Featured Article - January 2014
Short description: Homology modeling and metabolite docking predict in vitro enzymatic activities and physiological functions.
While the annotation of large-scale sequencing projects can often predict the function of uncharacterized proteins based on sequence homology, those predictions are prone to errors; thus, assigning valid function based on homology alone remains a challenge. To demonstrate the utility of computation-guided approaches to the prediction of enzymatic function, Jacobson and colleagues (PSI NYSGRC) developed a structure-guided strategy to discover a previously undocumented activity, 4R-hydroxyproline betaine 2-epimerase (Hyp-B 2-epimerase), and the associated catabolic pathway by which trans-4-hydroxy-L-proline-betaine (tHyp-b) is converted to α-ketoglutarate.
As a starting point, the authors used the structure of an uncharacterized member of the enolase superfamily, HpbD from marine bacterium Pelagibaca bermudensis, solved as part of the NYSGRC's efforts (PDB 2PMQ). A library of ∼87,000 ligands, including metabolites and potential enolase substrates, was used for in silico docking to HpbD's active site. The top scoring hits, as ranked by calculated binding affinities, contained several proline analogs, leading to the prediction that the enzyme was an amino acid racemase or epimerase.
Because bacterial genes encoding proteins in the same metabolic pathway are often clustered, the authors carried out homology modeling of an ABC-type transporter encoded by hpbJ, a gene located next to hpdD that had been previously annotated as binding glycine betaine or L-proline. The HpbJ model indicated that its ligand was indeed likely a betaine, and virtual docking of a library of betaines ranked tHyp-B as having the highest affinity. Modeling of a protein encoded by another hpbD neighbor—HpbB1— also predicted a betaine substrate.
Collectively, these results suggested a pathway in which HpbD uses tHyp-B as substrate in a 1,1-proton transfer (epimerase) reaction. The authors verified this prediction with enzymatic analyses and by solving the structure of HpbD bound to tHyp-B (PDB 4H2H). They further speculated that these proteins might constitute a catabolic pathway that degrades tHyp-B to α-ketoglutarate, with HpbD catalyzing the first reaction: epimerization of tHyp-B to cHyp-B.
Using gene deletions, the authors carried out functional analyses of the hpbD gene cluster in Paracoccus denitrificans and demonstrated that the proteins are part of a pathway that controls the intracellular concentration of tHyp-B. Under low salt growth conditions, tHyp-B was used as a carbon source, in a catabolic pathway beginning with the epimerization of tHyp-B and culminating in the production of α-ketoglutarate. In contrast, under high salt conditions, the pathway was inactivated, allowing for the accumulation of tHyp-B, which then acted as an osmoprotectant. The expression of HpbD was downregulated under high salt conditions, and this mode of regulation could serve as an osmolarity-dependent switch that determines either the accumulation of tHyp-B or its utilization as an alternative carbon and nitrogen source.
S. Zhao et al. Discovery of new enzymes and metabolic pathways by using structure and genome context.
Nature. 502, 698-702 (2013). doi:10.1038/nature12576