|
Figure 1. Reconstruction of the export of the singly spliced transcript in Xenopus oocytes. (A) Gene expression of HIV-1 is regulated by RNA export. (B) Schematic representation of pre-ftzRRE RNA used for the export analysis. (C) 32P-labeled pre-ftzRRE RNA was microinjected into the nucleus of Xenopus oocytes together with 32P-labeled U1ftz (elongated U1), U6RRE and U6Δss, in either the absence (lanes 1–8) or presence (lanes 9–14) of the purified recombinant Rev protein (160 fmol/oocyte), with either CTE (50 fmol/oocyte; lanes 5, 6, 11 and 12) or the CTE mutant M36, which does not bind TAP/NXF1 (CTEmut, 50 fmol/oocyte; lanes 7, 8, 13 and 14), or without the inhibitor (lanes 1–4, 9 and 10). RNA was extracted from nuclear (N) and cytoplasmic (C) fractions, immediately (0 h; lanes 1 and 2) or 1 h (1 h; lanes 3–14) after the injection, and were analyzed using 6% denaturing PAGE. (D) The same experiments as in (C) were performed in the absence (lanes 1–6) or presence (lanes 7–10) of Rev, except with BSA-NES (190 ng/oocyte; lanes 3, 4, 7 and 8) or BSA-mut (M10) NES, which does not bind CRM1 (190 ng/oocyte; lanes 5, 6, 9 and 10), or without the inhibitor (lanes 1 and 2). RNA export was analyzed immediately (0 h; lanes 1 and 2) or 1 h (1 h; lanes 3–10) after the injection. (E) Quantitation of the export of spliced ftzRRE RNA from three independent experiments performed as in (C) and (D). Averages and standard deviations with CTE (gray bars) or BSA-NES (black bars), or without inhibitors (none; white bars) in the absence (-) or presence (+Rev) of Rev are shown.
|
|
Figure 2. Inhibition of TAP-dependent RNA export by Rev. (A) The same 32P-labeled RNA mixture as in Figure 1 together with 32P-labeled m7G-capped U5ΔSm RNA was microinjected into the nucleus of oocytes. The effect of the wild-type Rev or Rev M10 mutant protein (160 fmol/oocyte) was examined as in Figure 1. (B) Quantitation of the export of spliced ftzRRE, elongated U1 and U5 RNAs from three independent experiments performed as in (A). Averages and standard deviations with M10 (black bars) or without proteins (buffer; white bars) are shown. (C and D) The same 32P-labeled RNA mixture as in (A), except that either pre-CDCRRE (C) or pre-betaRRE (D) was used instead of pre-ftzRRE, was microinjected into the nucleus. The effect of the wild-type Rev or Rev M10 mutant protein (160 fmol/oocyte) was examined as in (A). See also Supplementary Figure S1 for quantitation.
|
|
Figure 3. Effect of Rev on the association of mRNA binding proteins. (A) The same 32P-labeled RNA mixture as in Figure 1 was injected into the nucleus in the absence or presence of Rev. The nuclear fraction was prepared after 1 h, and GST pull-down was performed with glutathione beads that had been pre-bound with either the GST-TAP231 or GST protein. RNA precipitated with each type of bead was recovered and analyzed. The input lanes were loaded with 10% of each input mixture. (B) The nuclear fraction was prepared as in (A), and IP was performed with the anti-Aly/REF monoclonal antibody (11G5, αAly), anti-Y14 monoclonal antibody (4C4, αY14) or anti-Myc monoclonal antibody (9E10, αMyc) that had been pre-bound to Protein A-Sepharose beads. RNA precipitated with each antibody was recovered and analyzed. (C) The recombinant FLAG-UAP56 protein (50 fmol/oocyte) was pre-injected into the cytoplasm. After 16 h incubation, a second microinjection was performed into the nucleus with the same 32P-labeled RNA mixture as in Figure 1, except that pre-ftz RNA was used instead of elongated U1 RNA, in the absence or presence of Rev. IP was performed with 11G5, the anti-FLAG monoclonal antibody (M2, αFLAG), or 9E10. (D) Rev inhibits the association of TAP and the TREX complex with spliced ftzRRE RNA.
|
|
Figure 4. Effect of Rev on EJC formation. 32P-labeled pre-ftzRRE RNA (left) or intronless ftzRRE RNA (right) was microinjected into the nucleus in either the absence (A) or presence (B) of Rev. The nuclear fraction was prepared after 1 h, and RNase H digestion was performed with the antisense oligo DNAs A–D. RNA digestion was analyzed using 6% denaturing PAGE.
