Cell Surface Vimentin Is an Attachment Receptor for Enterovirus 71

  1. Po Tiena
  1. aCenter for Molecular Virology, CAS Key Laboratory of Pathogenic Microbiology and Immunology, Institute of Microbiology, Chinese Academy of Sciences, Beijing, People's Republic of China
  2. bAnhui University, Anhui, People's Republic of China
  3. cUniversity of the Chinese Academy of Sciences, Beijing, People's Republic of China
  1. K. Kirkegaard, Editor
  1. FIG 1

    Experiments showing specific interaction between EV71 and cellular vimentin. (A) Detection of specific interactions between EV71 BrCr virus particles and cellular vimentin in uninfected U251 cell lysates by immunoprecipitation assays and Western blotting performed as described in Materials and Methods. The figure shows a Western blot of the following precipitated protein samples using anti-EV71 VP1 and antivimentin antibodies: input, untreated cell lysate; control, cell lysate incubated with agarose beads treated with purified EV71 particles; anti-EV71, cell lysate incubated with anti-EV71 VP1 monoclonal antibody-conjugated agarose beads treated with purified EV71 particles; IgG, cell lysate incubated with mouse IgG-conjugated agarose beads treated with purified EV71 particles. (B) Detection of specific binding of EV71 with vimentin in EV71-infected U251 cells using coimmunoprecipitation assays. U251 cells were infected with EV71 and lysates prepared as described in Materials and Methods. Infected cell lysates were then incubated with either mouse IgG or vimentin monoclonal antibody (antivimentin)-conjugated agarose beads for 2 h at 4°C. The precipitated proteins were blotted with anti-EV71 VP1 and antivimentin antibodies. Input, EV71 and vimentin markers; control, proteins from cell lysate incubated with IgG-conjugated agarose beads; antivimentin, proteins from cell lysate incubated with vimentin monoclonal antibody-conjugated agarose beads. (C) Analysis of the specific interactions between various strains of EV71 viruses and U251 cellular vimentin by immunoprecipitation assays. CA16, EV71 BrCr, EV71 Hubei 09, and EV71 HeN 09, cell lysates incubated with antivimentin monoclonal antibody-conjugated agarose beads treated with purified CA16, EV71 BrCr, EV71 Hubei 09, and EV71 HeN 09 particles, respectively; control, cell lysate incubated with agarose beads treated with purified EV71 BrCr particles.

  2. FIG 2

    (A) Detection of vimentin expression on the cell surface of U251, RD, Vero, and HeLa cells by flow cytometry. Cells were fixed and incubated with either mouse IgG (black line) or antibody to vimentin (red line). The cells were then incubated with the fluorescent secondary antibody and subjected to flow cytometry analysis as described in Materials and Methods. y axis (counts) = cell counts; x axis (FL1-H) = fluorescence density. (B) Analysis of the cell surface distribution of vimentin and cell surface-bound EV71 by indirect immunofluorescence in U251 cells. Cells were infected with EV71 BrCr (+EV71) at 4°C for 1 h and then stained with specific antibody to either EV71 (red) or vimentin (green) and analyzed by confocal fluorescence microscopy. Bar = 20 μm. (C) Analysis of the distribution of cell vimentin by indirect immunofluorescence in U251 cells (−EV71). Cells were fixed and permeabilized as described in Materials and Methods. Cells were then stained with antibodies to vimentin (Vim; green fluorescence) and EV71 (red florescence) and subjected to confocal microscopy analysis. An overlay of the vimentin and EV71 florescence is also shown (merged). Cell morphology (phase) was assessed by light microscopy.

