on a roller apparatus for 7-10 days to allow the endothelial cells to establish a monolayer and become firmly attached. Endothelial cell coverage was confirmed by observing uptake of acetylated low density lipoprotem.While particular ~. I ' of the present invention have been illustrated and described herein, the present invention should not be limited to such illustlations and ' , It should be apparent that changes and, , r ' may be . ' and embodied as part of the present invention within the scope of the following claims.

Claims

The invention claimed is:

1. A blood contact surface which comprises:
a substratum comprising a subendothelial matrix layer that supports attachment and growth of applied cardiovascular endothelial cells;
a layer of applied cardiovascular endothelial cells attached to the subendothelial matrix layer;
wherein the cardiovascular endothelial cells serve as a direct blood contact surface; and wherein, in the absence of cells attached to the substratum, the substratum serves as a direct blood contact surface.

2. The blood contact surface of claim 1, wherein the subendothelial matrix layer is preserved.

3. The blood contact surface of claim 1, wherein the substratum is located in situ in a donor blood vessel

4. The blood contact surface of claim 3, wherein the subendothelial matrix layer is extracted from a donor blood vessel and applied to a synthetic component, the subendothelial matrix layer serving as a direct blood contact surface.

5. The blood contact surface of claim 4, wherein the subendothelial matrix layer is preserved.

6. The blood contact surface of claim 3, wherein the donor blood vessel is surrounded along its length by a sheath; and wherein the sheath is made of microporous polymeric material that is resistant to cell ingrowth, while being permeable to macromolecules.

7. The blood contact surface of claim 6, wherein the sheath is made of microporous polymeric material selected from a group consisting of polytetrafluoroethylene, polyethylene terephthalate, polypropylene, and fluorinated ethylene propylene, either alone or in combination,

8. The blood contact surface of claim 6, wherein the sheath is made of a combination of multiple layers of expanded polytetrafluoroethylene and fluorinated ethylene propylene

9. The blood contact surface of claim 1, wherein the subendothelial matrix layer is produced by in vitro tissue co-culture of endothelial cells and smooth muscle cells.

10. The blood contact surface of claim 9, wherein the endothelial cells are obtained from cardiovascular tissue.

11. The blood contact surface of claim 9, wherein the smooth muscle cells are obtained from cardiovascular tissue.

12. The blood contact surface of claim 9, wherein the subendothelial matrix layer produced by in vitro tissue co-culture of endothelial cells and smooth muscle cells is applied to a surface of a synthetic base material upon which a direct blood contact surface is to be established to form a subendothelial matrix layer thereon.

13. The blood contact surface of claim 12 wherein the synthetic base material comprises a tubular structure.

14. The blood contact surface of claim 12 wherein the synthetic base material comprises at least a portion of a heart valve.

15. The blood contact surface of claim 12 wherein the synthetic base material comprises a flexible sheet.

16. The blood contact surface of claim 12, wherein the synthetic base material is made of polymeric materials selected from a group consisting of polytetrafluoroethylene, polyethylene terephthalate, polypropylene, and fluorinated ethylene propylene, either alone or in combination.

17. The blood contact surface of claim 12 wherein the synthetic base material is expanded polytetrafluoroethylene.

18. The blood contact surface of claim 9, wherein the in vitro tissue co-culture of endothelial cells and smooth muscle cells is conducted on a surface of a synthetic base material upon which a direct blood contact surface is to be established.

19. The blood contact surface of claim 18 wherein the synthetic base material comprises a tubular structure.

20. The blood contact surface of claim 18 wherein the synthetic base material comprises at least a portion of a heart valve.

21. The blood contact surface of claim 18 wherein the synthetic base material comprises a flexible sheet.

22. The blood contact surface of claim 18, wherein the synthetic base material is made of polymeric materials selected from a group consisting of polytetrafluoroethylene, polyethylene terephthalate, polypropylene, and fluorinated ethylene propylene, either alone or in combination.

23. The blood contact surface of claim 18 wherein the synthetic base material is expanded polytetrafluoroethylene.

24. The blood contact surface of claim 1, wherein the substratum contains at least one protein in a group consisting of chondroitin sulfate proteoglycans, fibronectin, collagen Type I, collagen Type III, collagen Type IV, and elastin.

25. The blood contact surface of claim 1 wherein the surface is formed through the following process:
providing a donor blood vessel comprising an endothelial cell layer attached to a subendothelial matrix layer attached to a smooth muscle cell layer;
stripping the endothelial cell layer from the subendothelial matrix layer of the donor blood vessel to produce a subendothelial matrix substratum substantially free of endothelial cells;
applying cardiovascular endothelial cells to the substratum.

