Figure 1.
The role of splicing and Rev-dependent export on miRNA silencing.
(A) A schematic of the SV40 promoter, intron and 3′ UTR from which the Renilla luciferase (Rluc) was transcribed in the psiCHECK-2 vector (psiCHECK). “SD” and “SA” indicate the splice donor and splice acceptor sites, respectively. The 3′ UTR has three restriction sites (XhoΙ, PmeΙ and NotΙ). Each let-7 targeting sequence was inserted into the PmeΙ site, and the “SD” or Rev response element (RRE) was inserted into the XhoΙ or NotΙ sites in various combinations. (B) The silencing of the RNA and mutated sequences cloned into the PmeΙ restriction site in the Rluc 3′ UTR in the presence of Rev-HA was assessed. The Rluc activity was normalized to the firefly luciferase activity, and an empty vector C was used as a control. (C) The miRNA-mediated silencing of the spliced Rev-HA-exported RNAs was assessed. “RRE” and “SD” were inserted into the restriction sites (XhoΙ or NotΙ) in the Rluc 3′ UTR. The red arrow points to the vectors that presented altered Rluc activity in the presence of Rev-HA. The red arrowhead points to the Bulge-, 3×Bulge- or Perfect-containing constructs that carry a correctly oriented RRE and were silenced in the presence of Rev-HA. (D) A schematic of the SV40 promoter and intron region and the truncated promoters without an intron. (E) The effects of splicing and the presence of enhancers on miRNA-mediated silencing were analyzed using plasmids containing the SV40 promoter or its truncated versions. The blue arrow points to the Bulge-containing constructs without an intron that carry a correctly oriented RRE and were not silenced in the presence of Rev-HA. For each promoter tested, the empty vector C was used as a control. The Renilla/firefly luciferase value was assessed, and six independent experiments were performed and expressed as the mean ± S.D. as a percentage of the control. ***P<0.001.
Figure 2.
Analysis of Rev-dependent export on RNAs with mutations in splice donor site.
(A) A schematic of the psiCHECK-2 vector (psiCHECK) in which the locations of primers to amplify the intron region or exon region of the Rluc are illustrated. The splice donor and the juxtaposed exon sequences (SD) are shown. The sequence mutated from the psiCHECK vector is shown in red. The p5SD vector was exchanged with the fifth SD (6,722–6,730) of pNL4-3. The pmSD has mutations in the highly conserved SD sequences. (B) The nuclear and cytoplasmic level of G3PDH RNA in HeLa cells was analyzed by RT-qPCR. The cytoplasmic level was set to 100. (C) The nuclear and cytoplasmic level of U1 snRNA. The nuclear level was set to 100. (D) The nuclear and cytoplasmic level of firefly luciferase RNA in which the cytoplasmic level was set to 100. (E) The levels of Rluc RNAs transported into the cytoplasm were analyzed by RT-qPCR. The intron region and exon region were amplified using the primers shown in (A). The RNA levels amplified by primers in the exon region were set to 100. The gray arrowhead and arrow point to the intron-containing Rluc RNA levels in each transfected cell. (F) The let-7-mediated silencing of Rev-HA-exported RNAs from p5SD. The red arrow points to the vectors that presented altered Rluc activity in the presence of Rev-HA. The blue arrow points to the Bulge-containing constructs that carry a correctly oriented RRE and were not silenced in the presence of Rev-HA. (G) The let-7-mediated silencing of Rev-HA-exported RNAs from pmSD. The Renilla/firefly luciferase value was assessed, and the data presented are the mean ± S.D. of the percentage normalized to the empty vector C.
Figure 3.
Silencing of RNAs transcribed from the HIV-1 LTR.
