FIELD OF THE INVENTION

This invention relates to a process for modification and acti¬ vation of a polymer surface in order to immobilize proteins and other types of molecules to the polymer surface. The invention further relates to the polymers in modified form, and the polymers having immobilised on their surfaces molecules such as proteins.

BACKGROUND OF THE INVENTION

Polymer surfaces have wide applications as support for biolo¬ gical macromolecules in medicinal diagnostics, science and industry. Immunosorbent analysis, e. g. IRMA (Immuno Radio Metric Assay) , ELISA (Enzyme Linked Immuno Sorbent Assay) IFMA

(Immuno Fluoro Metric Assay)/FIA (Fluoro Immuno Assay) , and ILMA(Immuno Lu ino Metric Assay)/LIA (Luminescent Immuno Assay) , as well as various forms of chromatography and bio¬ logical reactors are examples of areas, where immobilization of e.g. antibodies, enzymes, DNA, etc. to the surface of organic polymers are utilized. Furthermore, polymer immobilization of low molecular weight reagents is essential for solid support synthesis of biological polymers such as peptides (E. T. Kaiser: "Synthetic Approaches to Biologically Active Peptides and Proteins Including Enzymes" Ace. Chem. Res. .22., 47-54

(1989)) and nucleic acids (M. H. Caruthers "Gene Synthesis Machines" Science 23_0, 281-285 (1985)).

Although non-covalent immobilization by adsorption is suffi- cient for many purposes this is often in- or nonefficient with certain molecules for which covalent attachment of the ligand to the polymer is required. Direct covalent attachment to con¬ ventional organic polymers (polystyrene, polyethylene, poly-

methyl methacrylate, etc.) is not possible, and it is there¬ fore necessary, often at great expense and difficulty, to prepare polymers from modified monomers (Advances in Polymer Synthesis (Polymer Science and Technology, vol. 31) B. M. Culbertson and J. E. McGroth, Plenum Press, New York 1985) , or to activate the conventional surface after polymerization. From a technical and economical point of view post polymeri¬ zation modification is preferable.

This has been recognised in the art for some time, and dif¬ ferent approaches have been proposed to solve this problem .

However, these approaches often involve the use of highly hazardous chemicals or procedures and/or one or several time consuming steps. As an example the preparation of Merrifields Peptide Resin involves the extremely carcinogenic chemical chloromethylmethyl ether. A further example is the introduc¬ tion of amino groups on the surface of polymers described in J. Virol. Methods, 3_, 155-165 (1981) which requires the use of the hazardous chemicals methanesulfonic acid, glacial acetic acid and fuming nitric acid, and lasts for two days.

Also, the solvents used in these methods often create an uncontrollable physical modification of the polymer surface because the polymer itself is soluble in the solvent. This produces a serious problem whenever spectrophotometrical determinations are subsequently used.

Polystyrene premodified during production with reactive groups, such as -OH, -S0 ₃ H, or -NH-, are available on the market. However, with these it is necessary to separately produce the desired articles from the specific modified polystyrene.

Use of electromagnetic radiation has been suggested in EP publication no. 155 252, according to which a polymer surface is activated by radiation grafting of vinyl monomers under highly specific conditions.

This process, however, is very time consuming, even with γ- radiation, which is the only applied radiation source, the reaction time is 10 to 12 hours. Also 7-radiation has tendency to discolour the polymer surface thereby rendering it unsuit- able for spectrophotometric measurements.

Another suggested use of electromagnetic radiation for such purpose is indicated in DE publication no. 34 35 744, accor¬ ding to which protein A is conjugated to a bifunctional spacer molecule containing arylazides, and subsequently binding the conjugate to the polymer surface by exposing the conjugate to light.

A still further suggested application using electromagnetic radiation is described in EP publication no. 319 957, wherein certain polycyclic photoreactive compounds are especially subjected to ultra violet light activating the photoreactive compound which subsequently reacts with groups on the polymer surface. Thereafter a protein may be coupled to the surface via a linker molecule.