  3. FIG 3

    Experiments showing the interaction between EV71 VP1 protein and vimentin. (A) Detection of binding between VP1 and vimentin using GST pulldown assays and Western blotting. The eluates obtained as described in Materials and Methods were blotted with antivimentin and anti-GST antibodies, and the resultant VP1 protein band observed is indicated. Input, untreated cell lysate; control, U251 cell lysate incubated with glutathione-Sepharose beads; VP1, 3C, and GST, U251 cell lysate incubated with VP1-GST, 3C-GST, or GST precombined glutathione-Sepharose beads, respectively. (B) Pulldown assays and Western blot analysis showing the interaction between VP1 and vimentin. GST or VP1-GST protein was incubated with antivimentin monoclonal antibody-conjugated agarose beads that were preincubated with vimentin protein. The precipitated proteins were blotted with antibodies to either vimentin or GST, and the resultant VP1-GST protein band observed is indicated. (C) Analysis of the binding of VP1 to vimentin by coimmunoprecipitation and Western blotting. U251 cells were transfected with a plasmid expressing VP1, lysed, and coimmunoprecipitated with vimentin antibodies (anti-Vim) or mouse IgG. The precipitated proteins were blotted with antibodies to vimentin and EV71 VP1. Input, cell lysate; control, agarose beads incubated either with no antibodies or with IgG; anti-Vim and mouse IgG, agarose beads incubated with vimentin antibody and mouse IgG, respectively. (D) Coimmunoprecipitation and Western blot analysis of the binding of vimentin to VP1, VP2, and VP3. U251 cells were transfected with plasmids expressing either VP1, VP2, or VP3. Coimmunoprecipitation was performed with Flag antibody. The precipitated proteins were analyzed by Western blotting using antibodies to vimentin and Flag. Control, precipitated proteins from cells with no plasmid transfection; VP1, VP2, VP3, precipitated proteins from cells transfected with VP1, VP2, VP3 plasmids, respectively. The figure shows only EV71 VP1 interacted with vimentin and not VP2 or VP3.

  4. FIG 4

    Experiments on the role of vimentin in the attachment of EV71 to U251 cells. (A) Analysis of the binding of EV71 or CA16 to U251 cells using competitive inhibition assays and Western blotting. Cells were infected with EV71 or CA16 at 4°C for 1 h after the virus inoculum was preincubated with vimentin or BSA at the concentrations indicated (μg ml−1). Control, cells infected with untreated virus. Infected cell lysates were subjected to Western blot analysis using antibodies to EV71, CA16, and β-actin (internal control). The figure shows the inhibition of EV71 binding but not CA16 binding after pretreatment of the virus inoculum with vimentin. (B) Flow cytometry analysis of the binding of EV71 to U251 cells after pretreatment of the virus inoculum with increasing concentrations of vimentin (5 μg ml−1, black heavy line; 20 μg ml−1, black dotted line) or BSA (20 μg ml−1, gray hairline). Black hairline (mock), infected cells with no vimentin or BSA added. Gray filled line (control), cells with no EV71 infection. x axis (FL1-H) = fluorescence density. (C) Analysis of the binding of EV71 to U251 cells using quantitative RT-PCR. Cells were infected with EV71 at 4°C for 1 h after the virus inoculum was preincubated with vimentin or BSA at the concentrations indicated. Control, cells infected with untreated EV71. (D) Infectivity of vimentin-pretreated EV71 in U251 cells. EV71 was pretreated with various doses of vimentin (Vim) at 4°C for 1 h prior to infecting U251 cells. The total virus yield at 48 h postinfection was determined. Virus titer from cells infected with EV71 that had no vimentin pretreatment was used as a reference (100%) to calculate the percent reduction in 50% tissue culture infective dose (TCID50) in the vimentin-pretreated groups. (E) CPE of EV71 infections viewed under the visible light phase microscope, showing representative fields of control uninfected U251 cells (uninfected), cells infected with either untreated EV71 (EV71), vimentin-pretreated EV71 (Vim; 20 μg ml−1) or control BSA-pretreated cells (BSA; 20 μg ml−1). (F) Analysis of the influence of vimentin on the binding of EV71 to U251 cells using quantitative RT-PCR. Cells were infected with EV71 at 4°C. Vimentin (20 μg ml−1) or BSA (20 μg ml−1) was then added to the cell culture at the indicated time postinfection. All cells were harvested at 60 min postinfection, and the binding of EV71 was then analyzed using quantitative RT-PCR.

  5. FIG 5

    Analysis of the role of vimentin in the attachment of EV71 to RD, HeLa, and Vero cells. (A) Western blot analysis of the inhibition of EV71 infection after pretreating the virus inoculum with vimentin (Vim) or BSA. Cells were infected with EV71 at 4°C for 1 h, lysed, and Western blotted with antibody to EV71 and β-actin (internal control) as described in Materials and Methods. Control, cells infected with untreated virus; BSA, cells infected with BSA-pretreated virus; Vim, cells infected with vimentin-pretreated virus. (B) Analysis of the binding of EV71 to cells using quantitative RT-PCR. Control, cells infected with untreated virus; mock, cells incubated with no virus. (C) Infectivity of vimentin-pretreated EV71 in RD, HeLa, and Vero cells. EV71 was pretreated with vimentin or BSA at 4°C for 1 h prior to infecting the cells. The total virus yield at 12, 24, and 48 h postinfection at 37°C was determined. Control, cells infected with untreated virus; BSA, cells infected with BSA-treated virus; Vim, cells infected with vimentin-treated virus; mock, cells incubated with no virus.

