Investigating RNAs Involved in Translational Control by NMR and SAXS - Biophysical Approaches to Translational Control of Gene Expression - page 144

Fig. 8.4 RDCs ( D ij ) between spins i and j provide long-range distance restraints for the average
orientation ( q ) of the internuclear vector relative to the magnetic field ( B o ). Here, g i and g j are the
gyromagnetic ratios of spins i and j , respectively, r ij is their internuclear distance, h is Plank's con-
stant, m o is the magnetic permittivity of vacuum, and the angular term is shown in brackets . RDCs
are not observed in isotropic solution conditions because the angular term averages to nearly zero.
Adapted from Getz et al. ( 2007 )

imparting a small degree of alignment to the sample, the dipolar coupling retains
information about the angle of a bond vector relative to the alignment tensor. RDC
values are determined by subtracting the measured dipolar coupling in isotropic
conditions from the coupling in partially aligned conditions. For RNA, alignment is
typically achieved by adding filamentous bacteriophage (Pf1) to the RNA (Prestegard
et al. 2000 ; Hansen et al. 1998 ) .

8.1.2.3 Solution Structures of Large RNAs

Solving solution structures of even moderately sized RNAs (50-100 nts) can be
extremely challenging due to spectral overlap and fast relaxation rates. Only a hand-
ful of RNA structures greater than 75 nucleotides have been solved using NMR.
These include the 77 nt hepatitis C viral (HCV) internal ribosome entry site (IRES)
domain II (Lukavsky et al. 2003 ), an 86 nt tetraloop receptor complex (Davis et al.
2005 ), a 101 nt core encapsidation signal of Moloney murine leukemia virus
(MoMuLV) (D'Souza et al. 2004 ) (see Sect. 8.2 ), a 102 nt ribosome-binding struc-
tural element (RBSE) in Turnip Crinkle Virus (TCV) (Zuo et al. 2010 ) , and the 140
nt dimeric genome packing signal in MoMuLV (Miyazaki et al. 2010 ) . Each of
these studies employed a different approach to reduce spectral overlap and improve
NMR relaxation rates associated with large macromolecules (Fig. 8.5 ) .
The most common approach for studying large RNAs has been the so-called divide
and conquer approach (reviewed in Clos et al. 2011 ). Here, the RNA is separated into
small, thermodynamically stable helical domains. By breaking the RNA into smaller
domains, spectral overlap is reduced and assignment of the domain is often feasible.
Unfortunately, the structure of the subdomain may change in the context of the larger,
more biologically relevant RNA, if it is involved in tertiary interactions (Tzakos et al.
2006 ). Therefore, it is critical to verify that the chemical shifts for each subdomain are
consistent with those observed in the context of the larger RNA.