Development and demonstration of a trinucleotide exchange method for the directed evolution of proteins

Every living thing contains thousands of different proteins that carry out most of the crucial jobs needed to maintain life. Proteins are synthesised as a linear sequence of amino acids, which then fold to form their functional 3D structures. All the information required for a protein to reach its functional shape is encoded in the linear sequence of amino acids, which in turn is encoded by the DNA sequence of the gene for that protein. The process of evolution involves changes to the amino acid sequence of a protein through changes to the gene. As a consequence, these changes can alter the properties of the protein and may translate into beneficial effects, allowing the organism to survive a particular environmental challenge. How changes to the amino acid sequence of a protein translate into changes in the properties of a protein is one of the most fundamental questions in biology. The advent of genetic engineering has enabled us to change the DNA sequence of genes at will and so change the nature of the linear sequence of amino acids in the protein. This has allowed us to understand how certain amino acids contribute towards the properties of a protein. It has also allowed us to modify the characteristics of proteins for use in unnatural environments such as for industrial applications. The 3D structure of a protein is highly complex and our understanding of it is still limited. Therefore our ability to predict how a particular designed mutation will affect the properties of the protein is also limited. Nature takes a different, less rational approach, by introducing mutations at random and selecting only those mutations that are of benefit. As nature has already been very successful in adapting proteins for a multitude of functions, the proposed research aims to copy the process of evolution in the laboratory by developing a method that can introduce mutations randomly into a gene in order to change the properties of a protein. We will attempt to modify the characteristics of a protein called TEM-1 beta-lactamase, one of the proteins responsible for resistance to antibiotics such as penicillin. The advent of bacterial resistance to the traditional antibiotics led to the development of new and improved penicillin-like antibiotics. Nature was quick to respond and variants of TEM-1 soon evolved to overcome the toxic effects of these new antibiotics. Using our new method, the project will investigate how and which mutations contribute to adapting TEM-1 to detoxify these new antibiotics, helping us to understand the natural process by which this occurs.

The aim of this project is to develop and demonstrate a new and generally applicable method for creating the molecular diversity that is a prerequisite for the directed evolution of proteins. Directed evolution is a powerful and successful protein engineering strategy that is based on the principles of natural Darwinian evolution. It involves the creation of molecular diversity by the introduction of mutations at random positions within a gene followed by the selection of appropriate protein variants with new or improved characteristics. Although a variety of methods exist for generating molecular diversity at the genetic level, there are several limitations. These include error, codon and amplification biases, sampling of a restricted set of amino acids and the masking of beneficial mutations by deleterious ones. To overcome these limitations, a new method called trinucleotide exchange (TriNEx) is proposed. TriNEx involves the removal of a single trinucleotide sequence at random positions throughout a gene of interest followed by its replacement with a second (randomised or defined) trinucleotide sequence. This will allow the generation of new protein variants with one amino acid substituted for any other amino acid at random positions throughout the polypeptide. It is a novel extension of a transposon-based directed evolution method recently developed in my laboratory to sample sequence and conformational diversity due to amino acid deletions. To demonstrate the feasibility of the method, the TriNEx concept will be applied to our model system, TEM-1 beta-lactamase. TriNEx will be used to construct a library of gene variants with trinucleotide substitutions spread throughout the gene encoding TEM-1. To show that TriNEx can introduce mutations that influence the properties of a protein, TEM-1 variants that have improved hydrolytic activity towards two normally poor substrates for the wild-type enzyme, will be selected from the library.