Human Papillomavirus 11 Early Protein E6 Activates Autophagy by Repressing AKT/mTOR and Erk/mTOR

We transfected an HPV11 E6 plasmid into HaCaT cells, H8 cells, and NHEK cells and established a stable cell line expressing the HPV11 E6 protein. Then, we confirmed that HPV11 E6 induces autophagy by suppressing the AKT/mTOR and Erk/mTOR pathways. In contrast to the high-risk HPV E6 genes, HPV11 E6 did not affect the expression of p53. To the best of our knowledge, this study represents the first direct in-depth investigation of the relationship between the LR-HPV E6 gene and autophagy, which may help to reveal the pathogenesis of LR-HPV infection.

ABSTRACT

Low-risk human papillomaviruses (LR-HPVs) are the causative agents of genital warts, which are a widespread sexually transmitted disease. How LR-HPVs affect autophagy and the specific proteins involved are unknown. In the current study, we investigated the impact of LR-HPV11 early protein 6 (E6) on the activity of the autophagy pathway. We transfected an HPV11 E6 (11E6) plasmid into HaCaT cells, H8 cells, and NHEK cells and established a stable cell line expressing the HPV11 E6 protein. The differences in autophagy activity and upstream regulatory pathways compared with those in the parent cell lines were investigated using a Western blot analysis of the total and phosphorylated protein levels and confocal microscopy of immunostained cells and cells transfected with an mCherry-green fluorescent protein-LC3 expression plasmid. We used short hairpin RNA (shRNA) to knock down 11E6 and showed that these effects require continued 11E6 expression. Compared with its expression in the control cells, the expression of HPV11 E6 in the cells activated the autophagy pathway. The increased autophagy activity was the result of the decreased phosphorylation levels of the canonical autophagy repressor mammalian target of rapamycin (mTOR) at its Ser2448 position (the mTOR complex 1 [mTORC1] phosphorylation site) and decreased AKT and Erk phosphorylation. Therefore, these results indicate that HPV11 E6 activates autophagy through the AKT/mTOR and Erk/mTOR pathways. Our findings provide novel insight into the relationship between LR-HPV infections and autophagy and could help elucidate the pathogenic mechanisms of LR-HPV.

IMPORTANCE We transfected an HPV11 E6 plasmid into HaCaT cells, H8 cells, and NHEK cells and established a stable cell line expressing the HPV11 E6 protein. Then, we confirmed that HPV11 E6 induces autophagy by suppressing the AKT/mTOR and Erk/mTOR pathways. In contrast to the high-risk HPV E6 genes, HPV11 E6 did not affect the expression of p53. To the best of our knowledge, this study represents the first direct in-depth investigation of the relationship between the LR-HPV E6 gene and autophagy, which may help to reveal the pathogenesis of LR-HPV infection.

INTRODUCTION

Human papillomaviruses (HPVs) are small double-stranded DNA viruses that include more than 200 genotypes (1, 2). Based on differences in their pathogenicity, the HPV family can be further classified into low-risk HPV (LR-HPV) types and high-risk HPV (HR-HPV) types (3). LR-HPVs, particularly HPV6 and HPV11, are etiologically associated with genital warts, such as condylomata acuminata (CA) (4), which have become one of the most widespread sexually transmitted diseases, with an annual global incidence of 160 to 289 cases per 100,000 (5, 6). Due to the infectious and proliferative nature of HPV, CA has become a serious threat to public health, and affected patients are prone to frequent recurrences and undergo multiple surgical procedures (7).

The HPV genome can be divided into the following three regions: the early (E) region, late (L) region, and noncoding long control region (7, 8). The E region encodes the six nonstructural proteins E1, E2, E4, E5, E6, and E7 (9, 10). Numerous studies have revealed that E1 (11, 12), E2 (13, 14), E4 (15, 16), and E5 (17, 18) are required for viral DNA replication, while E6 and E7 cooperate to affect the cell status (19, 20). The HR-HPV type E6 protein forms a trimeric complex with p53 and E6AP, which is a ubiquitin ligase (21, 22). This complex enhances p53 degradation through the ubiquitin-proteasomal route, resulting in the loss of p53-mediated cell cycle arrest and apoptotic functions (23). However, the pathogenic mechanisms of the LR-HPV types are distinct from those of the HR-HPV types (24, 25), and only a few studies have examined the possible functions of the LR-HPV E6 protein. The LR-HPV11 E6 protein (11E6) has previously been shown to associate with E6AP in vivo, and this protein should thus be able to target E6-associated proteins for degradation (26). Indeed, Bak degradation through the LR-HPV E6-E6AP complex has been observed (27).

Recently, studies investigating the interplay between HPV and autophagy have achieved much progress, but these studies have focused predominantly on the HR-HPV types (28). Autophagy is a conserved, catabolic process occurring in eukaryotic cells through which damaged or dysfunctional proteins, nucleic acids, and organelles are selectively isolated in double-membrane vesicles and targeted to lysosomes for degradation (29, 30). In addition, the cellular autophagic machinery can capture and degrade intracellular pathogens. Therefore, autophagy plays a crucial role in the host defense against viral infections (29, 31). The capture and degradation of intracellular viruses involve selective targeting by ubiquitin-dependent/independent mechanisms of recognition by a double-membrane phagophore, which subsequently engulfs the virus particles. The phagophore then matures into a phagosome and fuses with lysosomes for the degradation of its contents (32, 33). The canonical autophagy pathway is repressed by mammalian target of rapamycin (mTOR) complex 1 (mTORC1) and its upstream AKT or mitogen-activated protein kinase (MAPK) signaling cascades through its phosphorylation of Unc-51-like kinase 1 (ULK1) at Ser757; this process prevents ULK1 translocation to phagophore assembly sites (34, 35). Autophagy can be activated by the suppressive effects of AMPK and p53 on mTOR (36, 37).

