Figure 1.
Schematic representation of the intragenic cis-regulatory region of HIV-1.
The complete functional unit of the intragenic cis-regulatory region encompasses nt 4079 to nt 6026 and is composed of the 5103 fragment, the hypersensitive site HS7 and the 5105 fragment. The three AP-1 binding sites of fragment 5103 are indicated as well as the previously characterized binding sites of the HS7 region (site B which binds Oct-1 and a T cell specific complex termed B3 or the monocyte/macrophage lineage-specific complex PU.1, the GC-box which binds Sp1 and Sp3, site C which binds Oct-1 and a T-cell specific complex termed C3 and site D which binds unidentified complexes [6]).
Figure 2.
AP-1 transcription factors specifically bind in vitro to each of the three intragenic AP-1 sites of fragment 5103.
(A) The nucleotide sequence of the wild-type AP-1#1, AP-1#2 and AP-1#3 site oligonucleotides used as probes in our EMSAs are shown properly aligned with the AP-1 consensus sequence. The position of the AP-1 binding site is indicated by an arrow on each coding strand and mismatches in the AP-1 sites with respect to the AP-1 consensus sequence are designated by asterisks. The conservation of the intragenic AP-1 sites was assessed by comparing their sequences at the nucleotide level based on the full spectrum of HIV and SIV sequences compiled in the HIV compendium database (hiv-web.lanl.gov). Sequence logos that represent the frequency of the nucleotide present at each position in the intragenic AP-1 binding sites were generated based on these sequence analyses for each intragenic AP-1 site. (B) The AP-1#1, AP-1#2 and AP-1#3 oligonucleotide probes were incubated with nuclear extracts (10 µg) from mock-treated (lane 1) or PMA-treated (lanes 2 to 6) HeLa cells in the absence of competitor (lanes 1 and 2) or in the presence of a molar excess (5 fold) of a competitor corresponding to the homologous AP-1 site (lane 3), to the heterologous Sp1 consensus (lane 4; nucleotide sequence of the coding strand: 5′-ATTCGATCGGGGCGGGGCGAGC-3′), to the AP-1 consensus (lane 5; nucleotide sequence of the coding strand: 5′-CGCTTGATGACTCAGCCGGAA-3′) or to the mutated AP-1 consensus (lane 6; nucleotide sequence of the coding strand: 5′-CGCTTGATGACTTGGCCGGAA-3′, where mutations compared to the consensus are indicated in bold). The figure shows the specific retarded bands of interest, which are indicated by arrows. The terms C1, C2 and C3 refer to complexes 1, 2 and 3. (C) Nuclear extracts from PMA-treated HeLa cells (10 µg) were incubated, before the addition of the AP-1 probe, either with a purified rabbit IgG as a negative control (lane 1), or with an antibody directed against AP-1 family members including c-Fos (lane 2), FosB (lane 3), Fra-1 (lane 4), Fra-2 (lane 5), c-Jun (lane 6), JunB (lane 7) and JunD (lane 8), or with an antibody directed against other members of the B-ZIP family such as CREB (lane 9), CREM (lane 10), ATF-1 (lane 11), ATF-2 (lane 12), C/EBPα (lane 13), C/EBPβ (lane 14) and C/EBPδ (lane 15), or with an antibody directed against Ets-1 (lane 16). The figure shows the specific retarded bands of interest indicated by arrows. Supershifted complexes are indicated by asterisks. (D) Ten µg of nuclear extracts from PMA-treated HeLa cells were incubated with an antibody directed against c-Fos (lane 18), an antibody directed against JunB (lane 19) or a combination of both antibodies (lane 20). A purified rabbit IgG was used as a negative control (lane 17). The AP-1#1, AP-1#2 or AP-1#3 oligonucleotide probe was then added to the mixture. The figure shows the specific retarded bands of interest indicated by arrows. The supershifted complexes are indicated by asterisks.
Figure 3.
Identification of point mutations which abolish AP-1 transcription factors' in vitro binding to their respective binding sites located in fragment 5103.
