Interactions of Retroviral Proteins with Nucleic Acids

Division Of Basic Sciences - Nci

Investigators

Linked publications, trials & patents

Abstract

There appear to be several different modes of interaction between retroviral proteins and nucleic acids, each with important functional consequences for viral replication. First, an exquisitely specific recognition by the Gag polyprotein (the structural protein of the virus particle) selects the viral RNA for packaging during virus assembly. The interactions of retroviral Gag proteins with RNAs are remarkably complex. We believe that one role played by the viral RNA is to bring Gag proteins close together, as discussed in our project on Retrovirus Assembly and Maturation (ZIA BC 010511). At the same time, successful replication requires that viral genomic RNA (vRNA), which contains a packaging signal (termed "Psi"), be selected for encapsidation in preference to the thousands of cellular RNA species that are also potential substrates for packaging. Finally, both HIV-1 Gag and its cleavage product nucleocapsid (NC) possess nucleic acid chaperone activity, enabling them to catalyze rearrangements of nucleic acids to the conformation with the lowest free energy. It appears that this activity in Gag is responsible for annealing a tRNA molecule to vRNA, and the chaperone activity of NC plays several critical roles during reverse transcription. Our research is devoted to elucidating these diverse interactions at the molecular level. To better understand the relationship between HIV-1 RNA packaging and assembly, we are addressing the following questions: How does HIV-1 Gag discriminate between different RNAs, and how can we explain the selective packaging of Psi-containing RNA during particle assembly? _____ We have found that when Gag is expressed in vivo in the absence of vRNA, it can package almost any cellular mRNA. Similarly, addition of almost any nucleic acid to HIV-1 Gag will lead to virus-like particle (VLP) assembly in vitro. Thus, vRNA is in competition with mRNAs for packaging; the Psi packaging signal gives it an advantage in this competition. We are studying the binding of recombinant Gag to Psi-containing and control RNAs by using fluorescence correlation spectroscopy, microscale thermophoresis, SwitchSense instrumentation, and mass photometry. We find that Gag binds RNAs cooperatively. The affinity of Gag for Psi is only modestly higher than that for control RNA. Using several Gag mutants, we have found that binding to the control is largely attributable to the matrix domain. Notably, binding to Psi has a different character, as it is far more salt resistant than binding to control RNA, indicating a higher nonelectrostatic component. We are testing the idea that binding to Psi triggers the conformational shift that Gag undergoes to an assembly-ready state more rapidly or more efficiently than other RNAs. We have found that binding to Psi supports particle assembly more efficiently than binding to other RNAs. This would explain selective packaging of vRNA. The data also identify specific nucleotides within Psi that are required for this efficient particle assembly. We have extended these studies by a detailed analysis of the kinetics of binding of HIV-1 Gag protein to 145-base RNAs representing the HIV-1 packaging signal ("Psi") and selected mutants of Psi. These experiments were made possible by the SwitchSense instrument, which monitors the association and dissociation of a protein in solution to immobilized RNAs. One particularly useful feature of this instrument is that binding to two different RNAs can be monitored simultaneously in the same flow cell. The dimerization initiation site ("DIS") palindromic sequence in our RNAs was mutated to prevent their dimerization in these assays. We find that even though the affinity of Gag for different RNAs is similar at physiological ionic strength, it binds more rapidly to Psi than to scrambled RNAs (with the same size and base composition as the Psi RNA, but different sequence) or to mutants of Psi in which unpaired G residues, known to be important in the binding, are replaced. This differential binding was particularly prominent at high Gag concentrations, suggesting a degree of cooperativity in the binding. Psi consists of three stem-loop structures. It was of interest to determine whether the rapid binding to Psi could be localized to one or more of these stem-loops. To test this possibility, we also measured kinetics of binding to the stem-loop pairs SL1-SL2 and SL2-SL3. These experiments showed that Gag binds more rapidly to SL2-SL3 than to SL1-SL2 or controls. However, the Gag bound to individual stem-loops with indistinguishable kinetics. It thus appears that the rapid binding to Psi is due to its interaction with the structure or sequence of SL2-SL3, which acts as an integral unit in this interaction. Determination of the 3-dimensional structure of an RNA molecule is a challenging undertaking. One of the reasons this is so difficult is that structured RNAs are, in general, contain highly flexible linker regions connecting double helices. Because of the flexibility of these linkers, individual molecules in a solution will have different 3-dimensional structures even if they have the same nucleotide sequence. Many approaches to structure determination, such as cryo-electron microscopy, depend on averaging the signals from large numbers of molecules, and are therefore confounded by heterogeneity in the population being analyzed. We have participated, together with our colleague Yun-Xing Wang (SBL) in the development of procedures for the use of atomic force microscopy for RNA structural analysis; as this technique obtains information from individual molecules, rather than pooling information from many molecules, it can determine structures of RNAs despite variation in the population of molecules under study. Patents linked to this project: U.S. Patent #5,674,720: "Design and Construction of Noninfectious Human Retroviral Mutants Deficient in Genomic RNA"; issued October 7, 1997; Robert J. Gorelick, Larry O. Arthur, Alan Rein, Louis E. Henderson, and Stephen Oroszlan. This patent describes mutants of HIV-1 that are structurally normal but noninfectious; these mutants could potentially be considered as vaccine constituents. U.S. Patent #7,572,828: "Identification of Anti-HIV Compounds Inhibiting Virus Assembly and Binding of Nucleocapsid Protein to Nucleic Acid"; issued August 11, 2009; Robert Shoemaker, Michael Currens, Alan Rein, Ya-Xiong Feng, Robert Fisher, Andrew Stephen, Shizuko Sei, Bruce Crise, Louis Henderson, and Karen Worthy. This patent describes a class of compounds with anti-HIV-1 activity, which are under investigation for use in antiretroviral therapy.