Functional Evidence for the Involvement of Microtubules and Dynein Motor Complexes in TRIM5α-Mediated Restriction of Retroviruses

  1. Lionel Berthouxa
  1. aLaboratory of Retrovirology, Department of Medical Biology and BioMed Group, Université du Québec à Trois-Rivières, Trois-Rivières, Québec, Canada
  2. bHIV-1 RNA Trafficking Laboratory, Lady Davis Institute at the Jewish General Hospital and Department of Medicine, McGill University, Montréal, Québec, Canada
  3. cCentre de Recherche en Infectiologie, Centre de Recherche du CHU de l'Université Laval, Québec, Canada
  1. W. I. Sundquist, Editor
  1. FIG 1
    View larger version:
    FIG 1

    Drug concentration-dependent enhancement of permissiveness to infection by TRIM5α-restricted retroviral vectors in human and simian cells. HeLa cells (A) and FRhK-4 cells (B) were infected with single doses of the indicated viral vectors in the absence or presence of increasing amounts of nocodazole or paclitaxel (taxol) for 16 h. The amounts of viruses used were adjusted to obtain in the vicinity of 0.1 to 1% of infected cells in the absence of the drug and yielded 0.09% (N-MLVGFP), 0.38% (B-MLVGFP), 0.85% (HIV-1NL43-GFP), and 1.23% (SIVmac-GFP) infected cells. The percentages of infected (GFP-positive) cells were determined by flow cytometry at 2 days postinfection, and the results are presented as the fold changes in infectivity relative to the relevant untreated controls.

  2. FIG 2
    View larger version:
    FIG 2

    Pharmacological disruption of microtubules decreases endogenous TRIM5α-mediated retroviral restriction. (A to C) Effect of nocodazole (noc) and paclitaxel (txl) on restriction. Human HeLa cells (A) or macaque FRhK-4 cells (B and C) were infected with multiples doses of N-MLVGFP, B-MLVGFP, HIV-1NL43-GFP, or SIVmac-GFP as indicated. In panel C, FRhK-4 cells were infected with WT HIV-1CMV-GFP or with the CA-G89V mutant of this vector. Infections were performed for 16 h and in the absence of drug or in the presence of either nocodazole or paclitaxel. Nocodazole was used at 0.1 μM in HeLa cells and at 6 μM in FRhK-4 cells, and paclitaxel was used at 0.1 μM in HeLa cells and at 2 μM at FRhK-4 cells. The x axis in each graph represents the amounts of virus used expressed in infectious units (IU) based on infectious titers calculated for each virus in permissive feline CRFK cells. Infected (GFP-expressing) cells were detected by flow cytometry at 2 days postinfection. (D) IF microscopy analysis of microtubules in treated cells. HeLa and FRhK-4 cells were transfected with GFP-α-tubulin and 2 days later were subjected to 2-h drug treatments using the concentrations described above and then fixed. GFP fluorescence was observed by IF microscopy, along with DNA, which was stained using DAPI (blue staining). taxol, paclitaxel. The scale bars on the images represent 10 μm. A representative image from each condition is presented.

  3. FIG 3
    View larger version:
    FIG 3

    Pharmacological disruption of dynein motor function decreases endogenous TRIM5α-mediated retroviral restriction. (A and B) Dose-dependent effect of EHNA on restriction. Human HeLa cells (A) and macaque FRhK-4 cells (B) were infected for 16 h with a single dose of the indicated viruses in the presence of increasing concentrations of EHNA. The amounts of viruses used were the same as in Fig. 1. Infected (GFP-expressing) cells were detected by flow cytometry at 2 days postinfection. (C and D) Virus dose-dependent effect of EHNA on restriction. Human HeLa cells (C) and macaque FRhK-4 cells (D) were infected for 16 h with multiples doses of the indicated viruses in the presence or absence of EHNA at 600 μM (HeLa) or 1.2 mM (FRhK-4). The x axis represents the amounts of virus used expressed in IU based on the infectious titers calculated for each virus in permissive feline CRFK cells. Infected (GFP-expressing) cells were detected by flow cytometry 2 days later. (E) IF microscopy analysis of LAMP-1 distribution. HeLa and FRhK-4 cells were treated for 2 h with EHNA or left untreated and then fixed and stained for the lysosomal marker LAMP-1 (red) and DNA (blue). Cell edges are outlined, and examples of localization shift caused by impaired dynein function are indicated by white arrows. A representative image from each condition is presented. The scale bars in the image panels represent 10 μm.

