Investigating RNAs Involved in Translational Control by NMR and SAXS - Biophysical Approaches to Translational Control of Gene Expression - page 143

directly bonded and neighboring nuclei can be used to help assign the identity of the
proton. Thus, experiments that correlate a proton chemical shift to its directly
bonded nucleus, such as 2D heteronuclear single quantum coherence (HSQC), are
invaluable for proton resonance assignment (Hennig et al. 2001 ) . Experiments that
correlate multiple bonds can provide even more information. For example, triple
resonance HCN-type heteronuclear NMR experiments can be used to correlate
nucleobase aromatic protons to a particular ribose group by appropriately transfer-
ring magnetization through the glycosidic bond (Sklenář et al. 1998 ) .
The second strategy for proton resonance assignment is based on through-space
transfer of magnetization, via the NOE. The nucleotide structure contains a number
of short (<6 Å) interproton distances that give rise to NOEs. Additionally, RNA
structures are dominated by the A-form helical geometry, which gives rise to pre-
dictable NOE patterns that can be used to obtain resonance assignments of neigh-
boring nucleotides (Hennig et al. 2001 ). Because the intensity of the NOE is
distance-dependent ( r −6 ), NOEs obtained from these experiments are utilized as dis-
tance restraints for use in restrained molecular dynamics simulations to produce an
ensemble of structures that satisfy the restraints (Lukavsky and Puglisi 2005 ) .
Torsion angle restraints are also important for structure determination. The sugar-
phosphate backbone and ribose ring are characterized by six backbone torsion
angles and the ribose sugar pucker, respectively. Ribose sugar puckers in A-form
RNA are C3¢-endo, but in non-helical regions such as loops the ribose ring may
adopt a C2¢-endo conformation or undergo dynamic exchange between C3¢ - and
C2¢-endo conformations. The sugar pucker can be analyzed using 2D COSY or
TOCSY experiments (Hennig et al. 2001 ). Backbone torsion angle measurements
around the phosphodiester bond can be quantitatively measured (Nozinovic et al.
2010a, b ). However, experiments designed to quantitatively measure backbone tor-
sion angles are often impractical for RNAs larger than 30 nucleotides, because they
rely on the ability to resolve phosphorous chemical shifts, which are usually exten-
sively overlapped.
Ideally, an NMR structure should be constrained by a large number of long-range
NOEs between protons far apart in sequence (Allain and Varani 1997 ) . However,
the vast majority of NOEs occur between protons within a nucleotide (intraresidue)
or between neighboring nucleotides. While these NOEs are important for establish-
ing resonance assignments and defining local structure, they cannot sufficiently
define the global RNA structure. RNA NMR structure quality can therefore be
significantly improved by inclusion of long-range distance restraints. Such informa-
tion can be obtained through the measurement of residual dipolar couplings (RDCs)
(Prestegard et al. 2000 ; Bax et al. 2001 ). RDCs measure the orientation of a bond
vector relative to the overall alignment of a molecule within the magnetic field.
Therefore, RDCs provide angular restraints that improve both the local and global
structural quality (Tzakos et al. 2006 ) and can also be used to characterize RNA
dynamics (discussed in greater detail in Sect. 8.1.2.4 ).
RDCs are not normally observed in solution under isotropic conditions because
the overall tumbling of a molecule causes the angular term in the RDC equation
(Fig. 8.4 ) to average to a value that approaches zero (Tolman 2001 ) . However, by