Reduced structural rigidity of MDMX protein enhances binding to TP53 mRNA

1 Reduced structural rigidity of MDMX protein enhances binding to TP53 mRNA Martina Kucerikova1,2, Ondrej Bonczek1, Vanesa Olivares-Illana3, Andres RodriguezRodriguez3, Jose G. Sampedro3, Lenka Hernychova1, Vaclav Hrabal1,4, Pavlina Zatloukalova1, Radovan Krejcir1, Robin Fahraeus1,5,6, Philip J. Coates1, Borivoj Vojtesek1,7*, Lucia Martinkova1* 1 RECAMO, Masaryk Memorial Cancer Institute, 602 00, Brno, Czech Republic 2 National Centre for Biomolecular Research, Faculty of Science, Masaryk University, 625 00, Brno, Czech Republic 3 Instituto de Fisica, Universidad Autonoma de San Luis Potosi, 78290 San Luis Potosi, Mexico 4 Department of Experimental Biology, Faculty of Science, Masaryk University, 625 00, Brno, Czech Republic 5 Inserm UMRS1131, Institut de Genetique Moleculaire, Universite Paris Cite, Hopital St. Louis, 75010, Paris, France 6 Department of Medical Biosciences, Umea University, 901 87, Umea, Sweden 7 Laboratory of Growth Regulators, Institute of Experimental Botany, The Czech Academy of Sciences, 779 00, Olomouc, Czech Republic Correspondence: [email protected], [email protected] Abstract The two murine double minute (MDM) family members, MDM2 and MDMX are a well-established negative regulator of p53 activity. Under DNA damage conditions, MDM2 and MDMX are phosphorylated near their RING domains (serine 395 at MDM2 and serine 403 at MDMX) and switch to act as p53 positive regulators. MDMX binds to TP53 mRNA and acts as a chaperone for RNA structure, enabling MDM2 to bind. This interaction enhances TP53 mRNA translation, leading to increased p53 protein production. While the biological significance of this interaction has been described, the specific features of the MDMX-RNA interaction remain poorly understood. We used various MDMX protein Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 2 constructs to characterize binding to TP53 mRNA and identified that the interaction mediated by the RING domain is modulated by the presence of other domains. Hydrogen-deuterium exchange mass spectrometry (HDX-MS) and binding assays in high salt conditions and various pH demonstrate that the whole protein participates in RNA interaction, with the Cterminal domain likely providing the contact with RNA by electrostatic forces. We show that protein structural changes induced by the chelating agent EDTA or the reducing agent TCEP enhances RNA binding by promoting partial structural destabilization of the protein. Our findings suggest that the MDMX/TP53 mRNA interaction is complex, with the RING domain binding to RNA and being supported by the entire protein, which acts as a scaffold for the RNA interaction. These results contribute to a better understanding of MDMX´s role in TP53 mRNA binding and provide valuable insights for future investigation of the MDM2-MDMXTP53 mRNA complex, which is crucial for p53 stabilization and activation under DNAdamaging conditions. Keywords: MDMX-RNA interaction; TP53 mRNA; RING domain; HDX-MS mapping Introduction MDMX is an oncoprotein that is amplified and overexpressed in various cancers, leading to diminished p53 activity and promoting tumour growth and survival [1,2]. This aberrant expression is often mutually exclusive with TP53 mutations and MDM2 amplification, and is frequently associated with poor prognosis in cancer patients, highlighting its crucial role in tumorigenesis [1,2]. Under normal conditions, MDMX, in collaboration with its homolog MDM2, plays a key role in maintaining low levels of p53 protein by directing p53 for proteasomal degradation [1]. MDMX also exerts p53independent functions, including regulation of cell cycle, genomic instability, and metabolism [3–5]. During DNA damage, a phosphorylation cascade triggered by ataxia-telangiectasia mutated (ATM) kinase disrupts the p53-MDM2-MDMX interaction, resulting in increased p53 protein levels and activation [6]. ATM directly phosphorylates MDM2 at serine 395 and MDMX at serine 403. As a result, MDMX and MDM2 shift their function to act as positive regulators of p53 activity by directly binding TP53 mRNA and enhancing its translation [7]. In this context, MDMX acts as an RNA chaperone protein for TP53 mRNA and enables the efficient binding of MDM2 [8]. Recently, it was shown that the ATM kinase also binds TP53 mRNA under stress conditions, and MDMX prevents this interaction and subsequent interaction with Nijmegen breakage syndrome 1 (NBS1), a component of the MRN (MRE11– RAD50–NBS1) complex involved in DNA damage recognition and repair [9]. Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 3 Structurally, MDMX is an intrinsically disordered protein with several organized domains, including the p53-binding domain at its N terminus, the WWW element, the acidic domain, the zinc-finger domain, and the RING domain (Figure 1A). The RING domain, located at the C-terminus, is critical for heterodimerization with MDM2 [10]. Recent in cellulo studies have also described that homodimerization of MDMX influences its binding to p53 [11]. The RNA-binding activity of MDMX has been mapped to its RING domain [8]. Due to its structural organization, MDMX exhibits significant flexibility [12]. Therefore, studies of domain functions, complemented by full-length protein analysis, can provide a more comprehensive understanding. We characterized key aspects of the MDMX-RNA