(A) The wild-type and mutated AP-1#1, AP-1#2 and AP-1#3 oligonucleotide sequences are indicated as well as the corresponding underlying amino acid sequence of the viral reverse transcriptase. Base pairs which were modified in the mutated versions of the AP-1 binding sites relative to the wild-type versions are indicated by asterisks. (B) The AP-1#1, AP-1#2 and AP-1#3 oligonucleotide probes were incubated with nuclear extracts (10 µg) from mock-treated (lane 1) or PMA-treated (lanes 2 to 6) HeLa cells in the absence of competitor (lanes 1 and 2) or in the presence of increasing molar excesses (2 and 4 fold) of each respective homologous AP-1 oligonucleotide (lanes 3 and 4) or of the corresponding mutated AP-1 oligonucleotide (lanes 5 and 6). The figure shows the specific retarded bands of interest, which are indicated by arrows.
Figure 4.
The intragenic AP-1 binding sites are fully responsible for the PMA-dependent enhancer activity of fragment 5103.
HeLa cells were transiently transfected with 200 ng of the following constructs: the control vector pTK, the pTK-5103s-wt, pTK-5103as-wt, pTK-5103as-totmut, pTK-5103as-mut1, pTK-5103as-mut2, pTK-5103as-mut3 or pTK-5103as-mut1+2 reporter construct. All transfection mixtures additionally contained the pRL-TK (1 ng), in which a cDNA encoding the Renilla luciferase gene is under the control of the HSV TK promoter. At 24 h post-transfection, cells were mock-treated (-) or treated with PMA (+) (20 nM). Luciferase activities (Firefly and Renilla) were measured in cell lysates 48 h after transfection. Results are expressed as LuciferaseFirefly/[Proteins]*LuciferaseRenilla and presented as histograms indicating the luciferase activities of each construct relative to that of the control vector pTK, which was assigned an arbitrary value of 1 in absence of PMA. Means and standard errors of the means from three independent transfection experiments each performed in triplicate are indicated. The PMA induction of each TK construct is written in the upper part of the Figure (in fold). * indicates p<0.05 compared with the wild-type construct.
Figure 5.
Overexpression of the dominant negative A-Fos mutant impairs the PMA-dependent enhancer activity of fragment 5103.
(A) HeLa cells were transiently cotransfected with 100 ng of the pTK, pTK-5103as-wt or pTK-5103as-totmut reporter construct and with increasing amounts (0, 50 or 100 ng) of the dominant negative construct pCG-AFos. To maintain the same amount of transfected DNA and to avoid squelching artifacts, the different quantities of A-Fos expression vector cotransfected were complemented to 100 ng of DNA by using the empty vector pCG. Cells were additionally cotransfected with 1 ng of pRL-TK. Twenty-four hours post-transfection, cells were mock-treated (-) or treated with PMA (+). Luciferase activities (Firefly and Renilla) were measured in cell lysates 48 h after transfection. Results are expressed as LuciferaseFirefly/LuciferaseRenilla and presented as histograms indicating the relative luciferase activity of each construct with respect to the activity of the same reporter construct in the absence of both PMA and A-Fos, which was assigned an arbitrary value of 1. Means and standard errors of the means from three independent transfection experiments each performed in triplicate are indicated, with p<0.01 compared with the empty vector pTK indicated by an asterisk. (B) A-Fos mutant expression is efficient and not affected by PMA-treatment of the cells. HeLa cells were transiently transfected with increasing amounts (0, 2 or 4 µg) of the dominant negative construct pCG-AFos. To maintain the same amount of transfected DNA, the different quantities of A-Fos expression vector transfected were complemented to 4 µg of DNA by using the empty vector pCG. Twenty-four hours post-transfection, cells were mock-treated (mock) or treated with PMA (PMA) for one hour. Nuclear extracts were then prepared and analysed by western blot with an antibody directed against the N-terminal FLAG epitope of the A-Fos mutant.
Figure 6.
Binding and functional properties of intragenic AP-1 sites are independent of Tat protein expression.
