Featured Article - May 2014
Short description: Structural modeling and docking of substrates identifies a biochemical pathway that is validated experimentally.
Within high-throughput sequencing and structure data, many proteins have unknown function while others are misannotated. In a previous functional identification effort, researchers analyzed a bacterial protein with a solved three-dimensional (3D) structure, Pelagibaca bermudensis HpbD, by performing in silico docking of a library of metabolites within its active site. Follow-up studies confirmed the activity of pbHpbD with the identified metabolite, proline betaine (Pro-B), which functions as an osmoprotectant in high-osmolarity conditions, or as a source of carbon and nitrogen in others.
Cronan, Gerlt and colleagues (Enzyme Function Initiative) have now extended their previous study by identifying a protein with similar activity in a different species, Paracoccus denitrificans. With a sequence identity of 64%, modeling the structure of pdHpbD on that of pbHpbD allowed docking of Pro-B into the former's active site, using the in-house software Protein Local Optimization. X-ray structures of pdHpbD confirmed the modeling results (PDB 4J1O).
To find other proteins involved in the Pro-B degradation pathway, the authors searched the genomic neighborhood of hpbD for related genes and, based on P. bermudensis data, predicted and validated the Pro-B catalysis pathway for P. denitrificans using genetic complementation and metabolyte analyses. Further, when a deletional analysis implicated HpbR in substrate-mediated transcriptional regulation, a BLAST search using the substrate-binding domain identified a protein with little overall sequence identity, but one that enabled homology modeling of the substrate-binding site and successful docking of Pro-B. Subsequent RT-PCRn analyses confirmed HpbR as a regulator of Pro-B catabolic genes.
The authors advocate this approach for function discovery in bacterial enzymes, combining co-location of genes with in silico metabolite docking into known or modeled 3D structures, in addition to guidance and confirmation using genetic and metabolite analyses.
R. Kumar et al. Prediction and biochemical demonstration of a catabolic pathway for the osmoprotectant proline betaine.
MBio. 5, e00933-13 (2014). doi:10.1128/mBio.00933-13