Modulation of human immunodeficiency virus 1 replication by interferon regulatory factors

Abstract

Transcription of the human immunodeficiency virus (HIV)-1 is controlled by the cooperation of virally encoded and host regulatory proteins. The Tat protein is essential for viral replication, however, expression of Tat after virus entry requires HIV-1 promoter activation. A sequence in the 5' HIV-1 LTR, containing a binding site for transcription factors of the interferon regulatory factors (IRF) family has been suggested to be critical for HIV-1 transcription and replication. Here we show that IRF-1 activates HIV-1 LTR transcription in a dose-dependent fashion and in the absence of Tat. This has biological significance since IRF-1 is produced early upon virus entry, both in cell lines and in primary CD4+ T cells, and before expression of Tat. IRF-1 also cooperates with Tat in amplifying virus gene transcription and replication. This cooperation depends upon a physical interaction that is blocked by overexpression of IRF-8, the natural repressor of IRF-1, and, in turn is released by overexpression of IRF-1. These data suggest a key role of IRF-1 in the early phase of viral replication and/or during viral reactivation from latency, when viral transactivators are absent or present at very low levels, and suggest that the interplay between IRF-1 and IRF-8 may play a key role in virus latency.

Figure 1.

Effect of IRFs on HIV-1 LTR transactivation. (A) Jurkat cells were transiently cotransfected with the HIV-1 LTR-CAT (1 μg) and vectors (2 μg) expressing the indicated IRFs. IRF-3 5D and IRF-7* codify for the constitutively activated forms of IRF-3 and -7, respectively (references and 41). After 24 h, CAT activity was evaluated as indicated in the Materials and Methods. (B) Dose–response effect of the wild-type IRF-1 or its mutant deleted in the activation domain (Δ IRF-1) on HIV-1 LTR-directed gene expression. Cells were transfected as in A except that 0.8 μg of HIV-1 LTR-CAT and increasing amounts of the vectors expressing IRF-1 or Δ IRF-1 (0.8, 1.6, 3.2 μg) were used. (C) Effect of IRF-1 on mutated HIV-LTR constructs. Cells were transiently transfected with wild-type HIV-LTR or Δ₁ LTR in which the ISRE region is deleted (reference 9), Δ₂ LTR in which the NF-κB sites are mutated or Δ₃ LTR in which both sites are deleted/mutated. IRF-1–expressing vector was cotransfected as indicated. Results are shown as percentages of the CAT activity of the wild-type HIV-LTR in IRF-1–transfected cells. (D) IRF-1 cooperates with Tat to enhance HIV-1 LTR-CAT activity. Cells were transiently cotransfected with the HIV-1 LTR-CAT (1 μg), Tat (5 ng), or IRF-1 (1 μg) expressing vectors, either alone or in combination, as indicated. CAT activity was determined after 24 h. The results quantified by an Instant Imager are reported as mean levels ± SE from three separate experiments.

Figure 1.

Effect of IRFs on HIV-1 LTR transactivation. (A) Jurkat cells were transiently cotransfected with the HIV-1 LTR-CAT (1 μg) and vectors (2 μg) expressing the indicated IRFs. IRF-3 5D and IRF-7* codify for the constitutively activated forms of IRF-3 and -7, respectively (references and 41). After 24 h, CAT activity was evaluated as indicated in the Materials and Methods. (B) Dose–response effect of the wild-type IRF-1 or its mutant deleted in the activation domain (Δ IRF-1) on HIV-1 LTR-directed gene expression. Cells were transfected as in A except that 0.8 μg of HIV-1 LTR-CAT and increasing amounts of the vectors expressing IRF-1 or Δ IRF-1 (0.8, 1.6, 3.2 μg) were used. (C) Effect of IRF-1 on mutated HIV-LTR constructs. Cells were transiently transfected with wild-type HIV-LTR or Δ₁ LTR in which the ISRE region is deleted (reference 9), Δ₂ LTR in which the NF-κB sites are mutated or Δ₃ LTR in which both sites are deleted/mutated. IRF-1–expressing vector was cotransfected as indicated. Results are shown as percentages of the CAT activity of the wild-type HIV-LTR in IRF-1–transfected cells. (D) IRF-1 cooperates with Tat to enhance HIV-1 LTR-CAT activity. Cells were transiently cotransfected with the HIV-1 LTR-CAT (1 μg), Tat (5 ng), or IRF-1 (1 μg) expressing vectors, either alone or in combination, as indicated. CAT activity was determined after 24 h. The results quantified by an Instant Imager are reported as mean levels ± SE from three separate experiments.

