Human Papillomavirus Infectious Entry and Trafficking Is a Rapid Process

ABSTRACT

Previous studies have indicated that human papillomavirus (HPV) infectious entry is slow, requiring many hours after initial infection for the virus to gain entry into the nucleus. However, intracellular transport pathways typically are very rapid, and in the context of a natural HPV infection in a wounded epithelium, such slow intracellular transport would seem to be at odds with a normal viral infection. Using synchronized cell populations, we show that HPV trafficking can be a rapid process. In cells that are infected in the late S-early G₂/M phase of the cell cycle, HPV16 pseudovirion (PsV) reporter DNA gene expression is detectable by 8 h postinfection. Likewise, reporter DNA can be visualized within the nucleus in conjunction with PML nuclear bodies 1 h to 2 h postinfection in cells that are infected with PsVs just prior to mitotic entry. This demonstrates that endosomal trafficking of HPV is rapid, with mitosis being the main restriction on nuclear entry.

IMPORTANCE HPV infectious entry appears to be slow and requires mitosis to occur before the incoming viral DNA can access the nucleus. In this study, we show that HPV trafficking in the cell actually is very rapid. This demonstrates that in the context of a normal virus infection, the cell cycle state will have a major influence on the time it takes for an incoming virus to enter the nucleus and initiate viral gene expression.

INTRODUCTION

Papillomaviruses (PV) are small, nonenveloped DNA viruses that infect epithelial cells. Their icosahedral capsid is formed by two structural proteins: L1 and L2. PVs have oncogenic potential and are associated with multiple human cancers, including cervical cancer, other anogenital cancers, and a significant number of head and neck tumors (1). The HPV life cycle is intimately linked to epithelial differentiation of keratinocytes, with the virus believed to enter the undifferentiated proliferating basal compartment of the epithelium through microtraumas in the skin. As the cells differentiate, the viral genome is amplified, and ultimately new infectious virus particles are assembled in the nucleus of differentiated spinous and granular keratinocytes (1).

PVs enter basal cells via endocytosis, but the precise mechanism appears diverse and is dependent on the virus type (2, 3). Virus capsids then disassemble in the late endosomes and/or lysosomes in a pH-dependent manner. L2 interacts with components of the cellular sorting machinery, such as sorting nexin 17, which is required for the lysosomal escape of the L2-DNA complex (4). L2 then targets the viral DNA to the perinuclear region of the cell through a pathway that requires Dynein-mediated transport (5) and endocytic retromer components (6), where the L2-DNA complex eventually accumulates in the trans-Golgi network (7, 8). HPV entry into the nucleus requires mitosis, during which the barrier between nucleoplasm and cytosol is removed and the trans-Golgi network becomes dispersed, thereby allowing the L2-DNA complex to gain access to the nucleus (9, 10), where it is found in close proximity with PML bodies in which viral genome expression is believed to initiate (11).

While previous studies have shown that mitosis is an important element for completion of infectious virus entry, expression of a HPV-16 pseudovirion (PsV) reporter gene cannot be detected earlier than 16 h postinfection in asynchronously growing cells (3). However, classical endosomal cargo trafficking pathways typically are very quick (12), suggesting that cells infected with HPVs when close to mitosis allow nuclear entry and initiation of viral gene expression at much earlier time points. To investigate this possibility, we have monitored HPV-16 PsV infection in cells following cell cycle synchronization. We now show that intracellular transport of PsVs is remarkably quick and, depending upon the specific phase of the cell cycle in which infection takes place, viral nuclear entry can occur within 1 h to 2 h postinfection and reporter gene expression can be detected as little as 8 h postinfection. This demonstrates that the main restriction on nuclear entry of the L2-DNA complex and the initiation of viral gene expression is the completion of mitosis.

MATERIALS AND METHODS

Cell lines.

Spontaneously immortalized human keratinocyte cells (HaCaT) were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS), penicillin-streptomycin (100 U/ml), and glutamine (300 mg/ml).

PsV production.

HPV-16 PsVs containing a luciferase reporter plasmid were generated in 293TT cells as previously described (4, 13). PsVs containing packaged 5-ethynyl-2′-deoxyuridine (EdU)-labeled plasmid DNA were prepared by addition of 25 μM EdU to the 293TT cells at 24 h posttransfection and harvested after a further 24 h.

Cell synchronization and infection.

