Suppression of HIV-1 Nef translation by Sam68 mutant-induced stress granules and nef mRNA sequestration

Abstract

HIV-1 Nef plays important roles in HIV-1 replication and pathogenesis. It is translated from completely spliced HIV-1 RNA, and its expression is inherently regulated at the levels of viral DNA transcription and RNA splicing. Here we show that Sam68 cytoplasmic mutants potently suppress Nef expression. The suppression requires Sam68 domain aa 269-321 and is correlated with its ability to induce stress granules. In addition, the suppression is specific to Nef, and direct binding to nef mRNA 3'UTR confers the suppression specificity. Furthermore, nef mRNA is targeted to and enriched in these induced stress granules. Importantly, Nef suppression occurs in the context of HIV-1 infection of CD4+ T lymphocytes with little MHC I and CD4 downregulation. Taken together, these results demonstrate that stress granule induction and nef mRNA sequestration account for this translational suppression of Nef expression and offer a strategy for development of anti-HIV therapeutics to buttress our fight against HIV/AIDS.

Figure 1. Δ410 effects on HIV-1 replication and Nef expression

A–D. 293T cells were transfected with NL4-3.Luc(env) (A & C) or NL4-3.Luc(nef) (B & D), and Sam68 or Δ410, cell culture supernatants were collected for the RT activity assay 42 hr after transfection (A & B), and the cells were harvested for the luciferase activity assay (C & D). The data were mean ± SEM. E. 293T cells were transfected with NL4-3, and an increasing amount of HA-tagged Sam68 or Δ410. The cells were harvested for Western blot analysis using anti-Nef, anti-HA or anti-β-actin antibodies 42 hr after transfection. F. 293T cells were transfected with NL4-3, JRCSF, YU2, SG3.1, or LAI.2 and Sam68 or Δ410 and similar Western blot analysis was performed.

Figure 2. Effects of Sam68 deletion mutation on HIV-1 Nef expression and their subcellular localization

A. 293T cells were transfected with NL4-3 and GFP-tagged Sam68 or each of the mutants, followed by Western blot analysis for anti-Nef, anti-GFP or anti-β-actin. B & C. COS-7 cells were transfected with each of the GFP-tagged plasmids in triplicates. Cells were stained in 100 ng/ml DAPI for nuclei. The images were representative of each GFP fusion protein (B). Random 10 fields were counted for the total number of transfected cells and the cells exhibiting granule-like structures and expressed as the percentage of the granule-containing cells (C). The data were mean ± SEM.

Figure 3. Δ410 co-localization with stress granule markers G3BP, TIA-1, and hnRNP A1

COS-7 cells were transfected with HA-tagged Δ410 and GFP.G3BP (A), GFP.TIA-1 (B), or GFP.hnRNP A1 (C). Prior to fixation, the cells were treated with or without 0.5 mM sodium arsenite (ARS) for 1 hr. The cells were then stained with an anti-HA antibody followed by a phycoerythrin (PE)-conjugated secondary antibody. D & E. 293T cells were transfected with GFP.Δ410 only and treated with ARS as described above. The cells were then stained using an anti-TIA-1 antibody (D) or an anti-eIF3η antibody (E) and PE-conjugated secondary antibody. F. 293T cells were transfected with GFP.Δ 410, GFP.TIA-1 or GFP.G3BP and treated with ARS as above, or with 100 μg/ml cyclohexamine for 90 min. G. COS-7 cells were transfected with HA-tagged Δ410 and GFP.hdcp-1α. H. 293T cells were transfected with RFP.Sam68 or GFP.PABP and treated with ARS as above. In all transfections, cells were stained in DAPI for nuclei and the images were representative of each transfection. The co-localization composite was shown as “Merge”.

Figure 4. The specificity of Δ410 suppression to Nef expression over Tat and Rev

A & C. The secondary structure of 5′UTR and 3′UTR of all minigenes were predicted by the RNAfold program. The length in nucleotides for each 5′UTR and 3′UTR was given beneath each minigene and mutants. B & D. 293T cells were transfected with each of the minigenes, Sam68 or each of its mutants. Cells were harvested to determine Tat, Rev, or Nef expression by Western blot analysis.

Figure 5. Effects of Δ410-induced SG formation on localization of nef RNA transcripts

293T cells were transfected with GFP.Δ 410, NL4-3, or both. The cells were then processed for the fluorescence in situ hybridization as described in Experimental Procedures using a Cy5-labelled nef oligonucleotide. The images were representative of each co-transfection. Co-localization of Cy5-nef and Δ410 was shown in the column marked as “Merge”.

Figure 6. Δ410-induced SG formation and its effect on HIV-1 Nef expression in Jurkats and PBMC

A. Jurkats were transfected with GFP.G3BP or GFP.TIA-1 with RFP.Δ410 and processed for microscopic imaging 42 hr post transfection. Co-localization of GFP and RFP Δ410 was shown in the columns marked as “Merge” along with high mag “insets”. B & C. Jurkats were transfected and then infected with 50,000 cpm RT count of VSV-G pseudotyped NL4-3.Luc(env) viruses. The cells were harvested for Western blot analysis (B) and for cell surface staining for MHC I followed by flow cytometry analysis (C) 48 hr post infection. Negative and positive MHC I staining were performed in mock transfected cells. Only GFP-positive cells that were infected with VSV-G pseudotyped NL4-3.Luc(env) viruses were gated for MHC I expression. D. Human PBMC was cultured in the presence of PHA (3 μg/ml) and IL-2 (50 units/ml) for 4 days, transfected with GFP or GFP.Δ410 expression plasmid, infected with VSV-G pseudotyped NL4-3.Luc(env) viruses for 6 hr, and harvested for Western blot analysis. B & D. *, degraded GFP.Δ 410. All data were representative of three independent experiments.