Design and synthesis of small-molecule inhibitors targeting the SCFskp2 E3 ligase and the MDMX-p53 interaction for cancer therapy

Abstract

The SKP1-Cullin1-F-box (SCF) E3 ligases promote the ubiquitination and proteasomemediated degradation of regulatory proteins. Subunits of the SCF complex have shown oncogenic activity, including the F-box protein S-phase Kinase-associated Protein 2 (SKP2). The SCFSKP2 E3 ligase targets several cell cycle negative regulators, e.g. p27, enabling replicative immortality. The only marketed drug that targets the ubiquitinproteasome system is Bortezomib (Velcade; Millenium Pharmaceuticals) (15), which is used to treat multiple myeloma, but has numerous side effects, as the result of targeting a proteasome involved in the regulation of multiple proteins. Targeting an F-box protein is an attractive solution because the F-box protein defines E3 ligase selectivity and each E3 ligase regulates fewer proteins. To date, no small molecule targeting an F-box protein has entered clinical trials. A recently reported compound, (((3-(2,2- dimethyltetrahydro-2H-pyran-4-yl)-4-phenylbutyl)amino)methyl)-N,N-dimethylaniline (16a), has shown evidence to suggest it inhibits the SCFSKP2 ligase. The synthesis of 16a (diastereoisomeric mixture) was completed at Newcastle University and growth inhibitory activity was confirmed in HeLa cells using a sulforhodamine B assay. Both enantiomers of N,N-dimethyl-4-(((4-phenyl-3-(tetrahydro-2H-pyran-4- yl)butyl)amino)methyl)aniline (68a) were synthesised, in one of the first documented enantioselective syntheses of a molecule of this chemotype, and demonstrated similar growth inhibitory activity in HeLa cells to each other and to the racemate, suggesting the compounds have a non-specific mechanism of action. ID HeLa GI50 (μM) 68a 27 ± 4 (S)-68a 33 ± 4 (R)-68a 29 ± 2 vi Murine double minute 2 (MDM2) and its structurally related homologue MDMX (MDM4) negatively regulate the protein level and transcriptional activity of the tumour suppressor p53. Overexpression of MDM2 and MDMX has been observed in multiple human cancers and is associated with cell immortality. Co-crystallisation of the p53-MDM2 and p53-MDMX complexes showed that three p53 residues (F19, W23 and L26) are critical to the formation of these dimers and smallmolecule inhibitors function by competitively blocking the p53-binding sites on MDM2 or MDMX. Several small-molecule MDM2 inhibitors have entered clinical trials; however, no small-molecule MDMX inhibitor has reached the same stage. A recently discovered series of 2,4-disubstituted thiazoles have shown modest potency against MDM2 and MDMX. In silico modelling suggested the compounds interacted with the p53-binding domains in both proteins and extensive SAR studies around 4-(2-((4- fluorobenzyl)amino)thiazol-4-yl)benzene-1,2-diol (59a) were conducted. This series was extended to include benzenoid and pyrrole-based compounds, including 2’-(4- chlorophenoxy)-3’-((4-fluorophenyl)amino)-[1,1’-biphenyl]-4-ol (138c) and 2-(4- chlorobenzyl)-1-(4-chlorophenyl)-5-(4-fluorophenyl)-1H-pyrrole-3-carboxylic acid (220c). All three chemotypes demonstrated low micromolar activity against MDMX and MDM2 by ELISA. ID MDMX IC50 (μM) MDM2 IC50 (μM) 59a 24.7 12.1 138c 19.0 23.2 220c 14.6 11.2 Current efforts are focussing on potentiating MDMX potency in each of the above chemotypes and trialling various co-crystallisation conditions so as to understand the molecules’ binding mechanism and guide rational drug design.