Some viruses can antagonize different steps of the autophagy pathway or use the autophagy pathway to their advantage (31, 32). Regarding HPV, infection with an HPV16 pseudovirus has been shown to lead to the appearance of autophagosomes (38). Furthermore, in primary human keratinocytes, which have a higher basal autophagy level, HPV virions further activate autophagy, resulting in the inhibition of infection (39). In contrast, interactions with the keratinocyte cell line HaCaT resulted in the activation of the phosphatidylinositol 3-kinase (PI3K)/AKT/mTOR signaling pathway and the inhibition of autophagy, which was critical for early infection (40). Although the mechanism has not yet been established, the role of the early HPV proteins E6 and E7 in HPV-autophagy interactions has been indicated. Previously, HPV16 E7-expressing primary human foreskin keratinocytes have been shown to express high levels of LC3-II, which could be abrogated by the autophagy inhibitor 3-methyladenine (3-MA) (41). In addition, autophagy was induced in SiHa cells infected with an E6/E7 knockout virus, which was accompanied by reduced p62 expression (42).

In this study, we investigated the impact of the early protein E6 from LR-HPV11 on the activity of the autophagy pathway. Using an E6-expressing HaCaT cell line, we show that HPV11 E6 induces autophagy activity, which is the result of mTOR pathway suppression due to changes in the phosphorylation state of the PI3K/AKT and MAPK/Erk pathways.

RESULTS

Whole-genome expression profiling of HPV11 E6-expressing HaCaT cells.

To explore the impact of HPV11 E6 on HaCaT cells, we examined the whole-genome expression profiles of HPV11 E6-expressing and control cells by microarrays. In total, 2,967 genes were differentially expressed (Fig. 1A), including 1,032 genes that were upregulated in E6-expressing cells (red dots) and 1,935 genes that were downregulated in E6-expressing cells (blue dots). Further analysis revealed differences in the expression levels of autophagy-related genes between the E6-expressing cells and control cells (Fig. 1B). In particular, ATG10 exhibited a 2.8-fold induction in E6-expressing cells, which was further confirmed by quantitative real-time PCR (qRT-PCR) (Fig. 1C). To investigate whether the differences in ATG10 gene expression levels were also reflected at the protein level, a Western blot analysis was performed. Indeed, the E6-expressing cells exhibited higher ATG10 protein levels than the control cells (Fig. 1D). Since ATG10 catalyzes the conjugation of ATG12 to ATG5, free ATG12 and ATG12-ATG5 complexes were investigated by Western blotting. However, no differences in the ATG12 levels were observed between E6-expressing and control cells (Fig. 1D). Overall, these results indicate that HPV11 E6 might have an impact on the autophagy pathway.

FIG 1.

Autophagy-related genes are differentially expressed in HPV11 E6-expressing HaCaT cells. (A) Scatter plot of whole-genome expression profiles of HPV11 E6-expressing (y axis) and control (x axis) cells as determined by microarrays. Genes showing at least a 2-fold increase in expression in HPV11 E6-expressing cells are indicated in red, while genes showing at least a 2-fold reduction in expression are indicated in blue. (B) Heat map of expression profiles of autophagy-related genes in HPV11 E6-expressing and control HaCaT cells. Expression levels ranging from low (blue) to high (red) are indicated by color coding. (C) Quantitative real-time PCR analysis of ATG10 gene expression in HPV11 E6-expressing and control cells. Relative expression levels of ATG10 were normalized to the expression of actin. The graph represents the mean ± SEM from three independent experiments. (D) Western blot analyses of ATG10 and the ATG5-ATG12 complex in HPV11 E6-expressing and control cells. Actin was used as a loading control. Protein levels were quantified and normalized to the level of the actin control. The graphs represent the mean ± SEM from three independent experiments. Significant differences were identified by Student's t test. *, P < 0.05.

HPV11 E6 induces autophagy.

To investigate the activity of the autophagy pathway in HPV11 E6-expressing cells, the autophagy activity marker proteins LC3B, p62, and Beclin-1 were examined by Western blotting. The expression levels of both LC3B-II and Beclin-1 were higher in E6-expressing cells than in control cells, but no differences in p62 were observed (Fig. 2A and B, 3A and B, 4A and B). To further investigate the functional activity of the increased LC3B-II levels in the E6-expressing cells, confocal microscopy experiments were performed after immunostaining for LC3B. Indeed, E6-expressing cells showed the increased formation of LC3B puncta (Fig. 2C), which are a hallmark of autophagosome formation. To investigate whether the increased LC3B-II levels in the E6-expressing cells were the result of increased autophagy activity or inhibited recycling, autophagy flux was analyzed by inhibiting the autophagosome-lysosome fusion step with the V-ATPase inhibitor bafilomycin A1 (Baf-A1). Although both E6-expressing and control cells showed increased LC3B-II levels after the Baf-A1 treatment (Fig. 2D and E), LC3B-II levels in the E6-expressing cells were still higher, indicating that HPV11 E6 is indeed involved in autophagy induction. To determine whether these effects require continuous 11E6 expression, short hairpin RNA 1 (shRNA-1), shRNA-2, and shRNA-3 were used to knock down 11E6 expression in the 11E6-expressing stable cell line. The efficiency of 11E6 knockdown in the 11E6-expressing HaCaT cell line by shRNA-1, shRNA-2, and shRNA-3 is shown in Fig. 2F. Among these three shRNAs, the efficiency of 11E6 knockdown by shRNA-3 was the highest. The autophagy activity marker protein LC3B was examined by Western blotting after 11E6 knockdown by shRNA-3. Compared with that in 11E6-expressing cells, LC3B-II expression was reduced after 11E6 knockdown by shRNA-3 (Fig. 2G and H).

FIG 2.

