Editor's Summary
16 August 2007
Form finds function
Though it is often possible to infer the function of a newly discovered protein by comparing its sequence to those of other characterized proteins, it can be extremely difficult to predict the function of an enzyme that is unrelated to other studied proteins. Hermann et al. now show that it is possible to use a computational approach to predict the function of an enzyme of unknown activity. They 'docked' high-energy intermediate forms of thousands of candidate metabolites to the X-ray crystal structure of Tm0936, a member of the amidohydrolase superfamily. These experiments predicted that the enzyme would deaminate 5-methylthioadenosine and S-adenosylhomocysteine; this was borne out biochemically, and the X-ray crystal structure of one of the products bound to Tm0936 corresponded closely to the predicted structure. These results suggest that structure-based docking using high-energy forms of potential substrates may be a useful tool to annotate enzymes for function.
News and Views: Computational biochemistry: Models of transition
Is it possible to determine the role of an enzyme from its structure? The latest findings suggest that it is, and prove the point by predicting the substrate for an enzyme of unknown function.
JoAnne Stubbe
doi:10.1038/448762a
Article: Structure-based activity prediction for an enzyme of unknown function
Johannes C. Hermann, Ricardo Marti-Arbona, Alexander A. Fedorov, Elena Fedorov, Steven C. Almo, Brian K. Shoichet & Frank M. Raushel
doi:10.1038/nature05981
Abstract | Full Text | PDF (751K) | Supplementary information