Enantiomer discrimination by enzymes is a very accurate mechanism, which often involves few amino acids located at the active site. Lipase isoforms from Candida rugosa are very good enzymatic models to study this phenomenon as they display high sequence homology (> 80%) and their enantioselectivity is often pointed out. In the present work, we investigated three lipases from C. rugosa (Lip1, Lip3, and Lip4, respectively) towards the resolution of 2-bromo-arylacetic acid esters, an important class of chemical intermediates in the pharmaceutical industry. All exhibited a high enantioselectivity, with Lip4 preferring the R-enantiomer (E-value = 15), while Lip1 and Lip3 showed an S-enantioselectivity > 200. A combination of sequence and structure analysis of the three C. rugosa lipases suggested that position 296 could play a role in S- or R-enantiomer preference of C. rugosa lipases. This led to the construction by site-directed mutagenesis of Lip1 and Lip4 variants in which position 296 was, respectively, exchanged by a Gly, Ala, Leu, or Phe amino acid. Screening of these variants for their enantioselectivity toward 2-bromo phenyl acetic acid octyl esters revealed that steric hindrance of the amino acid residue introduced at position 296 controls both the enantiopreference and the enantioselectivity value of the lipase: bulkier is the amino acid at position 296, larger is the selectivity towards the S-enantiomer. To investigate further these observations at an atomic level, we carried out a preliminary modeling study of the tetrahedral intermediates formed by Lip1 and Lip4 with the (R, S)-2-bromo-phenylacetic acid octyl ester enantiomers that provides some insight regarding the determinants responsible for lipase enantiodiscrimination. Biotechnol. Bioeng. 2011;108: 1749-1756. (C) 2011 Wiley Periodicals, Inc.