Random and site-directed mutagenesis of bacterial luciferase: investigation of the aldehyde binding site

Abstract
Numerous luciferase structural gene mutants of Vibrio harveyi have been generated by random mutagenesis and phenotypically characterized [Cline, T. W., and Hastings, J. W. (1972) Biochemistry 11, 3359-3370]. All mutants selected by Cline and Hastings for altered kinetics in the bioluminescence reaction had lesions in the .alpha. subunit. One of these mutants, AK-20, has normal or slightly enhanced thermal stability and enhanced FMNH2 binding affinity but a much-reduced quantum yield of bioluminescence and dramatically altered stability of the aldehyde-C4a-peroxydihydroflavin-luciferase intermediate (IIA), with a different aldehyde chain length dependence from that of the wild-type luciferase. To better understand the structural aspects of the aldehyde binding site in bacterial luciferase, we have cloned the luxAB genes from the V. harveyi mutant AK-20, determined the nucleotide sequence of the entire luxA gene, and determined the mutation to be TCT.fwdarw.TTT, resulting in a change of serine .fwdarw. phenylalanine at position 227 of the .alpha. subunit. To confirm that this alteration caused the altered kinetic properties of AK-20, we reverted the AK-20 luxA gene by oligonucleotide-directed site-specific mutagenesis to the wild-type sequence and found that the resulting enzyme is indistinguishable from the wild-type luciferase with respect to quantum yield, FMNH2 binding affinity, and intermediate IIA decay rates with 1-octanal, 1-decanl, and 1-dodecanal. To investigate the cause of the AK-20 phenotype, i.e., whether the phenotype is due to loss of the seryl residue or to the properties of the phenylalanyl residue, we have constructed mutants with alanine, tyrosine, and tryptophan at .alpha.227. The luciferases with .alpha.227 Tyr and .alpha.227 Trp exhibited kinetic and bioluminescence properties comparable to those of the AK-20 luciferase (.alpha.227 Phe) and distinctly different from those of the wild-type luciferase (.alpha.224 Ser), whereas luciferase with .alpha.227 Ala was similar to the wild-type enzyme in kinetics and quantum yield, suggesting that the properties of AK-20 must be due more to the introduction of the bulky aromatic group at .alpha.227 than to the loss of the serine hydroxyl group at that position. The results obtained here do not support a role of .alpha.Ser227 in catalysis, since the .alpha.Ala227 mutant has nearly wild-type activity.