Nathalie Evrard-Todeschi

Structural and Functional Characterization of Nrf2 Degradation by the Glycogen Synthase Kinase 3/β-TrCP Axis

ABSTRACT The transcription factor NF-E2-related factor 2 (Nrf2) is a master regulator of a genetic program, termed the phase 2 response, that controls redox homeostasis and participates in multiple aspects of physiology and pathology. Nrf2 protein stability is regulated by two E3 ubiquitin ligase adaptors, Keap1 and β-TrCP, the latter of which was only recently reported. Here, two-dimensional (2D) gel electrophoresis and site-directed mutagenesis allowed us to identify two serines of Nrf2 that are phosphorylated by glycogen synthase kinase 3β (GSK-3β) in the sequence DSGISL. Nuclear magnetic resonance studies defined key residues of this phosphosequence involved in docking to the WD40 propeller of β-TrCP, through electrostatic and hydrophobic interactions. We also identified three arginine residues of β-TrCP that participate in Nrf2 docking. Intraperitoneal injection of the GSK-3 inhibitor SB216763 led to increased Nrf2 and heme oxygenase-1 levels in liver and hippocampus. Moreover, mice with hippocampal absence of GSK-3β exhibited increased levels of Nrf2 and phase 2 gene products, reduced glutathione, and decreased levels of carbonylated proteins and malondialdehyde. This study establishes the structural parameters of the interaction of Nrf2 with the GSK-3/β-TrCP axis and its functional relevance in the regulation of Nrf2 by the signaling pathways that impinge on GSK-3.

Phosphorylation-dependent structure of ATF4 peptides derived from a human ATF4 protein, a member of the family of transcription factors☆

ATF4 plays a crucial role in the cellular response to stress and the F-box protein beta-TrCP, the receptor component of the SCF E3 ubiquitin ligase responsible for ATF4 degradation by the proteasome, binds to ATF4, and controls its stability. Association between the two proteins depends on ATF4 phosphorylation of serine residues 219 and 224 present in the context of DpSGXXXpS, which is similar but not identical to the DpSGXXpS motif found in most other substrates of beta-TrCP. We used NMR spectroscopy to analyze the structure of the 23P-ATF4 peptide. The 3D structure of the ligand was determined on the basis of NOESY restraints that provide an hairpin loop structure. In contrast, no ordered structure was observed in the NMR experiments for the nonphosphorylated 23-ATF4 in solution. This structural study provides information, which could be used to study the beta-TrCP receptor-ligand interaction in docking procedure. Docking studies showed that the binding epitope of the ligand, is represented by the DpSGIXXpSXE motif. 23P-ATF4 peptide fits the binding pocket of protein beta-TrCP very well, considering that the DpSGIXXpSXE motif adopts an S-turning conformation contrary to the extended DpSGXXpS motif in the other known beta-TrCP ligands.

Contact-based ligand-clustering approach for the identification of active compounds in virtual screening

Evaluation of docking results is one of the most important problems for virtual screening and in silico drug design. Modern approaches for the identification of active compounds in a large data set of docked molecules use energy scoring functions. One of the general and most significant limitations of these methods relates to inaccurate binding energy estimation, which results in false scoring of docked compounds. Automatic analysis of poses using self-organizing maps (AuPosSOM) represents an alternative approach for the evaluation of docking results based on the clustering of compounds by the similarity of their contacts with the receptor. A scoring function was developed for the identification of the active compounds in the AuPosSOM clustered dataset. In addition, the AuPosSOM efficiency for the clustering of compounds and the identification of key contacts considered as important for its activity, were also improved. Benchmark tests for several targets revealed that together with the developed scoring function, AuPosSOM represents a good alternative to the energy-based scoring functions for the evaluation of docking results.

Structure of the Complex between Phosphorylated Substrates and the SCF β-TrCP Ubiquitin Ligase Receptor: A Combined NMR, Molecular Modeling, and Docking Approach

The binding of phosphorylated peptides to the receptor plays a major role in many basic cellular processes in a variety of pathological states. Human beta-TrCP is a key component of a recently characterized E3 ubiquitin ligase complex that regulates protein degradation through the ubiquitin-dependent proteasome pathway. Docking studies were carried out to explore the structural requirements for the beta-TrCP substrates. Docking studies were performed on the bound conformation of the phosphorylated peptides determined by NMR, whereas the beta-TrCP structure was derived by X-ray from Protein Data Bank. After the docking calculation, during which the peptides were conformationally restrained, the complex presented herein was analyzed in terms of ligand-protein interactions and properties of contacting surfaces. The structural requirements for phosphorylated substrates in interaction with beta-TrCP were explored and compared with experimental data from TRNOESY and STD NMR results. The analysis revealed that the bend of the peptide structures, which is indispensable for beta-TrCP recognition, aligns two charged phosphate groups and a central hydrophobic group in a favorable arrangement that leads to the burial of the peptide surface in the binding cleft upon complexation. Through docking simulations, we have identified different specific binding pockets of beta-TrCP according to the ligand in interaction. These data should be valuable in the rational design of a ligand to be used in therapeutic approaches.

