hAda3 degradation by papillomavirus type 16 E6 correlates with abrogation of the p14ARF-p53 pathway and efficient immortalization of human mammary epithelial cells

Abstract

Two activities of human papillomavirus type 16 E6 (HPV16 E6) are proposed to contribute to the efficient immortalization of human epithelial cells: the degradation of p53 protein and the induction of telomerase. However, the requirement for p53 inactivation has been debated. Another E6 target is the hAda3 protein, a p53 coactivator and a component of histone acetyltransferase complexes. We have previously described the role of hAda3 and p53 acetylation in p14ARF-induced human mammary epithelial cell (MEC) senescence (P. Sekaric, V. A. Shamanin, J. Luo, and E. J. Androphy, Oncogene 26:6261-6268, 2007). In this study, we analyzed a set of HPV16 E6 mutants for the ability to induce hAda3 degradation. E6 mutants that degrade hAda3 but not p53 could abrogate p14ARF-induced growth arrest despite the presence of normal levels of p53 and efficiently immortalized MECs. However, two E6 mutants that previously were reported to immortalize MECs with low efficiency were found to be defective for both p53 and hAda3 degradation. We found that these immortal MECs select for reduced p53 protein levels through a proteasome-dependent mechanism. The findings strongly imply that the inactivation of the p14ARF-p53 pathway, either by the E6-mediated degradation of p53 or hAda3 or by cellular adaptation, is required for MEC immortalization.

FIG. 1.

E6^Y54D is defective in p53 degradation and inhibits p53 acetylation, stabilization, and growth arrest induced by p14ARF. (A) hTERT MECs, E6^Y54D MECs, and hTERT MECs expressing E6^Y54D were infected with pWZL-hygro-p14ARF or control pWZL-hygro retrovirus, selected with hygromycin, and harvested at day 7 postinfection. Cell lysates corresponding to 30 μg protein were analyzed by Western blotting. Actin was used as a loading control. Note that E6^Y54D inhibits p53 stabilization induced by p14ARF. (B) U2OS cells were transfected with increasing amounts of pcDNA3-p14ARF and 1 μg pLXSN-16E6^Y54D or vector control. Cells were harvested 48 h posttransfection. p14ARF, p53, and p21cip1 proteins were detected by Western blotting, and E6 mRNA was detected by RT-PCR. Actin and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were used as loading controls. In U2OS cells, p14ARF induced the dose-dependent stabilization of p53 that was inhibited by E6^Y54D. (C) U2OS cells were transfected with 0.2 μg pcDNA3-p14ARF and 0.01, 0.05, 0.2, or 1.0 μg pLXSN-16E6^Y54D. Cells were harvested 48 h posttransfection. The expression of p14ARF, total p53, p21cip1, and E6^Y54D was detected as described for panel B. E6^Y54D inhibited p14ARF-induced p53 stabilization and the induction of p21cip1. (D) Cell lysates from panel B were normalized for total p53, and the acetylation of lysine 382 in p53 was detected by Western blotting (lanes 1 and 3, 28 μg protein; lane 2, 1.8 μg protein; lane 4, 20 μg protein). The amount of HPV16 E6^Y54D or empty vector (V) in the transfection was 1 μg. (E) U2OS cells were transfected with 0.5 μg pEGFPF+, 0.2 μg pcDNA3-p14ARF, and 1 μg pLXSN-E6^Y54D or empty vectors. Monolayer cells were harvested 48 h posttransfection and stained with propidium iodide, and the DNA content of EGFP-positive cells was analyzed by FACS.

FIG. 2.

hAda3 and p53 degradation by HPV16 E6 (16E6) mutants in vivo. H1299 cells were transfected with pcDNA3-Flag-hAda3 (1 μg), pCMV-p53 (1 μg), and pLXSN-HPV16 E6 DNA (3 μg). Cells were harvested 48 h posttransfection. The levels of p53 and Flag-hAda3 proteins were analyzed by Western blotting, and E6 mRNA was analyzed by RT-PCR. Actin and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were used as loading controls. C, pLXSN vector control.

FIG. 3.

