Abstract
During human papillomavirus (HPV) entry, the virus exploits COPI-dependent retrograde transport to cross the Golgi apparatus before reaching the nucleus to cause infection. How HPV enters the nucleus after exiting the Golgi is unclear, although mitotic nuclear envelope breakdown (NEB) appears important. Here, we show that importin-7 (IPO7), a nuclear pore import receptor, is present at the Golgi and promotes HPV infection. IPO7 knockdown inhibits infection and causes HPV to accumulate in the Golgi without reaching mitotic chromosomes, demonstrating that IPO7 promotes Golgi-to-nucleus transport of HPV. Golgi-to-nucleus transport of a cellular cargo also requires IPO7, suggesting that HPV hijacks a preexisting pathway for nuclear entry. Furthermore, the C-terminal nuclear localization sequence of HPV L2 protein, which overlaps its cell-penetrating peptide sequence, binds IPO7 directly in a COPI-dependent virus trafficking step. Together, these data identify a role for an importin in HPV infection and suggest that the canonical nuclear pore import machinery plays an unanticipated role in NEB-dependent nuclear entry.
HPV, a cancer-causing human pathogen, exploits Golgi-associated importin-7 for Golgi-to-nucleus trafficking to promote infection.
INTRODUCTION
The human papillomavirus (HPV) causes ~5% of all human cancers globally, including essentially all cervical cancer as well as a substantial fraction of other anogenital and nongenital cancers (1, 2). Although effective vaccines against HPV infection exist, vaccination rates remain persistently low. In addition, there are no specific antiviral agents against active HPV infection. Hence, a better understanding of the mechanism of HPV infection may lead to the development of effective therapies against HPV-induced diseases.
Structurally, HPV is a small nonenveloped DNA virus composed of 72 pentamers of the major capsid protein L1 and up to 72 copies of the minor capsid protein L2. The L1 and L2 proteins encapsidate the viral DNA genome which must be delivered to the nucleus of the host cell to cause infection (3, 4). HPV initiates entry through L1 binding to heparan sulfate proteoglycans on the host cell surface or in the extracellular matrix. This interaction imparts conformational changes to the capsid and allows subsequent furin-mediated cleavage of the N terminus of L2 (5–7). Via an unidentified entry receptor, the virus is endocytosed and reaches the endosome (8–12). Here, the acidic luminal environment triggers partial disassembly of the capsids (11, 12), exposing the C terminus and the putative transmembrane domain of L2 which in turn engages the endosome-localized membrane protein γ-secretase (13, 14). γ-Secretase deploys an unconventional chaperone activity to facilitate membrane penetration of L2 (13), a step that requires the cell-penetrating peptide (CPP) of L2 which is located near its C terminus (15, 16). Membrane penetration effectively inserts most of L2 across the endosome membrane, exposing a large portion of L2 to the cytoplasm (17, 18).
In this membrane-inserted topology, L2 recruits different cytosolic host factors that direct proper virus trafficking. For instance, the retromer sorting machinery, along with the dynein-BICD2 motor complex, interacts with L2 to enable trafficking of HPV from the endosome to the trans-Golgi network (TGN) (19–23). Upon arrival to the TGN, L2 further binds to the COPI (coat protein complex I) sorting complex, which transports HPV across the Golgi stack membranes in a retrograde manner before reaching the nucleus to cause infection (24).
How HPV reaches the nucleus after it exits the Golgi and the role of L2 in this step are not fully understood. Nuclear envelope breakdown (NEB) during mitosis appears to allow HPV to enter the nucleus (25–30). During mitosis, L2 mediates transport of the virus along microtubules, apparently in a vesicle-bound form, into the nucleus (31–36). L2 and the viral genome appear to be anchored in the nucleus by being tethered to mitotic chromatin (30).
Nuclear import of many cellular proteins involves importins, which typically bind to and transport cellular cargos containing a nuclear localization signal (NLS) from the cytosol to the nucleoplasm via the nuclear pores. After nuclear entry, binding of Ran–guanosine triphosphate to the importin-cargo complex triggers the release of the cargo from the importins (37). The Ran-binding protein 10 (RanBP10) and karyopherin alpha 2 (KPNA2), an importin α protein, have been implicated in nuclear entry of HPV during infection (36). In addition, other components of the classic nuclear import machinery, including additional importin α proteins and nuclear import receptors, are involved in nuclear import of the L2 protein when it is expressed in the absence of L1, but these proteins may be important for nuclear import of newly synthesized L2 before capsid assembly during the late stages of the viral life cycle, rather than participating in nuclear targeting of HPV during virus entry (38–41). How (or whether) the incoming HPV exploits this nuclear import machinery to gain nuclear entry is not known.
In this study, by using an artificial protein that inhibits HPV infection, unbiased proteomics, pharmacological inhibition, gene-targeting knockdown (KD), as well as biochemical and microscopy approaches, we showed that the importin β family member importin-7 (IPO7) is a Golgi-associated host factor that plays a critical role in HPV infection. Specifically, we found that KD of IPO7 causes the virus to accumulate in the Golgi and prevents HPV from entering the nucleus to reach the mitotic chromosomes, thereby inhibiting infection. These results indicate that this importin family member plays a key role in Golgi-to-nuclear transport of HPV. Transport of a cellular cargo from the Golgi to the nucleus also depends on IPO7, indicating that HPV hijacks a preexisting route to reach the nucleus. We further show that IPO7 binds directly to a C-terminal region of L2 that harbors overlapping nuclear localization and CPP sequences. Furthermore, the IPO7-HPV L2 interaction occurs after COPI-dependent retrograde trafficking during virus entry, and COPI depletion blocks binding between IPO7 and L2. Our study suggests that HPV exploits a nuclear pore import component in an unconventional way before NEB-dependent nuclear entry.
RESULTS
Study of a Golgi-localized traptamer reveals an important role of importin β members in HPV infection
We previously used a protein modulation screen to isolate four artificial FLAG-APEX2–tagged (FA) transmembrane proteins (called traptamers) that inhibit different HPV entry steps (21, 42). These traptamers are thought to bind to and compromise the activity of host factors required for HPV infection. When the FA-JX4 traptamer is stably expressed in HeLa cells, it localizes to the TGN (21) and the Golgi apparatus (as shown by colocalization with the cis-Golgi protein GM130; fig. S1A) and impairs HPV pseudovirus (PsV) infection by trapping HPV in these compartments, thereby preventing arrival of the encapsidated DNA in the nucleus (21). We hypothesized that FA-JX4 binds to and inhibits critical Golgi-localized host factors essential for HPV trafficking. To identify host proteins that bind directly or indirectly to FA-JX4 and participate in the late stages of HPV entry, we infected HeLa cells expressing FA-JX4 or the control FA (a noninhibitory traptamer lacking a transmembrane domain) and performed anti-FLAG immunoprecipitation (IP) followed by mass spectrometry (MS) analysis. For these experiments, we used HPV16 PsV that is composed of the high-risk HPV16 L1 and L2 capsid proteins (without epitope tags) encapsidating a reporter plasmid encoding the green fluorescent protein (GFP) in place of the HPV16 genome. In this system, expression of GFP indicates successful entry of the PsV DNA to the nucleus, which allows expression of the reporter plasmid. Notably, HPV16 PsV infection recapitulates infection by authentic HPV isolated from the stratified keratinocyte raft cultures and is widely used in studies of HPV entry (43).
At 24 hours postinfection (hpi), cell extracts were collected and subjected to IP using an anti-FLAG antibody to pull down FA or FA-JX4, and the coprecipitated material was analyzed by MS. A fraction of the precipitated material was analyzed by SDS–polyacrylamide gel electrophoresis (SDS-PAGE) followed by immunoblotting to confirm that similar levels of FA and FA-JX4 proteins were IPed (fig. S1B). The MS results showed that neither HPV16 L1 nor L2 capsid protein was present in the FA or FA-JX4 IPed sample (Fig. 1A; the complete data are available in table S1), suggesting that FA-JX4 does not directly interact with the virus to block its entry. Instead, several Golgi-associated COPI subunits were found in the FA-JX4 (but not FA) IPed sample (table S1), consistent with the observation that FA-JX4 localizes to the TGN/Golgi (21). The IPed FA-JX4 (but not FA) sample also contained multiple peptides from five members of the importin β family, namely, IPO1, IPO4, IPO5, IPO7, and IPO8, as well as a single peptide from IPO9 (Fig. 1A). Importin β family proteins are components of the nuclear import machinery that transports cargos from the cytosol to the nucleoplasm via the nuclear pores. However, importin β family proteins are not known to be associated with the Golgi, and their involvement in HPV infection has not been examined.
Fig. 1. IPO7 promotes HPV infection.
(A) Number of peptides corresponding to importin β family proteins identified by MS. Immunoprecipitation and sample preparation from HeLa S3 cells expressing traptamer FA-JX4 or control FA were performed as described in fig. S1B. All MS data are in table S1. (B) HeLa cells were infected with HPV16.L2F (MOI, ~0.4; GFP reporter plasmid) and treated at infection with 5 to 10 μM importazole (IPZ) or DMSO. At 48 hpi, GFP-positive cells were quantified by flow cytometry and normalized to DMSO (0 μM IPZ) controls. Data (n = 3) are shown as individual points, means, and SDs; statistical significance was determined versus DMSO-treated cells by two-tailed, unequal variance t test. **P < 0.01 and ***P < 0.001. (C) HeLa cells were transfected with 10 nM of the indicated siRNA for 48 hours and then infected (MOI, ~0.2) with HPV16.L2F (GFP reporter plasmid). At 48 hpi, GFP-positive cells were measured by flow cytometry and normalized to Scr siRNA controls. Data (n = 3) are analyzed and presented as in (B), with statistical significance determined versus Scr-treated cells. *P < 0.05 and **P < 0.01. (D) HeLa cells were transfected with siRNA-resistant HA-IPO7-mCherry or control HA-mCherry for 24 hours and then with Scr or IPO7 siRNA #2 for 48 hours and infected (MOI, ~0.2) with HPV16.L2F (GFP reporter plasmid). At 48 hpi, GFP expression in mCherry-positive cells was quantified by flow cytometry and normalized to the Scr siRNA + HA-mCherry control. Data (n = 3) are analyzed and presented as in (B), with statistical significance determined between samples. *P < 0.05 and ***P < 0.001. (E) As in (C), except HaCaT cells were analyzed. ***P < 0.001. (F) As in (C), except HPV5.L2F (MOI, ~0.4) was used. ***P < 0.001. (G) As in (C), except HPV18.L2F (MOI, ~0.4) was used. ***P < 0.001.
To determine whether importin β proteins are important in HPV infection, we first tested whether HPV infection was perturbed by importazole (IPZ), an importin β inhibitor (44). In these experiments, we used HPV16 PsV containing HPV16 L1, 3xFLAG-tagged HPV16 L2 (to detect L2), and a reporter plasmid encoding S-tagged GFP (GFP-S). Following infection with this PsV, referred to as HPV16.L2F PsV, infectivity can be monitored by the expression of GFP-S. Compared to HeLa cells treated with the control vehicle dimethyl sulfoxide (DMSO), addition of IPZ at the same time as HPV16.L2F PsV reduced GFP-S expression at 48 hpi in a concentration-dependent manner as assessed by monitoring GFP-S fluorescence via flow cytometry (Fig. 1B). IPZ inhibition of HPV infection was also observed by immunoblotting for GFP-S (fig. S2A; quantified in the graph below); as a control, the γ-secretase inhibitor XXI blocked HPV infection, as expected (45). IPZ also inhibited HPV infection in the human cervical squamous carcinoma SiHa cells (fig. S2B) and in human skin HaCaT keratinocytes (fig. S2C). These findings suggest that importin β proteins promote HPV infection in multiple cell types.
Because HPV nuclear entry requires mitosis (25–30), we examined whether IPZ treatment altered the cell cycle profile. A previous report showed that high concentrations of IPZ (i.e., 50 and 100 μM) can impair cell cycle progression (44). However, our analysis showed that treating cells with 10 μM IPZ, which inhibited HPV PsV infection, did not affect the distribution of cells in different stages of the cell cycle when compared to the control group (fig. S2D; quantified in the graph below). Thus, inhibition of HPV infection by IPZ is not due to impaired mitosis.
Because IPZ treatment for as short as 1 hour can inhibit nuclear import events (44), we conducted a time-course experiment to assess when the importin β proteins act during HPV entry. In HeLa cells, IPZ (10 μM) was added either at the same time as HPV16.L2F PsV (0 hpi) or at 8-hour intervals after virus addition, and GFP-S fluorescence was analyzed by flow cytometry. IPZ maintained its ability to reduce HPV infection even if the drug was added 16 hpi but lost its inhibitory effect when added 24 hpi (fig. S2E); again, similar data were obtained by immunoblotting to detect GFP-S expression (fig. S2F). Because HPV nuclear entry events occur approximately 24 hpi (26, 46), these findings are consistent with the hypothesis that importin β family proteins play a pivotal role during relatively late stages of HPV entry.
