Nuclear Import Strategies of High Risk HPV16 L1 Major Capsid Protein*

During the late phase of human papillomavirus (HPV) infection, the L1 major capsid proteins enter the nuclei of host epithelial cells and, together with the L2 minor capsid proteins, assemble the replicated viral DNA into virions. We investigated the nuclear import of the L1 major capsid protein of high risk HPV16. When digitonin-permeabilized HeLa cells were incubated with HPV16 L1 capsomeres, the L1 protein was imported into the nucleus in a receptor-mediated manner. HPV16 L1 capsomeres formed complexes with Kap α2β1 heterodimers via interaction with Kap α2. Accordingly, nuclear import of HPV16 L1 capsomeres was mediated by Kap α2β1 heterodimers, required RanGDP and free GTP, and was independent of GTP hydrolysis. Remarkably, HPV16 L1 capsomeres also interacted with Kap β2 and binding of RanGTP to Kap β2 did not dissociate the HPV16 L1·Kap β2 complex. Significantly, HPV16 L1 capsomeres inhibited the nuclear import of Kap β2 and of a Kap β2-specific M9-containing cargo. These data suggest that, during the productive stage of infection, while the HPV16 L1 major capsid protein enters the nucleus via the Kap α2β1-mediated pathway to assemble the virions, it also inhibits the Kap β2-mediated nuclear import of host hnRNP A1 protein and, in this way, favors virion formation.

Human papillomaviruses (HPVs) 1 are thought to be the primary causative agent in more than 90% of all cervical cancers, with HPV16 being the type most frequently found in these tumors. Approximately 500,000 new cases of cervical cancer are identified each year globally with nearly 300,000 deaths annually. About 85 genotypically distinct HPV genotypes have been isolated and characterized, with roughly half infecting the skin and the other half preferentially infecting oral/anogenital mucosal epithelial tissues. Mucosal HPVs have demonstrated varying degrees of oncogenic potential; high risk HPVs, such as types 16, 18, 31, and 45, are frequently detected in invasive cervical carcinomas, whereas the low risk types, such as types 6 and 11, are more often associated with benign exophytic condylomas (1).
HPVs are small, nonenveloped, icosahedral DNA viruses that replicate in the nucleus of squamous epithelial cells. The virion particles (52-55 nm in diameter) consist of a single molecule comprising 8 kb of double-stranded circular DNA contained within a spherical capsid composed of 72 homopentameric L1 capsomeres and of L2 minor capsid protein. The number of L2 molecules per capsid has been estimated at 36 (2) or 12 (3). L1 protein is capable of self-assembly both in vivo and in vitro into capsid-like structures, referred to as virus-like particles (4 -9). L1 is stable in two oligomeric configurations: homopentameric capsomeres and capsids composed of 72 capsomeres. The capsids can be disassembled into capsomeres quantitatively by an agent that reduces disulfide bonds and reassembled by removing the reducing agent (10,11). The L1 capsid proteins of HPVs seem to enter the nucleus of host cells twice during the virus life cycle: immediately after the virions infect the undifferentiated proliferating epithelial cells and again during the late productive phase when the newly synthesized L1 and L2 proteins assemble the replicated HPV genomic DNA into infectious virions inside the nuclei of terminally differentiated epithelial cells.
