AU608173B2 - Acid treated polyacrylic acid grafted fluorocarbon polymer surface for cell attachment - Google Patents

Description

0 2 0 1 I a L~iilili-~_i i OPI DATE 23/03/90 APPLN. ID 40782 89 pCr AOJP DATE 9 CT M PCT/AU89/00356 INTERNATIONAL AFFLICATIUN 1'L1 W S Uti tJ uu1 tKlA UU~A luN A l 'lY (PCT) (51) International Patent Classification 4 (11) International Publication Number: WO 90/02145 C08F 259/08, 8/36 Al (43) International Publication Date: 8 March 1990 (08.03.90) (21) International Application Number: PCT/AU89/00356 (74) Agent: MAXWELL, Peter, Francis; Peter Maxwell Associates, Blaxland House, 5-7 Ross Street, North Par- (22) International Filing Date: 22 August 1989 (22.08.89) ramatta, NSW 2151 (AU).

Priority data: (81) Designated States: AT (European patent), AU, BE (Euro- PJ 0020 22 August 1988 (22.08.88) AU pean patent), CH (European patent), DE (European patent), FR (European patent), GB (European patent), IT (European patent), JP, LU (European patent), NL (Eu- (71) Applicants (for all designated States except US): COMMON- ropean patent), SE (European pztent), US.

WEALTH SCIENTIFIC AND INDUSTRIAL RE- SEARCH ORGANISATION [AU/AU]; Limestone Avenue, Campbell, ACT 2601 TELECTRONICS Published PTY. LIMITED [AU/AU]; 7 Sirius Road, Lane Cove, With international search report.

NSW 2066 (AU).

(72) Inventors; and Inventors/Applicants (for US only) STEELE, John [AU/ AU]; 7 The Carriageway, North Rocks, NSW 2151 (AU).

JOHANSEN, Oddvar [AU/AU]; 16 Darnley Grove, Mulgrave, VIC 3170 JOHNSON, Graham [GB/ AU]; 10 Moombara Avenue, Peakhurst, NSW 2210 HODGKIN, Johnathon [AU/AU]; 28 Celia Street, Burwood, VIC 3125 (AU).

(54)Title: ACID TREATED POLYACRYLIC ACID GRAFTED FLUOROCARBON POLYMER SURFACE FOR CELL

ATTACHMENT

(57) Abstract A surface for the attachment and growth of cells is prepared by first grafting polyacrylic acid chains to a fluorocarbon polymer substrate so that its weight increases by between 0.1 and 20 The surface is then treated with concentrated sulphuric acid under such conditions that will separately decarboxylate, aromatize and sulphonate an effective proportion of the grafted polyacrylic acid chains before being dried, soaked in a concentrated acid and brought to a substantially neutral pH for cell attachment and growth thereupon. The surface may also be used as a human tissue implant.

e Acid treated polyacrylic acid grafted fluomrcarcn polymer surface for cell attachnmnt FIELD OF INVENTION: This invention relates to the chemical modification of polytetrafluoroethylene (PTFE, which is sold under the Registered Trade Mark TEFLON) and other fluorocarbon polymers, to produce a surface that is a suitable substratum for the attachment and growth of adherent animal cells.

The invention also relates to chemical procedures to achieve this modification of PTFE and other fluorocarbon polymers and to procedures for the preparation of the chemically modified surfsce for cell attachment. Particular attention is drawn to the ability of PTFE modified by these procedures to serve as a substratum for the attachment and growth of human endothelial cells, and so to the potential use of these chemically modified fluorocarbon polymers and* procedures in the preparation of implantable materials including vascular prostheses and percutaneous implants.

BACKGROUND ART: Information gained during in vitro cell culture experiments can profitably be used in the design or selection of materials for use in specific biomaterials such as vascular prostheses. The attachment and growth of endothelial cells and other anchorage-dependent animal cells during in vitro cell culture requires both a suitable substratum for cell attachment and a culture medium that contains either serum, or certain purified serum proteins.

I _i i i WO 90/02145 PCT/AU89/00356 -2- Concerning the chemical nature of the substratum, cells such as endothelial cells do not attach and grow well on hydrophobic surfaces such as nonwettable polystyrene (bacteriological plastic) or on PTFE which is commonly used in vascular prosthetic grafts. On the other hand, cells including endothelial cells also fail to adhere to many hydrophilic polymers, such as the hydrogel poly-2hydroxyethylmethacrylate, polyHEMA. Cells do attach and grow on polymers where the surface is composed of microdomains containing both hydrophobic and hydrophilic regions. The use of polymers with a microdomain structure of this nature is now the state of the art in the biomedical material area the polyurethanes sold under the Registered Trade Marks BIOMER and MITRATHANE). The perfluorosulphonate ionomer which is known by the Registered Trade Mark NAFION has recently been shown to be suitable for endothelial cell attachment and growth (International patent application PCT/AU88/00368; McAuslan et al., (1988) J. Biomed. Mater.

Res., 22,'963-976; Norris et al., (1988) Clinical Materials, 3,153-162; and may also fit this generalisation, in that as only 1 in every 8 monomer units is sulphonated, the large segments of uncharged chains may allow for both hydrophobic and hydrophilic interactions.

