Portal de Periódicos da CAPES

Hitting a Moving Target: Targeting Transient Protein States

2013; Elsevier BV; Volume: 21; Issue: 12 Linguagem: Inglês

10.1016/j.str.2013.11.003

1878-4186

Matthias R. Bauer, Frank M. Boeckler,

Protein Structure and Dynamics

In this issue of Structure, Bista and colleagues report that inhibitors of the MDM2/p53 interaction can be designed to interact with a transiently folded α-helical segment of the MDM2 lid region. This suggests that targeting transient protein states in PPI inhibitor design could be a promising strategy to improve affinity and/or selectivity profiles. In this issue of Structure, Bista and colleagues report that inhibitors of the MDM2/p53 interaction can be designed to interact with a transiently folded α-helical segment of the MDM2 lid region. This suggests that targeting transient protein states in PPI inhibitor design could be a promising strategy to improve affinity and/or selectivity profiles. Protein-protein interactions (PPI) are essential for the regulation of diverse cell functions. Their biological relevance in many pathological processes makes them attractive drug targets. Although several effective small molecule inhibitors of PPIs have been identified and are currently in clinical trials, drug discovery on PPI targets still presents a major challenge (Mullard, 2012Mullard A. Nat. Rev. Drug Discov. 2012; 11: 173-175Crossref PubMed Scopus (197) Google Scholar). The flat, large, and relatively featureless interaction surfaces of PPIs are usually difficult to target with small molecules. This typically results in low hit rates for screening efforts against PPI targets and indicates that the chemical space of existing compound libraries could be biased toward targets with classical binding pockets. Known PPI inhibitors are usually larger, more three-dimensional, and more rigid than typical drug-like molecules (Mullard, 2012Mullard A. Nat. Rev. Drug Discov. 2012; 11: 173-175Crossref PubMed Scopus (197) Google Scholar). Therefore, novel strategies for the identification and rational design of small molecule PPI inhibitors are required. The p53-MDM2 PPI is a promising target for the development of new anticancer drugs. The tumor suppressor p53 plays a key role in DNA repair, cell cycle arrest, and apoptosis (Brown et al., 2009Brown C.J. Lain S. Verma C.S. Fersht A.R. Lane D.P. Nat. Rev. Cancer. 2009; 9: 862-873Crossref PubMed Scopus (693) Google Scholar). MDM2 is one of the main negative regulators of p53. It promotes ubiquitination and degradation of p53 and inhibits its transcriptional activity. Structures of the p53-MDM2 complex show that the N-terminal domain of MDM2 interacts mainly with the hydrophobic residues Phe19, Trp23, and Leu26 of the p53 transactivation domain 1 (TAD1), which forms a nascent helical structure (Figure 1A). MDM2 levels are upregulated in many human cancers. Disrupting the p53-MDM2 interaction is therefore a promising strategy to activate p53 and induce cell cycle arrest or apoptosis in cancer cells (Joerger and Fersht, 2008Joerger A.C. Fersht A.R. Annu. Rev. Biochem. 2008; 77: 557-582Crossref PubMed Scopus (438) Google Scholar). Several small molecule inhibitors have been reported to inhibit the formation of the p53-MDM2 complex, including the nutlins, spirooxindole-containing compounds, and 1,4-diazepinedione inhibitors (Zhao et al., 2013Zhao Y. Bernard D. Wang S. BioDiscovery. 2013; 8: 4Crossref Google Scholar). A common feature of these compounds is that they mimic hydrophobic interactions of p53 residues Phe19, Trp23, and Leu26 with MDM2. Recently, Michelsen et al., 2012Michelsen K. Jordan J.B. Lewis J. Long A.M. Yang E. Rew Y. Zhou J. Yakowec P. Schnier P.D. Huang X. Poppe L. J. Am. Chem. Soc. 2012; 134: 17059-17067Crossref PubMed Scopus (50) Google Scholar described a new piperidinone inhibitor that binds to the p53 interaction site of MDM2 and additionally induces partial folding of the intrinsically disordered N terminus of MDM2. Nuclear magnetic resonance (NMR) and X-ray structures of the MDM2-inhibitor complex show that residues 14–16 form a short β strand through ligand-mediated contacts and the MDM2 lid residues 21–24 form a short α helix (Figure 1B). In the MDM2 apo structure, the lid region exists in a dynamic equilibrium between two states. The more frequently populated “closed” state (Figure 1C) blocks the p53 binding surface, whereas the “open” state remains highly flexible (Showalter et al., 2008Showalter S.A. Bruschweiler-Li L. Johnson E. Zhang F. Brüschweiler R. J. Am. Chem. Soc. 2008; 130: 6472-6478Crossref PubMed Scopus (71) Google Scholar). Targeting the flexible N terminus could be a new strategy for small molecule drug discovery on MDM2. In this issue of Structure, Bista et al., 2013Bista M. Wolf S. Khoury K. Kowalska K. Huang Y. Wrona E. Arciniega M. Popowicz G.M. Holak T.A. Dömling A. Structure. 2013; 21 (this issue): 2143-2451Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar present the first p53-MDM2 inhibitor that directly interacts with the transiently folded α helix of the N-terminal lid region. First, the authors identified the compound KK271 using a pharmacophore model of the key hydrophobic interactions in the p53-MDM2 complex. Surprisingly, the crystal structure of the MDM2-KK271 complex revealed that two KK271 molecules bind to MDM2. One molecule fills the Phe19, Trp23, and Leu26 subpockets, whereas the second molecule occupies a new binding site that is directly connected to the Leu26 subpocket. Conformational rearrangement of Tyr100, Tyr104, and Leu54 enables the formation of this “expanded Leu26 binding site,” which is mainly filled by the benzyl moiety of KK271 (Bista et al., 2013Bista M. Wolf S. Khoury K. Kowalska K. Huang Y. Wrona E. Arciniega M. Popowicz G.M. Holak T.A. Dömling A. Structure. 