AU8017498A - Collagen-like polymers with cell binding activity - Google Patents

Description

WO 98/52620 PCT/EP98/02857 Collaaen-like Polymers with Cell Binding Activity The present invention relates to "collagen-like polymers" which exhibit cell binding activity, to bulk materials having a coating of the polymers applied thereto and to methods of covalently binding the polymers to bulk materials. In particular, but by no means exclusively, the present invention relates to implants such as wound repair materials, implanted contact lenses, synthetic epikeratoplasties, synthetic skin or connective tissue, orthopaedic implants, prosthetic joints, synthetic arterial surfaces, synthetic neural tissue, prosthetic organs or other synthetic bioactive apparatus or materials which comprise the collagen-like polymers. In addition, the present invention relates to methods of treatment involving the implants which comprise the cell binding collagen-like polymers. The term "collagen" relates to a family of native proteins present in the extracellular matrix, and particularly connective tissues, ligaments, tendons and bones of animals including humans. Native collagen has a primary structure of repeating trimeric amino acid sequences. Within a helical region, which constitutes about 95% of collagen molecules, the amino acid glycine (Gly) occurs at every third position of a peptide trimer. Imino residues (1), either proline (Pro) or hydroxyproline (Hyp), occur in 56% of the trimers, 20% as Gly-X-1; 27% as Gly-l-Y; and 9% as Gly-l-l (where X and Y are amino acid residues other than Pro or Hyp). Pro usually occurs in the second position in the repeating trimer, while Hyp usually occurs in the last position. Full particulars of the structure of native collagens can be found in Bhatnagar, R. and Rapaka, R. (1976) Chapter 10 in Biochemistry of Collagen, R. Ramachandren Editor, Plenum Press, New York, pp 481-482, which is included herein by way of reference. Tripeptide sequences (Gly-X-Y) wherein X and Y are amino acid residues other than proline (Pro) or hydroxyproline (Hyp) make up 44% of the collagen amino acid trimers. Glutamic acid, leucine and phenylalanine occur mostly in the X position and threonine, glutamine, methionine, arginine and lysine occur mostly at the Y position. With the exception of alanine and serine which can be present at the X position, the X position amino acids have bulky side chains.

WO 98/52620 PCT/EP98/02857 -2 In the past it has been of interest to develop synthetic polymers which have collagen-like properties which can be used within biomaterials for clinical applications. For example use of collagen-like synthetic polymers has been contemplated in drug delivery devices, occular devices and wound healing materials. With these goals in mind, many synthetic peptides composed of the trimeric amino acid sequences Gly-Pro-Xaa and Gly-Xaa-Pro (where Xaa is any natural amino acid residue) have been prepared to mimic the collagen structures. Native collagens have a characteristic tertiary, secondary and primary structure and are made up of three polypeptide chains comprising repeating amino acid trimers. These chains are arranged in three extended left-handed spirals of about 3 residues per turn: the polyproline l-like chains, as described in Ridge (1955) Nature 176: 915, which is also included herein by way of reference, in its entirety. The polyproline Il-like chains are arranged in a parallel direction and intertwined to adopt a supercoiled, right-handed triple helix conformation (Bella et al (1994) Science 266: 75-81). In developing synthetic collagen like polymers, it is naturally important that this triple helical conformation can be effectively mimicked, as collagen-like activity is dependent upon this conformation. Implants for human and animal bodies are currently being developed at a dramatic rate. For example, implants may vary from a simple synthetic skin for treating burn, wound or aesthetic surgery patients, to synthetic ligaments, articular surfaces or prosthetic joints for orthopaedic patients. It is even possible to implant bioactive apparatus such as artificial organs which will fulfil the function of the heart or lungs etc. Materials which assist the repair of wounds, implantable contact lenses and synthetic epikeratoplasties, as well as other synthetic connective tissues, orthopaedic implants, synthetic arterial surfaces, synthetic neural tissues or other synthetic bioactive apparatus or materials are also being increasingly implanted into human or animal patients. A problem arises however, as many of these implantable materials need to be accepted by adjacent tissues so that cells will bind to, and tissues will colonise, the implanted material. This has been a problematic area in the past as implanted materials which are generally manufactured from metals or metal alloys, plastics polymers or ceramics will not encourage cell binding and tissue colonisation. In the past such problems, particularly in the field of corneal onlays, have been dealt with by adsorption or covalent attachment of proteins (such as collagen, fibronectin or laminin) to the surfaces of implantable polymers. A surface coating of such proteins will stimulate WO 98/52620 PCT/EP98/02857 -3 cellular attachment to the polymer material. It has also been proposed in the past that cell attachment to the surface of synthetic polymers or other implants may be stimulated by the covalent attachment of fragments of cell-adhesive molecules. This approach also has been problematic as such fragments of cell-adhesive molecules or proteins such as collagen, fibronectin or laminin are generally not fully synthetic, which as a result involves manufacturing difficulties and often high economic cost as well as safety problems for the implant patient. In addition, naturally occurring peptides when used to coat implantable materials are subject to enzymatic attack which will negatively impact upon the cell binding and tissue attachment activities of the peptide concerned. A number of recent papers by Goodman et al (Journal of the American Chemical Society, Vol. 118, No. 28, pp 5156-5157; Vol. 118, No. 43, pp 10359-10364; Vol. 118, No. 44, pp 10725-10732; Vol. 118, No. 44, pp 10928-10929 and Biopolymers, Vol. 39, 859-872 (1996)), each of which is included herein in its entirety by way of reference, relate to collagen-like polymers which include peptoid residues. The attraction of such collagen-like polymers which include peptoid residues is that they can be wholly synthetically prepared. As a result of the synthetic preparation they are safe for use in relation to human or animal implants. In addition, peptoid residue containing polymers are not as susceptible to enzymatic attack, which can adversely affect the activity of natural peptides when used to stimulate cell binding and tissue colonisation. Surprisingly, the present inventors have determined that some collagen-like polymer compounds which comprise peptoid residues are capable of initiating cell binding and tissue colonisation. In particular, the present inventors have identified that collagen-like polymers comprising at least nine adjacent repeats of the sequence Gly-Pro-Nleu have the ability to initiate cell binding and tissue colonisation. As used herein, Gly represents the amino acid glycine, Pro represents the amino acid proline and Nleu represents the peptoid residue N isobutylglycine, as described in Simon et al Proc. NatI. Acad. Sci. U.S.A. 1992, 89(20), 9367-9371, which is included herein by reference in its entirety. The present inventors have also determined means by which the collagen-like polymers according to the invention can be bound to implantable bulk materials, namely by covalent binding of terminal primary amine groups on the collagen-like polymer compounds with either aldehyde or carboxyl groups on a surface or surfaces of the implantable bulk WO 98/52620 PCT/EP98/02857 -4 material. Such covalent bonds are in the form of either Schiff base, amine or amide linkages. The latter two linkages are preferred as they enable irreversible immobilisation of the collagen-!ike polymers with respect to the bulk material. Accordingly therefore, it is an object of the present invention to overcome or at least substantially ameliorate problems associated with obtaining cell binding and tissue colonisation about implantable materials, as identified in the above prior art discussion. It is a further object of the present invention to provide collagen-like polymer compounds capable of initiating cellular attachment thereto. It is a further object of the present invention to provide peptoid-containing biopolymers bound to bulk materials that will improve the biocompatibility of the bulk material and which will have reduced suspectibility of biological degradation. It is a still further object of the present invention to provide biocompatible materials which can be stored at ambient temperature without significant degradation. It is a still further object of the present invention to provide implantable bulk materials bound to collagen-like polymers that will initiate cellular attachment thereto as well as providing methods of attachment of the polymers to the implantable bulk materials, and methods of treatment of patients requiring implants. Other objects of the present invention will become apparent from the following detailed description thereof. According to one embodiment of the present invention there are provided collagen-like polymer compounds capable of initiating cellular attachment thereto, said polymer compounds comprising one or more peptoid residues. According to another embodiment of the present invention there are provided collagen-like polymer compounds capable of initiating cellular attachment thereto, said polymer compounds comprising the sequence of formula I, wherein (y) denotes z 9: WO 98/52620 PCT/EP98/02857 -5 -(Gly-Pro-Nleu) ( - (1). According to a further embodiment of the present invention there is provided an implantable bulk material comprising one or more collagen-like polymer compounds as described above, covalently bound to a surface or surfaces thereof. According to a further embodiment of the present invention there is provided a method of binding biopolymer compounds, preferably collagen-like polymer compounds to an implantable bulk material, wherein said polymer compounds are capable of initiating cellular attachment, which polymer compounds comprise one or more peptoid residues, as well as primary amine groups, the method comprising the steps of: (a) providing aldehyde or carboxyl groups on a surface or surfaces of said implantable bulk material; (b) reacting said primary amine groups with aidehyde or carboxyl groups to form covalent bonds. According to a further embodiment of the present invention there is provided a method of binding biopolymer compounds to a bulk material, which biopolymer compounds comprise Kemp triacid adducts of biopolymer compounds comprising one or more peptoid residues, the method comprising the steps of: (a) providing amine groups on a surface or surfaces of said implantable bulk material; (b) reacting said amine groups with carboxylic acid groups on said Kemp triacid adducts to form covalent bonds. According to a further embodiment of the present invention there is provided a method of binding collagen-like polymer compounds to an implantable bulk material, wherein said polymer compounds are capable of initiating cellular attachment, which polymer compounds comprise the sequence of formula 1, as well as primary amine groups, the method comprising the steps of: WO 98/52620 PCT/EP98/02857 -6 (a) providing aldehyde or carboxyl groups on a surface or surfaces of said implantable bulk material; (b) reacting said primary amine groups with aldehyde or carboxyl groups to form covalent bonds. According to a further embodiment of the present invention there are provided implantable bulk materials comprising a copolymer which includes collagen-like polymer compounds that comprise one or more peptoid residues. According to a further embodiment of the present invention there are provided implanlable bulk materials comprising a copolymer which includes collagen-like polymer compounds that comprise the sequence of formula 1. According to a still further embodiment of the present invention there is provided a method of treatment of a human or animal patient requiring an implantation which comprises obtaining an appropriate implantable bulk material as described above and implanting said implantable bulk material into said patient. Throughout this specification, unless the context requires otherwise, the word "comprise", or variations such as "comprises" or "comprising", should be interpreted inclusively rather than exclusively. That is, these words will be understood to imply the inclusion of a stated element or integer or group of elements or integers but not the exclusion of any other element or integer or group of elements or integers. The essential feature of the polymer compounds is that they are collagen-like, exhibit cellular binding initiation characteristics and include one or more peptoid units. Preferably, but by no means essentially, the collagen-like polymer compounds according to the present invention include at least nine consecutive repeats of the sequence "-(Gly-Pro-Nleu)-" wherein Gly represents glycine, Pro represents proline and Nleu represents the peptoid residue N-isobutylglycine as disclosed in the Simon et al paper referred to above. A feature of polymers according to the invention is that they exhibit the triple helical tertiary conformation analogous to that of naturally occurring collagen proteins. In this way, the polymer compounds according to the present invention are considered to be "collagen-like".