SUMMARY OF THE INVENTION

This invention in its first aspect relates to a novel and efficient process by which polymer surfaces upon treatment with a Ce(IV) salt and electromagnetic radiation are chemically modified in such a way that proteins and other ligands can be bound more tightly, possibly even covalently, to the surface, and thereby be immobilized.

In its second aspect the invention relates to polymers in modified form prepared for the establishment of covalent bonds to proteins or other ligands in order to immobilize these ligands to the polymer surface.

» In a third aspect the invention relates to the modified polymers of the invention whereto proteins or other ligands have been immobilized.

In a fourth aspect the invention relates to the use of the method or the modified polymers in analytic techniques such as immuno sorbent analysis (IRMA, ELISA, IFMA/FIA, ILMA/LIA) , or more generally in polymer support technology.

In a fifth aspect the invention relates to a package compri¬ sing the unmodified polymer formed in a desired configuration for a specific desired end use in combination with a set of chemicals suitable for modifying the polymer surface in order to prepare it for immobilization of proteins or other ligands.

BRIEF DESCRIPTION OF THE DRAWINGS

In the drawings Fig. 1 shows the effect of activation with a Ce(IV) salt and electromagnetic radiation, and the effect of adding Berol ^® or ethanolamine on the binding of a protein to a polymer surface.

Fig. 2 shows the relation between the irradiation time and the relative binding of a ligand to a polymer surface.

Fig. 3 shows the relation between the concentration of a Ce(IV) salt and the relative binding.

Fig. 4 shows the binding at a fixed concentration of enzyme conjugated antibody as a function of the log of the concentra¬ tion of competing unlabelled species to be bound added to microtiter wells, and the effect of adding Berol ^® and ethanol¬ amine.

Fig. 5 shows the binding of biotinamine (sometimes abridged as BioNH ₂ ) to a polystyrene surface.

Fig. 6 shows the binding of spermidine to a number of polymer surfaces activated in accordance with the invention as a function of irradiation time.

Fig. 7 shows the binding of ¹⁵ I-labelled protein as a function of activation treatment.

Fig. 8 shows the effect of divinyl sulfone treatment on the binding of a protein to a CAN activated polymer surface.

Fig. 9 shows the saturation of avidin binding to biotinamine coated CAN plates. A constant amount of avidin-peroxidase con¬ jugate and various amounts of avidin was used, x-x-x: Biotin coated, CAN-activated plates,♦-♦-♦: CAN-activated plates 0-0-0: untreated plates.

Fig. 10 shows the influence of the concentration of CAN used in the activation on the binding of avidin peroxidase conjugate on CAN-activated plates (D-O-D) and on biotin-coated CAN-plates (x-x-x) .

Fig. 11 shows the effect of spermine on the binding of biotin¬ amine to CAN activated plates. The plates were (1) CAN-ac¬ tivated, (2) incubated with the indicated concentration of spermine, (3) washed, (4) coated with biotinamine, and (5) assayed with avidin peroxidase conjugate.

Fig. 12 shows the effect of mercaptoethanol on the binding of biotinmercaptane to CAN-activated plates. Experiment as in Fig. 11 except biotinmercaptane was exchanged for biotinamine and 2- mercaptoethanol for spermine.

Fig. 13 shows the effects of a mixture of spermine and mercap¬ toethanol on the binding of biotinamine or biotinmercaptane to CAN activated polystyrene plates. Experiments as in Figures 11 and 12 using 100 μg/ml spermine and 100 mg/ml(10%) 2-mercaptoethanol.

Fig. 14 shows the stability of CAN-activated plates. The plates were activated on day 0 and kept humid in closed plastic bags at 4°C (X) or 20°C ( ). At the times indicated the plates were coated with biotinamine and assayed with avidin-peroxidase conjugate.

Fig. 15 shows an example of local irradiation dependent activation of a polystyrene surface.

Fig. 16 shows the formula of varius biotin derivatives used.

