Rare human papillomavirus 16 E6 variants reveal significant oncogenic potential

Abstract

The aim of this study was to determine whether low prevalence human papillomavirus (HPV) 16 E6 variants differ from high prevalence types in their functional abilities. We evaluated functions relevant to carcinogenesis for the rarely-detected European variants R8Q, R10G and R48W as compared to the commonly detected L83V. Human immortalized keratinocytes (NIKS) stably transduced with the E6 variants were used in most functional assays. Low and high prevalence E6 variants displayed similar abilities in abrogation of growth arrest and inhibition of p53 elevation induced by actinomycin D. Differences were detected in the abilities to dysregulate stratification and differentiation of NIKS in organotypic raft cultures, modulate detachment induced apoptosis (anoikis) and hyperactivate Wnt signaling. No distinctive phenotype could be assigned to include all rare variants. Like L83V, raft cultures derived from variants R10G and R48W similarly induced hyperplasia and aberrantly expressed keratin 5 in the suprabasal compartment with significantly lower expression of keratin 10. Unlike L83V, both variants, and particularly R48W, induced increased levels of anoikis upon suspension in semisolid medium. R8Q induced a unique phenotype characterized by thin organotypic raft cultures, low expression of keratin 10, and high expression of keratins 5 and 14 throughout all raft layers. Interestingly, in a reporter based assay R8Q exhibited a higher ability to augment TCF/β-catenin transcription. The data suggests that differences in E6 variant prevalence in cervical carcinoma may not be related to the carcinogenic potential of the E6 protein.

Figure 1

Gene expression of HPV16 E6 variants in NIKS as defined by reverse transcriptase polymerase chain reaction (RT-PCR), Western blot and immunofluorescence. RT-PCR detected three bands representing mRNAs of the full length and the splice variants E6I and E6II (A). With the E6 antibody used (clone 6F4) and input of 120 μg of total protein, a band of the expected size was detected in extracts of NIKS transduced with all E6 proteins except R8Q. Loading of 200 μg total protein detected a faint band (data not shown). Western with the anti-HA antibody detected all the E6 proteins (B). The micrograph shows immunofluorescence against the HA-tag epitope. E6-specific staining is present in all transduced NIKS lines (C).

Figure 2

Cell cycle profile of HPV16 E6 variants. The distribution of cells in G1, S and G2/M phases, obtained by flow cytometry, is demonstrated in A (histograms from a representative experiment) and B (mean values bars). Data with and without actinomycin D (AD) treatment (0.5 nM for 24 h) are shown. Differences in G1/S ratio are depicted in C. Mean values of at least three independent experiments are presented as average (mean) ± standard deviation (SD). Significance was tested by a one-way ANOVA test. A P value of < 0.05 was considered significant.

Figure 3

p53 level in HPV16 E6 variants NIKS before and after treatment with actinomycin D. (A) ELISA showing the remaining units of p53 quantified according to the number of cells after AD treatment (0.5 nM for 24 h). Mean values of at least three independent experiments are presented as average (mean) values ± standard deviation (SD). Significance was tested by a one-way ANOVA test. A P value of < 0.05 was considered significant. (B) The corresponding Western blot showed only p53-specific bands for the empty vector NIKS. After AD treatment the amount of p53 was increased approximately threefold as seen with ELISA.

Figure 4

Keratinocyte differentiation induced in 3D raft cultures. (A) The panel depicts 3D raft cultures obtained from NIKS transduced with the indicated HPV16 E6 variants and vector. Micrographs include immunoftuorescence for the basal cell marker K5 (green), suprabasal cell marker K10 (red), the overlay of both and H+E staining. Rafts had been grown three times and sections were cut twice from each set of paraffin blocks to ensure reproducibility. (B) Calculation of K5 and K10 levels are shown for each HPV16 genotype and the empty vector raft as described in Materials and Methods. Values of area intensities for K5 and K10 in the basal and suprabasal regions from 4 sections are presented as mean ± standard deviation (SD). (C) Raft cultures were produced as described in (A). Micrographs showing immunoftuorescence of the basal cell marker K14 (Red) and nuclear staining of the same section (DAPI; blue) are shown. (D) Calculation of K14 levels in the basal and suprabasal regions were carried out as described in (B). Significance was tested by one way ANOVA test. A P value of < 0.05 was considered significant.

Figure 5

Modulation of apoptosis by HPV16 E6 variants. Flow cytometry was performed after a 16 h treatment in semisolid medium and apoptosis-alive (Annexin V-FITC+/propidium iodide-), apoptosis-dead (Annexin V-FITC+/propidium iodide+), intact-dead (Annexin V-FITC-/propidium iodide+) and intact-alive (Annexin V-FITC-/propidium iodide-) cells were calculated for each quadrant. (A) dot plots from a representative experiment. (B) Mean values of at least three independent experiments are presented as average (mean) values ± standard deviation (SD). Significance was tested by a one-way ANOVA test. A P value of < 0.05 was considered significant.

Figure 6

Hyperactivation of β-catenin/TCF-dependent transcription induced by Wnt-3a and HFz1 in the context of E6 variants. (A) HEK293T cells were co-transfected with TCF/LUC reporter (1 μg), β-Gal (0.1 μg), Hfz1 (0.3 μg), Wnt-3a (0.1 μg) and E6 variants as indicated. Each E6/variant was transfected at 2 amounts, 0.25, 0.5 μg plasmid DNA. (B) C33A cells were transfected as in indicated in A except that 0.5 Renilla luciferase DNA was transfected instead of β-Gal. Luciferase activities were measured 48 h following transfection. The histograms show fold activation relative to the control cells transfected with the vector DNA (pJS55). The average values of fold activation ± SD of the indicated plasmids are shown. Data are from at least 3 independent experiments with HEK293T and two experiments with C33A cells. Significance was tested by a one-way ANOVA test. A P value of < 0.05 was considered significant.