Targeted degradation of MERTK and other TAM receptor paralogs by heterobifunctional targeted protein degraders

Abstract

TAM receptors (TYRO3, AXL, and MERTK) comprise a family of homologous receptor tyrosine kinases (RTK) that are expressed across a range of liquid and solid tumors where they contribute to both oncogenic signaling to promote tumor proliferation and survival, as well as expressed on myeloid and immune cells where they function to suppress host anti-tumor immunity. In recent years, several strategies have been employed to inhibit TAM kinases, most notably small molecule tyrosine kinase inhibitors and inhibitory neutralizing monoclonal antibodies (mAbs) that block receptor dimerization. Targeted protein degraders (TPD) use the ubiquitin proteasome pathway to redirect E3 ubiquitin ligase activity and target specific proteins for degradation. Here we employ first-in-class TPDs specific for MERTK/TAMs that consist of a cereblon E3 ligase binder linked to a tyrosine kinase inhibitor targeting MERTK and/or AXL and TYRO3. A series of MERTK TPDs were designed and investigated for their capacity to selectively degrade MERTK chimeric receptors, reduce surface expression on primary efferocytic bone marrow-derived macrophages, and impact on functional reduction in efferocytosis (clearance of apoptotic cells). We demonstrate proof-of-concept and establish that TPDs can be tailored to either selectivity degrades MERTK or concurrently degrade multiple TAMs and modulate receptor expression in vitro and in vivo. This work demonstrates the utility of proteome editing, enabled by tool degraders developed here towards dissecting the therapeutically relevant pathway biology in preclinical models, and the ability for TPDs to degrade transmembrane proteins. These data also provide proof of concept that TPDs may serve as a viable therapeutic strategy for targeting MERTK and other TAMs and that this technology could be expanded to other therapeutically relevant transmembrane proteins.

Keywords: AXL; MERTK; TAM receptors; TYRO3; heterobifunctional targeted protein degraders; receptor down-regulation.

Figure 1

Targeted protein degraders (TPD) for targeting MERTK on murine BMDMs. (A) BMDMs are defined as CD11b+ F4/80+ cells by flow cytometry. (B) Histograms show that MERTK is expressed on murine BMDMs and is highly induced by treatment with 0.1 µM of dexamethasone, whereas Mertk KO BMDMs do not express MERTK. (C) MERTK levels in WT, dexamethasone treated and Mertk KO BMDMs are measured by flow cytometry (Bar graphs represent mean of MFI ± SD, analyzed by ordinary one way ANOVA, n=3 *=p<0.05) and (D) mRNA expression is measured by qRTPCR (Bar graphs represent mean relative expression ± SD, analyzed by ordinary one way ANOVA, n=3 *=p<0.05; only statistically significant results have been represented by asterisks (*), for all figures in this paper). (E) shows the chemical structures of degraders, KTX-335 consists of a MERTK ligand conjugated to a cereblon ligand by a linker. KTX-959 and KTX-214 are compounds where the cereblon ligand is methylated and unable to engage cereblon, hence they serve as negative controls for KTX-652 and KTX-978, respectively.

Figure 2

KTX-335 selectively degrades MERTK and suppresses efferocytosis. (A) Treatment of EGFR/TAM chimeric cell lines with KTX-335 shows degradation specific for MERTK and not for AXL or TYRO3. (Plot shows relative MFI of EGFR ± SD normalized to untreated, analyzed by two way ANOVA, Dunnett’s multiple comparison test, n=3, *=p<0.05, ** =p < 0.01, *** = p <0.001, **** = p < 0.0001, compared to untreated, also shown as 0 nM). (B) shows kinetics of degradation of MERTK on naïve BMDMs treated with KTX-335 for 1, 3, 6, 10, 18 and 24 hours. MERTK expression is measured by flow cytometry using an anti-MERTK-PE antibody and y- axis represents relative MFI (PE) ± SD normalized to the 0 hour time point (Ordinary One way ANOVA, n=4, * =p<0.05, ** =p < 0.01, *** = p <0.001, **** = p < 0.0001, compared to 0 hour time point). (C) KTX-335 reduces surface MERTK expression on naïve BMDMs (18 hours treatment) in a dose dependent manner up-to 100 nM, and demonstrating the hook effect at higher concentrations, (Plot shows relative MFI of MERTK-PE ± SD normalized to untreated, analyzed by one-way ANOVA, n=4, **** = p < 0.0001, compared to untreated). (D) KTX-335 also efficiently reduces MERTK expression on Dexamethasone induced BMDMs (Plot shows relative MFI of MERTK-PE ± SD normalized to untreated, analyzed by one-way ANOVA, n=4, * =p<0.05, ** =p < 0.01 compared to untreated). (E) KTX-335 reduces efferocytosis of apoptotic cells in naïve BMDMs, (Plot shows the mean percentage of pHrodo+ cells ± SD within the CD11b+ F4/80+ gate, analyzed by ordinary one way ANOVA, n=3, * =p<0.05, ** =p < 0.01, *** = p <0.001, **** = p < 0.0001. (F) Western blot shows rebound of MERTK in BMDMs post wash out of KTX-335 at 0, 3, 6, 24 and 72 hours.