Conformational analysis of josamycin, a 16-membered macrolide free in solution and bound to bacterial ribosomes

The potent 14-membered macrolide antibiotics displayed a strong NMR response in the TRNOESY experiments whereas their metabolites which do not retain antimicrobial activity gave essentially blank TRNOESY spectra. These TRNOESY experiments are here extended to a macrolide (josamycin) belonging to the 16-atom macrolide class in order to compare responses according to the macrocycle size. Analysis of transferred nuclear Overhauser effect (TRNOE) experiments resulted in a set of constraints for all proton pairs. These constraints were used in structure determination procedures based on molecular modelling to obtain a bound structure compatible with the experimental NMR data. This study allowed us to identify the binding structure at the ribosome for the active 16-membered macrolide. The different conformations existing in solution for this antibiotic in the free state will be compared and correlated with the binding conformation at the ribosome obtained with TRNOE experiments so as to establish a structure–activity relationship. The comparative results indicate that one conformation (called S3) pre-existing in the conformational equilibrium of macrocycle in solution and close to the major one S5, is preferred in the bound state.

Conformational analysis of 2-(carboxycyclopropyl)glycine agonists of glutamate receptors in aqueous solution using a combination of NMR and molecular modelling experiments and charge calculations

Two classes of glutamate receptors [metabotropic (group-II) and ionotropic (NMDA) subclasses] are characterized by the binding of α-(carboxycyclopropyl)glycine (CCG) isomers, (2S,3S,4S)-CCG (L-CCG-I) and (2S,3R,4S)-CCG (L-CCG-IV) which contain an embedded L-glutamate moiety in a partially restricted conformation [relative to the C(3)–C(4) bond]. The spatial orientation of the perceived functional groups have been elucidated by a conformational analysis in aqueous solution of L-CCG-I and L-CCG-IV using a combination of NMR experimental results, theoretical simulation of NMR spectra, mechanics and dynamics calculations. It was of interest to compare the charge distributions resulting from a number of quantum calculations on the cyclopropane ring. One important conclusion of the study is that the best theoretical model is the MD in solvent. This study shows clearly the preferred ‘t-A’ and ‘g+-B’ conformations of the C(3) aminocarboxymethyl side chain for L-CCG-I and L-CCG-IV, respectively. Weak pH-dependent effects on the structure of the principal L-CCG-I and L-CCG-IV conformers have been established in aqueous solution. The conformations may be grouped by the two backbone torsion angles, χ1 [α-CO2––C(2)–C(3)–C(4)] and χ2 [+NC(2)–C(3)–C(4)–γ-CO2–] and by the two characteristic distances between the potentially active functional groups, α-N+–γ-CO2– (d1) and α-CO2––γ-CO2– (d2). The conformational preferences in solution of L-CCG-I and L-CCG-IV are discussed in the light of the physical features known for specific metabotropic (ACPD) and specific ionotropic (NMDA) agonists, respectively.

Model of the Interaction between the NF-κB Inhibitory Protein p100 and the E3 Ubiquitin Ligase β-TrCP based on NMR and Docking Experiments

NF-κB is a major transcription factor whose activation is triggered through two main activation pathways: the canonical pathway involving disruption of IκB-α/NF-κB complexes and the alternative pathway whose activation relies on the inducible proteolysis of the inhibitory protein p100. One central step controlling p100 processing consists in the interaction of the E3 ubiquitin ligase β-TrCP with p100, thereby leading to its ubiquitinylation and subsequent either complete degradation or partial proteolysis by the proteasome. However, the interaction mechanism between p100 and β-TrCP is still poorly defined. In this work, a diphosphorylated 21-mer p100 peptide model containing the phosphodegron motif was used to characterize the interaction with β-TrCP by NMR. In parallel, docking simulations were performed in order to obtain a model of the 21P-p100/β-TrCP complex. Saturation transfer difference (STD) experiments were performed in order to highlight the residues of p100 involved in the interaction with the β-TrCP protein. These results highlighted the importance of pSer865 and pSer869 residues in the interaction with β-TrCP and particularly the Tyr867 that fits inside the hydrophobe β-TrCP cavity with the Arg474 guanidinium group. Four other arginines, Arg285, Arg410, Arg431, and Arg521, were found essential in the stabilization of p100 on the β-TrCP surface. Importantly, the requirement for these five arginine residues of β-TrCP for the interaction with p100 was further confirmed in vivo, thereby validating the docking model through a biological approach.