Proteins must carry nuclear localization signals (NLSs) to enter the nucleus. The first identified NLSs, now referred to as classic NLSs, fall into two categories: a monopartite NLS consisting of a simple sequence of 3-5 basic amino acid residues and a bipartite NLS consisting of a basic dipeptide upstream from a simple basic sequence. Other types of NLSs have since been identified in hnRNP proteins, ribosomal proteins, and U small nuclear RNPs. NLSs are recognized by adapters of the karyopherin ␣/importin ␣ (Kap ␣/Imp ␣) family and by import receptors of the karyopherin ␤/importin ␤ (Kap ␤/Imp ␤) family that shuttle between the nucleus and the cytoplasm. The basic paradigm for nuclear import is that the NLS cargo is bound (either directly or indirectly via an adapter) by an import receptor in the cytoplasm, translocated through the nuclear pore complex (NPC) and released inside the nucleus. The receptor and the adapter are then recycled back to the cytoplasm in a form competent for another round of import (12)(13)(14). The first identified nuclear import receptor was Kap ␤1, which functions together with a Kap ␣ adapter in nuclear import of proteins that contain classic monopartite or bipartite NLSs. Kap ␣ binds to the NLS of the cargo, whereas Kap ␤1 mediates docking at the NPC (15)(16)(17)(18)(19)(20)(21). There are six human Kap ␣ adapters, and it is likely that they have both distinct and overlapping specificities for classic basic NLSs. Kap ␤1 can also function without adapters when importing ribosomal proteins, cyclin B1, and viral proteins. Other members of the Kap ␤ family have been identified and shown to function in nuclear import of specific cargoes. All Kap ␤ receptors shuttle between the nu-cleus and the cytoplasm and bind to nucleoporins at the NPC and to the GTPase Ran in its GTP-bound form. The nucleotide state of Ran is regulated by cytoplasmic RanGAP and RanBP1, which accelerate GTP hydrolysis to form cytoplasmic RanGDP, and by the nuclear RanGEF (RCC1), which catalyzes nucleotide exchange to regenerate nuclear RanGTP. Binding of nuclear RanGTP to Kap ␤s (importins) causes the dissociation of the import complexes with release of the cargoes inside the nucleus. The Kap ␤s (importins/exportins) exit the nucleus in complex with RanGTP and, therefore, constantly deplete Ran from the nucleus (12)(13)(14). Ran is actively re-imported into the nucleus via the transport factor p10/NTF2, which binds specifically to RanGDP but not to RanGTP (22).
In this study we investigated the nuclear import of the L1 major capsid protein of high risk HPV16. We found that HPV16 L1 capsomeres enter the nucleus via the classic Kap ␣2␤1mediated nuclear import pathway. Importantly, we also discovered that both HPV16 and HPV45 L1 capsomeres interact with Kap ␤2 and inhibit nuclear import of Kap ␤2 and its specific M9-containing cargo. These data suggest that during the late productive stage of infection the newly synthesized L1 major capsid proteins can interact with Kap ␤2 and may inhibit major Kap ␤2-mediated nuclear import pathways and, thus, inhibit other nuclear events so as to optimize virion formation.

Preparation of Recombinant Human Nuclear Import Factors-His-
tagged Kap ␣2 (20), His-tagged Kap ␤1 (23), and His-tagged p10 (18) were expressed in Escherichia coli BL21(DE3) (3-h induction with 2 mM IPTG at 30°C), and the soluble His-tagged proteins were purified in their native state on Talon beads using a standard procedure. GST-Kap ␤1 (23) and GST-Kap ␤2 (24) were expressed in E. coli BL21(DE3) (3-h induction with 1 mM IPTG at 30°C), and the soluble GST fusion proteins were purified in their native state on glutathione-Sepharose beads using a standard procedure. To obtain cleaved Kap ␤2, the GST-Kap ␤2 fusion protein was incubated for 2 h at room temperature with a Tev enzyme that has a His tag (Invitrogen), and after cleavage the GST was removed by binding to glutathione-Sepharose beads and the Tev enzyme by binding to Talon beads, using standard procedures. Human Ran (25) was prepared as described previously (26). All proteins were checked for purity and lack of proteolytic degradation by SDS-PAGE and Coomassie Blue staining. The purified proteins were dialyzed in buffer A (20 mM HEPES-KOH, pH 7.3, 110 mM potassium acetate, 2 mM magnesium acetate, 1 mM EGTA, 2 mM dithiothreitol) containing protease inhibitors and stored in aliquots at Ϫ80°C until use.
Preparation of Recombinant HPV16 L1 Capsids and Capsomeres-The L1 capsids of HPV16 and HPV45 were generated in insect cells, purified as described (27,28), and stored at 4°C until use. Purity and lack of proteolytic degradation of the L1 proteins were always checked by SDS-PAGE, Coomassie Blue staining, and immunoblotting prior to use. L1 capsomeres were obtained by incubating the L1 capsids with 5% mercaptoethanol overnight at 4°C (11).
Antibodies-Rabbit polyclonal antiserum was raised against HPV16 L1 capsids as described (27,28). Murine antibodies raised against Kap ␣2/Rch-1 and Kap ␤2/transportin were from Transduction Laboratories; a mouse monoclonal antibody to Kap ␤1/p97 was from Affinity Bioreagents, Inc., and the goat anti-GST antibody was from Amersham Biosciences. HeLa cytosol from Cellex Biosciences Inc. was centrifuged and stored in aliquots at Ϫ80°C.