Autophagy is induced in HPV11 E6-expressing HaCaT cells. (A) Western blot analysis of the autophagy marker proteins LC3B, Beclin-1, and p62. Representative images of three independent biological replicates are shown, and actin was used as a loading control. (B) Quantitative analysis of the Western blots. The protein levels of LC3B-II, Beclin-1, and p62 were normalized to the level of the actin control. The data are expressed as the mean ± SEM from three independent experiments. Significant differences were identified by Student's t test. *, P < 0.05. (C) Immunostaining and confocal microscopy analysis of endogenous LC3B. The HaCaT cell nuclei were stained with DAPI (blue). (D) Western blot analysis of the LC3B-II levels after treatment with the V-ATPase inhibitor bafilomycin A1 (Baf-A1). Representative images of three independent biological replicates are shown, and actin was used as a loading control. (E) Quantitative analysis of the Western blots. The protein levels of LC3B-II were normalized to the level of the actin control. The data are expressed as the mean ± SEM from three independent experiments. Significant differences were identified by Student's t test. *, P < 0.05. (F) The relative expression of HPV11 E6 was analyzed using real-time PCR after transfecting the HPV11 E6-expressing HaCaT stable cell line with shRNA-1, shRNA-2, and shRNA-3. Significant differences were identified by Student's t test. *, P < 0.05. (G) Western blot analysis of LC3B levels following 11E6 knockdown by shRNA-3. (H) Quantitative analysis of the Western blots. The protein levels of LC3B-II were normalized to the level of the actin control. The data are expressed as the mean ± SEM from three independent experiments. Significant differences were identified by Student's t test. *, P < 0.05.

FIG 3.

HPV11 E6 induces autophagy by inhibiting the AKT/mTOR and Erk/mTOR pathways in the H8 cell line. (A) Western blot analysis of the autophagy marker proteins LC3B, Beclin-1, and p62. Representative images of three independent biological replicates are shown, and actin was used as a loading control. (B) Quantitative analysis of the Western blots. The protein levels of LC3B-II, Beclin-1, and p62 were normalized to the level of the actin control. The data are expressed as the mean ± SEM from three independent experiments. Significant differences were identified by Student's t test. *, P < 0.05. (C) Western blot analysis of phosphorylated and nonphosphorylated mTOR (Ser2448) and its substrate proteins 4EBP1 and 70S6K in HPV11 E6-expressing and control H8 cells (phosphorylated proteins are identified by the prefix “p”). Actin was used as a loading control. The ratio of phosphorylated and nonphosphorylated proteins was quantified. The graph represents the mean ± SEM from three independent repeats. Significant differences were identified by Student's t test. *, P < 0.05. (D) The two phosphorylation sites of ULK1, Ser757, and Ser555 were analyzed by Western blotting. Actin was used as a loading control. The ratio of phosphorylated and nonphosphorylated ULK1 was quantified. The graphs represent the mean ± SEM from three independent repeats. Significant differences were identified by Student's t test. *, P < 0.05. (E) Western blot analysis of phosphorylated and nonphosphorylated AKT and Erk. Actin was used as a loading control. The ratio of phosphorylated and nonphosphorylated AKT, Erk, and actin was quantified. The graphs represent the mean ± SEM from three independent repeats. Significant differences were identified by Student's t test. *, P < 0.05. (F) Western blot analysis of phosphorylated and nonphosphorylated AMPK and the total protein levels of p53. Actin was used as a loading control. The ratio of phosphorylated and nonphosphorylated AMPK and the ratio of p53 and actin were quantified. The graphs represent the mean ± SEM from three independent repeats.

FIG 4.

HPV11 E6 induces autophagy by inhibiting the AKT/mTOR and Erk/mTOR pathways in NHEK cells. (A) Western blot analysis of the autophagy marker proteins LC3B, Beclin-1, and p62. Representative images of three independent biological replicates are shown, and actin was used as a loading control. (B) Quantitative analysis of the Western blots. The protein levels of LC3B-II, Beclin-1, and p62 were normalized to the level of the actin control. The data are expressed as the mean ± SEM from three independent experiments. Significant differences were identified by Student's t test. *, P < 0.05. (C) Western blot analysis of phosphorylated and nonphosphorylated mTOR (Ser2448) and its substrate proteins 4EBP1 and 70S6K in HPV11 E6-expressing and control NHEK cells. Actin was used as a loading control. The ratio of phosphorylated and nonphosphorylated proteins was quantified. The graph represents the mean ± SEM from three independent repeats. Significant differences were identified by Student's t test. *, P < 0.05. (D) The two phosphorylation sites of ULK1, Ser757 and Ser555, were analyzed by Western blotting. Actin was used as a loading control. The ratio of phosphorylated and nonphosphorylated ULK1 was quantified. The graphs represent the mean ± SEM from three independent repeats. Significant differences were identified by Student's t test. *, P < 0.05. (E) Western blot analysis of phosphorylated and nonphosphorylated AKT and Erk. Actin was used as a loading control. The ratios of phosphorylated and nonphosphorylated AKT, Erk, and actin were quantified. The graphs represent the mean ± SEM from three independent repeats. Significant differences were identified by Student's t test. *, P < 0.05. (F) Western blot analysis of phosphorylated and nonphosphorylated AMPK and the total protein levels of p53. Actin was used as a loading control. The ratio of phosphorylated and nonphosphorylated AMPK and the ratio of p53 and actin were quantified. The graphs represent the mean ± SEM from three independent repeats.

HPV11 E6 induces autophagy flux and autolysosome formation.