DETAILED DESCRIPTION OF THE INVENTION

This invention in its first aspect relates to a method for modifying polymer surfaces, which method comprises a) application of a solution of a Ce(IV) salt to the surface of the polymer, whereby the surface is covered with the solution, b) irradiation of the solution covered surface with electromagnetic radiation, whereby the surface is activated, c) washing the activated surface once or repeatedly,

In connection with this invention polymers to be used are any suitable polymer, and may preferably be selected from the group comprising polystyrene, polyethylene, polypropylene, polyure- thane, polycarbonate, polyethylene terephthalate glycol, polyvinyl acetate, polyvinyl chloride, polyvinylpyrrolidone, polyacrylonitrile, polymethylmethacrylate, polytetrafluorethy- lene, butyl rubber, styrenebutadiene rubber, natural rubber, poly-4-methylpentylene, and polyesters.

The cerium(IV) salt used can be any Ce(IV) salt, but pre- ferably complex salts with a cation of the Ce(IV)X ₂ ⁶⁺ type, where X is a suitable cation, such as NH ₄ ⁺ , Li ⁺ , Na ⁺ , K ⁺ , or Cs ⁺ , pre¬ ferably NH ₄ ⁺ in combination with the appropriate number of one or more suitable anions, which preferably is nitrate, but any

other type of Ce(IV) salt capable of generating radicals upon being subjected to electromagnetic radiation is suitable. One specific example of a special type of activator salt is "double" salts such as CsNH ₄ [Ce(N0 ₃ ) ₆ ] which are just as useful for this invention as the "pure" salts.

For making up the salt solution any suitable solvent may be used. Suitable examples include water, dimethyl sulfoxide (DMSO) , acetonitrile, and acidic aqueous media. For the purpose of the invention acidic aqueous media are preferred. The acid should be chosen as one capable of stabilizing the radicals formed, and in connection with a nitrate salt a nitrogen containing acid, preferably HN0 ₃ , would be suitable.

The electromagnetic radiation used is typically ultra violet visible light, preferably near visible UV light, but other types of electromagnetic radiation capable of generating radicals from the Ce(IV) salt is suitable. This could be X- rays or 7-radiation.

The irradiation may be applied diffusely to the surface, but for certain uses local and/or directed application is pre¬ ferred, and in some instances even mandatory.

In the washing step various liquids may be used, such as water, or buffers.

It is also foreseen that an additional step (d) may be added to the above process comprising a chemical modification of the activated polymer surface. These further modifications of the activated surface are examplified by: i) hydrolysis reactions leading to alcohols, ii) elimination reactions resulting in alkenes, iii) oxidation reactions leading to epoxides, aldehydes, and carboxylic acids, and iv) addition reactions leading e.g. to halogenides.

The modification reagents include inter alia bases (NaOH) , aacciiddss ((HH ₂₂ SS00 ₄₄ )) ,, ppeerrooxxiiddeess ((HH ₂₂ 00 ₂₂ ,, RRCC00 ₂₂ OOHH)) ,, Cr ₂ 0 ₇ ^2" ' Mn0 ₄ ^" ' halogens (I ₂ , Br ₂ , Cl ₂ ) and pseudohalogens (BrCN)

Through this procedure activated polymer surfaces are obtained whereto proteins or other species can be efficiently immobili¬ zed, possibly through covalent binding.

It is well documented that irradiation of Cerium(IV) diammo- nium hexanitrate, Ce(NH ₄ ) ₂ (N0 ₃ ) ₆ , (CAN) , in acidic medium (nitric acid) gives rise to nitrate radicals (N0 ₃ ^* ) which are highly oxidizing (E. Baciocchi et al. "Cerium(IV) Ammonium Nitrate Catalyzed Photochemical Auto-oxidation of Alkylbenzenes" Tetrahedron Lett. 2J5, 3353-3356 (1985)). Treatment of toluene with CAN and long wavelength ultraviolet radiation (UVA) results in a high yield of benzyl nitrate which is easily hydrolyzed to benzyl alcohol (Baciocchi et al., supra) .

