Specific Guanosines in the HIV-2 Leader RNA are Essential for Efficient Viral Genome Packaging

Abstract

HIV-2, a human pathogen that causes acquired immunodeficiency syndrome, is distinct from the more prevalent HIV-1 in several features including its evolutionary history and certain aspects of viral replication. Like other retroviruses, HIV-2 packages two copies of full-length viral RNA during virus assembly and efficient genome encapsidation is mediated by the viral protein Gag. We sought to define cis-acting elements in the HIV-2 genome that are important for the encapsidation of full-length RNA into viral particles. Based on previous studies of murine leukemia virus and HIV-1, we hypothesized that unpaired guanosines in the 5' untranslated region (UTR) play an important role in Gag:RNA interactions leading to genome packaging. To test our hypothesis, we targeted 18 guanosines located in 9 sites within the HIV-2 5' UTR and performed substitution analyses. We found that mutating as few as three guanosines significantly reduce RNA packaging efficiency. However, not all guanosines examined have the same effect; instead, a hierarchical order exists wherein a primary site, a secondary site, and three tertiary sites are identified. Additionally, there are functional overlaps in these sites and mutations of more than one site can act synergistically to cause genome packaging defects. These studies demonstrate the importance of specific guanosines in HIV-2 5'UTR in mediating genome packaging. Our results also demonstrate an interchangeable and hierarchical nature of guanosine-containing sites, which was not previously established, thereby revealing key insights into the replication mechanisms of HIV-2.

Keywords: RNA; encapsidation; guanosine; retroviruses; untranslated region.

Fig. 1.

Approach used to examine the roles of unpaired guanosines in the HIV-2 5’UTR on RNA packaging. A. General structures of constructs and experimental protocol used to study RNA packaging. BSL, stem-loop sequences recognized by bacterial protein BglG; asterisk denotes inactivating mutation; stop with arrow indicates introduced stop codon. Poly A, polyadenylation signal; pro, promoter; LTR, long terminal repeat; RRE, Rev response element; p2A, p2A self-cleaving peptide. Asterisks indicate stop codons. Representative images of viral particles captured using fluorescence microscopy are shown. Merge and Shift, CeFP and YFP signals were overlaid and YFP signals were shifted to the right by 3 pixels to visualize colocalization. B. Schematic of HIV-2 leader RNA structure as described by Purzycka et al. 2011 (15). For simplicity, only the nucleotides within the single-stranded sites targeted for mutations are shown; red uppercase Gs indicate specific guanosines mutated; blue numbers (#1 to #9), denote sites of mutations; yellow box denotes position of Gag translation start. Site 3 is embedded within a 10-nucleotide palindromic sequence (pal). C. Sites of mutations in the context of the HIV-2 leader DNA sequence used in this study. Red uppercase G, guanosines targeted for mutation; blue numbers, individual sites. Indicated guanosines were mutated to adenosines, except in the case of site #2 in which the AGTAAG sequence was mutated to ACTAAA to avoid introducing a major polyadenylation signal and site #3 in which the GGAGTG sequence was mutated to AAATTG to avoid introducing an ATG upstream of the gag open reading frame. For reference, the PBS sequence is shown in bold and underlined, and the Gag ATG translation start codon is shown in bold uppercase with yellow highlight.

Fig. 2.

Effects of mutating putative unpaired guanosines in the HIV-2 5’UTR on RNA packaging and viral particle production. A. Proportions of viral particles containing HIV-2 RNA genome. Particles derived from HIV-2 constructs containing unaltered 5’UTR (WT) or mutated 5’ UTR in which guanosines in 9 target sites were mutated (M1-9). B. Particle production of WT and M1-9 constructs. WT particle production is set to be 100%; equal amounts of WT or M1-9 DNA were used in each transfection. Results shown are averages of three independent experiments; error bars indicate standard deviations and open circles indicate values obtained from each independent experiment. P-values were calculated using the unpaired t-test; n.s., not significant; ****, p-value < 0.0001.

Fig. 3.

Identifying sites in M1-9 that affect HIV-2 genome packaging. A. Effects of mutations in the first four (M1234) or last five (M56789) sites on HIV-2 RNA packaging. B. Effects of mutating sites 1,2, and 4 (M124) or site 3 (M3) on HIV-2 RNA packaging. Results shown are averages of at least three independent experiments. Error bars, standard deviations; open circles indicate values obtained from each independent experiment. P-values were calculated by one-way ANOVA with Bonferroni correction for multiple pairwise comparisons; n.s., not significant; ****, p-value < 0.0001.

Fig. 4.

Delineating the minimal number of mutations required to recapitulate the M1-9 phenotype. A. Effects of mutating site 3 in combination with another site on genome packaging efficiency. The RNA packaging efficiencies of double mutants were compared to that of M3. B. Effects of mutating sites 2 and 3 in combination with select single sites on RNA packaging. Statistical comparisons to M1-9 are shown for the mutants. All results are averages of at least three independent experiments; error bars, standard deviations, with open circles indicating values obtained from each individual experiment. P-values for all experiments were calculated by one-way ANOVA with Bonferroni correction for multiple pairwise comparisons: n.s., not significant; **, p-value < 0.01; ***, p-value < 0.005; ****, p-value < 0.0001.

Fig. 5.

Effects of sites 5, 7, and 8 mutations on genome packaging efficiency. A. Effect of combining sites 2 or 3 with sites 5,7, and 8 mutations on RNA packaging. B. Effects of restoring sites 5, 7, or 8 in M2578 and M3578 on RNA packaging. The same data sets on WT, M2578, and M3578 are shown in both A and B panels. All results are averages of at least three independent experiments; error bars, standard deviations, with open circles indicating values obtained from each individual experiment. P-values were calculated by one-way ANOVA with Bonferroni correction for multiple pairwise comparisons: n.s., not significant; *, p-value < 0.05; **, p-value < 0.01; ****, p-value < 0.0001.

Fig. 6.

Delineating the effects of guanosine mutations on HIV-2 RNA packaging. X-axis, number of guanosines mutated; y-axis, RNA packaging efficiency. Blue dots, mutants containing site 3 mutations; yellow dots, mutants without site 3 mutation but containing site 2 mutations; red dots, wild-type and a mutant (M56789) without sites 2 or 3 mutations.