To investigate the functional autophagy activity in E6-expressing cells, autophagosome maturation into autolysosomes was investigated after transfection with a plasmid expressing an mCherry-green fluorescent protein (GFP)-LC3B fusion protein. Since GFP is acid sensitive, its fluorescent activity is lost in the acidic environment of the autolysosome. Therefore, autophagosomes that mature into autolysosomes are visible as red puncta. Indeed, more red puncta were observed in E6-expressing cells than in control cells, indicating that the autophagosomes in E6-expressing cells could mature into autolysosomes. As a control, cells were treated with Baf-A1, which resulted in the inhibition of autolysosome formation in both E6-expressing and control cells (Fig. 5A and B). Similar results were obtained when using chloroquine (CQ) to inhibit autophagosome-lysosome fusion (Fig. 5C and D). These results indicate that the autophagy pathway induced in E6-expressing cells is functionally active and can form increased numbers of autolysosomes.

FIG 5.

Autolysosome formation is induced in HPV11 E6-expressing HaCaT cells. (A) Confocal microscopy analysis of HPV11 E6-expressing and control cells transfected with a vector expressing mCherry-GFP-LC3B. Cytosolic LC3 puncta are shown in yellow (mCherry-GFP), while lysosomal LC3 puncta are shown in red (mCherry only). The V-ATPase inhibitor bafilomycin A1 (Baf-A1) was used as a control. (B) Quantification of the ratio of red puncta (autolysosomes) to all puncta (autophagosomes plus autolysosomes). The graph represents the mean ± SEM from three independent repeats, and at least 100 cells were analyzed per repeat. Significant differences were identified by Student's t test. *, P < 0.05. (C) Confocal microscopy analysis of HPV11 E6-expressing and control cells transfected with a vector expressing mCherry-GFP-LC3B using the inhibitor chloroquine (CQ) as a control. (D) Quantification of the ratio of red puncta to all puncta. The graph represents the mean ± SEM from three independent repeats, and at least 100 cells were analyzed per repeat. Significant differences were identified by Student's t test. *, P < 0.05.

HPV11 E6 induces autophagy via Erk/mTOR and AKT/mTOR signaling.

The canonical autophagy pathway is regulated by mTORC1, which phosphorylates ULK1 and other non-autophagy-related proteins, such as 4EBP1 and 70S6K. In addition, mTOR contains four characterized phosphorylation sites, but mTOR is phosphorylated mainly at Ser2448 as a part of the mTORC1 complex. Therefore, the phosphorylation state of mTOR (Ser2448), 4EBP1, and 70S6K in E6-expressing and control cells was investigated. Western blot analyses showed that the levels of phosphorylated mTOR, 4EBP1, and 70S6K were lower in E6-expressing cells (Fig. 3C, 4C, and 6A). Furthermore, reduced levels of phosphorylated ULK1 at Ser757, which is a target of mTORC1, were observed in E6-expressing cells, while the phosphorylation levels at Ser555, which is a target of AMPK, remained unaltered (Fig. 3D, 4D, and 6B). Therefore, these results are consistent with the increased autophagy activity observed in E6-expressing cells. To further explore how HPV11 E6 affects the phosphorylation status of mTOR, we investigated the upstream pathways of mTOR. Compared with those in control cells, reduced phosphorylation levels of AKT and MAPK/Erk were detected in E6-expressing cells, but no differences in AMPK and p53 were observed (Fig. 3E and F, 4 E and F, and 6C). To find other targets to help explain how E6 modulates Erk and AKT, we analyzed our microarray data and found that Erk and AKT signaling was not indicated directly in the KEGG pathway analysis. However, the KEGG pathway analysis identified the MAPK signaling pathway, which includes the Erk signaling pathway. In total, 15 genes in the MAPK pathway were downregulated with fold changes of >2.0 between the two groups. The 15 downregulated genes were as follows: NFATC1, TGFB3, PLA2G4D, BDNF, FGF22, MAPK8IP2, FGFR1, FGF10, CACNA2D3, RPS6KA2, CACNB2, CACNA1S, CACNA1A, FGFR3, and MRAS (Fig. 6D). To identify the E6 site that can distinguish the activities of AKT from those of Erk, four plasmids with different mutant sites, R78A, CC67/137GG, L111Q, and W133R, were constructed based on previous research. To determine the effect on the AKT/mTOR pathway, Erk/mTOR pathway proteins, and autophagy levels induced by the mutant E6 proteins, phosphorylated and nonphosphorylated mTOR (Ser2448), AKT, Erk, ULK1 (Ser757), and LC3B were investigated in control HaCaT, 11E6-expressing HaCaT, and four mutant 11E6 gene stable cell lines by Western blotting. The results showed that, compared with the wild-type E6 genome, the four mutants had no changes in the AKT/mTOR or Erk/mTOR pathways or LC3B (Fig. 6E and F). To further confirm the role of AKT and Erk in the E6-mediated activation of autophagy, the specific chemical activators ethyl 2-amino-6-chloro-4-(1-cyano-2-ethoxy-2-oxoethyl)-4H-chromene-3-carboxylate (SC79; AKT) and phorbol 12,13-dibutyrate (PDBu; Erk) were used. The addition of SC79 overrode the inhibiting effects of HPV11 E6 on the phosphorylation levels of AKT and partially restored the phosphorylation levels of mTOR (Ser2448) and ULK1 (Ser757). The autophagy activity marker protein LC3B was partially restored as well (Fig. 7A). Similarly, the addition of PDBu restored the phosphorylation levels of Erk and partially rescued the phosphorylation levels of mTOR (Ser2448) and ULK1 (Ser757). In addition, the autophagy activity marker protein LC3B was partially restored (Fig. 7B). Therefore, HPV11 E6 appears to be involved in the specific activation of autophagy via the MAPK/Erk and AKT pathways.

FIG 6.

