Development of an online SPE–LC–MS-based assay using endogenous substrate for investigation of soluble epoxide hydrolase (sEH) inhibitors

摘要

Soluble epoxide hydrolase (sEH) is a promising therapeutic target for the treatment of hypertension, pain, and inflammation-related diseases. In order to enable the development of sEH inhibitors (sEHIs), assays are needed for determination of their potency. Therefore, we developed a new method utilizing an epoxide of arachidonic acid (14(15)-EpETrE) as substrate. Incubation samples were directly injected without purification into an online solid phase extraction (SPE) liquid chromatography electrospray ionization tandem mass spectrometry (LC–ESI–MS–MS) setup allowing a total run time of only 108 s for a full gradient separation. Analytes were extracted from the matrix within 30 s by turbulent flow chromatography. Subsequently, a full gradient separation was carried out on a 50X2.1 mm RP-18 column filled with 1.7 μm core–shell particles. The analytes were detected with high sensitivity by ESI–MS–MS in SRM mode. The substrate 14(15)-EpETrE eluted at a stable retention time of 96 ± 1 s and its sEH hydrolysis product 14,15-DiHETrE at 63 ± 1 s with narrow peak width (full width at half maximum height: 1.5 ± 0.1 s). The analytical performance of the method was excellent, with a limit of detection of 2 fmol on column, a linear range of over three orders of magnitude, and a negligible carry-over of 0.1% for 14,15-DiHETrE. The enzyme assay was carried out in a 96-well plate format, and near perfect sigmoidal dose–response curves were obtained for 12 concentrations of each inhibitor in only 22 min, enabling precise determination of IC₅₀ values. In contrast with other approaches, this method enables quantitative evaluation of potent sEHIs with picomolar potencies because only 33 pmol L−1 sEH were used in the reaction vessel. This was demonstrated by ranking ten compounds by their activity; in the fluorescence method all yielded IC₅₀ ≤ 1 nmol L−1. Comparison of 13 inhibitors with IC₅₀ values >1 nmol L−1 showed a good correlation with the fluorescence method (linear correlation coefficient 0.9, slope 0.95, Spearman’s rho 0.9). For individual compounds, however, up to eightfold differences in potencies between this and the fluorescence method were obtained. Therefore, enzyme assays using natural substrate, as described here, are indispensable for reliable determination of structure–activity relationships for sEH inhibition.