(A) HeLa cells were transiently transfected with the expression vector pTat72 encoding the one-exon form of Tat (72 amino acids named Tat72), the expression vector pTat101 encoding the two-exon form of Tat (101 amino acids or Tat101), or with the corresponding empty vector pREP9 used as a control. Twenty-four hours post-transfection, cells were mock-treated (-) or treated with PMA (+) for one hour. Nuclear extracts from these transiently transfected cells were prepared and used in EMSA experiments. The AP-1#1, AP-1#2 and AP-1#3 probes incubated with 10 µg of nuclear extracts from HeLa cells (lanes 1 and 2), HeLa cells expressing Tat72 (lanes 3 and 4) or HeLa cells expressing Tat101 (lanes 5 and 6), which were mock-treated (lanes 1, 3 and 5) or treated with PMA (lanes 2, 4 and 6). Retarded bands of interest are shown and indicated by arrows. (B) HeLa cells were transiently cotransfected with 100 ng of the pTK, pTK-5103as-wt or pTK-5103as-totmut reporter construct and with 200 ng of an expression vector encoding either the one-exon form of Tat (pTat72) or the two-exon form of Tat (pTat101) or of the corresponding empty vector pREP9. All transfection mixtures contained 1 ng of pRL-TK as a control of transfection efficiency. Twenty-four hours post-transfection, cells were mock-treated (-) or treated with PMA (+). Luciferase activities (Firefly and Renilla) were measured in cell lysates 48 h post-transfection. Results are expressed as LuciferaseFirefly/[Proteins]*LuciferaseRenilla and presented as histograms indicating the relative luciferase activities with respect to the activity of the control vector pTK in the absence of PMA, which was assigned an arbitrary value of 1. Means and standard errors of the means of one representative experiment from three independent transfection experiments each performed in triplicate are indicated. * indicates p<0.01 compared to the empty vector pTK.
Figure 7.
The AP-1 transcription factors c-Fos, JunB and JunD are recruited in vivo to the 5103 fragment region.
HeLa cells were transiently transfected with the wild-type pHIV-1 or the mutated pHIV-1-AP-1#totmut construct. Twenty-four hours post-transfection, cells were mock-treated (-) or treated with PMA (+). Twenty-four hours post-induction, cells were cross-linked for 10 min at room temperature with 1% formaldehyde. To detect chromosomal flanking regions, pellets were sonicated to obtain DNA fragments of an average size of 400 bp. Chromatin immunoprecipitations were performed with specific antibodies directed against c-Fos, JunB, Fra-1 or JunD. To test aspecific binding to the beads, a purified IgG was used as a control for immunoprecipitation. Quantitative PCR reactions were performed with oligonucleotide primers hybridizing either in the nuc-1 region (termed nuc-1), or in a region overlapping the three AP-1 binding sites of the 5103 fragment (termed 5103 fragment), or in the vpr gene (termed vpr gene) where no AP-1 binding sites have been previously reported. Fold enrichments were calculated as percentages of immunoprecipitated DNA following the formula “Immunoprecipitated DNA (IP)*100/Input DNA (INP)”. Values represent the means of triplicate samples and standard errors of the means are indicated. An experiment representative of three independent ChIP assays is shown. (B) Mutations in the intragenic AP-1 sites affect the PMA-inducible in vivo recruitment of RNA polymerase II to the HIV-1 5′LTR region. Chromatin immunoprecipitations were performed with a specific antibody directed against RNAPII and a purified IgG as a control. Quantitative PCR reactions were performed with the same oligonucleotide primers and fold enrichments were calculated as in panel (A). Means and standard errors of the means from one experiment representative of three independent ChIP assays are shown.
Figure 8.
Viral particles from mutant and wild-type virus stocks are similar at the protein and at the RNA levels.