HPV11 E6 induces autophagy by inhibiting the AKT/mTOR and Erk/mTOR pathways. (A) Western blot analysis of phosphorylated and nonphosphorylated mTOR (Ser2448) and its substrate proteins 4EBP1 and 70S6K in HPV11 E6-expressing and control cells. Actin was used as a loading control. The ratio of the phosphorylated and nonphosphorylated proteins was quantified. The graph represents the mean ± SEM from three independent repeats. Significant differences were identified by Student's t test. *, P < 0.05. (B) The two phosphorylation sites of ULK1, i.e., Ser757 and Ser555, were analyzed by Western blotting. Actin was used as a loading control. The ratio of phosphorylated and nonphosphorylated ULK1 was quantified. The graphs represent the mean ± SEM from three independent repeats. Significant differences were identified by Student's t test. *, P < 0.05. (C) Western blot analysis of phosphorylated and nonphosphorylated AKT, Erk, and AMPK and the total protein levels of p53. Actin was used as a loading control. The ratio of phosphorylated and nonphosphorylated AKT, Erk and AMPK and the ratio of p53 and actin were quantified. The graphs represent the mean ± SEM from three independent repeats. Significant differences were identified by Student's t test. *, P < 0.05. (D) Heat map of the expression profiles of MAPK pathway genes in HPV11 E6-expressing and control HaCaT cells. In total, 15 genes in the MAPK pathway were downregulated with fold changes of >2.0 between the two groups. Expression levels ranging from low (blue) to high (red) are indicated by color coding. (E) Western blot analysis of phosphorylated and nonphosphorylated mTOR (Ser2448), AKT, ULK1 (Ser757), and LC3B in the control HaCaT, 11E6-expressing HaCaT and four mutant 11E6 gene stable cell lines. Actin was used as a loading control. The ratio of phosphorylated and nonphosphorylated mTOR, AKT, and ULK1 was quantified. (F) The graphs represent the mean ± SEM from three independent repeats. Significant differences were identified by Student's t test. *, P < 0.05.

FIG 7.

The chemical activators SC79 and PDBu abolish the inhibitory effect of HPV11 E6 on the AKT/mTOR and Erk/mTOR pathways and autophagy level. (A) Western blot analysis of phosphorylated and nonphosphorylated mTOR (Ser2448), AKT, ULK1 (Ser757), and LC3B in HPV11 E6-expressing HaCaT and control cells stimulated with SC79. Actin was used as a loading control. The ratio of phosphorylated and nonphosphorylated mTOR, AKT, and ULK1 was quantified. The graphs represent the mean ± SEM from three independent repeats. Significant differences were identified by Student's t test. *, P < 0.05. (B) Western blot analysis of phosphorylated and nonphosphorylated mTOR (Ser2448), Erk, ULK1 (Ser757), and LC3B in HPV11 E6-expressing HaCaT and control cells stimulated with PDBu. Actin was used as a loading control. The ratio of phosphorylated and nonphosphorylated mTOR, Erk, and ULK1 was quantified. The graphs represent the mean ± SEM from three independent repeats. Significant differences were identified by Student's t test. *, P < 0.05.

DISCUSSION

During infection, HPV E6 and E7 play an important role in the pathogenic process (43, 44). Although E6 and E7 are both early proteins that are coexpressed during the course of HPV infection, the dissection of their individual contribution is important for determining their specific functions. Regarding HR-HPV E6, several of its functions and intracellular interactions leading to cellular transformation have already been determined. However, the functional characterization of the role of LR-HPV E6 and its contribution to the pathogenic process remain elusive (45). Given the distinct pathogenic processes of HR-HPVs and LR-HPVs, the function of E6 is also likely to be distinct. HR-HPV E6 has already been well established to interact with p53 and promote p53 ubiquitination and degradation via E6AP (46, 47). Although E6 in both LR-HPV and HR-HPV can bind p53 at its C-terminal binding site, this interaction does not induce p53 degradation (47). However, only HR-HPV E6 can interact with an additional binding site on p53 at its core, and this interaction is necessary for p53 degradation (48). Thus far, the functionality of the interaction between LR-HPV E6 and p53 remains unclear. To date, the only functional description of LR-HPV E6 is its ability to bind Bak and cause its degradation, which has also been observed for HR-HPV E6 (27). However, several additional HR-HPV E6-specific functionalities have already been described, including its ability to bind PDZ-containing proteins through its C-terminal PDZ binding motif (49); its activation of telomerase, which is an enzyme that adds telomere repeats to the ends of chromosomes (50); and its activation of AKT/mTORC1 signaling (51).

Similarly, research investigating the interactions between HPV and autophagy has focused primarily on HR-HPVs. For instance, autophagy was induced in HeLa cells inoculated with HR-HPV16 pseudovirions, which were subsequently captured in double-membrane-bound vesicles or autophagosomes (38). Infecting primary human foreskin keratinocytes with HPV16 virions also induced autophagy, which had an inhibitory effect on the infection, given that treatment with the autophagy inhibitor 3-MA significantly enhanced the infectivity of the HPV16 virions (39). In HPV16, E7 appears to be involved in autophagy activation since the expression of E7 in primary foreskin keratinocytes increased the expression levels of the autophagy marker protein LC3B-II, which was suggested to be related to triggering metabolic stress and which could be abolished by 3-MA (41). Importantly, these studies suggested that HPV16 or specific factors/proteins of HPV16 activate the autophagy pathway during infection in primary keratinocytes or epithelial cells; however, contradictory evidence has been shown during infection in the W12 cervical keratinocyte cell line. Autophagy was activated during infection only when the expression of early genes was depleted, suggesting that the function of early genes is involved in autophagy inhibition in W12 cells (42). Furthermore, infection in the HaCaT keratinocyte cell line showed that HPV16 pseudovirions induced rapid PI3K/AKT/mTOR pathway activation, which inhibited autophagy (40). In our study, we also used the HaCaT keratinocyte cell line, H8 cell line, and NHEK cells to ectopically express LR-HPV11 early protein E6 to elucidate its possible interactions and effects on the autophagy pathway. Our results show that HPV11 E6 activates autophagy by inhibiting AKT/mTOR and Erk/mTOR signaling. These results are in contrast to observations of HR-HPV16 pseudovirion infection in the HaCaT cell line or the ectopic expression of HPV16 E6 in primary human foreskin keratinocytes, which activated PI3K/AKT/mTOR signaling and inhibited autophagy (40, 51). However, our results are consistent with observations following the ectopic expression of HR-HPV16 E7 in primary foreskin keratinocytes (41), which similarly activated autophagy.

