A Nimbolide-Based Kinase Degrader Preferentially Degrades Oncogenic BCR-ABL

Abstract

Targeted protein degradation (TPD) and proteolysis-targeting chimeras (PROTACs) have arisen as powerful therapeutic modalities for degrading specific protein targets in a proteasome-dependent manner.However, a major limitation to broader TPD applications is the lack of E3 ligase recruiters. Recently, we discovered the natural product nimbolide as a covalent ligand for the E3 ligase RNF114. When linked to the BET family inhibitor JQ1, the resulting heterobifunctional PROTAC molecule was capable of selectively degrading BRD4 in cancer cells. Here, we show the broader utility of nimbolide as an E3 ligase recruiter for TPD applications. We demonstrate that a PROTAC linking nimbolide to the kinase and BCR-ABL fusion oncogene inhibitor dasatinib, BT1, selectively degrades BCR-ABL over c-ABL in leukemia cancer cells, compared to previously reported cereblon or VHL-recruiting BCR-ABL degraders that show opposite selectivity or in some cases inactivity. Further contrasting from cereblon or VHL-recruiting degradation, we show that BT1 treatment not only leads to BCR-ABL degradation, but also stabilizes the endogenous RNF114 substrate and tumor suppressor substrate p21. This leads to additional anti-proliferative effects in leukemia cancer cells beyond those observed with cereblon or VHL-recruiting BCR-ABL PROTACs. Thus, we further establish nimbolide as an additional general E3 ligase recruiter for PROTACs with unique additional benefits for oncology applications. We also further demonstrate the importance of expanding upon the arsenal of E3 ligase recruiters, as such molecules confer differing and unpredictable selectivity for the degradation of neo-substrate proteins.

Chemoproteomics-enabled discovery of covalent RNF114-based degraders that mimic natural product function

“…To further demonstrate the utility of our fully synthetic RNF114 recruiter EN219 in degrading other more challenging protein targets, we synthesized a degrader linking EN219 to the BCR-ABL inhibitor dasatinib, ML 2-23 and ML 2-22, bearing a longer PEG3 linker and a shorter C3 alkyl linker, respectively (Figures 4A and S4F). For this particular target, ML 2-23 with the longer linker showed more robust degradation of BCR-ABL in K562 leukemia cells compared with ML 2-22, consistent with previously observed structure-activity relationships of nimbolide-based BCR-ABL degraders (Figures 4B, 4C, and S4G) (Tong et al, 2020). Consistent with ML 2-23 engaging BCR-ABL in cells, we observed inhibition of CRKL phosphorylation, a downstream substrate of BCR-ABL signaling (Figures 4B and 4C).…”

“…Interestingly, EN219 showed preferential degradation of BCR-ABL compared with c-ABL, compared with several previous BCR-ABL degraders utilizing cereblon or VHL recruiters that showed opposite selectivity (Figures 4B and 4C) (Burslem et al, 2019;Lai et al, 2016). This preferential degradation was also observed with the equivalent nimbolide-based degrader (Tong et al, 2020). We further showed that this loss of BCR-ABL and c-ABL was not due to transcriptional downregulation of these genes, since BCR-ABL mRNA levels remained unchanged and we observed a likely compensatory increase in c-ABL mRNA levels (Figure S4H).…”

“…ii. The aforementioned crude carboxylic acid was dissolved in DCM (2 mL) and amine 7 (19.8 mg, 0.032 mmol), 1-[Bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxide hexafluorophosphate (HATU) (18.2 mg, 0.048 mmol), and N,N-Diisopropylethylamine (206.0 mg, 1.60 mmol) were added (Tong et al, 2020). The reaction mixture was stirred overnight with monitoring by TLC (20% Methanol in DCM).…”

“…Indeed, while over 600 different E3 ligases have been annotated, only a small handful of these potential targets have succumbed to the PROTAC strategy. Discovering additional and more synthetically tractable E3 ligase recruiters is thus an important topic in expanding the scope of TPD and may help to address resistance mechanisms (Bond et al, 2020;Ottis et al, 2019), promote differing selectivity or kinetic profiles of degradation (Bondeson et al, 2018;Huang et al, 2018;Tong et al, 2020), and lead to cell-type or locationspecific degradation.…”

A strategy to assess the cellular activity of E3 ligase components against neo-substrates using electrophilic probes

“…The successful ubiquitination of neo-substrates often reflects their ability to form de novo protein-protein interfaces with the recruited E3 ligase (Gadd et al, 2017;Nowak et al, 2018). For example, recent studies show that degradation by bifunctional multi-kinase degraders often depends on stable ternary complex formation, and that degradation selectivity may vary according to the recruited E3 ligase (Tong et al, 2020;Bondeson et al, 2018;Lai et al, 2015). We were therefore interested in extending COFFEE to additional neo-substrates by labeling VHL with dasatinib, a promiscuous kinase inhibitor that engages at least 10 kinases with an apparent Ki < 100 nM, in addition to its primary target, BCR-ABL (Klaeger et al, 2017;Montenegro et al, 2020;Leonard et al, 2016).…”