Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Aug 15.
Published in final edited form as: Virology. 2013 May 29;443(1):113–122. doi: 10.1016/j.virol.2013.04.031

Characterization of the transport signals that mediate the nucleocytoplasmic traffic of low risk HPV11 E7

Courtney H McKee 1, Zeynep Onder 1, Aditya Ashok 1, Rebeca Cardoso 1, Junona Moroianu 1,*
PMCID: PMC3758764  NIHMSID: NIHMS476726  PMID: 23725695

Abstract

We previously discovered that nuclear import of low risk HPV11 E7 is mediated by its zinc-binding domain via a pathway that is independent of karyopherins/importins (Piccioli et al., 2010. Virology 407, 100–109). In this study we mapped and characterized a leucine-rich nuclear export signal (NES), 76IRQLQDLLL84, within the zinc-binding domain that mediates the nuclear export of HPV11 E7 in a CRM1-dependent manner. We also identified a mostly hydrophobic patch 65VRLVV69 within the zinc-binding domain that mediates nuclear import of HPV11 E7 via hydrophobic interactions with the FG-repeats domain of Nup62. Substitutions of hydrophobic residues to alanine within the 65VRLVV69 sequence disrupt the nuclear localization of 11E7, whereas the R66A mutation has no effect. Overall the data support a model of nuclear entry of HPV11 E7 protein via hydrophobic interactions with FG nucleoporins at the nuclear pore complex.

Keywords: Human papillomavirus, HPV11 E7 protein, Nuclear localization signal, Nuclear export signal, FG-nucleoporin Nup62, CRM1 nuclear export receptor

Introduction

In the United States, infection with human papillomaviruses (HPVs) is more prevalent than all other sexually transmitted diseases combined. HPV infection is associated with more than 99% of cervical cancers, and a high percentage of anal, perianal, vulvar, penile and oropharyngeal cancers. Some 30 distinct HPV genotypes preferentially infect anogenital mucosal epithelial tissues and they demonstrate different degrees of oncogenic potential with some classified as “high risk” types, and others as “low risk” types. The high risk HPVs such as, types 16, 18, 31 and 45, are frequently detected in invasive cervical carcinomas (zur Hausen, 2000). The most common low risk HPV types, 6 and 11, cause benign anogenital warts, or condyloma accuminata, affecting 1–2% of young adults in many countries (Doorbar, 2006; zur Hausen, 2000, 2009).

The replication cycle of HPVs is linked to the differentiation program of host epithelial cells with the productive viral DNA amplification occurring in the upper layers of the differentiated epithelial cells. As HPVs depend on the replication machinery of the host cells they have evolved the E7 proteins to induce reentry into the S phase of the differentiated epithelial cells and establish the appropriate environment required to support viral DNA amplification. The high risk HPV16/18 oncoproteins bind and destabilize the retinoblastoma protein (pRB) and the RB-related pocket proteins, p130 and p107. Interestingly, the low risk HPV E7 proteins bind pRB with approx. ten-fold lower efficiency and destabilize it inefficiently (McLaughlin-Drubin and Munger, 2009). Both the high risk HPV18 E7 and the low risk HPV11 E7 bind and target nuclear p130 for degradation causing S-phase entry in the differentiated human keratinocytes (Genovese et al., 2008).

In addition to nuclear target proteins, HPV16 E7 and HPV11 E7 have also targets in the cytoplasm, such as the microtubule-associated N-end rule ubiqutin ligase p600 (Huh et al., 2005) and the nuclear mitotic apparatus protein (NuMA) (Nguyen and Munger, 2009). The E7 proteins can change the localization of steroid receptor co-activator (SRC-1) from the nucleus to the cytoplasm (Baldwin et al., 2006).

HPV16 E7 and HPV11 E7 are mostly nuclear when they are expressed transiently in different cell lines including HaCaT and U2OS (Fiedler et al., 2004; Guccione et al., 2002). We have previously discovered that nuclear import of HPV16 E7 and HPV11 E7 is mediated by a novel Ran-dependent pathway that is independent of karyopherins/importins and it is mediated by their unique zinc-binding domain located in the C-terminal domain, amino acids 39–98 (Angeline et al., 2003; Knapp et al., 2009; Piccioli et al., 2010). The zinc-binding domain consists of two copies of CysXXCys motifs separated by 29 amino acids. Mutagenesis of Cys residues in each of the two CysXXCys motifs involved in zinc binding changed the localization of the resultant EGFP-11E7 and EGFP-11cE7 to mostly cytoplasmic suggesting that the integrity of the zinc-binding domain is essential for the nuclear localization of 11E7 protein (Piccioli et al., 2010).

Nuclear import and export take place through the nuclear pore complexes (NPCs) that represent specialized channels spanning the nuclear envelope. There are some 30 different proteins called nucleoporins (Nups) that assemble to form the NPC. There are three major classes of Nups: transmembrane Nups (Poms) involved in anchoring the NPC at the nuclear envelope, structural Nups, and FG-Nups, containing phenylalanine–glycine (FG) repeats, and involved in nuclear import and export (Terry and Wente, 2009). The mechanism of translocation through the NPC involves low affinity hydrophobic interactions between karyopherins (importins and exportins) and the FG repeats in FG-Nups. Major FG-Nups of the mammalian NPC involved in nuclear import include Nup62 located at the central transport channel, Nup214 and Nup358 on the cytoplasmic fibers and Nup153 on the nucleoplasmic fibers of the NPC (Terry and Wente, 2009).

We have previously found that HPV11 E7 and HPV16 E7, similarly as the karyopherins/importins, can bind to phenyl-Sepharose beads, which mimic the specificity of FG nucleoporins at the NPC. These data support a model in which E7 proteins may enter the nucleus via direct low affinity hydrophobic interactions with FG nucleoporins in a similar manner as the karyopherins/importins (Piccioli et al., 2010).