The surface that is commonly used for animal cell attachment and growth in vitro is polystyrene, modified by one of a number of techniques to produce a surface that can promote cell attachment (tissue culture polystyrene). This L SWO90/02145 PC/AU89/00356 -3modification of polystyrene has been performed by treatment with sulphuric acid (Kennedy Axelrod, (1971) Immunology, 20,253-257); with chromic acid or with sulphuric acid and chromic acid (Klemperer Knox, (1977) Laboratory Practice 26(3), 179-180); or treatment with a corona discharge process (Maroudas, (1973) in "New Techniques in Biophysics and Cell Biology" H. Payne and B. J. Smith, eds) Wiley Interscience, London). These treatments are believed to introduce hydroxyl groups, and the surface concentration of hydroxyl groups must fall within a range for the polystyrene derivative to be suitable for cell attachment (Curtis et.

al., (1983) J. Cell Biology 97, 1500-1506; and Curtis et. al.

(1986) J. Cell Sci. 86,9-24). Carboxyl groups produced in the reactions appear to play only a small role in the cell adhesion to modified polystyrene (Curtis et al., 1986). Very few sulphonate groups are introduced into the surface (Curtis et al. 1983).

Somewhat different results as to the surface groups required for cell attachment were obtained in a study of cell attachment to polyHEMA by McAuslan et al (PCT/AU87/00043 and McAuslan et al., J. Biomed. Mater. Res., 1987). In that study it was shown that hydrolytic etching of polyHEMA with sulphuric acid converted the non-adhesive surface into a surface that is highly adhesive for cells. In that case, the improvement in adhesiveness of the hydroxyl-rich surf,.ce of polyHEMA surface for cells appeared to correlate with the partial introduction of carboxyl groups onto the surface.

WO 90/02145 PCT/AU89/00356 -4- Another aspect of the mechanism of adhesion of cells to polymeric surfaces is that serum adhesive proteins adsorbed to the surface contribute to the cellular attachment reaction. For tissue culture polystyrene, the serum component fibronectin (Fn) has been shown to support endothelial cell attachment. Recent results from Underwood et al. (Aust. New Zealand Soc. Cell Biol., 1988 Meeting, abstracts 1988) point to the adsorption to the polystyrene surface of a second serum component, vitronectins as being essential to the attachment of endothelial cells. The nature of the surface chemistry can have subtle effects on the conformation of the attached serum components with consequential effects on the biological potency of the adsorbed protein. Grinnel and Feld (1981) J. Biomed. Mater.

Res., 15, 363-381 and (1982) J. Biol Chem., 257, 4888-4893; have compared the binding of fibronectin to tissue culture polystyrene and biological potency of the bound fibronectin with the binding to hydrophobic unmodified polystyrene. That study showed that the ability of the fibronectin adsorbed to the tissue culture polystyrene surface to promote cell attachment was markedly greater than that of fibronectin adsorbed to the hydrophobic polystyrene surface. It follows that the suitability of a polymer surface for cell attachment is related to both the surface chemistry and to the ability of the surface to adsorb specific adhesive proteins (whether from the serum or as purified serum components) in an active conformation.

L 1 The luminal surface of natural blood vessels has an antithrombogenic character which is believed to be a direct consequence of the ability of the endothelial cells that line the vessel to resist thrombus formation. Synthetic vascular grafts have a markedly more thrombogenic surface and frequently fail because of spontaneous thrombosis. It is believed that if the surface of the graft can be covered with endothelial cells that function physiologically, these cells will form a naturally nonthrombogenic interface between the graft and the blood. The cells that are involved in such a process of endothelialisation could arise through spontaneous coverage from endogenous sources (migrating endothelial cells from cut edges of the adjacent blood vessel, or else from capillaries migrating from the perigraft tissue through the interstices of a porous graft) or by seeding of the graft with endothelium. One aspect of the design of vascular prostheses is therefore to ensure that endothelial cells can attach and grow on the surface, particularly where the graft is for use in small to medium-sized arteries that carry low blood flow. Thus the ability of the polymer surface to support endothelial cell attachment and growth is an important characteristic of the effectiveness of the prosthesis. Surfaces that are suitable for endothelial cell attachment and growth are likely to support ingrowth of other mesenchymal tissues, and so be suitable for general implant applications including the enhancement of wound closure and anchorage of percutaneous implants.

~c i r i r WO 90/02145 P~/AU89/00356 The failure of the hydrophobic surface of PTFE to adequately support cell attachment, including attachment of endothelial cells exposed to the shear forces involved in blood flow, is a limitation to the use of this material for vascular prostheses. If PTFE could be modified to produce a surface that supported enhanced endothelial cell attachment and growth, the modified surface could be expected to be more suitable than unmodified PTFE for the process of in vivo endothelialisation. PTFE that is modified to be superior to unmodified PTFE for the attachment of endothelial cells would certainly be preferable for use in the new approach (Herring, Gardner and Glover, (1978) Surgery, 84, 498-504) of preseeding grafts with endothelial cells prior to implantation.

While the introduction of strongly bonded surface carboxyl groups to normally hydrophobic fluorocarbon polymers by various high energy techniques of grafting is well known (eg Charpiro Jendrychowska-Bonamour (1980) Polymer Engineering and Science, 20(3), 202-205) we have found that the resultant surface poorly supports endothelial cell growth. It appears that the even distribution of carboxyl groups provided by tiese grafting methods is not beneficial to cell attachment, when quite low (around grafting levels or higher levels are used. In contrast to this finding and the lack of cell attachment to unmodified PTFE and other fluorocarbon polymers, it has now been found by the present inventors that certain acidic treatments of PTFE and other fluorocarbon polymers to which polyacrylic acid chains have been grafted produce surfaces having improved cell i lr WO 90/02145 PCT/AU89/00356 -7attachment and growth properties, without adversely affecting the physical properties of the materials.