IPO7 promotes HPV infection
We next used a small interfering RNA (siRNA)–mediated KD approach to further support a role of the importin β family proteins in HPV infection. We tested two different siRNAs against each of the six identified FA-JX4–interacting importin β proteins (Fig. 1A) to deplete the intended target in HeLa cells. Compared to cells transfected with the control scrambled (Scr) siRNA, all of the siRNAs except IPO4 siRNA #1 robustly reduced the level of their intended protein (fig. S3A). The KD cells were infected with HPV16.L2F PsV, and infectivity was assessed by flow cytometry for GFP fluorescence. Although the different siRNAs decreased infectivity by varying extents compared to control cells, KD of IPO5 or IPO7 caused the most severe inhibitory effect (Fig. 1C). IPO7 siRNA #2 caused the strongest inhibition of HPV infection, to an extent similar to cells depleted of the γ-secretase subunit PS1 (Fig. 1C) (13, 45). Immunoblotting for GFP-S after KD of IPO7 (using siRNA #2) also demonstrated a marked inhibition of infection by HPV16 PsV (fig. S3B, left). KD of IPO7 did not affect the distribution of cells in the cell cycle compared to control cells (fig. S3C; quantified in the graph below), suggesting that the inhibition of HPV infection by IPO7 KD is not caused by a defect in cell cycle progression. By contrast, IPO5 KD altered the cell cycle profile (fig. S3C; quantified in the graph below) and therefore may elicit unintended effects on HPV entry. Because depletion of IPO7 led to the most pronounced impairment of HPV infection without affecting cell cycle progression, for the remainder of this study, we focused on elucidating the role of IPO7 in HPV infection.
To establish that the IPO7 KD phenotype in HPV infection is not due to an off-target effect, we performed a KD-rescue experiment. Control or IPO7 KD cells were transfected with a plasmid expressing hemagglutinin (HA)–tagged mCherry (HA-mCherry) or HA-mCherry fused to (siRNA-resistant) IPO7 (HA-IPO7-mCherry) (fig. S3D). Cells were then infected with HPV16.L2F PsV, and flow cytometry was performed 2 days after infection. GFP fluorescence was used to monitor infectivity, and mCherry fluorescence was used as a measure of expression of the rescue construct. We found that HPV infection in cells expressing the control HA-mCherry was inhibited by IPO7 KD (Fig. 1D; compare the second to the first bar), as expected (Fig. 1C). In contrast, expression of HA-IPO7-mCherry almost entirely rescued the infectivity defect caused by IPO7 KD (Fig. 1D; compare the fourth to the second bar); expression of HA-IPO7-mCherry in control cells did not enhance infectivity (Fig. 1D; compare the third to the first bar). Restoration of HPV infection in IPO7 KD cells by expression of HA-IPO7-mCherry rules out off-target effects of the IPO7 siRNA on infectivity, establishing that IPO7 promotes HPV infection.
As assessed by flow cytometry (Fig. 1E) and immunoblotting (fig. S3E), IPO7 plays an important role in HPV infection in HaCaT cells as well as in HeLa cells. In addition, IPO7 is also required for efficient infection by HPV5 and HPV18 PsV in HeLa cells (Fig. 1, F and G by flow cytometry, and fig. S3B, center and right, by immunoblotting). Together, these results reveal a requirement of IPO7 in HPV infection in several cell lines and for several HPV types.
IPO7 promotes Golgi-to-nucleus transport of HPV
Our data thus far suggest that relatively late steps during HPV entry require IPO7, an established nuclear entry factor. To determine which HPV trafficking step was affected by IPO7 KD, we performed proximity ligation assay (PLA) under the KD condition. PLA is an immunocytochemistry-based assay that generates fluorescent signals only when two targeted proteins are within 40 nm of each other. We first examined whether transit of HPV through the Golgi is impaired upon IPO7 KD. At 24 hpi, we observed a PLA signal between L2-3xFLAG and GM130 in control cells infected with HPV16.L2F (containing a reporter plasmid that encodes luciferase instead of GFP-S to prevent interference with the PLA fluorescent signal) but not in uninfected control cells (fig. S4A; quantified in fig. S4B), demonstrating that HPV reaches the cis-Golgi compartment, as expected (24). Under IPO7 KD, the PLA signal between L2-3xFLAG and GM130 modestly increased (1.3-fold) when compared to control cells (fig. S4A; quantified in fig. S4B). Notably, when PLA for GM130 and L2-3xFLAG was carried out at 32 hpi, the PLA signal in the IPO7 KD cells was 3.1 times that of the signal in the control cells (Fig. 2A; quantified in Fig. 2B). These results demonstrate that in IPO7-depleted cells, HPV can transport across the Golgi stack to reach the cis-Golgi but accumulates in this compartment over time, suggesting that IPO7 is required for the exit of HPV from the cis-Golgi.
Fig. 2. IPO7 promotes Golgi-to-nucleus transport of HPV.
(A) HeLa S3 cells were transfected with 10 nM Scr or IPO7 siRNA #2 for 48 hours and then left uninfected or infected with HPV16.L2F (MOI, ~100). At 32 hpi, PLA (signals shown in green) was performed with antibodies recognizing FLAG (to detect L2) and GM130. Nuclei were stained with 4ʹ,6-diamidino-2-phenylindole (DAPI; blue). Similar results were obtained in two independent experiments. Scale bar, 10 μm. (B) PLA fluorescence intensity per cell (>100 cells) as in (A), shown as individual values, means, and SDs, with statistical significance determined by one-way analysis of variance (ANOVA). ***P < 0.001 and ****P < 0.0001. (C) As in (A), except infection at MOI of ~150 with HPV16.L2F-containing EdU-labeled genome. Cells were fixed 32 hpi, subjected to Click-iT EdU detection (green), and immunostained for Nesprin-2 (magenta). Scale bar, 10 μm. (D) Nuclear, extranuclear, and total EdU intensity per cell in multiple images as in (C). Using a cutoff defined as the mean intensity per uninfected cell plus one SD, all 89 control cells and 86 of 89 IPO7 KD cells were EdU positive. One-way ANOVA was used to determine statistical significance. ns, not significant; ****P < 0.0001. (E) As in (A), except infection at MOI of ~15. Cells were fixed at 32 hpi and immunostained for FLAG (green) and GM130 (magenta); DNA was stained by DAPI (blue). Cells displaying condensed chromosomes in DAPI staining and fragmented Golgi (GM130 puncta) were identified as mitotic cells. Similar results were obtained in two independent experiments. Scale bars, 10 μm. (F) Ratio of FLAG intensity on condensed chromosomes to total per mitotic cell (>27 mitotic cells) as in (E), shown as individual values, means, and SDs, with statistical significance determined by two-tailed, unequal variance t test. ****P < 0.0001.
By confocal microscopy of standard immunofluorescence staining, we observed that at 32 hpi, the L2-3xFLAG signals can be detected in the nucleus and the Golgi of infected control cells, as demonstrated by colocalization of L2-3xFLAG with 4ʹ,6-diamidino-2-phenylindole (DAPI)–stained nuclei and GM130, respectively (fig. S4C, center column, see zoom of merge). By contrast, in IPO7-depleted cells, L2-3xFLAG remained colocalized with GM130, with a low level in the nucleus (fig. S4C, right column, see zoom of merge). Similar results were observed when another cis-Golgi protein, Giantin, was used for the colocalization analysis (fig. S4D). As controls, the localization of GM130 in interphase cells (fig. S4C), as well as the overall GM130 level (fig. S4E), was unaffected following IPO7 KD. In addition, in uninfected mitotic cells that display condensed chromosomes, depletion of IPO7 did not interfere with Golgi fragmentation, as evidenced by the appearance of dispersed GM130 puncta (fig. S4F). Together, these findings suggest that depletion of IPO7 prevents HPV arrival to the nucleus by trapping the virus in the Golgi without impairing overall Golgi integrity.
To confirm the role of IPO7 in transport of HPV to the nucleus, we assessed the fate of HPV16.L2F PsV in which the pseudogenome (a luciferase reporter plasmid) is labeled with the nucleoside analog 5-ethynyl-2′-deoxyuridine (EdU) (47). In this experiment, the nuclear envelope is identified by using an antibody against Nesprin-2, a nuclear membrane protein. As expected, a low background EdU signal was observed in uninfected control cells (Fig. 2C; the nuclear, extranuclear, and total EdU fluorescence intensity was quantified in Fig. 2D). In control cells infected with the EdU-labeled virus for 32 hours, strong EdU signals appeared in the nucleus, while a relatively low level of signal was found outside of the nucleus (Fig. 2C; quantified in Fig. 2D). These results demonstrate successful trafficking of the incoming PsV and delivery of the encapsidated EdU viral genome into the nucleus. In IPO7 KD cells, the EdU signal in the nucleus was markedly reduced, and the signal accumulated outside of the nucleus (Fig. 2C; quantified in Fig. 2D), indicating that nuclear arrival of HPV is inhibited. These findings further demonstrate an important role of IPO7 in transport of HPV to the nucleus to cause infection.
After NEB during mitosis, HPV enters the nucleus and tethers to condensed chromosomes (30). We therefore tested the role of IPO7 in HPV association with condensed chromosomes in mitotic cells. Immunofluorescent staining for L2-3xFLAG showed colocalization of HPV L2 with condensed chromosomes as expected, which was markedly decreased in IPO7-depleted cells (Fig. 2E; quantified in Fig. 2F). These data are consistent with the PLA (Fig. 2, A and B) and genome-labeled HPV (Fig. 2, C and D) results, demonstrating that IPO7 plays a pivotal role in promoting Golgi-to-nucleus trafficking of HPV during entry.
IPO7 associates with the Golgi membrane and promotes Golgi-to-nucleus transport of a cellular cargo
Because HPV is trapped in the Golgi under IPO7 KD (Fig. 2, A and B, and fig. S4, A to D), a pool of IPO7 might associate with the Golgi where it engages HPV and delivers the virus to the nucleus. We therefore examined whether IPO7 associates with the Golgi by conducting cell fractionation followed by immuno-isolation of the Golgi membrane. Extracts of uninfected HeLa cells were generated by mechanical homogenization and layered over a discontinuous sucrose gradient. After ultracentrifugation, individual fractions were collected from the top of the gradient, and the samples were subjected to SDS-PAGE followed by immunoblotting to identify the organelle(s) contained in each fraction (Fig. 3A). The TGN protein TGN46 and the cis-Golgi protein GM130 were found predominantly in fractions #2 to #4, consistent with our previous report (24). By contrast, the cytosolic protein HSP90 was predominantly found in fraction #2, the endoplasmic reticulum (ER) luminal protein PDI and the ER membrane protein BAP31 in almost every fraction, the early endosome protein EEA1 in fractions #1 and #2, the nuclear pore protein NUP188 mostly in fraction #2 with low levels in fractions #4 and #5 and fractions #9 and #10, and the nuclear histone H3 in fraction #2 and #6 to #11. Thus, fractions #2 to #4 contain most of the TGN/Golgi material. Notably, IPO7 also showed a strong enrichment in fractions #2 to #4 (Fig. 3A), most similar to the GM130 pattern, suggesting that some IPO7 is localized to the TGN/Golgi compartment in uninfected cells.
Fig. 3. IPO7 associates with the Golgi membrane and promotes Golgi-to-nucleus transport of a cellular cargo.
(A) HeLa cells were mechanically homogenized and fractionated through a 0.5 to 1.6 M sucrose gradient. Portions of each fraction (#1 to #11, top to bottom) were immunoblotted with the indicated antibodies to identify organelle markers. Fractions #3 containing TGN46 and GM130 represents the Golgi-enriched fraction. (B) Fraction #3 collected from (A) was subjected to IP with an anti-GM130 antibody or a control IgG and then immunoblotted with the indicated antibodies. (C) HeLa S3 cells were transfected with 10 nM Scr or IPO7 siRNA #2 for 48 hours and then uninfected or infected with HPV16.L2F (MOI, ~100). PLA (green) for IPO7 and GM130 was performed at 30 hpi; nuclei were stained with DAPI (blue). Similar results were obtained in two independent experiments. Scale bar, 10 μm. (D) PLA fluorescence intensity per cell (>80 cells) as in (C), shown as individual values, means, and SDs, with statistical significance determined by one-way ANOVA. ***P < 0.001 and ****P < 0.0001. (E) Uninfected HeLa cells were transfected with the mNG2-SREBP1 reporter construct for 48 hours and then with 10 nM Scr or IPO7 siRNA #2 for another 48 hours. Samples were fixed and permeabilized, and nuclei were stained with DAPI. The nuclear import of mNG2-SREBP1 was indicated by colocalization of mNG2-SREBP1 fluorescence (green) and DAPI (blue). Similar results were obtained in two independent experiments. Scale bar, 10 μm. (F) Ratio of nuclear to total mNG2-SREBP1 intensity per cell (>26 mNG-positive cells) as in (E) is plotted as individual values, means, and SDs, with statistical significance determined by two-tailed, unequal variance t test. ****P < 0.0001.
To determine whether IPO7 associates with the Golgi, we used an antibody against GM130 to IP the Golgi membrane in fraction #3 collected from uninfected cells, a TGN/Golgi-enriched fraction with minimal cosedimentation of cytosolic proteins or proteins from other cellular compartments (Fig. 3A). Our results showed that IPO7 was coimmunoprecipitated (co-IPed) with anti-GM130 precipitated materials (Fig. 3B), confirming the cosedimentation results that a pool of IPO7 associates with the Golgi membrane. As expected, the Golgi-associated protein GRASP55 also coprecipitated with GM130 (Fig. 3B). When the Golgi IP experiment was repeated in the presence of Triton X-100, a detergent that solubilizes membranes, IPO7 did not coprecipitate with GM130 (fig. S5A). In these experiments, we estimate that at least 3% of IPO7 (i.e., 30% of 10% input IPO7) associates with the Golgi membrane. As a control, IP of the ER membrane protein BAP31 from fraction #3 did not pull down IPO7 regardless of the presence or absence of the detergent (fig. S5B). The loss of IPO7 coprecipitation with GM130 after detergent-mediated solubilization of the Golgi membrane suggests that the IPO7-GM130 interaction depends on the presence of the intact membrane. This indicates that IPO7 does not bind directly to GM130 but is instead part of a Golgi membrane–associated protein complex.
