Fig 1.
E6 proteins display different E6AP interaction profiles.
(A) Schematic of B42 transactivator-domain E6AP fusion proteins utilized in panel B. B42_LQELL contains E6AP residues 406–417. B42_LQELS also contains E6AP residues 406–417 but the double LL (residues 412 and 413) have been mutated to LS. B42_E6AP_Ub–_LQELL consists of ubiquitin ligase dead full-length E6AP. B42_E6AP_Ub–_LQELS is also ubiquitin ligase dead full-length E6AP, but is mutated in the LQELL motif to LQELS, and similarly B42_E6A_Ub-LQEAS as indicated. (B) E6 proteins from human and animal papillomaviruses have different requirements for interaction with E6AP. Bait yeast strains expressing the LexA DNA binding domain fused to E6 proteins from the listed papillomaviruses were mated to prey yeast expressing B42 transcriptional activation domain fusions illustrated in part A. HPV41 E6 preferentially binds MAML1 and not E6AP, and was used as a negative control for binding. The horizontal white line indicates development on parallel matched XGAL plates. H = Homo sapiens (human), Ps = Phocoena spinipinnis (Burmeister’s porpoise), Pph = Phocoena phocoena (harbor porpoise), Tt = Tursiops truncatus (bottlenose dolphin), Mm = Macaca mulata (rhesus monkey), Ss = Sus scrofa (wild boar), Um = Ursus maritimus (polar bear).
Fig 2.
The 16E6_L50A mutant enables characterization of E6-E6AP interactions.
(A) The E6 L50 residue is in close proximity to the double L residues in the E6AP LQELL peptide. The HPV16 E6 structure (PDB file 4GIZ) is depicted with the amino terminal zinc-structured domain in green, the carboxy terminal zinc-structured domain in blue, and the connecting alpha helix in yellow. The 16E6 protein is shown interacting with the E6AP LQELL peptide (in light pink), and the side chains of the double L residues are shown. The side chain of the E6 L50 residue is highlighted in red. The E6 L50 residue (red) makes a 3.7 Å side chain contact with the last leucine residue (L) in the E6AP LQELL peptide which is lost upon mutation of L50 to an alanine. (B) Mutation of 16E6 L50 residue to an alanine results in a high-risk E6 protein that is unable to bind the isolated E6AP LQELL peptide but retains association with full-length E6AP. Bait Yeast expressing LexA fused to either full length ubiquitin ligase dead E6AP (E6AP_Ub–) or the isolated E6AP LQELL peptide (E6AP_406–417) were mated to yeast co-expressing untagged 16E6_WT or the indicated 16E6 mutants containing a single amino acid change and p53.
Fig 3.
The amino terminal region of E6AP is required for 16E6 degradation of p53.
(A) Schematic of E6AP amino terminal truncations. Previously described E6 LQELL binding region is between amino acids 403 and 416, as depicted. Each E6AP amino terminal truncation retains the E6 LQELL binding region and the E6AP carboxy terminal HECT ligase domain. (B) Ability of 16E6_L50A to bind to E6AP requires E6AP residues 315–331. Yeast strains expressing the LexA DNA binding domain fused to ubiquitin ligase dead (Ub–) E6AP full length (FL), amino terminal E6AP truncations, or the isolated E6AP LQELL peptide (406–417). These bait yeasts were mated to yeast strains co-expressing p53, 16E6_WT, 16E6_L50A, or transactivator Gal4 (G4) fused to the PDZ protein PTPN3 as indicated. Positive controls include 16E6_WT co-expressed with p53 to ensure E6AP expression and G4-PTPN3 co-expressed with 16E6_WT to ensure 16E6_WT expression. Ability of 16E6_L50A to bind E6AP_Ub- and recruit p53 is lost when E6AP is truncated from residue 315 to residue 331. Horizontal black line indicates removal of an irrelevant sample. (C) Ability of E6 to stimulate ubiquitin ligase activity in the complex with p53 and PTPN3 requires E6AP amino