In this study we continued our analysis of the nucleocytoplasmic transport of low risk HPV11 E7 protein. We discovered and characterized a leucine-rich nuclear export signal (NES), 76IRQLQDLLL84, within the zinc-binding domain that mediates nuclear export of HPV11 E7 via a CRM1 pathway. We identified a mostly hydrophobic patch 65VRLVV69 within the zinc-binding domain that mediates nuclear import via hydrophobic interactions with the FG repeats domain of the central nucleoporin, Nup62. Substitutions of hydrophobic residues to alanine in this mostly hydrophobic patch 65VRLVV69 disrupt the nuclear localization of EGFP-11E7 and EGFP-11cE7, suggesting that their nuclear import is critically dependent on these residues. These data strongly support a model of nuclear entry of HPV11 E7 protein via hydrophobic interactions with FG nucleoporins at the NPC.

Results

The RJA nuclear export inhibitor partially restores the nuclear localization of HPV11 E7 cysteine mutants

We have previously discovered that HPV16 E7 oncoprotein is exported from the nucleus via a CRM1-mediated pathway (Knapp et al., 2009). To determine if HPV11 E7 can also undergo nuclear export we used transient transfections in HeLa cells with plasmids containing EGFP fused to 11E7 and mutants, and Ratjadone A (RJA), a specific nuclear export inhibitor that blocks CRM1-mediated nuclear export of proteins containing leucinerich NESs. RJA inhibition functions similarly to Leptomycin B (LMB): it covalently modifies a cysteine residue in the NES-binding pocket of CRM1 thereby inhibiting NES-binding (Hutten and Kehlenbach, 2007; Meissner et al., 2004). We analyzed the effect of RJA on the localization of two cysteine mutants, EGFP-11E7C58A and EGFP-11E7C91A, which were previously determined to have an altered cytoplasmic and pancellular localization due to the weakening of the NLS of 11E7. In contrast, the wild type EGFP-11E7 has a mostly nuclear localization (Piccioli et al., 2010). As a positive control for RJA, we used EGFP-16E7-NES, which contains a fusion of 16E7 with the strong leucine-rich NES of HIV Rev at its C-terminus (Knapp et al., 2009). Wild type EGFP-16E7 localizes to the nucleus, and addition of the NES of Rev changes its localization to cytoplasmic. Therefore, HeLa cells were transfected with EGFP-11E7 wild type, EGFP-11E7C58A, EGFP-11E7C91A, and EGFP-16E7NES plasmids and examined by confocal microscopy 24 h post transfection. RJA treatment was performed 20 h post transfection for 4 h. Wild type EGFP-11E7 localizes mostly to the nucleus in the absence or presence of RJA (Fig. 1A, panels A and C), with a slight increase in the percent of cells with nuclear localization from 80% to 90% in the presence of RJA (Fig. 1B). As expected, RJA changed the localization of EGFP-16E7NES from cytoplasmic to mostly nuclear in the great majority of cells (Fig. 1A, compare panels M and O, Fig. 1B). Interestingly, RJA also changed the localization of EGFP-11E7C58A and EGFP-11E7C91A from cytoplasmic and pancellular to mostly nuclear (Fig. 1A, compare panel E with G, and panel I with K). Quantitative analysis indicated that the effect of RJA was significant, with a change for EGFP-11E7C91A from 90% of the cells with cytoplasmic phenotype to 55% cells with mostly nuclear localization (and the rest of cells with pancellular localization) in the presence of RJA (Fig. 1B). RJA treatment also partially rescued the nuclear localization of EGFP-11E7C58A to 55% of the cells, with the remaining cells having a pancellular phenotype (Fig. 1B). These data suggest that HPV11 E7 can undergo CRM1-mediated nuclear export, which can be specifically inhibited by RJA.

Fig. 1.

Fig. 1

Open in a new tab

(A) The RJA nuclear export inhibitor partially restores the nuclear localization of EGFP-11E7 C58A and C91A mutants. HeLa cells were transfected with either EGFP-11E7 wild type (panels A–D), EGFP-11E7C58A (panels E–H), EGFP-11E7C91A (panels I–L), and EGFP-16E7NES (panels M–P) plasmids in the absence (−RJA) or in the presence of RJA (+RJA) and examined by confocal fluorescence microscopy at 24 h post-transfection. Panels A, C, E, G, I, K, M, and O represent the fluorescence of the EGFP and panels B, D, F, H, J, L, N, and P the DAPI staining of the nuclei. (B) Quantitative analysis of the effect of RJA on the localization of the EGFP-11E7 C58A and C91A mutants. The data from four experiments using EGFP-11E7, EGFP-11E7C58A, EGFP-11E7C91A, and EGFP-16E7NES plasmids, with or without RJA treatment, have been used for the quantitative analysis and the graphic representation. Mostly nuclear, black bars; pancellular, gray bars; cytoplasmic, dotted bars.

Analysis of HPV11 E7 revealed a potential leucine-rich NES (76IRQLQDLLL84) located in the C-terminus and homologous with the previously characterized NES of HPV16 E7 (Knapp et al., 2009). Therefore to further analyze the nuclear export of HPV11 E7 we examined the effect of RJA on the same two cysteine mutants in the context of its C-terminal domain, 11cE7. HeLa cells were transfected with EGFP-11cE7 wild type, EGFP-11cE7C58A, EGFP-11cE7C91A, and EGFP-16E7NES plasmids and examined by confocal microscopy 24 h post transfection. Wild type EGFP-11cE7 was localized to the nucleus of 80% of the transfected cells in the absence of RJA (Fig. 2A and B), and in its presence this nuclear localization increased to 90% cells (Fig. 2A and B). RJA treatment led to a partial recovery of the localization of EGFP-11cE7C91A mutant with approximately 10% of the cells having a mostly nuclear phenotype and 75% with pancellular localization in contrast with over 90% of cells having cytoplasmic localization in the absence of RJA (Fig. 2A, compare panels I and K; and Fig. 2B). The effect of RJA on the localization of EGFP-11cEC58A was more modest as in the absence of RJA the mutant already had a pancellular localization in 65% of the cells, with the rest exhibiting a cytoplasmic localization, while in the presence of RJA, 15% of the cells had a mostly nuclear phenotype and 80% had a pancellular localization (Fig. 2B). These data suggest that HPV11 E7 has a functional leucine-rich NES located within its C-terminal domain whose activity can be inhibited by RJA.