To further support the idea that a pool of IPO7 proteins associates with the Golgi membrane, we performed PLA to determine whether IPO7 and GM130 are in close proximity in intact cells. We found prominent PLA signals generated from antibodies against IPO7 and GM130 in control but not IPO7 KD cells (Fig. 3C; quantified in Fig. 3D), indicating that a fraction of IPO7 and GM130 is indeed proximal. The IPO7/GM130 PLA signal increased modestly in HPV-infected cells (Fig. 3D). Because IPO7 is dispersed throughout the cell (figs. S4F and S5C), only a fraction of IPO7 is likely proximal to the Golgi membrane. Colocalization analysis showed that 3.5% of IPO7 colocalized with GM130 in uninfected cells (fig. S5D), consistent with the co-IP results (fig. S5A). A total of 4.2% of IPO7 colocalized with GM130 in infected cells (fig. S5D), in agreement with the modest increase in IPO7/GM130 PLA signal found in infected cells compared to uninfected cells (Fig. 3D). There was no change in the nuclear distribution of IPO7 between uninfected and infected cells (fig. S5E). Together, these results demonstrate that a pool of IPO7 associates with the Golgi membrane and suggest that HPV may recruit or retain IPO7 at the Golgi to support trafficking beyond the Golgi.
Our finding that IPO7 promotes Golgi-to-nucleus transport of HPV prompted us to test whether cellular cargos that undergo Golgi-to-nucleus transport might similarly depend on IPO7. The sterol regulatory element–binding protein 1 (SREBP1) is an ER membrane–bound transcription factor that mobilizes from the ER to the Golgi when activated (48, 49). At the Golgi, SREBP1 is cleaved, generating an N-terminal soluble (cytosolic) transcription factor that transports to the nucleus to stimulate expression of genes involved in lipid synthesis (48, 50). To test whether IPO7 plays a role in transporting (Golgi-derived) soluble SREBP1 to the nucleus, HeLa cells were transfected with a reporter construct (mNG2-SREBP1), in which SREBP1 (lacking the N-terminal activation domain of SREBP1) is fused at the N terminus with two copies of the fluorescent protein mNeonGreen (51). In the presence of glucose [which activates SREBP1 (50–52)], mNG2-SREBP1 was found in the nucleus as expected (Fig. 3E, left column; quantified in Fig. 3F). By contrast, mNG2-SREBP1 did not localize to the nucleus under IPO7 KD (Fig. 3E, right column; quantified in Fig. 3F), indicating that IPO7 plays a critical role in delivering SREBP1 to the nucleus. These findings highlight IPO7 as a key mediator of Golgi-to-nucleus trafficking and underscore its essential role in the nuclear entry of HPV.
HPV binds to IPO7 during virus entry
Our data thus far identified IPO7 as (at least in part) a Golgi-associated protein that promotes HPV infection by enabling Golgi-to-nucleus transport of the incoming virus particle. We hypothesize that the Golgi-associated IPO7 interacts directly with the cytosol-exposed segment of the HPV L2 protein to deliver HPV to the nucleus. To determine whether HPV and IPO7 interact during entry, we first used PLA to assess whether HPV and IPO7 are in proximity in infected cells. HeLa S3 cells were uninfected or infected with HPV16.L2F PsV containing a luciferase reporter plasmid, and PLA for HPV L2 and IPO7 was performed at 24 hpi. Low background-level PLA signal between IPO7 and L2-3xFLAG was observed in uninfected control cells, but the infected control cells displayed strong PLA signals (Fig. 4A; quantified in Fig. 4B). Most of the PLA signal was found in the cytosol, with a low level at or near the nuclear membrane. As expected, when IPO7 KD cells were infected by HPV16.L2F, there was only low background-level PLA signal between IPO7 and L2-3xFLAG (Fig. 4A; quantified in Fig. 4B). By confocal immunofluorescent microscopy (fig. S6A), IPO7 signals were dispersed throughout the control cell but were markedly reduced in the IPO7 KD cells, as expected. However, HPV L2 was mostly in the cytosol at 24 hpi in both control and IPO7 KD cells. These results suggest that a pool of IPO7 is proximal to HPV outside of the nucleus late during entry.
Fig. 4. HPV binds to IPO7 during virus entry.
(A) HeLa S3 cells were uninfected or infected with HPV16.L2F (MOI, ~100). PLA (green) for IPO7 and FLAG (to detect L2) was performed at 24 hpi; nuclei were stained with DAPI (blue). Similar results were obtained in two independent experiments. Scale bar, 10 μm. (B) PLA fluorescence intensity per cell (>100 cells) as in (A), shown as individual values, means, and SDs, with statistical significance determined by one-way ANOVA. ****P < 0.0001. (C) Whole-cell extracts from uninfected or HPV16.L2F-infected (MOI, ~1) HeLa cells were collected at the indicated time points. A portion of the resulting extracts (input) was analyzed by immunoblotting for FLAG (to detect L2) and IPO7. The remaining extracts were subjected to IP with an anti-IPO7 antibody or control IgG and analyzed by immunoblotting for FLAG and IPO7. (D) HeLa cells were pretreated with 9 μM RO3306 or DMSO as a control for 24 hours and then uninfected or infected with HPV16.L2F (MOI, ~1), with RO3306 or DMSO maintained in the medium. At 24 hpi, cells were lysed and analyzed in (C). (E) HeLa cells treated with 10 nM Scr or a mixture of 5 nM COPA and 5 nM COPG1 (COPA/G1) siRNA were uninfected or infected with HPV16.L2F PsV (MOI, ~1). At 24 hpi, cells were lysed, and a portion of the extracts (input) was immunoblotted for FLAG, IPO7, γ-COP, or β-actin (as a loading control). The remaining extracts were subjected to IP with an anti-IPO7 antibody or a control IgG and analyzed as in (C). (F) As in (E), except cells were transfected with 10 nM Scr or IPO7 siRNA #2 before infection. Samples were subjected to IP with an anti–γ-COP antibody or control IgG and analyzed as in (E).
To test whether HPV L2 is in a physical complex with IPO7 during entry, HeLa cells were uninfected or infected with HPV16.L2F for 8, 16, and 24 hours. Cells were then harvested, and the resulting cell extract was subjected to IP using an anti-IPO7 antibody. The precipitated material was subjected to SDS-PAGE followed by immunoblotting. Anti-IPO7 pulled down L2-3xFLAG predominantly at 24 hpi, whereas low level background signals were detected in samples from earlier time points [or in the negative immunoglobulin G (IgG) control IP] (Fig. 4C, top). These findings suggest that IPO7 binds to HPV L2 at 24 hpi when the virus initiates nuclear entry, consistent with the PLA data (Fig. 4, A and B). To test whether the HPV L2–IPO7 interaction depends on mitosis, HeLa cells were treated with the control DMSO or the mitosis inhibitor RO3306 for 24 hours and the effect of the inhibitor on disrupting the cell cycle verified by flow cytometry (fig. S6B). DMSO- and RO3306-treated cells were then infected with HPV16.L2F for another 24 hours, and the co-IP experiments were carried out as above. We found a similar level of L2-3xFLAG co-IPed with IPO7 in the DMSO- and RO3306-treated cells (Fig. 4D, top), indicating that binding of IPO7 to HPV L2 does not require mitosis. In addition, because we rarely observe colocalization of IPO7 and HPV L2 on condensed chromosomes during mitosis (fig. S6C), most of the IPO7-L2 interaction likely occurs outside of the nucleus, consistent with the PLA results demonstrating that HPV L2 and IPO7 are proximal to each other mainly outside of the nucleus (Fig. 4A).
IPO7 was identified by MS as a host protein that specifically interacts with the FA-JX4 traptamer (Fig. 1A). We found that IP of IPO7 selectively pulled down FA-JX4 but not FA (fig. S6D, top) in HPV16.L2F-infected cells, in agreement with our IP-MS results (Fig. 1A). The IPO7/FA-JX4 interaction was also observed in uninfected cells. Because FA-JX4 localizes to the TGN/Golgi (fig. S1A) and traps the virus in this compartment (21), we hypothesized that FA-JX4 disrupts the ability of IPO7 to bind HPV L2. To test this, control FA- and FA-JX4–expressing cells were infected with HPV16.L2F PsV. Extracts prepared from cells 24 hpi were subjected to IP using an anti-IPO7 (or control IgG) antibody, and the precipitated material was evaluated by SDS-PAGE and immunoblotting for FLAG-tagged L2. The IPO7-HPV L2 interaction was readily detected in cells expressing the control protein FA but was markedly diminished in cells expressing FA-JX4 (fig. S6E, top). These results suggest that FA-JX4 inhibits the ability of HPV to engage IPO7, which is required to transport the virus to the nucleus, thus explaining, at least in part, how FA-JX4 impairs HPV infection.
The HPV L2–IPO7 interaction requires COPI
We previously reported that HPV recruits the COPI complex upon TGN arrival to enable transit across the Golgi stacks (24). The interaction between HPV L2 and COPI could be detected beginning at 16 hpi. In contrast, HPV L2 binds to IPO7 at 24 hpi, presumably after COPI has mediated the virus transport across the TGN/Golgi stacks. Therefore, we hypothesized that COPI is required for the HPV L2–IPO7 interaction. To test this, HeLa cells were transfected with Scr siRNA or siRNAs against COPA and COPG1 (to deplete the α- and γ1-COP subunits, respectively), followed by infection with HPV16.L2F. The cells were lysed at 24 hpi, and the resulting extracts were IPed with antibody recognizing IPO7. We found that anti-IPO7 pulled down L2-3xFLAG from extracts of control but not COPI KD cells (Fig. 4E, top), indicating that the IPO7-HPV L2 interaction requires COPI. In contrast, IPO7 depletion did not inhibit the HPV L2-COPI interaction (Fig. 4F, top): Anti–γ-COP immunoprecipitated HPV16 L2-3xFLAG at 24 hpi even in the absence of IPO7; the HPV L2-COPI interaction was modestly enhanced in the absence of IPO7. We did not detect an interaction between γ-COP and IPO7 in uninfected or infected cells (fig. S6F), suggesting that while COPI plays a role earlier in the process, it is not directly involved in IPO7-L2 binding. These findings strongly suggest that the HPV L2–IPO7 interaction occurs after COPI has facilitated the transit of HPV across the Golgi stacks.
IPO7 binds directly to the C-terminal NLS motif of HPV L2
We next determined the sequence in HPV L2 that interact with IPO7. IPO7 typically recognizes the NLS of cellular cargos to deliver them to the nucleus through the nuclear pore (53). Therefore, we hypothesized that HPV L2 engages IPO7 via an NLS in L2 to promote nuclear targeting of HPV. There are three predicted NLS motifs in HPV16 L2, each enriched with arginine (R) and lysine (K) residues that typically define an NLS motif (38, 41, 54). These putative NLS motifs are located at the N terminus (nNLS), central region (mNLS), and C terminus (cNLS) of L2 (Fig. 5A); the highly conserved cNLS, composed of six R/K amino acids in HPV16 (RKRRKR), is also termed the CPP because it exerts an intrinsic membrane penetration activity that mediates protrusion of L2 across the endosome membrane during the early steps of HPV entry (15). Because the nNLS is removed after cleavage by furin during early entry steps (55), this NLS is not likely to participate in nuclear transport events of the incoming HPV. We therefore tested whether the mNLS or the cNLS of L2 (Fig. 5A), which are expected to be exposed to the cytosol when L2 is inserted into the membrane (17, 18), interact with IPO7.
Fig. 5. IPO7 binds directly to the C-terminal NLS motif of HPV L2.
(A) Diagram of the NLSs in HPV16 L2. Three postulated NLSs are indicated as red boxes, with basic amino acids in the NLSs highlighted in red. The predicted transmembrane domain (TMD) spanning amino acids 45 to 67 is denoted as a black box. The RBS is indicated as a blue box, the amino acids required for retromer binding are highlighted in blue, and the corresponding RBS mutation to alanines are underlined in the DM peptide. The cNLS mutation to alanines are underlined in the 6A peptide. (B) Whole-cell extracts of uninfected HeLa cells were incubated with N-terminal biotin-tagged peptides containing residues S296–S316 or S434–R461 of HPV16 L2. The pull-down experiment without peptide was used as a negative control. Samples captured with streptavidin beads were analyzed by immunoblotting for IPO7 or β-actin. (C) Coomassie stain of the purified proteins. EGFP-FLAG, C-terminal FLAG-tagged EGFP. FLAG-IPO7, N-terminal FLAG-tagged IPO7. (D) N-terminal biotin-tagged peptides containing residues S434–R461 of HPV16 L2 (WT) and the corresponding DM and 6A mutant peptides as indicated in (A) were incubated with purified EGFP-FLAG or FLAG-IPO7. Samples captured with streptavidin beads were analyzed by Western blotting (WB) for FLAG. (E) HeLa cells were transfected with indicated DNA constructs for 24 hours to express WT HA-(Δ1–67) L2-3xFLAG (WT) or 6A HA-(Δ1–67) L2-3xFLAG (6A). The cells were lysed, and the resulting extracts were analyzed as in Fig. 4C. (F) HeLa cells treated with 20 nM Scr or KPNA2 siRNA were uninfected or infected with HPV16.L2F PsV (MOI, ~1). At 24 hpi, cells were lysed and analyzed as in Fig. 4E, except antibodies against FLAG, IPO7, KPNA2, or HSP90 (as a loading control) were used to analyze the input.