WO 98/52620 PCT/EP98/02857 -7 Collagen-like polymers containing peptoid residues are disclosed in WO 97/19106. The disclosure of these specifications is included herein in its entirety by way of reference. It is to be clearly understood that the collagen-like polymer compounds according to the present invention, may additionally include one or more or many other chemical constituents. For example other naturally occurring or non-naturally occurring amino or imino acid groups or other synthetic polymers (eg. hydrogel constituents) may be copolymerised with the collagen-like polymer compounds of the invention which include peptoid residues. In particularly preferred embodiments of the invention the collagen-like polymer compounds are terminated by an amine group or by the Kemp triacid (KTA) (cis,cis-1,3,5-trimethylcyclohexane-1 ,3,5-tricarboxylic acid, as referred to in Kemp and Petrakis J. Org. Chem. 1981, 46, 5140-5143, which is included herein in its entirety by way of reference). By the term "capable of initiating cellular attachment", it is intended to convey that the collagen-like polymer compounds according to the present invention, and implantable bulk materials having the collagen-like polymer compounds bound thereto, will encourage cellular binding and tissue colonisation thereto when located in an environment containing human or animal cells and/or tissues. It is also specifically envisaged that the collagen-like polymer compounds according to the present invention may include an extra terminal primary amine group, which can be utilised for the purposes of binding these collagen-like polymer compounds to materials to be used for implantation purposes. As a result, what can take place is that a biomateria! surface which comprises the collagen-like polymer compounds according to the present invention can be applied to an implantable bulk material in order that the implantable material will be appropriately established within the human or animal organism by attachment of cells and tissues thereto. The Kemp triacid adduct of three collagen-like polymer compounds of suitable composition and length is particularly advantageous because when the Kemp triacid is condensed with glycine or alanine it conveniently acts as a template for inducing a-helicity of an attached polymer (Muller et al., 1993, Chapter 33 in Perspectives in Medicinal Chemistry, Testa et al. Editors, Basel). By attaching on to the Kemp triacid adduct a further chemical group capable of reacting with a chemical group on the surface of a desired bulk material, it is possible to bind the Kemp triacid adduct covalently on to a bulk material and thus confer the WO 98/52620 PCT/EP98/02857 -8 ability to bind cells to bulk materials that do not possess such ability in the absence of surface-immobi!ized collagen-like polymer compounds or their Kemp triacid adducts. The Nleu residue as referred to above is an example of a peptoid residue. As outlined in the Simon et al paper, peptoid residues are a relatively new class of unnatural imino acids which comprise N-substituted glycine residues, wherein the substituents on the nitrogen atom are the a-positioned side chains of amino acids. Given that peptoid residues do not occur in nature, peptoid residues and the collagen-like polymers comprising them have improved resistance to enzymatic attack relative to naturally occurring amino acids and peptides. The present invention includes within its scope collagen-like polymer compounds which include one or more of the same or different peptoid residues. Collagen-like polymer compounds that include any of the peptoid residues are comprehended by the invention, as long as they exhibit the requisite cellular binding. The implantable bulk materials according to the present application are many and varied and include the following which are listed here by way of example: wound repair materials, synthetic skin or connective tissue, ocular implants such as implanted contact lenses and synthetic epikeratoplasties or corneal grafts, orthopaedic implants such as prosthetic joints or synthetic arterial surfaces, synthetic tendon or ligament tissues or materials used to secure bone or ligament in surgical procedures, synthetic neural tissue, prosthetic organs such as apparatus which will carry out the function of the heart, lungs, etc, components of blood contacting devices, immunoassays, antigen/antibody detection kits, affinity matrices etc., other synthetic bioactive apparatus such as heart pacemakers or other synthetic implantable materials. It is to be understood that the examples provided above are not intended to limit the scope of the invention in any way, and that the present invention relates to implantable bulk materials of all types which may need to be implanted into a human or animal body, and will require a surface coating of collagen-like polymers according to the invention to initiate cellular attachment. Another important aspect of the present invention relates to the means by which the collagen-like polymers according to the present invention can be bound to implantable bulk materials. As previously referred to this binding relates to covalent binding which will result WO 98/52620 PCT/EP98/02857 -9 in a permanent interaction between the collagen-like polymer compounds and the implantable bulk material. That is, whereas polymers or peptides simply physisorbed to an implantable bulk material would be likely to undergo exchange with other biomolecules within a cellular or tissue system, covalently bound polymer compounds will not be so affected. For example, it is anticipated that over time physisorbed polymers or proteins would be displaced from the surface of an implantable bulk material by large proteins such as albumin and immunoglobulins, whereas covalently attached polymer compounds are less likely to be so affected. In order to initiate covalent bonding between the collagen-like polymer compounds according to the invention and implantable bulk materials it is necessary for active groups on both of these entities to be either inserted or exploited. Preferably, active groups for the purposes of binding to implantable bulk materials will be present at or near a terminus of the collagen-like polymer compound. In this way it is less likely that binding of the collagen-like polymer compound to the implantable bulk material will adversely affect the confirmation of the collagen-like polymer compound. This does not mean to say however, that is not possible for active groups within the collagen-like polymer compound chain to be exploited for the purposes of binding to an implantable bulk material, although in such cases, it will be necessary to ensure that the activity and conformation of the collagen-like polymer compound is not adversely disrupted. If a mechanically and optionally optically suitable bulk material does not inherently contain on its surface chemical groups capable of reacting with reactive groups on the collagen-like polymer compounds, then a range of suitable chemical groups can be introduced into the surface of the bulk material by application of surface modification techniques known in the art. For instance surface amine groups can be provided by a plasma polymerization treatment using alkylamine vapours (Griesser and Chatelier, Journal of Applied Polymer Science, Vol. 48, pp 361-384, 1990). Particularly useful examples of chemical linkage between the collagen-like polymer compounds and the implantable bulk materials involve the exploitation of terminal primary amine groups on the collagen-like polymer compound which can be reacted with either aldehyde or carboxyl groups on a surface or surfaces of the implantable bulk material. Binding between termina! amine groups and surface aldehyde groups can be used to result WO 98/52620 PCT/EP98/02857 -10 in a Schiff base linkage whereas binding between surface carboxyl groups and terminal amine groups can be used to form an amide linkage between the collagen-like polymer and the implantable bulk materials. As is well known in the art, Schiff base linkages are of a reversible nature and it may therefore be advantageous for them to be reduced to amine linkages, by treatment with an appropriate reducing agent such as, for example, cyanoborohydride. It is also possible for collagen-like polymers according to the invention to be copolymerised with other units in order to produce implantable bulk materials. Naturally, it would be necessary for these other units to be carefully chosen on the basis of properties which would make them suitable for use in the particular bulk material concerned. For example, in the case of implantable contact lenses the units chosen to be combined with the collagen like polymers of the invention would need to exhibit satisfactory stability, permeability, strength, flexibility and optical charactistics for ocular implantation. The invention will now be described further with reference to the following non-limiting examples: Example 1: Cell Binding Activity of Collagen-like polymer Sequences (a) Synthetic collagen-like polymer sequences Polymer Sequence #1 HCI (Gly-Pro-Pro)-NH 2 #2 HCI (Gly-Pro-Nieu)-NH 2 #3 HCI (Gly-Nleu-Pro)-NH 2 #4 Ac-(Gly-Pro-Nleu)-NH 2 #95120 Ac-(Gly-Pro-Hyp)-NH2 #95121 Ac-(Gly-Pro-Nleu) 9

