In Vivo-Active Soluble Epoxide Hydrolase-Targeting PROTACs with Improved Potency and Stability

Abstract

Soluble epoxide hydrolase (sEH) is a bifunctional enzyme involved in fatty acid metabolism and a promising drug target. We previously reported first-generation sEH proteolysis-targeting chimeras (PROTACs) with limited degradation potency and low aqueous and metabolic stability. Herein, we report the development of next-generation sEH PROTAC molecules with improved stability and degradation potency. One of the most potent molecules (compound 8) exhibits a half-maximal degradation concentration in the sub-nM range, is stable in vivo, and effectively degrades sEH in mouse livers and brown adipose tissues. Given the role played by sEH in many metabolic and nonmetabolic diseases, the presented molecules provide useful chemical probes for the study of sEH biology. They also hold potential for therapeutic development against a range of disease conditions, including diabetes, inflammation, and metabolic disorders.

Figure 1

(A) Chemical structures of clinical candidate sEHis. (B) Chemical structures of first-generation sEH PROTACs 1a and ALT-PG2. (C) The sEHi scaffolds used in this study, t-TUCB and TPPU, and their X-ray crystal structures when complexed with human sEH (Left panel, t-TUCB, PDB: 6AUM. Right panel, TPPU, PDB: 4OD0). The solvent-exposed portions of the molecules are highlighted by dotted red circles.

Figure 2

Compounds 6 and 8 have improved stability compared to 1a and ALT-PG2. (A) Chemical stability in cell culture medium. ^aStability in cell culture medium was measured in DMEM. (B) In vitro metabolic stability. Ketanserin was used as an assay control. ^bMicrosomal stability was measured by incubating 500 nM compound with mouse liver microsomes (0.25 mg/mL) and NADPH (1 mM) solution for 0–45 min, in duplicate (Table S1). ^cValues measured without NADPH after incubation for 45 min.

Figure 3

Compounds 6 and 8 have greater sEH degradation potency than first-generation sEH PROTAC compounds 1a and ALT-PG2. (A) Concentration-dependent degradation of sEH induced by compounds 6 and 8 (0.03–10 nM). Differences between treatment groups were analyzed by one-way ANOVA with the Holm–Sidak test (****p < 0.0001, ***p < 0.001, **p < 0.01, *p < 0.04, n.s.: not significant). (B) Direct comparison of sEH degradation induced by 1a and ALT-PG2 vs compound 8. Values represent the mean ± SD (n = 3). Differences between treatment groups were analyzed by one-way ANOVA with the Holm–Sidak test (****p < 0.0001, *** p < 0.001, **p < 0.01).

Figure 4

Detailed characterization of sEH degradation induced by compound 8. (A) Degradation kinetics of sEH induced by 50 nM compound 8 treatment for 2–48 h. Whole-cell lysate was employed in immunoblotting. Values represent the mean ± SD (n = 3). Student’s t-test was performed, and p-values represent significant differences between compound treatment vs vehicle control (****p < 0.0001, ***p < 0.001). (B) Validation of sEH degradation mechanism. Treatment with compound 8 alone or in combination with (Top panel) Lenalidomide (cereblon ligand) or t-TUCB (sEHi), (Middle panel) proteasome inhibitor MG-132, or (Bottom panel) NEDD8-activating enzyme inhibitor MLN4942. Cells were lysed following 4 h treatment of compound 8. Whole-cell lysate was employed in immunoblotting. (C) Compound 8 selectively degraded cytosolic sEH but not peroxisomal sEH. Cytosol-selective sEH degradation was induced by 100 nM compound 8 treatment for 48 h. Cell organelles were fractionated into the cytosol-containing S10 fraction and peroxisome-containing P10 fraction, which were confirmed by immunoblotting with catalase (peroxisome marker) and GAPDH (cytoplasmic marker), respectively. Values represent the mean ± SD (n = 3). Statistical differences were analyzed by Student’s t-test. Differences were significant for S10 samples (****p < 0.0001) but not for P10 samples (n.s.). (D) Quantitative MS-based proteomic analysis indicated that sEH was one of the most significantly reduced proteins after compound 8 treatment. Data points representing sEH are shown in red. The vertical dashed lines and horizontal dashed line in the volcano plot represent log₂ (fold change) cutoff of ±1.0 (fold change = ± 2.0) and −log₁₀ (p-value) cutoff of 1.30 (p = 0.05, n = 3).

Figure 5

Compound 8 effectively reduces ER stress. HEK293T cells were treated with the indicated sEH modulators for 12 h prior to adding 1 μM Tg for an additional 24 h. Changes in ER stress were determined via immunoblotting with selected ER stress markers: (A) pIRE(S724)/IRE, (B) pPERK(T981)/PERK, and (C) pEIF2α(S51)/EIF2α. Values represent the mean ± SEM (n = 4). Differences between treatment groups were analyzed by one-way ANOVA with the Holm–Sidak test (****p < 0.0001, ***p < 0.001, **p < 0.01, *p < 0.04) or Student’s t-test (^$$$$p < 0.0001, ^$$$p < 0.004, ^$$p < 0.009, ^$p < 0.04). The original Western blots are included in the Supporting Information.

Figure 6

In vivo evaluation of compound 8. (A) Pharmacokinetic profile (Top panel) and pharmacokinetic parameters (Bottom panel) of compounds 8, 17, and 19 (Male CD1 mice, n = 3 per group, 10 mg/kg i.p.). Values represent the mean ± SD (n = 3) (Table S3). (B, C) Compound 8 induces sEH degradation in vivo. Male C57BL/6J mice (n = 3 per group) received a single i.p. injection of vehicle (Veh) or compound 8 at a dosage of 12 mg/kg. After 24 h, mice were sacrificed, and sEH levels in the liver (B) and BAT (C) were measured via immunoblotting. Values represent the mean ± SEM (n = 3). Student’s t-test was performed, and p-values represent significant differences between compound treatment vs vehicle control (**p < 0.01, *p < 0.04).