Fig 2: NISCH S-glutathionylation activates the PAK1-LIMK1-cofilin pathway.MDA-MB-231 cells expressing NISCH WT or C185S were incubated in low or high glucose (LG or HG) for 16 h. (a) Phosphorylation levels of PAK1, ERK1/2, and MEK1/2 (n = 3). (b) Phosphorylation levels of LIMK1 and cofilin (n = 3). (c) Paxillin phosphorylation level (n = 3). (d) A model for cell migration induced by NISCH glutathionylation. In a non-stressed condition, NISCH binds to its protein partners, including Rac1 and PAK1, thus suppressing Rac1-and PAK1-mediated cell migration. NISCH may bind and retain Rac1 and integrin-α5 in the cytosol or endosome. However, upon ROS production, NISCH is glutathionylated at C185, dissociating Rac1 and PAK1, but not integrin-α5. The activated PAK1 activates its downstream LIMK1 and inhibits cofilin through phosphorylation, thereby increasing actin dynamics and polymerization. The dissociated Rac1 increases its localization to the membrane, where it activates actin filament branching, induces lamellipodia and membrane ruffles, and increases cell migration. Data represent the mean ± SD. The statistical difference was analyzed by one-way ANOVA and Tukey’s post-hoc test (a–c), where *p < 0.03, **p < 0.002, ***p < 0.0002, ****p < 0.0001.

Fig 3: NISCH is selectively susceptible to glutathionylation at C185.(a) NISCH structure, domains, and selected cysteines, including C185 (stick model). The structure was retrieved from the AlphaFold database. PX, LRR, CC, and ITGBD were found in UniProt. TED, PH-like, and PH are found in the AlphaFold and Pfam databases. (b) The accessible surface area (ASA) of cysteines in NISCH. (c) C185 and its neighboring residues in NISCH orthologs. (d) The outline of the clickable glutathione approach. Upon incubating azido-Ala, cells expressing a glutathione synthetase mutant (GS M4) synthesize and produce clickable glutathione (N3-GSH), which forms glutathionylation with proteins. (e-f) Glutathionylation of NISCH WT and cysteine mutants in HEK293/GS M4 cells in response to hydrogen peroxide (H2O2). H2O2 was incubated for 15 min (n = 3). (g-h) Glutathionylation of NISCH WT and C185S in MDA-MB-231 cells expressing GS M4 in low glucose (LG) compared to high glucose (HG) conditions. Cells were incubated in LG or HG for 16 h (n = 3–4). The lysates were subjected to click chemistry using biotin-alkyne and analyzed by Western blot using FLAG antibody before (input) and after (eluted) pull-down with streptavidin-agarose. Data represent the mean ± SD. The statistical difference was analyzed by one-way ANOVA and Tukey’s post-hoc test (e, g-h), where *p < 0.03, **p < 0.002, ***p < 0.0002, ****p < 0.0001.

Fig 4: Engineered NISCH mutant suppresses cancer cell migration and invasion.(a) A scheme of the G-PROV approach that enables functional analysis of glutathionylation on the protein of interest (POI): dehydroglutathione (dhG) is used to introduce a non-reducible glutathione modification on the purified engineered NISCH construct (enNISCH). Subsequently, the glutathione-modified NISCH construct is delivered to cells via fusogenic liposome, where the effect of glutathionylation on NISCH can be analyzed. (b) dhG modification on purified enNISCH. Purified enNISCH WT and C185S were incubated with dhG. The dhG-mediated glutathione modification was analyzed by glutathione antibody (n = 2). (c) Analysis of enNISCH delivered to MDA-MB-231 cells via fusogenic liposome. After incubating fusogenic liposomes containing His-enNISCH constructs (modified or non-modified by dhG) for 2 h, lysates were analyzed by Western blot (n = 2). (d-e) The migration and invasion assays. After incubating fusogenic liposomes for 2 h, MDA-MB-231 cells were analyzed by the wound-healing migration (d, n = 6) and transwell invasion (e, n = 2) assays. (f) The colony formation by enNISCH constructs. Fusogenic liposomes containing enNISCH WT or C185S were incubated with MDA-MB-231 cells on plates, and the number of colonies formed was counted after 15 days. Cells were incubated in DMEM containing 25 mM Glc (d–e) or 5 mM Glc (f). The statistical difference was analyzed by one-way ANOVA and Tukey’s post-hoc test (b–e) or two-way ANOVA and Sidak’s post-hoc test (f), where *p < 0.03, **p < 0.002, ***p < 0.0002, ****p < 0.0001.

Fig 5: NISCH C185 S-glutathionylation regulates cell migration and invasion.(a-d) Migration analysis of MCF7 cells expressing NISCH WT or C185S. (a) Analysis of NISCH knockout (KO) in MCF7 cells (left, n = 3), and cell migration analysis (right, n = 3) after CRISPR/Cas9-based NISCH KO. (b) The wound-healing migration analysis. MCF7-NISCH KO (MCF7NIS−KO) cells re-expressing NISCH WT or C185S (MCF7-NISCH WT or C185S) were incubated in high glucose (HG, 25 mM) or low glucose (LG, 5 mM) (n = 6). The yellow color indicates the area without cells. (c) Single-cell migration analysis. MCF7-NISCH WT or C185S cells in LG or HG were monitored by confocal microscopy every 10 min (0–180 min). The images were combined to analyze the migration of individual cells (n = 8–10). (d) Transwell migration analysis. MCF7-NISCH WT or C185S cells were incubated in LG or HG for 24 h (n = 3). (e) Spheroid invasion analysis of MCF7-NISCH WT or C185S. Individual spheroids were incubated in HG or LG and monitored daily for 96 h (n = 10). Data represent the mean ± SD. The statistical difference was analyzed by one-way ANOVA and Tukey’s post-hoc test (a, c) or two-way ANOVA and Sidak’s post-hoc test (b, d-e), where *p < 0.03, **p < 0.002, ***p < 0.0002, ****p < 0.0001.