The present invention particularly relates to an application of computer-aided rational design techniques for the rapid development and optimization of polymers that act as synthetic receptors. Specifically the imprinted polymers are synthesized by a polymerization of individual monomers with or without cross-linker in the presence of template molecule (s), which can be small molecules such as drugs, pesticides, peptides as well as large molecules such as carbohydrates, nucleic acids and proteins.

Background Art Molecularly imprinted polymers (MIPs), materials with artificially created receptor-like recognition properties, have recently attracted significant attention as potential substitutes for unstable receptors and antibodies in affinity chromatography, membranes, capillary electrophoresis and sensor technology (US

patents 5110833,5587273,5756717,5728296,5786428 and 5849215). Among the factors limiting their practical application is the absence of a general procedure for polymer synthesis. Several attempts have been made to develop a general procedure for rational design of the imprinted polymers (Nicholls I. A. (1995): Thermodynamic consideration for the design of and ligand recognition by molecularly imprinted polymers, Chem. Lett., 1035-1036; Whitcombe M. J., Martin L., Vulfson E. N.

(1998): Predicting the selectivity of imprinted polymers.

Chromatography, 47,457-464; Takeuchi T., Fukuma D., Matsui J. (1999): Combinatorial molecular imprinting: an approach to synthetic polymer receptors. Anal. Chem., 71, 285-290). In the best cases they give rules or hints, indicating how the MIP should be made in order to possess a certain level of specificity. Thus it-is recommended that the polymerization should be performed in a hydrophobic solvent in order to produce a material able to interact with the template through electrostatic interactions. At the same time the choice of monomers, solvent and polymerization conditions in most cases depends on general principles, personal experience or information about similar systems. In some extreme cases it has been necessary to produce and investigate hundreds of polymers in order to optimize MIP monomer composition

(Takeuchi et al. op. cit).

Disclosure of Invention The present invention describes a new method for rational choice of the functional monomers for the preparation of MIPs.

The present invention describes a computer-aided rational design techniques for the rapid development and optimization of molecularly imprinted polymers, which includes screening of a virtual library of functional monomers for their interaction with the template molecule and selection of those monomers forming a strong complex with the template for polymer preparation. The procedure of monomer selection includes several stages (it is important to note that each individual step as well as their combination can be used separately for the design of MIPs and are covered by the present invention). First, a virtual library of molecular models of functional monomers is produced containing molecules that possess polymerizable residues and residues able to interact with a template, e. g. through electrostatic, hydrophobic van- der-Waals forces, dipole-dipole interactions and/or reversible covalent bonds. Secondly, a molecular model of the template molecule is prepared. Charges for each atom are calculated, and the structure of the template and

monomers refined using molecular mechanical methods.

Thirdly, each of the entries in the virtual library is probed for their possible interaction with the template molecule. The monomers giving the highest binding score represent the best candidates for polymer preparation.

Fourthly, copies of the most ideal hits are placed around the target. Simulated annealing is then used to simulate pre-arrangement of the functional monomers with template in the monomer mixture prior to polymerization. At the end of the program, the number and the position of the functional monomers are examined The type and quantity of the monomers participating in the complex with template indicate the type and ratio of the template and monomers in an optimized MIP composition. Finally a mixture of monomers corresponding to this composition is polymerised in the presence of the template to produce a MIP.

Main embodiments include: 1. A computer-aided rational design technique for the rapid development and optimization of molecularly imprinted polymers (MIPs), which includes design of a virtual library of functional monomers and screening of their interaction with a template molecule. Functional monomers should possess polymerizable residues and residues able to interact with the template, e. g. through electrostatic, hydrophobic van-der-Waals

forces, dipole-dipole interactions and/or reversible covalent bonds. The library of functional monomers may include one or more of vinyl monomers, allyl monomers, acetylenes, acrylates, methacrylates, amino acids, nucleosides, nucleotides, carbohydrates, phenols, heterocycles, aniline, and their derivatives.

2. Design of a molecular model of the template using molecular mechanical methods. The template molecule may be selected from biological receptors, nucleic acids, hormones, heparin, antibiotics, vitamins, drugs, cell components and components of viruses such as carbohydrates, saccharides, nucleoproteins, mucoproteins, lipoproteins, peptides and proteins, glycoproteins, glucosaminoglycanes and steroids.