  6. FIG 6

    Effect of pretreating host cells with vimentin antibodies on the binding of EV71 to the cells. (A) Western blot analysis of EV71 BrCr replication in U251 cells that had been preincubated with vimentin antibodies prior to infection. Cells were pretreated with the indicated concentrations of vimentin antibodies (μg ml−1) or rabbit IgG (isotype ab; 80 μg ml−1) before being infected with EV71 for 1 h at 4°C. Cells were then lysed and subjected to Western blot analysis with antibody to EV71 and β-actin (internal control). Control, cells incubated with EV71. (B) CPE of EV71 BrCr infection viewed under a visible light phase microscope, showing representative fields of the uninfected control U251 cells (uninfected), vimentin antibody-pretreated cells infected with EV71 (Vim ab [60 μg ml−1] and EV71), and rabbit IgG-pretreated cells infected with EV71 (isotype ab [60 μg ml−1] and EV71). (C) Analysis of the binding of EV71 BrCr to U251 cells using quantitative RT-PCR. Cells were pretreated with vimentin antibodies (Vim ab; 60 μg ml−1) or rabbit IgG (isotype ab; 60 μg ml−1) before infection as described above. Control, untreated cells incubated with EV71; mock, incubated cells with no EV71; isotype ab, isotype antibody-pretreated cells infected with EV71; Vim, vimentin antibody-pretreated cells infected with EV71. (D) Graphs showing virus titers in infected U251 cells and the corresponding culture medium at 0, 12, and 24 h after infection. The cells were pretreated with vimentin antibody (Vim ab; 60 μg ml−1) or rabbit IgG (isotype ab; 60 μg ml−1) before infection. Control, untreated cells infected with EV71; mock, uninfected mock-treated cells. (E) Analysis of the influence of vimentin antibody (Vim ab) and isotype antibody (isotype ab) on the binding of EV71 Hubei09 and HeN09 strains to U251 cells using quantitative RT-PCR. Control, untreated cells incubated with EV71.

  7. FIG 7

    Effect of the downregulation of vimentin expression in U251 cells on the efficiency of EV71 binding and replication in host cells. U251, C-U251, and VK-U251 are cells with no vector, cells with empty vector, and cells with the vimentin knockdown plasmid, respectively. (A) Flow cytometry analysis of cell surface vimentin expression in U251 (gray hairline), C-U251 (black hairline), and VK-U251 (gray heavy line) cells. Nonpermeabilized cells were fixed and stained with antibody specific to vimentin and subjected to flow cytometry analysis. Thick black line, VK-U251 cells stained with mouse IgG. y axis (counts) = cell counts; x axis (FL1-H) = fluorescence density. (B) Flow cytometry analysis of the binding of EV71 to VK-U251 and C-U251 cells. VK-U251 and C-U251 cells were incubated with EV71 for 1 h at 4°C and then fixed, stained with antibody to EV71, and subjected to flow cytometry. Black heavy line, VK-U251 cells with no EV71 infection; gray heavy line, C-U251 cells with no EV71 infection; gray dotted line, infected VK-U251 cells stained with EV71 antibody; gray hairline, infected U251 cells stained with EV71 antibody; black hairline, infected C-U251 cells stained with EV71 antibody. (C) Analysis of EV71 binding in U251, VK-U251, and C-U251 cells that had been infected with virus inoculum preincubated with vimentin before transfection by quantitative RT-PCR. Control, cells incubated with untreated EV71; BSA, cells infected with virus inoculum preincubated with BSA; Vim, cells infected with virus inoculum preincubated with vimentin; mock, cells mock treated with no virus infection. (D) Western blot analysis of EV71 binding and expression of SCARB2, PSGL-1, and vimentin in U251, C-U251, and VK-U251 cells. The respective cells were infected with EV71 at 4°C for 1 h and then lysed and blotted with antibodies to either EV71, SCARB2, PSGL-1, vimentin, or β-actin (internal control). (E) Analysis of CPE of EV71 infection in C-U251 and VK-U251 cells after infection at 37°C for 48 h using a visible light phase microscope. (F) Determination of virus titers in infected U251, C-U251, and VK-U251 cells and the corresponding culture supernatants at 12, 24, and 48 h after infection with EV71.