Figure 2.

IRF-1 mRNA is induced early upon HIV-1 infection and before expression of Tat. (A) Jurkat cells were infected with the HIV-1 strain IIIB (5,000 cpm/ml) and, at the indicated time points, total RNA was extracted and analyzed by RNase protection with a IRF-1–specific antisense riboprobe. 18S rRNA was used as a control of RNA loading. (B) mRNA relative fold-increase after normalization to the 18S RNA, quantified by Instant Imager. Mean values from three separate experiments are shown. (C) Total RNA extracted at 5 and 24 h after infection shown in A was analyzed by RT-PCR for the doubly-spliced (tat/rev) transcript as described in Materials and Methods.

Figure 2.

IRF-1 mRNA is induced early upon HIV-1 infection and before expression of Tat. (A) Jurkat cells were infected with the HIV-1 strain IIIB (5,000 cpm/ml) and, at the indicated time points, total RNA was extracted and analyzed by RNase protection with a IRF-1–specific antisense riboprobe. 18S rRNA was used as a control of RNA loading. (B) mRNA relative fold-increase after normalization to the 18S RNA, quantified by Instant Imager. Mean values from three separate experiments are shown. (C) Total RNA extracted at 5 and 24 h after infection shown in A was analyzed by RT-PCR for the doubly-spliced (tat/rev) transcript as described in Materials and Methods.

Figure 2.

IRF-1 mRNA is induced early upon HIV-1 infection and before expression of Tat. (A) Jurkat cells were infected with the HIV-1 strain IIIB (5,000 cpm/ml) and, at the indicated time points, total RNA was extracted and analyzed by RNase protection with a IRF-1–specific antisense riboprobe. 18S rRNA was used as a control of RNA loading. (B) mRNA relative fold-increase after normalization to the 18S RNA, quantified by Instant Imager. Mean values from three separate experiments are shown. (C) Total RNA extracted at 5 and 24 h after infection shown in A was analyzed by RT-PCR for the doubly-spliced (tat/rev) transcript as described in Materials and Methods.

Figure 4.

IRF-1 and IRF-2 bind the HIV-ISRE. (A) Nuclear cell extracts (20 μg) from Jurkat cells uninfected or infected with HIV-1 IIIB strain (5,000 cpm/ml) were prepared at different time points after infection and incubated with an oligonucleotide corresponding to the four tandem IRF-binding sites (C13). Supershift assays were performed in the presence of specific anti–IRF-1 and anti–IRF-2 antibodies as indicated. Binding complexes were resolved by PAGE and visualized by autoradiography. (B) DNA pull-down assays. Biotinylated oligodeoxynucleotides containing the wild-type or a mutated version of the HIV-ISRE (Materials and Methods), coupled to Streptavidin MagneSphere were incubated with nuclear extracts from Jurkat cells or primary CD4⁺ T cells infected with the HIV-1 IIIB strain (5,000 cpm/ml). The bound proteins were eluted from the beads by boiling in sample buffer and analyzed by WB with antibodies against IRF-1. INPUT indicates the level of endogenous IRF-1 in the uninfected and infected nuclear cell extracts (20 μg) at the indicated time points determined by WB analysis.

Figure 3.

HIV-1 infection induces IRF-1 mRNA expression in primary CD4⁺ T lymphocytes. (A) Purified CD4⁺ T cells were infected with the HIV-1 IIIB strain (5,000 cpm/ml) and IRF-1 mRNA evaluated by semiquantitative RT-PCR analysis at the indicated time points after infection. A representative experiment out of three performed is shown. The increase in IRF-1 mRNA accumulation begun to be evident between 5 and 12 h after infection and peaked between 24 and 48 h after infection depending on the donor. 26S RNA was used for normalization as described in Materials and Methods. (B) mRNA relative fold-increase after normalization to the 26S RNA quantified by Instant Imager. (C) Total RNA extracted at the indicated time points as in A was analyzed by RT-PCR for the doubly-spliced (tat/rev) transcript as described in Materials and Methods.