3. Screening of the virtual library of the functional monomers for their ability to form molecular complex with the template. The monomers giving the highest binding score represent the best candidates for polymer preparation.

4. Placing of the functional monomers (or monomers giving the highest binding score) around the template and using molecular mechanics in order to simulate pre- arrangement of the functional monomers with template in the monomer mixture prior to polymerization. At the end of the program, the number and the position of the

functional monomers are examined The type and quantity of the monomers participating in the complex with template indicate the type and ratio of the template and monomers in an optimized MIP composition.

5. The parameters of the modeling and screening (dielectric constants, temperature chosen for "annealing"procedure, type of interactions) could be modified in order to simulate the real polymerization or re-binding conditions.

BRIEF DESCRIPTION OF THE DRAWINGS: Figure 1. Depicts a virtual library of the functional monomers used for the development of MIP, specific for atrazine.

Figure 2. Shows the functional monomers from the virtual library giving the highest binding score when interacting with atrazine.

Figure 3. Depicts interactions between functional monomers and atrazine in an optimized monomer-template complex.

Figure 4. Recovery of the atrazine by polymers, based on itaconic acid, acrylamide and AMPSA. 100 ml of herbicide solution in water (10-9 M) were filtered through 100 mg of polymer and eluted with 90% methanol containing 10 mM HC1.

Figure 5. Depicts interactions between functional monomers and methylated creatinine in an optimized monomer-template complex.

Figure 6. Interaction between phthalic dialdehyde, mercaptan group (OPA reagents) and primary amine (I); thioacetal formation (II); formation of fluorescent complex between thioacetal and primary amine (III).

Figure 7. Depicts imprinting factors of the polymers MIP A, MIP B and MIP C calculated for creatinine and creatine. Imprinted factors were calculated using equation I = AF (MIP)/ AF (Blank) where AF is change in fluorescence for MIP and blank polymers, correspondingly.

Measurements were made in 100 mM sodium phosphate buffer, pH 7.0 (Ex = 320 nm; Em = 410 nm).

MODES FOR CARRYING OUT THE INVENTION The first embodiment describes the design of a virtual library of functional monomers which possess polymerizable residues and residues, able to interact with a template through electrostatic, hydrophobic van- der-Waals forces, dipole-dipole interactions or reversible covalent bonds. The library of functional monomers should contain at least two, and preferably more monomers which can be vinyl monomers, allyl monomers, acetylenes, acrylates, methacrylates, amino acids, nucleosides, nucleotides, sugars and saccharides, carbohydrates, phenols, heterocycles, aniline, and their derivatives. Preferable monomers are these able to interact with the template through non-covalent interactions and be polymerized through a radical mechanism.

The second embodiment describes the design of a molecular model of the template using molecular mechanical methods. The template is selected from a group including biological receptors, nucleic acids,

immunosuppressants, hormones, heparin, antibiotics, vitamins, drugs, cell components and components of viruses such as carbohydrates, lipids, saccharides, nucleoproteins, mucoproteins, lipoproteins, peptides and proteins, glycoproteins, glucosaminoglycanes and steroids.

The third embodiment describes screening the virtual library of functional monomers for their ability to form a molecular complex with the template. The monomers giving the highest binding score represent the best candidates for polymer preparation and can be used directly for the polymer synthesis as part of a monomer mixture.

The fourth embodiment describes a refining step for optimizing the monomer composition. It includes placing of one or more functional monomers, preferably the monomers giving the highest binding score around the template, and using molecular mechanics to simulate pre- arrangement of the functional monomers with template in the monomer mixture prior to polymerization. At the end of the program, the number and the position of the functional monomers are examined. The refining step has three goals: first, to evaluate the quantity of the monomer units which should be used in complexation with template; second, to determine if cocktails of monomers form a stronger complex with the template than individual monomers; third, to check possible interactions between the functional monomers in a monomer mixture, both positive (stabilizing complex) and negative (competing). The type and quantity of the monomers participating in the complex with template (first shell layer), or their combination with the monomers which interact with the first shell layer of monomers,

stabilizing complex (second shell layer) indicate the type and ratio of the template and monomers in an optimized MIP composition. All components of the refining process can be used and accounted for individually or in combination in the optimization of the monomer composition.

