The HPV16 oncogenes cause aberrant stem cell mobilization

Abstract

Human Papilloma Virus related epithelial cancers have been speculated to derive from virus-infected tissue stem cells. Stem cells also are thought to provide a reservoir of latently infected cells that can persist for long periods. In this study we have examined the effects of HPV16 E6 and E7 oncogenes on multipotent epithelial stem cells, using in vivo systems. Our results show that expression of HPV16 oncogenes reduces the number of bulge label-retaining cells within hair follicles at telogen suggesting aberrant mobilization, a result supported by increased mobilization upon acute anagen induction. Importantly the loss of relative quiescence, a hallmark feature of stem cells, occurs in the absence of a reduction in other stem cell markers. This points to an atypical stem cell compartment in the context of E6 and E7 expression. We hypothesize that this aberrant compartment may have important roles in the viral life cycle and/or ensuing carcinogenesis.

Keywords: E6; E7; HPV; Stem cells.

Fig. 1. Expression of the HPV16 oncogenes leads to reduced detection of LRCs in hair follicle bulge at telogen

(A) LRCs were labeled using a BrdU pulse administered shortly after birth and chased until second telogen. ~50 Hair follicles were selected from at least 3 mice of each genotype, NTG, E6, E7 and E6E7 mice. The mean number of BrdUrd positive cells per hair follicle bulge was quantified and plotted for each genotype (columns); bars, SD. All statistical comparisons were performed using a two-sided Wilcoxon rank sum test. (B) Representative immunofluorescent figures of hair follicles showing BrdU positive cells (red–white arrow). Counterstaining was done with DAPI (blue).

Fig. 2. Expression of the HPV16 oncogenes leads to more rapid mobilization of LRCs in response to acute anagen induction

(A) In order to induce anagen in mice where LRCs were labeled, TPA was applied on mice every 48 h for four times. H&E staining was performed on all tail hair follicles. Hair follicle length was quantified to verify effective anagen induction. (B) ~70 Hair follicles were selected from at least 3 mice of each genotype, NTG, E6, E7 and E6E7 mice. The mean hair follicle length in μm was measured and plotted for each genotype (columns); bars, SD. All statistical comparisons were performed using a two-sided Wilcoxon rank sum test. Statistical significance was also observed between NTG and transgenic mice under resting (no TPA) conditions. (C) In order to track the mobilization of LRCs in response to acute anagen induction, the percentage reduction of LRCs was tracked per genotype. At least 3 mice of each genotype at anagen (TPA) and telogen (no TPA), NTG, E6, E7 and E6E7 mice, were selected and hair follicle bulge regions were quantified. The relative reduction of BrdUrd positive cells per hair follicle bulge was plotted for each genotype (columns); bars, SD. All statistical comparisons were performed using a two-sided Wilcoxon rank sum test. Statistical significance was also observed between E6 and E6E7 mice.

Fig. 3. Increased mobilization of LRCs correlates with increased hair growth in mice expressing the HPV16 E6 and E7 oncogenes

(A) Backs of telogen mice of all genotypes were shaved and pictures were taken at days 0 and 8 after shaving. (B) ~50 Hair follicles were selected from at least 3 mice of each genotype, NTG, E6, E7 and E6E7 mice. The mean number of PCNA positive cells at the base of each hair follicle was quantified and plotted for each genotype (columns); bars, SD. All statistical comparisons were performed using a two-sided Wilcoxon rank sum test. (C) Representative immunofluorescent figures of hair follicles showing PCNA positive cells (red). Counterstaining was done with DAPI (blue).

Fig. 4. Other markers of bulge stem cells are not reduced in response to HPV16 oncogene expression

(A) Immunofluorescence was performed using a K15-specific antibody. ~50 Hair follicles were selected from at least 3 mice of each genotype, NTG, E6, E7 and E6E7 mice. The mean number of K15 positive cells of each hair follicle was quantified and plotted for each genotype (columns); bars, SD. All statistical comparisons were performed using a two-sided Wilcoxon rank sum test. Statistical significance was also observed between the NTG and the transgenic mice (no TPA) as well as between the various transgenics (TPA-treated). (B) Representative immunofluorescence of K15 staining (red) in the hair follicles of the genotypes examined. Counterstaining was done with DAPI (blue).

Fig. 5. The quiescence of the bulge stem cell population is affected upstream as well as downstream of the Nfatc1 pathway by HPV16 oncogene expression

(A–D) ~50 Hair follicles were selected from at least 3 mice of each genotype, NTG, E6, E7 and E6E7 mice. The mean number of Nfatc1 (A,B) and Cdk4 (C,D) positive cells of each hair follicle in telogen (no TPA) (A,C) and anagen (TPA) (B,D) conditions was quantified and plotted for each genotype (columns); bars, SD. All statistical comparisons were performed using a two-sided Wilcoxon rank sum test. (E,F) Representative immunofluorescense staining of (E) nuclear Nfatc1 and (F) Cdk4 positive cells is depicted by arrows. Counterstaining was done with DAPI (blue).