Discovery and Structure-Based Design of Potent Covalent PPARγ Inverse-Agonists BAY-4931 and BAY-0069

Abstract

The ligand-activated nuclear receptor peroxisome-proliferator-activated receptor-γ (PPARG or PPARγ) represents a potential target for a new generation of cancer therapeutics, especially in muscle-invasive luminal bladder cancer where PPARγ is a critical lineage driver. Here we disclose the discovery of a series of chloro-nitro-arene covalent inverse-agonists of PPARγ that exploit a benzoxazole core to improve interactions with corepressors NCOR1 and NCOR2. In vitro treatment of sensitive cell lines with these compounds results in the robust regulation of PPARγ target genes and antiproliferative effects. Despite their imperfect physicochemical properties, the compounds showed modest pharmacodynamic target regulation in vivo. Improvements to the in vitro potency and efficacy of BAY-4931 and BAY-0069 compared to those of previously described PPARγ inverse-agonists show that these compounds are novel tools for probing the in vitro biology of PPARγ inverse-agonism.

Ligand efficacy shifts a nuclear receptor conformational ensemble between transcriptionally active and repressive states

“…In contrast, crystal structures coupled to SAR studies only reveal the fully repressive LBD conformation and do not reveal the structural mechanism of graded repression. Recent studies have reported improved 2-chloro-5-nitrobenzamide PPARγ inverse agonists [24][25][26] ; these studies reported more elaborate R 1 group chemical modifications in contrast to the relatively minimal chemical modifications that we show here are sufficient to push the LBD conformational ensemble towards a fully repressive state. With the growing interest in developing PPARγ inverse agonists for bladder cancer therapeutics, including an analog similar to the 2-chloro-5nitrobenzamide scaffold that is progressing toward phase 1 clinical trials 37,38 , our findings demonstrate a platform that can structurally assess, explain, and potentially predict the activity of PPARγ compounds ranging from agonism to inverse agonism.…”

“…GW9662 is a transcriptionally neutral PPARγ ligand, whereas T0070907 is an inverse agonist that represses PPARγ transcription 21 . T0070907 and GW9662 share the same 2-chloro-5-nitrobenzamide scaffold, and substitutions at the amide R 1 group can be modified to elicit PPARγ agonism 22 or inverse agonism 21,[23][24][25][26] . Covalent PPARγ inverse agonists are currently being developed as potential therapeutics in bladder cancer where hyperactivation of PPARγ transcription occurs [27][28][29][30] .…”

“…Covalent PPARγ inverse agonists are currently being developed as potential therapeutics in bladder cancer where hyperactivation of PPARγ transcription occurs [27][28][29][30] . Crystal structures of PPARγ LBD bound to corepressor peptide and inverse agonists analogs of T0070907 [23][24][25][26] have provided insight into low-energy structural snapshots of PPARγ forced into a fully repressive state by the peptide. However, these static structures do not inform on the structural mechanism, or conformation-activity relationship, behind the differential graded activity observed within a ligand series during structure-activity relationship development, as they do not elucidate how compounds with graded activity influence the dynamic PPARγ LBD conformational ensemble.…”

Emerging and Re-emerging Warheads for Targeted Covalent Inhibitors: An Update

“…Compounds with (non-hetero) aryl-based S N Ar warheads, including the chloro­nitro­benz­amide GW9662 ( 102 , Figure A), have previously been described as covalent modulators of the peroxisome-proliferator activated receptor (PPAR) family of nuclear receptors (see the discussion in our previous Perspective). In a screening aimed at identifying inverse PPARγ agonists, researchers from Bayer obtained hit 103 (Figure A), an inverse agonist related to neutral antagonist GW9662, and pyridine analog T0070907 ( 104 ), which was shown to modify Cys313 (PPARγ isoform 2; see also the X-ray structure of related compound 105 in Figure B) in a similar manner. After optimization at the benz­oxazole-linked aryl residue, SAR studies at the warhead demonstrated that changing the nitro group to less electron-withdrawing substituents or replacement for different warheads, such as acrylamide or 2-chloro­pyridine, considerably reduces biological activity.…”

Unanticipated mechanisms of covalent inhibitor and synthetic ligand cobinding to PPARγ

“…Although these covalent ligands have similar pharmacological functions to orthosteric non-covalent antagonists and inverse agonists, GW9662 and T0070907 function through a different structural mechanism: they slow the rate of exchange between transcriptionally active and repressive conformations natively populated in the apo-LBD, with T0070907 having a more pronounced effect than GW9662 10,23 . In the transcriptionally repressive conformation stabilized by covalent inverse agonists, helix 12 adopts a solvent-occluded conformation within the orthosteric pocket that overlaps with orthosteric ligand binding poses 10,23–25 . These and other published studies have informed a ligand activation model whereby agonist binding to the orthosteric pocket either displaces helix 12 from a solvent occluded repressive conformation within the orthosteric pocket to a solvent exposed active conformation, or selects for an active helix 12 conformation from the dynamic LBD ensemble 26 .…”