In accordance with the present invention the inventors have applied this type of photochemistry in a more general manner by treating polymer surfaces with Ce(IV) salts, and electro¬ magnetic radiation, and tested the properties of the resulting surfaces in terms of their capacity to bind protein and low molecular weight amines.

Without being in any way bound to any specific theory concer¬ ning the mechanism underlying the results obtained, we believe that the activation of the polymer surface is obtained through the formation of radicals that are stabilized in the medium used for applying the Ce(IV) salt to the polymer surface, which radicals subsequent to their formation through a hydrogen abstraction process create activated groups in the polymer, which groups are susceptible to react with certain groups in a molecule possibly with the formation of a covalent bond.

In a further embodiment the activated group may be coupled to an intermediate molecule, hereinafter designated spacer molecule or spacer, which in accordance with the intended use

may be homofunctional, meaning that the spacer contains only reactive groups of one kind (e.g. -NH ₂ , -OH, or -COOH) or heterofunctional, meaning that the spacer contains reactive groups of more than one kind (e.g. -NH ₂ , -SH and -OH) . The spacer is usually of the so-called bifunctional type, meaning that the spacer contains two reactive groups, whereof one is intended to bind to the polymer surface function, and the other to the species to be immobilized. However, the spacer may also be multifunctional, meaning that one spacer is capable of binding more than one species to the polymer surface.

The spacer is used for several purposes, such as stabilizing the highly reactive activated group in the polymer in order to control and regulate the formation of the covalent bond, or in the case of a bulky ligand the spacer may provide a solution to problems connected with steric hindrance for the ligand, and thereby give the ligand "access" to the polymer surface.

A further modification of the activated groups and/or spacers can be obtained through the use of so-called linker reagents.

Suitable homobifunctional linker reagents for the purpose of this invention may be: divinyl sulfone, o-phenylenedimaleimi- de, dimethyl adipimidate, glutaraldehyde, glutaconaldehyde, carbodiimides, tolylene-2,4-diisocyanate, disuccinimidyl suberate, bis-oxiranes, bis-N-hydroxysuccinimide esters, etc., preferably divinyl sulfone.

Suitable heterobifunctional linker reagents may be maleimido- benzoic acid N-hydroxysuccinimide ester.

The above listings are in no way intended to be limiting for spacers and/or linker reagents suitable for the invention, but are only examples of suitable spacers and linker reagents.

In accordance with the preferred embodiment of the present invention a polystyrene surface, such as the wells in an ELISA microtiter plate, is treated with CAN in aqueous nitric acid

and radiated with UVA. It is believed that this generates nitrate radicals, which abstract hydrogen from the polymer resulting in nitrate ester formation, hydroperoxydation, hydroxylation, or other oxidations, etc.

In a variant of the process using divinyl sulfone as a linker reagent it is believed that one of the vinyl groups establishes chemical bonds to the activated groups on the polymer surface, and the other vinyl group is available for establishing covalent bonds to the desired ligand.

We do not yet know the chemical structure of the modified polystyrene surface but by drawing the analogy to the products formed with toluene, etc. we suggest that inter alia nitrate esters are formed on the surface. These may then react direct¬ ly with the amino groups of protein or after hydrolysis to alcohols be cross-linked to proteins by divinyl sulfone.

As indicated above the properties of the resulting surfaces were tested in terms of their capacity to bind protein both by adsorption and by covalent binding (cf. below) .

Adsorption of proteins to polystyrene is presumably mostly due to hydrophobic and electrostatic interactions which can be subdued by including a detergent in the coating mixture. For testing we chose Berol ^® EMU-043 which is non-ionic and chemi¬ cally inert for this purpose, and as shown in Fig. 1 relating to Example 1 below, this detergent inhibits adsorption of protein (antibody) to non-treated polystyrene. In order to test the non-covalent binding we suppressed the covalent binding by including ethanolamine which contains two of the functions (-NH ₂ , -OH) which are primarily responsible for covalent protein binding. Ethanolamine has no inhibitory effect on protein adsorption to unmodified polystyrene (Example 1, Fig. 1) .