interaction, providing new insights into how MDMX modulates TP53 mRNA and its potential roles. We confirmed that the RING domain is essential for TP53 mRNA interaction, while other domains play modulatory roles likely due to structural organization. Binding at various pH and high salt conditions, together with HDX-MS, revealed that protein-RNA binding is partially driven by electrostatic forces at the protein's C-terminus, but the entire protein participates in binding and stabilizes the interaction. Furthermore, we investigated how reducing TCEP and chelating conditions EDTA affect both the structural organization of MDMX and its TP53 mRNA-binding capacity, with conformational destabilization enhancing RNA interaction. The role of MDMX as an RNA-binding protein adds an important dimension to its regulatory repertoire. A deeper understanding of these interactions could lead to novel therapeutic approaches in cancer treatment. Material and methods Cloning and purification of protein constructs The original purification protocol [12] was modified as follows: cDNA of MDMX wild-type (WT) or S403D constructs were cloned into HIS-tagged vectors pET28B (Merck Millipore, Darmstadt, Germany) and pDEST17 (Thermo Fisher Scientific, Waltham, MA, USA). The Ser403-to-Asp mutation mimics ATM kinase–mediated phosphorylation at this site in response to DNA damage. The insoluble protein fraction was resuspended in guanidine buffer (6 M guanidine hydrochloride, 50 mM Tris (pH 8.0), 150 mM NaCl, 10 mM imidazole, 2 mM MgCl₂) and incubated at 37 °C for 90 min, with Turbonuclease from Serratia marcescens (Merck Millipore) added during the final 15 min of incubation to remove residual nucleic acids. The Ni-NTA Agarose (Qiagen, Hilden, Germany) was washed with 0.1 M Tris (pH 8), 2 M NaBr, 0.5 M NaCl, and 10 mM 2-mercaptoethanol. Proteins were eluted from the washed resin using 5 ml (5 fractions of 1 ml) of elution buffer (50 mM Tris (pH 8), 300 mM Imidazole, 10 µM ZnSO4·7H2O). Eluted proteins were dialyzed overnight at 4 °C in 50 mM Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 4 Tris (pH 8), 150 mM NaCl, and 10 µM ZnSO₄. After dialysis, proteins were concentrated using 10 kDa MWCO Amicon Ultra-4 centrifugal filters (Merck Millipore). Sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) SDS-PAGE was performed as described previously [13]. 300 ng of purified protein was loaded per lane for electrophoretic separation, and gels were stained with Coomassie brilliant blue for 30 min. Protein visualization was performed using ChemiDoc Imaging (BioRad, Hercules, CA, USA). Electrophoretic mobility shift assay (EMSA) In vitro transcription of TP53 mRNA was performed using the mMESSAGE mMACHINE™ T7 Transcription Kit (Thermo Fisher Scientific) according to the manufacturer’s instructions. The linearized pcDNA3-p53 plasmid (encoding nucleotides 1–240) [8] was used as the DNA template. A total of 100 ng of RNA was mixed with purified protein (in a 10 µL reaction) at RNA:protein molar ratios specified in the figures, using binding buffer (150 mM NaCl, 50 mM Tris, 1 mM TCEP, 10 µM ZnSO₄, pH 7.5) along with additional conditions outlined in the figures. Samples were incubated at room temperature for 20 min before being loaded onto 1.2% agarose gels. The gels were post-stained with GelRed nucleic acid stain (Biotium, Fremont, CA, USA) for 40 min. For protein detection following the GelRed staining, the same gels were fixed in 50% methanol and 7% acetic acid for 20 min, and subsequently stained overnight with SYPRO™ Ruby protein gel stain (Thermo Fisher Scientific). Both gels were imaged using ChemiDoc Imaging (Bio-Rad). Densitometric analysis was performed for all EMSA gels, and the intensity of free RNA was normalized to the RNA-only control. Data represent mean ± SEM from three independent biological replicates (n = 3). Statistical significance was determined using two-tailed Student’s t-test, with the following annotations: *p < 0.05, **p < 0.01, ***p < 0.001. In addition, some comparisons between specific conditions are indicated in the figures using alternative symbols: •p < 0.05, ••p < 0.01, •••p < 0.001. HDX-MS HDX-MS of the HDMX-S403D protein was conducted in both ligand-free and ligand-bound conformations with TP53 mRNA, following the experimental procedures described in [14,15]. Detailed methodology and HDX-MS data are available in the ProteomeXchange (PX) Consortium [16] via the Proteomics Identifications (PRIDE) [17] repository (Dataset identifier: PXD060074; Username: reviewer_px[email protected]; Password: LyazwVQgkNEC). Circular dichroism (CD) spectroscopy Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 5 Far-UV CD spectra of MDMX-S403D FL and S403D (322-490) were recorded using a Jasco J-1500 Spectrometer (Jasco Inc., Easton, MD, USA) in a wavelength range of 180 to 280 nm. Measurements were performed at a protein concentration of 0.5 mg/ml in phosphate buffer (pH 7.5), using quartz cuvettes with a 0.1 cm path length. Intrinsic fluorescence The emission fluorescence spectra of the proteins (at 0.5 mg/ml concentration) in phosphate buffer (pH 7.5) or binding buffer were recorded using an Agilent Technologies Cary Eclipse Fluorescence Spectrophotometer (Agilent Technologies, Santa Clara, CA, USA). Measurements were performed at room temperature in quartz cuvettes with a 1 cm path length. The excitation wavelength was set to 280 nm, and