  8. FIG 8

    Effect of pretreating U251 cells with SCARB2 antibodies on the binding of EV71 to the cells. (A) Flow cytometry analysis of the cell surface expression of U251, RD, Vero, HeLa, and Jurkat cells. Cells were fixed and incubated with either mouse IgG (black line) or antibody to vimentin (gray line) and then incubated with the fluorescent secondary antibody and subjected to flow cytometry analysis. y axis (counts) = cell counts; x axis = fluorescence density. (B) Quantitative RT-PCR analysis of the binding of EV71 to VK-U251 cells that were infected after preincubation with antibodies to either vimentin (Vim ab; 60 μg ml−1), PSGL-1 (PSGL-1 ab; 60 μg ml−1), SCARB2 (SCARB2 ab; 60 μg ml−1), or isotype IgG (isotype ab; 60 μg ml−1). SCARB2 ab + Vim ab = cells preincubated with SCARB2 antibody (60 μg ml−1) for 30 min at 37°C and then incubated with vimentin antibody for 30 min at 37°C. Vim ab + SCARB2 ab = cells preincubated with vimentin antibody for 30 min at 37°C followed by incubation with SCARB2 antibody for 30 min at 37°C. Control, EV71-infected cells without preincubation; mock, mock-treated cells with no virus. (C) Quantitative RT-PCR analysis of the binding of EV71 to U251 cells that were preincubated with antibodies to either vimentin, SCARB2, or isotype IgG as described above. (D) Determination of EV71 virus titers in infected U251 cells and the corresponding culture media at 12, 24, and 48 h after infection. The cells were preincubated with vimentin antibody (Vim ab; 60 μg ml−1), SCARB2 antibody (SCARB2 ab; 60 μg ml−1), or isotype IgG (control; 60 μg ml−1) before infection. Mock, uninfected cells. (E) Determination of EV71 virus titers in infected VK-U251 cells and the corresponding culture supernatants as described for panel D. Cells were preincubated with vimentin antibody (Vim ab), SCARB2 antibody (SCARB2 ab), or isotype IgG (isotype ab) before infection. Control, EV71-infected cells without preincubation.

  9. FIG 9

    (A) Analysis of the binding of EV71 in U251 cells after pretreating the virus inoculum with BSA (20 μg ml−1), mouse FC (20 μg ml−1), PSGL-1-FC (20 μg ml−1), SCARB2 (20 μg ml−1), or/and vimentin (20 μg ml−1) by using quantitative RT-PCR. Control, cells infected with EV71. (B) Virus titers in the supernatants and infected U251 cells after pretreating the virus inoculum with BSA (20 μg ml−1), mouse FC (20 μg ml−1), PSGL-1-FC (20 μg ml−1), SCARB2 (20 μg ml−1), or/and vimentin (20 μg ml−1). Twenty-four hours postinfection, cells and supernatants were collected and frozen and thawed three times. After centrifugation for 10 min at 1,000 × g, the supernatants were recovered. A total of 200 μl of supernatants were used for the RNA extraction and quantitative PCR analysis. Mock, uninfected cells.

  10. FIG 10

    Effect of pretreating RD, HeLa, Vero, and Jurkat cells with SCARB2 antibodies on the binding of EV71 BrCr strain. (A) Quantitative RT-PCR analysis of the binding of EV71 to RD, HeLa, and Vero cells that were preincubated with antibodies to vimentin (Vim ab; 60 μg ml−1), SCARB2 (SCARB2 ab; 60 μg ml−1), or isotype IgG (isotype ab; 60 μg ml−1). Control, EV71-infected cells without preincubation; mock, uninfected cells. (B) Determination of virus titers in EV71-infected RD, HeLa, and Vero cells and the corresponding culture supernatants. The cells were preincubated with vimentin antibody, SCARB2 antibody, or isotype IgG (control) before infection. (C) Quantitative RT-PCR analysis of the effect of pretreating U251 cells with SCARB2 antibodies (SCARB2 ab; 60 μg ml−1) on the binding of EV71 Hubei 09 and HeN 09 strains to U251 cells. Control, EV71-infected cells with no preincubation; mock, uninfected cells.