  4. FIG 4
    View larger version:
    FIG 4

    DHC depletion decreases TRIM5α-mediated retroviral restriction and has no effect on cells treated with nocodazole or paclitaxel. (A) Western blot analysis of DHC expression in HeLa cells 48 h after transfection of the indicated siRNAs. Actin was analyzed as a loading control. (B) IF microscopy analysis of LAMP-1. LAMP-1 was stained 72 h after siRNAs transfection (red), and DNA was stained using DAPI (blue). The cell edges are outlined, and a LAMP-1 distribution shift caused by impaired dynein function is indicated by a white arrow. A representative image is shown for each condition. The scale bars in the image panels represent 10 μm. (C) Effect of DHC knockdown on restriction. Human HeLa cells were transfected with siRNAs against DHC (siDHC) or against luciferase (siLuc) as a control. After 72 h, the cells were infected for 16 h with multiple doses of N-or B-MLVGFP. Infected cells were detected by flow cytometry 2 days after infection. The x axis shows the amounts of virus used expressed in CRFK infectious units. (D and E) Effect of combining siRNA transfections with 0.25 μM nocodazole (D) or 0.1 μM paclitaxel (taxol) (E). Infections were performed and analyzed as in panel C, and panels C to E all share the symbol key.

  5. FIG 5
    View larger version:
    FIG 5

    Inhibition of HIV-1 restriction by rhTRIM5α exogenously expressed in HeLa cells. (A) Effect of nocodazole (noc) and paclitaxel (txl) on the restriction of HIV-1 vectors at multiple virus doses. HeLa cells stably expressing FLAG-tagged rhTRIM5α (rhT5α) were infected for 16 h with multiple amounts of the indicated vectors in the presence or absence of 0.1 μM nocodazole or paclitaxel. Virus doses were normalized according to titers in CRFK cells. The percentages of infected cells were analyzed at 2 days postinfection. (B) Effect of DHC depletion on the restriction of HIV-1NL43-GFP. HeLa cells stably expressing rhTRIM5α (rhT5α) were transfected with siRNAs targeting DHC (siDHC) or, as a control, luciferase (siLuc). The efficiency of DHC knockdown was similar to that in Fig. 4A (not shown). Cells were then infected for 16 h with HIV-1NL43-GFP or SIVmac-GFP, and the percentages of infected cells were determined 2 days later. (C and D) Effect of nocodazole or paclitaxel treatment on the infectivity of HIV-1CMV-GFP in cells expressing HIV-1 restrictive or nonrestrictive TRIM5α alleles. HeLa cells transduced as indicated with FLAG-tagged huTRIM5α (huT5α), FLAG-tagged rhTRIM5α (rhT5α), or empty vector were treated or not with 0.1 μM nocodazole (C) or 0.1 μM paclitaxel (D). Cells were simultaneously infected in triplicates with single doses of HIV-1CMV-GFP adjusted to yield ∼1 to 10% of infected cells in the absence of the drug. Infectivities obtained in the absence of drugs are indicated in the figure as follows: (C) HeLa [vector], 1.23% ± 0.09%; HeLa [huT5α], 2.14 ± 0.81%; and HeLa [rhT5α], 0.82% ± 0.06%; and (D) HeLa [vector], 6.84% ± 0.14%; HeLa [huT5α], 6.58% ± 0.27%; and HeLa [rhT5α], 2.79% ± 0.19%. Infected cells were detected by flow cytometry at 2 days postinfection, and the results are presented as the fold changes relative to the relevant untreated controls (relative change in infectivity). (E) Effect of DHC depletion on restriction. HeLa cells stably expressing huTRIM5α or rhTRIM5α were transfected with siRNAs against DHC (siDHC) or against luciferase (siLuc) as a control and infected 72 h later in triplicates with single doses of the viral vector HIV-1CMV-GFP for 16 h. Virus doses were adjusted to obtain ∼1% of infected cells for the siLuc control and yielded 1.61% ± 0.29% infected HeLa [huT5α] cells and 0.30% ± 0.08% infected HeLa [rhT5α] cells. Infected cells were detected by flow cytometry 2 days postinfection and results are presented as the fold changes relative to the relevant untreated controls (relative change in infectivity). ***, P < 0.0001 (Student t test). (F) Western blotting of cells stably transduced with FLAG-tagged huTRIM5α and rhTRIM5α, with actin as a loading control.

