RNA secondary structure of the feline immunodeficiency virus 5'UTR and Gag coding region

Abstract

The 5' untranslated region (5'UTR) of lentiviral genomic RNA is highly structured, and is the site of multiple RNA-RNA and RNA-protein interactions throughout the viral life cycle. The 5'UTR plays a critical role during transcription, translational regulation, genome dimerization, reverse transcription priming and encapsidation. The 5'UTR structures of human lentiviruses have been extensively studied, yet the respective role and conformation of each domain is still controversial. To gain insight into the structure-function relationship of lentiviral 5'UTRs, we modelled the RNA structure of the feline immunodeficiency virus (FIV), a virus that is evolutionarily distant from the primate viruses. Through combined chemical and enzymatic structure probing and a thorough phylogenetic study, we establish a model for the secondary structure of the 5'UTR and Gag coding region. This work highlights properties common to all lentiviruses, like the primer binding site structure and the presence of a stable stem-loop at the 5' extremity. We find that FIV has also evolved specific features, including a long stem loop overlapping the end of the 5'UTR and the beginning of the coding region. In addition, we observed footprints of Gag protein on each side of the initiation codon, this sheds light on the role of the sequences required for encapsidation.

Figure 1.

FIV 5′UTR forms loose dimers in vitro. Three RNA constructs, the 5′UTR, the coding region and the full-length, were renatured and incubated under different conditions: M (Monomer: 0.1 magnesium), D (Dimer: 5 mM magnesium), P (Probing: 10 mM magnesium) or F (Footprint: Probing + 300 nM p55Gag), and then run under native (TBM) or semi denaturing (TBE) conditions (see ‘Materials and methods’ section). For each construct, the fast migrating band represents monomer, while the dimer runs above.

Figure 2.

Chemical and enzymatic probing of the PBS and polyA region in the 5′UTR. (A) Full length RNA was renatured and treated with DMS and CMCT as described in ‘Materials and methods’ section. The probing was conducted in presence or in absence of magnesium. For control 1, RNA was treated exactly like the probed RNA in presence of magnesium, except that the chemical reagent was omitted (elongation of an unmodified RNA with the MMLV RNAse H⁻). For DMS, Gag footprint on the RNA was established by conducting the probing in presence of 300 nM of protein (see ‘Materials and methods’ section). Reverse transcription products were separated on a 6% gel, along with a control and four sequencing lanes. Control 2 is a control for the sequencing reaction, it consist in a reverse transcription of an unmodified RNA with the AMV reverse transcriptase. Some interesting positions are indicated on the side of the autoradiogram. Note that chemical reagents induce a premature reverse transcriptase one nucleotide before the nucleotide hit. Therefore the chemically induced stops run one faster than the corresponding sequence and are therefore indicated with an asterisk. (B) Same that in A, except that RNA was incubated with RNAse V1. Only the condition with magnesium has been performed because divalent ions are required for the nuclease activity. The reverse transcriptase stop induced by RNAse cleavage runs at the same position that its corresponding sequence.

Figure 3.

Secondary structure model of FIV 5′UTR and gag coding region. Colours reflect the accessibility of the nucleotides to the different probes as indicated in the boxed area. ‘Weakly reactive’ refers to reverse transcriptase stops enhanced 2- to 3-fold compared to the untreated RNA, ‘Strongly reactive’ position are enhanced over 4-fold. The star denotes position protected by at least 2-fold in presence of magnesium. (A) Model of the full-length RNA. The polyA signal, the primer binding site and the self complementary sequence in the DIS loop are boxed. (B) Alternative secondary structure model for the region surrounding the initiation codon. The splice site donor sequence and the initiation codon are boxed.

Figure 4.

Chemical and enzymatic probing of the SL1 region—Gag footprint on SL1. For comments see Figure 2. Note that chemical reagents induce a premature reverse transcriptase one nucleotide before the nucleotide hit. Therefore the chemically induced stops run one nucleotide further than the corresponding sequence and are therefore indicated with an asterisk.

Figure 5.

Alternative models compatible with the probing data. (A) Alternative models for the PBS. (B) Alternative model for the SL7/SL8 region. SL7 can be elongated at its base, this results in the disruption of SL8. Legend and colour code are the same as that in Figure 3.

Figure 6.

HIV-1 Gag binds to FIV leader and coding region. Filter binding assays of HIV-1 Gag polyprotein binding to FIV full length (hollow circles), FIV 5′UTR (filled squares) or the coding region (hollow squares). To facilitate comparison, results have been normalized to the value of the maximum that corresponds to a plateau except for the 5′UTR. The maximum bindings observed were 57% for the full length, 62% for the coding region and 40% for the 5′UTR. Error bars represent the SD from the mean from the data of three independent experiments.

Figure 7.

Chemical of SL13 in the coding region—Gag footprint on SL13. For comments see Figure 2.

Figure 8.

Chemical and enzymatic probing of SL11 in the coding region—Gag footprint on SL11. For comments see Figure 2. Note that chemical reagents induce a premature reverse transcriptase one nucleotide before the nucleotide hit. Therefore the chemically induced stops run one nucleotide further than the corresponding sequence and are therefore indicated with an asterisk.

Figure 9.

Footprint of the Gag protein on the 5′UTR and gag coding region. Colours reflect the modification of the accessibility of each nucleotide induced by Gag binding. Protection or enhancement was at least 2-fold.