Construction and analysis of randomized protein-encoding libraries using error-prone PCR

Abstract

In contrast to site-directed mutagenesis and rational design, directed evolution harnesses Darwinian principles to identify proteins with new or improved properties. The critical first steps in a directed evolution experiment are as follows: (a) to introduce random diversity into the gene of interest and (b) to capture that diversity by cloning the resulting population of molecules into a suitable expression vector, en bloc. Error-prone PCR (epPCR) is a common method for introducing random mutations into a gene. In this chapter, we describe detailed protocols for epPCR and for the construction of large, maximally diverse libraries of cloned variants. We also describe the utility of an online program, PEDEL-AA, for analyzing the compositions of epPCR libraries. The methods described here were used to construct several libraries in our laboratory. A side-by-side comparison of the results is used to show that, ultimately, epPCR is a highly stochastic process.