Discovery Medicine, Volume 3, Number 18, October 2003 36 Targeted Mutagenesis in E. coli: A Powerful Tool for the Generation of Random Mutant Libraries Manel Camps and Lawrence A. Loeb* The Joseph Gottstein Memorial Cancer Research Laboratory, Department of Pathology, University of Washington, Seattle, WA 98195-7705, USA *Correspondence: laloeb@u.washington.edu M utagenesis is widely used in the fields of genetics, enzyme catalysis, ligand-binding recognition, meta- bolic regulation, control of gene expression, and to study mechanisms of DNA repair. Mutations are also introduced to optimize enzyme performance. This area has attracted renewed interest with the increased use of biocatalysts in chemical and pharmaceutical synthesis, in bioremediation, and in biotechnology. Individual mutations are introduced, based on detailed structural and functional information, to generate enzymes or other proteins with novel properties and to change the properties of regulatory sequences such as promoters and origins of replication (site-directed mutagenesis). Alternatively, variants of interest can be identified in large libraries harboring random substitutions (random mutagen- esis). Unlike site-directed mutagenesis, random mutagene- sis requires little or no previous information of the targeted genes. This approach, however, typically allows only a small fraction of all possible mutants to be analyzed. This is due to a number of factors including the inordinate num- bers of possible mutations, the need to identify individual mutants of interest from a large pool, the generation of non-functional mutants, and the nature of the genetic code. Libraries containing random mutations may be generated in vitro by treatment of DNA with mutagens, by mutagenic PCR, or by substituting portions of plasmid-encoded genes with oligonucleotides containing random substitutions. These protocols have few limitations on the number of mutations that may be introduced per gene. Random mutant libraries may also be generated in vivo by replicat- ing the DNA encoding the sequence of interest in a muta- genic cellular environment. Libraries generated in "muta- tor strains" tend to exhibit a wider spectrum of mutants given that no cloning steps are involved. However, the complexity of these libraries (i.e., the number of mutants containing multiple substitutions) is restricted by the fact that high levels of mutagenesis normally interfere with the survival of the host. To overcome this limitation, complex combinations of mutations may be built up by allowing the accumulation of mutations with a positive effect on the desired phenotype through iterative mutagenesis and selec- tion, a strategy known as "directed evolution." A system of random mutagenesis specifically designed to facilitate the directed evolution of proteins or of regulatory sequences has recently been developed (Camps et al., 2003). The system is delineated in the figure below. Mutagenesis is performed in E. coli cells transformed with 2-plasmids: one plasmid encodes an error-prone Pol I pro- tein ("mutagenic plasmid") and the second plasmid encodes the sequence to be mutagenized ("target plas- mid"). The mutagenic plasmid is Pol I-independent. Errors in the replication of the target plasmid by error-prone Pol I result in mutations. Provided that a complementation or drug selection can be established in the strain used for mutagenesis, a continuous positive selection should lead to the directed evolution of the target sequence. Alternatively, the libraries thus generated may be retrieved and retransformed into the appropriate indicator strain for selection or screening. In this case, directed evolution may be accomplished by iterating mutagenesis with selection/screening of the recovered library. The 2-plasmid system was first described by Fabret et al. (2000), which expressed a mutant of Pol I deficient in proofreading (exo - ) in a mismatch-deficient strain. In an advanced version of this system, we generated a mutant of Pol I containing Targeted Mutagenesis Mutagenic Plasmid D424A I709N A759R Target Plasmid Target Plasmid Target Plasmid Target Plasmid Chromosome Error-prone Pol I Target Plasmid Target Plasmid Target Plasmid Target Plasmid Chromosome Mutagenic Plasmid D424A I709N A759R Error-prone Pol I E coli Mutagenesis Repetitive plasmid recovery and transformation Directed evolution continuous selection 1 2 Transfection into indicator cells plasmid recovery