2013; 21 (this issue): 2143-2451Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar). Taking advantage of this “artifact,” design and synthesis of compounds targeting the enlarged leucine binding site yielded YH300, which is twice as potent as KK271. Modifications of KK271 include an additional 4-fluorophenyl group that is linked by a short oxymethylene linker to the core scaffold. The crystal structure of the MDM2-YH300 complex shows that the extended Leu26 subpocket is occupied by the 4-fluorophenyl moiety. Intriguingly, the N-terminal lid (residues 19–24) partially covers the 4-fluorophenyl moiety, creating a concave binding pocket on the MDM2 surface (Figure 1D). Stabilization of this flexible region with small molecules is an exciting example of exploiting the conformational adaptability of the lid and could be a blueprint for novel types of potent p53-MDM2 inhibitors. MDMX is another important negative regulator of p53 and a close structural homolog of MDM2. The N-terminal binding sites of MDMX and MDM2 are highly related and bind p53 in virtually identical binding modes (Zhao et al., 2013Zhao Y. Bernard D. Wang S. BioDiscovery. 2013; 8: 4Crossref Google Scholar). Recently, the endogenous inhibitor lithocholic acid was found to bind to MDM2 and more preferentially to MDMX (Vogel et al., 2012Vogel S.M. Bauer M.R. Joerger A.C. Wilcken R. Brandt T. Veprintsev D.B. Rutherford T.J. Fersht A.R. Boeckler F.M. Proc. Natl. Acad. Sci. USA. 2012; 109: 16906-16910Crossref PubMed Scopus (51) Google Scholar). Popowicz et al., 2010Popowicz G.M. Czarna A. Wolf S. Wang K. Wang W. Dömling A. Holak T.A. Cell Cycle. 2010; 9: 1104-1111Crossref PubMed Scopus (177) Google Scholar previously described the inhibitor WK298, which binds to the p53 binding site of MDMX and a secondary site in close proximity to the MDMX Leu26 subpocket. This secondary MDMX binding site resembles the expanded Leu26 pocket in the KK271-MDM2 structure. However, it is less concave than the corresponding MDM2 site and might be more difficult to target with rationally designed inhibitors (Popowicz et al., 2010Popowicz G.M. Czarna A. Wolf S. Wang K. Wang W. Dömling A. Holak T.A. Cell Cycle. 2010; 9: 1104-1111Crossref PubMed Scopus (177) Google Scholar). So far, there is no experimental evidence for a potential transient interaction between the N terminus and the N-terminal domain of MDMX. The MDM2 lid residues 16–24 are not conserved in MDMX, and secondary structure prediction shows that the corresponding region of MDMX has no helix propensity. It is therefore unlikely that the intrinsically disordered N terminus of MDMX can be stabilized in a similar way as observed in the YH300-MDM2 complex. Targeting the flexible MDM2 N terminus could be also useful for improving selectivity and toxicity profiles of inhibitors. Recently, Brown et al., 2013Brown C.J. Quah S.T. Jong J. Goh A.M. Chiam P.C. Khoo K.H. Choong M.L. Lee M.A. Yurlova L. Zolghadr K. et al.ACS Chem. Biol. 2013; 8: 506-512Crossref PubMed Scopus (155) Google Scholar reported that the cellular toxicity of small molecules like the nutlins is mainly p53 independent. Considering that the p53 TAD1 domain interacts also with other proteins (e.g., p300 and TAFIIb), it is not surprising that small molecules mimicking the p53 hotspot residues show off-target toxicity. Additionally, the high lipophilicity of many p53-MDM2 inhibitors may increase promiscuous binding to proteins. Using an extended pharmacophore model based on the YH300-MDM2 complex could be an interesting strategy for increasing inhibitor specificity. Evaluation of YH300 in cellular assays should help to assess its potential advantages over existing MDM2 inhibitors. How can this strategy of targeting transient protein states be extended to other PPI targets? The typical flat PPI binding interfaces without significant cavity depth require ligands that are generally larger, more three-dimensional, and more rigid than usual drug molecules. These are often inspired by nature, mimicking the natural peptide or protein structure of the PPI (Figures 2A–2D ). In contrast to regular binding pockets, the usually solvent-exposed binding site residues might be more flexible and can more easily adapt to ligands. It is difficult to probe such adaptable interfaces for transient protein states, starting from large protein mimetics of high molecular complexity based on their usually sophisticated synthetic accessibility (Figure 2E). Diversity-oriented compound libraries of low molecular complexity should be useful chemical probes for the detection of structural plasticity and the identification of such transient states (Figures 2F–2I). This could be another forte of fragment-based drug discovery, which has already been quite successfully applied for PPI targets. Transient Protein States in Designing Inhibitors of the MDM2-p53 InteractionBista et al.StructureOctober 24, 2013In BriefThe major role of MDM2 is to control tumor suppressor p53 protein levels. Bista et al. report structures of MDM2 in complex with two small-molecule compounds, inhibitors of the p53-MDM2 interaction. Both small molecules induce a hydrophobic pocket on the MDM2 surface. Full-Text PDF Open Archive