Fig 2: AHCYL1/IRBIT and tau are in close proximity in cultured cells.A, schematic representation of the PLA method. B, confocal images of 7-day-old mouse primary cortical neurons stained with Hoechst-33342 (blue), IRBIT (white), tau (red), and MAP2 (green). The scale bar represents 5 μm. C, confocal images of primary cortical neurons stained with Hoechst-33342 (blue) and PLA probe (red). Primary antibodies were against IRBIT and tau, with +crtl indicating the presence of species-specific secondary antibodies. The scale bar represents 10 μm. D, representative PLA (IRBIRT + tau) staining in cortical neurons. Hoechst-33342 was used to stain the nucleus. The scale bar represents 10 μm. E, confocal images of mouse-derived cortical neurons stained with Hoechst-33342 (blue), tau (white), IRBIT (green), and PLA (IRBIT + tau; red). The scale bar represents 5 μm. F, representative immunoblots of smNPCs transfected with scramble and siIRBIT. Antibodies against IRBIT and β-actin (as loading control) were used. M is protein molecular marker. G, representative confocal images of siRNA-transfected smNPCs, which were stained with PLA for IRBIT + tau. Antibody against KDEL was used as counterstaining of the ER. Hoechst-33342 was used to image nuclei. The scale bar represents 10 μm. Quantification of PLA dots is reported on the right (Mann–Whitney test, ∗∗∗∗p < 0.0001). AHCYL1/IRBIT, SAH hydrolase–like protein 1/IP3R-binding protein; ER, endoplasmic reticulum; KDEL, Lys–Asp–Glu–Leu; MAP2, microtubule-associated protein 2; PLA, proximity ligation assay; smNPC, small-molecule neural progenitor cell.

Fig 3: Scheme for the effect of DJ-1 on negatively regulating ferroptosis.There are two routes to generate the intracellular cysteine in cells: one is the cystine import though SLC7A11 and the other one is the transsulfuration pathway. Under basal conditions, the cystine import plays the dominate role in the GSH precursor cysteine generation. When the cystine import is blocked by erastin, the DJ-1 preserved transsulfuration pathway still could provide essential cysteine synthesis (refer to DJ-1 high with ferroptotic stress). DJ-1 depletion can markedly inhibit GSH levels only when the cystine import is blocked (refer to DJ-1 low with ferroptotic stress). The DJ-1 KD distinctly decreases Hcy levels by impairing SAHH activity via enhancing its interaction with AHCYL1. Decreased level of Hcy subsequently causes the depletion of GSH and eventually enhances ferroptotic cell death.

Fig 4: Tau overexpression alters neither mitochondria–ER contact sites nor mitochondrial respiration.A, representative confocal images of HA-YFP and GFP-tau overexpressing HEK293T cells stained with PLA (TOM20 + KDEL). Quantification of PLA dots is shown on the right. Data were obtained from two independent experiments and n = 19 cells/condition (unpaired t test, ns). The scale bar represents 10 μm. B and C, seahorse analysis and representative confocal images of HEK293T overexpressing mitochondrial-targeted GFP and GFP-tau. The scale bar represents 10 μm. D, schematic representation of the interaction between tau, IRBIT, and IP3R. As a result of tau binding, IRBIT dissociates from IP3R at the ER. ER, endoplasmic reticulum; HA, hemagglutinin; HEK293T, human embryonic kidney 293T cell line; IP3R, inositol 1,4,5-trisphosphate receptor; KDEL, Lys–Asp–Glu–Leu; ns, not significant; PLA, proximity ligation assay; TOM20, translocase of outer mitochondrial membrane 20.

Fig 5: Tau interferes with AHCYL1/IRBIT localization at the IP3R.A, PLA staining in HEK293T cells O/E HA-YFP or GFP-tau using primary antibodies against IRBIT and IP3R. Quantification of PLA dots is shown on the right. Data were obtained from two independent experiments and n = 15 cells/condition (unpaired t test, ∗∗∗∗p < 0.0001). The scale bar represents 10 μm. B, representative immunoblots of HEK293T cells overexpressing HA-htau40. M is protein molecular marker. Densitometries for IP3R and AHCYL1/IRBIT expression are on the right (n of experiments = 4; unpaired t test, ns). C, representative immunoblots of IP3R and AHCYL1/IRBIT in HeLa cells transfected with scramble and siIRBIT. M is protein molecular marker. Densitometry of IP3R expression is on the right (n of experiments = 4; unpaired t test, ns). D, representative confocal images of HA-YFP, GFP-tau, and GFP-IRBIT overexpressing HeLa cells, which were cotransfected with scramble and siIRBIT. Cells were stained with Hoechst-33342 (nuclei, blue), anti-GFP (green), and PLA (IRBIT + IP3R, red). Quantification of PLA dots is shown on the right (n of experiments = 3; ordinary one-way ANOVA, ∗p < 0.05, ∗∗∗∗p < 0.0001, ns). Representative confocal image of a negative PLA staining is included. The scale bar represents 10 μm. AHCYL1/IRBIT, SAH hydrolase–like protein 1/IP3R-binding protein; HA, hemagglutinin; HEK293T, human embryonic kidney 293T cell line; IP3R, inositol 1,4,5-trisphosphate receptor; ns, nonsignificant; O/E, overexpression; PLA, proximity ligation assay.