Análisis desde el Interaccionismo Simbólico

Although some of the major biological activities of E7 were described in previous sections, E7 has the potential to affect many other cellular processes. The effects exerted by E7 on the host cell are primarily carried out by binding numerous cellular factors. Understanding the structure of the HPV E7 proteins can provide great insight into the mechanisms behind E7’s ability to target a broad spectrum of cellular targets. In addition, the structure could also significantly aid in planning mutational studies that mechanistically address E7 functions and interactions. Potentially, structural studies could also help explain the functional differences between the high-risk and low-risk E7 proteins, and the malignant progression of lesions induced by high-risk HPVs. Ultimately, this information may suggest new antiviral strategies to interfere with the action of E7.

With those aims in mind, a solution structure of the high-risk HPV45 E7 protein was solved using Nuclear Magnetic Resonance (NMR) spectroscopy (FIG.1-6) (174). Experiments using full-length HPV45 E7 and a construct representing its CR3 region revealed that the N-terminus (amino acids, 1-54) of this oncoprotein is unstructured and flexible in solution, whereas the C-terminus (residues 55-106) folds autonomously into a well-structured zinc-binding domain. Results obtained from the NMR analysis are consistent with the intrinsic disorder of high-risk E7 proteins predicted by bioinformatic studies (229) and with a previously reported secondary structure of E7 proteins (228). The N-terminal region of E7 represents an example of an intrinsically disordered region that may undergo a localized conformational change upon interaction with its biological target (66). This is not surprising given the fact that intrinsic protein disorder is very common in cancer associated proteins, with approximately 79% of cancer-associated and 66% of cell-signaling proteins containing predicted regions of disorder of 30 residues or longer (120). The dynamic nature of the N-terminus may aid HPV E7 proteins in fulfilling some of their biological functions. In disordered domains, protein function is traced to short sequences called “linear motifs”. Indeed, the N-terminus of E7 contains multiple functional linear motifs that contribute to binding of cellular targets, with the LXCXE motif being a prime example (29, 78, 146, 188, 203). Linear motifs within the

F st n gr D up st w FIG. 1-6. R tructural sch ot shown as reen and blu Different view p 90° for B tructures we was originally Representat hematic of th s they are in ue, and the a ws were gen BOTTOM v ere generated y published tive 3-dimen he HPV45 E ntrinsically d alpha helices nerated by ro view, and l d from the P in Oncogene nsional stru E7 (residues disordered. s and beta sh otating the to left or coun PDB file #2 e (2006), 25 ucture of th 55-106) CR Structures o heets are lab op left struct nter clockwi 2F8B using P :5953-5959. he HPV45 R3 dimer. C of protomers beled in the t ture down 90 ise 90° for PyMOL sof . CR3 dimer CR1 and CR s are colour top left struc 0° for TOP v side view. ftware. (Stru r. A R2 are red in cture. view, The ucture

intrinsically disordered N-terminus may be beneficial by providing multiple binding sites, with high specificity but low affinity, extending the capacity of the interaction surface to bind diverse proteins and enhancing protein association rates (254). Indeed, the localization of the LXCXE motif within an intrinsically disordered domain is thought to provide the fast, diffusion-controlled interaction that allows viral proteins to outcompete physiological targets of pRb (27).

The solution structure of the C-terminal domain of HPV45 E7 is virtually identical to the X-ray crystallography structure of a CR3 construct derived from the low-risk HPV1 E7 protein (148). Both of these E7-CR3 domains form a well-structured zinc-binding domain with a unique β1β2α1β3α2 topology containing a zinc binding fold that is not found in any other solved structures (FIG.1-6); however, it has been proposed that the fold arose from a host PHD domain (45), which is also involved in protein-protein interactions. This region contains a C4-type zinc finger which co-ordinates one molecule of zinc ion. The distance between the two Cys-X-X-Cys motifs is too large to form a classical zinc finger structure that was first proposed for the Xenopus transcription factor IIIA (TFIIIA) (144, 166).

Both structural studies demonstrate that E7 assembles as a roughly globular, obligate zinc-dependent dimer. Conversely, other studies have suggested that E7 CR3 may adapt a range of conformations that contribute to the protein interaction repertoire of E7. It was shown that E7 CR3 protomers modulate binding to the AB domain of the retinoblastoma protein (28, 29) and are able to bind zinc, and yet other studies have shown that CR3 can also form large structural oligomers (4, 5, 39). It is therefore unclear whether E7 dimerization is of functional importance.

Although the co-ordinated zinc ions are not directly involved in E7 dimer formation, they are important for maintaining the folded state of the protomeric CR3. Indeed, the removal of zinc ions causes unfolding of CR3 and formation of large oligomeric structures. E7 forms a stable dimer through interactions of α1 helices of each protomer and β-sheet interactions between the β2 and β3 strands of opposing protomers. Dimerization leads to the formation of a contiguous hydrophobic core that is further stabilized by a subset of

residues that form intersubunit contacts. A sequence alignment of the CR3 domains from E7 proteins previously predicted that the E7 dimer is maintained in large part by a hydrophobic core, based on the observations that 7 hydrophobic residues in CR3 exhibit more than 90% conservation, and 5 more exhibit 70% conservation (228). Based on sequence analysis of more than 200 E7 sequences, the highest degree of sequence conservation within the CR3 region, aside from the invariant zinc-coordinating cysteins, belongs to the residues that form the hydrophobic dimerization interface (27). These findings suggest that the structural features of CR3 are likely to be very highly conserved among E7 proteins.

A large number of cellular targets have been reported to interact with E7 via the C- terminus; however, most studies have not attempted to map the residues on the surface which are necessary for the interaction with the particular target. On the contrary, studies which have endeavored to carry out mutational and mapping studies have frequently utilized mutants within CR3 which target residues critical for the formation of the hydrophobic core and/or dimerization. These studies include mapping of p21Cip1, p27Kip1, TBP, TAF110, Mi2β and pCAF (9, 26, 83, 126, 156, 157, 249). The loss of interaction or function observed in these studies could likely be attributed to gross structural effects upon E7 or to the loss of dimerization rather than to impairment of specific functions. This issue has consistently been raised with mutations in the coordinating cysteines that destroy protein stability (32, 184, 233) and impair transformation and transactivation functions (70, 159, 184, 215). To explore the function of CR3, a thorough structure- function analysis will need to be performed using a set of reagents which take into consideration that the CR3 domain of E7 is highly structured, and which aim to preserve the folded state.