To investigate other targets to explain how E6 modulates Erk and AKT, a KEGG pathway analysis identified the MAPK signaling pathway, which includes the Erk signaling pathway. In total, 15 genes in the MAPK pathway were downregulated with fold changes of >2.0 between the two groups. The 15 downregulated genes are as follows: NFATC1, TGFB3, PLA2G4D, BDNF, FGF22, MAPK8IP2, FGFR1, FGF10, CACNA2D3, RPS6KA2, CACNB2, CACNA1S, CACNA1A, FGFR3, and MRAS. FGF10 and FGF22 are members of the fibroblast growth factor (FGF) family, while FGFR1 and FGFR3 are members of the fibroblast growth factor receptor (FGFR) family. FGF10 is known to activate the Erk and AKT signaling pathways (52, 53). FGF22 is an important paralog of FGF10 (54, 55), and FGFR1 and FGFR3 are involved in the activation of Erk and AKT signaling (56, 57). TGFB3 is a secreted ligand of the transforming growth factor β (TGF-β) superfamily. TGFB3 is also known to activate the Erk and AKT signaling pathways (58, 59), while BDNF has been shown to be involved in Erk and PI3K/AKT signaling (60, 61). NFATC1 is a downstream gene of Erk signaling (62, 63), and RPS6KA2 is a downstream gene of Erk signaling (64). MRAS is a member of the Ras subfamily of GTPases and activates the Erk and AKT signaling pathways (65, 66). In addition, we found that MAPK8IP2 is related to Jun N-terminal protein kinase (JNK) signaling (67). We could not find any articles discussing the relationship between PLA2G4D and Erk or AKT signaling. CACNA2D3, CACNB2, CACNA1S, and CACNA1A are proteins involved in the voltage-dependent calcium channel complex. However, we found no relationship between these genes and Erk or AKT signaling. Overall, the FGF family, especially FGF10, might modulate Erk or AKT signaling.

To determine which E6 site can distinguish the activities of AKT and Erk, we constructed four plasmids with different mutation sites and established stable cell lines. The results revealed no differences in the AKT/mTOR or Erk/mTOR pathways or autophagy level between the wild-type 11E6 and mutated 11E6 groups. The four sites were selected based on relevant studies (26, 68). First, some studies have identified several amino acids that are conserved in the different forms of E6 (amino acids W133, CC67/137GG, and L111). These three sites have already been studied in the context of high-risk E6 proteins. The W133R mutation has been shown to inhibit the ability of HPV-16 E6 to both bind and target p53 for degradation, while the CC67/137GG mutation blocks p53- and p73-dependent transactivation (69, 70). The L111Q mutation eliminates most 16E6 activities, including the binding and degradation of p53 and binding to E6AP and E6BP (71). Another site with a conserved mutation, R78A, was explored in a few previous studies, but one paper showed that R78A can interact with E6AP (26). Few studies have focused on the relationship between these mutation sites and the AKT/mTOR and Erk/mTOR pathways. Thus, in the future, we could focus on identifying new sites in the E6 gene that are related to the Erk/mTOR and AKT/mTOR pathways.

Autophagy is a genetically well-defined signaling pathway that maintains cell homeostasis by recycling misfolded proteins and defective cellular organelles (72, 73). However, prolonged autophagy activation can eventually lead to cell death by self-digestion. Our findings show that LR-HPV11 E6 expression in a keratinocyte cell line activates the autophagic process. One possibility is that the ectopic expression of HPV11 E6 may increase the energy requirements and trigger metabolic stress, which is similar to the proposed outcome following the ectopic expression of HR-HPV16 E7 (41). Moreover, E6 can increase the numbers of unfolded and misfolded proteins in host cells (74), which activates autophagy. The increased autophagy levels could maintain host cell survival and help virus proliferation and expansion. However, whether autophagy is induced in CA remains to be determined. In conclusion, this study showed that the ectopic expression of LR-HPV11 E6 in HaCaT cells induces autophagy activity as a result of decreased AKT/mTOR and Erk/mTOR signaling. These results deepen our understanding of LR-HPV interactions with host cells and the mechanisms involved in HPV pathogenicity.

MATERIALS AND METHODS

Construction of the HPV11 E6-expressing HaCaT cell line.

The gene encoding E6 from HPV11 was amplified from vector pBR322 containing the full-length genome of HPV11 (ATCC 45151D) using primers HPV11 E6-F and HPV11 E6-R (Table 1) and cloned into vector pcDNA3.1. HaCaT human keratinocyte cells were cultured in Dulbecco's modified Eagle’s medium (DMEM; Gibco), and approximately 2.5 × 10⁶ cells were used for the nuclear transfection of 2.5 μg of the pcDNA3.1 (HPV11 E6) recombinant plasmid with an Amaxa cell line Nucleofector kit V according to the manufacturer’s recommendations (75, 76) and the Nucleofector program U-020 procedures. After nuclear transfection, the cells were resuspended in preheated culture medium and transferred to a 6-well plate (Corning, USA). After a 72-h incubation, G418 (1,000 μg/ml) was added, and the cells were further incubated for 4 weeks. Cell cultures were diluted to obtain monoclonal cells, and the expression of HPV11 E6 was confirmed by reverse transcription-PCR (RT-PCR).

TABLE 1.

