Entry of human papillomavirus type 16 by actin-dependent, clathrin- and lipid raft-independent endocytosis

Abstract

Infectious endocytosis of incoming human papillomavirus type 16 (HPV-16), the main etiological agent of cervical cancer, is poorly characterized in terms of cellular requirements and pathways. Conflicting reports attribute HPV-16 entry to clathrin-dependent and -independent mechanisms. To comprehensively describe the cell biological features of HPV-16 entry into human epithelial cells, we compared HPV-16 pseudovirion (PsV) infection in the context of cell perturbations (drug inhibition, siRNA silencing, overexpression of dominant mutants) to five other viruses (influenza A virus, Semliki Forest virus, simian virus 40, vesicular stomatitis virus, and vaccinia virus) with defined endocytic requirements. Our analysis included infection data, i.e. GFP expression after plasmid delivery by HPV-16 PsV, and endocytosis assays in combination with electron, immunofluorescence, and video microscopy. The results indicated that HPV-16 entry into HeLa and HaCaT cells was clathrin-, caveolin-, cholesterol- and dynamin-independent. The virus made use of a potentially novel ligand-induced endocytic pathway related to macropinocytosis. This pathway was distinct from classical macropinocytosis in regards to vesicle size, cholesterol-sensitivity, and GTPase requirements, but similar in respect to the need for tyrosine kinase signaling, actin dynamics, Na⁺/H⁺ exchangers, PAK-1 and PKC. After internalization the virus was transported to late endosomes and/or endolysosomes, and activated through exposure to low pH.

Figure 1. HPV-16 internalization events are fast and occur asynchronously over several hours.

To test the average internalization kinetics of infectious HPV-16 PsV, HeLa (A) or HaCaT (B) cells were infected with HPV-16 PsV at MOIs of 0.1 vs. 10 transducing units/cell, i.e. about 10 vs. 1,000 particles/cell (black vs. white circles, respectively). Virus was bound to cells at 4°C, warmed to 37°C and at indicated times post warming, i.e. post infection (p.i.), cells were treated with pH 10.5 for 1 min. to inactivate extracellular virus. 48 h p.i. cells were analyzed for GFP expression from the PsV reporter plasmid by flow cytometry. The amount of GFP expressing cells relative to untreated control is given as relative internalization in %. (C) To test the internalization kinetics of single particles, we infected HeLa cells with HPV-16 PsV labeled with AF488 (5,000 particles/cell, HPV-16 AF488, green). Cells were transfected 16 h prior to infection with a construct expressing clathrin light chain-mRFP (red) to investigate any potential cointernalization. Depicted are TIRFM images of an uninfected (C, above) and infected (C, below) cell. Kymographs to the right of the images depict the formation and internalization of a CCP or virus (encircled in the images, also shown in Video S1), respectively. Internalization was defined as the loss of fluorescent signal. The time span required for CCP formation or virus internalization was determined from the initial detection of fluorescence to complete loss of fluorescence for CCPs, or from the moment of confinement to loss of fluorescence for HPV-16, respectively. The results were quantified manually and plotted (C, far below).

Figure 2. HPV-16 entry is clathrin-, caveolin-, cholesterol- and flotillin-independent.