To test this possibility, we synthesized N-terminal biotinylated peptides composed of amino acids 296 to 316 of L2 [Biotin-(S296–S316) L2] that harbors the mNLS sequence or of amino acids 434 to 461 of L2 [Biotin-(S434–R461) L2] that contains the cNLS sequence (Fig. 5A). Extracts from uninfected HeLa cells were incubated with either of these peptides (or no peptide addition as the negative control), the peptides and any associated proteins were pulled down using streptavidin beads, and the precipitated material was subjected to SDS-PAGE and immunoblotting for IPO7. The peptide harboring cNLS [Biotin-(S434–R461) L2] but not the mNLS peptide pulled down IPO7 (Fig. 5B, top), whereas neither of the peptides pulled down the control protein β-actin (Fig. 5B, bottom). These results indicate that a short segment of L2 containing the cNLS is sufficient to pull down IPO7 from cell extracts.
To evaluate whether the IPO7-peptide interaction is direct, we performed a similar biotin-pull-down assay, with purified FLAG-tagged IPO7 (FLAG-IPO7) and the control EGFP-FLAG proteins instead of using the cell extract (Fig. 5C). We found that the Biotin-(S434–R461) L2 peptide (WT) effectively pulled down FLAG-IPO7 but not EGFP-FLAG (Fig. 5D), demonstrating that IPO7 binds to this peptide directly. However, if the basic RKRRKR cNLS sequence in the Biotin-(S434–R461) L2 peptide was mutated to AAAAAA generating the 6A mutant (Fig. 5A), then this peptide no longer pulled down FLAG-IPO7 (Fig. 5D). By contrast, if the retromer binding site (RBS) sequences (FYL/YYML) that lie upstream of the RKRRKR sequence (56) were mutated to A, generating the double mutant (DM) peptide unable to bind retromer, (Fig. 5A), this peptide demonstrated a similar ability to pull down FLAG-IPO7 as the WT peptide (Fig. 5D). We conclude that the highly basic cNLS/CPP at the C terminus of L2 is critical for engaging IPO7, consistent with a previous report showing that IPO7 recognizes sequences enriched in basic amino acids (53). In contrast, the RBS in this peptide is not required for IPO7 binding.
When HPV L2 is inserted into the endosome membrane during entry, most of L2 (i.e., sequences downstream of amino acid 67) is expected to be exposed to the cytosol (17, 18). We therefore tested whether an L2 protein (lacking the first 67 amino acids) fused to an HA-tag at its N terminus and a 3xFLAG-tag at its C terminus [i.e., WT HA-(Δ1–67) L2-3xFLAG], or the corresponding 6A mutant [i.e., 6A HA-(Δ1–67) L2-3xFLAG], can bind to IPO7. Cells expressing WT HA-(Δ1–67) L2-3xFLAG or 6A HA-(Δ1–67) L2-3xFLAG were lysed, and the extracts were subjected to IP using an anti-IPO7 antibody to pull-down endogenous IPO7 (Fig. 5E). Similar to the peptide binding results (Fig. 5D), we found that WT (but not 6A) HA-(Δ1–67) L2-3xFLAG co-IPed with IPO7. Because the highly basic cNLS/CPP motif resembles a classical NLS typically recognized by importin α, we asked whether the interaction between HPV16 L2 and IPO7 depends on KPNA2, an importin α previously implicated in HPV infection (36, 38). HeLa cells were treated with Scr or KPNA2 siRNA for 48 hours and either left uninfected or infected with HPV16.L2F. At 24 hpi, IP with an anti-IPO7 antibody showed that comparable amounts of L2-3xFLAG were coprecipitated in both control and KPNA2 KD cells (Fig. 5F), showing that KPNA2 is not required for the interaction between IPO7 and HPV16 L2. Together, these results support the hypothesis that the cNLS/CPP motif at the C terminus of HPV16 L2 is essential for the interaction of L2 with IPO7 during HPV cell entry.
DISCUSSION
Although HPV is responsible for ~5% of all human cancers, effective antiviral therapies remain elusive, due in part to an incomplete understanding of the virus entry mechanism. Early during entry, HPV undergoes endocytosis and reaches the endosome. In this compartment, the virus capsid protein L2 is inserted across the endosome membrane so that most of L2 is exposed to the cytosol (15, 17, 18). In this membrane-inserted topology, the cytosol-exposed region of L2 recruits critical cytosolic sorting factors that mediate proper trafficking to the nucleus for infection. For example, HPV exploits COPI, BicD2/dynein, and Rab6a to traffic from the TGN in a retrograde direction to cross the Golgi stacks (22–24). However, after traversing the Golgi, it is not known how HPV is targeted to the nucleus. Although several reports indicated that HPV can reach the ER (57–60), which is the typical destination of cellular cargos after retrograde Golgi transport, whether the ER is part of infectious entry of HPV is unknown. A recent study showed that impairing COPII-dependent ER trafficking only mildly disrupts HPV infection (61), suggesting that little HPV that reaches the ER and then exits this compartment via COPII-dependent transport likely only plays a minor role in HPV infection. In this study, we show that HPV relies on the Golgi-associated importin β member IPO7, a component of the nuclear pore import machinery, to target to the nucleus and cause infection. Depletion of IPO7 leads to the accumulation of HPV in the Golgi, preventing it from entering the nucleus and reaching mitotic chromosomes. In addition, we demonstrate that the Golgi-to-nucleus transport of cellular cargo also relies on IPO7, indicating that HPV exploits an existing pathway for its nuclear entry.
We previously reported that expression of an artificial transmembrane protein (named FA-JX4) traps HPV in the TGN/Golgi and blocks virus infection (21). On the basis of this observation, we hypothesized that FA-JX4 binds to a putative host factor that is required for infection, preventing the host factor from engaging HPV. Using an unbiased proteomic strategy, we identified IPO7 as an FA-JX4 interacting partner that supports infection. This binding is likely indirect because IPO7 is not a transmembrane protein. IPO7 was not identified in previous biochemical and genetic screens for HPV entry factors. Similarly, the Rab7 guanosine triphosphatase (GTPase) activating protein, TBC1D5, was identified as an HPV entry factor by studying a different inhibitory protein, FA-JX2, which traps incoming HPV in the endosome (42). These results emphasize the value of using orthogonal approaches to dissect HPV entry (62). The FA-JX4 IP-MS analysis revealed additional cellular components that may be involved in HPV infection. For example, members of the exportin family (including XPO1, CSE1L, XPOT, XPO4, XPO5, and XPO6) and the conserved oligomeric Golgi complex (such as COG1, COG2, COG3, COG5, COG6, and COG7) were also identified as FA-JX4–interacting partners (table S1). The potential roles of these host components in HPV infection require future investigation.
By using a cell fractionation approach and a Golgi-IP method, we found that a pool of IPO7 is associated with the Golgi membrane in cells. This finding is further supported by PLA analysis, which demonstrated that IPO7 is proximal to the Golgi marker GM130 in intact infected and uninfected cells. The strategic positioning of IPO7 at the Golgi is suited for HPV entry because incoming HPV traffics through the Golgi en route to the nucleus, and other manipulations that impair Golgi transit, such as COPI KD, inhibit infection. IPO7 lacks a transmembrane domain, so the mechanism by which IPO7 engages the Golgi membrane is unclear. It is possible that a Golgi-resident transmembrane/peripheral protein serves as an IPO7 binding partner, anchoring this importin β protein to the Golgi membrane. Localization of IPO7 to the Golgi likely serves a critical cellular function. This idea is supported by our finding that transport of the Golgi-derived soluble SREBP1 transcription factor to the nucleus depends on IPO7 and is consistent with previous studies demonstrating that importin β family proteins mediate nuclear entry of this family of transcription factors (49, 63, 64). Because IPO7 transports a cellular cargo from the Golgi to the nucleus, HPV exploits a preexisting Golgi-to-nucleus transport pathway during entry.
Functionally, we demonstrated that chemical inactivation of the importin β family proteins with IPZ robustly blocked HPV infection if added before 24 hpi, a time point when HPV has already initiated nuclear entry (26, 46). We then used RNA interference against importin β members that interacted with FA-JX4 to show that depletion of IPO7 (and other importin β family members) led to a substantial inhibition of HPV infection by different HPV types and in different cell lines, indicating that IPO7 plays a key role in infection. IPO5 but not IPO7 was identified in a large-scale siRNA screen for host factors involved in HPV infection (26). However, IPO5 KD disrupts the cell cycle (fig. S3C), consistent with a previous report (65), which raises the question of whether loss of IPO5 directly affects nuclear entry of HPV or rather inhibits HPV entry indirectly due to its impact on cell cycle progression.
We used labeled encapsidated DNA, immunofluorescent staining, and PLA for HPV capsid proteins to examine the fate of HPV in cells depleted of IPO7. These studies showed that in these cells, the virus is trapped in the Golgi and that the infectious virus cannot reach the nucleus. These observations provide a simple explanation for the inhibition of HPV infection by IPO7 KD and indicate that IPO7 is responsible for Golgi-to-nucleus transport of HPV. Published work from other laboratories showed that IPO7 also plays an important role in infection by other viruses. For example, IPO7 directly translocates the flavivirus core protein into the nucleus to promote infectious virus production, although viral genome replication occurs in the cytoplasm (66). Similarly, the retrovirus HIV-1 exploits IPO7 to maximize nuclear import of its genome (67), as well as to facilitate replication via interaction between the viral integrase and IPO7 (68).
Our time-course co-IP experiments demonstrated that HPV L2 is in a complex with IPO7 by 24 hpi, after the virus enters the TGN/Golgi, consistent with the idea that IPO7 acts after the virus has transited across the Golgi stacks. In contrast, HPV and retromer are in a complex by 8 hpi (19), and HPV and COPI are in a complex by 16 hpi (24). Our in vitro binding studies showed that IPO7 can bind to the cNLS/CPP located immediately downstream of the RBS of L2. However, cellular factors such as the retromer, Rab6a, and BicD2, which interact with the RBS at the endosome to ensure stable insertion of L2 and facilitate proper HPV trafficking to the TGN, may sterically hinder IPO7 from binding to the cNLS/CPP of L2. When most of the virus is localized at the endosome, IPO7 does not interact with HPV L2. These observations suggest that IPO7 binding during infection may depend on additional functions of L2 segments that become exposed to the cytoplasm only when the virus reaches the Golgi. Alternatively, L2/IPO7 binding may require other cellular proteins that are accessible only when the virus is at the Golgi. Depletion of COPI interfered with the ability of HPV L2 to interact with IPO7 in infected cells, suggesting that COPI-dependent transport of HPV from the TGN to the Golgi stacks and across the Golgi stack is upstream of IPO7 action during HPV entry.
IPO7 typically binds to highly basic NLSs, but our in vitro binding studies revealed that IPO7 does not bind to mNLS of L2. Recent work indicated that this element has only weak nuclear localization activity, and it has been reclassified as a nuclear retention or chromatin tethering signal (30, 41, 54). The inability of this element to bind IPO7 suggests that, in our assay, IPO7 is not binding promiscuously to all basic segments. The cNLS motif overlaps with a CPP sequence that mediates penetration of L2 across the endosome membrane (15), indicating that the C terminus of L2 is responsible for at least two independent steps during HPV entry. Because all known mutations in this basic segment inhibit L2 membrane protrusion and prevent entry of HPV into the retrograde pathway, we cannot assess the effects of these mutations on IPO7 binding or late entry events in infected cells. Our finding that FA-JX4 interferes with IPO7-HPV binding provides a simple explanation for the inhibitory effect of the traptamer on HPV trafficking.
Although classic nuclear entry of small cellular cytosolic cargoes typically relies on an importin α and β heterodimer for delivery to the nuclear pore (37), our MS analysis did not identify any importin α family members as FA-JX4–binding partners, suggesting that IPO7 targets HPV to the nucleus without assistance from importin α proteins. Importin β family proteins have been shown to be sufficient to drive nuclear entry of some cellular proteins, such as SREBPs (49, 63, 69–72). Although a previous study suggested that the importin α member KPNA2 supports HPV infection (36), the mechanism by which KPNA2 assists in HPV nuclear entry remains unclear. RanBP10, a protein associated with KPNA2, is known to facilitate the transport of HPV, presumably in a vesicle-bound form, along microtubules to mitotic chromatin (36). While our IP data indicate that KPNA2 is not involved in the interaction between HPV and IPO7, it remains possible that upon Golgi exit, IPO7 cooperates with KPNA2 and RanBP10 to target HPV to the nucleus. This question deserves future investigation.
NEB during mitosis is thought to enable entry of HPV into the nucleoplasm (25–30). However, it is largely unknown whether NEB-dependent nuclear entry of HPV involves the classic nuclear import machinery, including importin α/β, the Ran GTPase, or nucleoporins (NUPs) of the nuclear pore complex (NPC). NUPs appear to be dispensable for HPV infection because depletion of NUP153, the essential component of the NPC basket (73), does not affect HPV16 infection (26). Our study suggests that at least one component of the nuclear import machinery, IPO7, plays a pivotal role during or before NEB-mediated nuclear entry of viral components. We postulate that during HPV entry, the importin β member IPO7 targets the virus to the nuclear envelope without using the channel function of the NPC to gain nuclear entry. Instead, we propose that the IPO7-dependent targeting step positions the virus next to the nucleus so that upon NEB during mitosis, the virus can readily enter the nucleus. The association of the L2 protein with chromatin then traps the L2 protein and associates viral DNA in the nucleus when the nuclear envelope reforms (30, 41). IPO7 likely dissociates from L2 before HPV reaches the condensed mitotic chromosomes because our PLA results imply that the IPO7-L2 interaction occurs predominantly outside the nucleus, and most HPV L2 on mitotic chromosomes does not colocalize with IPO7, although rare instances of colocalization were observed (with an example shown in fig. S6C). This suggests that most of IPO7-L2 interaction occurs before nuclear arrival of HPV. However, if IPO7 helps to deliver HPV to the condensed chromosomes during mitosis, IPO7 may rapidly release HPV L2 upon chromosome tethering. Our results raise the possibility that some cellular cargos also use a combination of the canonical nuclear import machinery and NEB to gain nuclear entry.