Figure 3

KTX-652 is a MERTK and AXL dual degrader that potently inhibits efferocytosis. (A) Flow cytometric assay with EGFR/TAM chimeric cell lines shows that KTX-652 targets both MERTK and AXL (Plot shows relative MFI of EGFR ± SEM normalized to untreated, analyzed by mixed effects analysis, Dunnett’s multiple comparison test, n=5, *=p<0.05, ** =p < 0.01, *** = p <0.001, **** = p < 0.0001, compared to untreated). (B) KTX-652, but not KTX-959 reduces MERTK expression on naïve BMDMs (Plot shows MFI ± SD normalized to untreated, analyzed using two way ANOVA and Dunnett’s multiple comparisons test, n=4, * =p<0.05, ** =p < 0.01, *** = p <0.001, **** = p < 0.0001, compared to untreated). (C) KTX-652 but not KTX-959 reduces MERTK expression on dexamethasone treated BMDMs (Plot shows MFI ± SD normalized to untreated, two way ANOVA and Dunnett’s multiple comparisons test, n= 4,* =p < 0.05). (D) KTX-652 is effective at inhibiting efferocytosis of apoptotic cells at nM concentrations in BMDMs (Plot shows the mean of percentage of pHrodo+ cells ± SD within the CD11b+ F4/80+ gate, analyzed by two way ANOVA, Dunnett’s multiple comparison test, n=3, *=p<0.05, ** =p < 0.01, *** = p <0.001, **** = p < 0.0001). (E) KTX-652 degrades surface AXL on EO771 cells, measured by flow cytometry (Plot shows relative MFI ± SD normalized to untreated, analyzed by two way ANOVA and Dunnett’s multiple comparisons test, n= 4, *=p<0.05, ** =p < 0.01, *** = p <0.001, **** = p < 0.0001). (F) shows the functional consequence of AXL degradation in EO771 cells, shown by reduced proliferation of EO771 cells treated with 100 nM KTX-652 (Plot depicts % confluence as a function of time, as measured by Incucyte).

Figure 4

A single dose of KTX-652 causes dose and time-dependent MERTK degradation in mouse spleen. (A) Schematic showing workflow of PK and PD experiments. Single dose of KTX-652 was administered subcutaneously at 3 mg/kg, 30 mg/kg or 100 mg/kg. Plasma and spleens were collected at 0.25, 0.5, 3, 6, 24, 48, and 72 hours for PK and PD analysis. Pharmacokinetic profile in spleen (B) and plasma (C) show good exposure levels up to 72 hours for all 3 concentrations (Plots show mean ± SEM concentrations of KTX-652 in plasma and spleen, 3 mice per group). Western blot representation of MERTK reduction in spleen levels in spleen at 3 hours after administration of KTX-652 at doses, 3 mg/kg (D), 30 mg/kg (E) and 100 mg/kg (F). Reduction in MERTK levels is seen up to 6 hours, and increases at 24 hours for 3 mg/kg dose (D, G), and at 48 hours for 30 mg/kg dose (H). MERTK levels remain low for 72 hours at 100 mg/kg (I). (Quantification of western blots, represented by relative MERTK/B-ACTIN, analyzed by Sidak’s multiple comparisons test, * =p<0.05, ** =p < 0.01, *** = p <0.001, **** = p < 0.0001, 3 mice per group, per time point.

Figure 5

KTX-978 functions as a pan-TAM degrader (A) Treatment on EGFR/TAM chimeric cell lines shows that KTX-978 is a pan-TAM degrader (Plot shows relative MFI of EGFR ± SEM normalized to untreated, analyzed by mixed effects analysis, Bonferroni’s multiple comparison test, n=3, *=p <0.05, ** =p < 0.01, *** = p < 0.001, **** = p < 0.0001, compared to untreated). (B) KTX-978 induces targeted degradation of MERTK on peritoneal macrophages from C57BL/6 mice, while KTX-214 has a reduced effect. (Plot represents relative MFI ± SD normalized to untreated, analyzed by two way ANOVA, Dunnett’s multiple comparison test, n=3, *=p<0.05, ** =p < 0.01, *** = p <0.001, **** = p < 0.0001). (C) Similarly, KTX-978 and KTX-214 with lesser potency, reduces apoptotic cell efferocytosis in BMDMs (Plot shows the mean of percentage of pHrodo+ cells ± SD within the CD11b+ F4/80+ gate, analyzed by Sidak’s multiple comparisons test, n=3 *=p<0.05, ** =p < 0.01, *** = p <0.001, **** = p < 0.0001). (D) KTX-978 degrades surface AXL on EO771 cells, up to 100 nM measured by flow cytometry (Plot shows relative MFI ± SD normalized to untreated, analyzed by two way ANOVA and Dunnett’s multiple comparisons test, n= 3, * =p<0.05, ** =p < 0.01, *** = p <0.001, **** = p < 0.0001). (E) shows the functional consequence of AXL degradation in EO771 cells, shown by reduced proliferation of EO771 cells treated with 100 nM KTX-978 (Plot depicts % confluence as a function of time, as measured by Incucyte).

Figure 6

KTX-978 is a potent degrader of MERTK in spleen and tumor. (A) Plasma concentration of KTX-978 was maintained above 10 ng/mL following a single subcutaneous dose of 10 mg/kg up to 72 hours. (B) MERTK degradation in spleen after 6 hours of treatment with 30 mg/kg and 100 mg/kg dose shown by western blotting. Pharmacodynamic profile showing sustained degradation of MERTK in spleen at doses 30 mg/kg and 100 mg/kg after 6, 24 and 48 hours. (C) and in the CT-26 tumor at doses 30 mg/kg and 100 mg/kg for 24 hours (D). (Quantification of western blots, represented by relative MERTK/B-ACTIN, analyzed by Sidak’s multiple comparison’s test, n=9, ** =p < 0.01, *** = p <0.001, **** = p < 0.0001).