Treatment of the polystyrene surface with CAN and UVA has several effects on its ability to bind protein. Firstly, total binding is decreased by approximately 70%. Secondly, about 60%

of the residual binding is not affected by the presence of Berol ^® , and thirdly, the residual binding is reduced to about 40% by the inclusion of ethanolamine (Example l. Fig. 1) .

We infer from these results that the physical and chemical properties of the polystyrene surface have been significantly altered by the treatment with CAN and UVA. The results fur¬ thermore strongly indicate that more than 60% of the protein is covalently bound to the surface.

Inclusion of a linker or coupling activator, such as inserting a divinyl sulfone coupling step and an extra washing step between steps c) and d) in the procedure resulted in an increase of approximately 50% in the quantity of protein immobilized in the presence of Berol ^® (cf. Fig. 8) , and the quality of the results in terms of a decrease in standard deviation between parallel experiments was improved. Therefore divinyl sulfone treatment was included as an embodiment of the invention in the characterization experiments described below.

Another consequence is that treatment with a linker agent is included in preferred embodiments of the methods of the in¬ vention, meaning that the invention further relates to a method for modifying polymer surfaces, which method comprises a) application of a solution of a Ce(IV) nitrate and a nitrogen containing acid to the surface of the polymer, b) irradiation of the surface with long wavelength ultraviolet light, c) washing the surface once or repeatedly, e) incubation of the polymer surface with a linker agent, and f) washing the surface once or repeatedly.

Through this procedure polymer surfaces are obtained that may be better suited for having proteins or other species immobi¬ lized thereto than the surfaces that are not treated with a linker agent.

For this embodiment it is also foreseen that an additional step (d) may be added to the above process comprising a chemical modification of the activated polymer surface. These further modifications of the activated surface are examplified as above.

The invention in a further aspect relates to methods of immobi¬ lizing proteins or other species to a polymer surface, which methods in addition to steps (a) to (c), (d) or (f) above comprise the following steps: g) application of a protein or other species to the activated, and washed surface, and h) washing the surface.

The species used may be any type of molecule, such as a pro¬ tein (an enzyme, an antibody, an antigen) , a peptide, a nucleic acid (DNA or RNA) , a carbohydrate, a lipid, an amino acid, a nucleoside, an amine, a thiol, an alcohol, or whatever it is desired to immobilize on the polymer surface, including coupling of catalysts, fluorescent compounds, and/or inorganic moieties, possibly through chelating agents.

In the context of this invention the species to be immobilized is also meant to encompass cells, virus, microorganisms, and the like.

The protein or other species which was applied in step (g) is thus immobilized on the polymer surface.

In order to characterize the reaction leading to modification of the polystyrene surface, we have varied the metal salt, and its concentration, the duration and wavelength of irradiation and the protein concentration, and analyzed the ability of the surface to bind protein (enzyme-conjugated antibody) both in the absence and presence of inhibitor (Berol ^® /aminoethanol) .

The effect of varying the concentration of CAN is shown in Fig. 3. It is seen that the degree of modification reaches a plateau at approx. 4 mM CAN concentration. Analogously, the maximum degree of modification is reached within 10 minutes of irradi- ation with the equipment used (Fig. 2) .

The formation of nitrate radicals by irradiation of CAN can be detected by flash photolysis (R. W. Glass and T. W. Martin: "Flash Generation and Decay Kinetics of the Nitrate Radical in Aqueous Nitric Acid Solutions" J. Am. Chem. Soc. 9_2, 5084-5093 (1970)) .

Since it appears as if an important feature of this invention is the formation of radicals, especially nitrate radicals, the modification of the invention can also be effectuated by treatment of the polymer surface with nitrate radicals in the vapour phase.

In order to screen for other transition metal salts, which might be useful for polymer surface modification, flash photolysis experiments were performed with nitrates of prac¬ tically all metals and metaloids except Tc and the radioactive actinides in their most usual oxidation state capable of forming nitrates. In no other case than Ce(IV) could nitrate or other radicals be detected. Accordingly, photochemical modifi¬ cation of the polystyrene surface with salts like Cu(N0 ₃ ) ₂ or Fe(N0 ₃ ) ₃ did not change the protein binding properties of the surface in the ELISA. These results support a radical mecha¬ nism.