blank measurements (without protein) were recorded and subtracted from the experimental spectra. Results Domain-specific contributions to the RNA-binding activity of the protein To characterize the RNA-binding properties of MDMX, we prepared recombinant proteins in both the wild-type (WT) form and a phosphomimetic mutation at serine 403 (S403D), which enhances affinity for TP53 mRNA [8]. For the S403D mutant, we generated four MDMX variants: full-length (FL; 1–490), a truncated construct lacking the N-terminal domain (128– 490), a construct lacking the C-terminal region including the RING domain (1–436), and a Cterminal fragment containing only the RING domain (322–490). In addition, two WT constructs were produced: full-length (FL; 1–490) and the RING-containing C-terminal variant (322–490) (Figure 1A). All recombinant proteins were analysed using EMSA. The TP53 mRNA fragment (1–240), corresponding to the first 240 nucleotides of the full-length p53 coding sequence, was synthesized by in vitro transcription. This region includes the internal ribosome entry site (IRES) for the p53/47 isoform and has previously been identified as the MDMX-interacting region [8]. Binding assays were performed using 100 ng of TP53 mRNA (nucleotides 1–240), hereafter referred to as “RNA”, and increasing concentrations of recombinant MDMX proteins. The RNA-protein formation was indicated as a reduction of free RNA and for protein variants FL (Figure 1B and 1F) and truncation 128-490 (Figure 1C) as a shifted gel band. The C-terminal constructs 322-490 (Figure 1E) migrated poorly into gel, often appearing as a smear or remaining trapped in the gel wells. This behavior is likely due to the higher pI and structural properties of the construct, although the possibility of aggregation cannot be entirely ruled out. Moreover, protein conformational changes may reduce RNA accessibility in complex, thereby hindering its detectability by staining under Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 6 certain conditions. Therefore, we evaluated the reduction of free RNA rather than the signal intensity of the protein–RNA complex to compare the RNA-binding affinity of all protein variants (Figure 1H), as well as protein-RNA affinity under other conditions (shown in later figures). As expected, in the absence of protein (lane 1:0), only free RNA was detected, serving as a reference for the unbound RNA signal (Figure 1B-G). MDMX-S403D FL (Figure 1B) showed enhanced binding to TP53 mRNA, as evidenced by a reduction in free RNA and the appearance of a shifted RNA–protein complex at a 1:25 RNA:protein molar ratio, with near-complete formation of the complex observed at 1:100. The S403D variant lacking the N-terminal domain, MDMX-S403D (128–490), exhibited reduced RNA binding (Figure 1C), suggesting that the N-terminal region modulates or facilitates the RNA interaction. The MDMX-S403D (1–436) variant, which lacks the RING domain, did not bind TP53 mRNA at the tested concentrations (Figure 1D), indicating that the RING domain is essential for RNA interaction. In contrast, the C-terminal variant MDMX-S403D (322–490), containing the RING domain, reduced free RNA across all tested RNA:protein ratios, indicating higher affinity than full-length MDMX-S403D (Figure 1E), as previously shown by RNA-ELISA [8]. However, smear rather than discrete bands are visible as complexes, leaving open the possibility that protein aggregation contributes to the observed binding. We also compared RNA binding of the two WT forms of MDMX, full-length (1–490) (Figure 1F) and the RINGcontaining C-terminal variant (322–490) (Figure 1G), and confirmed previous findings [8] that WT proteins bind TP53 mRNA less efficiently than their phosphomimetic counterparts (Figure 1B and 1E). Although potential aggregation of the C-terminal variant constructs may contribute to the observed binding pattern, the findings nonetheless suggest that the RING domain plays an important role in the MDMX–TP53 mRNA interaction. The S403D phosphomimetic mutation enhances binding affinity, whereas other domains potentially mask or modulate the RNA-binding interface. Mapping the RNA interaction site of the MDMX protein MDMX is a highly flexible protein due to disordered regions between its structured domains [12] implying that certain protein or RNA interactions may exert allosteric effects that regulate its interactome [18]. To identify the RNA-binding site on MDMX-S403D FL (1-490) and to study its allosteric effect, we performed HDX-MS at three time intervals: 10 s (Figure S1A), 120 s (Figure 2A), and 1 800 s (Figure S1B). By this method, direct nucleic acid - protein interactions are seen as deuteration suppression, and rearrangements of protein structure due to ligand binding are seen as increased deuteration for structure opening or decreased deuteration for structure closing. Our results showed no specific RNA-binding site within the RING domain, but small deuteration changes were observed throughout the entire protein, with the most noticeable opening in the N-terminal domain (100-120 aa) (Figure 2A), Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 7 supporting the notion that the entire protein participates, either directly or indirectly, in RNA binding. The RING domain (440–490) exhibits low deuteration (Figure 2A), suggesting masking by other protein domains. The inability to identify a direct RNA