  6. FIG 6
    View larger version:
    FIG 6

    Nocodazole and paclitaxel inhibit TRIM5α-mediated restriction of nonpseudotyped HIV-1. (A) MAGI cells transduced with FLAG-tagged huTRIM5α (huT5α), FLAG-tagged rhTRIM5α (rhT5α), or the empty vector were treated or not with 0.25 μM nocodazole (noc) or 0.1 μM paclitaxel (txl). As a control, cells were also infected in the presence of the fusion inhibitor heparin (20 μg/ml). Cells were infected in triplicates with single doses of HIV-1NL43-IRES-GFP. The amounts of virus used were adjusted to obtain ∼0.1% of infected cells in the absence of the drug and yielded 0.22% ± 0.025% infected MAGI [vector] cells, 0.16% ± 0.03% infected MAGI [huT5α] cells, and 0.043% ± 0.006% infected MAGI [rhT5α] cells. Supernatants were replaced with fresh medium containing heparin at 16 h postinfection in order to prevent reinfections. Infected (GFP-positive) cells were detected by flow cytometry 2 days postinfection, and the results are presented as the fold changes in the percentages of infected cells relative to the relevant untreated controls (relative change in infectivity). * and **, P = 0.0134 and P = 0.0068, respectively (Student t test). (B) Western blot analysis of FLAG-huTRIM5α and FLAG-rhTRIM5α expression in stably transduced MAGI cells. Actin was analyzed as a loading control.

  7. FIG 7
    View larger version:
    FIG 7

    Effect of treatments on TRIM5α stability and cell viability. (A) Effect on rhTRIM5α protein levels. HeLa cells stably expressing FLAG-tagged rhTRIM5α were treated with nocodazole (0.1 μM), paclitaxel (labeled “taxol” in panel A; 0.1 μM), EHNA (600 μM) and concomitantly treated with cycloheximide. Alternatively, cells were transfected with luciferase- or DHC-targeting siRNAs like before and treated with cycloheximide 72 h later. The efficiency of DHC knockdown was similar to that in Fig. 4A (not shown). Protein lysates were prepared at the indicated time points following the beginning of drug treatments. Lysates were subjected to Western blotting to detect FLAG-rhTRIM5α and actin. The results of one representative experiment of three independent experiments are shown. (B) Analysis of FLAG-rhTRIM5α protein turnover. FLAG-rhTRIM5α bands detected by Western blotting were quantified by densitometry and normalized to actin levels and then to the value obtained at the “0” time point; ctl, control; noc, nocodazole; txl, paclitaxel. Average data from three independent experiments with standard deviations are shown. The P values indicated on the graphs were calculated by using a Student t test for the three conditions at which an effect could be observed: EHNA treatment (1 h) and DHC siRNA transfection (3 and 4.5 h). (C) Cell viability assay. HeLa and FRhK-4 cells were treated with multiple concentrations of the indicated drugs for 16 h and cell viability was then determined using the XTT assay and normalized using the value obtained in the absence of drug as a reference. Vertical lines indicate the drug concentrations used in the present study, for each combination of drug and cell line. taxol, paclitaxel.

