Protein engineering

Abstract

Relating primary sequence to three-dimensional structure has long been the holy grail of structural biology and appears to be far from achievement. Within grasp however, is the use of intuitive or unintuitive methodology to modify existing known protein structures to achieve the desired effect. We use protein engineering as a general term for the design of proteins with useful or valuable properties. The technique has become possible due to our increasing knowledge of detailed protein structures, which in turn highlights potential for improving key facets of protein structure; for example, the mutation of specific residues with a view to improving binding or catalysis. This rational design (Section 2.1) requires the scientist to have a detailed prior knowledge of the protein to attempt to make specific informed changes to the sequence to exert the desired effect. The technique is quite straightforward, involving mutation at the genetic level followed by expression and characterization. This site-directed mutagenesis approach is discussed in Section 2.1.1. However, rational mutations do not always generate the desired effect. This has invariably led to computer-based approaches for protein design. These are designed to save time in identifying mutations that generate the desired effect of low energy structures, and aim for lower the sequence conformation space that is required in the search. To simplify the procedure, these algorithms are based on approximations that require less processing time. Unfortunately, approximations can also lead to false positives which do not yield the predicted desired effect at the protein level. Computer aided protein engineering strategies are discussed in Section 2.1.2. The second protein engineering approach, known as directed evolution relies on a selection system to pick from a range of variants. This involves the construction of protein libraries that contain a wealth of randomized positions. The generation of libraries is discussed extensively in the chapter "Directed Protein Evolution" in this book. Many of these residues will be intuitively predicted to have the desired result, while for other residues the outcome of the change may not be known. By screening these mutations at the protein level for their desired function, sequences conforming to the best molecule for the desired role can be screened. This has the advantage over site-directed mutagenesis or computer-based design that you obtain exactly what you select for. Theory of library-based design strategies is discussed in Section 2.2, and includes a discussion as well as published examples of the phage display (2.2.1.), ribosome display (2.2.2.), and yeast two-hybrid systems (2.2.3) that have been used to screen protein libraries. Also discussed are the advantages and pitfalls of working with any one of these techniques. Protein-fragment complementation assay (PCA) systems are discussed (2.2.4) along with several examples of the screening system in action, as well as methods of cell surface display (2.2.5). Finally, in vitro compartmentalization methods are discussed (2.2.6). The chapter closes (Section 3) with a range of examples for each of the techniques highlighted. © 2008 Humana Press.