Figure 3.

HIV-1 infection induces IRF-1 mRNA expression in primary CD4⁺ T lymphocytes. (A) Purified CD4⁺ T cells were infected with the HIV-1 IIIB strain (5,000 cpm/ml) and IRF-1 mRNA evaluated by semiquantitative RT-PCR analysis at the indicated time points after infection. A representative experiment out of three performed is shown. The increase in IRF-1 mRNA accumulation begun to be evident between 5 and 12 h after infection and peaked between 24 and 48 h after infection depending on the donor. 26S RNA was used for normalization as described in Materials and Methods. (B) mRNA relative fold-increase after normalization to the 26S RNA quantified by Instant Imager. (C) Total RNA extracted at the indicated time points as in A was analyzed by RT-PCR for the doubly-spliced (tat/rev) transcript as described in Materials and Methods.

Figure 3.

HIV-1 infection induces IRF-1 mRNA expression in primary CD4⁺ T lymphocytes. (A) Purified CD4⁺ T cells were infected with the HIV-1 IIIB strain (5,000 cpm/ml) and IRF-1 mRNA evaluated by semiquantitative RT-PCR analysis at the indicated time points after infection. A representative experiment out of three performed is shown. The increase in IRF-1 mRNA accumulation begun to be evident between 5 and 12 h after infection and peaked between 24 and 48 h after infection depending on the donor. 26S RNA was used for normalization as described in Materials and Methods. (B) mRNA relative fold-increase after normalization to the 26S RNA quantified by Instant Imager. (C) Total RNA extracted at the indicated time points as in A was analyzed by RT-PCR for the doubly-spliced (tat/rev) transcript as described in Materials and Methods.

Figure 5.

GST-pull down assays. The indicated IRFs (IRF-1, IRF-2, IRF-3, IRF-4, IRF-7, IRF-8, a COOH-terminal deleted mutant of IRF-1 [Δ IRF-1]), and Tat, were translated in vitro in the presence of ³⁵[S] methionine as indicated in Materials and Methods and incubated with recombinant GST–Tat or GST-IRF-1 fusion proteins immobilized on glutathione-sepharose. Input corresponds to 10% of the ³⁵[S]-labeled proteins used in the binding experiments. The complexes were resolved by PAGE and detected by autoradiography. Binding of ³⁵[S]-labeled proteins to beads containing only GST protein is also shown. Quantitation of incorporated radioactivity was performed by Instant Imager.

Figure 6.

IRF-1 and Tat associate intracellularly. (A) In vitro–translated IRF-1 (lanes 1–3) and nuclear cell extracts from Jurkat cells treated with IFN-γ (lanes 4–6) or control medium (lanes 7–9) were incubated with purified GST–Tat fusion protein or GST alone. Bound proteins were then analyzed by WB using anti–IRF-1 polyclonal antibody as described in Materials and Methods. 10% of the extract used for binding assays is shown in lane 7 (untreated cells) and lane 4 (IFN-γ-treated cells). The slowly migrating IRF-1^p* band observed in IFN-γ–treated cell extracts is due to the phosphorylation induced by IFN-γ (reference 64). Extra bands in lanes 7–9 are not specific. (B) 293 HEK cells were transfected with the expression plasmids encoding IRF-1 or Tat, alone or in combination. Whole cell extracts (300 μg) were immunoprecipitated with anti–IRF-1 antibodies (αIRF-1). Immunoprecipitated complexes were separated by 10% SDS-PAGE and subsequently probed with anti-Tat antibodies (αTat) as indicated. Whole cell extracts (10 μg) were separated on 10% or 15% SDS-PAGE and probed with anti-Tat or anti–IRF-1 antibodies.

Figure 6.