Preparation of GST-NLS HPV16L1 Fusion Proteins-The L1 major capsid protein of HPV16 contains two classic NLSs at its C terminus: a monopartite NLS (AKRKKRKL) and a bipartite type that overlaps with the monopartite NLS (KRKATPTTSSTSTTAKRKKRKL) (29). The monopartite NLS (mpNLS) of HPV16 L1 and the overlapping bipartite/ monopartite NLS (bpNLS) of HPV16 L1 were separately fused to a GST reporter protein. For making the GST-bpNLS construct the forward and reverse oligonucleotides, phosphorylated at 5Ј-ends, were annealed, and the overhanging areas were filled in using a DNA polymerase I, large Klenow fragment. The resulting DNA fragments were blunt endligated with T4 DNA ligase, double-digested with EcoRI and XhoI, and inserted into pGEX4T-1 (Amersham Biosciences) that was double cut with the same enzymes. For making the GST-mpNLS construct, unphosphorylated oligonucleotides were designed with the corresponding restriction-cutting sites, annealed, and inserted directly into the pGEX4T-1 vector as described above. The GST-bpNLS and GST-mpNLS constructs were transformed in XL1 blue bacteria and confirmed by automated sequence analysis (BioServe Biotechnologies Sequencing Department). The M9-GST construct was a gift from Dr. Gideon Dreyfuss. For protein expression, the constructs were used to transform E. coli BL21(DE3) bacteria. After induction of E. coli BL21(DE3) with 1 mM IPTG for 3 h at 37°C, the GST-NLS fusion proteins were purified in their native state on glutathione-Sepharose using a standard procedure. The purified proteins were checked by SDS-PAGE and Coomassie Blue staining, dialyzed in buffer A, and stored in aliquots at Ϫ80°C until use.
In Vitro Nuclear Import Assays-Digitonin-permeabilized HeLa cells have been extensively used by many laboratories to investigate different nuclear import pathways mediated by mammalian Kap ␤s/importins (20, 22, 30 -38). HeLa cells are cervical carcinoma cells that contain HPV18 DNA integrated into the cellular genome with consequent disruption of several viral genes but preservation of the E6 and E7 genes. Although HeLa cells are HPV18-positive, this does not affect the import properties of their nuclear pore complexes, as demonstrated by the fact that the majority of nuclear import pathways have been identified and characterized with HeLa cells (20, 22, 30 -38). We have previously used these in vitro nuclear import assays in HeLa cells to investigate import of HPV11 L1 and HPV45 L1 major capsid proteins (38,39). In the present study, subconfluent HeLa cells, grown on poly-L-lysine-coated glass coverslips for 24 h, were permeabilized with 70 g/ml digitonin for 5 min on ice and washed with buffer A. Digitonin permeabilizes the plasma membrane but leaves the nuclear envelope intact. As a consequence, digitonin-permeabilized HeLa cells retain intact import-competent nuclei but are largely depleted of cytosolic transport factors, as indicated by the absence of NLS-mediated nuclear import in the presence of buffer alone and in the absence of either exogenous cytosol or recombinant Kaps plus RanGDP. It should be noted that p10 is retained in the digitonin-permeabilized HeLa cells in sufficient amounts to support NLS-mediated nuclear import (Refs. 32, 34, and 35 and Fig. 2). Unless otherwise specified all import reactions contained the energyregenerating system (0. Final import reaction volume was adjusted to 20 l with buffer A. For visualization of nuclear import, the HPV16 L1 protein and the GST-NLS fusion proteins were detected by immunofluorescence with specific antibodies, as previously described (38). The nuclei were identified by DAPI staining. Nuclear import was analyzed with a Nikon Eclipse TE 300 microscope, which has a fluorescence attachment, and a Sony DKC-5000 charge-coupled device camera.
In-solution Binding Assays-Two types of binding assays were performed. First, GST-NLS HPV16L1 fusion proteins, immobilized on glutathione-Sepharose beads (2 g of protein/10 l of beads) were incubated under rotation for 30 min at room temperature with the purified Kaps (2 g of protein) in buffer A (total volume, 40 l). Control experiments for binding specificity consisted of incubating GST immobilized on glutathione-Sepharose beads with the purified Kaps. After incubation, the beads were washed three times with buffer A, and the bound proteins were eluted with SDS-PAGE sample buffer and analyzed by SDS-PAGE followed by Coomassie Blue staining.
