Localized Random Polymerase Chain Reaction Mutagenesis

Since the invention of the polymerase chain reaction (PCR) for in vitro amplification of DNA using thermostable DNA polymerase (1 ), an impressive assortment of powerful modifications of the method have been further developed (2 ,3 ). Until recently, surprisingly little attention has been paid to application of PCR as a tool for localized random mutagenesis. Taq DNA polymerase (from Thermus aquaticus ), the most commonly used thermal stable DNA polymerase in PCR, exhibits a relatively high single base misincorporation rate of 10⁻⁴ since it lacks detectable 3–5′ proofreading exonuclease activity (4 ,5 ). Depending on the conditions, including pH and relative concentra tions of MgCl₂ and deoxynucleotide triphosphates, the base substitution error rates of Taq DNA polymerase can vary between 1/3200 and 1/ 300,000 (5 ). Base substitution error rates on the order of 1/10,000 are common using standard PCR conditions (PE Cetus, Norwalk, CT kit, 4). Other factors that influence misincorporation events during PCR include the length of the sequence, the number of amplification cycles, and the initial number of template copies (6 ). Thus, PCR products obtained using Taq DNA polymerase contain substitution mutations at low but appreciable frequencies. Although it is generally desirable to minimize the misincorporation activity in conventional PCR amplification, the mutagenic activity of Taq DNA polymerase can be successfully exploited to rapidly isolate random mutations through-out an entire or defined region of a gene.