Cancer-Associated Mutations Perturb the Disordered Ensemble and Interactions of the Intrinsically Disordered p53 Transactivation Domain

Abstract

Intrinsically disordered proteins (IDPs) are key components of regulatory networks that control crucial aspects of cell decision making. The intrinsically disordered transactivation domain (TAD) of tumor suppressor p53 mediates its interactions with multiple regulatory pathways to control the p53 homeostasis during the cellular response to genotoxic stress. Many cancer-associated mutations have been discovered in p53-TAD, but their structural and functional consequences are poorly understood. Here, by combining atomistic simulations, NMR spectroscopy, and binding assays, we demonstrate that cancer-associated mutations can significantly perturb the balance of p53 interactions with key activation and degradation regulators. Importantly, the four mutations studied in this work do not all directly disrupt the known interaction interfaces. Instead, at least three of these mutations likely modulate the disordered state of p53-TAD to perturb its interactions with regulators. Specifically, NMR and simulation analysis together suggest that these mutations can modulate the level of conformational expansion as well as rigidity of the disordered state. Our work suggests that the disordered conformational ensemble of p53-TAD can serve as a central conduit in regulating the response to various cellular stimuli at the protein-protein interaction level. Understanding how the disordered state of IDPs may be modulated by regulatory signals and/or disease associated perturbations will be essential in the studies on the role of IDPs in biology and diseases.

Keywords: NMR; atomistic simulation; biolayer interferometry; conformational equilibrium; intrinsically disordered proteins.

Figure 1.. p53-TAD domain structure, interactions and regulation.

(A) Domain structure of p53: TAD (transactivation domain), PRD (proline-rich domain), DBD (DNA-binding domain), TD (tetramerization domain), and NRD (negative regulatory domain). The sequence of human p53-TAD is shown with the selected cancer-linked mutation positions marked in red and the amino acid substitutions shown in green. The helical segments within TAD, AD1 and AD2, are shown below with the interacting partners, HDM2 and CBP/p300. (B) p53 regulation is mediated by HDM2 and CBP/p300. Top: In unstressed cells, HDM2 binds to AD1 and negatively regulates p53 stability by polyubiquitination (UB) of NRD. Bottom: Under genotoxic stress, phosphorylation of TAD inhibits HDM2-p53-TAD interaction and promotes the formation of a complex with CBP/p300, which leads to acetylation (AC) of NRD, and to activation and stabilization of p53.

Figure 2.. Binding of the p53-TAD variants to HDM2 (residues 17–125), TAZ1, and TAZ2.

100 ng/μL of p53-TAD^WT and its variants were applied to Ni-NTA BLI biosensor tip prior to the introduction of 3 μM HDM2(17–125) (A), 30 μM TAZ1 (B), or 300 nM TAZ2 (C). BLI signal response was set at 0 at the beginning of the association phase (time 0 s). The dissociation phase was initiated at 300 s for HDM2 and TAZ2 or at 120 s for TAZ1. Each trace represents the average response from two independent experiments. The control traces labeled “Non-Specific” were obtained without p53-TAD on the biosensor tip.

Figure 3.. Dynamic binding interface between p53-TAD and TAZ2.

All 20 members of the NMR ensemble (PDB: 5HPD) are shown for p53-TAD (green, with AD1 and AD2 colored in purple and orange, respectively), while a single copy of TAZ2 (model 1 of the ensemble) is shown as silver molecular surface. Side chains of p53-TAD residues K24, N29, N30, D49 and W53 are shown in stick representation.

Figure 4.. Chemical shift perturbations of p53-TAD variants.

(A) Overlay of p53-TAD ¹⁵N-HSQC spectra for WT (orange), K24N (blue), N29K/N30D (purple), D49Y (pink), and W53G (green). Non-overlapped positions indicate changes in backbone amide chemical environment due to a mutation. (B) Chemical shift perturbation of each assigned p53-TAD residue referenced to their original WT assignment. A stronger chemical shift perturbation indicates a greater impact upon a residue’s chemical environment due to mutation.

Figure 5.. Effects of p53-TAD variants on dynamic properties.

The NHNOE/NONOE Ratio (A) and J(ω_N)/J(ω_H) ratio (B) for each p53-TAD variant was calculated on per residue basis with K24N (blue), N29K/N30D (purple), D49Y (pink) and W53G (green). For each p53-TAD variant, an overlay of the WT data (black) is included as the reference. Error bars represent 95% confidence interval. Higher J(ω_N)/J(ω_H) and NHNOE/NONOE ratios indicate a greater steric restriction, while lower values for both correlate with a greater flexibility.

Figure 6.. Conformational properties of free p53-TAD variants obtained from atomistic simulations.

(A) Residue helical propensity of four p53-TAD mutated variants in comparison with WT. Both α helices and 3₁₀ helices were included in this calculation, since the probability of forming 3₁₀ helices is nonnegligible in force field a99SB-disp. (B) Radius of gyration (R_g) probability distribution of four p53-TAD variants in comparison with WT. (C) Probability distribution of R_g of p53-TAD residues 41–61 for four protein variants in comparison with WT. The shaded areas indicate uncertainties, which were estimated by dividing each trajectory into three equal segments and calculating the standard deviation of the mean values.

Figure 7.. Conformational distributions of free p53-TAD^WT and its variants.

The simulated structural ensembles of p53-TAD were projected onto the first two PCs derived from PCA analysis. Heatmaps show the negative logarithm of the probability distribution. Values in the parenthesis are percentages of variance captured in each direction. Ten representative structures, along with their average radius of gyration, are shown for a few highly populated states, where the peptide is shown in backbone trace, with the color changing from red at the N-terminus to blue at the C-terminus.