Second, GST-Kap ␤1 or GST-Kap ␤2 immobilized on glutathione-Sepharose beads (2 g of protein/10 l of beads) was incubated for 30 min under rotation at room temperature with HPV16 L1 or HPV45 L1 capsomeres (2 g of protein) in buffer A containing 0.25% Tween 20 (binding buffer). In some experiments, 2 g of either Kap ␣2 or RanGTP was added to the incubation mixture, as indicated in the figure legends. Control experiments for binding specificity consisted of incubating GST immobilized on glutathione-Sepharose beads with the L1 capsomeres. The bound proteins were eluted with SDS-PAGE sample buffer and analyzed by SDS-PAGE followed by Coomassie Blue staining.

RESULTS
Disassembly of L1 Capsids Is Required for Nuclear Import of HPV16 L1 Major Capsid Protein-We had previously established for the L1 major capsid proteins of HPV11 and HPV45 that capsid disassembly is required for their nuclear import (38,39). Therefore, the nuclear import of the L1 major capsid protein of HPV16 was investigated in either capsid or capsomere conformation. Digitonin-permeabilized HeLa cells were incubated with either HPV16 L1 capsomeres (Fig. 1A) or HPV16 L1 capsids (Fig. 1B) in the presence of cytosol and an energy-regenerating system. As previously found for HPV11 and HPV45, the HPV16 L1 capsomeres were imported into the nuclei of digitonin-permeabilized cells in the presence of cytosol (Fig. 1A), whereas the HPV16 L1 capsids were not (Fig. 1B).
HPV16 L1 Major Capsid Protein Enters the Nucleus via the Kap ␣2␤1-mediated Pathway-It has been shown previously that the L1 major capsid protein of HPV16 contains two classic NLSs at the C terminus: a monopartite NLS and a bipartite type that overlaps with the monopartite NLS (29). Because HPV16 L1 contains classic basic NLSs, we investigated whether nuclear import of HPV16 L1 major capsid protein can be mediated by the Kap ␣2␤1 heterodimers. Digitonin-permeabilized HeLa cells were incubated with HPV16 L1 capsomeres in the presence of either . We found that nuclear import of HPV16 L1 was efficiently mediated by Kap ␣2␤1 heterodimers (Fig. 2, C and D) but not by Kap ␤1 (Fig. 2, A and  B). Addition of exogenous p10 did not increase nuclear import of HPV16 L1 capsomeres in agreement with previous results that sufficient endogenous p10 is retained in the digitoninpermeabilized HeLa cells to support NLS-mediated import (32,34,35).
We next tested for nuclear import two GST fusion proteins containing either the overlapping bipartite/monopartite NLS of HPV16 L1 (GST-bpNLS HPV16L1 ) or the monopartite NLS of HPV16 L1 (GST-mpNLS HPV16L1 ). Both the GST-bpNLS HPV16L1 and the GST-mpNLS HPV16L1 were imported into the nucleus in the presence of cytosol or in the presence of Kap ␣2␤1 heterodimers plus RanGDP but not in the presence of transport buffer alone or Kap ␤1 plus RanGDP (Fig. 3 and data not shown). As a control, GST itself was not imported into the nucleus in the presence of cytosol (data not shown).
In agreement with the nuclear import data, the bpNLS HPV16L1 interacted directly with Kap ␣2 (Fig. 4A, lane 1) and formed a trimeric complex with the Kap ␣2/Kap ␤1 heterodimer (Fig. 4A, lane 2). The weak binding of Kap ␤1 alone to the GST-bpNLS HPV16L1 (Fig. 4A, lane 3) is assumed to be nonspecific, because it is equivalent to the interaction of Kap ␤1 with the GST alone (Fig. 4A, lane 4). The mpNLS HPV16L1 also interacted directly with Kap ␣2 and formed a trimeric complex with the Kap ␣2/Kap ␤1 heterodimer (data not shown).
To investigate the direct interactions between HPV16 L1 capsomeres and Kap ␣2␤1 heterodimers, we used in-solution binding assays with the GST-Kap ␤1 immobilized on glutathione-Sepharose beads via the GST (Fig. 4B). The GST-Kap ␤1-containing beads were incubated for 30 min at room temperature with HPV16 L1 capsomeres alone or in the presence of Kap ␣2. Binding of HPV16 L1 capsomeres to the Kap ␤1containing beads was strongly increased in the presence of Kap ␣2 (Fig. 4B, compare lanes 1 and 2) indicating the formation of a complex: L1 capsomere⅐Kap ␣2⅐Kap ␤1. Moreover, binding of RanGTP to Kap ␤1 inhibited binding of the Kap ␣2⅐L1 capsomere complex to Kap ␤1 or promoted its dissociation (Fig. 4B,  compare lanes 2 and 3). These binding results support the nuclear import data indicating that the Kap ␣2␤1 heterodimers can mediate nuclear import of HPV16 L1 capsomeres.