-NH

2 #95122 KTA-[Gly-(Gly-Pro-Nieu) 9

-NH

2 13 1 #95123 KTA-[Gly-(Gly-Pro-Hyp)-NH 2

]

3 #95124 Ac-(Gly-Pro-Hyp)-NH 2 #95106 Ac-(Giy-Pro-Nleu)-NH 2 #96118 H-(Gly-Pro-NIeu) 2

-NH

2 WO 98/52620 PCT/EP98/02857 - 11 Polymer Sequence #96119 H-(Gly-Pro-Nleu) 3

-NH

2 #96120 H-(Gly-Pro-Nleu) 4

-NH

2 #96121 H-(Gly-Pro-Nieu) 5

-NH

2 #96122 H-(Gly-Pro-Nieu) 7

-NH

2 #96123 H-(Gly-Pro-Nieu) 9

-NH

2 #96124 HCI (Gly-Nleu-Pro)-NH 2 #96127 Ac-(Gly-Nleu-Pro) 3

-NH

2 #96128 Ac-(Gly-Nleu-Pro) 6

-NH

2 #96129 Ac-(Gly-NIeu-Pro)jo-NH 2 (b) Determination of the cytotoxicity of synthetic collagen-like polymer sequences: Cells isolated from bovine cornea were used to evaluate the biocompatibility and cytotoxicity of a series of the collagen-like polymer sequences. The assay used, which involves adding the collagen-like polymer to an established culture of adherent cells, tests for any gross toxicity of the collagen-like polymers in soluble form. Stromal fibroblasts were seeded into replicate wells of a 96-well tissue culture polystyrene (TCPS) culture tray and incubated at 370C for four hours to allow them to attach. This was carried out in the absence of any collagen-like polymer. The seeding culture medium was then replaced with fresh culture media containing serial dilutions of each of the collagen-like polymers, from a final concentration of 1.0 mg/ml down to zero. The cells were incubated at 370C for forty hours in the collagen-like polymer-containing culture media. The culture media were then removed and MTS solution added to each well and incubated for a further four hours at 370C to determine the viability of the attached cells. The relative numbers of cells per well was determined colorimetrically on an ELISA plate reader at an absorbance wavelength of 490 nm. The results are shown in Tables 1 (a)-(i). When the collagen-like polymers were added to the culture medium for cells that were already adhered to the culture substratum, and cell culture continued in the presence of the collagen-like polymers for a period of forty hours, there were no apparent ill effects detected. The relative numbers of cells still viable after exposure to the collagen-like polymers were equivalent to that found for cells not exposed (the histogram "bars" with 0 mg/mi in Tables 1 (a)-(i)). These results indicated that the collagen-like polymers do not have any cytotoxic effect. In particular, the -(Gly-Pro- WO 98/52620 PCT/EP98/02857 -12 Nleu) 9 -collagen-like polymer sequence does not have any cytotoxicity. This non-cytotoxicity result indicated that the collagen-like polymers may be used in a cell culture assay to search for cell-binding activity: Tables 1 (a)-(i) show assays of attachment of bovine stromal fibroblasts seeded onto tissue culture polystyrene, then exposed to polymers #1-4 and 95120-95124, after four hours incubation. Polymer concentration (mg/ml) is plotted against absorbance at 490nm. Table 1 (a) Table 1 (b) Concentration Absorbance s.d. Concentration Absorbance s.d. (mg/ml) (490 nm) (mg/ml) (490 nm) 1.000 0.312 0.014 1.000 0.400 0.030 0.500 0.328 0.011 0.500 0.405 0.039 0.250 0.345 0.013 0.250 0.414 0.029 0.125 0.343 0.003 0.125 0.409 0.026 0.063 0.334 0.012 0.063 0.409 0.025 0.032 0.343 0.008 0.032 0.410 0.019 0.016 0.328 0.015 0.016 0.411 0.029 0.008 0.346 0.024 0.008 0.422 0.022 0.004 0.315 0.025 0.004 0.406 0.024 0.002 0.316 0.029 0.002 0.393 0.041 0.000 0.390 0.042 0.000 0.390 0.042 WO 98/52620 PCT/EP98/02857 - 13 Table 1 (c) Table 1 (d) Concentration Absorbance s.d. Concentration Absorbance s.d. (mg/ml) (490 nm) (mg/ml) (490 nm) 1.000 0.368 0.026 1.000 0.422 0.009 0.500 0.384 0.016 0.500 0.439 0.014 0.250 0.389 0.025 0.250 0.430 0.030 0.125 F0.397 0.024 0.125 0.441 0.017 0.2 0.0100 0.063 0.388 0.025 0.063 0.443 0.00 0.032 0.379 0.029 0.032 0.429 0.013 0.016 0.378 0.021 0.016 0.432 0.004 0.008 0.377 0.020 0.008 0.425 0.003 0.004 0.385 0.028 0.004 0.440 0.009 0.002 0.372 0.019 0.002 0.426 0.010 0.000 0.390 0.042 0.000 0.390 0.042 Table 1 (e) Table 1 (f) Concentration Absorbance s.d. Concentration Absorbance s.d. (mg/ml) (490 nm) (mg/ml) (490 nm) 1.000 0.751 0.012 1.000 0.799 0.018 0.500 0.855 0.028 0.500 0.868 0.015 0.250 0.813 0.016 0.250 0.780 0.018 [0.125 0.798 0.023 0.125 0.778 0.008 0.063 0.815 0.011 0.063 0.807 0.022 0.032 0.876 0.024 0.032 0.886 0.021 0.016 0.780 0.037 10.016 0.782 0.014 0.008 0.812 0.012 10.008 0.814 0.031 0.004 0.801 0.012 0.004 0.751 0.027 0.002 0.777 10.016 0.0020.2007 [0.000 1 0.931 1 0.029 10.000 0.761 0.06 WO 98/52620 PCT/EP98/02857 - 14 Table 1(g) Table 1(h) Concentration Absorbance s.d. Concentration Absorbance s.d. (mg/ml) (490 nm) (mg/ml) (490 nm) 1.000 0.741 1 0.052 1.000 0.803 0.033 0.500 0.853 0.016 0.500 0.814 0.044 0.250 0.857 0.029 0.250 0.815 0.012 0.125 0.857 0.028 0.125 0.826 0.002 0.063 0.857 0.012 0.063 0.800 1 0.022 0.032 0.808 0.025 0.032 0.818 0.010 0.016 0.777 0.027 0.016 0.759 0.009 0.008 0.772 0.017 0.008 0.784 0.022 0.004 0.764 0.012 0.004 0.777 0.012 0.002 0.761 0.004 0.002 0.769 0.027 10.000 0.931 0.029 0.000 0.761 0.067 Table 1 (i) Concentration Absorbance s.d. (mg/ml) (490 nm) 1.000 0.769 0.070 0.500 0.820 0.006 0.250 0.822 0.029 0.125 0.794 0.017 0.063 0.780 0.021 0.032 0.769 0.041 0.016 0.788 0.031 0.008 0.785 0.048 0.004 0.799 0.020 0.002 0.818 0.018 0.000 0.820 0.060 WO 98/52620 PCT/EP98/02857 -15 (c) Detection of cell-binding activify of synthetic collagen-like polymer sequences, by inhibition of cell adhesion: The synthetic collagen-like polymer sequences were assayed for a specific binding reaction at the cell surface, consistent with the specific sequence having cell-binding activity. This assay involves using the collagen-like polymer in solution and adding the collagen-like polymer at the time of inoculating the cells, to determine whether the collagen-like polymer acts as an inhibitor of the initial adhesion of cells to the culture substratum. If a collagen-like polymer has a cell-binding activity that enables it to effectively compete for cell membrane receptor sites then the collagen-like polymer may inhibit cell attachment, and so a reduced cell attachment is evidence of that collagen-like polymer having cell-binding activity. These studies were conducted with corneal stromal fibroblasts and with corneal epithelial cells. Stromal fibroblasts were seeded into replicate wells of a 96-well TCPS culture tray in culture medium containing the same seria! dilution range of collagen-like polymers as described in section (b) above. The cells were incubated for twenty five hours at 370C, then the culture media were removed and replaced with MTS as previously described. The relative numbers of cells attached to the substratum at the end of this period, for the situations where there was no collagen-like polymer present or collagen-like polymer was present, were determined colorimetricaily as described in (b) above. For the set of polymers #1-#4, there were no differences in the relative number of cells detected in the presence of the collagen-like polymers when compared to the number found in the control wells without collagen-like polymers (Tables 2(a)-(j)). In the second set of 5 collagen-like polymer sequences tested, however, two of the collagen-like polymers were capable of reducing cell attachment at the highest concentrations of 1 mg/ml and 0.5 mg/ml (Tables 2(g) and 2(h)). These two collagen-like polymers were polymer #95121 (a single chain with the sequence Ac-(Gly-Pro-Nieu) 9