HaCaT cells were seeded in a 12-well plate at a density of 0.75 × 10⁵ cells/well (aphidicolin treatment) or 0.5 × 10⁵ cells/well (thymidine treatment). After adherence, HaCaT cells were incubated for 24 h with 1 μg/ml aphidicolin (Sigma) to induce arrest at the G₁/S boundary. The cells then were released by washing with phosphate-buffered saline (PBS) and fresh DMEM was added. Cells were infected with HPV16 PsVs (1 h at 4°C) at a concentration of 12 ng/ml at different times after the release as indicated in the text. Cells were harvested at different time points (0 h, 4 h, 8 h, 12 h, and 16 h), and infection was monitored by luminometric analysis of firefly luciferase activity using a luciferase assay system kit (Promega).

For synchronization experiments using a double thymidine block, the cells were incubated with 2 mM thymidine (Sigma) for 16 h. The cells then were washed with PBS, and fresh DMEM was added for a further 9 h, after which 2 mM thymidine (Sigma) again was added for 16 h. The cells then were released by washing with PBS and infected 3 h postrelease (S phase) with HPV16 PsVs. One hour after infection at 4°C, the cells were washed with PBS, fresh medium was added, and cells were incubated for a further 8 h. After that time, cells were collected and infection was monitored by luminometric analysis as described above. Experiments were done in triplicate, and each sample was measured three times. Throughout, equal amounts of protein were used in the assays, which was ascertained by prior measurement of total protein concentration in the different cell extracts.

Flow cytometry.

Cell cycle analysis was done with fluorescence-activated cell sorter (FACS) analysis by measuring DNA content using propidium iodide staining as described previously (14).

PsV trafficking assay.

HaCaT cells seeded on coverslips were grown overnight and synchronized using aphidicolin block as described above. At different times after release from G₁/S (0 h and 10 h), cells were prechilled to 4°C, infected with EdU-labeled PsVs, and incubated for 1 h at 4°C with agitation to allow viral attachment. Cells then were washed, DMEM was replaced, and cells were transferred to 37°C. At different time points after attachment (1 h and 3 h), cells were washed with PBS and fixed in PBS plus 3.7% paraformaldehyde for 15 min at room temperature. EdU and PML staining were done as previously described using a Click-iT EdU imaging kit (Molecular Probes) and anti-PML primary antibodies (1:100; Santa Cruz) (4). Cells were counterstained with 4′,6-diamidino-2-phenylindole (DAPI), washed in water, and mounted on glass slides. Slides were visualized using a Zeiss Axiovert 100M microscope (Zeiss) attached to an LSM 510 confocal unit or a Leica DMLB fluorescence microscope equipped with a Leica photo camera (A01M871016).

Neutralization assay.

Neutralization assays were performed by incubation of PsVs with neutralizing antibody H16.V5 at a final concentration of 1:1,000 or preimmune antibody for 1 h at 4°C. Neutralized PsVs then were added to HaCaT cells, and infectivity was measured at different time points as described above. Mock infections also were performed as additional negative controls.

RESULTS AND DISCUSSION

Previous studies have shown that mitosis is needed to complete HPV infection and that the expression of a PsV reporter gene cannot be detected earlier than 16 h postinfection in asynchronously growing cells (3, 10, 11). However, intracellular transport mechanisms are relatively fast (12), suggesting that cells infected with HPVs when close to mitosis allow detection of viral gene expression and nuclear entry at much earlier time points postinfection.

To first identify the earliest time point at which PsV reporter gene expression can be detected in asynchronously growing HaCaT cells, HPV-16 PsVs carrying luciferase were used to infect HaCaTs, and cell extracts were collected at different times postinfection. Luciferase activity was measured, and the results depicted in Fig. 1A show the first detection of luciferase activity at 16 h postinfection, increasing to a maximum at 26 h postinfection (Fig. 1A). These results are very much in agreement with those of previous studies (3) and indicate that in asynchronous cultures, infectious HPV entry and initiation of gene expression is a relatively slow process.

FIG 1.