Nuclear Import of HPV16 L1 Major Capsid Protein Does Not Require GTP Hydrolysis by Ran-Previous studies have suggested that GTP hydrolysis by Ran is not required for Kap
␣2␤1-mediated nuclear import of specific cargoes (37,38). We investigated whether nuclear import of HPV16 L1 capsomeres and GST-mpNLS HPV16L1 are also independent of GTP hydrolysis by Ran by comparing their nuclear import in the presence of Kap ␣2 plus Kap ␤1 plus RanGDP and either GTP or the nonhydrolyzable GTP analogues, GTP␥S and GMP-PNP. Both HPV16 L1 capsomeres and the GST-mpNLS HPV16L1 were efficiently imported in the presence of either GTP, GTP␥S, or GMP-PNP (Fig. 5). These data suggest that GTP hydrolysis by Ran is not required for the nuclear import of HPV16 L1 major capsid protein.
In the presence of Kap ␣2, the binding of HPV16 L1 capsomeres to Kap ␤2 was strongly reduced (Fig. 6B, compare lanes 1  and 2), suggesting that Kap ␣2 competes with Kap ␤2 for binding to the L1 capsomeres. The bpNLS of HPV16 L1 that interacts with Kap ␣2 (Fig. 4A) did not bind to Kap ␤2 (data not shown). This suggests that the Kap ␤2-binding site is either different from the Kap ␣2-binding site or, if it overlaps with the Kap ␣2 site, it is longer.
The interaction between HPV16 L1 capsomeres and Kap ␤2 prompted us to investigate whether HPV16 L1 capsomeres can enter into the nucleus via a Kap ␤2-mediated pathway. Therefore, digitonin-permeabilized cells were incubated with either HPV16 L1 capsomeres or the M9-GST-positive control in the presence of Kap ␤2 plus RanGDP. We observed that Kap ␤2 plus RanGDP failed to promote nuclear import of HPV16 L1, whereas the M9-GST control was efficiently localized to the nucleus (data not shown).
Because HPV16 L1 bound to Kap ␤2 but failed to be imported, we questioned whether this interaction results in an import incompetent complex that blocks nuclear import of Kap ␤2. To answer this question, digitonin-permeabilized HeLa cells were incubated with either GST-Kap ␤2 or GST-Kap ␤2 plus RanGDP in the absence or presence of HPV16 L1 capsomeres. As expected, GST-Kap ␤2 entered the nucleus either in the absence or presence of RanGDP ( Fig. 7A and data not  shown). Interestingly, in the presence of HPV16 L1 capsomeres, nuclear import of GST-Kap ␤2 was strongly inhibited (Fig. 7, compare A and B) indicating that binding of HPV16 L1 capsomeres to GST-Kap ␤2 interferes with nuclear import of GST-Kap ␤2. In contrast, the weak interaction noted between HPV16 L1 capsomeres and GST-Kap ␤1 (Fig. 4B) did not inhibit nuclear import of GST-Kap ␤1 (data not shown). Significantly, we have found that, in the presence of HPV16 L1 capsomeres, Kap ␤2-mediated nuclear import of its specific M9-GST cargo was inhibited (Fig. 7, compare C and D). These results suggest that, although HPV16 L1 capsomeres do not use Kap ␤2 to enter into the nucleus, they inhibit Kap ␤2 nuclear import and, consequently, all Kap ␤2-mediated import pathways.