IRF-1 and Tat associate intracellularly. (A) In vitro–translated IRF-1 (lanes 1–3) and nuclear cell extracts from Jurkat cells treated with IFN-γ (lanes 4–6) or control medium (lanes 7–9) were incubated with purified GST–Tat fusion protein or GST alone. Bound proteins were then analyzed by WB using anti–IRF-1 polyclonal antibody as described in Materials and Methods. 10% of the extract used for binding assays is shown in lane 7 (untreated cells) and lane 4 (IFN-γ-treated cells). The slowly migrating IRF-1^p* band observed in IFN-γ–treated cell extracts is due to the phosphorylation induced by IFN-γ (reference 64). Extra bands in lanes 7–9 are not specific. (B) 293 HEK cells were transfected with the expression plasmids encoding IRF-1 or Tat, alone or in combination. Whole cell extracts (300 μg) were immunoprecipitated with anti–IRF-1 antibodies (αIRF-1). Immunoprecipitated complexes were separated by 10% SDS-PAGE and subsequently probed with anti-Tat antibodies (αTat) as indicated. Whole cell extracts (10 μg) were separated on 10% or 15% SDS-PAGE and probed with anti-Tat or anti–IRF-1 antibodies.

Figure 7.

IRF-8 but not IRF-2 inhibits the IRF-1-mediated and Tat-mediated HIV-1 LTR activity. (A and B) Transient cotransfections were performed with the HIV-LTR CAT reporter construct (1 μg) and IRF-1, IRF-2, IRF-8 (1 μg), or Tat (5 ng) expression vectors, respectively, as indicated. CAT activity was quantified 48 h after transfection. (C) RNase protection assay with a IRF-8–specific antisense riboprobe on total RNA extracted from Jurkat cells transfected with an empty vector or an IRF-8–expressing vector. IRF-8 indicates the transcript of the transduced gene and IRF-8e the endogenous recognized transcript. 18S RNA was used as a control of RNA loading and tRNA as a control of specificity. (D) Jurkat cells constitutively expressing IRF-8 or the empty vector were transiently transfected with the Tat-expressing vector (20 ng) or IRF-1–expressing vector (1 μg) along with the HIV-LTR reporter construct and CAT activity quantified as described in B. The results quantified by an Instant Imager are reported as mean levels ± SE from three separate experiments.

Figure 7.

IRF-8 but not IRF-2 inhibits the IRF-1-mediated and Tat-mediated HIV-1 LTR activity. (A and B) Transient cotransfections were performed with the HIV-LTR CAT reporter construct (1 μg) and IRF-1, IRF-2, IRF-8 (1 μg), or Tat (5 ng) expression vectors, respectively, as indicated. CAT activity was quantified 48 h after transfection. (C) RNase protection assay with a IRF-8–specific antisense riboprobe on total RNA extracted from Jurkat cells transfected with an empty vector or an IRF-8–expressing vector. IRF-8 indicates the transcript of the transduced gene and IRF-8e the endogenous recognized transcript. 18S RNA was used as a control of RNA loading and tRNA as a control of specificity. (D) Jurkat cells constitutively expressing IRF-8 or the empty vector were transiently transfected with the Tat-expressing vector (20 ng) or IRF-1–expressing vector (1 μg) along with the HIV-LTR reporter construct and CAT activity quantified as described in B. The results quantified by an Instant Imager are reported as mean levels ± SE from three separate experiments.

Figure 7.

IRF-8 but not IRF-2 inhibits the IRF-1-mediated and Tat-mediated HIV-1 LTR activity. (A and B) Transient cotransfections were performed with the HIV-LTR CAT reporter construct (1 μg) and IRF-1, IRF-2, IRF-8 (1 μg), or Tat (5 ng) expression vectors, respectively, as indicated. CAT activity was quantified 48 h after transfection. (C) RNase protection assay with a IRF-8–specific antisense riboprobe on total RNA extracted from Jurkat cells transfected with an empty vector or an IRF-8–expressing vector. IRF-8 indicates the transcript of the transduced gene and IRF-8e the endogenous recognized transcript. 18S RNA was used as a control of RNA loading and tRNA as a control of specificity. (D) Jurkat cells constitutively expressing IRF-8 or the empty vector were transiently transfected with the Tat-expressing vector (20 ng) or IRF-1–expressing vector (1 μg) along with the HIV-LTR reporter construct and CAT activity quantified as described in B. The results quantified by an Instant Imager are reported as mean levels ± SE from three separate experiments.

