Degradation of p53 by human Alphapapillomavirus E6 proteins shows a stronger correlation with phylogeny than oncogenicity

Abstract

Background: Human Papillomavirus (HPV) E6 induced p53 degradation is thought to be an essential activity by which high-risk human Alphapapillomaviruses (alpha-HPVs) contribute to cervical cancer development. However, most of our understanding is derived from the comparison of HPV16 and HPV11. These two viruses are relatively distinct viruses, making the extrapolation of these results difficult. In the present study, we expand the tested strains (types) to include members of all known HPV species groups within the Alphapapillomavirus genus.

Principal findings: We report the biochemical activity of E6 proteins from 27 HPV types representing all alpha-HPV species groups to degrade p53 in human cells. Expression of E6 from all HPV types epidemiologically classified as group 1 carcinogens significantly reduced p53 levels. However, several types not associated with cancer (e.g., HPV53, HPV70 and HPV71) were equally active in degrading p53. HPV types within species groups alpha 5, 6, 7, 9 and 11 share a most recent common ancestor (MRCA) and all contain E6 ORFs that degrade p53. A unique exception, HPV71 E6 ORF that degraded p53 was outside this clade and is one of the most prevalent HPV types infecting the cervix in a population-based study of 10,000 women. Alignment of E6 ORFs identified an amino acid site that was highly correlated with the biochemical ability to degrade p53. Alteration of this amino acid in HPV71 E6 abrogated its ability to degrade p53, while alteration of this site in HPV71-related HPV90 and HPV106 E6s enhanced their capacity to degrade p53.

Conclusions: These data suggest that the alpha-HPV E6 proteins' ability to degrade p53 is an evolved phenotype inherited from a most recent common ancestor of the high-risk species that does not always segregate with carcinogenicity. In addition, we identified an amino-acid residue strongly correlated with viral p53 degrading potential.

Figure 1. HPV E6-mediated p53 degradation.

A) HA-tagged p53 levels were visualized by Western blot after co-transfection with HA-tagged HPV E6 from HPV11, HPV16 and HPV18 into C-33A cells. Lanes 1, 3, 5 and 7 show results from co-transfection of HA-p53 and vectors indicated at the top of the figure. Lanes 2, 4, 6 and 8 show the p53 levels after treatment with MG132, as indicated at the top of the figure. β-tubulin was visualized as a loading control. (B) Half life of HA-p53 in transfected 293T cells. 293T cells were transfected with HA-p53 and control (pQCXIN) or E6 ORFs (HPV11, HPV18, HPV53, HPV56 and HPV66). The band intensities were determined from the scanned Western blot using ImageQuant and the signals at time 0 were defined as 100%. The band intensities of the indicated time points were normalized to time 0.

Figure 2. Phylogeny of alpha-HPVs and degradation of p53 by HPV E6 ORFs.

HPV E6 activity on HA-p53 steady state levels were determined by Western blot using the assay shown in Figure 1A. The phylogenetic tree at the left was constructed using the combined early gene sequences as previously reported . Epidemiological carcinogenicity was extrapolated from recent reviews , and is indicated in the column labeled “carcinogenic risk”: ++, highly oncogenic; +, oncogenic; +/−, probably oncogenic; −, not significantly associated with cervix cancer; NA, insufficient data. The p53 levels after co-transfection with E6 from each type indicated on the left are shown in the far right column labeled, “p53” (with and without MG132) and the results are summarized in the column labeled“↓p53”. Endogenous β-tubulin (far right column) represents a loading control. The alpha-HPV species groups are indicated by brackets with a number to the right. The empty vector control, pQCXIN is shown at the bottom.

Figure 3. Alignment of HPV alpha E6 ORFs.

The alignment of all 27 E6 ORFs tested in Figure 2 is shown. HPV types degrading p53 are shown at the top of the alignment and those not degrading p53 are shown at the bottom. The shaded region represents a proposed E6-AP binding domain , . The amino acid sequences of the E6 ORFs were aligned using Clustal X (version 1.81) . The amino acid at position 31 (arrow, bold) was associated with p53 degradation (p<0.01). “_” indicate gaps, whereas “.” indicates identical residues with the HPV16 E6 amino acid sequence shown in the top row.

Figure 4. HPV E6 mutagenesis and degradation of p53.

(A) The steady state levels of HA-p53 co-transfected with wild type or mutant HPV71, 90 and 106 HA-E6 in C-33A cells are shown in the row labeled HA-p53. The unmodified E6 ORFs are indicated with “WT” and the mutated E6 ORFs are indicated with the replacement amino acid at the top of the figure. β-tubulin is shown as a loading control in the bottom row. (B) Representative images of double immuno-fluorescence experiments. HA-tagged HPV E6 constructs are green, whereas the endogenous p53 is labeled red. In the p53 image, arrows indicate the location of the E6 expressing cells. Absence of p53 signal (red) is the result of degradation. In the merged images, yellow signifies a non-degrader, whereas green only (HPV E6) is indicative of a p53 degrader. Cell nuclei are detected with DAPI and are shown in the bottom row. Each HPV E6 construct is indicated at the bottom of each column.