contact site may reflect limitations of the experimental settings, including restricted accessibility of the RING domain under the assay conditions, the potential absence of a well-defined binding interface, or the inherent instability, transient nature, and possible non-specificity of the interaction, all of which could prevent reliable detection. We used increasing salt concentrations in the binding assay to elucidate the nature of the interaction between MDMX and RNA. The C-terminal construct MDMX-S403D (322–490) reduced nearly all available RNA at 150 mM NaCl (Figure 2B, lane 2, quantified in Figure 2D) occurring along with the smear on the gel that suggests RNA-protein complex formation. However, as the salt concentration increased (starting at 250 mM NaCl), the amount of free RNA also increased, suggesting that higher salt concentration disrupted electrostatic interactions required for RNA–protein complex formation (Figure 2B, lanes 4, 6 and 8). In contrast, the RNA–protein complexes formed by MDMX-S403D FL were resilient to increasing salt concentrations (150–600 mM), remaining stable even at higher ionic strength (Figure 2C, quantified in Figure 2D). This suggests that the interaction is supported by additional, non-electrostatic forces. Since salt concentration may influence the binding efficiency of GelRed stain to RNA (Figure S2), each EMSA gel included free RNA controls loaded at the corresponding NaCl concentration. Altogether, these findings suggest that the C-terminal construct domain binds RNA via electrostatic forces that are not detectable by HDX-MS, while other MDMX domains may partially shield the RING interface and contribute to stabilization of the RNA-protein complex. The effect of pH on the binding of MDMX-S403D to RNA To evaluate further the electrostatic contribution of protein-RNA interaction, we examined how the variations in pH (7.0, 7.5, and 8.0) will influence the RNA binding. The MDMXS403D FL protein has an isoelectric point (pI) of 5.1, whereas the C-terminal construct MDMX-S403D (322–490) has a pI of 7.6. The pH shift from 7.0 to 8.0 primarily affects cysteine and histidine residues located mainly in the C-terminus region of MDMX, altering their protonation state and resulting in a more negative overall charge (Figure S3) [19,20]. For the MDMX-S403D FL construct (Figure 3A, quantified in 3C), no significant difference in RNA binding was observed across the tested pH range (Figure 3A, lanes 2-4, upper panel). RNA alone exhibited a consistent migration pattern at pH 7.5 (lane 1) and pH 8.0 (lane 8). The MDMX-S403D (322–490) construct (Figure 3B, quantified in 3C) showed small but consistent reduction in RNA binding as pH increased (Figure 3B, lanes 2-4, upper panel), corresponding to reduced protonation at higher pH values leading to increased negative Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 8 charge and weakened electrostatic interactions. To determine whether altered protein migration contributed to the observed effects, we stained the agarose gels with SYPRO Ruby (Figure 3A and 3B, lower panels). Across all tested pH conditions, the migration of both constructs remained unchanged, whether in free form (lanes 5–7) or as part of an RNA–protein complex (lanes 2–4), indicating that differences in EMSA results are not due to altered protein mobility. Moreover, these results show that the C-terminal truncation (322– 490) alone penetrates poorly into the gel wells (Figure 3B, lower panel, lanes 5-7) and is only drawn efficiently into the gel in the presence of RNA (lane 2-4), appearing as a smear, whereas the full-length protein penetrates the gel more efficiently and forms discrete complexes visible as a band (Figure 3A, lower panel). These results reflect the electrostatic nature of RNA binding by the C-terminal region and highlight the role of the entire protein in stabilizing the RNA interaction across a broader pH range. Conditions influencing accessibility of the MDMX RING domain regulate RNA binding We hypothesized that MDMX binds RNA via its RING domain, with other domains potentially masking the interaction interface. To characterize the structural features within the RING domain that are critical for RNA interaction, we examined the role of zinc coordination and disulfide bond integrity in MDMX-S403D FL binding. The RING domain of MDMX consists of a C2H2C4 structure with two cysteines (C1 and C2), two histidines (H3 and H4), and four cysteines (C5–C8), which coordinate two Zn²⁺ions essential for stabilizing its folded state [21,22]. Moreover, MDMX contains an additional Zn-finger domain spanning residues 290– 322. To investigate the role of zinc ions and their chelation by EDTA in RNA binding, we performed EMSA using the MDMX-S403D FL protein (Figure 4A, quantified in 4B) and the C-terminal variant MDMX-S403D (322–490) (Figure S4A, quantified in S4B) in the presence of zinc concentrations promoting proper protein folding (10 µM Zn²⁺), an excess of free Zn²⁺ (100 µM), and 1 mM EDTA combined with varying zinc concentrations. For MDMX-S403D FL, excess zinc ions revealed decrease in RNA binding seen as free RNA increase (Figure 4A, upper panel, lane 3). In contrast, the addition of EDTA enhanced RNA binding significantly, suggesting that zinc chelation by EDTA facilitates interaction with RNA (Figure 4A, upper panel, lane 4 and 5, quantified in Figure 4B). Similar trends in RNA