Accordingly, the present invention is centred on the development of processes for the chemical modification of PTFE and other fluorocarbon polymers, to produce an implantable surface that supports the attachment and growth of animal (including human) tissue cells, such as fibroblasts and other mesenchymally-derived cells, epithelial cells and endothelial cells. Where it is endothelial cells in contact with the fluoropolymer surface, the surface produced by this process would support the attachment and growth of the endothelial cells into a confluent surface. The attached endothelial cells then present at the blood interface an antithrombogenic surface which inhibits undesirable platelet interactions.

DISCLOSURE OF THE INVENTION: It is an object of the present invention to provide a material useful in implantable prostheses which will substantially overcome the disadvantages of the prior art in that it has improved biocompatibility arising from enhanced cell attachment properties and antithrombogenicity.

According to the invention there is provided a process for preparing a surface for the attachment and growth of cells, said process comprising the steps of:i) grafting polyacrylic acid chains to a fluorocarbon polymer substrate so that the weight of the fluorocarbon polymer substrate increases by between 0.1% and WO 90/02145 PCT/AU89/00356 -8ii) treating the fluorocarbon polymer surface produced by step with concentrated sulphuric acid at a sufficiently high temperature and for a time to separately decarboxylate, aromatize and sulphonate an effective proportion of the grafted polyacrylic acid chains, iii) drying the surface produced by step (ii), iv) soaking the surface produced by step (iii) in a concentrated acid, v) neutralizing the surface produced by step (iv).

Optionally, the grafted substrate produced by step (i) may be left to soak in room temperature sulphuric acid prior to the treatment of step This is best done overnight.

Where required, the neutralized surface produced by step may be treated with serum, cell attachment factors derived from serum or connective tissue (such as fibronectin or vitronectin) or with tissue growth factors.

The preferred method of grafting polyacrylic acid chains to the substrate in step is by gamma-irradiation grafting. Other means of grafting, such as laser grafting, may be employed where appropriate. Where gamma-irradiation grafting is employed, step is preferably carried out according to the method of Charpiro (reviewed in Charpiro Jendrychowska-Bonamour, 1980) which involves treatment of the polymer surface with a water soluble inhibitor in the grafting solution.

WO 90/02145 PCT/AU89/00356 -9- The time required to seperately decarboxylate, aromatize and sulphonate an effective proportion of the grafted polyacrylic acid chains of the surface in step (ii) will depend on the temperature at which the sulphuric acid treatment of the polymer is carried out, but it is preferred that the treatment of step (ii) is with sulphuric acid at 105 0 C for 2 hours. It is understood by the skilled artisan that an effective proportion of grafted polyacrylic acid chains refers to that proportion of same that will lead to effective attachment and growth of cells on the surface of the invention.

Preferably, the drying that is required in step (iii) is carried out by heating at 105 0 C for 4 hours, while the soaking in concentrated acid that is required in step (iv) is preferably carried out with concentrated nitric acid at room temperature for 4 hours.

Neutralization of the relatively acidic surface produced by step is preferably carried out by washing with phosphate buffered saline at pH 7.4.

According to another aspect of the invention, there is provided a surface for the attachment and growth of cells whenever prepared by any one of the aforementioned p:ocesses.

According to a further aspect of the invention, there is provided a method of promoting cell attachment and growth on a surface comprising preparing a surface according to any one of the aforementioned processes and exposing said surface to cells.

WO 90/02145 PCT/AU89/00356 The preferred fluorocarbon polymers of the substrate include polytetrafluoroethylene, fluorinated ethylene propylene, polychlorotrifluoroethylene and polyvinylidine fluoride which may readily be gamma-irradiation grafted with polyacrylic acid chains. The polyacrylic acid chains that are grafted onto the substrate in this manner may be the product of reaction of the substrate with acrylic acid or acrylate ester.

Where gamma-irradiation grafting has been employed in step the chemical changes to the surface produced after step in which step concentrated nitric acid has been used, have been studied by electron spectroscopy for chemical analysis (ESCA). This study shows that small concentrations of surface sulphur groups (probably sulphonic acid) at a binding energy of.167 eV (for S 2p) are present and that the number of these increases with the time and temperature of acid treatment. Preferably, the surface produced after step (iv) will contain about 1 sulphur atom per 75 carbon atoms.

Fourier Transform Infrared (FTIR) studies of the treated PTFE surfaces (by Attenuated Total Reflectance ATR) showed that considerable decarboxylation of the acid groups at 1710 cm 1 had occurred in the 1 micron surface layer, but that other non-ionizable, oxidized species had partly taken their place (broad peaks at 1720 -1670 cm-1).

