Structural And Functional Studies Of A Human E3Ubiquitin Ligase MDM2with A Ribosomal Protein RPL11

Inactivation of tumor suppressor genes and activation of oncogenes could lead to tumorigenesis. The well-studied tumor suppressor p53plays a critical role in tumor suppression. p53is a transcription factor that can up-regulate many downstream genes that are involved in cell cycle arrest, DNA repair, and apoptosis. Under normal conditions, the p53is strictly inhibited mainly by an E3ubiquitin ligase, MDM2, which inhibits its transcription activity and also mediates p53ubiquitination and degradation. The MDM2gene is also a p53target, thus they form an MDM2-p53feedback loop. There are different factors controlling the feedback loop in response to different stresses. DNA damage mediated by ATM (Ataxia telangiectasia mutated)/ATR (ATM Rad3related kinase) kinases, oncogenic stress by ARF (Alternating reading frame) induction and ribosomal stress through ribosomal proteins represent three major p53-activating pathways (DNA damage response pathway, ARF-MDM2-p53pathway, RPs-MDM2-p53pathway) that all disrupt MDM2imposed negative control of p53. In the RPs-MDM2-RPLll pathway, RPLll was reported to have the most significant role in p53activation among all the ribosomal proteins. However, it is unclear what is molecular mechanism governing the inhibition of MDM2by RPL11.Through extensive screening of MDM2truncated variants, the human MDM229o-437-RPLll1-178complex was obtained. The MDM2-RPL11crystals were acquired by vapor diffusion in sitting drops. Determined by molecular replacement, the MDM2zinc finger domain in the complex adopted a fold consisting of a310helix followed by two pairs of anti-parallel β-strands, and the overall structure of RPLll formed a hand-like shape consisting of6a-helices,7β-sheets and several loops. The four anti-parallel P-sheets formed a concave surface to which MDM2was tightly bound. The total buried surface area between MDM2and RPLll was large and there were many interactions at the interface, including electrostatic, hydrogen bonds, hydrophobic, and π-π stacking interactions. Two zinc fingers were observed in the MDM2-RPL11complex, one was intramolecular in the MDM2C4zinc finger and another was intermolecular between MDM2and RPLll. The experimental results emphasized the importance of maintaining the integrity of both zinc fingers in MDM2-RPL11interactions. Moreover, we also demonstrated that the acidic domain of MDM2is essential for RPL11binding through structure analysis and structure-guided mutagenesis. We further confirmed the key residues in MDM2-RPL11interactions by pull-down in vitro and co-immunoprecipiation in vivo. RPL11lost the ability to rescue MDM2mutation-induced p53degradation in vivo. The structure-guided mutagenesis showed that mutants of these key residues in MDM2and RPL11cause tumor development in cancers. Considering RPL11is an essential component of60S ribosome subunit, we further made a structure comparison between MDM2-RPL11complex and60S RPL11-28S rRNA in H. sapiens. The results indicated that RPLll contacts MDM2in a manner similar to that of28S rRNA. We further performed both electrophoretic mobility shift assay and microscale thermophoresis experiments, and the results strongly suggested that MDM2mimics nucleic acid binding to RPLll. Moreover, structure analysis demonstrated that both MDM2and RPL11undergo conformational changes after binding to each other. These conformational changes allowed RPL11to tightly bind MDM2. MDM2and MDMX share highly conserved functional domains, however, RPL11can selectively bind to MDM2, but not MDMX. The structure-based gain-of-function and loss-of-function mutations confirmed that the C4zinc finger domain is responsible for the exclusive binding preference of RPL11for MDM2, but not MDMX.hi this study, we reveal how the binding of RPL11to MDM2leads to inhibition of its ubiquitin ligase activity and subsequent p53activation, thus provide a structural basis for potential anti-tumor drug development.