Quaternary structure of p53: The light at the end of the tunnel

Abstract

Eukaryotic transcription factors belong to a class of proteins that often have folded domains linked by intrinsically disordered segments. These modular proteins have an inherent problem of being resistant to crystallization and the concern that even if they were crystallized the resultant structure might not be representative of the one in solution because of crystal packing effects. They are also difficult to study by NMR because they are generally so large. A notable member of this class of proteins is the tumor suppressor p53, which is the key protein in the cell's defenses against cancer (1–3). p53 is ≈40% intrinsically disordered. Of 393 residues of each chain of its tetrameric structure, only residues 100–300, the core domain (CD) that binds sequence specifically to DNA, and residues 324–355, the tetramerization domain (Tet), are folded. The structures of the two domains were determined by x-ray crystallography and NMR methods. The CD adopts an Ig-like β-sandwich that provides a scaffold for a DNA binding surface (4, 5). The Tet forms a symmetrical tetramer made of two tight dimers stabilized by an antiparallel β-sheet and helix–helix interactions (6–9). The sequences of the p53 DNA binding sites or response ele-ments have four subsites, 5 bp long each, so that four CDs in tetrameric p53 can bind to one response element. Okorokov et al. (10) recently proposed a model for the structure of murine p53 on the basis of cryoelectron microscopy and single-particle analysis. Their model was very unusual in that it is composed of a skewed cube with the Tet being dissociated, so that four p53 monomers interact via their N and C termini. The arrangement of the CDs in this structure is not compatible with all four binding to the four subsites of the DNA response element, suggesting an alternative binding mode that has yet to be proven.