acids 310–315. Bait yeast were transfected with the LexA DNA binding domain fused to WT E6AP containing the same truncation endpoints as described in B. These yeast strains were mated to the same yeast as described in B and diploids selected. Upon truncation of E6AP from amino acid 300 to 315, the interaction of p53 and PTPN3 with E6 was restored as indicated by the appearance of blue dye. Horizontal black line indicates removal of irrelevant samples. (D) Requirement of E6AP amino acids 310–320 for E6 ability to initiate p53 degradation in the presence of E6AP_WT is recapitulated in E6AP-null 8B9 cells. Plasmids encoding the indicated FLAG_E6AP_WT truncations (1.25 ug), human p53 (0.5 ug), HA-GFP (0.01 ug), and either untagged 16E6_WT or 16E6_L50A (2 ug), as indicated, were transiently transfected into murine E6AP-null 8B9 cells and p53 and E6 expression were analyzed by western blot. 16E6_WT requires E6AP amino acids 315–320 to initiate E6AP-mediated degradation of p53 while 16E6_L50A requires E6AP amino acids 310–315. A single representative blot is shown. Vertical black line indicates removal of an irrelevant sample. UT = untransfected. Quantitation of protein expression from triplicate experiments as shown below the blots. E6AP stabilization of 16E6 (black line) is not required for p53 (gray line) degradation by the E6-E6AP complex. p53 levels and E6 levels are normalized to co-transfected HA_GFP. HA_GFP normalized E6 expression levels are further normalized to 16E6_WT protein levels in the presence of full length E6AP (lane 4 in panel D). HA_GFP normalized p53 protein expression levels are normalized to p53 levels in the presence of 16E6_WT with no co-expressed E6AP protein (lane 2 in panel D). The means of triplicate independent experiments ± standard errors are shown. E. Diverse E6 proteins require E6AP region 300–403 for ubiquitin ligase function. The indicated LexA_E6 fusions from Fig 1 are expressed together with transactivator fusions of E6AP active for Ubiquitin ligase activity (WT) or Ubiquitin minus (Ub-) in columns, either full length E6AP_1–875, or amino-terminal truncations 300–875 or 403–875. White vertical line indicates the division from 2 matched plates.
Fig 4.
Co-immune precipitations between 16E6, fragments of E6AP, and human p53.
(A) Untagged 16E6_WT (4 ug), human p53 (2.9 ug), HA_GFP (0.1 ug), and the indicated FLAG_E6AP_Ub– truncations (3 ug) were co-transfected into E6AP-null 8B9 cells and harvested 18 hrs. later in 0.5X IGEPAL lysis buffer as described in methods. Western blots of input samples are clustered at the bottom and FLAG-immunoprecipitated (IP) samples are clustered at the top. Input was 5% of the immunoprecipitated sample size. The 16E6_WT IP blot shows a short and a long exposure; the overexposure is necessary to see 16E6_WT in lanes 7 and 9–11 (possibly due to lower expression of FLAG_E6AP truncations in those lanes). Ub– indicates a ubiquitin ligase dead E6AP mutant. UT = untransfected. (B) Amino-terminal E6AP truncations maintain ability to be loaded with ubiquitin. HA_Ubiquitin (6 ug) and FLAG_E6AP (3 ug; either ubiquitin ligase dead (Ub–) or ubiquitin ligase active (WT)) plasmids were co-transfected into E6AP-null 8B9 cells and harvested 18 hrs. later in 0.5X IGEPAL lysis buffer as described in the methods. Western blots of input samples are clustered at the bottom and HA-immunoprecipitated samples are clustered at the top. Input was 5% of the immunoprecipitated sample size. E6AP_Ub– indicates a ubiquitin ligase dead E6AP mutant. Black line indicated removal of an irrelevant sample.
Fig 5.
No single amino acid point mutation within E6AP region 310–320 prevents 16E6_WT from initiating degradation of p53.