-NH

2 ) and polymer #95122 (a triplet of the same sequence coupled to KTA)). The inhibitory effect was reduced at collagen-like polymer concentrations lower that 0.5 mg/ml. The inhibitory effect was specific to this sequence as other collagen-like polymers tested which contained long or short chains of the alternative sequence -(Gly-Pro-Hyp),-did not reduce cell attachment. Under the above assay conditions, therefore, none of the synthetic collagen-like polymers tested interfered with normal metabolic processes (see Section b above) but two of them demonstrated a cell- WO 98/52620 PCT/EP98/02857 -16 binding activity (based on their ability to interfere with the attachment of bovine corneal fibroblasts to a TCPS cell culture substrate) when present at collagen-like polymer concentrations > 0.25 mg/ml; these two collagen-like polymers containing the sequence -(Gly-Pro-Nleu)n-. Tables 2(a)-(i) show assays of attachment of bovine stromal fibroblasts seeded onto tissue culture polystyrene in the presence of polymers #1-4 and 95120-95124. Polymer concentration (mg/ml) is plotted against absorbance at 490nm. Table 2(a) Table 2(b) Concentration Absorbance s.d. Concentration Absorbance s.d. (mg/ml) (490 nm) (mg/ml) (490 nm) 1.000 0.312 0.014 1.000 0.400 0.030 0.500 ]0.328 0.011 0.500 0.405 0.390 0.250 0.345 0.013 0.250 0.414 0.029 0.125 0.343 0.003 0.125 0.409 0.026 0.063 0.334 0.012 0.063 0.409 0.025 0.032 0.343 0.008 0.032 0.410 0.019 0.016 0.328 0.015 0.016 0.410 0.029 0.008 0.346 0.024 0.008 0.422 0.022 0.004 0.315 0.025 0.004 0.406 0.024 0.002 0.316 0.029 0.002 0.393 0.041 0.000 0.390 0.042 0.0000.390 0.042 WO 98/52620 PCT/EP98/02857 -17 Table 2(c) Table 2(d) Concentration Absorbance s.d. Concentration Absorbance s.d. (mg/ml) (490 nm) (mg/ml) (490 nm) 1.000 0 368 0.026 1.000 0.422 0.009 0.500 0.384 0.016 0.500 0.439 0.014 0.250 0.389 0.025 0.250 0.430 0.003 0.125 0.379 0.024 0.125 0.441 0.017 0.063 0.388 0.025 0.063 0.443 0.005 0.032 0.379 0.029 0.032 0.429 0.013 0.016 0.378 0.021 0.016 0.432 0.004 0.008 0.377 0.020 0.008 0.425 0.006 0.004 0.385 0.028 0.004 0.440 0.009 0.002 0.372 0.019 0.002 0.426 0.010 0.000 0.390 0.042 0.000 0.390 0.042 Table 2(e) Table 2(f) Concentration Absorbance s.d. Concentration Absorbance s.d. (mg/ml) (490 nm) (mg/ml) (490 nm) 1.000 0.616 0.036 1.000 0.091 0.013 0.500 0.600 0.018 0.500 0.300 0.014 0.250 0.609 0.017 0.250 0.437 0.026 0,25 0.607 0.011 0.503 0.008 0.063 0.608 0.011 0.063 0.518 0.015 0.032 0.622 0.021 0.032 0.470 0.004 0.016 0.603 0.062 0.016 0.492 0.017 0.008 0.597 0.007 10.008 0.485 0.017 0.004 0.604 0.024 0 004 0.482 0.022 0.002 0.589 0.007 0.002 0.473 0.018 0.000 0.569 0.047 10.000 0.478 0.018 WO 98/52620 PCT/EP98/02857 -18 Table 2(g) Table 2(h) Concentration Absorbance s.d. Concentration Absorbance s.d. (mg/ml) (490 nm) (mg/ml) (490 nm) 1.000 0.222 0.059 1.000 0.706 0.104 0.500 0.388 0.069 10.500 0.679 0.038 0.250 0.610 0.052 10.250 0.626 0.049 0.125 0.678 0.010 0.125 0.695 0.029 0.063 0.674 0.014 0.063 0.688 0.038 0.032 0.671 0.024 0.032 0.696 0.054 0.016 0.667 0.019 0.016 0.660 0.016 0.008 0.651 0.030 0.008 0.655 0.020 0.004 T0.652 0.032 0.004 0.647 0.042 0.002 0.652 0.013 0.002 0.660 0.025 0.000 0.632 J_0.002 0.000 0.650 0.009 Table 2(i) Concentration Absorbance s.d. (mg/ml) (490 nm) 1.000 0.615 0.023 0.500 0.616 0.020 0.250 0.607 0.029 0.125 0.597 0.041 0.063 0.581 0 0.032 0.588 0.029 0.016 0.727 0.013 0.008 0.698 0.041 0.004 0.675 0.029 0.020.740 0.036 .000.709 0.0261 The ability of collagen-like polymers containing the sequence -(Gly-Pro-Nleu)n- to inhibit the initial attachment of cells was also determined for corneal epithelial cells. Bovine corneal WO 98/52620 PCT/EP98/02857 -19 epithelial cells at passage 1 were seeded onto TCPS in the presence of serially diluted (1:2) collagen-like polymer (1 mg/ml maximal) concentrations and cultured for 24 hours in either serum-free culture medium, or culture medium containing 20% (v/v) serum (depleted of the cell adhesive glycoproteins, fibronectin and vitronectin, Underwood and Bennett, 1989). Removal of fibronectin and vitronectin from the serum ensured that exogenous cell attachment factors would not play a significant role in the assay and mask any effects contributed by the collagen-like polymers. The number of adherent cells was determined colorimetrically using an MTS-based method and then measuring absorbance at 490 nm on a plate reader. Corneal epithelial cell attachment was reduced by 85% in medium containing 1.0 mg/ml of Polymer #95121 (single chain with the sequence Ac-(Gly-Pro-Nleu) 9

-NH

2 -) as compared to serum-free medium alone. Similarly, this collagen-like polymer reduced by 50% the attachment of corneal epithelial cells to TCPS, when cultured in medium containing 20% (v/v) depleted serum (Table 3(b)). Serum-free medium containing 0.5 mg/ml of polymer #95121 reduced cell attachment by 20%. Polymer #95122, comprised of a KTA template linked to three [Gly-(Gly-Pro-Nleu) 9

-NH

2 ] chains, was also found to inhibit attachment of corneal epithelial cells, when employed in serum-free culture medium and at the highest concentrations. Serum-free culture medium containing 1.0 mg/mI of polymer #95122 reduced corneal epithelial cell attachment by 27%, whilst a collagen-like polymer concentration of 0.5 mg/ml resulted in a 19% reduction. The effect was diminished at the lower concentrations (Table 3(c)). Polymer #95106, containing the single sequence Ac-Gly-Pro-Nleu-NH 2 , was included as a short sequence control for the other two collagen-like polymers. There was no inhibition of cell attachment at any concentration of this collagen-like polymer, in either culture medium (Table 3(a)). Tables 3(a)-(c) show assays of attachment of bovine corneal epithelial cells to tissue culture polystyrene after seeding in the presence of Ac-Gly-Pro-Nleu-NH 2 (a), Ac-(Gly-Pro-Neu) 0

-NH

2 (b) and KTA-[Gly-(Gly-Pro-Neu)-NH 2

]

3 . Polymer concentration (mg/ml) is plotted against absorbance at 490nm.