changes were observed for the MDMX-S403D (322–490) variant (Figure S4A and S4B), and upon EDTA addition, this was accompanied by disappearance of the smear that had indicated RNA– protein complex formation (Figure S4A, upper panel, lines 4 a 5). It has been proposed that EDTA may induce unfolding and potential aggregation of RING finger domains [23]. To examine this effect, we used SYPRO Ruby staining to evaluate protein mobility in agarose gels, in both RNA-bound and RNA-free forms. MDMX-S403D FL formed two bands even in Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 9 the absence of RNA (Figure 4A, lower panel, lanes 6-8). No significant change in mobility was observed upon Zn²⁺treatment (lanes 3 and 7), indicating no major structural changes occurred. However, in the presence of 1 mM EDTA, we observed reduction of the lower protein band in both RNA-bound (lanes 4 and 5) and RNA-free forms (lane 8), suggesting a structural alteration of the protein. By overlaying GelRed and SYPRO Ruby staining, we identified the upper band as the RNA-binding form (Figure S4E), consistent with EDTAinduced structural rearrangement that promotes RNA interaction. The C-terminal variant MDMX-S403D (322–490) showed no mobility changes in response to zinc ions (Figure S4A, lower panel, lanes 2, 3, 6 and 7). However, addition of 1 mM EDTA induced accumulation of the protein in the gel wells accompanied by the disappearance of the smear, suggesting protein aggregation (Figure S4A, lower panel, lanes 4, 5, and 8). Together, these results indicate that zinc ions interfere with RNA binding, while EDTA enhances binding, presumably by chelating zinc and unmasking the interface. However, EDTA probably weakens the structural stability of the protein, particularly of the C-terminal domain. These findings highlight the delicate balance between structural integrity and accessibility in the regulation of RNA binding by the MDMX RING domain. The MDMX protein contains cysteine residues, primarily located in the zinc finger and RING domains, which can form disulfide bonds involved in protein folding and oligomerization. However, under reducing conditions prevalent in the cellular environment, disulfide bonds become unstable. To investigate the role of disulfide bonds in RNA binding, we used increasing concentration of TCEP, a potent reducing agent, and analysed RNA-binding capacity using EMSA (Figure 4C and quantified in 4D). RNA–protein complex formation was observed under native conditions without TCEP (lane 2) and increased upon treatment with 1 mM TCEP (lane 3). However, at 5 mM TCEP, complex formation by MDMX-S403D FL was disrupted, indicated by the reappearance of free RNA (lane 4). SYPRO Ruby confirmed that 1 mM TCEP did not alter the migration of MDMX-S403D FL (Figure 4C, lower panel, lane 3), whereas 5 mM TCEP probably caused aggregation, with protein retained in the well (lane 4). Similar aggregation was observed for the RNA-free protein at 5 mM TCEP (lane 7). For MDMX-S403D (322–490) (Figure S4C and quantified in S4D), we observe decrease in free RNA upon 1mM and 5 mM TCEP addition. However, the SYPRO staining reveals (Figure S4C, lower panel) that 1 mM TCEP enhanced gel penetration of both RNA-bound and RNAfree forms (lanes 3 and 6), indicating disulfide bond disruption. In contrast, 5 mM TCEP again led to protein retention in the well suggesting aggregation (lanes 4 and 7). These results suggest that excessive disulfide bond formation can partially obstruct RNA binding, whereas their complete absence disrupts protein integrity. In summary, disulfide bonds are essential for both the RNA-binding ability and the structural stability of MDMX. Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 16 probably regulated by an induced fit mechanism. Bioscience Reports 42, BSR20220186 https://doi.org/10.1042/BSR20220186 13Müller, P., Ceskova, P. and Vojtesek, B. (2005) Hsp90 Is Essential for Restoring Cellular Functions of Temperature-sensitive p53 Mutant Protein but Not for Stabilization and Activation of Wild-type p53. Journal of Biological Chemistry 280, 6682–6691 https://doi.org/10.1074/jbc.M412767200 14Uhrik, L., Wang, L., Haronikova, L., Medina-Medina, I., Rebolloso-Gomez, Y., Chen, S., et al. (2019) Allosteric changes in HDM2 by the ATM phosphomimetic S395D mutation: implications on HDM2 function. Biochemical Journal 476, 3401–3411 https://doi.org/10.1042/BCJ20190687 15Nekulová, M., Wyszkowska, M., Friedlová, N., Uhrík, L., Zavadil Kokáš, F., Hrabal, V., et al. (2024) Biochemical evidence for conformational variants in the anti-viral and prometastatic protein IFITM1. Biological Chemistry 405, 311–324 https://doi.org/10.1515/hsz2023-0327 16Deutsch, E.W., Bandeira, N., Perez-Riverol, Y., Sharma, V., Carver, J.J., Mendoza, L., et al. (2023) The ProteomeXchange consortium at 10 years: 2023 update. Nucleic Acids Research 51, D1539–D1548 https://doi.org/10.1093/nar/gkac1040 17Perez-Riverol, Y., Bai, J., Bandla, C., García-Seisdedos, D., Hewapathirana, S., Kamatchinathan, S., et al. (2022) The PRIDE database resources in 2022: a hub for mass spectrometry-based proteomics evidences. Nucleic Acids Research 50, D543–D552 https://doi.org/10.1093/nar/gkab1038 18Medina‐Medina, I., Martínez‐Sánchez, M., Hernández‐Monge, J., Fahraeus, R., Muller, P. and Olivares‐Illana, V. (2018) p53 promotes its own polyubiquitination by enhancing the HDM2 and HDMX interaction. Protein