Fig. 2.

Fig. 2

Open in a new tab

(A) The RJA nuclear export inhibitor partially restores the nuclear localization of EGFP-11cE7 C58A and C91A mutants. HeLa cells were transfected with either EGFP-11cE7 wild type (panels A–D), EGFP-11cE7C58A (panels E–H), EGFP-11cE7C91A (panels I–L), and EGFP-16E7NES (panels M–P) plasmids in the absence (−RJA) or in the presence of RJA (+RJA) and examined by confocal fluorescence microscopy at 24 h post transfection. Panels A, C, E, G, I, K, M, and O represent the fluorescence of the EGFP and panels B, D, F, H, J, L, N, and P the DAPI staining of the nuclei. (B) Quantitative analysis of the effect of RJA on the localization of EGFP-11cE7 C58A and C91A mutants. The data from six experiments using EGFP-11cE7, EGFP-11cE7C58A, EGFP-16E7NES plasmids, and four experiments using EGFP-11cE7C91A plasmid, with or without RJA treatment, have been used for the quantitative analysis and the graphic representation. Mostly nuclear, black bars; pancellular, gray bars; cytoplasmic, dotted bars.

HPV11 E7 has a leucine-rich NES and mutation of critical leucine residues inhibit its nuclear export function

We have previously determined that there is a leucine-rich NES (76IRTLEDLLM84) located within the zinc-binding domain of HPV16 E7 oncoprotein that mediates its nuclear export in a CRM1-dependent manner (Knapp et al., 2009). Comparison of the sequences of HPV16 E7 NES with the homologous sequences in other HPV types showed that the hydrophobic residues are conserved in other HPV types, including HPV11 E7 (76IRQLQDLLL84). In order to investigate if this homologous NES of 11E7 is functional, we mutated the leucine residues, one at position 79 and three at positions 82–84 in the context of EGFP-11E7C91A and EGFP-11E7C58A mutants and analyzed the localization of these double mutants 24 h post-transfection in HeLa cells. The localization of both EGFP-11E7C91A/L79A and EGFP-11E7C91A/82LLL/AAA double mutants was changed in comparison with that of EGFP-11E7C91A, which had a cytoplasmic localization in 90% of the transfected cells (Fig. 3A, panels D and F, and Fig. 3B). The EGFP-11E7C91A/L79A had a mostly nuclear phenotype in some 20% of the cells and pancellular localization in approximately 75% of the cells (Fig. 3A, panels G and I, and Fig. 3B). Similarly, EGFP-11E7C91A/82LLL/AAA displayed a mostly nuclear localization in 20% of the cells and pancellular localization in approximately 75% of the cells (Fig. 3A, panels J and L, and Fig. 3B).

Fig. 3.

Fig. 3

Open in a new tab

(A) Substitutions of critical leucine residues to alanine in a potential cNES change the localization of the EGFP-11E7 C91A mutant. HeLa cells were transfected with either EGFP-11E7 wild type (panels A–C), EGFP-11E7C91A (panels D–F), EGFP-11E7C91A/L79A (panels G–I), and EGFP-11E7C91A/82LLL/AAA (panels J–L) plasmids and examined by confocal fluorescence microscopy at 24 h post transfection. Panels A, D, G, and J represent the fluorescence of the EGFP; panels B, E, H, and K the DAPI staining of the nuclei, and panels C, F, I and L the MERGE. (B) Quantitative analysis of the effect of mutations of critical leucine residues in the cNES on the localization of the EGFP-11E7 C91A mutant. The data from four experiments using EGFP-11E7, EGFP-11E7C91A, EGFP-11E7C91A/L79A, and EGFP-11E7C91A/82LLL/AAA plasmids have been used for the quantitative analysis and the graphic representation. Mostly nuclear, black bars; pancellular, gray bars; cytoplasmic, dotted bars. (C) Quantitative analysis of the effect of mutations of critical leucine residues in a potential cNES on the localization of the EGFP-11E7 C58A mutant. The data from four experiments using EGFP-11E7, EGFP-11E7C58A, EGFP-11E7C58A/L79A, and EGFP-11E7C58A/82LLL/AAA plasmids have been used for the quantitative analysis and the graphic representation. Mostly nuclear, black bars; pancellular, gray bars; cytoplasmic, dotted bars.

The localization of EGFP-11E7C58A/L79A and EGFP-11E7C58A/82LLL/AAA double mutants was also changed in comparison with the EGFP-11E7C58A single mutant (Fig. 3C). However, the effect of the leucine mutations on localization was more modest as EGFP-11E7C58A had already pancellular localization in approximately 70% of the cells and a cytoplasmic localization in only 30% of the cells (Fig. 3C). The EGFP-11E7C58A/L79A double mutant was localized to the nucleus in some 20% of the cells and throughout the cell (pancellular) in approximately 75% of the cells (Fig. 3C). EGFP-11E7C58A/82LLL/AAA was nuclear in some 10% of the cells and pancellular in approximately 85% cells (Fig. 3C). An immunoblot analysis of the expressed EGFP-11E7 and mutants showed that all the fusion proteins were expressed at similar levels and were not degraded (data not shown).

Overall, analysis of the localization of the 11E7 double mutants containing mutations of leucine residues in the NES suggests that the 76IRQLQDLLL84 sequence is indeed a functional NES for 11E7, mediating its nuclear export.

HPV11 E7 interacts with CRM1 nuclear export receptor

The identification of a functional leucine-rich NES mediating 11E7 nuclear export prompted us to analyze the interaction of 11E7 with CRM1 nuclear export receptor. HeLa cell lysates containing EGFP-11cE7, EGFP-16E7NES (as a positive control), or EGFP (as a negative control) were incubated with GST-CRM1 or GST immobilized on glutathione-Sepharose beads and the bound proteins were analyzed by immunobloting with a GFP antibody. As expected, EGFP-16E7NES bound to GST-CRM1 and not to GST itself (Fig. 4, lanes 2, 5 and 8); EGFP did not bind to either one (Fig. 4, lanes 3, 6 and 9). Interestingly, EGFP-11cE7 also bound to GST-CRM1 and not to GST itself (Fig. 4, lanes 1, 4 and 7). These data indicate that 11E7 interacts with CRM1 nuclear export receptor via its leucine-rich NES located in its C terminal domain (11cE7).