(A) Schematic of full length E6AP protein. The amino acids located between 310 and 320 (inclusive) are depicted, as is the location of the HECT ligase domain and the active cysteine residue (C843), responsible for ubiquitination of p53. (B) Full length E6AP containing single amino acid point mutations within the 310–320 region still degrade p53 in the presence of 16E6_WT. Plasmids encoding the indicated FLAG_E6AP (1.25 ug), human p53 (0.5 ug), HA_GFP (0.01 ug), and either 16E6_WT or 16E6_L50A (2 ug) were co-transfected into E6AP-null 8B9 cells. Δ310–320 is full length E6AP_WT deleted of residues 310–320 (inclusive). Amino terminal E6AP truncation 320–875 was used as a negative control for p53 degradation in the presence of either 16E6_WT or 16E6_L50A. Vertical black line indicates samples were run on two different western blots. UT = untransfected.
Fig 6.
Low-risk and high-risk E6 proteins require the same amino terminal region of E6AP to initiate degradation of cellular substrates.
NHERF1 degradation lost in the presence of a ubiquitin active truncation E6AP 314–875 with both (A) low-risk 11E6 and (B) high-risk 16E6. The listed FLAG_E6AP_WT truncations were co-transfected with HA_NHERF1 (0.75 ug), HA_GFP (0.08 ug), and FLAG_11E6_WT (2 ug; panel A) or 16E6_WT (2 ug; panel B) into E6AP-null 8B9 cells and HA_NHERF1 expression analyzed by western blot. HA_GFP was co-transfected as a transfection control. Bar graphs below western blots indicate quantitation of HA_NHERF1 protein levels first normalized to HA_GFP to account for transfection variability and then normalized to HA_NHERF1 protein levels in the presence of full length E6AP with no co-expressed E6 (lane 2). E6AP amino terminal truncation to residue 314 displays a lack (with 16E6_WT) or loss (with 11E6_WT) of targeted NHERF1 degradation. The shown experiment is representative of four separate experiments.
Fig 7.
E6AP LQELL is important for efficient degradation of cellular substrates by both high and low-risk E6 proteins.
HA_GFP (0.1 ug), human p53 (0.15 ug) or HA_NHERF1 (0.3 ug), and each of the listed FLAG_E6AP variants (0.2 ug) were co-transfected with either untagged 16E6_WT or FLAG_11E6_WT (0.2 ug) in E6AP-null 8B9 cells. E6AP_Ub– indicates a ubiquitin ligase dead E6AP protein that was used as a control. E6AP_LQEAS contains two single amino acid point mutations in the E6 LQELL binding region within E6AP. HA_NHERF1 and p53 protein levels in the presence of the indicated E6 and E6AP proteins were normalized to HA_GFP. The bar graph below the blot represents quantification of either HA_NHERF1 protein levels (black bar) or human p53 protein levels (gray bar) in the presence of the indicated E6 and E6AP proteins. UT = untransfected. Blots from a single representative experiment are shown with quantitation from the average and standard deviation of 4 experimental replicates. *, P< 0.05; ** P<0.01 by Student’s t test.
Fig 8.
The extreme N-terminus of E6 may play a role in E6AP binding.
(A and B) Yeast strains expressing the indicated LexA DNA binding domain fusions were mated to prey yeast transfected with B42 transactivator fused E6AP constructs depicted in Fig 1A, transactivator-fused PTPN3, p53, or unfused E6AP_Ub- plus p53 as indicated. LexA_LQELL_16E6 is a triple fusion of LexA to the LQELL peptide of E6AP then to 16E6; in this construction, the LQELL peptide interacts with 16E6 in cis and competes for the binding of B42_LQELL expressed in trans [39]. 16E6_WT was used as a positive control for expression of E6AP proteins, although none of the E6 proteins interacted with B42_LQEAS. The low-risk 11E6 protein is unable to interact with PTPN3 due to a lack of a PDZ binding motif. Deletion of residues 1–8 of 16E6 or 1–9 of 11E6 impairs the Type II interaction of 16E6 with E6AP.
Fig 9.