In further characterization of the binding of species to polymer surfaces according to the invention, the effect of the concentration of the species to be bound was also investigated, and it was found that protein binding to the modified poly- εtyrene surface was saturated at a protein concentration of approx. 3 μg/ml which is similar to that measured for the unmodified surface.

The activation of the polystyrene surface with CAN and UVA also resulted in a dramatic increase (>40 fold) in the ability of the surface to bind amines and mercaptanes. This is illustrated by the binding of biotinamine (6-biotinylaminohexane-l-amine) and biotinmercaptane (6-biotinylaminohexane-l-thiole) to microtiter wells as detected by the subsequent binding of avidin-peroxidase conjugate (Fig. 2, 5, 9, and 12), and by the binding of spermidine to polystyrene, polyethylene, polypropy¬ lene, or polymethylmethacrylate tubes (Fig. 6) .

As a special case, it is feasible to chemically/photochemically equip macromolecules with an -SH or -NH ₂ linker whereby they can be immobilized to the activated surface as illustrated by the binding of thiolated DNA to polystyrene tubes (Table 1) .

Table 1

Binding of Thiolated DNA to Polystyrene Tubes

Treatment

none

CAN

It is particularly noteworthy that the amount of avidin-peroxi¬ dase bound to CAN activated polystyrene via biotinamine (Fig. 5, column 5) equals that bound to untreated polystyrene via adsorption (in the absence of detergent; Fig. 5, column 1) . Fig. 8 shows that inclusion of the linker, divinylsulfone (DVS) , increases the Berol ^® resistant (covalent) protein binding to CAN activated polystyrene by approximately a factor of two.

Just as is the case with binding of IgG to CAN-activated microtitet wells, the binding of avidin to biotin coated CAN- activated plates is saturated at elevated concentrations of avidin (Fig. 9) , and this saturation occurs at a concentration of -1 μg/ml, i.e. similar to that observed with IgG.

The optimum CAN concentration for activation for biotinamine coating is also similar (~2 mM) (Fig. 10) to that determined for IgG coating indicating that the same underlying photochemistry is responsible for the activation in both cases. The "chemical specificity" of the CAN activated polystyrene surface was studied using the biotin derivatives (1) biotin-N-hydroxy succinimide (biotin active ester) , (2) biotinamine, (3)t-Boc-biotin amine, (4) biotinhexole, (5) biotinheptane, (6) biotinmercaptane, and (7) biotindisulfide shown in Fig. 16, where X is biotinyl, and the quenchers spermine and mercaptoethanol. Some of the results from these studies -are presented in Table 2 below.

TABLE 2 Specificity of CAN activated plates

Biotin derivative % Binding

(2) 100

(3) 37

(4) 9 (5) 6

(6) 83

(7) 57

As can be seen from Table 2 only biotin derivatives containing -NH ₂ or -SH groups bind to the activated surface whereas derivatives containing -OH or protected -NH ₂ groups do not. Accordingly, the binding of biotinamine is inhibited by spermine (Fig. 11) and the binding of biotinmercaptane by mercaptoethanol (Fig. 12) . It is noteworthy that the binding of biotinamine is more strongly inhibited by spermine than by mercaptoethanol and similarly that the binding of biotinmercap¬ tane is more strongly inhibited by mercaptoethanol than by spermine (Fig. 13) . These results show that apparently at least two types of modifications occur at the surface of which one seemingly reacts somewhat better with amines and the other better with mercaptanes.

In Table 3 below the reproducibility of the binding of avidin- peroxidase conjugate to polystyrene plates is shown.