Fig. 4.

Fig. 4

Open in a new tab

HPV11 cE7 interacts with CRM1 nuclear export receptor. HeLa cells were transfected with EGFP-11cE7, EGFP-16E7NES or EGFP and the cell lysates were prepared 24 h post-transfection, subjected to SDS-PAGE and transfer to nitrocellulose and the blots probed with GFP antibody (lanes 1, 2 and 3). GST-CRM1 (lanes 4–6) and GST (lanes 7–9) immobilized on glutathione-Sepharose were incubated with the cell lysates and the bound proteins were eluted, subjected to SDS-PAGE and transfer to nitrocellulose and analyzed by immunobloting with a GFP antibody (EGFP-11cE7, lanes 4 and 7; EGFP-16E7NES, lanes 5 and 8; EGFP, lanes 6 and 9).

HPV11 E7 interacts with the FG nucleoporin Nup62

We have previously determined that HPV11 E7 enters the nucleus via a pathway mediated by its zinc-binding domain that is independent of karyopherins/importins (Piccioli et al., 2010). Moreover, previous phenyl-Sepharose binding results showed that 11E7 and 11cE7 can bind to phenyl-Sepharose (Piccioli et al., 2010), which mimics the specificity of FG repeats of nucleoporins and binds all the karyopherins (importins and exportins) (Ribbeck and Gorlich, 2002). These data suggested that 11E7 may be imported into the nucleus via low affinity hydrophobic interactions with the FG-nucleoporins at the NPC in a similar way as the karyopherins. To examine this hypothesis we chose the FG nucleoporin Nup62 that has been shown to interact with karyopherins and has a central location at the transport channel of the NPC (Solmaz et al., 2011). In these experiments we used Nup62N, containing the FG domain of Nup62 with six FG repeats, and Nup62C, representing the C-terminal domain of Nup62 lacking any FG repeats. Therefore, HeLa cell lysates containing EGFP-11E7, EGFP-11cE7, and EGFP (negative control) were incubated with GST-Nup62N, GST-Nup62C, or GST (negative control) immobilized on glutathione-Sepharose and the bound proteins were analyzed by immunobloting with a GFP antibody. As a positive control, we analyzed the binding of Kap β2/transportin nuclear import receptor to GST-Nup62N. As expected Kap β2 bound to GST-Nup62N, and not to GST-Nup62C or GST itself (Fig. 5, lanes 8, 12 and 16). Significantly, both EGFP-11E7 and EGFP-11cE7 bound to GST-Nup62N (Fig. 5, lanes 5 and 6) and not to the GST itself (Fig. 5, lanes 13 and 14). Moreover, EGFP-11E7 and EGFP-11cE7 did not bind to GST-Nup62C (Fig. 5, lanes 9 and 10). EGFP did not bind either to GST-Nup62N, GST-Nup62C or GST itself (Fig. 5, lanes 7, 11 and 15). These data indicate that HPV11 E7 protein can bind specifically to the FG-repeats domain of Nup62 via its zinc-binding domain.

Fig. 5.

Fig. 5

Open in a new tab

HPV11 E7 interacts via its zinc-binding domain with the FG domain of Nup62. HeLa cells were transfected with EGFP-11E7, EGFP-11cE7, and EGFP and the cell lysates were prepared 24 h post-transfection and probed with GFP antibody (lane 1, EGFP-11E7; lane 2, EGFP-11cE7; lane 3, EGFP). HeLa cells lysate was also probed for Kap β2 (lane 4). GST-Nup62N (lanes 5–7), GST-Nup62C (lanes 9–11) and GST (lanes 13–15) immobilized on glutathione-Sepharose were incubated with the cell lysates and the bound proteins were eluted and analyzed by immunobloting with a GFP antibody (lanes 5, 9 and 13, EGFP-11E7; lanes 6, 10 and 14, EGFP-11cE7; lanes 7, 11 and15, EGFP). As a positive control, binding of Kap β2 to GST-Nup62N, GST-Nup62C and GST was also analyzed (lanes 8, 12 and 16).

Identification of a patch of hydrophobic residues required for the nuclear localization of HPV11 E7

The surface hydrophobicity of nuclear import receptors is sufficient to provide access to the NPC and the translocation process involves non-specific interactions between small hydrophobic patches on the surface of the receptors/karyopherins and the FG repeats in nucleoporins (Naim et al., 2009). Significantly, 11cE7 domain is very rich in nonpolar, hydrophobic residues, with some of them arranged in hydrophobic patches, which could mediate the low affinity hydrophobic interactions with the FG-nucleoporins at the NPC. We identified a mostly hydrophobic patch in the 11E7 zinc binding domain, 65VRLVV69, and replaced all five residues with alanines in the context of EGFP-11E7 and EGFP-11cE7. We also generated two single mutants by substituting the arginine at position 66 with alanine in the context of EGFP-11E7 and EGFP-11cE7. HeLa cells were transfected with the wild type EGFP-11E7 and EGFP-11cE7 and the mutant plasmids and the localization of all the expressed proteins was analyzed by confocal fluorescence microscopy. As expected EGFP-11E7 and EGFP-11cE7 had a mostly nuclear localization in the great majority of cells (Fig. 6A and B), whereas EGFP itself had pancellular localization (Fig. 6B). Significantly, both the EGFP-11E7VRLVV/AAAAA and EGFP-11cE7VRLVV/AAAAA mutant had a mostly cytoplasmic localization in approximately 97% cells (Fig. 6A and B). In contrast, both the EGFP-11E7R66A and EGFP-11cE7R66A mutants were mostly in the nucleus in approximately 98% of the cells (Fig. 6A and B). These data strongly suggest that the hydrophobic residues are essential for the nuclear localization of 11E7 and 11cE7 whereas the positively charged arginine residue does not play any role in this process.