The sixth embodiment describes"fitting"of the modeling and screening parameters (dielectric constants, temperature chosen for"annealing"procedure, type of interactions) to real polymerization or re-binding conditions. If, for example, the template is not soluble in organic solvents then the dielectric constant could be changed from vacuum to water or the constant of the solvent used for polymerization. The temperature in the "annealing"procedure can be also adjusted to the one applied during polymerization or the re-binding step.

The present invention will now be further described in detail by reference to the following, examples, which are intended to illustrate some of the possibilities, but are in no way intended to limit the scope of the invention.

Example 1. Computer aided design of an MIP specific for atrazine. a) The workstation used to generate molecular models and perform simulation was a Silicon Graphics Octane running the IRIX 6.4 operating system. The workstation was configured with two 195 MHz reduced instruction set processors, 712 MB memory and a 12 Gb fixed drive. The system was used to execute the software packages SYBYL (Tripos). The virtual

library contained the most frequently used functional monomers able to form non-covalent interactions with the template atrazine. These are methacrylic acid, acrylic acid, itaconic acid, acrolein, acrylamido-(2-methyl)-propane sulfonic acid (AMPSA), 2-vinylpyridine, 4-vinylpyridine, 1- vinylimidazole allylamine, N, N'-diethyl aminoethyl methacrylamide (DEAEM), acrylonitrile, 2- hydroxyethyl methacrylate, p-divinylbenzene, m- divinylbenzene, urocanic acid, urocanic ethyl ester, ethylene glycol dimethacrylate, N, N'-methylene-bis- acrylamide, styrene, acrylamide and, where appropriate, their protonated or deprotonated forms (see Figure 1).

The LEAPFROG algorithm was used to analyze binding between monomers and template. 100,000 iterations were completed in approximately 8 hours, results from each run were examined and the empirical binding score was rank to the binding interactions between the functional monomers and atrazine. The best five are shown in Figure 2. Their calculated binding scores were: Itaconic Acid-37.05 kcal. mol-1 Acrylamide-35.78 kcal. mol-1 2-Acryl-2-methyl-1-propanesulphonic

acid-32.64 kcal. mol-1 N', N-Methylbisacrylate-31.63 kcal. mol-1 Acrylic acid-29.61 kcal. mol-1 b) The top three monomers (itaconic acid, acrylamide and 2-acryl-2-methyl-1-propanesulfonic acid) were next"packed"around the template using a solvation experiment. Simulated annealing was used to optimize the organization of the monomers around the template molecule. Annealing conditions were fixed as 1000K- 10K. Equilibrium length was determined as 10000 ps.

Energy minimization was performed to a 0.01 kcal mol- convergence for each iteration. At the end of the program, the number and the position of the functional monomers were examined (Figure 3). The type and quantity of the monomers participating in the complex with template indicate'the type and ratio of the template and monomers in an optimized MIP composition. Accordingly to simulation atrazine forms in the model composition complex with 2 molecules of itaconic acid (first shell layer). The rest of the monomers are displaced from the complex as result of a competition process. At the same time, attention can also be paid to three molecules of acrylamide, which form a second shell layer interacting with itaconic acid from the first shell

layer. The optimization of the MIP composition gives a template-monomers ratio 1 (atrazine): 2 (itaconic acid): 3 (acrylamide).

The calculated energy of the complex is based primarily on two figures:-29.0 kCal Mol-1 for steric factors and-35.1 kCal Mol-1 for electrostatic interactions. c) The synthesis of polymers was made with solutions of corresponding monomers, as identified by computer simulation, using dimethylformamide as solvent and 80% cross-linker (ethyleneglycol dimethacrylate).

Compositions are shown in Table 1.