Table 3 Reproducibility of Binding to Polystyrene Plates PS/AvPo - Tween ^® PS/AvPo + Tween ^® CAN/AvPo + Tween ^® CAN/BioNH ₂ /AvPo + Tween ^®

PS: Polystyrene; AvPo: Avidinperoxidase conjugate; ±Tween ^® : protein binding in the presence/absence of detergent Tween ^® 20. CAN: CAN activated plates. CAN/BioNH ₂ : CAN activated, biotin- coated plates.

From Table 3 it is seen that the reproducibility in terms of interwell/interplate variation is equal to or better than untreated polystyrene.

The CAN activated plates are stable for at least 30 days when kept humid at 4°C or room temperature as assayed by their ability to bind biotinamine (Figure 14) ,

Finally the ability to activate surfaces locally by irradiation control was demonstrated. Irradiation of a lid from a micro- titer plate was performed through a mask spelling "NOVO" . The activated lid was coated with biotinamine and "developed" by X- ray autoradiography after coating with ¹²⁵ I-streptavidin (40 μCi/μg, Amersham) . Fig. 15 clearly demonstrates the present inventions ability to perform such activation.

Example l

Binding of IgG Horse Radish Peroxidase Complex to ELISA Plates

Activation of polystyrene ELISA plates

To each well in a polystyrene ELISA plate (NUNC, Denmark no. 439454) was added 100 μl 3.7 mM CAN in 2 M HN0 ₃ (Merck, pro analysi) . The plate was irradiated at 20°C through the bottom for 9 minutes with light from five fluorescent light tubes

(Phillips TLD 15 Watt/05, λ « 365 nm) at a distance of 2 cm.

The wells were subsequently washed with water and incubated with 100 μl 0.42 M divinyl sulfone (Aldrich) in 0.5 M Na ₂ C0 ₃

(Merck, pro analysi) , pH 11, for 30 minutes. The wells were washed three times with 400 μl water after which they were used for coating with protein.

Protein coating of plates

To each well of activated ELISA plates was added 100 μl 0.15 M PBS (phosphate buffered saline, 153 mM Na ⁺ ; 4.2 mM K ⁺ ; 9.6 mM phosphate; pH 7.4) containing the desired amount of protein. Following incubation (usually 2 hours at 20°C) the wells were washed three times with PBS with 0.05% Tween ^® , and the amount of bound protein was measured.

Quantification of bound protein

The IgG horse radish peroxidase (HRP) conjugated protein was quantitated by measuring the peroxidase activity in 100 mM sodium citrate buffer (Merck, reinst) pH 5.2, containing 0.03% H ₂ 0 ₂ , and 0.5 mg/ml 1,2-phenylenediamine, dihydrochloride (OPD) tablets (Dakopatts) as chromogenic substrate. 100 μl mixture was added to each well and the plate was incubated at 20°C for 10 minutes. The enzymatic reaction was terminated by addition of 100 μl 2 N H ₂ S0 ₄ and the colour reaction was quantified by measuring E ₄₉₂ using an automated ELISA scanner (EAR 400, SLT- LAB INSTRUMENTS, Austria).

The results from these determinations are shown in Fig. 1.

As indicated earlier in the specification this Example shows that the method of this invention reduces the adsorption capacity of a polystyrene surface by approximately 70%, whereas the capacity to bind covalently to the protein is improved to approximately 10% of the maximum adsorption capacity.

Figs. 2, 3, and 4 were constructed from data obtained in experiments parallel to this example by varying different parameters of the method of the invention.

Fig. 3 indicates that the optimum transition group metal salt concentration (CAN) is about 4 mM, and Fig. 2 indicates that an optimum radiation time for that salt with the present equipment would be from 8 to 10 minutes.

Example 2

Binding of ¹²⁵ iodinated IgG to ELISA plates

Activation and coating of the plates was performed in a manner similar to that described in Example 1.

²⁵ iodination of IgG

Iodination was performed with Na ¹²⁵ I (carrier free, Amersham) and iodo-beads (Pierce) according to the manufacturers (Pierce) recommendation. The specific activity of the labeled IgG was « 0.5 μCi/μg.

