Stereospecific assignments in proteins using exact NOEs

Abstract

Recently developed methods to measure distances in proteins with high accuracy by "exact" nuclear Overhauser effects (eNOEs) make it possible to determine stereospecific assignments, which are particularly important to fully exploit the accuracy of the eNOE distance measurements. Stereospecific assignments are determined by comparing the eNOE-derived distances to protein structure bundles calculated without stereospecific assignments, or an independently determined crystal structure. The absolute and relative CYANA target function difference upon swapping the stereospecific assignment of a diastereotopic group yields the respective stereospecific assignment. We applied the method to the eNOE data set that has recently been obtained for the third immunoglobulin-binding domain of protein G (GB3). The 884 eNOEs provide relevant data for 47 of the total of 75 diastereotopic groups. Stereospecific assignments could be established for 45 diastereotopic groups (96 %) using the X-ray structure, or for 27 diastereotopic groups (57 %) using structures calculated with the eNOE data set without stereospecific assignments, all of which are in agreement with those determined previously. The latter case is relevant for structure determinations based on eNOEs. The accuracy of the eNOE distance measurements is crucial for making stereospecific assignments because applying the same method to the traditional NOE data set for GB3 with imprecise upper distance bounds yields only 13 correct stereospecific assignments using the X-ray structure or 2 correct stereospecific assignments using NMR structures calculated without stereospecific assignments.

Compiled data set of exact NOE distance limits, residual dipolar couplings and scalar couplings for the protein GB3

“…These values are used here since a potentially wrong stereoassignment would not have an impact. The couplings of residue 52 are also in disagreement with the data set in reference [19] , the X-ray structure [8] , and our eNOE-based stereospecific assignment [20] , all of which suggest a single rotamer state. On the other hand, the 3 J C’,Cγ and 3 J N,Cγ couplings for residue 52 in reference [2] appear to be slightly averaged over at least two rotamer states.…”

How accurately do force fields represent protein side chain ensembles?

“…The other test case in our benchmark is the third IgG-binding domain of protein G. GB3 is a 56 residue protein extensively studied experimentally by NMR 41,62,[80][81][82][83] and, hence, often used for FF validation studies. 23,26,36,37,39,52 In a recent publication, Bax and co-workers estimated the conformational distribution of side chains in GB3 using a model-free analysis of RDCs.…”

The Exact Nuclear Overhauser Enhancement: Recent Advances

“…Figure 5 b shows the impact of supplementing eNOEs with gn-eNOEs on the 165-residue enzyme cyclophilin A, which resulted in a much tighter bundle than with eNOEs alone or with conventional NOEs. We have also developed a method for stereospecific assignments for the majority of relevant diastereotopic groups by comparing eNOE-derived distances to protein structure bundles calculated without stereospecific assignments, making it possible to obtain more detailed structural and dynamical information from NOEs [ 42 ].…”