Full-length E6AP outcompetes cis LQELL binding to 16E6_WT in vivo.
Plasmids encoding the indicated FLAG_E6AP_WT (0.5 ug), human p53 (0.5 ug), HA_GFP (0.1 ug), and the listed 16E6 proteins (0.5 ug) were transiently transfected into E6AP-null 8B9 cells and p53 protein expression was analyzed by western blot. LQELL_16E6_WT has an intramolecular interaction such that the 16E6 protein binds to the amino terminally fused LQELL E6AP peptide [39]; scramble_16E6_WT contains a same sized amino terminal fusion with a scrambled LQELL sequence. Unfused 16E6_WT serves as a control to ensure that scramble_16E6_WT degrades p53 as well as untagged 16E6_WT, indicating the amino terminal fusion is not disrupting the fold of the E6 protein. Full-length E6AP outcompeted the LQELL_16E6_WT intramolecular interaction resulting in p53 degradation while the amino terminal E6AP truncation 300–875 did not outcompete the LQELL_16E6_WT interaction. Mean p53 protein levels +/- standard deviations from 4 separate experiments were quantitated and normalized to HA_GFP to account for transfection efficiency. For comparison purposes, the HA_GFP normalized p53 levels were then normalized to p53 protein levels in the presence of full-length E6AP with no co-expressed 16E6 (lane 2). UT = untransfected. Blots from a single representative experiment are shown with quantitation from the average and standard deviation of 4 experimental replicates. *** P<0.001 by Student’s t test.
Fig 10.
E6AP regions in addition to the LQELL motif in both the amino and carboxy terminus are important in mediating 16E6 binding.
(A) E6AP residues 121–127 are important for enabling 16E6_L50A interaction and p53 recruitment. At the top, there is a schematic of E6AP and E6AP truncation mutants with yeast hybrid associations shown below. E6AP amino terminal truncations lack the HECT ligase domain, but retain the LQELL E6 binding motif. The indicated yeast expression plasmids were introduced into a LexA-responsive reporter strain by mating. Co-expression of 16E6_WT and p53 with each LexA_E6AP serves as a positive control for E6AP expression and folding. Co-expression of Gal4 (G4) transactivator fused PTPN3 with 16E6_WT and the various LexA_E6AP truncations ensures that 16E6_WT is being expressed and can recruit the PDZ protein PTPN3 to the E6AP protein. Vertical black line indicates removal of irrelevant samples. (B) E6AP residues 521–497 are important in mediating 16E6_L50A interaction with E6AP to recruit p53. At the top is a schematic of E6AP and E6AP truncation mutants examined for interaction with 16E6_L50A + p53. E6AP truncations lack the HECT ligase domain but retain the LQELL E6 binding motif. Yeast containing three hybrid plasmids expressing the indicated proteins were mated, selected, and analyzed for interaction as in part A. The PDZ protein PTPN3 co-expressed with 16E6_WT and each E6AP truncation demonstrated ability of 16E6 to bind the E6AP protein and recruit PTPN3.
Fig 11.
The low-risk 11E6 binding to E6AP requires E6AP residues located in the HECT ligase domain.
(A) Schematic of full length E6AP and E6AP truncations examined in panel B. Each C-terminal E6AP truncation contains the E6AP LQELL E6 binding motif at the amino terminus. (B) Bait yeast transfect with plasmids encoding LexA DNA binding domain fused 16E6_WT, 11E6_WT, or MmPV1E6_WT were mated to yeast containing B42 transactivator fused E6AP truncations as indicated. LexA_16E6_WT served as a positive control for E6AP expression and folding as 16E6 should interact with each E6AP truncation because they all contain the LQELL motif. 11E6 requires E6AP residues 561–595 to interact with E6AP.
Fig 12.
Summary schematic: Identified regions in E6AP important in mediating E6 binding and cellular protein degradation.
The herein identified and described regions within E6AP are indicated as being important in mediating E6 binding and/or in E6-mediated degradation of cellular proteins.