(A) Equivalent amounts of viral particles (as assessed by p24 ELISA assays) from the wild-type and each AP-1 mutant virus stocks were pelleted by centrifugation, lysed in Laemmli buffer and analyzed by Western blotting with an anti-HIV-1 immunoglobulin. The bands corresponding to the HIV-1 glycoprotein gp160, reverse transcriptase p66/p51, integrase p32 and capsid p24 proteins are indicated. MW, molecular weight (indicated in kDa). (B) Viral RNAs from equal amounts of viral particles were digested with DNase I and subsequently reverse-transcribed with random primers. First-strand viral cDNAs were then quantified by qPCR with primers hybridyzing in the TAR region (as described in the Materials and Methods section). An arbitrary value of 1 was assigned to the result obtained with the wild type virus stock HIV-1. Means and standard errors of the means from two independent experiments each performed in triplicate are indicated. (C) Mutations in the intragenic AP-1 binding sites affect HIV-1 expression. TZM-bl cells (6×103 cells) were infected or not with equal amounts of wild-type HIV-1 or mutant virus stocks. At 72 h post-infection, TZM-bl cells were lysed and luciferase activity was measured in cell lysates. Results are presented as histograms indicating the LuciferaseFirefly activity of the TZM-bl cells following infection with wild-type versus mutant viruses. An arbitrary value of 100% was attributed to the result obtained with the wild-type HIV-1. Means and standard errors of the means from one representative from three independent experiments each performed in triplicate are indicated. * indicates p<0.05 compared to the wild-type virus HIV-1.
Figure 9.
The AP-1 binding sites located in the 5103 fragment are important for viral replication.
Jurkat (A and C) or U937 (B and D) cells were infected with equivalent amounts of p24 concentration of wild-type (HIV-1), totally mutated (HIV-1-AP-1totmut), partially mutated (HIV-1-AP-1#1+2mut) or individually mutated (HIV-1-AP-1#1mut, HIV-1-AP-1#2mut and HIV-1-AP-1#3mut) viral infectious stocks as described in the Materials and Methods section. (A and B) Viral production in cell supernatants was quantified by measuring p24 antigen concentration in the culture supernatants at different times following infection. The experiment shown is representative of at least 3 independent infection experiments (each performed in triplicate). The means are presented and the variation for a given mutant between different experiments was <15% in each case. (C and D) Total RNA was extracted from infected cells at days 5, 10 and 15 post-infection, digested with DNase I and reverse-transcribed using random primers. First-strand cDNAs were analyzed by qPCR with the comparative Ct (ΔΔCt) quantification method using the two following sets of primers: initiated transcripts (TAR primers) and elongated transcripts (Tat primers) were quantified using β-actin to normalize the results. Means and standard errors of the means from two independent experiments each performed in triplicate are indicated. * and ** indicate p<0.05 and p<0.1 compared to the wild-type virus HIV-1 at day 15 post-infection.
Figure 10.
The splicing pattern of HIV-1 transcripts is unaffected by the mutations introduced in the intragenic AP-1 binding sites.
Total RNA was extracted from infected Jurkat or U937 cells (with wild-type or mutant virus stocks) at 5 days post-infection. After DNase I treatment, total RNA was reverse-transcribed with random primers and first strand cDNAs were quantified with primers designed to quantify full-length unspliced viral mRNAs, singly-spliced viral mRNAs and multiply-spliced viral mRNAs. The total amount of viral mRNAs for each virus was arbitrarily attributed a value of 100% and the proportion of each type of transcript is presented as histograms indicating the means and standard errors of the means from two independent experiments performed in duplicate.
Figure 11.
Single-round HIV-1 infections of MDMs are affected by mutations in the intragenic AP-1 binding sites.
Monocyte-derived macrophages from three healthy donors were isolated and individually infected with VSV-G pseudotyped HIV-1NL4.3 viruses (HIV-1VSV, HIV-1-AP-1#totmutVSV, HIV-1-AP-1#1mutVSV, HIV-1-AP-1#2mutVSV, HIV-1-AP-1#3mutVSV, HIV-1-AP-1#1+2mutVSV). Production of p24 in the culture supernatant was measured by ELISA at day 10 post-infection. Results are presented as histograms indicating the p24 production level of each mutant virus compared to the wild-type virus HIV-1VSV, which was assigned an arbitrary value of 100%, and correspond to results obtained from three independent donors in order to take into account the variability that may exist between donors. Means and standard errors of the means are indicated. * indicates p<0.05 compared to the wild-type virus HIV-1VSV.