26. The process of claim 25 which further comprises:
stripping the endothelial cell layer from the subendothelial layer by treating with a stripping solution.

27. The process of claim 26 wherein the stripping solution is an aqueous solution of ammonium hydroxide at a ranging from about 0.01M to about 0.5M.

28. The process of claim 25 that further comprises:

stripping the endothelial cell layer by freeze-thaw treating the endothelial layer.

29 The process of claim 25 which further comprises:
stripping the endothelial cell layer by air dry treating the endothelial layer.

30. The process of claim 25, which further comprises:
removing the subendothelial matrix layer from the donor blood vessel and applying the subendothelial matrix layer to a synthetic component, the subendothelial matrix layer serving as a direct blood contact surface.

31. The process of claim 30, which further comprises:
preserving the subendothelial matrix layer.

32. The process of claim 25 which further comprises:
treating the subendothelial matrix layer and any associated vascular tissue to reduce calcification by saturating the subendothelial matrix layer and any associated vascular tissue with a solution that predominantly extracts lipid components.

33. The process of claim 32 wherein the solution used to reduce calcification is a solution of chloroform and methanol.

34. The process of claim 31 which further comprises:
treating the subendothelial matrix layer with a fixative solution to preserve the subendothelial matrix layer.

35. The process of claim 34 wherein the fixative solution is a solution of ethanol.

36. The process of claim 34 wherein the fixative solution is a solution of glutaraldehyde.

37. The blood contact surface of claim 9 wherein the surface is formed through the following process:
co-culturing endothelial cells and smooth muscle cells in vitro to establish an endothelial cell layer overlying a smooth muscle cell layer;
incubating the endothelial cell layer with the smooth muscle cell layer for a sufficient length of time to produce a subendothelial matrix layer between the endothelial cell layer and the smooth muscle cell layer;
stripping the endothelial cell layer from the subendothelial matrix layer to produce a subendothelial matrix substratum substantially free of endothelial cells;

applying cardiovascular endothelial cells to the substratum to form a cardiovascular endothelial cell layer thereon.

38. The process of claim 37 which further comprises:
stripping the endothelial cell layer from the subendothelial layer by treating with a stripping solution.

39. The process of claim 38 wherein the stripping solution is an aqueous stripping solution of ammonium hydroxide at a ranging from about 0.01M to about 0.5M.

40. The process of claim 37 that further comprises:
stripping the endothelial cell layer by freeze-thaw treating the endothelial layer.

41. The process of claim 37 which further comprises:
stripping the endothelial cell layer by air dry treating the endothelial layer.

42. The process of claim 37 which further comprises:
providing a tubular artificial vascular prosthesis having an interior surface and an exterior surface;
removing the subendothelial matrix layer from the smooth muscle cell layer; and applying the subendothelial matrix layer to the interior surface of the tubular artificial vascular prosthesis to provide the subendothelial matrix substratum.

43. The process of claim 37 which further comprises:
treating the subendothelial matrix layer with a fixative solution to preserve the subendothelial matrix layer.

44. The process of claim 43 wherein the fixative solution is a solution of ethanol.

45. The process of claim 43 wherein the fixative solution is a solution of glutaraldehyde.

46. The blood contact surface of claim 37 wherein the surface is formed by co-culturing endothelial cells and smooth muscle cells in vitro on a surface of an artificial vascular prosthesis that will serve as a direct blood contact surface to establish an endothelial cell layer overlying a smooth muscle cell layer.

47. The blood contact surface of claim 46 wherein the artificial vascular prosthesis is a tubular vascular prosthesis having an interior surface and an exterior surface wherein the in in vitro co-culture is conducted on the interior surface of the tubular vascular prosthesis.

48. A method of producing a blood contact surface which comprises:
providing a donor blood vessel comprising a layer of endothelial cells attached to a subendothelial matrix layer attached to a layer of smooth muscle cells;
treating the blood vessel so as to remove the endothelial cells while leaving the subendothelial matrix layer in situ to form a substratum;
employing the substratum as a direct blood contact surface in an appliance, in the absence of cells attached to the substratum;
growing vascular endothelial cells on the substratum to form a vascular endothelial cell layer; and employing the vascular endothelial cell layer as a direct blood contact surface in an appliance.

49. The method of claim 48 which further comprises:
employing the substratum as an appliance comprising an artificial vascular prosthesis obtained from a donor blood vessel.

