Direct HPV E6/Myc interactions induce histone modifications, Pol II phosphorylation, and hTERT promoter activation

Abstract

Human Papillomavirus Viruses (HPVs) are associated with the majority of human cervical and anal cancers and 10-30% of head and neck squamous carcinomas. E6 oncoprotein from high risk HPVs interacts with the p53 tumor suppressor protein to facilitate its degradation and increases telomerase activity for extending the life span of host cells. We published previously that the Myc cellular transcription factor associates with the high-risk HPV E6 protein in vivo and participates in the transactivation of the hTERT promoter. In the present study, we further analyzed the role of E6 and the Myc-Max-Mad network in regulating the hTERT promoter. We confirmed that E6 and Myc interact independently and that Max can also form a complex with E6. However, the E6/Max complex is observed only in the presence of Myc, suggesting that E6 associates with Myc/Max dimers. Consistent with the hypothesis that Myc is required for E6 induction of the hTERT promoter, Myc antagonists (Mad or Mnt) significantly blocked E6-mediated transactivation of the hTERT promoter. Analysis of Myc mutants demonstrated that both the transactivation domain and HLH domain of Myc protein were required for binding E6 and for the consequent transactivation of the hTERT promoter, by either Myc or E6. We also showed that E6 increased phosphorylation of Pol II on the hTERT promoter and induced epigenetic histone modifications of the hTERT promoter. More important, knockdown of Myc expression dramatically decreased engagement of acetyl-histones and Pol II at the hTERT promoter in E6-expressing cells. Thus, E6/Myc interaction triggers the transactivation of the hTERT promoter by modulating both histone modifications, Pol II phosphorylation and promoter engagement, suggesting a novel mechanism for telomerase activation and a new target for HPV- associated human cancer.

Keywords: RNA polymerase II; histones; oncoproteins; papillomaviruses; telomerase.

Figure 1. HPV E6 associates with Myc in vivo and in vitro

(A) Diagram of Myc structure and schematic representation of the deletion constructs of Myc. MBI: Myc box I; MBII: Myc box II; MBIII: Myc box III; NLS: nuclear localization signal; b: basic; HLH: helix-loop-helix domain; LZ: leucine zipper. (B) HPV E6 interacts with Myc in vivo. Extracts of 293T cells transfected with pcDNA-Myc and/or pCMV3xFlag-E6 as indicated with “+” and “-”, respectively, were immunoprecipitated with the Flag antibody (central) or Myc antibody (N-262, Santa Cruz) and blotted with Myc (N-262, up) and Flag (M2, down) antibodies. Lysates were also analyzed with straight WB, as shown in left. (C) HPV E6s interact with Myc in vitro. The wt Myc was made by IVT, and the same amount of the IVT proteins were subjected to GST alone, GST-16E6 or GST-6BE6 pulldown assays. The captured products were separated on 4-20% SDS-PAGE and blotted with monoclonal anti-Myc antibody (9E10). The lower panel indicates the equal amount of GST-E6 products. The low risk HPV E6 (6BE6) has a relative lower affinity to associate with Myc compared to high risk E6 (16E6).

Figure 2. E6 increases Myc phosphorylation without induction of Myc expression

(A) Levels of total, soluble and unsoluble Myc proteins in keratinocytes expressing either LXSN or E6. RIPA lysates from keratinocytes expressing LXSN or E6 were centrifuged at 14,000 rpm at 4°C. RIPA lysates (total, indicated as T), supernatants (soluble, indicated as S1), pellets (unsoluble, chromatin-associated, indicated as S2) were blotted with 9E10 monoclonal Myc antibody. NS indicates non-specific band. E6 does not alter levels of total, soluble, or unsoluble Myc in keratinocytes. (B) E6 induces Myc phosphorylation. Western blot analysis for endogenous phosphorylated c-Myc proteins in pLXSN- and E6-transduced HFK cells. Immunoblot analysis was performed on whole cell lysates using Thr58/Ser62 phospho-c-Myc-specific antibody. The protein blot was then stripped and reprobed for total c-Myc to demonstrate equal expression and loading.

Figure 3. Knockdown of either Myc or Max decreases telomerase activity in E6E7 immortalized cells or tumor-derived cell lines

(A) Dissociation of Myc/Max complex blocks dramatically telomerase in E6/E7 immortalized cells. The HFKs expressing E6/E7 were treated with 4 nM of 10058-F4 for 16 hrs. Cell lysates were subjected to quantitative real time TRAP assay (QRT-TRAP) and a significant decrease in telomerase activity after 10058-F4 treatment was observed compared to that in control cells. (B) siRNAs decrease expression of Myc and Max proteins. Cells were transfected with either control siRNA (24 and 48 hours), or Myc specific (24 hours), or Max specific (48 hours) siRNA duplexes in 6-well plates and cell lysates were used for Western blot. Beta actin was blotted as internal controls. (C) siRNAs against Myc or Max decrease telomerase activity. The above same siRNA treated cells were lysated with TRAP buffer and QRT- TRAP assay was performed as described in Materials and Methods. Decreased telomerase activities in HFKs expressing E6/E7 were observed after treatment with either Myc or Max specific siRNA duplex. (D) Myc is required for E6-induced telomerase activity in HFKs. Myc specific siRNA duplex sufficiently knockdowns Myc expression as shown in lower panel with RT-PCR and significantly decreases E6-induced telomerase activity (upper panel) with regular TRAP assays.