Quantification of bound protein

¹²⁵ I-labelled protein was quantified by autoradiography of the microtiter plates using an Agfa Curix RP1 X-ray film and a 7 mm steel mask between the plate and the film. The autoradiogram was scanned at 550 nm with a Shimadzu CS930 densitometer scanner, and the results are shown in Fig. 7.

Example 3

Binding of Biotinamine (6-biotinylaminohexane-l-amine) to ELISA Plates.

Activation of the plates was performed as in Example l, optionally omitting the divinyl sulfone step. Incubation with biotinamine (10 μg/ml in PBS) was done for 30 minutes at 20°C.

The wells were subsequently washed three times with PBS, 0.05% TWEEN ^® 20, incubated for 120 minutes with avidinperoxidase conjugate (364 Dakopatts, diluted 1:4000 in PBS, 0.05% TWEEN ^® 20) , washed, and quantified as in Example 1. The results are shown in Figs. 2 and 5. Fig. 2 shows the effect of irradiation with (φ) or without (Q) biotinamine treatment.

Example 4

Binding of Biotinmercaptane (6-biotinylaminohexane-l-thiol) to ELISA plates This was performed analogously to Example 3 except biotinmercaptane was exchanged for biotinamine.

Example 5

Binding of Spermidine to Polystyrene, Polyethylene, Polypropy¬ lene, and Polymethylene Tubes.

Activation of Tubes

One ml of 3.7 mM CAN in 2 M HN0 ₃ was filled into each tube. The tubes were then irradiated from the bottom for the desired length of time using the light source described in Example 1.

The tubes were washed with water and incubated for 16 hours at 20"C with ³ H-spermidine (1 μg/ml, 10 ⁵ cpm/μg, New England Nuclear) dissolved in 1 ml NaHC0 ₃ , pH 9. The tubes were washed three times with water and the amount of radioactivity deter¬ mined by scintillation counting in Insta-Gel ^® (Lu ac) . Fig. 6

shows the results in terms of ³ H-spermidine binding as a function of the time of irradiation.

Example 6

Binding of Thiolated DNA to Polystyrene tubes

A mixture of a ³² P-end labeled DNA fragment (90 bp EcoRI-PvuII fragment of plasmid pUC19 (Nielsen et aJL. Biochemi- stry 27 _. , (1988) 6338-6343) 100 μg calf thymus DNA and 10 μg psoralen-disulfide (Eisner et al Analytical Biochem. 149, (1985) 578-581) in 1 ml 10 mM Tris-HCl, 1 mM EDTA buffer was irradiated for 60 min with light a source used for CAN ac¬ tivation. 10 μl 0.1 m dithiothreitol was added and the DNA was precipitated with 2 ml ethanol, 0.2 M sodium acetate. The precipitate was isolated by centrifugation washed with ethanol dried and redissolved in 1.2 ml H ₂ 0. Polystyrene tubes were activated as described in Example 5 and incubated with 200 μl 10 mM Tris-HCl, pH 7.4, 1 mM EDTA buffer containing 10 μg of the above thiolated DNA for 16 hrs. at 4°C. The tubes were washed 3 times and the amount of bound DNA measured by "Cere- ncov counting of ³² P in a scintillation counter.

Conclusion

We have shown that treatment of a polymer surface, such as a polystyrene surface, with a Ce(IV) salt, such as a nitric acidic solution of cerium ammonium nitrate, and electromagnetic radiation, such as long wavelength ultraviolet radiation, modified the surface in a way that it reduces its capacity to bind proteins by adsorption by 70% and at the same time allows apparently covalent attachment of protein with an efficiency corresponding to approx. 10% of the non-covalent adsorption observed with a non-treated polymer surface.

It was furthermore found that the activation method of the invention drastically increases (>40 fold) the capacity of the

treated surface to bind low molecular weight amines and mercaptanes.

This type of surface modification could be useful for a variety of techniques based on solid support "chemistry" such as immunosorbent assays, affinity column chromatography, enzyme immobilization (e.g. in fermenting) , therapeutic instruments (e.g. dialysis) , etc.