Articles are listed most recent first.
Optimal isotope labeling for NMR protein structure determination.
M. Kainosho et al. (2007). Nature 440:52.
Designer labels (News & Views). S. J. Opella. 440:40.
A new technology is described called "stereo-array isotope labeling" or
SAIL. In contrast to prior random fractional deuteration methods, it
involves synthesizing the protein's amino acids such that each
has not more than one covalently bonded (signal-generating)
1H -- the remaining hydrogens being (silent)
2H (deuterium). Spectra are simplified yet sharpened.
Accuracy is improved, yet proteins twice as large as usual can
Maltose-binding protein (41 kD) was solved to a resolution of about
1 Å (2D21). By June, 2005, only 1% of NMR structures in the PDB
were for proteins >25 kD. Only eight cases >25 kD in BioResMagBank
had backbone and sidechain chemical shift assignments more than 70%
Traditional biomolecular structure determination by NMR spectroscopy
allows for major errors.
Sander B. Nabuurs, Chris. A. E. M. Spronk, Geerten W. Vuister, and
PLoS Computational Biology
Open Access Full Text.
Several published NMR structures are identified containing major
errors. It is concluded that current methods for quality
assessment of NMR-based models permit such errors at a worrisome
"It is interesting to note here that the three erroneous
structures described in this paper stem from premier protein NMR groups,
all involved in the development of structure validation and refinement
methodologies [refs], and that these methodologies either failed or were
not or incorrectly applied in identifying the serious errors present in
these structure ensembles."
Also interesting, they found no evidence for a difference in the likely
error rate between structural genomics groups vs. individual research groups.
Simultaneous determination of protein structure and dynamics.
Kresten Lindorff-Larsen, Robert B. Best, Mark A. DePristo, Christopher M.
Dobson, and Michele Vendruscolo
"We present a protocol for the expermental determination of ensembles
of protein conformations that represent simultaneously the native
[average] structure and its associated dynamics." The method is termed
Dynamic Ensemble Refinement (DER). Information about structural
variablility in solution on a pico- to nanosecond timescale was
obtained from NMR relaxation experiments. "Because the NOE-derived
distances are averages over the large ensemble of molecules ..., we
enforce them on an average calculated over a computer-generated
ensemble of structures, rather than on a single conformer."
Despite the use of unrestrained molecular dynamics simulations in DER,
when backbone residual dipolar couplings and side-chain scalar couplings
were back-calculated from the resulting ensemble of models, they
agreed well with experimental values. Since the latter were not used in
model building, this provides validation for the DER results.
analysis of ubiquitin in solution revealed that while the core
backbone has solid-like rigidity, some of the surface backbone as well
as all sidechains have liquid-like mobility, both on the
surface and more surprisingly in the hydrophobic core.
Nevertheless, the core was as tightly packed in the DER models as
in conventional X-ray crystallographic or NMR models. The DER ensemble
was better able to serve as a basis for predicting the known effects
of mutations on stability than were conventional crystal or NMR models.
Feedback to Eric Martz.