Novel Virtual Screening Protocol Based on the Combined Use of Molecular Modeling and Electron-Ion Interaction Potential Techniques To Design HIV-1 Integrase Inhibitors Cristina Tintori, § Fabrizio Manetti, § Nevena Veljkovic, # Vladimir Perovic, # Jo Vercammen, ‡ Sean Hayes, | Silvio Massa, § Myriam Witvrouw, | Zeger Debyser, | Veljko Veljkovic, # and Maurizio Botta* ,§ Dipartimento Farmaco Chimico Tecnologico, Universita ` degli Studi di Siena, Via Alcide de Gasperi 2, I-53100 Siena, Italy, Center for Multidisciplinary Research and Engineering, Institute of Nuclear Sciences VINCA, P.O. Box 522, 11001 Belgrade, Serbia, Molecular Medicine, Katholieke Universiteit Leuven and IRC KULAK, Kapucijnenvoer 33, B-3000 Leuven, Belgium, and Laboratory of Biomolecular Dynamics, Katholieke Universiteit Leuven, Celestijnenlaan 200G, B-3001 Leuven, Belgium Received February 26, 2007 HIV-1 integrase (IN) is an essential enzyme for viral replication and represents an intriguing target for the development of new drugs. Although a large number of compounds have been reported to inhibit IN in biochemical assays, no drug active against this enzyme has been approved by the FDA so far. In this study, we report, for the first time, the use of the electron-ion interaction potential (EIIP) technique in combination with molecular modeling approaches for the identification of new IN inhibitors. An innovative virtual screening approach, based on the determination of both short- and long-range interactions between interacting molecules, was employed with the aim of identifying molecules able to inhibit the binding of IN to viral DNA. Moreover, results from a database screening on the commercial Asinex Gold Collection led to the selection of several compounds. One of them showed a significant inhibitory potency toward IN in the overall integration assay. Biological investigations also showed, in agreement with modeling studies, that these compounds prevent recognition of DNA by IN in a fluorescence fluctuation assay, probably by interacting with the DNA binding domain of IN. INTRODUCTION HIV-1 integrase (IN) is an essential enzyme for viral replication and represents an intriguing target for the development of new drugs. 1-3 IN catalyzes the integration of the viral DNA, previously transcribed by reverse tran- scriptase (RT), into the chromosomes of the host cell. The integration occurs in two temporally and spatially separated reactions, known as 3′-processing and strand transfer. The 3′-processing occurs in the cytoplasm, where IN binds the viral DNA and then removes a dinucleotide from each strand at the 3′-end adjacent to a conserved CA sequence. The protein-DNA complex is then transported into the nucleus where the strand transfer reaction takes place, and the 3′- ends of the viral DNA are covalently linked to the 5′-ends of the host DNA. 4 Although many compounds have been reported to inhibit HIV-1 IN in biochemical assays, 5-8 no drug active against this enzyme has been approved by the FDA so far. Inhibitors active toward IN can be divided into two groups: compounds able to inhibit the 3′-processing reaction, by interfering with the interaction of IN with viral DNA, and compounds that preferentially inhibit the strand transfer reaction. The well- known diketo acids (DKAs) 9-11 and their recent deriva- tives 12-13 selectively inhibit the strand transfer reaction. Therefore, DKAs are referred to as IN strand transfer inhibitors (INSTIs). They have good ex vivo activities against HIV-1 replication, and their mechanism of action, thought not yet entirely explained, is likely to be a consequence of the interaction between the acid moiety and metal ion(s) within the IN active site, resulting in a functional sequestra- tion of the critical metal cofactor(s). 14 Moreover, it was reported that they bind IN after the protein has formed a complex with viral DNA. 15 Unlike DKAs, styrylquinolines (SQs) 16,17 and pyranodipyrimidines (PDP) 18,19 are two classes of compounds characterized by the ability to inhibit the DNA binding of IN at low concentrations, and they are referred to as IN binding inhibitors (INBIs). Within the process of drug discovery, computational methodologies are a well-established and essential tool. In fact, new technologies, such as combinatorial chemistry and high throughput screening (HTS), which allow the synthesis of millions or possibly billions of compounds, require chemists to confront an unimaginably large and diverse “chemical landscapes”. Attempts must therefore be made to introduce a variety of computational techniques that allow chemists to reduce the huge molecular libraries to a more manageable size. Virtual screening (VS, also referred to as in silico screening) makes use of computational models to evaluate a specific biological activity of compounds in order to filter either existing databases or virtual libraries leading to the identification of molecules with activity against the target of interest. There are two basic approaches to VS, which depend on the available information about the 3D * Corresponding author fax: (+39) 0577 234333; e-mail: botta@unisi.it. § Universita ` degli Studi di Siena. # Institute of Nuclear Sciences VINCA. | Molecular Medicine, Katholieke Universiteit Leuven and IRC KULAK. ‡ Laboratory of Biomolecular Dynamics, Katholieke Universiteit Leuven. 1536 J. Chem. Inf. Model. 2007, 47, 1536-1544 10.1021/ci700078n CCC: $37.00 © 2007 American Chemical Society Published on Web 07/03/2007