The effect of high-frequency random mutagenesis on in vitro protein evolution:: A study on TEM-1 β-lactamase

For a number of years a major limitation in genetic analysis of protein function has been the inability to introduce multiple substitutions at distant sites that would enable the selection of clusters of mutations required for improved or novel biological functions. In order to achieve this, we have recently developed a novel mutagenesis procedure in which the triphosphate derivatives of a pyrimidine (6-(2-deoxy-beta-D-ribofuranosyl)-3,4-dihydro-8H-pyrimido-[4,5-c][1,2]oxazin-7-one; dP) and a purine (8-oxo-2'-deoxyguanosine; 8-oxodG) nucleoside analogue are employed in DNA synthesis reactions in vitro. The procedure allows control of the mutational load and can yield frequencies of amino acid residue substitutions at least one order of magnitude greater than those previously achieved. Here we report the results of an experiment in which we have hypermutated the bacterial enzyme TEM-l beta-lactamase and selected small pools (<1.5 x 10(5)) of clones for enzymatic activity against the beta-lactam antibiotic cefotaxime. The experiment resulted in the isolation of a number of TEM-1 mutants with greatly improved activity against cefotaxime. Among these, clone 3D.5 (E104K:M182T:G238S) exhibited a minimum inhibitory concentration for cefotaxime 20,000-fold higher than wild-type TEM-1 and a catalytic efficiency (k(cat)/K-m) 2383 times higher than the wild-type enzyme. Thus, small pools of hypermutated sequences enabled the selection of one of the most active extended beta-lactamases described so far. These results argue against the accepted view that multiple rounds of low-rate mutagenesis and stepwise selection are essential for in vitro protein evolution and extend the scope of directed molecular evolution to proteins for which no genetic selection is available. (C) 1999 Academic Press.