Figure 1.
Alignment of patient-derived RRE variants used for functional analysis.
Multiple nucleotide alignment of the full-length patient-derived RRE variants used for prediction of secondary structures and functional analyses. Several full-length RREs derived from two patients (patient 10: P10; and patient 5: P5) were amplified from plasma samples, cloned and sequenced. Boxes highlight the nucleotides coding ENF resistance mutations G36V/D, V38A and N43D for patient 10; and Q40H and L45M for patient 5. The characteristic five stem-loop regions of the RRE are identified on top of the sequence. Variants were designated according to the patient number, amino acid change in gp41 and clone number. The shaded areas indicate the reported high-affinity binding site of Rev located in the stem II. The NL4-3 RRE sequence is included for comparison.
Figure 2.
Predicted secondary structures of patient-derived RRE variants.
The prediction of the secondary structure of our RRE sequences was generated including the 329-nucleotide sequence of the patient-derived RREs in the RNA Fold Web server. The five well-defined stem-loop structures, including the branched stem-loop IIB that is critical for the binding of the Rev protein, were identified for most of the sequences. A) Predicted secondary structure of RRE from patient 10. Predicted secondary structure of the complete RRE of a BL clone from patient 10 is shown. The nucleotides encoding the amino acid changes associated with ENF resistance are underlined and a dotted box encloses the high-affinity Rev binding site. Stems loops II-III-IV and V for representative samples for each substitution present in Patent 10 are shown. Nucleotide changes present in the RRE variants with regard to the BL clone are marked with filled red circles. B) Representative RRE structures obtained from samples of patient 5.
Figure 3.
Binding and multimerization of selected RRE variants using electro-mobility shift assay (EMSA).
32P-labeled RRE RNA was synthesized in vitro and a constant amount was incubated in a binding reaction with increasing amounts of Rev protein (20 ng, 40 ng and 80 ng) or without Rev The resulting complexes were analyzed by electrophoresis in a native polyacrylamide gel. The gel was dried and examined using a phosphorimager. Binding pattern observed after incubation of RRE variants is shown in A) from patient 10, and B) from patient 5.
Figure 4.
Predicted secondary structures of RRE variants constructed by site-directed mutagenesis.
RRE variants with nucleotide substitutions at position 40 sRRE40 (Q40H), at position 45 sRRE45 (L45M) or without any, WT (40Q-45L), were generated by site-directed mutagenesis using as a template a RRE40-45 clone (Q40H-L45M). The sequences of these constructed clones were subjected to RNA fold analyses as described in Figure 2. Nucleotides that were substituted are highlighted in the filled red circles.
Figure 5.
In vitro Rev-RRE binding assay.
Electro-mobility shift assay (EMSA) in the absence or in the presence of 20 ng, 40 ng or 80 ng of Rev protein with the different RRE RNA variants (RREWT, sRRE40, sRRE45 and RRE40-45). The reaction products were separated in a polyacrylamide gel and the quantification of the gel shifts are displayed.
Figure 6.
293T cells were co-transfected with the constructed RRE variants, pDM628 or pDM628ΔRRE with or without pCMV-Rev. The export levels of the different variants were quantified by luminescence and corrected by the background signal from the luminometer noise or due to Rev-independent transport. This correction was performed to each sample by subtracting the luminescence that was measured when the cells were transfected without Rev (replaced with pcDNA 3.0). And finally, corrected luminescence values were calculated as the fold-change increase, which was performed by dividing the corrected luminescence of each plasmid by the corrected luminescence of the pDM628 plasmid. A) Rev-RRE mediated export from RNA variants containing changes at positions 36, 38 and 43, evaluated in the presence of Rev (ratio 1:5, Rev:RRE). B) Cytoplasmic export of RRE variants with changes at positions 40 and 45. Rev-dependent transport of the pDM628-based RRE variants: WT (40Q-45L), sRRE40 (Q40H), sRRE45 (L45M) and the double mutant RRE40-45 (Q40H-L45M); in the presence of three different concentrations of pCMV-Rev (200 ng, 20 ng or 2 ng per well). Data represent the mean +/− SEM of 3 independent experiments performed in triplicate transfections.
Figure 7.
Functional analyses of RRE with changes at positions 40 and 45 in lymphoid cells.
The lymphoid cell line MT-4 was co-electroporated with the NL4-3 hemigenomic plasmids p83-2 and with the p83-10 constructed variants, which only differed in their RRE region. A) Fold increase of cytosolic unspliced RNA in infected cells. Levels of unspliced RNA in the cytoplasmic fractions were determined by qRT-PCR. The fold change was calculated by the relative quantitation method 2(−ΔΔCt). GAPDH was used for normalization and RREWT (40Q-45L) as a calibrator. Data (mean +/− SEM) is derived from three independent experiments, with triplicate samples in each PCR. B) Quantification of the transcriptional activity per infected cell. The ratio between the levels of unspliced HIV RNA and total HIV DNA (2(−ΔΔCt) Unspliced HIV-1 RNA/2(−ΔΔCt) Total HIV-1 DNA) was calculated. Total HIV DNA content was determined in cells by qPCR using the same approach as described with the unspliced RNA levels. Data (mean +/− SEM) is derived from three independent experiments, with triplicate samples in each PCR. C) P24 protein production in cell-free supernatants. The p24 present in the supernatant of the cultures was quantified by ELISA from the same time-point that the RNA and DNA levels were determined. Data (mean +/− SEM) is derived from two independent experiments. D) Normalized p24 protein production. Raw p24 values were normalized to copies of total HIV-1 DNA to correct for differences in electroporation efficiency and gp41 function. Data (mean +/− SEM) is derived from two independent experiments.