Comparative rate of HPV16 PsVs infectious entry in asynchronous and synchronous HaCAT cells. (A) Asynchronously growing HaCaT cells were infected with HPV-16 PsVs carrying a luciferase reporter plasmid and were harvested at different time points postinfection (0 h, 4 h, 8 h, 13 h,16 h, 19 h, 22 h, 26 h, and 48 h), and the luciferase activity was measured. The results are expressed as the means from at least three independent experiments, and the standard deviations are shown. (B) Aphidicolin-synchronized HaCaT cells were infected with HPV16 PsVs carrying a luciferase reporter plasmid 7 h postrelease from G₁/S (time point 0 h). Cells were harvested at different time points postinfection (0 h, 4 h, 8 h, 12 h, and 16 h) and luciferase activity measured. The results are expressed as the means from at least three independent experiments, and the standard deviations are shown. (C) Cell cycle analysis of the cells shown in panel B stained with propidium iodide and analyzed by flow cytometry. Note that the cells were mostly arrested in G₁ following aphidicolin treatment (panel I), at the time of infection 7 h postrelease they were in S phase (panel II), and had entered mitosis by 4 h postinfection (panel III). A significant proportion of cells had reentered G₁ at the 8-h time point (panel IV), when luciferase activity was first detected. (D) HPV16 PsV infections were performed as described for panel B, except that PsVs were preincubated with H16.V5 neutralizing antibody or preimmune (PI) antibody prior to infection and luciferase activity measured 8 h postinfection. The results are expressed as the means from at least three independent experiments, and the standard deviations are shown. Also shown are mock-infected cells.

We then proceeded to investigate whether reporter gene activity could be detected at earlier time points postinfection if cells were exposed to virions close to mitosis. HaCaT cells were synchronized in G₁/S with aphidicolin and then were released from the G₁/S block. After 7 h, the cells were infected with HPV16 PsVs (at 0 h). Cells were harvested at different times postinfection (0 h, 4 h, 8 h, 12 h, 14 h, and 16 h) and analyzed for luciferase activity and for the corresponding cell cycle profiles. The results demonstrate that when cells predominantly in S phase are infected with PsVs (Fig. 1C, panel II), the luciferase activity can be detected first by 8 h postinfection (Fig. 1B), which is significantly quicker than that in asynchronously infected cells. As can be seen, no luciferase activity could be detected at 4 h postinfection, a time when the majority of the cells still were passing through mitosis (Fig. 1C, panel III), while at the 8-h time point a significant proportion of the cell population had passed back into the G₁ phase of the cell cycle, consistent with completion of mitosis (Fig. 1C, panel IV).

To verify that the luciferase signal observed at 8 h postinfection was a result of infectious virus entry, the assay was repeated but the PsVs were preincubated with the anti-L1 PsV neutralizing antibody H16.V5 (15). As can be seen from Fig. 1D, prior incubation of the PsVs with H16.V5 antibody abolished the luciferase activity at the 8-h time point. In order to confirm these results using a different method of cell synchronization, the neutralization assay was repeated and infections performed on HaCaT cells that had been synchronized following a double thymidine block. The PsVs were incubated with H16.V5 antibody and used to infect HaCaT cells 3 h postrelease from the thymidine block. Luciferase activity then was measured again at the 8-h time point postinfection. As can be seen in Fig. 2A, there were readily detectable levels of reporter gene activity at this time point, and this was abolished by prior incubation of the PsVs with the neutralizing antibody. Consistent with the results shown in Fig. 1, PsV infection at 3 h postrelease from the thymidine block occurred while the majority of the cell population was in S phase (Fig. 2B, panel II), and no luciferase activity was detectable at 4 h postinfection (data not shown), when most of the cells were in G₂M (Fig. 2B, panel III). However, luciferase was detected at 8 h postinfection when the majority of the cell population had reentered G₁ (Fig. 2B, panel IV). These results demonstrate that HPV PsV infectious entry and subsequent reporter gene expression is much quicker when cells are infected during the late S or the early G₂/M phase of the cell cycle. This is perfectly consistent with previous reports suggesting that infectious entry into the nucleus requires mitosis (9, 10), but it emphasizes that the intracellular trafficking of the virus is much faster than previously assumed.

FIG 2.

HPV16 PsV reporter gene expression can be detected 8 h postinfection. (A) Double thymidine block-synchronized HaCaT cells were infected with HPV16 PsVs carrying a luciferase reporter plasmid 3 h postrelease from G₁ arrest using PsVs that were preincubated with H16.V5 neutralizing antibody or preimmune (PI) antibody. After 8 h the cells were harvested and luciferase activity measured. The results are expressed as the means from at least three independent experiments, the standard deviations are shown, and a mock infection control also was included. (B) Cell cycle analysis of the HaCaT cells used for panel A. Note growth arrest in G₁ following the double thymidine block (panel I), entry into S phase 3 h postrelease (panel II) and at the time of infection (time 0 h), entry into G₂M 4 h postinfection (panel III), and reentry into G₁ 8 h postinfection (panel IV) at the time of the detection of luciferase activity for panel A.

