Biosensors and Bioelectronics 22 (2006) 153–163 Towards the design of highly selective recognition sites into molecular imprinting polymers: A computational approach Susanna Monti a , Chiara Cappelli b , Simona Bronco b , Paolo Giusti c , Gianluca Ciardelli d, a Istituto per i Processi Chimico-Fisici (IPCF-CNR), Area della Ricerca, via G. Moruzzi 1, I56124 Pisa, Italy b PolyLab-CNR Pisa c/o Dipartimento di Chimica e Chimica Industriale, Universit` a di Pisa, via Risorgimento 35, I-56126 Pisa, Italy c Dipartimento di Ingegneria Chimica, Chimica Industriale, Scienza dei Materiali, Universit` a di Pisa, via Diotisalvi, 2 I-56126 Pisa, Italy d Dipartimento di Meccanica, Politecnico di Torino, corso Duca degli Abruzzi 24, I-10129 Torino, Italy Received 27 October 2005; received in revised form 8 May 2006; accepted 9 May 2006 Available online 23 June 2006 Abstract A computational approach to simulate the formation of possible imprinted polymers in acetonitrile solution for theophylline (THO) is proposed, using combined molecular dynamics (MD), molecular mechanics (MM), docking and site mapping computational techniques. Methacrylic acid (MAA) and methylmethacrylate (MMA) monomers are used to simulate possible homo and copolymer structures. The model is able predict binding affinity and selectivity when considering THO analogues, such as caffeine, theobromine, xanthine and 3-methylxanthine. Comparison with available experimental data is proposed. © 2006 Elsevier B.V. All rights reserved. Keywords: Molecular imprinting; Molecular dynamics; Molecular mechanics 1. Introduction Comprehension at the molecular level of the mechanisms and the nature of the complex phenomena involved in the formation and recognition capability of molecularly selective, robustly sta- ble, imprinted polymers is a hard goal to achieve. The principle of molecular imprinting (Whitcombe and Vulfson, 2001; Mosbach and Ramstrom, 1996; Takeuchi and Matsui, 1996) is based on cross-linking polymerization in pres- ence of functional monomers around template molecules which are usually low molecular weight compounds such as sug- ars (Byrne et al., 2002; Wizeman and Kofinas, 2001), steroids (Cheong et al., 1998), amino acid derivatives, drugs (Levi et al., 1997) but also polypeptides (Rachkov and Minoura, 2001), proteins (for a review, see Ciardelli et al., 2005) and DNA double-stranded structures (Slinchenko et al., 2004). After poly- merization the template molecule is removed and a cavity of specific shape and functionalities complementary to the target analyte remains in the polymer. These synthetic receptors have a Corresponding author. Tel.: +39 011 5646920; fax: +39 011 5646999. E-mail address: gianluca.ciardelli@polito.it (G. Ciardelli). great variety of applications, in fact they can be used for selective solid phase extraction, as catalysts, as specific detection layers, as artificial antibodies. The research in the area of molecularly imprinted poly- mers (MIP) offers an example of interdisciplinary investigation, which involves both theoretical and experimental approaches in chemistry, physics, engineering and biology. It has been demon- strated that the success of the imprinting procedure crucially depends on the creation of strong monomer/template complexes and on the stability of the molecular assemblies during the entire polymerization reactions. Significant effort has been devoted to developing rational experimental approaches to design MIPs for the perspective application (Steinke et al., 1995) and to screen their selectivity (Allender et al., 2000). Lanza and Sellergren (1999) tested six functional monomers to rapidly screen the parameters of importance to reach a desired level of binding affinity and selectivity. It was demonstrated that a key role on the recognition prop- erties of MIPs was played by the functional component, and the polymer showing the highest selectivity was prepared using methacrylic acid as functional monomer; this holds also with respect to molecules of medical and biological interest. A rea- sonable alternative approach to predict the performances of 0956-5663/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.bios.2006.05.017