Fig. 6.

Fig. 6

Open in a new tab

(A) Mutations of hydrophobic residues within the zinc-binding domain disrupt the nuclear localization of EGFP-11E7 and EGFP-11cE7. HeLa cells were transfected with EGFP-11E7 (panels A, D and G), EGFP-11E7VRLVV/AAAAA (panels B, E and H), or EGFP-11E7R66A (panels C, F and I) plasmids, and the localization of the expressed proteins was analyzed by confocal fluorescence microscopy. Panels A, B and C represent the fluorescence of the EGFP and panels D, E, and F the DAPI staining of the nuclei, and panels G, H and I the MERGE. (B) Quantitative analysis showing that mutations of hydrophobic residues disrupt the nuclear localization of 11E7 and 11cE7. The data from four experiments using EGFP-11E7, EGFP-11E7VRLVV/AAAAA, EGFP-11E7R66A, EGFP-11cE7, EGFP-11cE7VRLVV/AAAAA, EGFP-11cE7R66A and EGFP plasmids have been used for the quantitative analysis and the graphic representation. Mostly nuclear, black bars; pancellular, gray bars; cytoplasmic, dotted bars. (C) Mutations of hydrophobic residues inhibit the interaction of 11E7 and 11cE7 with the FG domain of Nup62. HeLa cells were transfected with EGFP-11E7 (lane 1), EGFP-11E7VRLVV/AAAAA (lane 2), EGFP-11cE7 (lane 3), EGFP-11cE7VRLVV/AAAAA (lane 4), and EGFP (lane 5) and the cell lysates were prepared 24 h post-transfection and probed with a GFP antibody. HeLa cells lysate was also probed for Kap β2 (lane 6). GST-Nup62N (lanes 7–11) and GST (lanes 13–17) immobilized on glutathione-Sepharose were incubated with the cell lysates and the bound proteins were eluted and analyzed by immunobloting with a GFP antibody (lane 7, EGFP-11E7; lane 8, EGFP-11E7VRLVV/AAAAA; lane 9, EGFP-11cE7; lane 10, EGFP-11cE7VRLVV/AAAAA; lane 11, EGFP). As a positive control, binding of Kap β2 to GST-Nup62N and GST was also analyzed (lanes 12 and 18).

Significantly, the VRLVV/AAAAA mutation that disrupted the nuclear localization also inhibited the interaction of EGFP-11E7 and EGFP-11cE7 with the FG domain of Nup62 (Fig. 6C, compare lane 7 with 8, and lane 9 with 10). The positive binding control, Kap β2 bound to GST-Nup62N and not to GST itself (Fig. 6C, lanes 12 and 18), whereas EGFP did not bind to either one (Fig. 6C, lanes 11 and 18). In contrast, the R66A mutation did not affect the interaction of EGFP-11E7 and EGFP-11cE7 with the FG domain of Nup62 (data not shown). This is in agreement with the localization data showing that the R66A mutation did not affect the nuclear localization of EGFP-11E7 and EGFP-11cE7 (Fig. 6A and B). Overall, the data strongly suggest that the hydrophobic residues in the VRLVV sequence play an essential role in the hydrophobic interaction of the zinc-binding domain with the FG domain of Nup62 leading to nuclear import of 11E7.

Significantly, this hydrophobic patch is conserved in the E7 proteins of many other HPVs including mucosal and cutaneous, high and low risk types. Interestingly, HPV6 E7 containing the L67R mutation within its LRLCV hydrophobic sequence binds p130 in vitro, but loses its ability to target p130 for degradation and to decrease/delay differentiation in vivo (Zhang et al., 2006). We generated the EGFP-11E7L67R and EGFP-11cE7L67R mutants and analyzed their localization in comparison with the wild type proteins. We found that both EGFP-11E7L67R and EGFP-11cE7L67R mutants had a predominantly cytoplasmic localization (Fig. 7, panels B and J, and panels D and L) in the majority of cells (90.72 ± 1.85%, and 81.6 ± 1.23%, respectively) with the rest of transfected cells having a pancellular localization; this is in contrast with the mostly nuclear localization of the wild type proteins (Fig. 7, panels A and I, and panels C and K). These data indicate that the hydrophobic leucine 67 residue is essential for the nuclear import and localization of 11E7, and its substitution with a positively charged arginine residue disrupts 11E7 nuclear import.

Fig. 7.

Fig. 7

Open in a new tab

The L67R mutation disrupts the nuclear localization of EGFP-11E7 and EGFP-11cE7. HeLa cells were transfected with EGFP-11E7 (panels A, E and I), EGFP-11E7L67R (panels B, F and J), EGFP-11cE7 (panels C, G and K), or EGFP-11cE7L67R (panels D, H and L) plasmids, and the localization of the expressed proteins was analyzed by confocal fluorescence microscopy. Panels A–D represent the fluorescence of the EGFP, panels E–H the DAPI staining of the nuclei, and panels I–L the MERGE.

Discussion

In this study we continued our analysis of the nucleocytoplasmic traffic of low risk HPV11 E7 protein. We mapped and characterized a leucine-rich NES in the C terminal domain of low risk HPV11 E7 protein (76IRQLQDLLL84), which has a high degree of homology with the previously characterized NES sequence of HPV16 E7, 76IRTLEDLLM84 (Knapp et al., 2009). Changing critical leucine residues at position 79 or 82–84 to alanine inhibited the nuclear export activity of this HPV11 E7 NES. As the wild type HPV11 E7 exhibits a nuclear localization, these NES mutations were introduced in the context of two previously characterized zinc-binding deficient mutants, C58A and C91A, that have cytoplasmic and pancellular localization, due to weakening of the NLS of HPV11 E7 protein (Piccioli et al., 2010). Addition of the secondary NES mutations partially rescued the nuclear phenotype of the zinc-binding deficient mutants by inhibition of their nuclear export. The nuclear rescue did not reach the wild type level and this most likely is due to the weakening of nuclear import by the C58A and C91A mutations.

