Fig 1: Clinical significance of the study. (A) Correlation between FABP3 and intracellular cir93 in LUAD tissues (n = 250). The relevant information of LUAD patient is summarized in Supplementary Table S1. (B-C) The correlations between taurine and intracellular cir93 (B, n = 250) and between taurine and FABP3 (C, n = 250) in LUAD tissues. The relevant information of LUAD patient is summarized in Supplementary Table S1. (D-E) The correlations between NAT and intracellular cir93 (D, n = 50) and between NAT and FABP3 (E, n = 50) in LUAD tissues. The relevant information of LUAD patient is summarized in Supplementary Table S5. (F) Expression of cir93 and FABP3 in PDX#1 and PDX#2. The mice were euthanized at 32 days after passage. (G) The levels of 4-HNE in PDX#1 and PDX#2 following PKE administration (20 mg/kg, 2 weeks) in mice. The mice were euthanized at 32 days after passage. Scale bar, 100 µm. (H) Tissue slices from PDX#1 and PDX#2 were treated with DMSO or erastin (10 µmol/L, 24 h), in the presence or absence of Fer-1 (1 µmol/L, 24 h) ex vivo, followed by staining with PI. The mice were euthanized at 32 days after passage. Scale bar, 500 µm. (I) OS of PDX-bearing mice (n = 8/group). The mice were euthanized at 80 days after passage. (J) OS of LUAD patients with indicated cir93 and FABP3 expression levels (n = 41/group). The relevant information of LUAD patient is summarized in Supplementary Table S9. (K-L) Co-administrating GW4869 improved PKE efficacy in A549 cell-based CDX mouse models. The mice were euthanized at 36 days after injection. Representative images of tumors at the end of experiments, in which mice were administrated with DMSO or PKE (20 mg/kg), in the presence or absence of GW4869 (2 mg/kg), are shown in panel K, and the tumor growth curves were graphed in panel L. Scale bar, 5 mm. (M) OS curves in A549 cell-based CDX mice following administrating with DMSO or PKE administration (20 mg/kg, 3 weeks), in the presence or absence of GW4869 (2 mg/kg, 3 weeks). The mice were euthanized at 80 days after injection. (N) MDA concentrations in the tumors from panel K-L. (O-Q) Cell proliferation in WT, cir93-/- or FABP3-/- A549 cells treated with or without DMSO, PKE (10 µmol/L), GW4869 (10 µmol/L) and Lipro-1 (1 µmol/L) for the indicated time, as measured by a CCK8-based assay. The data are shown as the mean ± SD from three biological replicates (including IB). *P < 0.05, **P < 0.01 indicate statistical significance. Data in A-E were analyzed using Spearman's rank correlation analysis. Data in I, J, M were analyzed using log-rank tests. Data in L, O-Q were analyzed using a two-way ANOVA test. Data in N were analyzed using a one-way ANOVA test. Abbreviations: LUAD: lung adenocarcinoma, FABP3: fatty acid-binding protein 3, cir93: circRNA_101093, NAT: N-arachidonoyl taurine, PDX: patient-derived xenograft, DMSO: dimethyl sulfoxide, PKE: piperazine erastin, 4-HNE: anti-4-hydroxynonenal, DAPI: 2-(4-Amidinophenyl)-6-indolecarbamidine dihydrochloride, PI: propidium iodide, Fer-1: ferrostatin-1, CDX: cell-derived xenograft, OS: overall survival, PKE: piperazine erastin, MDA: malondialdehyde, WT: wild type, Lipro-1: liproxtratin-1, SD: standard deviation, IB: immunoblotting, ANOVA: analysis of variance
Fig 2: Schematic representation of the study. Briefly, exosome secreted by LUAD cells is essential to elevate intracellular cir93 in LUAD cells. Intracellular cir93‐FABP3 interactions are critical to upregulate FABP3 to reduce global AA via reactions with taurine. Moreover, the function of NAT (the product of taurine and AA) in preventing AA incorporation into the plasma membrane via suppressing ACSL4, LPCAT3 and PLTP was revealed in this study. Therefore, the roles of exosome and cir93 in reducing lipid peroxidation and desensitizing LUAD cells to ferroptosis were established and might help with the development of treatments for LUAD in the future. Abbreviations: LUAD: lung adenocarcinoma, cir93: circRNA_101093, FABP3: fatty acid‐binding protein 3, AA: arachidonic acid, NAT: N‐arachidonoyl taurine, TLX2: T‐cell leukemia homeobox 2, ACSL4: acyl‐CoA synthetase long‐chain family member, LPCAT3: lysophosphatidylcholine acyltransferase 3, PLTP: phospholipid transfer protein
Fig 3: Pathways validation of proteomic data. The protein expression levels of (A) platelet glycoprotein 4 (CD36) and (B) fatty acid-binding protein (FABP3) were measured by Western blot and normalized to beta-actin (internal control) in myocardium from control and Western diet (WD) groups (n = 4–6/group). A representative Western blot and quantification of protein levels are shown. Images of blots have been cropped; the full-length blots are presented in Supplementary Fig. S2. Cardiac levels of (C) malondialdehyde (MDA) and (D) carbonylated protein of control and WD groups (n = 8–13/group). Values are mean ± SD or median (Min-Max). Student’s t-test (in A, B, and D) or Mann-Whitney U-test (in C) for independent samples.