WO 98/52620 PCT/EP98/02857 - 20 Table 3(a) Serum Free Medium Serum Present Concentration Absorbance s.d. Concentration Absorbance s.d. (mg/ml) (490 nm) (mg/ml) (490 nm) 1.000 0.503 0.025 1.000 10.663 0.002 0.500 0.526 0.020 0.500 0.652 0.043 0.250 0.545 0.006 0.250 0.651 0.056 0.125 0.559 0.010 0.125 0.643 0.004 0.063 0.543 0.011 10.063 0.642 0.008 0.032 0.539 0.018 0.032 0.655 0.027 0.016 0.550 0.010 0.016 0.675 0.032 0.008 0.545 0.018 0.008 0.632 0.036 0.004 0.551 0.026 0.004 0.596 0.036 0.002 0.546 0.022 0.002 0.592 0.009 0.000 0.580 0.018 0.000 0.704 0.008 Table 3(b) Serum Free Medium Serum Present Concentration Absorbance s.d. Concentration Absorbance s.d. (mg/ml) (490 nm) (mg/ml) (490 nm) 1.000 0.094 0.030 1.000 0.324 0.030 0.500 0.478 0.013 0.500 0.570 0.017 0.250 0.524 0.025 10.250 0.615 0.006 0.125 0.515 0.030 0.125 0.618 0.018 0.063 0.514 0.028 0.063 0.629 0.015 0.032 0.494 0.024 0.032 0.623 0.024 0.016 0.512 0.017 0.016 0.636 0.010 10008 0.500 0.025 0.008 0.640 0.017 01004 0.513 0.014 0.004 0.643 0.044 0.002 0.515 0.013 0.002 0.644 0.019 [0.602 0.027 0000 0.630 0.012 0.000 [ 0.000__ _______ _______ ____ WO 98/52620 PCT/EP98/02857 - 21 Table 3(c) Serum Free Medium Serum Present Concentration Absorbance s.d. Concentration Absorbance s.d. (mg/ml) (490 nm) (mg/ml) (490 nm) 1.000 0.430 0.035 1.000 0.616 0.037 0.500 0.480 0.008 0.500 0.656 0.008 0.250 0.514 0.019 0.250 0.662 0.021 0.125 0.515 0.015 0.125 0.652 0.027 0.063 0.491 0.026 0.063 0.662 0.016 0.032 0.530 0.019 0.032 0.665 0.006 0.016 0.545 0.016 0.016 0.667 0.003 0.008 0.549 0.026 0.008 0.647 0.004 0.004 0.528 0.017 0.004 0.640 0.019 0.002 0.544 [0.016 0.002 0.644 0.007 0.000 0.591 0.017 0.000 0.630 0.007 These results show that collagen-like polymers containing the -(Gly-Pro-Nleu)r- sequence effectively inhibited the attachment of corneal epithelial cells to tissue culture polystyrene, both in the presence and absence of serum. Therefore, the collagen-like polymer sequence interfered with the attachment of both corneal stromal fibroblasts and corneal epithelial cells, indicating that this sequence has cell-binding activity. Furthermore, the efficacy of the alternative sequence, -(Gly-Nleu-Pro)n-, was determined. This alternative sequence -(Gly-Nleu-Pro)j 0 - collagen-like polymer was prepared on a laboratory scale by the group at University of California, San Diego (UCSD) and under their direction on a commercial scale by BioResearch of San Diego. Corneal epithelial cells were cultured for seven days and thereafter the relative numbers in the cultures were determined by colorimetrically measuring the amount of formazan product produced by the cells after exposure to MTT tetrazolium salt (Promega cell proliferation assay). The following free collagen-like polymers were assessed for the ability to inhibit attachment of the epithelial cells to TCPS when the cells were seeded in the presence of collagen-like polymer containing culture medium: WO 98/52620 PCT/EP98/02857 - 22 Polymer Sequence #96119 H-(Gly-Pro-Nleu) 3

-NH

2 #96121 H-(Gly-Pro-Nleu) 5

-NH

2 #96127 Ac-(Gly-Nleu-Pro) 3

-NH

2 #96128 Ac-(Gly-Neu-Pro)6-NH2 #96129 Ac-(Gly-Nleu-Pro) 10

-NH

2 Two commercially available "RGD"-containing peptides, that have cell-binding activity and are known to promote cell attachment (Pierschbacher and Ruoslahti 1984, 1987), were used as controls for the attachment competition assays: i) Gly-Arg-Gly-Asp-Ser-Pro-Cys (RGDSPC, cat #PO1 1), obtained from Telios Pharmaceuticals, and ii) Gly-Arg-Gly-Asp-Ser-Pro-Lys (RGDSPK, cat#12145-017), obtained from Life Technologies Inc. Corneal epithelial cell attachment to tissue culture polystyrene (TCPS) was not inhibited by the presence of any of the -(Gly-Nleu-Pro)n- collagen-like polymers (Tables 4(a)-(c)). Corneal epithelial cell attachment was rot inhibited by either short sequence repeats (Ac-(G!y-NIeu-Pro) 3

-NH

2 , polymer #96127, Table 4(a), or longer ones (Ac-(Gly-Nleu-Pro),o

NH

2 (Bioresearch commercial version), Table 4(b), either under serum-free or serum containing conditions. Only the GRGDSPK control peptide (Tables 4(c); 5(c)) showed inhibition of corneal epithelial cell attachment to TCPS in this experiment, reducing the number of cells that adhered in the absence of serum (represented by a reduction in the absorbance reading at 595 nm) by approximately 30%. Collagen-like polymers with the alternative -(Gly-NIeu-Pro)r- sequence that were used in the cell attachment competition assays consistently failed to reduce corneal epithelial cell attachment under both serum-free and serum-containing conditions. These collagen-like polymers were used to determine whether the number of repeats of the polymer sequence, -(Gly-Pro-Nleu)n- affects cell-binding activity, evident as corneal epithelial cell attachment, when the cells are seeded in the presence of that sequence.

WO 98/52620 PCT/EP98/02857 - 23 Corneal epithelial cell attachment to TCPS was also not affected in the presence of the shorter, i.e. < 9, -(Gly-Pro-NIeu)n- sequences (n=2 to n=7, polymers #96118 to #96122). The representative histograms, see Tables 5(a)-(c), show there was no measurable reduction in the number of cells attached in the presence of collagen-like polymer, relative to the control wells containing no collagen-like polymer. Tables 4(a)-(c) show assays of attachment of bovine corneal epithelial cells to tissue culture polystyrene after seeding in the presence of Ac-(Gly-Nleu-Pro) 3

-NH

2 (a), Ac-(Gly-Nleu-Pro)lo-NH2 (b) and Gly-Arg-Gly-Asp-Ser-Pro-Lys. Polymer concentration (mg/ml) is plotted against absorbance at 595nm. Tables 5(a)-(c) show assays of attachment of bovine corneal epithelial cells to tissue culture polystyrene after seeding in the presence of H-(Gly-Pro-Nleu) 3