7 r- WO 90/02145 PCT/AU89/00356 -11- ESCA studies comparing the grafted surface produced by step with the surface after step (iv) further showed that the surface carboxyl groups at binding energy of 288 eV are present at 1/5 of the equivalent levels of CF groups at binding energy of 292 eV in the surface after step The level of these surface carboxyl groups decreases during treatment at room temperature and at 105 0 C with sulphuric acid but is not significantly modified during the drying step (iii). Some re-oxidation of carbon to give further carboxyl groups occurs during step After step the carboxyl group signal had diminished to approximately 70% of the peak height of the original carboxyl peak (288 eV) ob--rved after step The ion exchange capabilities dropped from 1.2.meq/g in the original grafted films to 0.82 meq/g for 10 mins treatment, 0.75 meq/g for 18 mins and 0.26 meq/g for the optimum treatment period of over 1 hr. The time to reach the measured equilibrium ion exchange capabilities increased from 1 hr to 24 hrs due to this surface decarboxylation.

Ultraviolet and visible spectroscopy indicated that aromatization of the polyacrylic acid graft occurred, with an increase on treatment in the broad adsorption peak from below 250 nm to above 430 nm; adsorption values above 1.0 in the PTFE (Quinton cannula connector) tubes moved from about 305 nm to 550 nm. In line with the above changes, the ESCA studies showed that the ratio of unfluorinated carbon atoms carbons from the acrylic acid grafted on during step produced by step with concentrated sulphuric acid at a sufficiently high temperature and for a time to separately decarboxylate, aromatize and sulphonate an effective proportion of the grafted polyacrylic acid chains, /2 WO 90/02145 PCT/AU89/00356 -12decreased when compared to fluorinated carbons, during the treatment with sulphuric acid in step The number of sulphur atoms increased from negligible to be about 1 in of the C atoms following step and remained at this level in subsequent treatments during steps (iii), (iv) and None of these changes occurred with similar treatments of ungrafted PTFE film or tubing.

BRIEF DESCRIPTION OF THE DRAWINGS: Fig 1 is a schematic view of the flow test system used in the Examples, Fig 2 a) is a photomicrograph of ovine carotid arterial endothelial (OCAE) cells grown on modified sample #1 of a surface of the invention, Fig 2 b) is a photomicrograph of OCAE cells grown on modified sample #2 of a surface of the invention, Fig 2 c) is a photomicrograph of OCAE cells grown on unmodified PTFE surfaces, Fig 2 d) is a photomicrograph of OCAE cells grown on modified sample #3a of a surface of the invention, Fig 3 a) is a photomicrograph of human umbilical artery endothelial (HUAE) cells grown for 24 hours after seeding onto mPTFE in serum-free medium, Fig 3 b) is a photomicrograph of HUAE cells grown for 24 hours after seeding onto mPTFE in medium containing serum, Fig 3 c) is a photomicrograph of HUAE cells grown for 24 hours after seeding onto mPTFE in medium containing Vitronectin-stripped serum, II I I- WO 90/02145 PC/AU89/00356 -13- Fig 3 d) is a photomicrograph of HUAE cells grown for 24 hours after seeding onto TCP in serum-free medium, Fig 3 e) is a photomicrograph of HUAE cells grown for 24 hours after seeding onto TCP in medium containing serum, Fig 3 f) is a photomicrograph of HUAE cells grown for 24 hours after seeding onto TCP in medium containing Vitronectin-stripped serum, Fig 4 is a graph of HUAE cell growth over 5 days on various surfaces of Example 3, Fig 5 a) is a photomicrograph of HUAE cells grown for days on PTFE, Fig 5 b) is a photomicrograph of HUAE cells grown for days on mPTFE, Fig 6 a) is a photomicrograph of HUAE cells grown for days on Fibronectin-coated TCP, Fig 6 b) is a photomicrograph of HUAE cells grown for days on Fibronectin-coated PTFE, Fig 6 c) is a photomicrograph of HUAE cells grown for days on Fibronectin-coated mPTFE, and Fig 6 d) is a photomicrograph of HUAE cells grown for days on mPTFE not precoated with Fibronectin.

DESCRIPTION OF PREFERRED EMBODIMENTS: 3 3 3 3 3 3 In order that the invention may be more readily understood and put into practical effect, reference will now be made to the following examples.

EXAMPLE 1 Chemical modification of PTFE film and modification of PTFE tubes (TEFLON Cannula Connector tubes) WO 90/02145 PCT/AU89/00356 -14- A portion of polytetrafluoroethylene film (200 microns thick) was cut into 5cm 2 pieces and soaked in a solution of acrylic acid, 10% in methylene chloride, for 2 days. The samples were removed and air dried before being placed in a suitable glass vessel containing a grafting solution of lOg of acrylic acid in 90g of distilled water containing 0.03% CuC12. The samples were then irradiated for 16 hrs at a dose rate of approximately 0.07 MRad/hr (total dose 1.12 MRad) in a Cobalt 60 source, to give 1.25% by weight of grafted polyacrylic acid on the PTFE. The grafted film was then cleaned in distilled water, dried and then placed in concentrated sulphuric acid at 105 0 C for 90 mins. The light brown film was then washed in distilled water, dried at 0 C for 16 hrs, and finally cleaned in concentrated nitric acid by soaking for 24 hrs. Prior to tissue culture studies the pieces were extensively washed in sterile phosphate buffered saline (PBS) pH 7.2. These pieces of modified PTFE film were subsequently used in Example 3.

