Diagrammatic representation of the CAT-RRE reporter system (Phillips et al., 1992). (A) In the pRRE(+) construct, splice acceptor (SA) and splice donor (SD) sites flank a CAT gene and a FIV-Ch RRE which is in the sense orientation. In the pRRE(−) construct, FIV-Ch RRE was inserted in an antisense orientation. (B) The cloned rev plasmid and FIV-Ch plasmid were transfected into CrFK cells, respectively. Cell lysate was harvested at 24 hours and analyzed. The western blot confirms the expression of the cloned rev gene using expression of Rev encoded by FIV-Ch virus as a control.

Mutational analysis of SA1 and SA2 within the RRE of the full-length FIV-Ch. CrFK cells in 24 well plates were transfected with full length FIV-Ch molecular clones (1.25ug/well) containing RRE mutations in SA1 and SA2. The cell supernatants and proteins in the cell lysates were harvested 24 hours post-transfection. The released virus was assessed by RT activity assay in the cell supernatants and expression of the viral structural protein p24 was assessed by western blot assay. Western blot membranes were probed with a monoclonal anti-p24 antibody and polyclonal rabbit anti-Rev antibody. β-actin was used as a loading control. The values are obtained from triplicate experiments that were normalized to wild-type FIV-Ch levels (set at 100%). Immunoreactivity (IR) represents p24 expression measured on western blot. (A) Depiction of tested RRE mutants with modifications in SA1 indicated. Without changing the amino acid (aa) sequence of Env, SA1/54, SA1/55 mutants tested the nucleotide identity and/or structural importance of an “A A” internal loop, and SA1/56, SA1/57 mutants test the importance of a “G C” base pair. Triple base pair compensatory mutations test importance of SA1, in which two aa of the Env were concurrently changed by nucleotide modification. (B) Depiction of RRE mutations within SA2. Single base pair and triple base pair compensatory mutations, without and with changes to the amino acid sequence of Env, respectively, were designed to test the importance of SA2 for FIV replication. Error bars show standard deviations.

Mutational analysis of SA5 and SA6 within the RRE of the full-length FIV-Ch. (A) Depiction of RRE mutations within SA5. Triple base pair compensatory mutational analysis, without changing the amino acid sequence of Env, was designed to test the importance of SA5 for FIV replication. (B) Depiction of RRE mutations within SA6. Compensatory mutational analyses of two “A U” base pairs within SA6 were performed without changing the amino acid sequence of Env, which test the importance of SA6 for FIV replication. In addition, SA6 and SA5 deletion were also carried out to test importance of the two top stem-loops.

Secondary structure of RRE affects FIV replication. CRFK-transfection generated wild-type FIV-Ch and FIV-ST3/45 viruses were normalized with RPMI medium (15% FBS, 1% pen/strep) to the same RT activity level (~9000 cpm) and subsequently used to infect a feline T-lymphoblastoid cell line (MYA-1) (800 μl virus solution + 200 μl of ~2×10⁶ cells per well) in 24 well plates for a time course experiment. At 6 hours post-infection, supernatants were removed and 1 ml/well of fresh RPMI medium was added. On day 2 post-infection, 170μl/well of MYA-1 cell supernatant was harvested for the RT activity assay and 200μl/well of fresh medium was added. Sampling was repeated daily. A mock control was performed in parallel. At days 4, 8 and 12, total RNA and proteins were extracted from MYA-1 cells for viral mRNA semi-quantitative real time RT-PCR and p24 western blot assay, respectively. (A-C) show viral env, gag, and pol transcript levels, respectively. (D) Viral p24 expression corresponding to FIV-Ch and FIV-SA3/45 on day 4, 8 and 12 is shown. (E) Likewise, RT activity of FIV-Ch and FIV-SA3/45 strains are displayed for the entire 12 day time course with marked differences between viruses.

FIV-Ch genome and computer-predicted secondary structure of FIV-Ch RRE. (A) Schematic representation of genome organization and relative positions of open reading frames (ORF) and RRE RNA of FIV-Ch. The FIV-Ch genome is represented linearly with ORF regions depicted as boxes; the rev gene is depicted as a dotted box, the LTR is depicted by a solid thick line, the RRE RNA depicted as empty box. (B) The computer -predicted secondary structure of RRE contains two hairpins: stem-loop A (SLA) and stem-loop B (SLB). The end of the env ORF is indicated by an arrow. Nucleotides are numbered by their position within the full-length proviral (FIV-Ch) plasmid. (C) Secondary structure of RRE SLA in this study. The stop codon “UGA” of the env gene present in LA4 is denoted in gray, italic lettering.

Detection of secondary structure of SLA in the CAT-RRE reporter system. Each stem of SLA within the FIV-Ch RRE in the CAT-RRE reporter system was modified with a compensatory mutagenesis and/or deletion strategy for functional assessment of the secondary structure of the RRE. The CAT-RRE plasmids were co-transfected with the FIV-Ch proviral plasmid (FIV-Ch) or the cloned FIV rev plasmid (Rev). Expression concentrations of the CAT reporter protein were analyzed by ELISA. For quantification of CAT expression, the concentration of wild-type FIV-Ch CAT was set at 100% and mutant concentration were normalized to this value for (A) SA6, (B) SA5, (C) SA4, (D) SA3, (E) SA2 and (F) SA1. Relative CAT protein expression levels are presented as histograms. The values represent the means of results from duplicate transfections across two separate experiments.

Mutational analysis of SA3 and SA4 within the RRE of the full-length FIV-Ch. (A) Depiction of RRE mutations within SA3. Single base pair and triple base pair compensatory mutations, without and with changes to the amino acid sequence of Env, respectively, were designed to test the importance of SA3 for FIV replication. (B) Depiction of RRE mutations within SA4 as in panel A. In addition, the structural importance of an “A” bulge was tested by inserting a “U” on its opposite side in the predicted RRE secondary structure to create an “A U” base pair and consequently remove the “A” bulge.

Viral RNA structure-based mutagenesis to evaluate a putative FIV encoded antisense protein (ASP). Compensatory mutational analysis was performed on a single “C G” base pair within SA3 of RRE which represents the 3^rd nucleotide of the putative ASP start codon. This mutation was generated such that it did not alter the predicted amino acid sequence of the overlapping env. (A) A schematic figure shows the positions of the env gene, RRE RNA and the putative ASP ORF. The putative ASP was predicted to be a 103 aa hydrophobic protein (Briquet et al., 2001). (B) Analysis of RRE mutants with modifications in SA3 within the full length FIV-Ch virus context as in Figs. 4, 5, and 6. A series of mutants were generated which tested the effect of disruption of base pairing within SA3 and restoration of base pairing by compensatory mutations, concurrently with or without a silenced ASP start codon. “+” and “−” represent activated and silenced start codon of ASP, respectively.