The cNES of HPV11 E7 mediates its nuclear export in a CRM1-dependent manner, being specifically inhibited by the RJA nuclear export inhibitor. Moreover, we determined that HPV11 E7 interacts with CRM1 nuclear export receptor via its cNES located in the C-terminal domain, 11cE7. We obtained similar results for the interaction between the cNES of 16cE7 and CRM1 (data not shown).

Overall, the data suggest that HPV11 E7 like HPV16 E7 shuttles between the nucleus and cytoplasm, which would be expected from their functions, as these E7 proteins have cellular targets both in the nucleus and in the cytoplasm. The different interactions of E7 proteins with their nuclear and cytoplasmic targets could modulate their nucleocytoplasmic traffic and consequently their steady-state intracellular localization in different physiological conditions and during the viral cycle. Interestingly, the balance between nuclear import and nuclear export of wild type HPV11 cE7 favors nuclear localization, whereas for HPV16 cE7 it favors cytoplasmic localization (Knapp et al., 2009; Piccioli et al., 2010).

We have previously established that the zinc-binding domain of HPV11 E7 mediates its nuclear import via a pathway independent of karyopherins/importins and suggested that this pathway may involve hydrophobic interactions between the zinc-binding domain and FG nucleoporins (Piccioli et al., 2010). In this study we found that HPV11 E7 indeed interacts via its zinc-binding domain with the FG domain (containing six FG repeats) of Nup62, a nucleoporin located in the midplane of the nuclear pore complex and involved in nuclear import and export (Solmaz et al., 2011). Moreover, we identified a mostly hydrophobic patch 65VRLVV69 within the zinc-binding domain that is essential for the nuclear localization of 11E7 and 11cE7. Mutations of the hydrophobic residues to alanine completely disrupted the nuclear localization and the EGFP-11E7VRLVV/AAAAA and EGFP-11cE7VRLVV/AAAAA mutants had cytoplasmic localization. In contrast, the R66A mutation did not affect the nuclear localization of EGFP-11E7R66A and EGFP-11cE7R66A, indicating that this positively charged residue does not play any role in nuclear import of HPV11 E7. Significantly, the VRLVV/AAAAA mutations inhibited the interaction of both 11E7 and 11cE7 with the FG domain of Nup62, suggesting that the hydrophobic residues in this sequence play a key role in the hydrophobic interaction of the zinc-binding domain with Nup62. This inhibitory effect correlated very well with the complete disruption of the nuclear localization of EGFP-11E7 and EGFP-11cE7. Overall, these data strongly suggest that low risk HPV11 E7 is imported into the nucleus via hydrophobic interactions of its zinc-binding domain with the FG domain of Nup62 at the nuclear pore complex. It is likely that HPV11 E7 may interact with other FG nucleoporins at the NPC, like Nup153 and Nup358. Preliminary data suggest that HPV11 E7 indeed binds via its zinc-binding domain to Nup153.

The mechanism of nuclear import of low risk HPV11 E7 via hydrophobic interactions of its zinc-binding domain with the FG domain of Nup62 is conserved with that of the high risk HPV16 E7 oncoprotein (manuscript in preparation). Significantly, we have found that the homologous hydrophobic sequences in HPV16 E7 and HPV8 E7, LRLCV and LRLFV, are also essential for the nuclear import of these E7 proteins, and mutation of the hydrophobic residues to alanine inhibit both nuclear import and the interaction with Nup62N (in preparation). Interestingly, the HPV6 E7L67R mutant, although it binds p130 in vitro, it loses its ability to target p130 for degradation and to decrease/delay differentiation (Zhang et al., 2006). As we found that the L67R mutation disrupts the nuclear localization of 11E7, it can be speculated that nuclear localization of HPV6 E7 may be required for its ability to target p130 for degradation in vivo (Zhang et al., 2006).

This mechanism of nuclear import of HPV E7 proteins via direct hydrophobic interactions of their zinc-binding domain with FG nucleoporins is unique in comparison with other HPV proteins like L1 and L2 capsid proteins and E6 oncoprotein, that use different karyopherins/importins to enter the nucleus (Bordeaux et al., 2006; Darshan et al., 2004; Klucevsek et al., 2006; Le Roux and Moroianu, 2003; Nelson et al., 2002).

The TAX protein of human T lymphotropic virus type 1 also enters the nucleus via direct hydrophobic interactions of its zinc-binding domain with the FG nucleoporin Nup62 at the nuclear pore complex (Tsuji et al., 2007). Moreover, mutations in the zinc-binding domain that block the nuclear import and localization of TAX also inhibit its interaction with Nup62 (Tsuji et al., 2007). There are also cellular proteins like PU.1 transcription factor, that enter the nucleus via a pathway independent of karyopherins, mediated by direct binding to the FG-nucleoporins Nup62 and Nup153 (Zhong et al., 2005).

Materials and methods

Mutagenesis to generate different EGFP-11E7 and EGFP-11cE7 mutants

EGFP-11E7 and EGFP-11cE7 wild type, and EGFP-11E7C58A, EGFP-11E7C91A, EGFP-11cE7C58A, EGFP-11cE7C91A mutant plasmids were obtained previously (Piccioli et al., 2010). The EGFP-11E7C58A/L79A and EGFP-11E7C58A/82LLL/AAA, EGFP-11E7C91A/L79A and EGFP-11E7C91A/82LLL/AAA double mutants were generated using the Quik-Change™ Site-Directed Mutagenesis Kit (Stratagene) with EGFP-11E7C58A and EGFP-11E7C91A as templates and corresponding mutagenesis primers for the substitutions in the NES.

The EGFP-11E7VRLVV/AAAAA, EGFP-11cE7VRLVV/AAAAA, EGFP-11E7R66A, EGFP-11cE7R66A, EGFP-11E7L67R and EGFP-11cE7L67R mutant plasmids were generated using the QuikChange™ Site-Directed Mutagenesis Kit (Stratagene) with EGFP-11E7 and EGFP-11cE7 as templates and the corresponding mutagenesis primers.