The ability of the HPV16 L1 to interact and inhibit nuclear import of Kap ␤2 led us to test whether this activity is specific only for the L1 major capsid protein of high risk HPV16 or if it is shared with the L1 capsid proteins of other HPV types. We therefore investigated whether the L1 capsomeres of high risk HPV45 interact with Kap ␤2 and found that they did (Fig. 8,  lane 1). Moreover, binding of RanGTP to Kap ␤2 did not dissociate the HPV45 L1⅐Kap ␤2 complexes (Fig. 8, lane 2). Also, as for HPV16 L1 capsomeres, HPV45 L1 capsomeres efficiently inhibited nuclear import of Kap ␤2 and of its M9-GST cargo (Fig. 9). These data suggest that inhibition of Kap ␤2-mediated nuclear import pathways is an activity shared at least by the L1 major capsid proteins of HPV16 and HPV45. DISCUSSION In this study we investigated the nuclear import pathways for the L1 major capsid protein of high risk HPV16. The nuclear import assays were carried out in digitonin-permeabilized HeLa cells, which are commonly used to investigate nuclear import pathways mediated by different mammalian Kaps (20, 22, 30 -38). As previously established for the L1 major capsid proteins of HPV11 and HPV45 (38, 39), the intact HPV16 L1 capsids cannot enter the nuclei. This is not surprising because the capsid diameter (55 nm) is greater than the maximal func- tional diameter of the nuclear pore (39 nm) through which active transport occurs (46). In addition, recent x-ray crystallographic analysis of 12-capsomere icosahedral particles of HPV16 L1 suggests that the C terminus of HPV16 L1 might be oriented inside the particle and not exposed outside the capsid (47). During the import reactions in digitonin-permeabilized cells in the presence of cytosol it is possible that the L1 capsomeres might be further disassembled into monomers. The inability of intact capsids to enter the nuclei would suggest that during the initial stages of infection disassembly of the capsids would be required before the nuclear import of incoming L1 proteins can occur. The exposure of incoming HPV capsids to the reducing environment of the cytoplasm could lead to the cleavage of the stabilizing disulfide bonds and, as a consequence, disassembly of the capsids. During the late productive stage of infection the interaction between the newly synthesized L1 proteins and the corresponding Kaps could prevent the assembly of L1 into capsids in the cytoplasm. X-ray crystallographic analysis of 12-capsomere icosahedral particles of HPV16 L1 indicates that the H4 ␣-helical loop, which is adjacent to the C-terminal NLS, is involved in intercapsomere interactions (47). Binding of the one or more corresponding host Kaps to the NLS of HPV16 L1 would interfere with intercapsomere interactions and prevent capsid assembly in the cytoplasm. Once the L1 proteins enter the nucleus and are released from their complexes with the Kaps, they can assemble into capsids.
We found that HPV16 L1 capsomeres can interact efficiently with Kap ␣2␤1 heterodimers via Kap ␣2. The HPV16 L1⅐Kap ␣2⅐Kap ␤1 complex is translocated through the nuclear pore complex into the nucleus in the presence of RanGDP and free GTP. In the nucleus, binding of RanGTP to Kap ␤1 dissociates the import complex with release of HPV16 L1⅐Kap ␣2. We had previously established that the L1 proteins of low risk HPV11 and high risk HPV45 can interact with Kap ␣2 and enter the nucleus via the classic Kap ␣2␤1-mediated pathway (38,39). Six Kap ␣ isoforms have been identified in higher eukaryotes, and they all bind to the same Kap ␤1. The import specificity of Kap ␣ isoforms has been analyzed for proteins containing classic monopartite and bipartite NLSs (48). When the protein cargoes were tested singly, they were imported by all six Kap ␣ isoforms with similar efficiencies. The only exception was RCC1, which showed strong preference for the importin ␣3 isoform. The import efficiency varied for Kap ␣ isoforms when the protein cargoes were presented simultaneously (48). Because HPV L1 proteins contain basic monopartite and/or bipartite NLSs, it is likely that they could interact with other Kap ␣ isoforms in addition to Kap ␣2 and exploit these interactions to enter into the nucleus via a classic Kap ␣␤1-mediated pathway.
As previously shown for HPV11 L1 and HPV45 L1 capsomeres (38), we have found that HPV16 L1 capsomeres can also interact with HPV-DNA via their C-terminal NLSs (data not shown). This would suggest that, during the late productive stage of viral infection, when the newly synthesized proteins are imported into the nuclei, the interaction of L1 with the viral DNA inside the nucleus would favor the release of Kap ␣2 from the L1⅐Kap ␣2 complexes.