(A) HeLa cells were transfected with siRNA oligos directed against the AP2μ-subunit (AP2), clathrin heavy chain (CHC), or as control against luciferase. Cells were infected with HPV-16 PsV or SFV and infection was scored 24 or 6 h p.i. by flow cytometry, respectively. Depicted are the mean infection percentages relative to the luciferase control. (B) HeLa cells were pretreated for 30 min with chlorpromazine in the indicated concentrations. Cells were infected with HPV-16 PsV (black), SFV (grey), or SV40 (white) in the presence of inhibitor. Inhibitors were exchanged for HPV-16 and SV40 with NH₄Cl or DTT, respectively. Infection was scored by automated microscopy and image analysis. Depicted are results normalized to solvent treated control cells. See also Figure S2, and material and methods. (C) HeLa cells were pretreated overnight with nystatin/progesterone in the indicated concentrations. Cells were infected with HPV-16 PsV (black), SFV (grey), or SV40 (white) and analyzed as in (B). (D) HeLa cells were transfected with siRNA oligos directed against caveolin-1 or with the AllStar negative control (Qiagen). 48 h after siRNA transfection cells were infected with HPV-16 PsV (black) or VSV (grey). Infection was scored as in (B) and depicted relative to cells transfected with the AllStar negative control. (E) Mouse embryonic fibroblasts from caveolin-1 knockout mice or parental mice were infected with HPV-16 PsV at different MOI. Infection was scored by flow cytometry as described in material and methods. Depicted are infection percentages. (F) HeLa cells were transfected with siRNA oligos directed against flotillin-1 or −2 or with the AllStar negative control (Qiagen). 48 h after siRNA transfection cells were infected with HPV-16 PsV (black) or SV40 (white). Infection was scored as in (C) and depicted relative to cells transfected with the AllStar negative control. (G, H) Merge images of a single cLSM slice of caveolin-1 (J, red) or flotillin-1 (K, red) HPV-16 AF488 (green) fluorescence of HeLa cells infected for 8 h. Boxed regions were enlarged as insets. (J) HeLa cells were infected with HPV-16 AF488 for 2 or 8 h, fixed with paraformaldehyde and immunostained for caveolin-1 or flotillin-1. Samples were analyzed by confocal laser scanning microscopy (cLSM) and the colocalization indices were determined as described in material and methods. Asterisk indicates a significant Costes P-value.

Figure 3. HPV-16 entry is dynamin-2-independent.

HeLa cells were transfected with constructs expressing wildtype (WT, black) or the DN K44A (grey) dynamin-2-GFP. 24 h post transfection cells were infected with HPV-16 PsV dsRed (A), SFV (B), or SV40 (C), infection was scored by flow cytometry as described in Figure S2B in relation to the level of overexpressed dynamin-2 (low, medium high). Depicted are results normalized to GFP expressing control cells. (D) HeLa cells were pretreated for 30 min with dynasore in the indicated concentrations. Cells were infected with HPV-16 PsV (black), SFV (grey), or SV40 (white) and analyzed as in Figure 2C.

Figure 4. HPV-16 entry requires actin dynamics but not Rho-GTPase or Arf6 signaling.

HeLa cells were pretreated for 30 min (A, B, G) or overnight (C) with cytochalasin D (A), jasplakinolide (B), C3 toxin/toxin B (C), and EIPA (G) in the indicated concentrations. Cells were infected with HPV-16 PsV (black), SFV (grey), or SV40 (white) and analyzed as in Figure 2C. Alternatively, HeLa cells were transfected with Cdc42, Rac1, RhoA (D, E), or Arf6 (F) fused to mRFP. Either the wildtype (WT, black), the DN (grey), or the CA (white) mutant were used. 24 h post transfection cells were infected with HPV-16 PsV (D) or VV (E). Infection was scored by flow cytometry as in Figure 3A–C. Depicted are results normalized to mRFP expressing control cells.

Figure 5. Signaling events required during HPV-16 entry.

(A–H) HeLa cells were pretreated for 30 min with genistein (A), H-7 (B), okadaic acid or orthovanadate (C), iressa (D), calphostin C (E), wortmannin (F), PI-103 (G), or IPA-3 (H) in the indicated concentrations. (A–C, E–G) Cells were infected with HPV-16 PsV (black), SFV (grey), or SV40 (white) and analyzed as in Figure 2C. (D) Cells were infected with HPV-16 PsV (black), SFV (grey), or VV (white). Infection was scored by automated microscopy and image analysis as in Figure 2C (HPV-16, VV) or by flow cytometry (SFV). (H) Cells were infected with HPV-16 PsV (black), SV40 (grey) or VV (white). Infection was scored by automated microscopy and image analysis as in Figure 2C (HPV-16, VV) or by flow cytometry (SV40). (I) HeLa cells were transfected with siRNA oligos directed against PAK-1 or with the AllStar Negative control (Qiagen). 48 h after siRNA transfection cells were infected with HPV-16 PsV (black), VSV (grey), and VV (white). Infection was scored as in Figure 2C and depicted relative to cells transfected with the AllStar negative control.

Figure 6. HPV-16 cell binding and internalization.