MATERIALS AND METHODS
Antibodies and inhibitors
Antibodies and inhibitors used in this study are listed in Table 1.
Table 1. Antibodies and inhibitors.
IF, immunofluorescence; N/A, not applicable.
| Antibodies | |||
|---|---|---|---|
| Antigen | Species | Catalog no./source | Application |
| IPO1 | Rabbit | 10077-1-AP; Proteintech | WB |
| IPO4 | Rabbit | 11679-1-AP; Proteintech | WB |
| IPO5 | Mouse | sc-55527; Santa Cruz Biotechnology | WB |
| IPO7 | Mouse | sc-365231; Santa Cruz Biotechnology | WB, IF, PLA, and IP |
| IPO7 | Rabbit | 28289-1-AP; Proteintech | IP and PLA |
| IPO8 | Mouse | sc-398854; Santa Cruz Biotechnology | WB |
| IPO9 | Rabbit | SAB4200155; Millipore Sigma | WB |
| GRASP55 | Mouse | sc-271840; Santa Cruz Biotechnology | WB |
| FLAG | Mouse | F3165; Millipore Sigma | WB, IF, IP, and PLA |
| FLAG | Rabbit | 14793S; Cell Signaling | PLA |
| FLAG | Rabbit | F7425; Millipore Sigma | IF and WB |
| γ-COP | Rabbit | 12393-1-AP; Proteintech | WB and IP |
| PS1 | Rabbit | 5643S; Cell Signaling | WB |
| TGN46 | Rabbit | 13573-1-AP; Proteintech | WB |
| GM130 | Rabbit | ab52649; Abcam | WB, IF, PLA, and IP |
| EEA1 | Rabbit | C45B10; Cell Signaling | WB |
| BAP31 | Rat | MA3-002; Invitrogen | WB, IP, and IF |
| PDI | Mouse | ab2792; Abcam | WB |
| Histone H3 | Rabbit | 9715S; Cell Signaling | WB |
| GFP | Mouse | 632380; Takara Bio Clontech | WB |
| β-Actin | Rabbit | 4967S; Cell Signaling | WB |
| HSP90 | Mouse | sc13119; Santa Cruz Biotechnology | WB |
| HA | Rat | ROAHAHA; Roche | WB and IP |
| Nesprin-2 | Rabbit | ab314872; Abcam | IF |
| NUP188 | Rabbit | NBP1-28748; Novus | WB |
| KPNA2 | Rabbit | 10819-1-AP; Proteintech | WB |
| Giantin | Rabbit | 22270-1-AP; Proteintech | IF |
| Other antibodies | Species | Catalog no./source | Application |
| Anti-mouse IgG peroxidase | Goat | A4416; Millipore Sigma | WB |
| Anti-mouse IgG peroxidase | Goat | W4021; Promega | WB |
| Anti-rabbit IgG peroxidase | Goat | A4914; Millipore Sigma | WB |
| Anti-rabbit IgG peroxidase | Goat | W4011; Promega | WB |
| Anti-rat IgG peroxidase | Rabbit | A5795; Millipore Sigma | WB |
| Clean-blot IP detection reagent (HRP) | N/A | 21230; Thermo Fisher Scientific | WB |
| Anti-mouse Alexa Fluor 488 | Goat | A11029; Thermo Fisher Scientific | IF |
| Anti-mouse Alexa Fluor 594 | Goat | A11032; Thermo Fisher Scientific | IF |
| Anti-mouse Alexa Fluor 647 | Goat | A32728; Thermo Fisher Scientific | IF |
| Anti-rabbit Alexa Fluor 568 | Donkey | A10042; Thermo Fisher Scientific | IF |
| Anti-rabbit Alexa Fluor 594 | Goat | A11037; Thermo Fisher Scientific | IF |
| Anti-rabbit Alexa Fluor 647 | Goat | A21244; Thermo Fisher Scientific | IF |
| Anti-rat Alexa Fluor 647 | Goat | A21247; Thermo Fisher Scientific | IF |
| Normal rabbit IgG | Rabbit | 30000-0-AP; Proteintech | IP |
| Inhibitors | |||
| Compound | Solvent | Catalog no./source | Concentration |
| XXI (inhibiting γ-secretase) | DMSO | 565790; Millipore Sigma | 1 μM |
| Importazole (inhibiting importin β) | DMSO | 401105; Millipore Sigma | 1–10 μM |
| RO3306 (inhibiting CDK1) | DMSO | S7747; Selleckchem | 9 μM |
DNA constructs
For PsV production, the p5sheLL.L2F, p16sheLL.L2F, and p18sheLL.L2F constructs were modified from p5sheLL, p16sheLL, and p18sheLL plasmids (gifts from J. Schiller, National Cancer Institute, Rockville, MD; Addgene plasmids #46953, #37320, and #37321), respectively, as previously described (24). The pcDNA3.1 plasmid expressing GFP with a C-terminal S-tag was used as the pseudoviral genome (reporter plasmid) as previously described (24, 74). The pCINeo-Gluc construct generated in the previous study (24) was used as the reporter of HPV16 PsV for the PLA assay and EdU staining. For KD-rescue experiments, pCMVTNT-HA-mCherry generated in the previous study (24) was used as the control vector. The IPO7 coding sequence was amplified from pUC19-mIPO7 (MG5A1798-U, Sino Biological) and cloned into the pCMVTNT-HA-COPG1-mCherry plasmid (24) to replace COPG1. Site-directed mutagenesis was used to introduce siRNA-resistant mutations in the IPO7 siRNA #2 target site. For recombinant protein production, the siRNA-resistant IPO7 coding sequence was amplified from pCMVTNT-HA-IPO7-mCherry and cloned into pFLAG-CMV2 vector in-frame with the 5′ FLAG to encode FLAG-IPO7. The SREBP1 nuclear import reporter construct pcDNA3.1-EF1A-mNeonGreen2x-SREBP1-P2A-Puro (Addgene plasmid #228401) was described previously (51). pCMVTNT-HA-(Δ1–67) L2-3xFLAG construct was generated in the previous study (24), and the 6A mutation replacing RKRRKR was introduced by site-directed mutagenesis. pcDNA3.1(-)-EGFP-FLAG was described previously (75). All plasmids constructed in this study were verified by sequencing.
Cell culture
HeLa [American Type Culture Collection (ATCC), catalog no. CCL-2], HeLa-S3 (ATCC, catalog no. CCL-2.2), and SiHa (ATCC, catalog no. HTB-35) cells were obtained from ATCC. HaCaT cells were purchased from AddexBio Technologies. Human embryonic kidney (HEK) 293TT cells were obtained from C. Buck (National Cancer Institute, Rockville, MD). HeLa cells (ATCC, catalog no. CCL-2) were used throughout this study unless otherwise specified. HeLa-tTA clonal cells derived from HeLa S3 stably expressing FA or FA-JX4 were described previously (21). All cell lines were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS), penicillin, and streptomycin and incubated at 37°C and 5% CO2. Cell lines were authenticated by using the ATCC cell line authentication service.
HPV PsV production
HPV PsVs were produced by cotransfecting HEK 293TT cells with a reporter plasmid and p16sheLL, p16sheLL.L2F, p5sheLL.L2F, or p18sheLL.L2F using polyethyleneimine (PEI; Polysciences). Packaged PsVs were purified by density gradient centrifugation in OptiPrep (Millipore Sigma) as described (76, 77). Purified PsVs were subjected to SDS-PAGE followed by staining using SimplyBlue SafeStain (Invitrogen) to assess the quality and quantity of L1 and L2. PsV produced using the standard protocol consistently achieves infection rates of around 90% when 10 μl of a typical PsV preparation is used in a 12-well format, suggesting an approximate ratio of one infectious virus per 10 particles.
Immunoprecipitation and mass spectrometry
Cells stably expressing FA or FA-JX4 were seeded in 15-cm plates and grown to ~80% confluency and infected with HPV16 PsV (containing ~25 μg L1). At 24 hpi, cells from three 15-cm plates were pooled and lysed with 1% Triton X-100 in HN buffer [50 mM Hepes, 150 mM NaCl, and 1 mM phenylmethylsulfonyl fluoride (PMSF)] on ice for 15 min. The resulting extracts were centrifuged at 16,100g at 4°C for 10 min, and the supernatants were then incubated with anti-FLAG M2 antibody (~8 μg antibody per 1 ml of lysate) (F3165; Millipore Sigma) at 4°C overnight. The immune complex was captured with protein G–coated magnetic beads (Invitrogen, 10003D) at 4°C for 2 hours. Beads were washed on ice four times with 0.1% Triton X-100 in HN buffer. Bound proteins were eluted twice with 3xFLAG peptide (250 μg/ml; Millipore Sigma) in HN buffer containing 0.1% Triton X-100 at room temperature for 30 min. After the 3xFLAG peptide elution, the beads were incubated with 1× SDS sample buffer at 95°C for 10 min to elute the material still bound to the beads. A portion of the 3xFLAG peptide eluate was analyzed by SDS-PAGE followed by immunoblotting to confirm that equivalent amounts of traptamers (FA or FA-JX4) were precipitated. The remaining eluate was treated with 10% trichloroacetic acid and incubated on ice for 10 min. The sample was subjected to centrifugation, and the precipitated material washed twice with ice-cold acetone. The precipitate was subject to mass spectrometry analysis at the Taplin Mass-Spectrometry Core Facility (Harvard Medical School). Liquid chromatography tandem mass spectrometry was performed using an Orbitrap mass spectrometer (Thermo Fisher Scientific).
HPV infectivity
HeLa, SiHa, or HaCaT cells were treated as indicated and infected with HPV16.L2F, HPV5.L2F, or HPV18.L2F PsV [multiplicity of infection (MOI), ~0.3] containing a GFP reporter construct. Where indicated, inhibitors (1 μM XXI or 1 to 10 μM IPZ) were added at the time of infection or after infection at 8-hour time intervals. An equivalent volume of the carrier solvent DMSO (0.1%, v/v) was added to the control cultures. At 48 hpi, cells were washed with phosphate-buffered saline (PBS) and lysed in HN buffer containing 1% Triton X-100 on ice for 10 min. Cells were centrifuged at 16,100g at 4°C for 10 min. The resulting supernatant was incubated with SDS sample buffer containing 2-mercaptoethanol, denatured by incubating at 95°C for 10 min, and analyzed by SDS-PAGE followed by immunoblotting for GFP-S expression using an antibody recognizing GFP. Alternatively, infected cells were harvested at 48 hpi by trypsinization, resuspended in ice-cold PBS containing 2% FBS and DAPI (0.1 μg/ml), followed by the flow cytometry analysis performed with a Bio-Rad ZE5 cell analyzer (University of Michigan Flow Cytometry Core Facility). After gating for size and singlets, the population of DAPI-negative cells (~1 × 104 live cells) was analyzed for GFP fluorescence. For KD-rescue experiments, trypsin without phenol red was used to prevent interference of mCherry signals, and the infectivity was determined as GFP-positive population among mCherry-positive cells.
Immunoblotting
Protein samples were separated via SDS-PAGE and transferred to nitrocellulose membranes (Amersham, Millipore Sigma). The membranes were then subjected to incubation with primary antibodies in tris-buffered saline (TBS) supplemented with 3% skim milk and 0.2% Tween 20 at 4°C overnight and then to incubation with horseradish peroxidase (HRP)–conjugated secondary antibodies or Clean-blot IP detection reagent (HRP) (Pierce, Thermo Fisher Scientific) in TBS/3% skim milk/0.2% Tween 20 at room temperature for 1 hour. HRP substrates (Immobilon, Millipore Sigma) were used to generate the chemiluminescence signals that were detected by exposing to x-ray films (Ece Scientific, Thermo Fisher Scientific).
siRNA transfection
siRNA oligos used in this study are listed in the Table 2. Twelve-well plates were seeded with 7 × 104 HeLa, 4 × 105 HeLa S3, or 5 × 104 HaCaT cells per well and simultaneously reverse-transfected with 10 nM of indicated siRNA by using Lipofectamine RNAiMAX (Thermo Fisher Scientific) according to the manufacturer’s instructions.
