TRPM8 inhibits EC migration independently of the channel’s pore function. (Ai) Plot of the percentage of migration in the wound area of silenced HMECs (siTRPM8 in gray and CNTRL in black). (Aii) Plot of the percentage of migration in the wound area of silenced HUVECs (siTRPM8 in gray and CNTRL in black). (Bi) Representative photographs of a wound-healing assay taken at two different time points (t = 0 h, top; t = 8 h, bottom). BTECs transfected with 2 µg of a GFP control vector (CNTRL), 2 µg TRPM8, or TRPM8^Y905A were treated with control medium or 10 µM icilin. Bars, 50 µm. (Bii) Plot of the percentage of migration of BTECs overexpressing TRPM8 (TRPM8, in gray) TRPM8^Y905A (TRPM8^Y905A, in red; TRPM8^Y905A + icilin, in green) or GFP as a control (CNTRL, in black). §, significant difference relative to control. (C) TRPM8 inhibits in vitro vascular network formation. (Ci) Quantification of tubulogenesis in HMECs silenced for TRPM8 (HMEC/siTRPM8) or not (HMECs). (Cii) Quantification of tubulogenesis of BTECs overexpressing TRPM8 (BTEC/TRPM8) or not (BTECs). *, P < 0.05. (D) TRPM8 inhibits EC in vitro sprouting. Representative bright-field micrographs of control (CNTRL) or VEGF-stimulated (VEGF) EC spheroids, in the presence or absence of 10 µM icilin showing reduced vascular sprouting in spheroids. Bars, 50 µm. The bar graphs show quantification of EC spheroid sprouting area, revealing that stimulation of TRPM8 by icilin reduced the sprouting area. Data are expressed as mean ± SEM. *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001.