(A, B) HeLa cells were seeded in optical bottom 96-well plates to result in a confluent monolayer at the day of experimentation. Cells were pretreated with chlorate (A, overnight) or further small compound inhibitors (B, 30 min) in the indicated concentrations. Subsequently, HPV-16 AF488 (A, B) or HPV-16 AF488 preincubated with the indicated concentrations of heparin (A) was bound to cells for 1 h, after which cells were fixed with paraformaldehyde. Binding of HPV-16 to cells was analyzed by determining cell associated fluorescence in a sensitive plate reader. The cell associated fluorescence is depicted relative to solvent treated control cells. (C) HeLa cells were pretreated with small compound inhibitors (concentrations as in A, B). Subsequently, cells were infected with HPV-16 PsV in the presence of inhibitor. Cells were treated either treated with high pH buffer at 12 h p.i. to inactivate extracellular virus and infection was continued in the absence of inhbitor (infectious internalization) or the inhibitor was washed out and replaced by NH₄Cl. (infection). Note that orthovanadate, bafilomycin, and nocodazole inhibit infection but not internalization. The amount of GFP expressing cells relative to solvent treated control cells is given as relative internalization in %. (D) HeLa cells were pretreated with small compound inhibitors for 30 min (concentrations as in A, B). Subsequently, cells were infected with HPV-16 AF594 in the presence of inhibitor. At 6 h p.i., trypan blue (0.008%, w/v) was infused to quench the fluorescence of extracellular virus. Images depict the same cell prior and after trypan blue infusion.

Figure 7. Ultrastructural morphology of HPV-16 internalization.

HeLa (A–G) or HaCaT (H–K) cells were infected with HPV-16 PsV (10–20,000 particles/cell) for different times prior to fixation by glutaraldehyde. Cells were either untreated (A–F; H–J) or treated with cytochalasin D (10 µg/ml) throughout infection (G, K). Cells were processed for thin section electron microscopy according to standard procedures. Depicted are representative micrographs for 2–4 h p.i. (A–F; H–J) or 6 h p.i. (G, K). Scale bars represent 100 nm.

Figure 8. Kinetics of HPV-16 acid-activation during entry.

(A) HeLa or HaCaT cells were infected with HPV-16 with different multiplicities of infection (MOI, in transducing units/cell). At different times p.i. NH₄Cl (straight lines) or Bafilomycin A1 (striated line) was added, and infection was continued in the presence of the drug. (B) HeLa were infected with HPV-16 at MOI 0.1 (10 particles/cell) in the presence of NH₄Cl (straight line). At different times p.i., the drug was washed out as indicated, and infection was continued in the absence of the drugs. (C) HeLa were infected with HPV-16 as in (B) in the presence of NH₄Cl. 8 h p.i., the drug was washed out, and infection was continued in the absence of the drug for indicated times, after which NH₄Cl was re-added and infection was continued in the presence of the drug, thus creating a time window of drug absence. (A–C) Infection was scored 48 h p.i by flow cytometry. Depicted are results normalized to untreated control cells.

Figure 9. Intracellular vesicular transport of HPV-16 leads virus particles to the late endosomal/lysosomal compartments.

(A) HeLa cells were pretreated for 30 min with nocodazole in the indicated concentrations. Cells were infected with HPV-16 PsV (black), SFV (grey), or SV40 (white) and analyzed as in Figure 2C. (B, C) HeLa cells were transfected with Rab5 (B) or Rab7 (C) fused to a mRFP tag. Either the wildtype (WT, black), the DN (grey), or the CA (white) mutant were used. 24 h post transfection, cells were infected with HPV-16 PsV, SFV, VSV, or IAV as indicated. Infection was scored by flow cytometry as in Figure 3A–C. Depicted are results normalized to mRFP expressing control cells. (D) HeLa cells were infected with HPV-16 AF488 for 2 or 8 h, in the presence (white) or absence (black) of 20 mM NH₄Cl, fixed with paraformaldehyde and immunostained for EEA1, LAMP-1, or giantin. Samples were analyzed by confocal laser scanning microscopy (cLSM) and the relative amount of the colocalization index was determined as described in material and methods. Asterisk indicates a significant Costes P-value. (E–G) Merge images of a single cLSM slice of EEA1 (E), LAMP-1 (F), or giantin (G) (red) and HPV-16 AF488 (green) fluorescence of HeLa cells infected for 8 h.

Figure 10. Ultrastructural morphology of HPV-16 intracellular trafficking.

HeLa cells were infected with HPV-16 PsV (10–20,000 particles/cell) for 12 h (A, B) or 18 h (C, D) prior to fixation by glutaraldehyde. Cells were processed for thin section electron microscopy according to standard procedures. Scale bars represent 250 nm.