Fig. 1: Engineering of a genetically encoded CRAC channel inhibitor binder (CRAB).

a Schematic illustration of CRAC activation and CRAB-mediated inhibition. Left: ER Ca²⁺ depletion triggers conformational changes, oligomerization, and translocation of STIM1 to ER-PM contact site, where its exposed SOAR domain engages and gates ORAI channels to permit Ca²⁺ influx. This activates calmodulin-dependent calcineurin, leading to NFAT dephosphorylation and nuclear translocation to drive gene expression. Right: PM-anchored CRAB competes with endogenous ORAI for STIM1-SOAR after store depletion, thereby preventing STIM1-ORAI coupling to suppress CRAC channel activation. Abbreviations: CRAC calcium release-activated calcium channel, CRAB CRAC channel inhibitory binder, PM plasma membrane, ER endoplasmic reticulum, SOAR STIM1 Orai-activating region, NFAT nuclear factor of activated-T cells. Created in BioRender. Liu, X. (2026) https://BioRender.com/enohm3h. b Confocal images of HeLa cells co-expressing YFP-SOAR (green) with cytosolic mCherry-tagged ORAI3 C-terminal peptide (mCh-O3CT, red; top), PM-anchored O3CT (mCh-O3CT-CAAX; middle), or PM-anchored O3CT mutant G290V (bottom). Scale bar, 10 µm. c Quantification of PM-to-cytosol fluorescence intensity ratio (F_PM/F_Cyto) of YFP-SOAR from panel b. YFP-SOAR expressed alone served as the control. Data represent mean ± sem. n = 30 cells from three independent biological replicates. ** P = 0.006; ****P < 0.0001 (Two tailed Welch’s t-test). d Deep mutational scanning heatmap illustrating the relative extent of YFP-SOAR colocalization with plasma membrane-anchored O3CT variants. Scale bar, 0 (white, no interaction) to 2 (red, strong interaction), with 1 (blue) indicating intermediate colocalization. n = 10 cells per condition were analyzed across three independent biological replicates. e An AlphaFold-predicted model structure of the O3CT peptide in complex with the SOAR domain of STIM1, with inset highlighting the G290V mutation (red) in O3CT. f Comparison of ΔΔG of mutation values for the CRAB-SOAR complex. ΔΔG of mutation values were calculated with Rosetta 3.14 across 50 iterations for WT and each of the Q289Y, G290R, and G290V mutants. Error bars are shown as ± SD and *P = 0.013; ****P < 0.0001 (Two-tailed Welch’s t-test). g Monitoring Ca²⁺ influx in HeLa cells stably expressing GCaMP6s and transiently co-expressing the indicated O3CT variants. HeLa cells were subjected to TG- induced ER store depletion in a Ca²⁺-free buffer, followed by re-addition of 2 mM extracellular Ca²⁺ to trigger Ca²⁺ influx. Data are shown as mean ± sem. n = 30 cells from three independent biological replicates. h Quantification of peak GCaMP6s intensities of cells shown in panel (g). Data are shown as mean ± sem. n = 30 cells from three independent biological replicates. ****P < 0.0001; *P = 0.0463, (Two tailed Welch’s t-test). i Quantification of nuclear GFP intensities in HeLa cells stably expressing NFAT_1-460-GFP alone (control) or with transient co-expression of the indicated O3CT variants following TG-induced ER store depletion. Data are shown as mean ± sem. n = 95 cells from three independent biological replicates. ****P < 0.0001 (Two tailed Welch’s t test). j Confocal images of HeLa cells stably expressing NFAT_1-460-GFP (green) and transiently co-expressing cytosolic mCherry-tagged O3CT (mCh-O3CT, red) or PM-anchored CRAB (mCh-PM-CRAB, red) in the absence (top panels) or presence of TG treatment (bottom panels). Notably, cells expressing mCh-PM-CRAB exhibited abrogation of NFAT nuclear translocation (marked by asterisks) regardless of ER store depletion. Representative images selected from three independent biological replicates. Scale bar, 20 µm. Source data are provided as a Source Data file.