  8. FIG 8
    View larger version:
    FIG 8

    DHC depletion, paclitaxel treatment, and nocodazole treatment increase the stability of HIV-1 cores in permissive or restrictive conditions. (A) HeLa cells stably expressing rhTRIM5α (rhT5α) or transduced with the empty vector were transfected with the indicated siRNAs and 72 h later infected with HIV-1NL43-GFP or an entry-incompetent version of this vector lacking the VSV G envelope protein. Supernatants were replaced by fresh medium 2 h postinfection, and cells were lysed after an additional 4 h of incubation. Precleared lysates were layered on sucrose cushions, and particulate viral CA cores were pelleted by ultracentrifugation. Proteins in postcentrifugation pellets (PEL), supernatants (SUP), and whole-cell lysates (WCL) were analyzed by Western blotting to detect CA (p24). Bands corresponding to p24 were quantified by densitometry and plotted as pellet/WCL and pellet/supernatant ratios, relative to the nonrestrictive untreated control. Average data from three independent experiments with standard deviations are shown. The fold changes and P values (calculated using a Student t test) are shown for specific pairs of data. (B) Same as panel A, but the cells were treated or not with 0.1 μM paclitaxel (taxol) during the 2 h of infection and during the 4 h of incubation. (C) Same as panel B, but cells were treated or not with 0.1 μM nocodazole (noc).

  9. FIG 9
    View larger version:
    FIG 9

    Influence of DHC depletion and nocodazole treatment on the localization and dynamics of rhTRIM5α cytoplasmic bodies (CBs). (A and B) IF microscopy. HeLa cells transduced with FLAG-tagged rhT5α and control cells transduced with the empty vector were seeded on glass coverslips and 24 h later treated or not with 2 μM nocodazole for 6 h (A) or transfected with 40 nM siRNAs targeting DHC or Luc and 48 h later seeded on glass coverslips and incubated for an additional 24 h (B). Cells were fixed and immunostained for FLAG (red). DNA was stained with Hoechst 33342 (blue). Representative images are shown. (C) Sizes of CBs. All CBs were outlined in cells from a minimum of five randomly chosen fields, and their area was calculated using an image analysis software (AxioVision). Red bars show the standard errors of the mean (SEM). ***, P ≤ 0.0001 (Student t test). (D) The relative localization of all TRIM5α CBs from a minimum of five randomly chosen cells was calculated using the formula x/(x + y), where x is the shortest distance to the nucleus, and y is the shortest distance to the cell's edge. SEM are shown as red bars. ***, P ≤ 0.0001. (E to G) Colocalization of rhT5α CBs and microtubules. (E) HeLa cells transduced with FLAG-tagged rhTRIM5α were transfected with siRNAs targeting Luc or DHC, and 24 h later the cells were additionally transfected with a plasmid expressing GFP–α-tubulin. The next day, cells were seeded on glass coverslips and grown for 24 h. Cells were fixed and immunostained for FLAG (red) and DNA (blue). (F) Same as panel E, except that the cells were transduced with the empty vector and transfected with luciferase-targeting siRNA. Representative images are shown. Boxes in the top right corner show enlarged regions outlined on the images, and white arrows show examples of GFP-tubulin/FLAG-TRIM5α colocalizations. (G) Events of colocalization were quantified as a percentage of the total CBs in 10 randomly chosen fields. Bars represent the mean values from analyzed cells with SEM (**, P < 0.005).

| Table of Contents

This Article

  1. Accepted manuscript posted online 5 March 2014, doi: 10.1128/JVI.03717-13 J. Virol. vol. 88 no. 10 5661-5676
  1. AbstractFree
  2. » Figures
  3. Full Text
  4. PDF

Article Usage Stats

  1. Article Usage Statistics

current issue

    August 2017, volume 91, issue 15
  1. Current Issue
  1. Alert me to new issues of JVI