These results indicate that PsV reporter DNA can gain access to the nucleus relatively quickly postinfection if infection occurs close to mitosis. To identify the minimum time required for PsV DNA to gain access to the nucleus, we used HPV16 PsVs containing reporter plasmid DNA labeled with the thymidine analog 5-ethynyl-2′deoxyuridine (EdU). HaCaT cells were synchronized with aphidicolin as described above and then released from the G₁/S block and infected with PsVs either immediately, when the cells were in G₁ (Fig. 1C, panel I), or after 10 h, when the cells were still passing through G₂M (Fig. 1C, panel III). The infected cells were fixed at 1 h postinfection, and EdU was detected in an azide-alkyne reaction. Cells then were stained for PML, a major component of ND10 domains, where HPV genomes typically are found in an established HPV infection (11). As can be seen from Fig. 3A, EdU-labeled DNA encapsidated by HPV-16 PsVs was detected only in the cytoplasm in cells infected immediately after release from G₁ and fixed 1 h postinfection. In contrast, in cells that were infected 10 h postrelease from G₁ and fixed 1 h postinfection, EdU staining was readily detected within the nucleus, and colocalization with PML could be observed (Fig. 3B). Similar results also were obtained when cells were analyzed by confocal microscopy. As can be seen from Fig. 4A, there is no colocalization of EdU with PML when cells are infected immediately after release and fixed 1 h postinfection, while in cells infected 10 h postrelease and analyzed 1 h postinfection there is clear colocalization of PsV reporter DNA with PML, as shown from the zeta stack analysis (Fig. 4B). In cells actually undergoing mitosis, PML bodies become fewer and larger (16), and in several instances, EdU-labeled PsV reporter DNA shows clear colocalization with PML at such sites, using both epifluorescence (Fig. 3B) and confocal microscopy with zeta stacks (Fig. 5). Taken together, these results demonstrate that HPV PsV reporter DNA can gain access to the nucleus as soon as 1 to 2 h postinfection if the infected cells are close to mitotic entry, and that association with PML also can occur at this time.

FIG 3.

Detection of HPV16 PsV reporter DNA in proximity with PML 1 h postinfection. Aphidicolin-synchronized HaCaT cells were infected with HPV16 PsVs carrying EdU-labeled luciferase reporter plasmid 1 h (A) or 10 h (B) postrelease from G₁/S. After a further 1 h at 37°C, the cells were fixed and processed for the detection of EdU-labeled DNA (red), PML (green), and DAPI (blue). Note the appearance of perinuclear accumulation of EdU-labeled DNA shown in panel A but colocalization with nuclear PML shown in panel B, as indicated by the arrows and the expanded field.

FIG 4.

Confocal image analysis showing HPV16 PsV reporter DNA in conjunction with PML. Aphidicolin synchronized HaCaT cells were infected with HPV16 PsVs carrying EdU-labeled luciferase reporter plasmid 1 h (A) or 10 h (B) postrelease from G₁/S. After a further 1 h at 37°C, the cells were fixed and processed for the detection of EdU-labeled DNA (red) and PML (green). Shown are the zeta stacks demonstrating cytoplasmic PsV EdU-labeled DNA shown in panel A and colocalization with PML in two different fields (upper and lower) in panel B, with an expanded view of one of those fields.

FIG 5.

Confocal image analysis showing HPV16 PsVs reporter DNA in conjunction with PML in mitotic cells. Aphidicolin-synchronized HaCaT cells were infected with HPV16 PsVs carrying EdU-labeled luciferase reporter plasmid 10 h postrelease from G₁/S. After a further 3 h at 37°C, the cells were fixed and processed for the detection of EdU-labeled DNA (red) and PML (in green). Shown are the zeta stacks demonstrating PsV EdU-labeled DNA and colocalization with PML in two different fields (upper and lower) in cells that are undergoing mitosis.

While previous studies have shown a requirement for mitosis for the L2-DNA complex to gain entry into the nucleus, there has been little information on the speed with which HPV virions and their packaged genomes are trafficked within the cell. We provide compelling evidence that reporter gene expression can be detected as little as 8 h postinfection if cells are infected in late S phase. Likewise, PsV reporter DNA can be found associated with PML nuclear bodies as little as 1 to 2 h postinfection and, most strikingly, this also can be seen clearly in cells which are undergoing mitosis and where nuclear envelope breakdown already has occurred, suggesting that the association of the L2-DNA complex with PML nuclear bodies first appears during mitosis. Current studies are focused on identifying the cellular components that are required for recruiting the L2-DNA complex to these nuclear domains.