(A) A schematic of the HIV-1 LTR. The important transcriptional elements in the U3 region, characteristic secondary structures and major functional regions are shown. “M/E” indicates the modulator and enhancer regions, and “P” indicates the promoter region. Several truncated promoters, U3I, U3, EP, PI, PT, P and NFm, which is identical to P but carries mutations in the two NF-κB binding sites that render them nonfunctional, were generated as shown. The SV40 promoter and intron region of the psiCHECK were replaced with various regions of the LTR. (B) The RNAs transcribed by promoters derived from several regions of the HIV-1 LTR were exported by Rev-HA and analyzed to determine the effects of miRNA-mediated interference. The blue arrow points to the Bulge-containing constructs that carry a correctly oriented RRE and were not silenced in the presence of Rev-HA. (C) The schematics of the pU3I, pU3IN, psiCHECK and pSVEPI constructs. In the pU3IN and pSVEPI constructs, the region downstream of the TAR (light blue region) of pU3I and the region containing the chimeric intron from the psiCHECK were exchanged with each other. (D) The effects of splicing on miRNA-mediated silencing were analyzed using chimeric promoters derived from the LTR and SV40 promoters. The red arrowhead points to the Bulge-containing constructs that carry an intron and a correctly oriented RRE and were silenced in the presence of Rev-HA. The blue arrow points to the Bulge-containing constructs without an intron that carry a correctly oriented RRE and were not silenced in the presence of Rev-HA. The Renilla/firefly luciferase value was assessed in each graph. The empty vector C was used as a control in each promoter setting, and the results are presented as the mean ± S.D. as a percentage of the control from three independent experiments. “pcDNA” denotes the pcDNA3.1(+) plasmid. ***P<0.001, **P<0.005.
Figure 4.
The effects of mutations in suppressive sequences in the pol and env-nef regions.
(A) The pol and env-nef regions of pNL4-3 are illustrated. The cis-acting instability elements (INS) are also illustrated below the genome. The arrows indicate the mutated sites in the pol (sites “a” and “b”) and env-nef (sites “c”, “d” and “e”) regions. The black (>40%) and gray (40–20%) bars represent the repressed region. The number and bar pattern corresponds to the graph below. (B) The effects of mutations introduced into the repressive sequences in the pol region (sites “a” and “b”) were assessed in Jurkat cells. The “1 m” indicates the vector that has the mutated sequence introduced into the repressive site. (C) Mutational assay of the suppressive sequences (sites “c” and “d”) in the env-nef region. (D) Mutational assay of the suppressive sequences (site “e”) in the nef region. The Rluc activity was normalized to the firefly luciferase activity, and the data shown are the mean ± S.D. of the percentage of the activity of the empty psiCHECK-2 vector (psiCHECK). (E) Pattern mutations and the effects of each dissected sequence in the pol (sites “a” and “b”) and env-nef (sites “c”, “d” and “e”) regions. The gray characters represent unchanged residues. Asterisks denote mutations that did not change any amino acids (see Table S1 for details).
Figure 5.
The effects of combining multiple mutations in the pol and env-nef silencing regions.
(A) The pol and env-nef regions of pNL4-3 and the position of the cis-acting instability elements (INS) are illustrated. The arrows indicate the mutated sites in the pol (sites “a” and “b”) and env-nef (sites “c”, “d” and “e”) regions. The number and bar pattern corresponds to the graph below. (B) The effects of multiple mutations in the pol region (sites “a” and “b”; elements 1 and 7) were assessed in Jurkat cells. Six independent experiments were performed. The bar patterns correspond to the mutational patterns shown in (F). The “2 m” indicates that the vector has two mutation sites in the repressive sequence. (C) The effects of multiple mutations in the pol region were assessed in M4C8 cells. (D) The effects of multiple mutations introduced into the env-nef region (sites “c”, “d” and “e”; elements 15 and 16) were assessed in Jurkat cells. Four independent experiments were performed. The bar patterns correspond to the mutational patterns shown in (G). The “1 m”, “2 m” and “3 m” indicate that the vector has one, two and three mutation sites individually. The red and blue bars indicate the more derepressed mutational patterns. (E) The effects of multiple mutations introduced into the env-nef region were assessed in M4C8 cells. The psiCHECK was used as a control, and the results are expressed as the mean ± S.D. as a percentage of the control. ***P<0.001, **P<0.005 and *P<0.05. (F) Combinations of mutated patterns in the pol region and the bar patterns in graph B and C. (G) Combinations of mutated patterns in the env-nef region and the bar patterns in graph D and E. The gray characters represent unchanged residues. The asterisk denotes the mutational pattern used for the generation of mutant viruses.
Figure 6.
The characterization of suppressive sequences in the pol and env-nef regions.