The mutant plasmids were transformed into XL1-blue competent cells and extracted using Quantum PrepR Plasmid MiniPrep kit (BioRad). All purified mutant plasmids were checked by sequencing (Eurofins MWG) and maintained in stock cultures at −80 °C before use in transfection experiments.

Transient expression of EGFP fusion proteins and confocal microscopy analysis

HeLa cells (ATCC) were plated on 12 mm poly-L-lysine-coated glass coverslips to 50% confluency 24 h prior to transfection. Cells in each well were transfected with the corresponding EGFP fusion plasmid (as indicated in the figure legends) and the FuGENE 6 reagent (Roche Applied Science, IN). Media was changed to DMEM with 10% FBS and 1% penicillin–streptomycin after 6 h and the cells were fixed 24 h after the initial transfection with 3.7% formaldehyde in PBS (10 min). Coverslips were mounted using Vectashield-DAPI mounting medium (Vector Labs, CA) to visualize the nuclei by DAPI staining. The slides were examined by fluorescence microscopy using a Leica TCS Sp5 broadband confocal microscope and pictures were taken using the Leica LAS AF software (Leica Microsystems).

Quantification of the intracellular localization phenotypes of the different EGFP-11E7 and EGFP-11cE7 proteins defined as mostly nuclear, pancellular or mostly cytoplasmic was performed on all the transfection experiments. The data from at least four experiments were used for the quantitative analysis and the graphical representations display average values with standard deviation.

Immunoblot analysis of the expressed EGFP-11E7, EGFP-11cE7 and different mutants performed with a GFP antibody indicated that the proteins were expressed at similar levels and not degraded.

Ratjadone A treatment of transfected cells

A 10 ng/ml solution of Ratjadone A (RJA) in DMEM+ was added to the transfected cells 20 h post transfection for 4 h. A parallel transfection received no RJA treatment, only fresh DMEM+ at 20 h post transfection for 4 h. Cells were then fixed and the localization of different EGFP fusion proteins analyzed as described above. The experiments presented in the figures and used for the quantitation analysis were performed in parallel.

Preparation of GST-CRM1, GST-Nup62N and GST-Nup62C

The GST-CRM1 plasmid was kindly provided by Dr. Jorgen Kjems and the GST-Nup62N (containing aa 1–265 of Nup62) and GST-Nup62C (containing aa 178–522 of Nup62) plasmids (Zhong et al., 2005) were kindly provided Dr. Nabeel Yaseen. For protein expression and purification the GST-CRM1, GST-Nup62N and GST-Nup62C plasmids were used to transform E. coli BL21 CodonPlus. After induction of the bacteria with 1 mM IPTG for 3 h the GST-fusion proteins were purified in their native state on Glutathione-Sepharose beads using a standard procedure.

Binding assays

Before the binding assays GST-CRM1 and GST immobilized on glutathione-Sepharose beads were analyzed by SDS-PAGE and Coomassie staining. This analysis showed that the proteins were intact and bound to the beads at similar levels. HeLa cell lysates containing the expressed EGFP-11cE7, EGFP-16E7NES (as a positive control), or EGFP (as a negative control) and 1 mM GTP-γs were incubated with either GST-CRM1 or GST immobilized on glutathione-Sepharose beads for 1 h at 4 °C. After washing the beads to remove nonspecific binding the bound proteins were eluted with SDS-sample buffer, subjected to SDS-PAGE and analyzed by immunobloting with a GFP antibody.

Before the binding assays GST-Nup62N, GST-Nup62C and GST immobilized on glutathione-Sepharose were analyzed by SDS-PAGE and Coomassie staining. This analysis revealed for GST-Nup62N and GST-Nup62C the presence of the intact proteins bound at similar levels to the beads, and also some degradation. HeLa cell lysates containing the expressed EGFP-11E7, EGFP-11cE7, and EGFP (negative control) were incubated with GST-Nup62N (containing the FG domain of Nup62), GST-Nup62C (lacking FG repeats) or with GST immobilized on glutathione-Sepharose for 1 h at 4 °C. After washing the beads to remove nonspecific binding the bound proteins were analyzed by immunobloting with a GFP antibody. As a positive control for binding to Nup62N, we analyzed the binding of Kap β2/transportin nuclear import receptor to GST-Nup62N by incubating HeLa cell lysate with GST-Nup62N, GST-Nup62C or GST and the bound proteins were analyzed by immunobloting with a Kap β2 antibody.

Acknowledgments

We thank Dr. Jorgen Kjems for the GST-CRM1 plasmid and Dr. Nabeel Yaseen for the GST-Nup62N and GST-Nup62C plasmids. We also thank Dr. Joshua Rosenberg for excellent technical assistance with confocal fluorescence microscopy and Jeremy Eberhard for help with mutagenesis. Courtney McKee was a Beckman Scholar and she was supported by a Beckman fellowship. This work was supported by a grant from the National Institutes of Health (R01 CA94898) to Junona Moroianu.