In the initial stage of infection the incoming L1 major capsid proteins are transported into the nuclei of proliferating epithelial cells, whereas in the late productive phase the newly synthesized L1 proteins are transported into the nuclei of terminally differentiated keratinocytes. It is not known if there are differences in the expression of different Kaps between the proliferating versus terminally differentiated epithelial cells. However, it has been reported that Kap ␤1, Kap ␣2 (and other Kap ␣ isoforms), and Ran are ubiquitously expressed in various human adult tissues, with lower levels of Kap ␤1 found only in spleen and lower levels of Kap ␣2 found only in brain and liver (48). This would suggest that HPV L1 major capsid proteins could enter the nucleus via a classic Kap ␣␤1-mediated pathway during both the initial and late stage of infection. Although the role of nuclear import of L1 major capsid proteins in assembling the virions during the late stage of infection is clearly established, the role of L1 nuclear import during the initial phase of infection is unclear. It is not known if, after disassembly of the incoming virions in the cytoplasm, some of the L1 capsomeres would remain associated with the L2 minor capsid protein. If this is the case the question to be addressed is: do these L1⅐L2 complexes play a role in facilitating nuclear import of the viral DNA?
Remarkably, we also discovered that the L1 capsomeres of high risk HPV16 and HPV45 can interact with Kap ␤2 and inhibit nuclear import of Kap ␤2 and of its specific M9-GST cargo. Moreover, preliminary results suggest that the L1 capsomeres of low risk HPV11 share these activities. 2 Together these data suggest that during the late stage of HPV infection the L1 major capsid proteins can interact with Kap ␤2 and inhibit Kap ␤2-mediated nuclear import pathways of the host epithelial cells. The main cargoes transported by Kap ␤2 are hnRNP A1 and A2, and the interaction with Kap ␤2 is mediated by the shuttling sequence M9 (24, 34, 40 -43). It has been shown that nucleocytoplasmic shuttling of hnRNP A1 is strongly correlated with mRNA nuclear export (49). Hence, in the late stage of viral infection the interaction of newly synthesized L1 proteins with Kap ␤2 may inhibit Kap ␤2-mediated nuclear import of hnRNP A1 and, as a consequence, affect mRNA nuclear export. As HPV L1 capsomeres are efficiently imported into the nuclei of digitonin-permeabilized cells in the presence of HeLa cytosol, the higher affinity interaction of L1 capsomeres is with Kap ␣2␤1 heterodimers leading to nuclear import. This is in agreement with the fact that Kap ␣2 competed with Kap ␤2 for binding to HPV16 L1 capsomeres (Fig.  6B) and with the immunoisolation assays with HPV45 L1 capsomeres in which Kap ␣2, but not Kap ␤2, was isolated (38). This suggests that, during the late phase of the viral life cycle, when the L1 proteins are synthesized, some of the L1 proteins could interact with Kap ␤2 and inhibit Kap ␤2-mediated nuclear import of hnRNP A1 and, consequently, affect mRNA export. Because little is known about the stoichiometry of L1 capsomere levels versus Kap ␣2␤1 heterodimers and Kap ␤2 during different stages of HPV infection, further studies in vivo are required to analyze if L1 major capsid proteins inhibit the nuclear import of hnRNP A1 and A2 during the late stage of the viral life cycle.
The interaction of HPV L1 major capsid proteins with Kap ␤2 raises the question of whether L1 proteins interact with additional members of the Kap ␤ family. Preliminary data indicate that HPV16 L1 can also interact with Kap ␤3 and inhibit Kap ␤3 nuclear import. This suggests that HPV L1 major capsid proteins could also interact with other members of the Kap ␤ superfamily and, perhaps, inhibit other major nuclear import pathways of host cells.
There are many studies demonstrating that viruses exploit the nuclear import pathways of host cells for entry of their macromolecules into the nucleus. However, only a few examples of viral infections that inhibit nuclear import and/or export pathways of host cells have been reported. Poliovirus infection inhibits nuclear import of hnRNP A1, K, and C causing their relocalization in the cytoplasm. The mechanism of inhibition seems to involve the proteolytic degradation of Nup153 and p62, affecting traffic through the nuclear pore complex (50). Also, the matrix protein of the vesicular stomatitis virus inhibits different nuclear import and export pathways, and the inhibitory effect of the matrix protein seems to involve binding to the nucleoplasmic nucleoporin Nup98 (51,52). In both cases the viruses are targeting components of the nuclear pore complex. We propose that the strategy used by HPVs involves targeting the import receptor, Kap ␤2, and blocking Kap ␤2mediated nuclear import of hnRNP A1 and A2. Further studies are required to understand the role of inhibition of Kap ␤2mediated nuclear import pathways of host cells during the late stage of infection when the virions are formed and released.