50. The method of claim 49 which further comprises:
surrounding the artificial vascular prosthesis along its length with a sheath comprising a material which permits the passage of macromolecules through the sheath, while resisting ingrowth of cells from tissue surrounding the artificial vascular prosthesis when implanted.

51 The method of claim 48, which further comprises:
extracting the subendothelial matrix layer from a donor blood vessel and applying the subendothelial matrix layer to a synthetic component, the subendothelial matrix layer serving as a direct blood contact surface.

52. The method of claim 51, which further comprises:
preserving the subendothelial matrix layer.

53. The method of claim 48 which further comprises:
treating the subendothelial matrix layer and any associated vascular tissue layers to reduce calcification with a solution that predominantly extracts lipid components.

54. The method of claim 53 wherein the solution used to reduce calcification is a solution of chloroform and methanol.

55. The method of claim 52 which further comprises:
treating the subendothelial matrix layer with a fixative solution to preserve the subendothelial matrix layer.

56. The method of claim 55 wherein the fixative solution is a solution of ethanol.

57. The method of claim 55 wherein the fixative solution is a solution of glutaraldehyde.

58. A method of producing a blood contact surface which comprises:
co-culturing endothelial cells and smooth muscle cells in vitro to establish an endothelial cell layer overlying a smooth muscle cell layer, incubating the co-culture until a subendothelial matrix layer is formed between the donor endothelial cell layer and a smooth muscle cell layer by the interaction of the smooth muscle cells and the donor endothelial cells;
treating the co-cultured endothelial cell layer and the smooth muscle cell layer so as to remove the donor endothelial cells thereby forming a subendothelial matrix substratum;
employing the substratum as a direct blood contact surface in an appliance, in the absence of cells attached to the substratum;
growing vascular endothelial cells on the substratum to form an endothelial cell layer; and employing the vascular endothelial cell layer as a direct blood contact surface in an appliance.

59. The method of claim 58 which further comprises: removing the subendothelial matrix layer from the smooth muscle cell layer and applying the subendothelial matrix substratum to a surface of an artificial vascular prosthesis upon which the blood contact surface is to be located.

60. The method of claim 58 which further comprises:
co-culturing the endothelial cells and the smooth muscle cells in vitro on a surface of an artificial vascular prosthesis upon which the blood contact surface is to be located, in order to establish the endothelial cell layer overlying the smooth muscle cell layer.

61. The blood contact surface of claim 60 wherein the artificial vascular prosthesis is a tubular vascular prosthesis having an interior surface and an exterior surface wherein the in vitro co-culture is conducted on the interior surface of the tubular vascular prosthesis.

62. The method of claim 58 that further comprises:
employing endothelial cells that are vascular endothelial cells.

63. The method of claim 58 which further compnses:
employing smooth muscle cells that are vascular smooth muscle ceils.

64. The method of claim 58 which further comprises:
treating the subendothelial matrix layer with a fixative solution to preserve the subendothelial matrix layer.

65. The method of claim 64 wherein the fixative solution is a solution of ethanol.

66. The method of claim 64 wherein the fixative solution is a solution of glutaraldehyde.

67. An artificial vascular prosthesis which comprises:
a de-endoghelialized subendothelial matrix layer preserved with a fixative, the substratum serving as foundation upon which vsscular endothelial cells attach and grow to form a vascular endothelial cell layer;
wherein the vascular endothelial cell layer serves as a direct blood contact surface; and wherein, in the absence of cells attached to the preserved subendothelial matrix layer, the preserved subendothelial matrix layer serves as a direct blood contact surface.

68. The artificial vascular prosthesis of claim 67 wherein, the subendothelial matrix layer is obtained from a donor blood vessel and preserved in situ in the donor blood vessel.

69. The artificial vascular prosthesis of claim 67 wherein, the subendothelial matrix layer is produced by in vitro co-culture of endothelial cells and smooth muscle cells and applied to a surface of an artificial vascular prosthesis which will serve as a direct blood contact surface.

70. The artificial vascular prosthesis of claim 67 wherein, the subendothelial matrix layer is produced by in vitro co-culture of endothelial cells and smooth muscle cells conducted on a surface of an artificial vascular prosthesis which will serve as a direct blood contact surface.

71. The artificial vascular prosthesis of claim 67, wherein, prior to preservation of the de-endothelialized subendothelial matrix layer, the de-endothelialized subendothelial matrix layer and any vascular tissue layers externally attached to the de-endothelialized subendothelial matrix layer are treated to reduce calcification.