Abbreviations

HPV

human papillomavirus

NLS

nuclear localization signal

NES

nuclear export signal

EGFP

enhanced green fluorescent protein

GST

glutathione-S-transferase

References

  1. Angeline M, Merle E, Moroianu J. The E7 oncoprotein of high-risk human papillomavirus type 16 enters the nucleus via a nonclassical Ran-dependent pathway. Virology. 2003;317:13–23. doi: 10.1016/j.virol.2003.08.024. [DOI] [PubMed] [Google Scholar]
  2. Baldwin A, Huh KW, Munger K. Human papillomavirus E7 oncoprotein dysregulates steroid receptor coactivator 1 localization and function. J Virol. 2006;80:6669–6677. doi: 10.1128/JVI.02497-05. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bordeaux J, Forte S, Harding E, Darshan MS, 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. 2006;80:8259–8262. doi: 10.1128/JVI.00776-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Darshan MS, 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. 2004;78:12179–12188. doi: 10.1128/JVI.78.22.12179-12188.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Doorbar J. Molecular biology of human papillomavirus infection and cervical cancer. Clin Sci (Lond) 2006;110:525–541. doi: 10.1042/CS20050369. [DOI] [PubMed] [Google Scholar]
  6. Fiedler M, Muller-Holzner E, Viertler HP, Widschwendter A, Laich A, Pfister G, Spoden GA, Jansen-Durr P, Zwerschke W. High level HPV-16 E7 oncoprotein expression correlates with reduced pRb-levels in cervical biopsies. FASEB J. 2004;18:1120–1122. doi: 10.1096/fj.03-1332fje. [DOI] [PubMed] [Google Scholar]
  7. Genovese NJ, Banerjee NS, Broker TR, Chow LT. Casein kinase II motifdependent phosphorylation of human papillomavirus E7 protein promotes p130 degradation and S-phase induction in differentiated human keratinocytes. J Virol. 2008;82:4862–4873. doi: 10.1128/JVI.01202-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Guccione E, Massimi P, Bernat A, Banks L. Comparative analysis of the intracellular location of the high- and low-risk human papillomavirus oncoproteins. Virology. 2002;293:20–25. doi: 10.1006/viro.2001.1290. [DOI] [PubMed] [Google Scholar]
  9. Huh KW, DeMasi J, Ogawa H, Nakatani Y, Howley PM, Munger K. Association of the human papillomavirus type 16 E7 oncoprotein with the 600-kDa retinoblastoma protein-associated factor, p600. Proc Nat Acad Sci U S A. 2005;102:11492–11497. doi: 10.1073/pnas.0505337102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Hutten S, Kehlenbach RH. CRM1-mediated nuclear export: to the pore and beyond. Trends Cell Biol. 2007;17:193–201. doi: 10.1016/j.tcb.2007.02.003. [DOI] [PubMed] [Google Scholar]
  11. Klucevsek K, Daley J, Darshan MS, Bordeaux J, Moroianu J. Nuclear import strategies of high-risk HPV18 L2 minor capsid protein. Virology. 2006;352:200–208. doi: 10.1016/j.virol.2006.04.007. [DOI] [PubMed] [Google Scholar]
  12. Knapp AA, McManus PM, Bockstall K, Moroianu J. Identification of the nuclear localization and export signals of high risk HPV16 E7 oncoprotein. Virology. 2009;383:60–68. doi: 10.1016/j.virol.2008.09.037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Le Roux LG, Moroianu J. Nuclear entry of high-risk human papillomavirus type 16 E6 oncoprotein occurs via several pathways. J Virol. 2003;77:2330–2337. doi: 10.1128/JVI.77.4.2330-2337.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. McLaughlin-Drubin ME, Munger K. The human papillomavirus E7 oncoprotein. Virology. 2009;384:335–344. doi: 10.1016/j.virol.2008.10.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Meissner T, Krause E, Vinkemeier U. Ratjadone and leptomycin B block CRM1-dependent nuclear export by identical mechanisms. FEBS Lett. 2004;576:27–30. doi: 10.1016/j.febslet.2004.08.056. [DOI] [PubMed] [Google Scholar]
  16. Naim B, Zbaida D, Dagan S, Kapon R, Reich Z. Cargo surface hydrophobicity is sufficient to overcome the nuclear pore complex selectivity barrier. EMBO J. 2009;28:2697–2705. doi: 10.1038/emboj.2009.225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Nelson LM, Rose RC, Moroianu J. Nuclear import strategies of high risk HPV16 L1 major capsid protein. J Biol Chem. 2002;23:23. doi: 10.1074/jbc.M200724200. [DOI] [PubMed] [Google Scholar]
  18. Nguyen CL, Munger K. Human papillomavirus E7 protein deregulates mitosis via an association with nuclear mitotic apparatus protein 1. J Virol. 2009;83:1700–1707. doi: 10.1128/JVI.01971-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Piccioli Z, McKee CH, Leszczynski A, Onder Z, Hannah EC, Mamoor S, Crosby L, Moroianu J. The nuclear localization of low risk HPV11 E7 protein mediated by its zinc binding domain is independent of nuclear import receptors. Virology. 2010;407:100–109. doi: 10.1016/j.virol.2010.07.045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Ribbeck K, Gorlich D. The permeability barrier of nuclear pore complexes appears to operate via hydrophobic exclusion. EMBO J. 2002;21:2664–2671. doi: 10.1093/emboj/21.11.2664. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Solmaz SR, Chauhan R, Blobel G, Melcak I. Molecular architecture of the transport channel of the nuclear pore complex. Cell. 2011;147:590–602. doi: 10.1016/j.cell.2011.09.034. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Terry LJ, Wente SR. Flexible gates: dynamic topologies and functions for FG nucleoporins in nucleocytoplasmic transport. Eukaryot Cell. 2009;8:1814–1827. doi: 10.1128/EC.00225-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Tsuji T, Sheehy N, Gautier VW, Hayakawa H, Sawa H, Hall WW. The nuclear import of the human T lymphotropic virus type I (HTLV-1) tax protein is carrier- and energy-independent. J Biol Chem. 2007;282:13875–13883. doi: 10.1074/jbc.M611629200. [DOI] [PubMed] [Google Scholar]
  24. Zhang B, Chen W, Roman A. The E7 proteins of low- and high-risk human papillomaviruses share the ability to target the pRB family member p130 for degradation. Proc Nat Acad Sci U S A. 2006;103:437–442. doi: 10.1073/pnas.0510012103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Zhong H, Takeda A, Nazari R, Shio H, Blobel G, Yaseen NR. Carrier-independent nuclear import of the transcription factor PU.1 via RanGTP-stimulated binding to Nup153. J Biol Chem. 2005;280:10675–10682. doi: 10.1074/jbc.M412878200. [DOI] [PubMed] [Google Scholar]
  26. zur Hausen H. Papillomaviruses causing cancer: evasion from host-cell control in early events in carcinogenesis. J Natl Cancer Inst. 2000;92:690–698. doi: 10.1093/jnci/92.9.690. [DOI] [PubMed] [Google Scholar]
  27. zur Hausen H. Papillomaviruses in the causation of human cancers—a brief historical account. Virology. 2009;384:260–265. doi: 10.1016/j.virol.2008.11.046. [DOI] [PubMed] [Google Scholar]

RESOURCES