Regulation of FSP1 myristoylation by NADPH: A novel mechanism for ferroptosis inhibition

Abstract

Excitotoxicity is a prevalent pathological event in neurodegenerative diseases. The involvement of ferroptosis in the pathogenesis of excitotoxicity remains elusive. Transcriptome analysis has revealed that cytoplasmic reduced nicotinamide adenine dinucleotide phosphate (NADPH) levels are associated with susceptibility to ferroptosis-inducing compounds. Here we show that exogenous NADPH, besides being reductant, interacts with N-myristoyltransferase 2 (NMT2) and upregulates the N-myristoylated ferroptosis suppressor protein 1 (FSP1). NADPH increases membrane-localized FSP1 and strengthens resistance to ferroptosis. Arg-291 of NMT2 is critical for the NADPH-NMT2-FSP1 axis-mediated suppression of ferroptosis. This study suggests that NMT2 plays a pivotal role by bridging NADPH levels and neuronal susceptibility to ferroptosis. We propose a mechanism by which the NADPH regulates N-myristoylation, which has important implications for ferroptosis and disease treatment.

Keywords: Excitotoxicity; FSP1; Ferroptosis; Myristoylation; NADPH; Neurodegenerative diseases.

Fig. 1

KA-induced excitotoxicity triggered ferroptosis. (A) RNA sequencing data from primary cortical neurons treated with KA (100 μM, 8 h) were collected from the dataset GSE111434. Pathway enrichment analysis was performed on the differentially expressed genes (DEGs) to confirm their potential function, with color indicating the significance of enrichment results and the bar chart length indicating the enrichment score. (B) The identified DEGs were compared with the ferroptosis dataset from the FerrDb database, and 237 differentially expressed genes related to ferroptosis were identified. The heat map of differentially expressed genes shows 10 upregulated genes and 10 downregulated genes, with color indicating gene expression levels. (C) TEM images of the perinuclear region of striatal neurons from KA (0.625 nmol, 48 h)-treated mice (n = 3). Images c-d are representative areas indicated by the red wireframes in a-b. Images e-f show representative mitochondria indicated by red arrows in c-d. a-b, scale bar = 5 μm; c-d, scale bar = 2 μm; e-f, scale bar = 500 nm. (D) Mitochondrial area frequency in the perinuclear region. More than 90 mitochondria were calculated from 5 pictures of 3 mice per group. (E) The frequency of mitochondrial area in each interval was counted. **P < 0.01 vs control. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 2

Exogenous NADPH supplementation alleviated KA-induced ferroptosis. (A) Mice were injected with NADPH (2.5 mg/kg) 30 min before KA (0.625 nmol) injection, and NADPH was administered every 24 h. The striatum was dissected for TEM detection 48 h after KA injection (n = 3). TEM images of the perinuclear area of striatal neurons are shown. Images d-f are representative areas indicated by the red wireframes in a-c. Images g-i show representative mitochondria indicated by red arrows in d-f. a-c, scale bar = 5 μm; d-f, scale bar = 2 μm; g-i, scale bar = 500 nm. (B) Mitochondrial area frequency in the perinuclear region. More than 90 mitochondria were calculated from 5 pictures of 3 mice per group. (C) The frequency of mitochondrial area in each interval was counted. (D) NADPH levels were detected after treating mouse primary cortical neurons with KA (100 μM), RSL3 (1 μM), or Fer-1 (8 μM) for 0,2,4,8,16 h. (E) Cultures were pretreated with 10 μM NADPH for 4 h and then treated with 100 μM kA for 4 h. Representative flow cytometry analysis images of C11-BODIPY fluorescence are shown. (F) Quantification of ferroptosis-positive cells (n = 3). (G) BODIPY™ 581/591C11 was employed to label lipid peroxidation. The fluorescence intensity of reduced BODIPY (red), oxidized BODIPY (green) and Hoechst (blue) was observed using confocal microscopy, scale bar = 20 μm ***P < 0.001 vs control; ^#P < 0.05 vs KA-treated group, ^###P < 0.001 vs KA-treated group. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 3