Twenty cannula connector tubes (Catalog no. 11150-002 from Quinton Instrument Co., Seattle Washington 98121, described as tubes of plain non-etched PTFE of 25mm length, 2.8mm 3.4mm were soaked in a solution of acrylic acid, 10% in methylene chloride, for 2 days before being air dried and pla,'ed in the grafting solution of acrylic acid in 0.03% aqueous CuCl 2 solution. These were gammairradiation treated at a dose rate of approximately 0.07 MRad/hr for 16 hrs (total dose 1.12 MRad) to give a 1.5% by -i ~XY P WO.90/02145 PCT/AU89/00356 weight graft on the tubes. The grafted tubes were then cleaned in distilled water, dried, allowed to soak overnight in sulphuric acid at room temperature and then placed in concentrated sulphuric acid at 1050C for varying periods of time, to give surfaces with different characteristics.

Treatment with sulphuric acid for 90 minutes was used to produce mPTFE tube sample whereas treatment for minutes was used to produce mPTFE tube sample The tubes had different colours depending upon the treatment time, varying from light brown in colour (mPTFE tube sample to a darker brown (mPTFE tube sample These tubes were then washed in distilled water, dried at 90 0 C for 16 hrs and finally treated with concentrated nitric acid by soaking for 24 hrs. Other grafted cannula connector tubes were treated with sulphuric acid at 105 0 C for 2 hr, then dried at 105 0 C for 4 hr, then soaked with room temperature nitric acid for 4 hr and then washed in distilled water (these are mPTFE tube samples Prior to tissue culture studies all the tube samples were extensively washed in sterile PBS.

These modified PTFE tubes were subsequently used in Example 2.

EXAMPLE 2 Attachment and Growth of Ovine Endothelial Cells i On Modified PTFE Tubes Methods PTFE Cannula connector tubes which were modified by the procedure specified in Example 1 above were sonicated in *14 im=ulmc e a u x-ruwn .n viaro is polystyrene, modified by one of a number of techniques to produce a surface that can promote cell attachment (tissue culture polystyrene). This WO 90/02145 PCT/AU89/00356 16 acetone and washed in 70% ethanol. The mPTFE tubes required pH equilibration prior to use, by incubation with several changes of serum-free culture medium until a stable pH was detected. (No further change in the colour of the pH indicator in the medium over a period of more than 1 hr was deemed sufficient evidence of pH stability). Some of the mPTFE samples #3a were pre-coated with fibronectin (Fn) from bovine plasma by being asceptically filled with a solution of ug/ml Fn in PBS, plugged and incubated at 370C for 1 hr prior to cell seeding (these are mPTFE tube samples #3b).

An ovine carotid arterial endothelial cell culture (OCAE) was established after the methodology of Jaffe ((1984) in "Biology of EndothelialCells" Jaffe, ed) Martinus Nijhoff, Boston), and routinely maintained in McCoy (modified) medium supplemented with 20% foetal bovine serum, ug/ml penicillin and 100 ug/ml streptomycin and passaged using trypsin-versene. For experimental work cells were used between passage 5 and passage 12 (inclusive).

Preequilibrated mPTFE tubes were individually placed into sterile, screw-cap polystyrene vials, then 9ml of growth medium containing 2 x 106 cells was added to each vial.

The cell suspension was gassed with a mixture of 5% C02 in air and the vial tightly sealed. The vials were then placed inside a TCP roller bottle and firmly held in position by packing. The loaded bottle was then rotated at 1 r.p.m. on a roller at 37 0 C. The culture medium was replenished at 24 hr and 72 hr and the tubes removed for subsequent flowtesting after 5 days.

WO 90/02145 PCT/AU89/00356 -17- Visualisation of progressive cell growth necessitated the removal of a 5mm long end section of selected tubes which were then fixed in a 2.5% of glutaraldehyde in PBS, (GLUT), stained with Eosin Y and viewed under fluorescence microscopy. The cells were observed using an Olympus IMT microscope with a reflected light fluorescence attachment.

Other tubes supporting cell attachment and growth were cultured for 6 hr in culture medium consisting of Dulbecco's modified Eagle's medium containing glutamine, 3 mg/l methionine and 25 uCi/ml of 3bS-methionine, then further incubated with McCoys 5A medium with serum and supplements for a further 15 hr. The tubes containing the metabolicallylabelled cells were briefly washed in PBS then inserted into the flow test system as shown in Figure 1.

The flow test system of Fig 1 includes a gas cylinder having a cylinder outlet 11 that passes through gas flow regulator 12 and delivers gas to a vacuum flask 13 having an air pressure manometer 14. A tube 15 feeds from the cell culture medium 16 in the flask 13 to a water bath 17 held at 37 0 C. A bath outlet tube 18 has a media pressure manometer 19 connected thereto and downstream of the tube 18 is a flow adaptor tubing 20 and the subject graft 21 (in this case, the prepared tubes). Further downstream is a flow probe 22 connected to a flow meter 23. The tube 18 terminates at a three-way tap 24 from where the tube 18 forks to a pair of glass fibre filters 25 and 26. Material not captured by the filters is collected in the flask 27.

r r WO 90/02145 PCT/AU89/00356 -18- The tubes were subjected to increasing flow rates of a medium consisting of McCoys 5A medium containing 20 mM Hepes buffer (pH 7.2) and 20% foetal bovine serum at 37 0

C

for the specified time periods. Cells released from the tube and that were collected on the downstream filters were quantitated by radioactive determination (liquid scintillation counting). Following the flow studies, the tube was removed and bisected, then half of the tube was examined for adherent cells by microscopic techniques and the cells on the other half were removed using trypsin-versene and the radioactivity in the released cells was determined.