  11. FIG 11

    Experiments on the effect of pretreatment of RD, HeLa, Vero, and Jurkat cells with PSGL-1 antibodies on the binding of EV71 Hubei 09 strain. (A) Analysis of the binding of EV71 to RD, HeLa, and Vero cells preincubated with antibodies to vimentin, PSGL-1, or isotype IgG (60 μg ml−1) by using quantitative RT-PCR as described in Materials and Methods. Control, cells infected with EV71. (B) Quantitative RT-PCR analysis of the effect of pretreatment of Jurkat cells with PSGL-1 antibodies on the binding of EV71. Control, cells infected with virus; mock, cells with no virus infection. (C) Virus titers in the supernatants and infected Jurkat cells that were preincubated with vimentin antibody, PSGL-1 antibody, or isotype IgG (control; 60 μg ml−1) before infection. The data are shown as mean virus titers ± SD based on three independent experiments.

  12. FIG 12

    Analysis of the role of vimentin in EV71 binding in mouse 3T3 cells. (A) Detection of vimentin expression on the cell surface of mouse 3T3 cells. Cells were fixed and incubated with either mouse IgG (black line) or antibody to vimentin (gray line). The cells were then incubated with the fluorescent secondary antibody and subjected to flow cytometry analysis as described in Materials and Methods. y axis (counts) = cell counts; x axis = fluorescence density. (B) Flow cytometry analysis of the binding of EV71 to 3T3 cells. Cells were incubated with EV71 at an MOI of 20 at 4°C for 1 h, fixed, and incubated with either mouse IgG (black line) or antibody to EV71 (gray line). (C) Quantitative RT-PCR analysis of the binding of EV71 to 3T3 cells that were preincubated with antibodies to vimentin, SCARB2, PSGL-1, or isotype IgG (60 μg ml−1) and then infected with EV71. Control, EV71-infected cells without preincubation; mock, uninfected cells. (D) Determination of virus titers in EV71-infected 3T3 cells and the corresponding culture supernatants at 12, 24, and 48 h postinfection. The cells were preincubated with vimentin antibody, SCARB2 antibody, or isotype IgG (control) before infection. Mock, uninfected cells. (E) Coimmunoprecipitation and Western blot analysis of the interaction between mouse vimentin and EV71 VP1 as described in Materials and Methods. 3T3 cells were transfected with plasmids expressing VP1. Coimmunoprecipitation was performed with Flag antibody. The precipitated proteins were analyzed by Western blotting using antibodies to vimentin 1 and Flag. Control, proteins from cell lysate incubated with agarose beads; VP1, proteins from cell lysate incubated with Flag-conjugated agarose beads; IgG, proteins from cell lysate incubated with IgG-conjugated agarose beads. (F) Detection of vimentin expression on the cell surface of mouse 3T3 (black dotted line), siC-3T3 (gray line), and siVim-3T3 (black line) cells. Cells were fixed and incubated with either mouse IgG (shaded area) or antibody to vimentin. The cells were then incubated with the fluorescent secondary antibody and subjected to flow cytometry analysis as described in Materials and Methods. (G) Analysis of the binding of EV71 to 3T3, siC-3T3, and siVim-3T3 cells using quantitative RT-PCR.

  13. FIG 13

    Identification of the domain of vimentin involved in the interaction with EV71 VP1. (A) Analysis of various truncated vimentin fragments for interaction with VP1. The His-tagged vimentin and its fragments (aa 1 to 466, 1 to 230, 231 to 466, 1 to 115, 116 to 230, 1 to 56, and 57 to 115) were subjected to a pulldown assay using the GST-VP1-coupled glutathione-Sepharose beads or GST beads as described in Materials and Methods. Proteins coprecipitating with the beads were analyzed by immunoblotting using anti-His antibody or anti-GST antibody. (B) Effect of pretreating EV71 inoculum with full-length vimentin and truncated vimentin on the binding of the virus to U251 cells. EV71 was pretreated with vimentin fragments at a concentration of 20 μg ml−1 at 37°C for 1 h prior to infecting U251 cells. The cells were then infected with EV71 at 4°C for 1 h. Quantitative RT-PCR analysis was then performed to determine the EV71 RNA levels as described in Materials and Methods. Control, cells infected with untreated EV71.

| Table of Contents