Table 1. Composition of the computational polymers specific for atrazine. Polymer Atrazine Functional EGDM1 DMF2 Initiators monomer AMPSA-215 mg 1044 mg AMPSt 9 ml 9 ml 50 mg MIP AMPSA--1044 mg AMPSA 4 ml 4 ml 50 mg Blank AA-MIP 215 mg 284.32 mg AA 2 ml 2 ml 50 mg AA-Blank-284. 32 mg AA 2 ml 2 ml 50 mg IA-MIP 215 mg 520 mg IA 3 ml 3 ml 50 mg lA-Blank-520 mg IA3 ml3 ml50 mg

1EGDM-ethylene glycol dimethacrylate 2DMF-dimethylformamide 31, 1'-Azobis (cyclohexanecarbonitrile) The polymerisation was initiated by adding 1% azobis (cyclohexane carbonitrile) and heating of the monomer mixture during 12 hours. The resulting polymer was ground, sieved and sedimented in acetone, giving a suspension with an average particle size of 45-106 ; J. m. 100 ml of herbicide solution in water (10-9 M) were filtered through 100 mg of polymer and eluted with 90% methanol containing 10 mM HC1. The herbicide concentration was measured using test-system based on the thylakoid membranes and Hill reaction (Piletskaya E. V. et al. (1999). Anal. Chem. Acta, 391,1-7). A good correlation was found between the polymer affinity and the monomer binding score (Figure 5).

It was found that the MIPs designed using the computational approach had superior binding and recognition properties in comparison with a MIP

designed using the conventional approach (based on methacrylic acid as a functional monomer).

Additional polymer was prepared as described before using a template-monomers ratio 1 (atrazine): 2 (itaconic acid): 3 (acrylamide). This polymer has shown an atrazine binding similar to IA-based MIP with 100% efficiency of atrazine recovering (data not shown).

Example 2. Computer aided design of the MIP specific for creatinine. a) The virtual library used for the design of an MIP specific for creatinine contained 20 monomers as used in Example 1 and one additional monomer- polymerizable thioacetal formed by allyl mercaptan and o-phthalic dialdehyde, which was necessary for creatinine sensing. Creatinine analogue-N, N'- dimethylcreatinine was used as a template instead of creatinine in order to prevent covalent interactions between the template and the functional monomers. The computer simulation was performed in the same way as described in Example 1.

The optimized composition contains template-monomers ratio 1 (N, N'-dimethylcreatinine) : 3 (thioacetal) : 1 (urocanic ethyl ester) (Figure 5).

b) Three polymers were synthesized using a conventional selection of monomers (with cross-linker only and with methacrylic as functional monomer) and computational selection. Three additional blank polymers were prepared in the absence of template.

The composition of the monomer mixtures is shown in Table 2.

O-phthalic dialdehyde (563 mg) and allyl mercaptan (330 mg) were dissolved in 2 ml of DMSO and mixed with methylated creatinine (131 mg). Functional monomers, methacrylic acid (258 mg) (MIP B) or urocanic acid ethyl ester (138 mg) (MIP C) and azobis (cyclohexane carbonitrile) (ACC) (50 mg) were added to the monomers, the solution was purged with nitrogen and left for 1 hour at room temperature. The polymerization was initiated by heating overnight at 80 °C. The resulting polymer was ground, sieved and sedimented in acetone, giving a suspension with an average particle size of 5 J. m. Three additional blank polymers were prepared in the absence of template.

Table 2 Composition of polymers designed using conventional and computational approaches

Polymer AM OPA MA Urocanic Creatinine acid ethyl ester MIP A 330 ul 563 mg--113 mg (pmol) (mol) (mol) Blank A 330 ul 563 mg--- (mol)(mol) MIP B 330 ul 563 mg 258 ul-113 mg (Rmol) (mol) J. mol) (pmol) Blank B 330 ul 563 mg 258 ul-- (pmol)(pmol) (pmol) MIP C 330 ul 563 mg 138. 13 113 mg (pmol) (pmol) mg (mol) (pmol) Blank C 330 ul 563 mg 138. 13 (pmol)(pmolj mg (pmol) AM-allyl mercaptan OPA-o-phtalic dialdehyde; MA-methacrylic acid; All the monomer mixtures contained 2 ml of EDMA- ethylene glycol dimethacrylate and 2 ml of DMSO- dimethyl sulfoxide with 50 mg of AIBN-azobis (isobutyronitrile). c) Polymer recognition properties were analyzed by

measuring flucrescence changes generated by a polymer suspension following the reaction between primary amines and polymer thioacetal leading to formation of an isoindole complex (Figure 6). The imprinting factor, calculated for MIPs, clearly indicates the superior selectivity of the polymer designed using the computational approach in comparison with polymers designed using a traditional approach (Figure 7).