Results OCAE cell growth was studied on mPTFE tubes that had four levels of surface modification (mPTFE samples #2, #3a and #3b, which had the same level of acrylic acid graft but which had subsequently been exposed to different postgraft modification conditions, as earlier specified. The cell growth on these tubes was compared to that on unmodified PTFE cannula connectors.

mPTFE tube samples #3a and #3b which were exposed to longer periods of past-grafting modification than was mPTFE tube sample supported good OCAE cell growth. The cell monolayer on sample #1 reached confluent levels by 3 days (see Figure 2a), and at 5 days showed no signs of cell layer delamination or clumping. However on mPTFE tube sample the cells did not reach confluence by day 5, and many of the cells tended to clump together, whilst others displayed a I ,I I- applications including the enhancement of wound closure and anchorage of percutaneous implants.

p.

WO 90/02145 PCT/AU89/00356 -19spindle-like morphology (see Figure 2b for morphology after 3 days). The unmodified PTFE tubes supported very little OCAE cell attachment and growth, with those few cells that were present showing very poor cell spreading (see Figure 2c).

mPTFE tube sample #3a supported -ood cell attachment and growth of OCAE cells, (See figure 2d) and a similar result was achieved with mPTFE sample #3b (not shown).

These results demonstrate that chemical modification of PTFE tubes in the manner described for tube samples #3a and #3b produced surfaces that were markedly better than unmodified PTFE tubes for initial OCAE cell attachment and growth.

Flow Testing mPTFE tubes were prepared similarly to tube sample #1, seeded with OCAE cells which formed a confluent layer after days growth on their luminal surface, and were tested in an in vitro flow system (See Table 1, experiment No.s 1 to 6).

The OCAE cells withstood the shear force treatments of up to 4 dynes/cm 2 with a minimum of 73% of the cells remaining attached to the tube, and with less than 7% of the cells being lost from the surface in 4 of the 6 experiments, mPTFE tubes prepared similarly to tube samples #3a and #3b were seeded with OCAE cells, and these cells withstood shear force treatments of up to 20 dynes/cm 2 (Experiment No.s 7 to of Table 1) which is equivalent to shear force levels found in arterial blood vessels. A minimum of 94.3% of the cells a I 4 1 iLuu.ng ana ine iacK or cell attachment to unmodified PTFE and other fluorocarbon polymers, it has now been found by the present inventors that certain acidic treatments of PTFE and other fluorocarbon polymers to which polyacrylic acid chains have been grafted produce surfaces having improved cell I t WO 90/0214PCT/AU89/0035 6 WO 90/02145 remained attached to tube sample #3a as shown in Experiment No.s 7 to 10, whilst a minimum of 92.5% of the cells remained attached to sample #3b as shown in Experiment No.s 11 to These experiments show that endothelial cells form strong attachment to the mPTFE surface.

WO 90/02145 W090/2145PCT/AU89/00356 -21- Thble 1.

eteian of CCAE cells an nF1' tube urr~e 1a 73.0 27.0 2 a 98.1 1.9 3a 81.0 19.0 4 b 93.9 6.1 b 97.6 2.4 6 c 98.1 1.9 7 d 97.4 2.6 8 d 94.7 5.3 9 d 96.7 3.3 d 94.3 5.7 11 d 97.7 2.3 12 d 96.4 3.6 13 d 92.5 14 d 97.8 2.2 d 97.1 2.9 Experiment No.s 1 to 6 utilize tube sample #I, experiment No.s 7 to 10 utilize tube sample #3a and experiment No.s 11 to 15 utilize tube sample #3b.

po.±ymer surrace with a water Soluble inhibitor in the grafting solution.

WO 90/02145 PCF/AJ89/OO356 22 Cells were subjected to the following flow protocols: a 10 min at 14 ml/min (0.8 dynes/cm 2 then 10 min at 38 mi/mmn (2.3 dynes/cm 2 then 10 min at mi/mmn (4.0 dynes/cm 2 b 5 min at 38 mi/mmn then 10 min at 56 mi/mmn then 9 min at 70 ml/min.

c 10 min at 38 ml/min then 10 min at 56 mi/mmn then 6 min at 70 ml/min.

d 10 min at 66 mi/mmn (4.0 dynes/cm 2 then 10 min at 132 ml/min (8 dynes/cm 2 then 10 min at 198 mi/mmn (12 dynes/cm 2 then 10 min at 264 mi/mmn (16 dynes/cm 2 then 10 min at 330 mi/mmn dynes/cm 2 it r -r WO 90/02145 PC/AU9/00356 -23- EXAMPLE 3 Attachment and Growth of Human Endothelial Cells On Modified PTFE Film Methods Sheets of unmodified, virgin PTFE and PTFE modified by the procedure specified in Example 1 above, (mPTFE), were cut into squares of' approximately 5mm x 5mm, sonicated in acetone then washed extensively in 70% ethanol for 2 hr. The mPTFE required pH equilibration and stabilisation prior to use, and equilibration was achieved with several changes of serum-free growth medium until a stable pH was detected. Those samples to be coated with fibronectin (Fn) were each placed into a 22mm diameter TCP well (12-well cluster dish) and covered with 1 ml of a solution of 40 ug/ml Fn in PBS and incubated at 370C for 1 hr.

