Conformational Stability of the N-Terminal Region of MDM2

Abstract

MDM2 is an E3 ubiquitin ligase which is crucial for the degradation and inhibition of the key tumor-suppressor protein p53. In this work, we explored the stability and the conformational features of the N-terminal region of MDM2 (N-MDM2), through which it binds to the p53 protein as well as other protein partners. The isolated domain possessed a native-like conformational stability in a narrow pH range (7.0 to 10.0), as shown by intrinsic and 8-anilinonapthalene-1-sulfonic acid (ANS) fluorescence, far-UV circular dichroism (CD), and size exclusion chromatography (SEC). Guanidinium chloride (GdmCl) denaturation followed by intrinsic and ANS fluorescence, far-UV CD and SEC at physiological pH, and differential scanning calorimetry (DSC) and thermo-fluorescence experiments showed that (i) the conformational stability of isolated N-MDM2 was very low; and (ii) unfolding occurred through the presence of several intermediates. The presence of a hierarchy in the unfolding intermediates was also evidenced through DSC and by simulating the unfolding process with the help of computational techniques based on constraint network analysis (CNA). We propose that the low stability of this protein is related to its inherent flexibility and its ability to interact with several molecular partners through different routes.

Keywords: MDM2; circular dichroism; conformational stability; constraint network analysis; differential scanning calorimetry; fluorescence; intrinsically disordered protein.

Figure 1

Structure of MDM2, as modeled by AlphaFold (entry Q00987). The three small well-folded domains are shown (sphere representation): N-terminal region N-MDM2 (red), including the p53 binding domain; zinc finger domain (blue); C-terminal RING domain (yellow). For N-MDM2 (circled), the details of the secondary structure are shown (inset on the left, having a different orientation), with α-helices in red and β-strands in blue.

Figure 2

pH-induced structural changes in N-MDM2 followed by spectroscopic techniques: (A) variations in the intrinsic fluorescence of N-MDM2 monitored by the changes in the <λ>. (B) changes in intensity of the ANS binding followed at 480 nm for N-MDM2. The ANS concentration was 100 µM. (C) changes in molar ellipticity at 222 nm, from far-UV spectra of N-MDM2. Spectra were acquired in either 1 cm pathlength (fluorescence) or 0.1 cm pathlength cells (CD); buffer concentration was 10 mM in all cases. The curves represent the variation in a spectroscopic parameter (raw ellipticity at 222 nm for CD, and the <λ> or the intensity at a particular wavelength for fluorescence) vs. the pH.

Figure 3

Structural changes in N-MDM2 induced by pH and GdmCl denaturation, followed by hydrodynamic methods (SEC). (A) changes in the elution volume (mL) of N-MDM2 as the pH was varied. The bed volume of the column was 18.98 mL, and at pH values < 4.0, the proteins eluted at volumes larger than the bed volume; these experimental elution volumes have not been represented. (B) changes in the elution volume of the domain as the GdmCl concentration was modified. The points correspond to elution volume of the protein species appearing at any concentration (red circles), at [GdmCl] ≥ 0.25 M (blue squares), and at [GdmCl] ≥ 2.5 M (green squares). The bars are the standard deviations from three different measurements for each pH or GdmCl concentration.

Figure 4

Differential scanning calorimetry and thermal denaturations followed by an external probe. (A) differential scanning calorimetry. The thermogram corresponding to the thermal unfolding of N-MDM2 was analyzed with a two-transition model. More complex models performed worse and did not converge. (Inset) Using the estimated thermodynamic parameters for each transition (T_m,i, ΔH_i(T_m,i), and ΔC_P,i), the populations of the different conformational states were calculated: native state (continuous black line), first intermediate state (dashed black line), second intermediate state (continuous gray line), and unfolded state (dashed gray line). (B) thermal denaturations followed by the fluorescence of the SYPRO Orange. Two unfolding traces are shown, corresponding to two different concentrations: 14 μM (circles) and 25 μM (squares). The analysis was performed employing the two-state (single-transition) model to estimate T_m, ΔH(T_m), and ΔC_P.

Figure 5

GdmCl-denaturation of N-MDM2 followed by spectroscopic techniques. The curves represent the variation in a spectrocopic parameter (raw ellipticity at 222 nm for CD and the <λ> for fluorescence) vs. the concentration of denaturant. (A) changes in the ellipticity at 222 nm (far-UV CD). (B) changes in the intrinsic fluorescence monitored by <λ>, after excitation at 280 nm. (C) changes in ANS fluorescence monitored by <λ>, after excitation at 370 nm.

Figure 6

Global flexibility indexes of N-MDM2, as a function of the reaction coordinate of the unfolding process. (A) normalized number of inner degrees of freedom in the protein, Φ, also known as the floppy mode density. (B) cluster configuration entropy, Σ, describing the disorder of the network. (C) rigidity order parameter, Π, indicating the number of constraints in the network.

Figure 7

Local flexibility indexes of N-MDM2 as a function of the residue. (A) percolation index, P, and rigidity index, R, both related in different ways to protein rigidity. (B) frequency of unfolding nuclei, F, which is proportional to the probability of a residue to behave as a weak spot for the protein stability.

Figure 8

Mapping of the two local indexes P and R onto the native structure of N-MDM2 as a function of the reaction coordinate of the unfolding process. The main transitions found in the analysis of the two key global indexes, in the ensemble of networks created from the input protein structure, are indicated: the cluster configuration entropy Σ (transition at 1.83 kcal mol⁻¹) and the rigidity order parameter Π (2.16 kcal mol⁻¹).