Generation of noble-gas binding sites for crystallographic phasing using site-directed mutagenesis

In recent years, the use of noble-gas complexes for the de novo phasing of protein structures has proven to be a useful alternative to selenomethionine and more traditional derivatives, largely owing to improvements in methods for incorporating noble gases within protein crystals. Advantages of noble-gas derivatives include a high degree of isomorphism when using xenon for multiple isomorphous replacement (MIR) and an easily accessible absorption edge when using krypton for multiwavelength anomalous dispersion (MAD) phasing. One problem with this approach is that not all proteins contain cavities which bind noble gases. Even in proteins which do bind noble gases, the resulting derivative may not be sufficient for phasing. Using T4 lysozyme as an example, it is illustrated how this limitation might be overcome by using `large-to-small' mutations to introduce potential noble-gas binding sites. Wild-type T4 lysozyme contains a single xenon-binding site. By truncating leucine and phenylalanine residues to alanine, it is possible to generate additional noble-gas binding sites within the core of the protein. Combining rotating-anode data from xenon complexes of wild-type and mutant lysozymes yields MIR phases which compare favorably with those determined from a selenomethionine MAD experiment conducted at a synchrotron. Experience with T4 lysozyme suggests that a leucine-to-alanine substitution made at random in a protein of unknown structure has about a 30% chance of providing a useful derivative. This procedure holds promise for the determination of unknown protein structures, especially when selenomethionine-containing protein is not available or when access to a synchrotron is limited.