|
|
Figure 5. Effect of Rev on the export of intronless RNA containing RRE. (A and B) The same experiments as in Figure 1C and D, respectively, were performed, except that intronless ftzRRE RNA was used instead of pre-ftzRRE RNA and incubation was performed for 1.5 h. (C) Quantitation of the export of ftzRRE RNA in (A) and (B). (D) IP was performed as in Figure 3B, except that intronless ftzRRE RNA was used instead of pre-ftzRRE RNA. (E) Rev inhibits the association of TAP and the TREX complex with intronless ftzRRE RNA.
|
|
Figure 6. Effect of TAP-p15 overexpression on HIV-1 expression. (A) Genome organization of HIV-1 NL4-3ΔRev. The start codon of the Rev gene was mutated from ATG to ACG. (B) HEK293T cells in a 6-well plate (70% confluent) were transfected with pcDNA3-GFP (1 μg), pNL4-3ΔRev (1 μg), and either pCI-neo (0.1 μg) (−) or pCI-FLAG-Rev (0.1 μg) (+). After 24 h, cells were collected and proteins from cell pellets were analyzed by SDS-PAGE and western blotting with rabbit anti-Gagp55 antiserum. UT: untransfected cells were used as a control. (C) HEK293T cells in a 6-well plate (70% confluent) were transfected with pcDNA3-GFP (1 μg), pNL4-3ΔRev (1 μg), pCI-FLAG-Rev (0.1 or 1 μg), and either pcDNA5 (1 μg) (−) or pcDNA5-FLAG-TAP (0.6 μg) plus pcDNA5-FLAG-p15 (0.3 μg) (+). 0.9 μg of pCI-neo was added for 0.1 μg of pCI-FLAG-Rev to equalize the amount of plasmid DNAs. Supernatants and cells were collected after 24 h. RNA from cell pellets was subjected to semi-quantitative RT-PCR. PCR products were analyzed by electrophoresis in a 2% agarose gel. (D) Quantitation of the relative level of RNAs from three independent experiments performed as in (C). GFP mRNA was used for normalization. 0.1 μg of the pCI-FLAG-Rev sample (lane 1) was set to 1. Averages and standard deviations are shown. (E) Protein from the cell pellets in (C) was analyzed by SDS-PAGE and western blotting with rabbit anti-Gagp55 antiserum or the anti-GFP antibody. UT: untransfected cells were used as a control. U2AF65 was a loading control. (F) The filtrated media from (C) were immunoprecipitated with rabbit anti-Gagp55 antiserum and detected by western blotting with the monoclonal anti-Gagp24 antibody.
|
|
Figure 7. Molecular mechanism of remodeling of export RNPs by Rev. (A) 32P-labeled U6Δss RNA together with either pre-ftzRRE RNA (RRE) or pre-ftzERR RNA (ERR) was microinjected into the nucleus in either the absence or presence of Rev. The nuclear fraction was prepared after 1 h (lanes 1–4), and RNase H digestion was performed with the antisense oligo DNA E (lanes 5–8). The reaction mixture was immunoprecipitated with the anti-T7 antibody (lanes 9–12). RNAs were analyzed using 6% denaturing PAGE. (B) Purified recombinant CBP80 (3 μg) was pulled down by GST, GST-T7-Rev, GST-Aly or GST-hnRNP A1 (GST-A1) (1 μg each) in the presence of RNase A (1 mg/ml). Pulled down CBP80 was separated by SDS-PAGE and detected by western blotting. (C) Recombinant CBP80 (3 μg) was pulled down by GST or GST-Aly (1 μg each) in the presence of RNase A (1 mg/ml) and in the absence or presence of T7-Rev (0.2, 1, 5 or 25 μg), and was analyzed as in (B). (D) 32P-labeled U6Δss, m7G-capped 350 nt-ERR RNA and m7G-capped 260 nt-RRE RNA were incubated with or without recombinant T7-Rev (30 nM) or CBC (30, 100, 300 nM). The reaction mixture was immunoprecipitated with the anti-T7 antibody. Precipitated RNAs were analyzed by 8% denaturing PAGE.
|
|
Figure 8. A model of how the HIV-1 Rev protein remodels viral export RNPs. Not all RNA binding proteins are shown for the sake of clarity. See the Discussion section for details.
|