Technical Highlight - September 2012
Short description: A new protocol for automated NMR protein structure determination minimizes instrument time and efficiently produces high-quality structures.
NMR structure determination of proteins with 150 residues or less has become increasingly automated, and several distinct protocols are now available. Compared to the traditional interactive approach, automated protocols are less biased and more efficient, yet often require some manual input during the data flow. Bottlenecks are typically encountered during either chemical shift assignment or compilation of distance restraints from nuclear Overhauser enhancement (NOE) data.
Wüthrich and colleagues (PSI JCSG) have developed a streamlined protocol, J-UNIO, based on automated projection spectroscopy (APSY) and NOE experiments for data acquisition and the modular UNIO software for resonance assignment and structure calculation. The authors selected the 115-residue protein YP_926445.1 from the Gram-negative bacterium Shewanella amazonensis to illustrate the J-UNIO procedure. Assessment of an NMR profile of signal intensities from a microscale sample indicated that the APSY experiments would succeed, and a UNIO module was used to assign the backbone chemical shifts. The authors next used another UNIO module to generate side-chain assignments from the same 1H-1H NOE data sets that provide the distance constraints for automated calculation of an intermediate “Structure A” for interactive validation. Further NOE assignments and distance restraints were then obtained, allowing calculation of a high-resolution “Structure V” ensemble which was rigorously validated using several quality criteria. The YP_926445.1 protein contains a novel fold used to generate a new Pfam family, which now includes members from over 50 bacterial species.
The authors applied the J-UNIO protocol to 17 targets from the JCSG pipeline ranging in size from 67 to 149 residues. To maximize success rates, the authors were relatively generous with instrument time and acquired high-resolution APSY data resulting in backbone assignments that were generally between 81–96% complete and contained very few errors. For the set of 17 proteins, extensions of the assignments to near completion was achieved in only a few hours of interactive analysis. The authors caution that while up to 5% erroneous NOE-based side chain assignments are possible, these are permissive with respect to calculating the correct fold. They further note that the J-UNIO protocol starts from the raw data, unlike most other automated procedures. Further comparisons of automated protocols will determine which will find broad applicability in the protein NMR community.
P. Serrano et al. The J-UNIO protocol for automated protein structure determination by NMR in solution.
J. Biomol. NMR. 53, 341-354 (2012). doi:10.1007/s10858-012-9645-2