Inhibition of p14ARF-induced p53 activation and growth arrest by HPV16 E6 (16E6) mutants that are defective in p53 degradation. (A) U2OS cells were transfected with 0.2 μg p14ARF and 1 μg wild-type or mutant HPV16 E6 or empty vector DNA (V). The expression of p14ARF, p53, and p21cip1 proteins and HPV16 E6 mRNA was detected as described in the legend to Fig. 1B. Asterisks indicate HPV16 E6 mutants defective in both p53 and hAda3 degradation. The graph shows the levels of p53 and p21cip1 quantified using an LAS1000+ luminescent image analyzer (Fuji) and normalized to actin. Levels in control cells transfected with p14ARF (lane 2) were set to 1. Note that E6 and E6^F2V completely blocked the p14ARF-induced accumulation of p53 and p21cip1. (B) Cell lysates from the experiments shown in panel A were normalized for total p53, and the acetylation of lysine 382 in p53 was detected as described in the legend to Fig. 1D. The graph shows the levels of acetylated p53 that were quantified and normalized to total p53 as described for panel A. (C) U2OS cells were transfected with GFP-, p14ARF-, and HPV16 E6-expressing plasmids or control vectors (V). The DNA content of propidium iodide-stained cells was determined by FACS as described in the legend to Fig. 1E. GAPDH, glyceraldehyde-3-phosphate dehydrogenase; Ac-K382-p53, K382-acetylated p53.

FIG. 4.

Low-risk genital HPV E6s do not induce the degradation of hAda3 and do not inhibit the p14ARF-induced accumulation of p53 and p21cip1. (A) H1299 cells were transfected with Flag-tagged hAda3, p53, and HPV6b, HPV6vc, HPV11, or HPV16, as described in the legend to Fig. 2. The expression of Flag-hADA3 and p53 was tested by Western blotting. Note that, in contrast to HPV16 E6, the HPV6 and HPV11 E6 do not induce hAda3 degradation. (B) U2OS cells were transfected with p14ARF, HPV6b E6, HPV6vc E6, HPV16 E6^F2V, or empty vector (V), as described in the legend to Fig. 3A. The expression of p14ARF, p53, and p21cip1 proteins was detected by Western blotting, and E6 mRNA was detected by RT-PCR. Note that HPV6 E6 and HPV11 E6 do not inhibit the accumulation of p53 and p21cip1 proteins induced by p14ARF. HPV16 E6^F2V was used for comparison. Actin was used as a loading control.

FIG. 5.

MECs immortalized by hAda3 and p53 degradation-defective E6 mutants maintain the expression of p14ARF but select for the proteosomal degradation of the p53 protein. (A) The expression of p14ARF, hTERT, and HPV16 E6 mRNA was tested by RT-PCR in parental 76N primary MECs, cells infected with pLXSN vector, or late-passage HPV16 E6-immortalized MECs (lanes 3 to 7). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the loading control. hTERT-immortalized MECs (lane 8) were used for comparison. MCF7 and SAOS2 cell lines were used as PCR standards for hTERT and p14ARF, respectively. MCF7 is a p14ARF-negative, telomerase-positive breast cancer cell line; SAOS2 is a p14ARF-positive, telomerase-negative osteosarcoma cell line. Note that the hTERT-immortalized MECs down-regulate p14ARF compared to the regulation of p14ARF by parental primary 76N MECs. In contrast, 76N MECs immortalized by wild-type and mutant HPV16 E6 activate hTERT expression and maintain p14ARF. (B) Cell lysates corresponding to 30 μg protein were tested by Western blotting for p53 and HDM2 proteins. Actin was used as a loading control. In comparison to E6^F2V- or E6^Y54D-immortalized MECs, the E6^L37S and E6^L110Q-immortalized cells exhibit reduced p53 protein levels. Parental primary 76N MECs and wild-type HPV16 E6-immortalized MECs were included for comparison. Note that E6^L37S- and E6^L110Q-immortalized MECs did not overexpress HDM2. (C) Western blotting of E6^L37S- and E6^L110Q-immortalized MECs treated for 4 h with DMSO (lanes 1 and 3) or 10 μM proteasome inhibitor MG132 (lanes 2 and 4). Actin was used as a loading control. 16E6, HPV16 E6.