Molecular recognition studies of chloramphenicol acetyltransferase by substrate modification.

This thesis was a study of the molecular recognition process of chloramphenicol acetyltransferase (CAT). In an approach which is complementary to site-directed mutagenesis, the substrate, chloramphenicol (CM), has been modified in a variety of ways to allow quantitative evaluations of the energetics of individual enzyme-ligand interactions to be made. Nineteen diastereo- and enantiopure chloramphenicol analogues have been prepared and their kinetic parameters as substrates for CAT determined. These compounds were derived from L- and D-amino acids and chloramphenicol base, (1R, 2R)-2-amino-1-(p-nitrophenyl)-propan-1,3-diol. The asymmetric synthesis of a series of 2-amino 1,3-diols via alternative routes was investigated. The first started with the N-BOC-protected amino acid D-serine and involved conversion of the acid to a ketone with organolithium reagents followed by asymmetric reduction to give the diol. For the second route N,O-diprotected D-serinal was converted to the amino diols via reaction with organometallic reagents. Addition of [RCu]MgX2 (X=Br or Cl) to the aldehyde led to essentially enantiopure single diastereomers with the threo-configuration. From kinetic studies of the CAT reaction with the modified ligands the apparent binding energies, Gapp, of each of the substituents of chloramphenicol has been evaluated. Examination of analogues in which the p-nitrophenyl group of chloramphenicol was replaced by alkyl groups suggested that the aryl binding pocket is a local hydrophobic region approximately 1.5 times more hydrophobic than n-octanol. There is a linear free energy relationship between the specificity constant, kcat/Km, of these ligands and the hydrophobicity of the alkyl substituent. The incremental apparent binding energy for each methylene group bound in this pocket is estimated to be 1.0 kcal mol-1 and it is proposed that this binding site can only enclose an alkyl chain up to C4. A large number of chloramphenicol analogues were found to be substrates for CAT. This is believed to be due to the dominant hydrophobic interactions in the CAT:CM complex and the lack of many direct hydrogen bonds, which for some ligands apparently leads to alternative binding modes.