Exogenous NADPH supplementation upregulated the membrane localization of FSP1. (A, B) HT22 neurons were pre-incubated with 10 μM NADPH for 4 h and then treated with 50 nM RSL3 for 6 h. Representative images and semiquantitation of western blots for detecting FSP1 expression in the plasma membrane fraction and cytoplasm fraction are shown (n = 3). (C, D) The striatum of mice treated with 0.625 nmol KA for 48 h was recruited for western blots. Representative images and semiquantitation of western blots for detecting FSP1 expression in the plasma membrane fraction and cytoplasm fraction are shown (n = 3). (E, F) Mice were injected with NADPH (2.5 mg/kg) 30 min before KA (0.625 nmol) injection, and NADPH was administered every 24 h. The striatum was collected for western blotting detection 48 h after KA injection. Representative images and semiquantitation of western blots for detecting FSP1 expression in the plasma membrane fraction and cytoplasm fraction are shown (n = 3). (G) Mice were injected with NADPH (2.5 mg/kg) 30 min before KA (0.625 nmol) injection, and NADPH was administered every 24 h. The striatum was collected for western blotting detection 48 h after KA injection (n = 3). Representative immunofluorescence images of FSP1 (green), Na⁺/K⁺-ATPase (red) and Hoechst (blue) are shown, scale bar = 10 μm. (H) NADH consumption assay (340 nm) in PBS buffer using purified FSP1 in combination with CoQ1. The striatum of mice treated with 0.625 nmol kA for 48 h was recruited to purify FSP1 (n = 3). ^&P < 0.05 vs RSL3-treated group, *P < 0.05 vs control, **P < 0.01 vs control, ***P < 0.001 vs control; ^#P < 0.05 vs KA-treated group, ^###P < 0.001 vs KA-treated group. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 4

Exogenous NADPH supplementation promoted N-myristoylation of FSP1. (A) Primary cortical neurons were pre-incubated with 10 μM NADPH for 4 h and then treated with 100 μM kA for 8 h. Myristoylated proteins are labeled using click chemistry. Representative fluorescence images of FSP1 (green), Myristoylation (red) and DAPI (blue) are shown (n = 3), scale bar = 5 μm. (B) Mice were injected with NADPH (2.5 mg/kg) 30 min before KA (0.625 nmol) injection, and NADPH was administered every 24 h. The striatum was collected 48 h after KA injection. FSP1 was purified by immunoprecipitation, followed by detection of NMT2 and FSP1 expression by western blots. (C) After siRNA transfection into HT22 cells for 48 h, myristoylated proteins are labeled using click chemistry. Representative fluorescence images of FSP1 (green), Myristoylation (red) and DAPI (blue) are shown (n = 3), scale bar = 10 μm. The region of interest (ROI) refers to the frame selection to enlarge the image. (D) After siRNA transfection into HT22 cells for 48 h, the cultures were pre-incubated with NADPH (10 μM, 4 h) or NAC (2.5 mM, 4 h), then treated with 50 nM RSL3 for 6 h. Representative flow cytometry analysis images of C11-BODIPY fluorescence are shown (n = 3). (E) Quantification of ferroptosis-positive cells. *P < 0.05, **P < 0.01, ***P < 0.001. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 5

Interaction of murine NMT2 with compound NADPH. (A) NADPH structure. (B) Structure of murine NMT2. (C) CDOCKER interaction energy of the optimal docking structure. (D) Active surface action. (E) 2D plot of the interaction (hydrogen bonds marked as dashed lines).

Fig. 6

The Arg-291 mutation in NMT2 abolished the anti-ferroptosis effect of NADPH. (A) After plasmid transfection into HT22 cells for 48 h, the expression of GST-NMT2 was detected by western blots. (B–D) The cultures were pre-incubated with 10 μM NADPH for 4 h and then treated with 50 nM RSL3 for 6 h. Representative flow cytometry analysis images and quantification of ferroptosis-positive cells of C11-BODIPY fluorescence are shown. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 7

NADPH bound to Arg-291 of NMT2 to enhance its myristictransferase activity. (A) Prokaryotic expression GST-NMT2, GST-NMT2 (p.S289G), GST-NMT2 (p.L290G), GST-NMT2 (p.R291G), GST-NMT2 (p.V292G), GST-NMT2 (p.A293G) were purified by affinity chromatography, and the purification efficiency was verified by Coomassie brilliant blue staining. (B) 2', 5'-ADP pull down was performed to verify the affinity of GST-NMT2 protein and its mutants to NADPH. (C) The activity of the myristoyltransferase GST-NMT2/GST-NMT2 (p.R291G) was determined by relative fluorescence intensity in a measurement system with or without 1 mM NADPH. (D) After plasmid transfection into HT22 cells for 48 h, the cultures were pre-incubated with 10 μM NADPH for 4 h and then treated with 50 nM RSL3 for 6 h. Representative fluorescence images of FSP1 (green), membranes and lipids (red) and DAPI (blue) are shown (n = 3), scale bar = 10 μm. ROI refers to the frame selection to enlarge the image. (E, F) After the plasmid transfection, HT22 cells were incubated with 10 μM NADPH for 4 h and then treated with 50 nM RSL3 for 6 h. Representative images and semiquantitation of western blots for detecting FSP1 expression in the plasma membrane fraction and cytoplasm fraction are shown (n = 3). **P < 0.01. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)