Science 27, 976–986 https://doi.org/10.1002/pro.3405 19University of Freiburg Patchwork: Visualization of protein sequences with functional features https://patchwork.biologie.uni-freiburg.de/index.php 20Hernychova, L. (2025) Schematic representation of predicted charge distribution changes along the MDMX protein. BioRender. https://BioRender.com/9gxyfjv 21Linke, K., Mace, P.D., Smith, C.A., Vaux, D.L., Silke, J. and Day, C.L. (2008) Structure of the MDM2/MDMX RING domain heterodimer reveals dimerization is required for their ubiquitylation in trans. Cell Death Differ 15, 841–848 https://doi.org/10.1038/sj.cdd.4402309 22Kostic, M., Matt, T., Martinez-Yamout, M.A., Dyson, H.J. and Wright, P.E. (2006) Solution Structure of the Hdm2 C2H2C4 RING, a Domain Critical for Ubiquitination of p53. Journal of Molecular Biology 363, 433–450 https://doi.org/10.1016/j.jmb.2006.08.027 23Nyborg, J.K. and Peersen, O.B. (2004) That zincing feeling: the effects of EDTA on the behaviour of zinc-binding transcriptional regulators. Biochemical Journal 381, e3 https://doi.org/10.1042/BJ20041096 24Medina-Medina, I., García-Beltrán, P., De La Mora-de La Mora, I., Oria-Hernández, J., Millot, G., Fahraeus, R., et al. (2016) Allosteric Interactions by p53 mRNA Govern HDM2 E3 Ubiquitin Ligase Specificity under Different Conditions. Molecular and Cellular Biology 36, 2195–2205 https://doi.org/10.1128/MCB.00113-16 Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 17 25Almeida, F.C.L., Sanches, K., Pinheiro-Aguiar, R., Almeida, V.S. and Caruso, I.P. (2021) Protein Surface Interactions—Theoretical and Experimental Studies. Front. Mol. Biosci. 8, 706002 https://doi.org/10.3389/fmolb.2021.706002 26Chen, S.-J. (2008) RNA Folding: Conformational Statistics, Folding Kinetics, and Ion Electrostatics. Annu. Rev. Biophys. 37, 197–214 https://doi.org/10.1146/annurev.biophys.37.032807.125957 27Haronikova, L., Olivares-Illana, V., Wang, L., Karakostis, K., Chen, S. and Fåhraeus, R. (2019) The p53 mRNA: an integral part of the cellular stress response. Nucleic Acids Research 47, 3257–3271 https://doi.org/10.1093/nar/gkz124 28Popowicz, G., Czarna, A. and Holak, T. (2008) Structure of the human Mdmx protein bound to the p53 tumor suppressor transactivation domain. Cell Cycle 7, 2441–2443 https://doi.org/10.4161/cc.6365 29Pazgier, M., Liu, M., Zou, G., Yuan, W., Li, C., Li, C., et al. (2009) Structural basis for high-affinity peptide inhibition of p53 interactions with MDM2 and MDMX. Proc. Natl. Acad. Sci. U.S.A. 106, 4665–4670 https://doi.org/10.1073/pnas.0900947106 30Huang, Y., Li, W., Zhou, Y., Bai, J., Li, N., Su, Z., et al. (2025) Strategies for p53 Activation and Targeted Inhibitors of the p53-Mdm2/MdmX Interaction. Cells 14, 583 https://doi.org/10.3390/cells14080583 31Pairawan, S., Zhao, M., Yuca, E., Annis, A., Evans, K., Sutton, D., et al. (2021) First in class dual MDM2/MDMX inhibitor ALRN-6924 enhances antitumor efficacy of chemotherapy in TP53 wild-type hormone receptor-positive breast cancer models. Breast Cancer Res 23, 29 https://doi.org/10.1186/s13058-021-01406-x 32Wu, W., Xu, C., Ling, X., Fan, C., Buckley, B.P., Chernov, M.V., et al. (2015) Targeting RING domains of Mdm2–MdmX E3 complex activates apoptotic arm of the p53 pathway in leukemia/lymphoma cells. Cell Death Dis 6, e2035–e2035 https://doi.org/10.1038/cddis.2015.358 33Lama, R., Xu, C., Galster, S.L., Querol-García, J., Portwood, S., Mavis, C.K., et al. (2022) Small molecule MMRi62 targets MDM4 for degradation and induces leukemic cell apoptosis regardless of p53 status. Front. Oncol. 12, 933446 https://doi.org/10.3389/fonc.2022.933446 34Holmstrom, E.D., Liu, Z., Nettels, D., Best, R.B. and Schuler, B. (2019) Disordered RNA chaperones can enhance nucleic acid folding via local charge screening. Nat Commun 10, 2453 https://doi.org/10.1038/s41467-019-10356-0 35Woodson, S.A., Panja, S. and Santiago-Frangos, A. (2018) Proteins That Chaperone RNA Regulation. Microbiol Spectr 6, 6.4.21 https://doi.org/10.1128/microbiolspec.RWR0026-2018 36Hernández-Monge, J., Rousset-Roman, A.B., Medina-Medina, I. and Olivares-Illana, V. (2016) Dual function of MDM2 and MDMX toward the tumor suppressors p53 and RB. Genes Cancer 7, 278–287 https://doi.org/10.18632/genesandcancer.120 Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 18 Figure 1: TP53 mRNA binding to phosphomimetic and wild-type MDMX protein constructs. (A) Schematic representation of MDMX protein constructs carrying a phosphomimetic mutation at serine 403 (MDMX-S403D), shown in four variants: full-length MDMX-S403D FL (1–490), a truncated form lacking the N-terminal domain MDMX-S403D (128–490), a construct lacking the C-terminal region including the RING domain MDMX-S403D (1–436), and a C-terminal fragment containing only the RING domain MDMX-S403D (322–490). The domain architecture includes the p53-binding domain (p53BD, pink), WWW domain (blue), acidic domain (AD, green), zinc-binding domain (ZD, red), and the RING domain (yellow). (B–E) EMSA analysis of TP53 mRNA binding to the four MDMX-S403D variants shown in (A). (F, G) EMSA analysis of TP53 mRNA binding to wild-type MDMX constructs: (F) fulllength MDMX WT FL (1–491) and (G) the C-terminal RING-containing fragment MDMX WT (322–490). All recombinant proteins were incubated with in vitro transcribed TP53 mRNA (nucleotides 1–240) at RNA:protein molar ratios of 1:25, 1:50, and 1:100. Free