-NH

2 (a), H-(Gly-Pro-Nleu) 5

-NH

2 (b) and Gly-Arg-Gly-Asp-Ser-Pro-Lys. Polymer concentration (mg/ml) is plotted against absorbance at 595nm. Table 4(a) Serum Free Medium Serum Present Concentration Absorbance s.d. Concentration Absorbance s.d. (mg/ml) (490 nm) (mg/ml) (490 nm) 1.000 0.244 0.005 1.000 0.264 0.028 0.500 0.232 0.010 0.500 0.287 0.036 0.250 0.231 10.002 0.250 0.282 0.018 0.125 0.244 0.010 0.125 0.290 10.032 0.063 0.241 0.009 1 0.307 0.038 0.032 0.239 0.009 0.032 0.293 0.021 0.016 0.228 0.016 0.016 10.294 0.004 0.008 0.226 0.003 0.008 0.312 0.004 0.004 0.235 0.009 0.004 0.305 0.008 0.002 0.233 0.009 0.002 0.313 0.018 0.000 0.239 0.006 0.000 0.311 0.015 WO 98/52620 PCT/EP98/02857 - 24 Table 4(b) Serum Free Medium Serum Present Concentration Absorbance s.d. Concentration Absorbance s.d. (mg/ml) (490 nm) (mg/ml) (490 nm) 1.000 0.286 0.011 1.000 0.271 0.009 0.500 0.272 0.007 0.500 0.314 0.015 0.250 0.276 0.007 0.250 0.340 0.012 0.125 0.284 0.017 0.125 0.340 0.010 0.063 0.268 0.013 0.063 0.337 0.015 0.032 0.252 0.012 0.032 0.320 0.028 0.016 0.242 0.006 0.016 0.311 0.021 0.008 0.240 0.013 0.008 0.316 0.017 0.004 0.235 0.005 0.004 0.309 0.015 0.002 0.241 0.003 0.002 0.312 0.012 0.000 0.249 0.003 0.000 0.326 0.014 Table 4(c) Serum Free Medium Serum Present Concentration Absorbance s.d. Concentration Absorbance s.d. (mg/ml) (490 nm) (mg/ml) (490 nm) 1.000 0.168 0.025 1.000 0.312 0.019 0.500 0.213 0.015 0.500 0.335 0.018 0.250 0.237 0.011 0.250 0.346 0.016 0.125 10.264 0.006 0.125 0.355 0.013 0.063 0.277 0.004 0.063 0.350 0.012 0.032 10.277 0.004 0032 0.349 0.011 0.016 10.268 0.003 0.016 0.334 10.023E 0.008 0.278 0.029 0.008 0.340 0.021 0.004 10.263 0.014 0.004 0.331 0.008 0.002 0.266 0.016 0.002 0.339 0.006 0.000 0.263 0.005 0.000 0.351 0.016 WO 98/52620 PCT/EP98/02857 - 25 Table 5(a) Table 5(b) Concentration Absorbance s.d. Concentration Absorbance s.d. (mg/ml) (490 nm) (mg/ml) (490 nm) 1.000 0.060 0.005 1.000 0.073 0.003 0.500 0.072 [0.004 0.500 0.073 0.007 0.250 0.072 0.002 0.250 0.069 0.001 0.125 0.065 10.002 0.125 0.067 0.004 0.063 0.065 0.003 0.063 0.065 0.003 0.032 0.063 [0.002 0.032 0.064 0.004 0.016 0.057 0.003 0.016 0.061 0.005 0.008 0.056 0.002 0.008 0.058 0.004 0.004 0.061 0.002 0.004 0.065 0.001 0.002 0.059 0.004 0.002 0.060 0.002 0.000 0.063 0.001 0.000 0.065 0.008 Table 5(c) Concentration Absorbance s.d. (mg/ml) (490 nm) 1.000 0.041 0.004 0.500 0.044 0.003 0.250 0.049 0.001 0.125 0.045 0.002 0.063 0.050 0.001 0.032 0.053 0.004 0.016 0.061 0.003 0.008 0.059 0.005 0.004 0.051 0.005 0.002 0.058 0.003 0.000 0.063 0.008 These results show that the collagen-like polymer having the sequence of formula 1, wherein (y) denotes 2: 9: WO 98/52620 PCT/EP98/02857 - 26 -(Gly-Pro-Nleu) (y) - (1) has cell-binding activity, as indicated by: i) Sequences containing -(Gly-Pro-Nleu)n- are effective at inhibiting cell attachment in vitro providing the number of repeats ("n") is at least nine. This occurs in both singlet and KTA triplet versions, and was demonstrated with the two different cell types epithelial cells and fibroblasts; ii) This effect is quite specific to the -(Gly-Pro-Nleu) sequence, as the alternative sequence -(Gly-Nleu-Pro)n-, including where n is up to ten repeats long, have no effect on corneal epithelial cell attachment in vitro. References relevant to this example are as follows: Pierschbacher, M.D. and Ruoslahti, E., (1984) Proc.Natl.Acad.Sci.USA, 81:5985; Pierschbacher, M.D. and Ruoslahti, E., (1987) J.Biol.Chem, 262:17294; Underwood, P.A. and Bennett, F.A. (1989) J.Cell.Sci.,93:641, which are each included herein by reference in their entirety. Example 2: Chemical Linkage of Collagen-like polymers to Implantable Bulk Material Collagen-like polymers or the Kemp triacid adduct of collagen-like polymers can be covalently immobilized permanently onto synthetic surfaces by a number of covalent interfacial linking strategies which are known in the art. For instance, it is practical to utilize collagen-like polymer molecules that have been equipped with a terminal primary amine group (Structure 1), 0 0 H-N N N NH 2 H 0 Structure 1 : Structure of amine-endfunctionalized collagen-like polymer molecule. since such amine-functionalized collagen-like polymer molecules can be immobilized by a number of interfacial covalent reaction schemes onto surfaces that possess various WO 98/52620 PCT/EP98/02857 - 27 chemical groups capable of reacting with primary amine groups. Examples of surfaces that can covalently immobilize amine-endfunctionalized collagen-like polymers comprise: (a) Activation with reactive surface aldehyde groups If the implantable bulk material does not contain reactive surface aldehyde groups, such aldehyde groups can be provided using established surface modification techniques, e.g., reacting surface hydroxyl groups with glutaraldehyde (Gotoh et al., Journal of Membrane Science, 41, 1989, 291-303 ; Zhuang and Butterfield, Journal of Membrane Science, 66, 1992, 247-257), or providing surface aldehyde groups by gas plasma methods. After amine terminated collagen-like polymer molecules have been reacted with surface aldehyde groups to form a Schiff base linkage, this linkage is then optionally stabilized with sodium cyanoborohydride, to result in a -NH-CH 2 - linkage, to confer better resistance to slow, thermally induced reversion back to amine and aldehyde groups. (b) Activation with reactive surface carbonyl groups By way of example, in a first step, a mechanically and optionally optically suitable implantable bulk material is coated with a thin plasma polymer containing amine groups (Griesser and Chatelier, Journal of Applied Polymer Science: Applied Polymer Symposium, 46, 1990, 361-384); intermediate reagent molecules containing at least two carboxylic acid groups, or at least two functional groups capable of forming carboxylic acid groups upon e.g. hydrolysis (for instance, succinyl chloride, PEG-dicarboxylate, carboxymethyldextran, hyaluronic acid, PAMAM-COOH starburst dendrimer) are then reacted with the amine surface to form interfacial amide linkages via EDC/NHS catalysis, a procedure leaves free carboxylic acid and/or carboxylate groups oriented away from the synthetic material for subsequent reaction with amine-endfunctionalized collagen-like polymers to form an amide linkage between collagen-like polymer and the bulk material. This situation, although already disclosed to a person skilled in the art by the above explanation, is further illustrated in Figures 7 and 8 of the priority document AU 6862/97 of the present invention, the disclosure of said Figures being incorporated herein by reference. Similarly, interfacial linking reaction schemes for covalently attaching the Kemp triacid adduct can be designed on the basis of general knowledge of established chemical reactions. For example, the KTA adduct can form amide linkages with an alkylamine plasma modified surface or a bisamino-PEG spacer attached to an aldehyde containing surface.

WO 98/52620 PCT/EP98/02857 - 28 Such linkage reaction schemes between collagen-like polymer and the bulk material have been illustrated in Figures 1 to 8 of the priority document AU 6842/97 of the present invention, the disclosure of said Figures being incorporated herein by reference. A full description of representative collagen-like polymer immobilization reactions and results describing the biological activities of surfaces containing immobilised collagen-like polymer sequences is included below. Example 3: Coupling of Peptoide Comprising Collagen-like polymers to FEP This section describes the procedures used for coupling the synthetic collagen-like polymer sequences onto a surface that does not itself support cell colonisation: fluorinated ethylene propylene, FEP. Surface analyses results that demonstrate that the collagen-like polymers were immobilised onto the surface are described. Biological assessment of the surface containing the immobilised synthetic collagen-like polymer sequences was performed by measuring the attachment and growth of bovine corneal epithelial cells on the surfaces over a seven days period. 1. Methods (a): Description and characterisation of coupling of synthetic collagen-like polymer sequences to FEP substrates The following collagen-like polymers were used in these experiments: Polymer Sequence #95104 HCI (Gly-Pro-Nieu)-NH 2 #95105 HCI (Gly-Nleu-Pro)-NH 2 #96118 H-(Gly-Pro-Nleu) 2

-NH

2 #96119 H-(Gly-Pro-Nleu) 3

-NH

2 #96120 H-(Gly-Pro-Nleu) 4

-NH

2 #96121 H-(Gly-Pro-Nieu)s-NH2 #96122 H-(Gly-Pro-Nleu)-NH 2 #96123 H-(Gly-Pro-Nleu)-NH 2 #96124 HCI (Gly-Nleu-Pro) 9

-NH

2 WO 98/52620 PCT/EP98/02857 - 29 Acetaldehyde plasma polymer (AApp) or heptylamine plasma polymer (Happ), deposited on FEP, were used to achieve linkage of synthetic collagen-like polymer sequences through the presence of the free amine group on the collagen-like polymer molecule. a) Acetaldehyde Plasma Polymer: Surface activation was undertaken by exposing the FEP to acetaldehyde monomer in a RF plasma. The process results in the formation of aldehyde groups on the surface which can then react to form a collagen-like polymer-surface linkage. Sodium cyanoboro hydride (NaCNBH 3 ), in excess, was used as a reducing agent. b) Heptylamine Plasma Polymer: Surface activation using heptylamine process vapour in a RF plasma produces amine groups on the surface, which are then succinylated to generate carboxyl groups on the surface. The amine containing collagen-like polymer is coupled to the carboxyl groups using EDC/NHS. In an alternative scheme, amines on the heptylamine plasma polymer treated FEP surface are exposed to starburst dendrimers as intermediates in the collagen-like polymer coupling methodology. The multiple (32) carboxylic acid groups on each dendrimer molecule provide a higher surface density of carboxylic acid groups than the succinylation approach and provide a high degree of in-built redundancy. Acetaldehyde Plasma Polymer deposition: Fluorinated ethylene propylene (FEP) Teflon 100 Type A DuPont (1 2.7mm in width), was used as received and plasma polymer deposited in a custom-built plasma apparatus, employing a Javac DDL 300 rotary high vacuum pump and an ENI HPG-2 High Frequency plasma generator. Acetaldehyde monomer was used as purchased from Aldrich, 99%, cat# 11,007-8, Mw 44.05). Treatment conditions (reactor base pressure (Pb) monomer pressure (Pm), plasma ignition power (P) and plasma frequency (F), plasma treatment duration (T), monomer flux and pressure during treatment) were all carefully monitored and controlled. Standard conditions were used, being Pb= 0.025 Torr, Pm = 0.300 Torr, P = 5 Watts, F = 125 kHz, T = 60 sec. Heptylamine Plasma Polymer deposition: Heptylamine monomer was used as purchased from Aldrich (99%, Cat# 12,680-2, M.Wt. 115.2). A similar plasma apparatus was used with the following conditions Pb = 0.015 Torr, Pm = 0.130 Torr, P = 20 Watts, F = 200 kHz, T = 20 sec. To succinylate the HApp surface, freshly prepared samples (< 2 hours old) were placed in 25ml of a 10% (v/v) solution of succinyl chloride in acetone for 1 hour, rinsed twice in WO 98/52620 PCT/EP98/02857 - 30 acetone then soaked in acetone for 1 hour. The samples were then rinsed two times in water and soaked for 1 hour in MilliQ water. For the attachment of the starburst dendrimer freshly prepared heptylamine surfaces were exposed to a solution (1 mg/ml) of the dendrimer (PAMAM Generation 2.5, Aldrich Cat# 41241-4, M.Wt. 6011, 32 carboxyls / dendrimer) for 1 hour to allow the carboxyl containing dendrimer to react with the amines on the surface. EDC (1 -ethyl-3-(3-dimethylaminopropyl)carbodiimide (Sigma, Cat#E0388) and NHS (N-hydroxysuccinimide) (Aldrich, Cat# 13067-2, MW 115.09) were then added at 19.2 and 11.5 mg/I respectively and mixed rapidly. During the next 60 minutes bubbles which form on the underside of the FEP are frequently removed. The reaction is allowed to proceed overnight at room temperature. Collagen-like polymer Linkage/Detection Collagen-like polymer coating solution was prepared by dissolving synthetic collagen-like polymer sequence in sterile phosphate buffered saline (PBS), pH 7.4, to either 50 [g/ml end concentration for acetaldehyde linkage; or in sterile MilliQ water for heptylamine linkage (phosphate interferes with the EDC/NHS reaction). The polymer #96123 was used as a model polymer for attachment studies. FEP samples with freshly prepared AApp surfaces (< 2 hrs old) were left to react with this coating solution (overnight at 40C and subsequently 2 hr at room temperature). Excess NaCNBH 3 (Sigma, 90%, Cat# S-8628, lot# 33H3669, Mw 62.84) was added. Thereafter samples were washed (3 x 5 minutes) in sterile MilliQ water and air-dried prior to analysis by XPS. Succinylated and starburst dendrimer HApp surfaces were left to react with the collagen-like polymer containing solution for 1 hour. This allows the amine containing collagen-like polymer to adsorb to the carboxyl groups on the surface. EDC and NHS are then added at 19.2 and 11.5 mg/ respectively and mixed rapidly. During the next 60 minutes, bubbles which form on the underside of the FEP are frequently removed. The reaction is allowed to proceed overnight at room temperature. The FEP samples are washed three times with water, then soaked in water for 1 hour and air-dried prior to analysis by XPS. To assess the effect of the reducing agent on the strength of the collagen-like polymer-FEP bonding, samples were prepared with and without the use of NaCNBH 3 or EDC/NHS and analysed (XPS) before and after autoclaving (15 minutes, 120 PSI). Samples were prepared and are identified under the following codes: WO 98/52620 PCT/EP98/02857 -31 DCP Code UCSD Code Sample KM1:42:1 96118 FEP + AApp + NaCNBH3 + H-(Gly-Pro-Nleu)2-NH2 KM1:25:1 96119 FEP + AApp + NaCNBH 3 + H-(Gly-Pro-Nieu)3-NH 2 KM1:42:2 96120 FEP + AApp + NaCNBH 3 + H-(Gly-Pro-Nleu)4-NH 2 KM1:25:2 96121 FEP + AApp + NaCNBH 3 + H-(Gly-Pro-NIeu)s-NH 2 KM1:42:3 96122 FEP + AApp + NaCNBH 3 + H-(Gly-Pro-NIeu)-NH 2 KM1:50:1 96123 FEP + AApp + NaCNBH 3 + H-(Gly-Pro-NIeu)-NH 2 KM1:25:3 96124 FEP + AApp + NaCNBH 3 + H-(Gly-Nleu-Pro) 9