A human umbilical artery endothelial cell culture (HUAE) was established and grown in 75cm 2 tissue culture polystyrene (TCP) flasks coated with Fn. Fn coating was achieved by incubating the flasks with 5 ml solution of Fn in PBS at 37 0 C for 1 hr prior to cell seeding.

The cells were routinely maintained in a growth medium consisting of an equal mixture of McCoy 5A (modified) and BM86-Wissler media supplemented with 30% v/v foetal bovine serum, 40ng/ml fibroblast growth factor, 60 ug/ml endothelial cell growth supplement, 20 ug/ml insulin, 60 ug/ml penicillin and 100 ug/ml streptomycin. The cells were routinely passaged using trypsin-versene, and for experimental work cells were used between passage 15 and passage (inclusive).

I;

WO 90/02145 PC/AU9/00356 -24- For cell growth studies, pre-equilibrated samples were individually placed into 22 mm diameter TCP wells and 2 ml of the growth med-im earlier used to maintain the HUAE cell culture containing 5 x 10 4 cells was added to each well.

In some experiments of this Example, the serum adhesive glycoprotein fibronectin was removed from the foetal bovine serum component of the growth medium prior to use of the growth medium in the cell growth studies by passage over a gelatin-Sepharose affinity column. Serum treated on a gelatin-Sepharose column was confirmed to be free of fibronectin by immunoassay of the fibronectin content. In other experiments, the serum adhesive glycoprotein vitronectin was similarly removed by passage over an affinity column consisting of immobilized anti-vitronectin monoclonal antibody. The sera that were depleted of vitronectin by this affinity technique were confirmed to have been exhaustively stripped of vitronection by immunossay of vitronectin content.

TCP and Fn-coated TCP were used as control surfaces, with particular attention being paid to Fn coated TCP, which is known to support good HUAE cell attachment and growth.

Duplicates of each test sample were fixed in GLUT at 24, 72 and 120 hr, then stained with a 0.05% aqueous solution of Eosin Y, and viewed by fluorescence microscopy. Ten random fields per test sample were photographed and the mean and standard error of cell number per cm 2 were determined from photographic prints.

WO .90/02145 Results The number of HUAE ci as viewed 24 hr after cell that on TCP whereas for un attached was 74% of that o Table 2.

PCT/AU89/00356 ells attached to the mPTFE surface seeding was approximately 88% of modified PTFE the number of cells n TCP see Table 2.

HUAE cells attached/cm2 after 24 hr.

Substrate Mean cell no./cm2

TCP

PTFE

mPTFE TCP/Fn PTFE/Fn mPTFE/Fn 2012 (141) 1489 (179) 1771 (333) 3817 (292) 3037 (202) 4000 (343) The morphology of the HUAE cells on mPTFE was generally quite elongated and spindle-like. This morphology indicated that although some cell spreading processes had occurred, the cells had not spread to form the well spread morphology that is typical of well attached endothelial cells. The morphology of the HUAE cells on the mPTFE was generally similar to that on TCP after 24 hr. By comparison, HUAE cells that had been seeded and attached on unmodified PTFE were mainly truncated or rounded up, clearly showing that the attachment of individual cells on the mPTFE was better than on the unmodified PTFE surface.

WO 90/02145 26 PC/AU89/00356 The role that the serum component of culture medium may play in the attachment of HUAE cells to mPTFE and to tissue culture plastic was determined. When the culture medium in which the HUAE cells were seeded did not contain serum, the HUAE cells did not spread on either mPTFE or TCP surfaces (see Figure 3a and The role that serum adhesive glycoproteins fibronectin and vitronectin may play in the attachment of HUAE cells to mPTFE was determined by selectively depleting the serum component of the culture medium of fibronectin or vitronectin by passage over an affinity chromatography matrix. Selective removal of fibronectin from the serum component of the culture medium did not interfere with HUAE cell attachment and spreading onto mPTFE or onto TCP. Selective removal of vitronectin from the culture medium caused reduced cell attachment to the TCP surface with the attached cells being unable to spread onto the surface (Fig 3e as compared to Fig 3f). When HUAE cells were seeded onto the mPTFE surface in medium containing serum depleted of vitronectin, the HUAE cells did attach and spread, to an extent that was equivalent to that on mPTFE with intact serum (compare Fig 3b with The importance of vitronectin, which is also known as serum spreading factor, epibolin or 70K spreading factor, has previously been reported for polymer .rfaces such as tissue culture polystyrene (see Grinnell (1976) Exp. Cell Res., 97, 265- 274 and (1977) Exp. Cell Res., 110, 175-190; Underwood and Bennett (1989) J. Cell Science (in press, 1989)) and

'V

WO 90/02145 PCT/AU89/00356 -27- Nafion (International patent application PCT/AU88/00368).

These results indicate that vitronectin is not essential for attachment of HUAE cells to the mPTFE surface, unlike the TCP surface. These results do not exclude the possiblity that serum vitronectin adsorbed onto the mPTFE surface may contribute to HUAE cell attachment and spreading onto mPTFE when the culture medium contains serum.

Subsequent growth over a 5 day period was monitored (see Figure 4) and Figures 5a) and b) show the morphology on PTFE and mPTFE respectively after 3 days growth. These studies showed that cell growth occurred slowly on the mPTFE surface, whereas on unmodified PTFE there was very poor cell growth, and cell numbers reduced to a point where very few cells were visible after 72 hr, and no cell attachment was evident after 5 days.

