Abstract

This is a report of a de novo design study that aimed to identify novel structures capable of inhibiting the human immune deficiency virus (HIV-1) protease ligand binding pocket (HIV-1 PR_LBP). Baseline information regarding ligand binding modality and affinity was obtained through analysis of the pdb crystallographic depositions describing the HIV-1 PR enzyme complexed with small molecule inhibitors currently available on the Protein Data Bank (PDB). Molecular visualisation and modelling was carried out using SYBYL^® 1.1, and in silico predicted ligand binding affinity (LBA) was quantified using XSCORE_V1.3. The de novo design phase of the study was based on the utilisation of the bound co-ordinates of Lopinavir. This particular HIV-1 protease (HIV-1 PR) inhibitor was selected as a template owing to its superior in vivoactivity and unique binding modality. Based on literature derived data, the cyclic urea moiety of Lopinavir was retained as a seed fragment, overlaid onto its counterpart moiety and planted into the HIV-1 PR_LBP with growth being allowed according to defined parameters utilising the genetic algorithm embedded in the GROW module of LIGBUILDER^®V1.2. The result was the identification of 200 de novo designed structures with a predicted in silico ligand binding affinity (LBA) (pK_d) ranging between 9.63 and 10.00. A smaller cohort (n = 35) was also Lipinski rule of 5 complaint. The implication of this study consequently is that this series of novel structures may be compiled into a library that may be of utility in high throughput screening (HTS) processes and future iterative optimisation.

Key words: High-throughput screening (HTS), protein data bank (PDB) depositions, ligand binding pocket (LBP), ligand binding affinity (LBA), human immune deficiency virus (HIV)-1 protease (HIV-1 PR), acquired immune deficiency syndrome (AIDS).