Primers used in this study

Primer	Sequence (5′–3′)
HPV11 E6-F	CCGATGGAAAGTAAAGATGCCTCC
HPV11 E6-R	CCGTTAGGGTAACAAGTCTTCCAT
ATG10-F	ATAAGATGCGACTGCTACAGGG
ATG10-R	CAATGCTCAGCCATGATGTGAT
β-actin-F	GGCGGCACCACCATGTACCCT
β-actin-R	AGGGGCCGGACTCGTCATACT

Construction of the HPV11 E6-expressing H8 cell line.

The gene encoding E6 from HPV11 was amplified from vector pBR322 containing the full-length genome of HPV11 (ATCC 45151D) using primers HPV11 E6-F and HPV11 E6-R (Table 1) and cloned into vector pcDNA3.1. The H8 cells were cultured in DMEM (Gibco), and approximately 1 × 10⁶ cells were used for the transfection of 2.5 μg of the pcDNA3.1 (HPV11 E6) recombinant plasmid with the Lipofectamine 3000 reagent according to the manufacturer’s recommendations. After a 72-h incubation, G418 (800 μg/ml) was added, and the cells were further incubated for 4 weeks. The cell expression of HPV11 E6 was confirmed by reverse transcription-PCR.

Construction of the HPV11 E6-expressing NHEK cells.

The gene encoding E6 from HPV11 was amplified from vector pBR322 containing the full-length genome of HPV11 (ATCC 45151D) using primers HPV11 E6-F and HPV11 E6-R (Table 1) and cloned into vector pcDNA3.1. NHEK human keratinocyte cells were cultured in RPMI 1640 medium (1640; Gibco), and approximately 7 × 10⁵ cells were used for the nuclear transfection of 2.5 μg of the pcDNA3.1 (HPV11 E6) recombinant plasmid with an Amaxa human keratinocyte Nucleofector kit according to the manufacturer’s recommendations and the Nucleofector program T-024 procedures. After nuclear transfection, the cells were resuspended in preheated culture medium and transferred to a 6-well plate (Corning, USA). After a 72-h incubation, G418 (1,000 μg/ml) was added, and the cells were further incubated for 4 weeks. The cell expression of HPV11 E6 was confirmed by reverse transcription-PCR.

Cell culture.

HaCaT (including 11E6-expressing HaCaT and control HaCaT) and H8 (including 11E6-expressing H8 and control H8) cells were cultured at 37°C in the presence of 5% CO₂ using DMEM (Gibco) containing 10% fetal bovine serum (FBS; Gibco, Australia). HPV 11E6-expressing NHEK and control NHEK cells were cultured at 37°C in the presence of 5% CO₂ using RPMI 1640 (Gibco) containing 10% FBS (Gibco, Australia). Approximately 1 × 10⁶ cells were seeded in a 6-well plate and incubated for 24 h, before the samples were collected for RNA extraction, Western blot analyses, and confocal microscopy. For the experiments involving chemical modulators, HaCaT cells (including 11E6-expressing HaCaT and control HaCaT cells) were incubated with bafilomycin A1 (Baf-A1; 25 nM, 6 h), chloroquine (CQ; 50 mM, 6 h), SC79 (10 μM, 1 h), or PDBu (1 μM, 30 min) before sampling. For the confocal microscopy and immunofluorescence staining experiments, cells were seeded on glass slides, placed in 6-well plates, and, as needed, transfected with a plasmid expressing mCherry-GFP-LC3B using the nuclear transfection method.

RNA extraction and cDNA synthesis.

Total RNA was isolated from the HPV11 E6-expressing and control cell lines using the TRIzol reagent according to the manufacturer's protocol (Invitrogen, Life Technologies, USA). For the microarray analyses, RNA was amplified and labeled using a low-input Quick Amp labeling kit (one-color; catalog number 5190-2305; Agilent Technologies, Santa Clara, CA, USA), and the labeled cDNA was purified using an RNeasy minikit (catalog number 74106; Qiagen, GmbH, Germany). For the PCR-based methods, cDNA was generated using a HiFi Script cDNA synthesis kit (CWBIO, Shanghai, China).

Whole-genome gene expression profiling by microarrays.

In total, 600 ng of Cy3-labeled cDNA from the HPV11 E6-expressing HaCaT and control cell lines was hybridized to microarray slides using a gene expression hybridization kit (catalog number 5188-5242; Agilent Technologies, Santa Clara, CA, USA) and a hybridization oven (catalog number G2545A; Agilent Technologies, Santa Clara, CA, USA). After 17 h of hybridization, the slides were washed in staining dishes (catalog number 121; Thermo Shandon, Waltham, MA, USA) using a gene expression wash buffer kit according to the manufacturer’s instructions (catalog number 5188-5327; Agilent Technologies, Santa Clara, CA, USA). The slides were subsequently scanned using an Agilent microarray scanner (dye channel, green; scan resolution, 3 μm; photoelectric multiplication tube (PMT), 100%; and 20 bit; catalog number G2565CA; Agilent Technologies, Santa Clara, CA, USA). The data were extracted using Feature Extraction software (version 10.7; Agilent Technologies, Santa Clara, CA, USA), and the raw data were normalized using a quantile algorithm from the limma package in the R program (44).

Semiquantitative RT-PCR.

Reverse transcription-PCR (RT-PCR) was performed using a 50-μl mixture containing Premix Taq (CWBIO, Shanghai, China), 5 μl of cDNA, and 2 μl of HPV11 E6 or β-actin PCR primers (Table 1). PCR was carried out as follows: for HPV11 E6, 1 cycle of 94°C for 5 min, followed by 30 cycles of 94°C for 30 s, 60°C for 30 s, and 72°C for 1 min and a final cycle at 72°C for 7 min; for β-actin, 1 cycle of 94°C for 5 min, followed by 30 cycles of 94°C for 30 s, 55°C for 30 s, and 72°C for 1 min and a final cycle of 72°C for 7 min. The amplified products were visualized on a 1% agarose gel.

qRT-PCR.