For HUAE cells the cell attachment, cellular morphology and growth on TCP that had been coated with Fn was better than on uncoated TCP (Table Similarly, Fn-coating of mPTFE enhanced the attachment of HUAE cells as evident after 24 hr (Table This had the effect that the Fn-coated mPTFE and Fn-coated TCP surfaces supported approximately twice the number of attached cells than that seen on the uncoated surfaces. The Fn-coated mPTFE displayed a similar capacity as Fn-coated TCP for HUAE cell attachment, whilst the Fn-coated PTFE supported approximately 80% of that cell number. The morphology of the cells after 24 hr culture on Fn-coated mPTFE was markedly more spread than the cells on WO 90/02145 PC/AU89/00356I -28the mPTFE which had not been precoated with Fn, and was generally similar to the morphology of cells cultured on Fncoated TCP. During growth over the subsequent 72 hr, the Fncoated mPTFE maintained parity with the control Fn-coated TCP at close to confluence (See Figure 4 and for cellular morphology, compare Figure 6a with Figure 6c). However, the Fn-coated PTFE sample, although maintaining good cellular morphology (Figure 6b), exhibited a reduction in cell numbers, dropping from 75% of that of the Fn-coated TCP to only 50% after 5 days. This showed the Fn-coated PTFE surface to be only marginally better than mPTFE that had not been coated with Fn (Figure 4 and Fig 6d). Both the Fncoated mPTFE and the Fn-coated TCP continued to increase in cell number at comparable rates of growth during 5 days.

The present invention is centred on the evaluation and design of materials which provide a suitable support for endothelial cells. The endothelial cells should provide an antithrombogenic surface as they do in vivo. By successfully growing endothelial cells, derived from either ovine carotid artery and human umbilical artery, on the surface of modified PTFE film and modified PTFE tubes, we have shown that the modified PTFE surface has good cell supportive characteristics. The results indicate that this chemical method of modification of PTFE may be used to improve the cell adhesive characteristics of PTFE, and this chemically modified PTFE surface would be superior to unmodified PTFE for use in implantable materials including vascular prostheses and percutaneous implants.

procedure specified in Example 1 above were sonicated in I c WO 90/02145 PCT/AU89/00356 -29- It will be obvious tothose skilled in the art that the technique described in these Examples for enhancing the cell supportive characteristics of PTFE is applicable to expanded PTFE (sold under the registered trade mark GORE-TEX) and other fluorocarbon polymers.

The foregoing describes only some of the embodiments of the present invention and modifications, obvious to those skilled in the art,, an be made thereto without departing from the scope and ambit of the invention.

Claims (14)

1. A process for preparing a surface for the attachment and growth of cells, said process comprising the steps of:- i) grafting polyacrylic acid chains to a fluorocarbon polymer substrate so that the weight of the fluorocarbon polymer substrate increases by between 0.1% and ii) treating the fluorocarbon polymer surface produced by step with concentrated sulphuric acid at a sufficiently high temperature and for a time to separately decarboxylate, aromatize and sulphonate an effective proportion of the grafted polyacrylic acid chains, iii) drying the surface produced by step (ii), iv) soaking the surface produced by step (iii) in a concentrated acid, v) neutralizing the surface produced by step (iv).

2. A process according to claim 1 further including the step of treating the neutralized surface with any one of serum, cell attachment factors derived from serum or connective tissue, or tissue growth factors.

3. A process according to claim 2 wherein the cell attachment factor is selected from the group comprising fibronectin or vitronectin. the cells did not reach confluence by day 5, and many of the cells tended to clump together, whilst others displayed a WO 90/02145 PCT/AU89/00356 -31-

4. A process according to any one of claims 1 to 3 wherein the polyacrylic acid chains are grafted onto the fluorocarbon polymer substrate by any one of gamma-irradiation grafting or laser grafting means.

A process according to any one of claims 1 to 4 wherein step (ii) is carried out at a temperature of 1050C for 2 hours.

6. A process according to any one of claims 1 to 5 wherein step (iii) is carried out at a temperature of 105 0 C for 4 hours.

7. A process according to any one of claims 1 to 6 wherein step (iv) is carried out by soaking in concentrated nitric acid at room temperature for 4 hours.

8. A process according to any one of claims 1 to 7 wherein step is carried out by washing with phosphate buffered saline at pH 7.4.

9. A process according to any one of claims 1 to 8 wherein the fluorocarbon polymer substrate is selected from the group comprising polytetrafluoroethylene, fluorinated ethylene propylene, polychlorotrifluoroethylene and polyvinylidine fluoride.

A surface for the attachment and growth of cells whenever prepared by the process of any one of claims 1 to 9.

11. A surface according to claim 10 which contains about 1 sulphur atom per 75 carbon atoms.

12. A tissue implant incorporating the surface of claim or claim 11. WO 90/02145 PCT/AU89/00356 -32-

13. A method of promoting cell attachment and growth on a surface, said method comprising:- preparing a surface according to the process of any one of claims 1 to 9, and exposing said surface to cells.

14. A method according to claim 13 wherein the cells to be attached and grown are selected from the group comprising endothelial, epithelial, fibroblast and other mesenchymally- derived cells. A method of promoting cell attachment and growth on a surface, said method being substantially as hereinbefore described with reference to the Examples.