Theoretical evaluation of electronic properties of small heterocycles to be integrated into (3D-(Q)) SAR analyses

Nowadays, computational chemists involved in drug design are facing a dilemn. They are tempted to take full advantage of the evolution of computers (in terms of power and diversity (from workstations to PC's)) and have to adapt their methodologies to these new tools. At the same time, if they want to participate to the process of drug design, they need to produce rapidly accurate theoretical results (including electronic parameters such as molecular electrostatic potential, (Mulliken or potential derived) atomic charges, frontier orbitals (HOMO, LUMO) topologies and energies, dipole moment, ...) to be included in the elaboration of pharmacophores and in (Q)SAR studies. In this contribution, we evaluate the electronic properties derived from ab initio calculations (Restricted Hartree-Fock level) with a particular emphasis on methods by which such calculations can be carried out rapidly and reliably. Several calculations of those properties have been performed at different levels of sophistication [STO-3G(*), 3-21G(*), 4-31G(*), 6-31G(*). 6-3 IG**, 6-311G(*)] and will be discussed in this work. The compounds used for this study are small heterocycles containing N, O and S atoms (Fig. 1) observed in molecules presenting a pharmacological behavior. All of these compounds present a non trivial electronic distribution. The problem of producing a reliable geometry as input for those fragments has also been adressed and geometry optimization procedures (based on both molecular and quantum mechanic energy calculations) were investigated with respect to experimental data. From this experience, it comes out that most computational efforts have to be given for the evaluation of the electronic properties, the 3-21G basis set being a good compromise, while geometries can be successfully obtained from force-field based (eg cff91) MM calculations when no X-ray staicture are available. This procedure is now to be extended to a larger range of molecules.