(A) The effect of each anti-Drosha siRNA (siRNA1 and siRNA2) on Drosha protein level was evaluated by western blot using an anti-Drosha antibody. The reduced amount of Drosha protein in the Jurkat cells was confirmed at 48 h after treatment with anti-Drosha siRNAs compared with the control or mock treatment. After stripping, the membrane was reprobed with an anti-ß actin antibody. (B) Schematic representation of the experimental procedures for RNAi experiments in Jurkat cells. (C) The effects of the anti-Drosha siRNA at the repressed “a ” and “b” sites in the pol region (element 1 in Fig. 5A) and sites “c”, “d” and “e” in the env-nef region (element 15 in Fig. 5A) were assessed in Jurkat cells (black bars). The derepressive effect was confirmed in each region relative to the corresponding vectors with mutations in the repressive sites (mutation 1-2m-2 in the pol region (lattice bars) and 15-3m-2 in the env-nef region (blue bars); see also Fig. 5B–G). The bar patterns correspond to the mutational patterns shown in Fig. 5. The Renilla/firefly luciferase value was assessed, and the data shown are the mean percentages ± S.D. of the mutated vector from six independent experiments. ***P<0.001. (D) The Jurkat cells were transfected with each vector, and RNA-immunoprecipitation (IP) using anti-human Ago2 (hAgo2) antibody was performed. The input and the IP samples were adjusted equal volume and the same volume was loaded and analyzed by western blot. The levels of the immunoprecipitated Renilla luciferase mRNAs compared to Firefly luciferase mRNAs produced from the vectors (Vectors 1 and 15) and the mutated vectors (Vectors 1-2m-2 and 15-3m-2) were analyzed by RT-qPCR. The normalized values of the Renilla/firefly luciferase levels are shown. The each mutated vector was set to 100.
Table 1.
Putative microRNAs predicted to target pol or env-nef region.
Figure 7.
The effects of the mutations on viral replication and Gag processing.
(A) pNL4-3 plasmids with mutations in pol, env-nef or both were constructed (marked with asterisks in Fig. 5F and G). Jurkat cells were transfected with the resulting plasmids or pNL4-3 (“C”). The amount of p24 produced at 24 h after transfection was analyzed. The amount of p24 produced by the pNL4-3-transfected cells at 24 h post-transfection was set at 100%, and the p24 production in each tested line was expressed as the mean ± S.D. percent of this amount. The bar patterns correspond to the mutational patterns shown in the bottom right and in Fig. 5. (B) Mutant viruses or NL4-3 (“C”) were obtained by transfecting 293T cells with each plasmid, and the resulting viruses were normalized based on the amount of p24. Jurkat cells were infected with 100 ng of p24-normalized virus. The amount of p24 produced by NL4-3-infected cells at 6 days post-infection was set at 100%, and each p24 amount was expressed as the mean ± S.D. percent of this amount. (C) M4C8 cells were infected with 10 ng of p24-normalized virus and analyzed as described in (B). (D) TZM-bl cells were infected with 10 ng of p24-normalized virus. The firefly luciferase activities were assessed at 40 h post-infection and the luciferase activity by NL4-3-infected cells was set at 100%. The (−) indicates the activity of the mock-treated cells. (E) M4C8 cells were infected with 10 ng of p24-normalized virus. The amount and processing of Gag in the infected cells were analyzed at 5 days post-infection by western blot using the HIV-1 p24 Gag monoclonal antibody. The membrane was then stripped and reprobed with an anti-ß actin antibody as a loading control.
Figure 8.
The effect of Rev-dependent export for Let-7 family members and suppressive sequences within HIV-1 in Jurkat cells.
(A) The secondary structure of the miRNA/mRNA duplex of let-7 family members targeting the Bulge sequence was predicted using the RNA-hybrid program. (B) The relative RNA expression levels in the let-7 family members from Jurkat cells were calculated following normalization to the internal control RUN6B. (C) The silencing of RNAs containing the let-7 targeting-sequences in Jurkat cells. (D) The effects of the Bulge and BulgeMut sequences on RNA export by Rev-HA in Jurkat cells. The blue arrow points to the Bulge-containing constructs and the black arrow points to the BulgeMut-containing vectors. Three independent experiments were performed. **P<0.005, *P<0.05. (E) The effect of Rev-mediated export on RNAs in which the suppressive sequences identified in the pol (element 1 in Fig. 5A) and env-nef (element 15 in Fig. 5A) regions were inserted into the Rluc 3′ UTR was assessed in Jurkat cells (black bars, blue arrows). The corresponding vectors with mutations in the repressive sites (lattice and blue bars, black arrows) were also assessed as well (mutation 1-2m-2 in the pol region and 15-3m-2 in the env-nef region; Fig. 5B–G). The bar patterns correspond to the mutational patterns shown in Fig. 5. The Renilla/firefly luciferase value was assessed in each graph. The empty vector C was used as a control, and the results are presented as the mean ± S.D. as a percentage of the control. “pcDNA” denotes the pcDNA3.1(+) plasmid.