Garry C. King*, Zhijian Xi+ and Malgorzata J. Michnicka+
*School of Biochemistry and Molecular Genetics, The University of New South Wales, Sydney 2052 Australia, and +Department of Biochemistry and Cell Biology, Rice University, Houston TX 77005 USA.
The TAR RNA element plays a central role in the control of HIV-1 transcription. This structure nominally consists of a stem-tribulge-stem-hexaloop motif which interacts with the HIV-1 Tat protein and a number of cellular factors. The major question of interest is the extent to which the local conformation and dynamics of the RNA determine its protein recognition functions.
A medium resolution 1H-NMR structure of [[Delta]]TAR, a 29mer fragment derived from the full-length RNA, suggested that the bulge and loop regions may have relatively well-defined conformations. However, detailed structures would prove very difficult to obtain due to the relatively small number of experimental constraints available in these regions. Given this systemic limitation, less direct structural probes such as dynamics measurements and site-directed mutagenesis were employed to obtain a better description of TAR conformational behaviour.
Uniform isotope labelling and 2D 13C-1H NMR methods were used to examine the local atomic dynamics of [[Delta]]TAR. The influence of carbon-carbon coupling in ribose spin systems was minimised by the use of constant-time variants of known pulse sequences. Observed T1, T2 and NOE values were found to vary significantly, covering the ranges 356-531 ms, 21-77 ms, and 0.00-0.62 respectively. Order parameter analyses using three variants of the Lipari-Szabo formalism yielded S2 order parameters covering the range between 0.4 and 1.0, indicating a distribution of residue mobilities in a folded RNA similar to that observed for proteins. No evidence for the rapid sugar repuckering events predicted by molecular dynamics simulations was obtained. Loop residues C30 and A35 exhibited the lowest order parameters, and are least important for binding of TAR by cognate proteins, suggesting that residue entropy may play a more significant role in RNA recognition than previously appreciated.
1H-NMR analysis of a 17mer derived from the naturally-occurring TAR mutant HIVRF was also performed. These results, consistent with the dynamics measurements, suggest that the TAR loop does adopt a compact, relatively rigid structure.