Table 2. siRNA oligos for gene KDs.
| siRNA | Sequence | Supplier |
|---|---|---|
| Scr | Predesigned | QIAGEN; Catalog no. 1027281 |
| IPO1 | #1 CUGCUUUACAGAAUCUGGU | Sigma-Aldrich |
| #2 CUGUUAAGGAUGUUCCAAA | ||
| IPO4 | #1 CUGUACUGGGCCUCUCCUA | Sigma-Aldrich |
| #2 GCAACAACACUAUUUAGAU | ||
| IPO5 | #1 CUCUACAGCUAAGUCUAAA | Sigma-Aldrich |
| #2 CUUAAUGUCACCCACCAGA | ||
| IPO7 | #1 GCAAUUGCAGCUUUGUAUU | Sigma-Aldrich |
| #2 GCAAUAUAUGGCUCCUCGA | ||
| IPO8 | #1 CAAUUGCUGCCUUGUACUA | Sigma-Aldrich |
| #2 GAUAUUCCUGCCUCGUAUU | ||
| IPO9 | #1 GGUUUAAGAUGGAGGUCCU | Sigma-Aldrich |
| #2 CAAACCUGCUCUAGAGUUU | ||
| PS1 | UCAAGUACCUCCCUGAAUG | Sigma-Aldrich |
| COPA | CCAUUGAUCCCACUGAGUUCA | Sigma-Aldrich |
| COPG1 | CUUGUAAUCUGGAUCUGGA | Sigma-Aldrich |
| KPNA2 | GGUGAAACUGAGAAACUUA | Sigma-Aldrich |
DNA transfection
For KD-rescue experiments, HeLa cells were grown to ~80% confluency in 10-cm plates and then transfected with 3 μg of pCMVTNT-HA-mCherry or 8 μg of pCMVTNT-HA-IPO7-mCherry plasmids for 24 hours using the transfection reagent FuGENE HD (Promega). The cells were then trypsinized, and 1.5 × 105 cells per well were seeded in six-well plates and reverse-transfected with 10 nM Scr siRNA or IPO7 siRNA #2 using the transfection reagent Lipofectamine RNAiMAX (Thermo Fisher Scientific) for 48 hours. Attached cells were washed with PBS twice and infected with HPV16.L2F (MOI, ~0.3) for another 48 hours. Infectivity was analyzed by flow cytometry to determine the fraction of GFP-expressing cells among mCherry-expressing cells. A portion of cells was collected for immunoblot analyses to verify the expression of rescue constructs and IPO7 KD. For recombinant protein production, HEK 293TT cells were seeded in 10-cm plates for 24 hours to reach ~80% confluency and then transfected with 5 μg of indicated plasmids for 48 hours using PEI. For the expression of mNG2-SREBP1, 1 × 104 HeLa cells were seeded onto glass coverslips in 24-well plates for 24 hours and transfected with 0.5 μg of pcDNA3.1-EF1A-mNeonGreen2x-SREBP1-P2A-Puro by using PEI for 48 hours. Cells were then transfected for 48 hours with 10 nM Scr siRNA or IPO7 siRNA #2 by using the transfection reagent Lipofectamine RNAiMAX. For the expression of HA-(Δ1–67) L2-3xFLAG, HeLa cells were seeded in 10-cm plates for 24 hours to reach ~80% confluency, and then PEI was used to transfect the cells for 24 hours with 2.5 μg of pCMVTNT-HA-(Δ1–67) L2-3xFLAG or 5 μg of pCMVTNT-HA-(Δ1–67) L2-3xFLAG-6A.
Cell cycle analysis
DNA content was measured using Hoechst 33342 staining and flow cytometry analysis. Nonpermeabilized cells were incubated with Hoechst 33342 (10 μg/ml; Biotechne) at 37°C for 30 min. Cells were then washed with PBS, harvested by trypsinization, and resuspended in ice-cold PBS containing 2% FBS. Flow cytometry was performed with a Bio-Rad ZE5 cell analyzer (University of Michigan Flow Cytometry Core Facility). The percentage of cells in G1, S, or G2/M phases was determined from control cell cultures using FlowJo software (BD).
Sucrose gradient fractionation
Sucrose gradient fractionation was carried out to isolate TGN/Golgi-containing cellular fractions as described (24). Briefly, HeLa cells were plated in 15-cm plates and grown to ~80% confluency. Two 15-cm plates per preparation were processed as follows. Cells were trypsinized and resuspended in HBS buffer [10 mM Hepes (pH 7.2), 1 mM Mg(OAc)2, 0.25 M sucrose, 1 mM EDTA, 1 mM dithiothreitol (DTT), and 1 mM PMSF] and homogenized by passing 31 times through an 8-μm clearance ball bearing homogenizer on ice (Isobiotech). The homogenate was centrifuged at 2000g for 30 min at 4°C, and the resulting supernatant (~1.2 ml) was layered over a sucrose gradient composed of 1 ml of 1.6 M, 1 ml of 1.4 M, 2 ml of 1.2 M, 3 ml of 1.0 M, 2 ml of 0.8 M, and 1 ml of 0.5 M sucrose in magnesium, EGTA and β-glycerophosphate (MEB) buffer [50 mM tris-HCl (pH 7.4), 50 mM KCl, 20 mM β-glycerophosphate, 15 mM EGTA, 10 mM MgCl2, and 1 mM DTT] supplemented with 250 mM KCl. The sucrose gradients were ultracentrifuged at 30,000 rpm in a SW41Ti swinging-bucket rotor (Beckman) at 4°C for 20 hours. After centrifugation, 11 fractions were collected from top to bottom. A portion of each fraction was analyzed by SDS-PAGE followed by immunoblotting to identify the TGN/Golgi-containing fractions.
Proximity ligation assay
A total of 2 × 105 HeLa S3 cells were seeded onto glass coverslips in a 24-well plate. For IPO7 KD experiments, cells were transfected with 10 nM siRNA for 48 hours. The cells were infected with HPV16.L2F PsV (MOI, ~100) or left uninfected. At 24, 30, or 32 hpi, cells were fixed with 4% paraformaldehyde (PFA), permeabilized with 0.1% saponin, and incubated at 4°C overnight with the combination of rabbit and mouse antibodies as listed in Table 3.
Table 3. Combination of antibodies used in PLA.
| PLA targets | Rabbit antibody | Mouse antibody |
|---|---|---|
| IPO7 and HPV16 L2-3xFLAG | Anti-FLAG (1:200 dilution) | Anti-IPO7 (1:100 dilution) |
| GM130 and IPO7 | Anti-GM130 (1:200 dilution) | Anti-IPO7 (1:100 dilution) |
| GM130 and HPV16 L2-3xFLAG | Anti-GM130 (1:200 dilution) | Anti-FLAG (1:500 dilution) |
PLA was performed with Duolink reagents (Millipore Sigma) as described previously (78). Briefly, cells were incubated with PLA probes in a humidified chamber at 37°C for 1 hour. Ligation was performed at 37°C for 45 min, and amplification was performed for 3 hours at 37°C. Coverslips were mounted with mounting medium containing DAPI (Abcam, ab104139) and visualized by confocal fluorescence microscopy (Zeiss LSM 980 or LSM 800). Images from single Z-planes were processed by ZEN software (Zeiss) and quantitatively analyzed by Fiji software (79) to measure the total fluorescence intensity per cell in each sample. All PLA experiments were performed independently at least two (in most cases three) times with similar results.
EdU labeling and immunofluorescence
To produce EdU-labeled HPV16 PsV, 100 μM EdU (Thermo Fisher Scientific, C10337) was added to the HEK 293TT cell growth medium at 6 hours after transfection of p16SheLL.L2F and the Gluc reporter plasmid. PsV was purified from transfected cells as described above. HeLa cells were uninfected or infected (MOI, ~150) with EdU-labeled HPV16.L2F PsV. At 32 hpi, cells were fixed in 4% PFA and permeabilized with 1% saponin. The cells were incubated with the Click-iT reaction mixture using the Alexa Fluor 488 Imaging Kit (Thermo Fisher Scientific, C10425) according to the manufacturer’s instruction. The cells were then incubated with antibodies recognizing anti–Nesprin-2 antibody and the appropriate secondary antibodies.
For immunofluorescence of GM130 and the traptamer FA or FA-JX4, 1 × 105 HeLa S3 or cells stably expressing FA or FA-JX4 per well were seeded onto glass coverslips in a 12-well plate for 48 hours. For immunofluorescence of L2-3xFLAG, IPO7, Giantin, and/or GM130, 7 × 104 HeLa cells per well were reverse-transfected with 10 nM Scr or IPO7 siRNA #2 and seeded onto glass coverslips in a 12-well plate for 48 hours. For immunofluorescence of BAP31 in mNG2-SREBP1–expressing cells, 1 × 104 HeLa cells per well were seeded onto glass coverslips in 24-well plates for 24 hours and transfected with 0.5 μg of pcDNA3.1-EF1A-mNeonGreen2x-SREBP1-P2A-Puro for 48 hours, followed by another transfection with 10 nM Scr siRNA or IPO7 siRNA #2 for 48 hours. Cells were left uninfected or infected with HPV16.L2F PsV and fixed at indicated time point with 4% PFA for 10 min, permeabilized, and blocked in DMEM supplemented with 0.1% Triton X-100 and 10% FBS for 30 min at room temperature. Samples were incubated overnight at 4°C with indicated antibodies in DMEM/0.1% Triton X-100/10% FBS. Alexa Fluor–conjugated secondary antibodies (Life Technologies) were 1:2000 diluted in DMEM/0.1% Triton X-100/10% FBS and incubated at room temperature for 1 hour.
Coverslips were mounted with mounting medium containing DAPI (Abcam, ab104139) and visualized by confocal fluorescence microscopy (Zeiss LSM 800 confocal laser scanning microscope with a Plan-Apochromat 40×/1.4 oil differential interference contrast M27 objective or Leica Stellaris 5 confocal microscope with HC PL APO CS2 63×/1.4 oil immersion objective lens). Confocal images from single Z-planes are presented throughout the manuscript.
Immunoprecipitation
For Golgi membrane IP, cell fraction #3 collected from the sucrose gradient was mixed with an anti-GM130 antibody (~1 μg/μl) or a nonspecific rabbit normal IgG antibody (~1 μg/μl) and incubated at 4°C overnight. Where indicated, the overnight incubation was supplemented with 1% Triton X-100. Protein G–coated magnetic beads (Thermo Fisher Scientific, 10003D) were preequilibrated with low-DTT HBS buffer [10 mM Hepes (pH 7.2), 1 mM Mg(OAc)2, 0.25 M sucrose, 1 mM EDTA, 0.1 mM DTT, and 1 mM PMSF] and used to capture the immune complex at 4°C for 30 min. After incubation, the beads were washed on ice once with low-DTT HBS buffer, three times with low-DTT HBS buffer supplemented with 300 mM NaCl, and once with low-DTT HBS buffer. The samples were eluted with 1× SDS sample buffer containing 2-mercaptoethanol at 95°C for 10 min. As a control experiment, a rat anti-BAP31 antibody (~1 μg/μl) was used to isolate the ER membrane, with a rat anti-HA antibody used for the control IP.
For IP of IPO7 from the whole cell extract, cells were grown to ~80% confluency in 10-cm plates and infected with HPV16.L2F PsV (MOI, ~1) or left uninfected. At the indicated time after infection, cells from one 10-cm plate were washed with PBS three times and lysed in 400 μl of 1% Triton X-100 in HN buffer and then incubated on ice for 10 min. After centrifugation at 16,100g at 4°C for 10 min, the resulting supernatant was incubated with anti-IPO7 antibodies (~25 μg/ml) at 4°C overnight. The immune complex was captured with protein G–coated magnetic beads (Thermo Fisher Scientific, 10003D; or New England Biolabs, S1430S) at 4°C for 30 min. Beads were washed on ice once with 1% Triton X-100 in HN buffer, followed by one wash with 1% Triton X-100 in high-salt HN buffer (50 mM Hepes, 500 mM NaCl, and 1 mM PMSF), two washes with 1% Triton X-100 in HN buffer, and elution in 1× SDS sample buffer containing 2-mercaptoethanol at 95°C for 10 min. For IPO7 IP experiments following COPA/G1 KD, cells were treated with 10 nM Scr siRNA or a mixture of 5 nM COPA and 5 nM COPG1 siRNAs for 24 hours before infection. For IPO7 IP experiments following KPNA2 KD, cells were treated with 20 nM Scr or KPNA2 siRNA for 48 hours before infection. γ-COP IP experiments were performed as previously described (24), except cells were transfected with 10 nM Scr or IPO7 siRNA #2 for 48 hours before infection. For IPO7 IP experiments after RO3306 treatment, cells were grown to ~80% confluency in 10-cm plates and pretreated with 9 μM RO3306 dissolved in DMSO or DMSO alone (0.09%, v/v) for 24 hours. Cells were then infected or left uninfected for 24 hours with RO3306 or DMSO maintained in the medium and harvested for IPO7 IP as described above. For IPO7 IP experiments following ectopic expression of HA-(Δ1–67) L2-3xFLAG (WT or 6A), cells were transfected with the indicated construct for 24 hours and then harvested for IPO7 IP as described above, except that after capturing the immune complexes, the beads were washed on ice once with 1% Triton X-100 in HN buffer, followed by four washes with 1% Triton X-100 in high-salt HN buffer before elution.
Biotin peptide pull-down assay
Peptides consisting of the HPV16 L2 sequence were synthesized with an N-terminal biotin tag (GenScript). Peptides were dissolved in DMSO or water according to the manufacturer’s instruction, and peptide stocks (5 mg/ml) were aliquoted and stored at −80°C. A total of 1 × 106 HeLa cells were plated in 6-cm dishes. Twenty-four hours later, cells were collected by trypsinization, washed with PBS, and lysed with 165 μl of Hepes buffer [1% Triton X-100, 20 mM Hepes (pH 7.4), 50 mM NaCl, 5 mM MgCl2, 1 mM DTT, and 1 mM PMSF]. The lysate was incubated on ice for 45 min followed by centrifugation at 14,000g at 4°C for 15 min. The resulting supernatant was incubated with 10 μg of indicated biotinylated peptide or an equivalent volume (2 μl) of DMSO as a negative control at 4°C for 2 hours. Streptavidin magnetic beads (Pierce, 88817) preequilibrated with Hepes buffer were added to the lysates and incubated at 4°C for 1 hour. The beads were captured and washed on ice three times with Hepes buffer and incubated at 95°C for 10 min in 1× SDS sample buffer containing 2-mercaptoethanol. For the biotin peptide pull-down assay using purified FLAG-IPO7, the same conditions were used except ~100 ng of FLAG-IPO7 or ~80 ng of EGFP-FLAG instead of cellular extracts was incubated with 10 μg of indicated biotinylated peptide and captured with streptavidin magnetic beads.