Figure 8.

Inhibition of HIV-1 replication in IRF-8–expressing Jurkat cells. Jurkat cells stably transfected with the IRF-8 (lanes 2, 4, and 6) or the RcCMV (control vector) (lanes 1, 3, and 5) were infected with the HIV-1 IIIB strain at an infectious dose corresponding to 1,000 or 5,000 cpm/ml of RT activity. (A) Cells were collected after 24 and 48 h and total RNA analyzed by RT-PCR, as described in Materials and Methods. (B) HIV-p24 antigen production. After 48, 72 and 144 h, p24 antigen accumulation was determined in the cell supernatants as indicated in Materials and Methods.

Figure 8.

Inhibition of HIV-1 replication in IRF-8–expressing Jurkat cells. Jurkat cells stably transfected with the IRF-8 (lanes 2, 4, and 6) or the RcCMV (control vector) (lanes 1, 3, and 5) were infected with the HIV-1 IIIB strain at an infectious dose corresponding to 1,000 or 5,000 cpm/ml of RT activity. (A) Cells were collected after 24 and 48 h and total RNA analyzed by RT-PCR, as described in Materials and Methods. (B) HIV-p24 antigen production. After 48, 72 and 144 h, p24 antigen accumulation was determined in the cell supernatants as indicated in Materials and Methods.

Figure 9.

IRF-8 inhibits the binding of IRF-1 to immobilized Tat. Recombinant GST–Tat fusion protein was immobilized on glutathione agarose beads and incubated with the indicated ³⁵[S]-labeled IRFs as described in Materials and Methods. Input corresponds to 10% of the ³⁵[S]-labeled proteins used in the binding experiments; in lanes 1, 5, and 6 are shown the in vitro labeled IRF-8, IRF-2, and IRF-1, respectively. In lane 2, ³⁵[S]-labeled IRF-1 and IRF-8 were incubated together for 15 min at room temperature before the addition of GST–Tat fusion protein beads. Lane 3 shows the binding of ³⁵[S]-labeled IRF-1 alone to GST–Tat beads. In lane 4, ³⁵[S]-labeled IRF-1 and IRF-2 were preincubated together for 15 min at room temperature before the addition of GST–Tat fusion protein beads.

Figure 10.

IRF-1 overexpression releases the inhibition of HIV-1 replication by IRF-8. Bulk populations of Jurkat cells stably transfected with IRF-8 were engineered to constitutively express IRF-1. (A) RNase protection with a IRF-1–specific antisense riboprobe on total RNA extracted from IRF-8–expressing cells, transfected with an empty vector (RcIRF-8/Rc), or with an IRF-1–expressing vector (RcIRF-8/IRF-1). 18S RNA was used as a control of RNA loading and tRNA as a control of specificity. (B) Cells were infected with the HIV-1 IIIB strain as in Fig. 8 and after 48, 72, and 144 h, HIV p24 antigen accumulation was determined in the cell supernatants as indicated in Materials and Methods.

Figure 10.

IRF-1 overexpression releases the inhibition of HIV-1 replication by IRF-8. Bulk populations of Jurkat cells stably transfected with IRF-8 were engineered to constitutively express IRF-1. (A) RNase protection with a IRF-1–specific antisense riboprobe on total RNA extracted from IRF-8–expressing cells, transfected with an empty vector (RcIRF-8/Rc), or with an IRF-1–expressing vector (RcIRF-8/IRF-1). 18S RNA was used as a control of RNA loading and tRNA as a control of specificity. (B) Cells were infected with the HIV-1 IIIB strain as in Fig. 8 and after 48, 72, and 144 h, HIV p24 antigen accumulation was determined in the cell supernatants as indicated in Materials and Methods.