RNA (ratio 1:0) served as control. Reactions were performed in binding buffer (150 mM NaCl, 50 mM Tris, 1 mM TCEP, 10 µM ZnSO₄, pH 7.5). RNA–protein complexes and free RNA were visualized on agarose gels by GelRed staining. Brackets indicate smears corresponding to protein–TP53 mRNA complexes. (H) Quantification of free RNA band intensity from EMSA gels shown in (B–G). For each RNA:protein ratio (1:25, 1:50, 1:100), free RNA intensity was normalized to the RNA-only control (1:0). Colored curves represent individual constructs: (B) MDMX-S403D FL (1–490), red; (C) MDMX-S403D (128–490), blue; (D) MDMX-S403D (1– 436), green; (E) MDMX-S403D (322–490), purple; (F) MDMX WT FL (1–490), black; (G) MDMX WT (322–490), grey. Data represent mean ± SEM (n = 3 biological replicates). Figure 2: Characterization of the MDMX-S403D interaction with TP53 mRNA by HDXMS and EMSA under varying salt concentrations. (A) HDX-MS analysis of full-length MDMX carrying a phosphomimetic mutation at serine 403 (MDMX-S403D FL) in complex with TP53 mRNA. Relative fractional uptake across amino acid residues is shown for free MDMX-S403D FL (orange) and the MDMX-S403D FL–TP53 mRNA complex (blue) after 120 s deuterium exposure. Non-exchangeable proline residues are indicated as gaps in the profile. (B, C) EMSA analysis of TP53 mRNA binding to (B) MDMX-S403D (322–490) and (C) MDMX-S403 FL under increasing NaCl concentrations (150, 250, 400, and 600 mM). Protein–RNA complexes (lanes 2, 4, 6, 8) and corresponding free RNA controls (lanes 1, 3, 5, 7) were visualized on agarose gels by GelRed at an RNA:protein molar ratio of 1:50. Brackets indicate smears corresponding to protein–TP53 mRNA complexes. (D) Quantification of free RNA band intensity derived from the EMSA gels in (B) and (C). Data for MDMX-S403D FL are shown as empty columns (from gel C) and for MDMX-S403D (322–490) as dotted columns (from gel B), normalized to their respective RNA-only controls. Data represent mean ± SEM (n = 3 biological replicates; ***p < 0.001). Additional statistical comparisons are shown in the figure and alternatively indicated as •p < 0.05 and •••p < 0.001. Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 19 Figure 3: Effect of pH on TP53 mRNA-MDMX interaction. (A, B) EMSA showing the effect of pH (7.0, 7.5, and 8.0) on TP53 mRNA binding to (A) MDMX-S403D FL and (B) MDMX-S403D (322–490). Proteins were incubated with TP53 mRNA (lanes 2–4) or without RNA (lanes 5–7) in binding buffer, adjusted to the indicated pH values. Free RNA controls were included at pH 7.5 and 8.0 (lanes 1 and 8). The RNA:protein molar ratio was 1:50 in all reactions. Brackets indicate smears corresponding to protein– TP53 mRNA complexes. RNA and RNA–protein complexes were visualized on GelRedstained agarose gels (upper panels), and total protein was detected using SYPRO Ruby staining of the same gels (lower panels). (C) Quantification of free RNA band intensity derived from the EMSA gels in (A) and (B). Data for MDMX-S403 FL are shown as white columns (from gel A) and for MDMX-S403D (322–490) as grey columns (from gel B), normalized to their respective RNA-only control. Data represent mean ± SEM (n = 3 biological replicates; *p < 0.05, ***p < 0.001). Additional statistical comparisons are shown in the figure and alternatively indicated as •p < 0.05. Figure 4: Modulation of TP53 mRNA–MDMX-S403D FL interaction by zinc, EDTA, and reducing conditions. (A) EMSA analysis of TP53 mRNA binding to MDMX-S403D FL construct in the presence of Zn²⁺and EDTA. Proteins were incubated with TP53 mRNA (lanes 2–5) or without RNA (lanes 6–8) in binding buffer (150 mM NaCl, 50 mM Tris, 1 mM TCEP, pH 7.5) and either 10 µM or 100 µM ZnSO₄, in the presence of EDTA (dotted columns) or absence of EDTA (empty columns). Free RNA (lane 1) served as a control. (B, D) Quantification of free RNA band intensity from gels shown in (A) and (C), respectively, normalized to their corresponding RNA-only control. Data represent mean ± SEM (n = 3 biological replicates; ***p < 0.001). Additional statistical comparisons are shown in the figure and alternatively indicated as •p < 0.05 and ••p < 0.01. (C) EMSA showing the effect of the reducing agent TCEP on TP53 mRNA to MDMX-S403D FL construct. Proteins were incubated with TP53 mRNA (lanes 2–4) or without RNA (lanes 5–7) in binding buffer containing 0, 5, or 10 mM TCEP. Free RNA (lane 1) served as a control. In panels (A) and (C), RNA–protein complexes were visualized by GelRed staining (upper panels), and total protein was detected using SYPRO Ruby staining of the same gels (lower panels). The RNA:protein molar ratio was 1:50 in all reactions. Figure 5: Effects of TCEP and EDTA on conformation of MDMX-S403D FL. (A) CD spectroscopy of MDMX-S403D FL measured in 10 mM sodium phosphate buffer (pH 7.5, without NaCl) across the wavelength range 180–280 nm. (B) Intrinsic fluorescence emission spectra of MDMX-S403D FL measured in RNA-binding buffer (150 mM NaCl, 20 mM Tris-HCl, pH 7.5) upon excitation at 280 nm. In both (A) and (B), measurements were performed under four conditions: untreated (black line), with 1 mM TCEP (black dashed line), with 5 mM TCEP (grey line), and with a combination of 1 mM EDTA and 1 mM TCEP (red line). Figure S1: HDX-MS analysis of MDMX-S403D FL interaction with TP53 mRNA. HDX-MS analysis of full-length MDMX carrying a phosphomimetic mutation at serine 403 (MDMX-S403D FL) in