Preparation of recombinant proteins
Two 10-cm plates of HEK 293TT cells transfected with an expression plasmid for 48 hours were harvested and lysed in 1.2 ml of HN buffer containing 1% Triton X-100 on ice for 20 min. Following centrifugation at 16,100g at 4°C for 10 min, the resulting supernatant was incubated with anti-FLAG M2 antibody–conjugated agarose beads (Millipore Sigma, A2220) at 4°C for 2 hours. The beads were recovered by centrifugation and washed on ice twice with HN buffer containing 1% Triton X-100 supplemented with 300 mM NaCl and 1 mM adenosine triphosphate, once with HN buffer containing 1% Triton X-100, and then once with HN buffer containing 0.1% Triton X-100. The recombinant proteins were eluted twice with 3xFLAG peptide (150 μg/ml; Millipore Sigma) in HN buffer containing 0.1% Triton X-100 at room temperature for 30 min. A portion of the sample was analyzed with BSA as the standard by SDS-PAGE followed by SimplyBlue SafeStain (Invitrogen) to assess the quality and quantity of the purified recombinant proteins.
Statistical analysis
Quantitation of HPV infectivity by flow cytometry
Flow cytometry data were processed by Everest software (Bio-Rad). After gating for size and singlets, the population of DAPI-negative cells (~1 × 104 cells per condition) was analyzed for GFP or mCherry fluorescence. Data are presented as the means and SDs of three independent experiments. A two-tailed, unequal variance t test was used to determine statistical significance.
Quantitation of cell cycle by flow cytometry
DNA content was measured by Hoechst 33342 staining. The percentage of cells in G1, S, or G2/M phases was determined from control cell cultures using FlowJo software (BD). Data from three independent experiments are presented as the percentage of cells in each phase. A two-tailed, unequal variance t test was used to determine statistical significance for each cell cycle phase compared to control cells.
Quantitation of PLA and EdU images
For EdU images, the total area of each cell was manually defined using Fiji software (79) based on the baseline fluorescence in EdU staining. Nesprin-2 staining identified the nuclear membrane to define the nuclear area. The extranuclear fraction was determined by subtracting the nuclear area from the total area. EdU fluorescence intensity of nuclear, extranuclear, and total fractions was then measured using Fiji software on a per-cell basis. PLA images were also analyzed using Fiji software to measure the total fluorescence intensity per cell in each sample. Approximately 80 or more cells from 10 representative images were quantified for each condition. All PLA and EdU-staining experiments were performed independently at least two times with similar results. Because it is difficult to directly compare the quantitative results between experiments, results from one representative independent experiment are shown with data points, means, and SDs. One-way analysis of variance (ANOVA) was used to determine statistical significance. Data visualization and graphing were performed using Prism 10.4.2 (GraphPad Software).
Quantitation of mNG2-SREBP1 nuclear import
CellProfiler (80) was used to identify fluorescence signals for mNG2-SREBP1, as well as for non-nuclear (highly saturated BAP31 immunostaining) and nuclear (DAPI) areas. After color separation, the “IdentifyPrimaryObjects” module detected these signals based on predefined size and intensity thresholds. Manual selection was used to identify mNeonGreen-positive cells within the field for quantification. The “RelateObjects” module then associated the target signals with either non-nuclear or nuclear regions in each mNG2-SREBP1–expressing cell, while the “MeasureObjectIntensity” module retrieved the fluorescence intensity of mNG2-SREBP1. The ratio of fluorescence intensity of mNG2-SREBP1 in the nucleus to that in the combined non-nuclear and nuclear areas (whole cell) in at least 26 cells was then plotted using Prism 10.4.2 (GraphPad Software). A two-tailed, unequal variance t test was used to determine statistical significance.
Quantitation of IPO7 subcellular localization in Golgi and nucleus
CellProfiler was used to evaluate IPO7 localization within the nucleus and cis-Golgi compartments from multichannel fluorescence images. The DAPI channel was used to segment nuclei with the IdentifyPrimaryObjects module, while whole-cell outlines were identified on the basis of highly saturated GM130 staining following global thresholding and object identification. Cells touching the image border were automatically excluded from further analysis to ensure that only intact cells were analyzed. Nuclei were associated with their corresponding cell outlines using the RelateObjects module.
Golgi association was assessed by quantifying colocalization between IPO7 and GM130 signals within individual cells using the “MeasureColocalization” module. Manders’ colocalization coefficient, calculated with Costes automatic thresholding, was used as the metric for analysis. At least 147 cells from more than 20 images per condition were analyzed. For nuclear localization, the MeasureObjectIntensity module was used to quantify IPO7 fluorescence within individual nuclei, and intensity values were normalized to nuclear area to yield mean nuclear IPO7 intensity. Only nuclei with an area greater than 35,000 pixel2 (at 0.071 μm per pixel using a 63× objective) were included in the analysis to ensure that only intact, regular nuclei were quantified. At least 148 nuclei from more than 20 images per condition were analyzed. Statistical significance was determined using a two-tailed, unequal variance t test, and the results were plotted using Prism 10.4.2.
Quantitation of FLAG intensity on mitotic chromosomes
CellProfiler was used to identify immunofluorescence signals for FLAG (L2), as well as for nonchromosomal (highly saturated GM130 signals) and chromosomal (DAPI) areas. After color separation, the IdentifyPrimaryObjects module detected these signals based on predefined size and intensity thresholds. Manual selection was used to identify mitotic cells based on the presence of condensed chromosomes in DAPI staining within the field for quantification. The RelateObjects module then associated the target signals with either nonchromosomal or chromosomal regions in each mitotic cell, while the MeasureObjectIntensity module retrieved the immunofluorescence intensity of FLAG. The ratio of the immunofluorescence intensity of FLAG in the nucleus to that in the combined nonchromosomal and chromosomal areas (whole cell) in at least 27 cells was then plotted using Prism 10.4.2, using a two-tailed, unequal variance t test to determine statistical significance.
Acknowledgments
We would like to thank P. Coulombe (University of Michigan) for sharing the Zeiss LSM 800 confocal microscope. We also thank members of the Tsai laboratory, the DiMaio laboratory, and the Chelsey Spriggs laboratory (University of Michigan) for helpful discussion.
Funding: This work was supported by the National Institutes of Health grant R01AI150897 (B.T. and D.D.), National Institutes of Health grant R35CA242462 (D.D.), National Institutes of Health grant AI064296 (B.T.), and National Institutes of Health grant F31AI152365 (M.C.H.).
Competing interests: The authors declare that they have no competing interests.
Author contributions: Conceptualization: T.-T.W. and B.T. Methodology: T.-T.W., Y.T., and M.C.H. Investigation: T.-T.W., Y.T., M.C.H., and E.T.H. Supervision: B.T. and D.D. Writing—original draft: T.-T.W. and B.T. Writing—review and editing: T.-T.W., Y.T., M.C.H., E.T.H., D.D., and B.T.
Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials.
Supplementary Materials
This PDF file includes:
Figs. S1 to S6
Table S1
REFERENCES AND NOTES
- 1.de Martel C., Plummer M., Vignat J., Franceschi S., Worldwide burden of cancer attributable to HPV by site, country and HPV type. Int. J. Cancer 141, 664–670 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Malagon T., Franco E. L., Tejada R., Vaccarella S., Epidemiology of HPV-associated cancers past, present and future: Towards prevention and elimination. Nat. Rev. Clin. Oncol. 21, 522–538 (2024). [DOI] [PubMed] [Google Scholar]
- 3.Campos S. K., Subcellular trafficking of the papillomavirus genome during initial infection: The remarkable abilities of minor capsid protein L2. Viruses 9, 370 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Buck C. B., Cheng N., Thompson C. D., Lowy D. R., Steven A. C., Schiller J. T., Trus B. L., Arrangement of L2 within the papillomavirus capsid. J. Virol. 82, 5190–5197 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kines R. C., Thompson C. D., Lowy D. R., Schiller J. T., Day P. M., The initial steps leading to papillomavirus infection occur on the basement membrane prior to cell surface binding. Proc. Natl. Acad. Sci. U.S.A. 106, 20458–20463 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Richards R. M., Lowy D. R., Schiller J. T., Day P. M., Cleavage of the papillomavirus minor capsid protein, L2, at a furin consensus site is necessary for infection. Proc. Natl. Acad. Sci. U.S.A. 103, 1522–1527 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Becker M., Greune L., Schmidt M. A., Schelhaas M., Extracellular conformational changes in the capsid of human papillomaviruses contribute to asynchronous uptake into host cells. J. Virol. 92, e02106-17 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Day P. M., Lowy D. R., Schiller J. T., Heparan sulfate-independent cell binding and infection with furin-precleaved papillomavirus capsids. J. Virol. 82, 12565–12568 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Smith J. L., Campos S. K., Wandinger-Ness A., Ozbun M. A., Caveolin-1-dependent infectious entry of human papillomavirus type 31 in human keratinocytes proceeds to the endosomal pathway for pH-dependent uncoating. J. Virol. 82, 9505–9512 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bergant Marusic M., Ozbun M. A., Campos S. K., Myers M. P., Banks L., Human papillomavirus L2 facilitates viral escape from late endosomes via sorting nexin 17. Traffic 13, 455–467 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Selinka H. C., Giroglou T., Sapp M., Analysis of the infectious entry pathway of human papillomavirus type 33 pseudovirions. Virology 299, 279–287 (2002). [DOI] [PubMed] [Google Scholar]
- 12.Bienkowska-Haba M., Williams C., Kim S. M., Garcea R. L., Sapp M., Cyclophilins facilitate dissociation of the human papillomavirus type 16 capsid protein L1 from the L2/DNA complex following virus entry. J. Virol. 86, 9875–9887 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Inoue T., Zhang P., Zhang W., Goodner-Bingham K., Dupzyk A., DiMaio D., Tsai B., gamma-Secretase promotes membrane insertion of the human papillomavirus L2 capsid protein during virus infection. J. Cell Biol. 217, 3545–3559 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Crite M., DiMaio D., Human papillomavirus L2 capsid protein stabilizes γ-secretase during viral infection. Viruses 14, 804 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Zhang P., Monteiro da Silva G., Deatherage C., Burd C., DiMaio D., Cell-penetrating peptide mediates intracellular membrane passage of human papillomavirus L2 protein to trigger retrograde trafficking. Cell 174, 1465–1476.e13 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kamper N., Day P. M., Nowak T., Selinka H. C., Florin L., Bolscher J., Hilbig L., Schiller J. T., Sapp M., A membrane-destabilizing peptide in capsid protein L2 is required for egress of papillomavirus genomes from endosomes. J. Virol. 80, 759–768 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.DiGiuseppe S., Keiffer T. R., Bienkowska-Haba M., Luszczek W., Guion L. G., Muller M., Sapp M., Topography of the human papillomavirus minor capsid protein L2 during vesicular trafficking of infectious entry. J. Virol. 89, 10442–10452 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Bronnimann M. P., Chapman J. A., Park C. K., Campos S. K., A transmembrane domain and GxxxG motifs within L2 are essential for papillomavirus infection. J. Virol. 87, 464–473 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lipovsky A., Popa A., Pimienta G., Wyler M., Bhan A., Kuruvilla L., Guie M. A., Poffenberger A. C., Nelson C. D., Atwood W. J., DiMaio D., Genome-wide siRNA screen identifies the retromer as a cellular entry factor for human papillomavirus. Proc. Natl. Acad. Sci. U.S.A. 110, 7452–7457 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Pim D., Broniarczyk J., Siddiqa A., Massimi P., Banks L., Human papillomavirus 16 L2 recruits both retromer and retriever complexes during retrograde trafficking of the viral genome to the cell nucleus. J. Virol. 95, e02068-20 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Xie J., Zhang P., Crite M., Lindsay C. V., DiMaio D., Retromer stabilizes transient membrane insertion of L2 capsid protein during retrograde entry of human papillomavirus. Sci. Adv. 7, eabh4276 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Speckhart K., Choi J., DiMaio D., Tsai B., The BICD2 dynein cargo adaptor binds to the HPV16 L2 capsid protein and promotes HPV infection. PLOS Pathog. 20, e1012289 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Choi J., Speckhart K., Tsai B., DiMaio D., Rab6a enables BICD2/dynein-mediated trafficking of human papillomavirus from the trans-Golgi network during virus entry. MBio 15, e0281124 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Harwood M. C., Woo T. T., Takeo Y., DiMaio D., Tsai B., HPV is a cargo for the COPI sorting complex during virus entry. Sci. Adv. 9, eadc9830 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Pyeon D., Pearce S. M., Lank S. M., Ahlquist P., Lambert P. F., Establishment of human papillomavirus infection requires cell cycle progression. PLOS Pathog. 5, e1000318 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Aydin I., Weber S., Snijder B., Samperio Ventayol P., Kuhbacher A., Becker M., Day P. M., Schiller J. T., Kann M., Pelkmans L., Helenius A., Schelhaas M., Large scale RNAi reveals the requirement of nuclear envelope breakdown for nuclear import of human papillomaviruses. PLOS Pathog. 