The relative ATG10 gene expression in the HPV11 E6-expressing and control HaCaT cell lines was measured by quantitative real-time PCR (qRT-PCR), and the relative expression of HPV11 E6 was analyzed using real-time PCR after the HPV11 E6-expressing HaCaT stable cell line was transfected with shRNA-1, shRNA-2, and shRNA-3 (Table 2). Reaction systems with a total volume of 10 μl including 1 μl of cDNA, 5 μl of 2× SYBR Premix Ex Taq I (TaKaRa, Japan), 0.4 μl of forward primer, 0.4 μl of reverse primer, 0.2 μl of carboxy-X-rhodamine, and 3 μl of double-distilled H₂O were run on an ABI 7500 system (Applied Biosystems, USA). The reactions were carried out in 96-well plates at 95°C for 2 min, followed by 40 cycles of 95°C for 10 s and 58°C for 30 s. The relative ATG10 mRNA levels and the relative expression of HPV11 E6 were normalized to those of β-actin.

TABLE 2.

shRNA sequences targeting the HPV11 E6 gene

Name	Sequence
shRNA-1	CCAGTTGTGCAAGACGTTT
shRNA-2	CCTACAGTAGAAGAAGAAA
shRNA-3	GCGAGACAACTTTCCCTTT

Immunofluorescence assays and confocal laser scanning microscopy.

Cells were washed with phosphate-buffered solution (PBS) and fixed with 4% paraformaldehyde (Bio-Rad) for 15 min. For the immunofluorescence assays, the cells were permeabilized with 0.1% Triton X-100 (Solarbio, China) for 10 min. Then, the fixed permeabilized cells were blocked with 10% goat serum in 1% bovine serum albumin for 60 min at room temperature and incubated overnight at 4°C with polyclonal rabbit anti-LC3B IgG (1:100 dilution; Sigma-Aldrich, MO, USA). The cells were subsequently incubated with fluorescein isothiocyanate-conjugated Alexa Fluor 555 donkey anti-rabbit IgG (1:500 dilution; Beyotime Biotechnology) for 2 h in the dark. Finally, the cell nuclei were stained with 4,6-diamidino-2-phenylindole (DAPI) (1:1,000 dilution; Beyotime Biotechnology) for 10 min at room temperature, and images were obtained under a Zeiss Axiovert confocal microscope (Zeiss).

Western blot analysis.

Cells from the HPV11 E6-expressing group and the control group were collected and lysed with radioimmunoprecipitation assay lysis buffer (1% phenylmethylsulfonyl fluoride; Fdbio, China) for 30 min. After centrifugation for 20 min (10,000 × g), the supernatants were collected, and the protein concentrations were determined with a bicinchoninic acid kit (Beyotime, China). The protein samples were run on a 12% polyacrylamide SDS-PAGE gel and transferred onto polyvinylidene fluoride (PVDF) membranes (0.2 μm PVDF; Bio-Rad, USA) using a wet transblotting apparatus (Bio-Rad, USA). After blocking with 10% nonfat milk (BD Difco, USA) for 1 h, the membranes were incubated overnight with the following primary antibodies: monoclonal rabbit anti-phospho-mTOR (Ser2448; Cell Signaling Technology, Danvers, MA, USA), monoclonal rabbit anti-phospho-mTOR (Cell Signaling Technology, Danvers, MA, USA), anti-phospho-70S6 kinase (Cell Signaling Technology, Danvers, MA, USA), monoclonal rabbit anti-70S6 kinase (Cell Signaling Technology, Danvers, MA, USA), monoclonal rabbit anti-phospho-4E-BP1 (Thr37/46; Cell Signaling Technology, Danvers, MA, USA), monoclonal rabbit anti-4E-BP1 (Cell Signaling Technology, Danvers, MA, USA), monoclonal rabbit anti-Erk (Cell Signaling Technology, Danvers, MA, USA), monoclonal rabbit anti-phospho-Erk (Thr202/Tyr204; Cell Signaling Technology, Danvers, MA, USA), monoclonal rabbit anti-AKT (Cell Signaling Technology, Danvers, MA, USA), monoclonal rabbit anti-phospho-AKT (Ser473; Cell Signaling Technology, Danvers, MA, USA), monoclonal rabbit anti-ULK1 (Cell Signaling Technology, Danvers, MA, USA), monoclonal rabbit anti-phospho-ULK1 (Ser555; Cell Signaling Technology, Danvers, MA, USA), monoclonal rabbit anti-phospho-ULK1 (Ser757; Cell Signaling Technology, Danvers, MA, USA), monoclonal rabbit anti-AMPKα (Cell Signaling Technology, Danvers, MA, USA), monoclonal rabbit anti-phospho-AMPKα (Ser172; Cell Signaling Technology, Danvers, MA, USA), monoclonal rabbit anti-p53 (Cell Signaling Technology, Danvers, MA, USA), monoclonal rabbit anti-ATG12 (Cell Signaling Technology, Danvers, MA, USA), polyclonal rabbit anti-Beclin-1 (Cell Signaling Technology, Danvers, MA, USA), polyclonal rabbit anti-ATG10 (Novus Biological, USA), polyclonal rabbit anti-LC3B (Sigma-Aldrich, MO, USA), monoclonal mouse anti-β-actin (Sigma-Aldrich, MO, USA), and monoclonal rabbit anti-p62 (Abcam, Cambridge, UK). After the membranes were washed three times for 10 min each time with PBS, they were incubated for 2 h at room temperature with horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG (H+L; Beyotime, China) or HRP-conjugated goat anti-mouse IgG (H+L, Beyotime, China). The specific protein bands were detected with enhanced chemiluminescence (ECL) reagents (Bio-Rad) using a blot scanner.