complex with TP53 mRNA. Relative fractional uptake across amino Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 20 acid residues is shown for free MDMX-S403D FL (orange) and the MDMX-S403D FL–TP53 mRNA complex (blue) after 10 s (A) and 1800 s (B) of deuterium exposure. Nonexchangeable proline residues are indicated as gaps in the profiles. Figure S2: Effect of NaCl concentration on free TP53 mRNA mobility. EMSA analysis of free TP53 mRNA in binding buffer and either 150 mM or 600 mM NaCl, as indicated. No protein was added to the reaction. RNA was visualized on agarose gels by GelRed staining. Figure S3: Schematic representation of pH-dependent charge distribution in MDMX. Schematic representation of predicted charge distribution changes along the MDMX protein sequence at pH 7.0 versus pH 8.0. Positively charged residues are shown in blue, negatively charged residues in red, and amino acids affected by the pH shift are highlighted in yellow. The amino acid sequence of MDMX and its predicted structural changes under varying pH conditions were visualized using Patchwork (https://patchwork.biologie.uni-freiburg.de/). The final schematic representation was then prepared in BioRender. Available from: https://BioRender.com/9gxyfjv. Agreement number: ED28BFG4CD, License number: 66587182. Figure S4: Modulation of TP53 mRNA–MDMX-S403D (322-490) interaction by zinc, EDTA, and reducing conditions. (A) EMSA analysis of TP53 mRNA binding to MDMX-S403D (322-490) construct in the presence of Zn²⁺and EDTA. Proteins were incubated with TP53 mRNA (lanes 2–5) or without RNA (lanes 6–8) in binding buffer (150 mM NaCl, 50 mM Tris, 1 mM TCEP, pH 7.5) and either 10 µM or 100 µM ZnSO₄, in the presence of EDTA (dotted columns) or absence of EDTA (empty columns). Free RNA (lane 1) served as a control. (B, D) Quantification of free RNA band intensity from gels shown in (A) and (C), respectively, normalized to their corresponding RNA-only control. Data represent mean ± SEM (n = 3 biological replicates; *p < 0.05, **p < 0.01). Additional statistical comparisons are shown in the figure and alternatively indicated as •p < 0.05. (C) EMSA showing the effect of the reducing agent TCEP on TP53 mRNA to MDMX-S403D (322-490) construct. Proteins were incubated with TP53 mRNA (lanes 2–4) or without RNA (lanes 5–7) in binding buffer containing 0, 5, or 10 mM TCEP. Free RNA (lane 1) served as a control. In panels (A) and (C), brackets indicate smears corresponding to protein–TP53 mRNA complexes. RNA–protein complexes were visualized by GelRed staining (upper panels), and total protein was detected using SYPRO Ruby staining of the same gels (lower panels). The RNA:protein molar ratio was 1:50 in all reactions. (E) Overlay of EMSA gel stained with GelRed and SYPRO Ruby showing TP53 mRNA binding to MDMX-S403D FL in the presence of Zn²⁺and EDTA, corresponding to (A). The upper band corresponds to the RNA–protein complex, while the lower band represents unbound protein. Figure S5: Effects of TCEP and EDTA on conformation of MDMX-S403D (322–490). Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 21 (A) CD spectroscopy of MDMX-S403D (322-490) measured in 10 mM sodium phosphate buffer (pH 7.5, without NaCl) across the wavelength range 180–280 nm. (B) Intrinsic fluorescence emission spectra of MDMX-S403D FL measured in RNA-binding buffer (150 mM NaCl, 20 mM Tris-HCl, pH 7.5) upon excitation at 280 nm. In both (A) and (B), measurements were performed under four conditions: untreated (black line), with 1 mM TCEP (black dashed line), with 5 mM TCEP (grey line), and with a combination of 1 mM EDTA and 1 mM TCEP (red line). (C) CD spectroscopy of MDMX-S403D FL measured in 10 mM sodium phosphate buffer (pH 7.5, without NaCl) across a wavelength range of 180– 280 nm. The spectra show the protein in the absence of EDTA (black line) and in the presence of 1 mM EDTA (red line). (D) EMSA analysis of TP53 mRNA binding MDMXS403D FL in the presence of TCEP and EDTA. Reactions were carried out at an RNA:protein molar ratio of 1:50 in binding buffer containing 1 mM TCEP (lane 3), 1 mM EDTA (lane 4), or a combination of both (lane 5). Free RNA (lane 1) and untreated protein (lane 2) served as controls. RNA and RNA–protein complexes were visualized on GelRedstained agarose gels (upper panel), and total protein was detected using SYPRO Ruby staining of the same gel (lower panel). (E) Quantification of free RNA band intensity for MDMX-S403 FL treated with only TCEP (empty columns) and with a combination of TCEP and EDTA (dotted columns), normalized to the RNA-only control. Data represent mean ± SEM (n = 3 biological replicates; *p < 0.05, **p < 0.01). Additional statistical comparisons are shown in the figure and alternatively indicated as •p < 0.05 and •••p < 0.001. Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 22 Figure 1 Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 23 Figure 2 Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 24 Figure 3 Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646 25 Figure 4 Downloaded from http://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20253646/980042/bsr-2025-3646.pdf by guest on 21 November 2025 Bioscience Reports. This is an Accepted Manuscript. You are encouraged to use the Version of Record that, when published, will replace this version. The most up-to-date-version is available at https://doi.org/10.1042/BSR20253646