10, e1004162 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Calton C. M., Bronnimann M. P., Manson A. R., Li S., Chapman J. A., Suarez-Berumen M., Williamson T. R., Molugu S. K., Bernal R. A., Campos S. K., Translocation of the papillomavirus L2/vDNA complex across the limiting membrane requires the onset of mitosis. PLOS Pathog. 13, e1006200 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Rizzato M., Mao F., Chardon F., Lai K. Y., Villalonga-Planells R., Drexler H. C. A., Pesenti M. E., Fiskin M., Roos N., King K. M., Li S., Gamez E. R., Greune L., Dersch P., Simon C., Masson M., Van Doorslaer K., Campos S. K., Schelhaas M., Master mitotic kinases regulate viral genome delivery during papillomavirus cell entry. Nat. Commun. 14, 355 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.DiGiuseppe S., Luszczek W., Keiffer T. R., Bienkowska-Haba M., Guion L. G., Sapp M. J., Incoming human papillomavirus type 16 genome resides in a vesicular compartment throughout mitosis. Proc. Natl. Acad. Sci. U.S.A. 113, 6289–6294 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Aydin I., Villalonga-Planells R., Greune L., Bronnimann M. P., Calton C. M., Becker M., Lai K. Y., Campos S. K., Schmidt M. A., Schelhaas M., A central region in the minor capsid protein of papillomaviruses facilitates viral genome tethering and membrane penetration for mitotic nuclear entry. PLOS Pathog. 13, e1006308 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Keiffer T. R., Soorya S., Sapp M. J., Recent advances in our understanding of the infectious entry pathway of human papillomavirus type 16. Microorganisms 9, 2076 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.DiGiuseppe S., Bienkowska-Haba M., Sapp M., Human papillomavirus entry: Hiding in a bubble. J. Virol. 90, 8032–8035 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Ozbun M. A., Campos S. K., The long and winding road: Human papillomavirus entry and subcellular trafficking. Curr. Opin. Virol. 50, 76–86 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Florin L., Becker K. A., Lambert C., Nowak T., Sapp C., Strand D., Streeck R. E., Sapp M., Identification of a dynein interacting domain in the papillomavirus minor capsid protein l2. J. Virol. 80, 6691–6696 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Keiffer T. R., DiGiuseppe S., Guion L., Bienkowska-Haba M., Zwolinska K., Siddiqa A., Kushwaha A., Sapp M. J., HPV16 entry requires dynein for minus-end transport and utilizes kinesin Kif11 for plus-end transport along microtubules during mitosis. J. Virol. 99, e0093724 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Lai K. Y., Rizzato M., Aydin I., Villalonga-Planells R., Drexler H. C. A., Schelhaas M., A Ran-binding protein facilitates nuclear import of human papillomavirus type 16. PLOS Pathog. 17, e1009580 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Wing C. E., Fung H. Y. J., Chook Y. M., Karyopherin-mediated nucleocytoplasmic transport. Nat. Rev. Mol. Cell Biol. 23, 307–328 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Darshan M. S., Lucchi J., Harding E., Moroianu J., The l2 minor capsid protein of human papillomavirus type 16 interacts with a network of nuclear import receptors. J. Virol. 78, 12179–12188 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Bordeaux J., Forte S., Harding E., Darshan M. S., Klucevsek K., Moroianu J., The l2 minor capsid protein of low-risk human papillomavirus type 11 interacts with host nuclear import receptors and viral DNA. J. Virol. 80, 8259–8262 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Klucevsek K., Daley J., Darshan M. S., Bordeaux J., Moroianu J., Nuclear import strategies of high-risk HPV18 L2 minor capsid protein. Virology 352, 200–208 (2006). [DOI] [PubMed] [Google Scholar]
- 41.DiGiuseppe S., Bienkowska-Haba M., Hilbig L., Sapp M., The nuclear retention signal of HPV16 L2 protein is essential for incoming viral genome to transverse the trans-Golgi network. Virology 458-459, 93–105 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Xie J., Heim E. N., Crite M., DiMaio D., TBC1D5-catalyzed cycling of Rab7 is required for retromer-mediated human papillomavirus trafficking during virus entry. Cell Rep. 31, 107750 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Biryukov J., Meyers C., Papillomavirus infectious pathways: A comparison of systems. Viruses 7, 4303–4325 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Soderholm J. F., Bird S. L., Kalab P., Sampathkumar Y., Hasegawa K., Uehara-Bingen M., Weis K., Heald R., Importazole, a small molecule inhibitor of the transport receptor importin-β. ACS Chem. Biol. 6, 700–708 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Karanam B., Peng S., Li T., Buck C., Day P. M., Roden R. B., Papillomavirus infection requires gamma secretase. J. Virol. 84, 10661–10670 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Day P. M., Weisberg A. S., Thompson C. D., Hughes M. M., Pang Y. Y., Lowy D. R., Schiller J. T., Human papillomavirus 16 capsids mediate nuclear entry during infection. J. Virol. 93, e00454-19 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Salic A., Mitchison T. J., A chemical method for fast and sensitive detection of DNA synthesis in vivo. Proc. Natl. Acad. Sci. U.S.A. 105, 2415–2420 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Goldstein J. L., DeBose-Boyd R. A., Brown M. S., Protein sensors for membrane sterols. Cell 124, 35–46 (2006). [DOI] [PubMed] [Google Scholar]
- 49.Sato R., SREBPs: Protein interaction and SREBPs. FEBS J. 276, 622–627 (2009). [DOI] [PubMed] [Google Scholar]
- 50.Cheng C., Ru P., Geng F., Liu J., Yoo J. Y., Wu X., Cheng X., Euthine V., Hu P., Guo J. Y., Lefai E., Kaur B., Nohturfft A., Ma J., Chakravarti A., Guo D., Glucose-mediated N-glycosylation of SCAP is essential for SREBP-1 activation and tumor growth. Cancer Cell 28, 569–581 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Vercruysse T., Vanstreels E., Jacquemyn M., Boland S., Kilonda A., Allasia S., Vandecaetsbeek I., Klaassen H., Versele M., Chaltin P., Marchand A., Daelemans D., Ibetazol, a novel inhibitor of importin β1-mediated nuclear import. Commun. Biol. 7, 1560 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Guillet-Deniau I., Pichard A. L., Kone A., Esnous C., Nieruchalski M., Girard J., Prip-Buus C., Glucose induces de novo lipogenesis in rat muscle satellite cells through a sterol-regulatory-element-binding-protein-1c-dependent pathway. J. Cell Sci. 117, 1937–1944 (2004). [DOI] [PubMed] [Google Scholar]
- 53.Panagiotopoulos A. A., Polioudaki C., Ntallis S. G., Dellis D., Notas G., Panagiotidis C. A., Theodoropoulos P. A., Castanas E., Kampa M., The sequence [EKRKI(E/R)(K/L/R/S/T)] is a nuclear localization signal for importin 7 binding (NLS7). Biochim. Biophys. Acta Gen. Subj. 1865, 129851 (2021). [DOI] [PubMed] [Google Scholar]
- 54.Mamoor S., Onder Z., Karanam B., Kwak K., Bordeaux J., Crosby L., Roden R. B., Moroianu J., The high risk HPV16 L2 minor capsid protein has multiple transport signals that mediate its nucleocytoplasmic traffic. Virology 422, 413–424 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Bronnimann M. P., Calton C. M., Chiquette S. F., Li S., Lu M., Chapman J. A., Bratton K. N., Schlegel A. M., Campos S. K., Furin cleavage of L2 during papillomavirus infection: Minimal dependence on cyclophilins. J. Virol. 90, 6224–6234 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Popa A., Zhang W., Harrison M. S., Goodner K., Kazakov T., Goodwin E. C., Lipovsky A., Burd C. G., DiMaio D., Direct binding of retromer to human papillomavirus type 16 minor capsid protein L2 mediates endosome exit during viral infection. PLOS Pathog. 11, e1004699 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Campos S. K., Chapman J. A., Deymier M. J., Bronnimann M. P., Ozbun M. A., Opposing effects of bacitracin on human papillomavirus type 16 infection: Enhancement of binding and entry and inhibition of endosomal penetration. J. Virol. 86, 4169–4181 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Laniosz V., Dabydeen S. A., Havens M. A., Meneses P. I., Human papillomavirus type 16 infection of human keratinocytes requires clathrin and caveolin-1 and is brefeldin a sensitive. J. Virol. 83, 8221–8232 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Zhang W., Kazakov T., Popa A., DiMaio D., Vesicular trafficking of incoming human papillomavirus 16 to the Golgi apparatus and endoplasmic reticulum requires gamma-secretase activity. MBio 5, e01777–e01714 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60.Morante A. V., Baboolal D. D., Simon X., Pan E. C., Meneses P. I., Human papillomavirus minor capsid protein L2 mediates intracellular trafficking into and passage beyond the endoplasmic reticulum. Microbiol. Spectr. 10, e0150522 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61.Day P. M., Thompson C. D., Weisberg A. S., Schiller J. T., The COPII transport complex participates in HPV16 infection. Viruses 17, 616 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 62.Xie J., DiMaio D., Traptamer screening: A new functional genomics approach to study virus entry and other cellular processes. FEBS J. 289, 355–362 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 63.Nagoshi E., Imamoto N., Sato R., Yoneda Y., Nuclear import of sterol regulatory element-binding protein-2, a basic helix-loop-helix-leucine zipper (bHLH-Zip)-containing transcription factor, occurs through the direct interaction of importin beta with HLH-Zip. Mol. Biol. Cell 10, 2221–2233 (1999). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 64.Lee S. J., Sekimoto T., Yamashita E., Nagoshi E., Nakagawa A., Imamoto N., Yoshimura M., Sakai H., Chong K. T., Tsukihara T., Yoneda Y., The structure of importin-beta bound to SREBP-2: Nuclear import of a transcription factor. Science 302, 1571–1575 (2003). [DOI] [PubMed] [Google Scholar]
- 65.Li M., Li X., Chen S., Zhang T., Song L., Pei J., Sun G., Guo L., IPO5 mediates EMT and promotes esophageal cancer development through the RAS-ERK pathway. Oxid. Med. Cell. Longev. 2022, 6570879 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 66.Itoh Y., Miyamoto Y., Tokunaga M., Suzuki T., Takada A., Ninomiya A., Hishinuma T., Matsuda M., Yoneda Y., Oka M., Suzuki R., Matsuura Y., Okamoto T., Importin-7-dependent nuclear translocation of the Flavivirus core protein is required for infectious virus production. PLOS Pathog. 20, e1012409 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67.Zaitseva L., Cherepanov P., Leyens L., Wilson S. J., Rasaiyaah J., Fassati A., HIV-1 exploits importin 7 to maximize nuclear import of its DNA genome. Retrovirology 6, 11 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 68.Ao Z., Huang G., Yao H., Xu Z., Labine M., Cochrane A. W., Yao X., Interaction of human immunodeficiency virus type 1 integrase with cellular nuclear import receptor importin 7 and its impact on viral replication. J. Biol. Chem. 282, 13456–13467 (2007). [DOI] [PubMed] [Google Scholar]
- 69.Jakel S., Albig W., Kutay U., Bischoff F. R., Schwamborn K., Doenecke D., Gorlich D., The importin beta/importin 7 heterodimer is a functional nuclear import receptor for histone H1. EMBO J. 18, 2411–2423 (1999). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 70.Jakel S., Gorlich D., Importin beta, transportin, RanBP5 and RanBP7 mediate nuclear import of ribosomal proteins in mammalian cells. EMBO J. 17, 4491–4502 (1998). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 71.Nagoshi E., Yoneda Y., Dimerization of sterol regulatory element-binding protein 2 via the helix-loop-helix-leucine zipper domain is a prerequisite for its nuclear localization mediated by importin beta. Mol. Cell. Biol. 21, 2779–2789 (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 72.Freedman N. D., Yamamoto K. R., Importin 7 and importin alpha/importin beta are nuclear import receptors for the glucocorticoid receptor. Mol. Biol. Cell 15, 2276–2286 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 73.Walther T. C., Fornerod M., Pickersgill H., Goldberg M., Allen T. D., Mattaj I. W., The nucleoporin Nup153 is required for nuclear pore basket formation, nuclear pore complex anchoring and import of a subset of nuclear proteins. EMBO J. 20, 5703–5714 (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 74.Harwood M. C., Dupzyk A. J., Inoue T., DiMaio D., Tsai B., p120 catenin recruits HPV to γ-secretase to promote virus infection. PLOS Pathog. 16, e1008946 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 75.Bagchi P., Torres M., Qi L., Tsai B., Selective EMC subunits act as molecular tethers of intracellular organelles exploited during viral entry. Nat. Commun. 11, 1127 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 76.Buck C. B., Pastrana D. V., Lowy D. R., Schiller J. T., Efficient intracellular assembly of papillomaviral vectors. J. Virol. 78, 751–757 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 77.Buck C. B., Thompson C. D., Pang Y. Y., Lowy D. R., Schiller J. T., Maturation of papillomavirus capsids. J. Virol. 79, 2839–2846 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 78.Lipovsky A., Zhang W., Iwasaki A., DiMaio D., Application of the proximity-dependent assay and fluorescence imaging approaches to study viral entry pathways. Methods Mol. Biol. 1270, 437–451 (2015). [DOI] [PubMed] [Google Scholar]
- 79.Schindelin J., Arganda-Carreras I., Frise E., Kaynig V., Longair M., Pietzsch T., Preibisch S., Rueden C., Saalfeld S., Schmid B., Tinevez J. Y., White D. J., Hartenstein V., Eliceiri K., Tomancak P., Cardona A., Fiji: An open-source platform for biological-image analysis. Nat. Methods 9, 676–682 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 80.Kamentsky L., Jones T. R., Fraser A., Bray M. A., Logan D. J., Madden K. L., Ljosa V., Rueden C., Eliceiri K. W., Carpenter A. E., Improved structure, function and compatibility for CellProfiler: Modular high-throughput image analysis software. Bioinformatics 27, 1179–1180 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figs. S1 to S6
Table S1
Data Availability Statement
Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials.