Development of a 3D Structural Model for the HIV-1 5'UTR RNA

Nikolai Ulyanov, University of California, San Francisco
Basic Biomedical Sciences
Innovative, Developmental, Exploratory Award (IDEA)

A project is proposed to develop a working low-resolution model for the three-dimensional structure of 5-prime-untranslated region (5-prime-UTR) of the HIV-1 genomic RNA. 5-prime-UTR is a part of viral RNA that does not code for proteins, but is critical for many steps of the viral life cycle. This region is very strongly conserved among HIV-1 strains, and it is highly structured. Many structural elements and their functional roles have been characterized in 5-prime-UTR. However, our understanding of a spatial relationship between these elements is essentially one-dimensional or two-dimensional at best; it is based only on the proximity of secondary structure elements in the primary structure (sequence). A truly three-dimensional model of the 5-prime-UTR structure can advance our understanding of complex interactions taking place in this region, help refine existing and define new targets for anti-HIV therapeutic strategies. The project will be mostly computational in its nature; the miniCarlo software will be used to calculate the model. A specific hypothesis will be tested that the existing computational techniques can be used to successfully predict de novo the three-dimensional fold of the multi-domain 5-prime-UTR RNA of HIV-1. The success of the prediction will be judged on agreement with experimental structural information derived from the fluorescence resonant energy transfer (FRET) and small angle x-ray scattering (SAXS) data acquired for the 5-prime-UTR transcript in vitro. FRET experiments register exchange of energy between fluorescent probes attached to specific sites on 5-prime-UTR, which can be used to estimate the distances between these sites; experimental distances can be directly compared to the distances predicted by the model. SAXS data are sensitive to the overall shape of the molecule. The scattering curve can be calculated from the predicted model and compared with the experimental curve. If the hypothesis is proven incorrect, we will fall back on refining the model using FRET data as experimental restraints. A long-term goal is to refine this model based on acquired low-resolution structural data (SAXS and extended FRET measurements) and to extend the modeled system to study the dimerization of the 5-prime-UTR (including the primary and a possible alternative dimerization sites), mechanism of the dimer maturation, possible effects of downstream RNA sequences on the 5-prime-UTR structure, and effects of various proteins known to interact with 5-prime-UTR (starting with nucleocapsid protein, tat, and human cyclin T1).