-NH

2 KM1:25:4 FEP + AApp + NaCNBH 3 + Collagen KM1:25:5 FEP + AApp KM1:25:6 96119 FEP + HApp + EDC/NHS + H-(Gly-Pro-Nleu) 3

-NH

2 KM1:25:7 96121 FEP + HApp + EDC/NHS + H-(Gly-Pro-Nleu) 5

-NH

2 KM1:25:8 96124 FEP + HApp + EDC/NHS + H-(Gly-Pro-Nleu)-NH 2 KM1:25:9 FEP + HApp + EDC/NHS + Collagen KM1:25:10 FEP+ HApp KM1:44:4 FEP + AApp + NaCNBH 3 + Collagen KM1:44:5 FEP + AApp The collagen (Vitrogeni 00, Collagen Corp, Palo Alto, CA, USA) coupled surfaces were included as positive controls in the cell colonization assay as this combination has been previously shown to support good BCEp cell attachment and growth. Characterization of the coating procedures was conducted using X-ray photoelectron spectroscopy (XPS) (Kratos AXIS HSi, with monochromator), which allowed for the quantitative analysis of surface elemental composition. Fluorine (FEP), oxygen and carbonyl moieties (plasma polymer) and nitrogen (polymer and heptylamine plasma polymer) were used as indicators of surface modification. 1. Methods (b): Cell culture evaluation of the surfaces containing synthetic collagen-like polymer sequences WO 98/52620 PCT/EP98/02857 - 32 Bovine corneal epithelial (BCEp) cell attachment and growth to collagen-like polymers coupled to FEP Synthetic collagen-like polymer sequences were covalently coupled to FEP film using the above procedures. To measure the amount of cell growth support provided by the immobilized collagen-like polymers, BCEp cells were cultured for seven days. After this culture period, the relative numbers determined by either metabolically labelling the cells at day six with : 5 S-methionine and measuring the amount of incorporated 35 S following a further culture period of several hours, or by colorimetrically measuring the amount of formazan produced by the cells after exposure to MTT tetrazolium salt (Promega cell proliferation assay). The culture medium contained 20 % (v/v) foetal bovine serum (FBS) depleted of the cell adhesive glycoproteins, fibronectin and vitronectin (Underwood and Bennett, 1989). Removal of fibronectin and vitronectin from the serum ensured that exogenous cell attachment factors would not play a significant role in the assay, or mask any effects contributed by the coupled collagen-like polymers. 2. Results and Discussion (a): Description and characterisation of coupling of collagen-like polymers to model substrates: Acetaldehyde Plasma: XPS analysis (Table A) of acetaldehyde plasma polymers on FEP showed the absence of a fluorine signal, indicating that the Aapp film coating was at least 10-1 5nm thick (sampling depth of XPS). The incorporation of oxygen and carbonyls is a further indication of surface activation by AApp deposition. When bringing these surfaces in contact with a solution containing the polymer # 96123 (50 [ig/ml), incorporation of nitrogen (indicative of the polymer) was detected. This incorporation occurs in both the presence and absence of the NaCNBH 3 from the coating solution. To assess the strength of the polymer bonding, coatings with and without the NaCNBH 3 were autoclaved. In both the presence and absence of the reducing agent, some nitrogen signal is lost in the autociaving process (indicating removal of unbound polymer and possibly the thermal effects of the harsh autoclaving step), but a higher nitrogen signal is retained in the presence of the NaCNBH 3 ; this observation is in accord with the known lesser thermal stability of a Schiff base compared with an amine interfacial bond. When collagen-like polymer is allowed to adsorb to the FEP surface without AApp, no nitrogen was found on the surface (see Table A); thus, physisorption is highly ineffective and thus not an option when one wants to immobilize collagen-like polymers of the present invention onto FEP-based bulk materials.