Figure 4. Interference with the balance of Myc/Max network abolishes E6 induction of hTERT promoter

Primary HFKs were transfected with the hTERT core promoter and either wt Myc or its mutants, or together with E6. The pRL-CMV Renilla reniformis reporter plasmid was also transfected into the cells to standardize for transfection efficiency. Luciferase activity was measured 24 hours after transfection using the Dual luciferase reporter assay system (Promega). Relative fold activation reflects the normalized luciferase activity induced by Myc or its mutants with or without E6 compared to the normalized activity of vector control. The value of pGL3B-empty vector activity was set to 1. Error bars show the standard deviation from at least three independent experiments. (A) Myc antagonists, Mad and Mnt, block E6 induced hTERT promoter. (B) Myc partner, Max, reduces E6-mediated hTERT promoter activity. (C) C-terminal of Myc is required for E6-mediated hTERT promoter activity. Myc mutants missing C-terminal, which is responsible for dimerization with Max and binding to DNA, inhibit E6 induction of hTERT promoter. (D) Myc transactiviation domain (MBII, not MBI) is critical for E6-mediated hTERT promoter activity. Myc mutant missing MBII, not MBI, decreases E6 induction of the hTERT promoter.

Figure 5. HPV16 E6 induces chromatin changes and phosphorylation of the RNA Polymerase II (Pol II) CTD at the hTERT promoter

(A) E6 induces chromatin changes at the hTERT promoter. ChIP lysates from HFK-LXSN and HFK-E6 cells were precipitated with antibodies against Ac-H3, Ac-H4, Tri-Me K4-H3, Di-Me-K9 H3, or Ac-K9 H3, and mouse and rabbit IgG mixture were also included as a negative control. E6 significantly increases histone acetylation (Ac-H3, Ac-H4) at the hTERT promoter. The transcription permissive histones (Tri-Me-K4 H3 and Ac-K9 H3) are increased in E6 expressing cells compared to HFK with vector. However, a transcriptional repressive histone DiMeK9-H3 is slightly decreased in E6 expressing cells. (B) RNA Polymerase II is preloaded at the hTERT promoter prior to its activation. Antibody against RNA Pol II and IgG were used for IP and precipitates were subjected to real time PCR. Pol II binds to the hTERT promoter in HFK before its activation and E6 does not significantly alter Pol II bound to the hTERT promoter. (C) E6 induces Pol II phosphorylation at serine 2 (S2) of CTD. Antibodies against Pol II CTD and S2 phosphorylated CTD were used for quantitative ChIP assays. Upon expression of E6, Pol II protein is increased only slightly (1.5 fold) on the hTERT promoter. More significantly, the Pol II protein shows a 5-fold increase in phosphorylation of its CTD serine 2 residue.

Figure 6. Myc is required for E6-induced chromatin changes at the hTERT promoter

HFK-E6AU1 cells were infected with retroviruses expressing pGL2 shRNA or Myc shRNA and selected with puromycine for 4-5 days. (A) shRNA mediated knockdown of Myc expression. Total cellular RNA was isolated with Trizol (Invitrogen) and QRT-PCR was performed with Myc specific primers. GAPDH mRNA was used to normalize Myc expression. (B) E6 expression does not change in Myc shRNA expressing cells. The above cells were lysated with 2x SDS buffer and subjected to Western blot with antibodies against Myc, AU1 or beta-actin. Myc is decreased in Myc shRNA expressing cells and E6 expression remains identical in both pGL2 and Myc shRNA expressing cells. (C) Myc shRNA decreases telomerase activity in E6 expressing cells. The above same shRNA treated cells were lysated with TRAP buffer and QRT-TRAP assay was performed as described in Materials and Methods. A decreased telomerase activity in HFKs expressing E6AU1 were observed after treatment with Myc shRNA, not pGL2 shRNA. (D) Myc is required for E6 induced chromatin changes at the hTERT promoter. HFKs expressing E6AU1 or empty vector, LXSN, were lysated with ChIP buffer and precipitated with antibodies against Ac-H3, Ac-H4, AU1, or Pol II. Mouse and rabbit IgG mixture were also included as a negative control. Myc shRNA significantly descreases histone acetylation (Ac-H3, Ac-H4) and slightly decreases E6 and Pol II at the hTERT promoter in E6 expressing cells.

Figure 7. A working model for regulation of the hTERT promoter by E6 and Myc proteins

Results from the current and previous studies are summarized to illustrate the possible mechanism for regulation of the hTERT promoter by the HPV E6 and Myc oncoproteins. Myc and RNA polymerase II are pre-existing on the “silent” hTERT promoter. E6 induced epigenetic changes may “push” the “paused Pol II complex” on the promoter, leading an active transcription.