WO 98/52620 PCT/EP98/02857 - 33 Sample Description i DCP % % Code Carbon Oxygen Nitrogen Fluorine FEP + AApp KM51:1 83.2 15.6 0.1 0.1 FEP + Aapp + Peptoid KM51:3 80.9 16.1 2.1 0.1 FEP + Aapp + Peptoid + Autoclave KM51:4 81.6 16.1 1.3 0.2 FEP + Aapp + Peptoid + NaCNBH 3 KM51:5 80.4 15.9 2.8 0.1 FEP + Aapp + Peptoid + NaCNBH 3 KM51:6 81.9 15.4 1.7 0.1 + Autoclave FEP + Peptoid KM51:9 35 0 0.1 64.7 FEP + Peptoid + Autoclave KM51:10 32.8 0 0 67.1 Table A: Results of XPS analysis of attachment of polymer 96123 to acetaldehyde plasma polymer. Quantitative analysis of XPS results suggests that the density of polymer attachment is in the range of 1 molecule per 1.8 -7.2 nm 2 . This value appears reasonable, as we would not expect formation of an oriented, close-packed, covalently coupled polymer layer to occur; rather, a scaling theory description of the attached polymers would seem more appropriate, considering partial folding of the polymer molecules. Heptylamine Plasma: Following heptylamine plasma polymer treatment, the fluorine signal from the FEP was not detected and nitrogen was detected, indicative of the presence of the heptylamine plasma polymer with a thickness >1 Onm. Following succinylation and incubation of HA coated FEP in a solution of polymer # 96123, an increase in the nitrogen signal is observed both in the presence and absence of EDC/NHS (Table B). Following autoclaving, nitrogen signal is lost from surfaces, both plus and minus EDC/NHS, (removal of bound polymer) but a higher nitrogen signal is retained in the presence of EDC/NHS. The interpretation of these results is complicated by the presence of nitrogen, in both the heptylamine plasma polymer and in the polymer. High resolution analysis of the XPS N 1s signa!, however, shows that after attachment of the collagen-like polymer there is a marked increase in the amide to amine nitrogen signal. The collagen-like polymer molecule contains amide nitrogen bonds and the result is indicative of covalent polymer attachment WO 98/52620 PCT/EP98/02857 - 34 Sample Description DCP % % % Code Carbon Oxygen Nitrogen Fluorine HApp + succinylation KM12:1 82.4 9.8 5.4 0.1 HApp + succinylation + Peptoid KM12:2 82.8 10.3 6.6 0.1 96123 HApp + succinylation + Peptoid KM12:3 81.2 11.2 6.1 0.1 96123 + Autoclave HApp + succinylation + Peptoid KM12:4 80.4 10.9 8.0 0.1 96123 + EDC/NHS I HApp + succinylation + Peptoid KM12:5 81.0 11.4 7.1 0.1 96123 + EDC/NHS + Autoclave I Table B: Results of XPS analysis of attachment of polymer 96123 to heptylamine plasma polymer. These results show that synthetic collagen-like polymers are present on the FEP surface following activation, using both acetaldehyde and heptylamine plasma polymerisation. Following autoclaving, collagen-like polymers remain attached to the surface. Surfaces with attached collagen-like polymers were prepared using both attachment strategies and evaluated for support of cell colonisation in vitro as discussed below. 2. Results and Discussion (b): Cell culture evaluation BCEp Cell Growth on Immobilised Collagen-like polymers: BCEp cells were seeded onto plasma-modified FEP surfaces that had covalently coupled collagen-like polymers attached. The cells were cultured for seven days and relative numbers determined by either metabolically labelling the cells at day six with 3 5 S-methionine and measuring the amount of incorporated 3 "S, or by colorimetrically measuring the amount of formazan produced by the cells after exposure to MTT tetrazolium salt (Promega cell proliferation assay). Table 6 shows that the surfaces containing the covalently attached synthetic collagen-like polymer sequences supported cell colonisation at levels similar to that of the positive control surface, tissue culture polystyrene (TCPS). Based on these results, immobilization of collagen-like WO 98/52620 PCT/EP98/02857 - 35 polymers onto an FEP surface through a strategy that involves activation of the FEP surface through deposition of a plasma film does produce a surface that provides enhanced BCEp cell attachment. Table 6 Material Relative cell no. (% of that to TCPS) s.d. A 100.0 2.6 B 100.8 2.6 C 140.1 5.2 D 92.9 5.2 E 103.9 7.9 F 96.1 7.9 G 75.4 7.9 H 99.7 20.9 1 73.8 2.6 J 76.2 5.2 [K 12.6 0.0 Example 4: Corneal epithelial cell attachment and growth on covalently immobilised -Gly-Pro-Nieu- and -Gly-Nieu-Pro- sequences The ability of the collagen-like peptoid sequences to support the attachment and growth of cells, when these sequences were covalently attached to a surface, was tested as follows. Disaggregated bovine corneal epithelial cells (at cell culture passage 2) were seeded onto a series of surfaces. These surfaces were constructed on top of fluoroethylene propylene (FEP) film, by the deposition of a crosslinked polymer (by RF plasma treatment using heptylamine, HApp, monomer during the plasma treatment). Onto this plasma-deposited polymer surface was covalently coupled a carboxymethyldextran (CMD) spacer molecule and then various terminal residues. included in these terminal residues were two of the collagen-like peptoid sequences, H-(Gly-Pro-Nleu) 10 -Gly-Pro-NH 2 (GPNI) and H-(Gly-Nieu Pro) 1 o-NH 2 (GNIP). Other surfaces included the following control surfaces: bovine collagen type I, bovine plasma fibronectin, polylysine, bovine serum albumin. The intermediate surface construct, the CMD alone surface, was also included as a control. A tissue culture WO 98/52620 PCT/EP98/02857 - 36 polystyrene (TCPS) surface was used as a positive control surface, as this surface supports the attachment and growth of cells. The corneal epithelial cells were cultured in a culture medium supplemented with 20% (v/v) of a treated foetal calf serum. This serum had previously been depleted of fibronectin and vitronectin, which are cell attachment glycoproteins. As a result of this treatment process, this serum that does not in itself support cell attachmentand so any cell attachment activity that is observed in the test can be ascribed to an activity of the material surface itself with the cells. The culture medium was replenished every second day, and so any unattached cells were removed at days 2, 4 and 6 of culture. After seven days culture, the number of cells attached to the various surfaces was measured using a formazan-dye based assay (using MTT). Cell numbers were expressed as a mean percentage (± standard deviation of the mean) of that number found attached to the TCPS control surface, which was set as 100%. The CMD alone surface provided minimal (4 ± 1%) support of cell attachment and growth. Similarly, the control surface that contained coupled bovine serum albumin or coupled polylysine supported only a low level of activity (22 ± 6% for the albumin surface; 27 ± 12% for the polylysine surface). The surface that contained coupled fibronectin supported effective cell attachment and growth (71 ± 14%) as expected, and the coupled collagen I surface supported cell attachment and growth (61 ± 3%). The immobilised collagen-like peptoid sequence H-(Gly-Pro-Nieu) 10 -Gly-Pro-NH 2 very effectively supported the attachment and growth of corneal epithelial cells, at the level of 52 ± 4 % of that found on the TCPS control surface. However the alternative collagen-like peptoid sequence, H-(Gly-NIeu-Pro)1o

NH

2 , did not support cell attachment and growth, with activity of 6 ± 2% of that found on the TCPS control. It is concluded that the immobilised collagen-like peptoid sequence H-(Gly Pro-Nieu) 1 o-Gly-Pro-NH 2 is active for support of cell attachment and growth.

Claims (20)

1. A biocompatible material comprising a bulk material which has covalently bound to a surface or surfaces thereof one or more biopolymer compounds, said biopolymer compounds comprising at least one peptoid residue.

2. The biocompatible material according to claim 1 wherein said bulk material comprises metals, metallic alloys, plastics polymers or ceramics.

3. The biocompatible material according to either claim 1 or claim 2 wherein said biopolymer compounds comprise multimeric adducts of peptoid residue comprising biopolymers.

4. The biocompatible material according to claim 3 wherein said multimeric adducts are Kemp triacid adducts.

5. The biocompatible material according to any one of claims 1 to 4 wherein said biopolymer compounds comprise primary amine groups.

6. Collagen-like biopolymer compounds capable of initiating cellular attachment thereto, said polymer compounds comprising one or more peptoid residues.

7. Polymer compounds according to claim 6 capable of initiating cellular attachment thereto, said polymer compounds comprising the sequence of formula I, wherein (y) denotes a 9: -(Gly-Pro-Nleu) (Y) - (1).

8. Polymer compounds comprising Kemp triacid adducts of the polymer compounds claimed in either claim 6 or claim 7.

9. Polymer compounds according to any one of claims 6 to 8, comprising primary amine groups. WO 98/52620 PCT/EP98/02857 - 38

10. The biocompatible material according to claim 4 wherein said covalent binding is between aldehyde or carboxyl groups on said surface or surfaces, and said primary amine groups.

11. The biocompatible material according to claim 10 wherein said covalent binding comprises Schiff base linkages, optionally further reduced to amine linkages.

12. The biocompatible material according to claim 10 wherein said covalent binding comprises amide linkages.

13. The biocompatible material according to any one of claims 1 to 12, which comprises a wound repair material, a contact lens, a synthetic epikeratoplasty, synthetic skin or connective tissue, an orthopaedic implant, prosthetic joint or synthetic arterial surface,synthetic neural tissue, a prosthetic organ, components of blood contacting devices, antigen/antibody detection kits, immunoassays, affinity matrices or other synthetic bioactive apparatus or material.

14. A method of binding biopolymer compounds to a bulk material, which biopolymer compounds comprise at least one peptoid residue, as well as primary amine groups, the method comprising the steps of: (a) providing aldehyde or carboxyl groups on a surface or surfaces of said bulk material; (b) reacting said primary amine groups with aldehyde or carboxyl groups to form covalent bonds.

15. The method according to claim 14 wherein said covalent bonds comprise Schiff base linkages, optionally further reduced to amine linkages.

16. The method according to claim 14 wherein said covalent bonds comprise amide linkages.

17. An implantable bulk material comprising one or more polymer compounds according to either claim 8 or claim 9, covalently bound to a surface or surfaces of said bulk material. WO 98/52620 PCT/EP98/02857 - 39

18. A method of binding collagen-like polymer compounds to an implantable bulk material wherein said polymer compounds are capable of initiating cellular attachment, which polymer compounds comprise one or more peptoid residues, as well as primary amine groups, the method comprising the steps of: (a) providing aldehyde or carboxyl groups on a surface or surfaces of said implantable bulk material; (b) reacting said primary amine groups with aldehyde or carboxyl groups to form covalent bonds.

19. A method of binding collagen-like polymer compounds to an implantable bulk material wherein said polymer compounds are capable of initiating cellular attachment, which polymer compounds comprise the sequence of formula 1, wherein (y) denotes Z 9: -(Gly-Pro-Nieu) (y) - (1), as well as primary amine groups, the method comprising the steps of: a) providing aldehyde or carboxyl groups on a surface or surfaces of said implantable bulk material; (b) reacting said primary amine groups with said aldehyde or carboxyl groups to form covalent bonds.

20. A method of binding collagen-like polymer compounds to an implantable bulk material, wherein said polymer compounds are capable of initiating cellular attachment, which polymer compounds comprise Kemp triacid adducts of polymer compounds comprising the sequence of formula 1, wherein (y) denotes a 9: -(Gly-Pro-Nieu) (y) - (1), the method comprising the steps of (a) providing amine groups on a surface or surfaces of said implantable bulk material; (b) reacting said amine groups with carboxylic acid groups on said Kemp triacid adducts to form covalent bonds.