Fig 4: cir93-FABP3 interactions were essential for reducing lipid peroxidation and desensitizing LUAD cells to ferroptosis. (A) Venn diagram showing that FABP3 was the only protein that was simultaneously upregulated by and interacted with cir93, as identified via RNA pull-down followed by proteomics analyses in A549 cells and direct proteomics in both A549 and H1975 cells. Proteins regulated by cir93 were defined as those with fold changes>1.5 following treatments and with a P value < 0.05. (B) Representative IB images of FABP3 in H1975 cells with or without cir93 overexpression, and in A549 cells with or without anti-cir93 treatment. (C) FABP3 expression and cell death in WT and FABP3-/- H1975 cells with or without overexpressing cir93 prior to the treatment of DMSO, erastin (10 µmol/L, 24 h) or RSL3 (5 µmol/L, 24 h). (D) Prediction of the cir93-FABP3 interactions by the catRAPID database (http://service.tartaglialab.com/page/catrapid_group). (E) RNA pull-down experiments in H1975 cells expressing WT or mutant cir93 using WT or mutant cir93 probes, as indicated, and the interaction of FABP3 was revealed by IB. (F) RIP experiments performing by anti-Myc and IgG antibodies in WT and FABP3-/- H1975 cells reconstituting with indicated Myc-tagged FABP3 constructs. (G) Click chemistry experiments demonstrating changes in AA incorporation in control cells and H1975 cells overexpressing FABP3, FABP3?P#4 or cir93?R#2. Scale bar, 50 µm. (H) Changes in global AA in control cells and H1975 cells with indicated treatments. (I) Percentages of cir93 and FABP3 mutations in LUAD (n = 30). The relevant information of LUAD patient is summarized in Supplementary Table S6. (J) 4-HNE levels in LUAD tissue with or without cir93 and FABP3 mutations, as indicated. Scale bar, 200 µm. (K) The R#2 of cir93 and P#4 of FABP3 were essential for cir93 and FABP3 to reduce erastin- and RSL3-induced cell death in H1975 cells. Cell death was measured in H1975 cells pre-overexpressing WT, mutant cir93 or FABP3 followed by treatment with DMSO, erastin (10 µmol/L, 24 h) or RSL3 (5 µmol/L, 24 h). (L-M) Representative IB images for FABP3 following co-incubation with or without LUAD plasma exosome in FABP3-/- H1975 cells reconstituting with FABP3WT or FABP3?P#4 (L). The incidences for plasma exosome with capacity to alter FABP3 with different fold changes (FC) are shown in panel M. The relevant information of healthy individual and LUAD patient is summarized in Supplementary Table S3. The data are shown as the mean ± SD from three biological replicates (including IB). ** P < 0.01 indicates statistical significance. NS, non-significance. Data in C, F, G, H, K were analyzed using one-way ANOVA tests. Data in M were analyzed using a ?2 test. Abbreviations: cir93: circRNA_101093, FABP3: fatty acid-binding protein 3, NC: negative control, DMSO: dimethyl sulfoxide, RSL3: ras-selective lethal small molecule 3, NS: non-significance, WT: wild type, IB: immunoblotting, RIP: RNA Binding Protein Immunoprecipitation, AA: arachidonic acid, ConA: Concanavalin A, Mut: mutant, 4-HNE: anti-4-hydroxynonenal, FC: fold changes, LUAD: lung adenocarcinoma, h: hours, SD: standard deviation, ANOVA: analysis of variance
Fig 5: NAT suppressed AA incorporation into the plasma membrane. (A-B) AA incorporation into the plasma membrane, as measured by a click chemistry-based method in control and H1975 cells infected with shRNAs targeting against CDO1 and CSAD followed by treatment with or without cir93 or FABP3 overexpression plasmids. Representative images are shown in panel A, and data were graphed in panel B. Scale bar, 50 µm. (C) NAT was measured in WT and FABP3-/- H1975 cells with or without cir93 overexpression. (D) NAT in control and A549 cells treated with indicated anti-cir93. (E) NAT in control cells and H1975 cells infected with shRNAs targeting against CDO1 and CSAD followed by treatment with or without cir93 or FABP3 overexpression plasmids. (F) NAT in H1975 cells following co-incubation with plasma exosome from healthy individuals (n = 100) and LUAD patients (n = 100). The relevant information of healthy individual and LUAD patient is summarized in Supplementary Table S7. (G-H) Incorporation of AA into the plasma membrane in H1975 cells treated with DMSO or NAT (20 µmol/L, 24 h), as measured by a click chemistry-based method. Representative images are shown in panel G, and data were graphed in panel H. Scale bar, 20 µm. (I-J) Cell death (I) and lipid ROS generation (J) were measured in control cells and H1975 cells treated with erastin (10 µmol/L, 16-24 h) or RSL3 (5 µmol/L, 16-24 h) alone or in combination with NAT (20 µmol/L, 16-24 h). (K) Schematic presentation of how AA incorporation into the membrane and the enzymes involved. (L) Representative IB images of ACSL4, LPCAT3 and PLTP in H1975 cells treated with DMSO, or increasing concentration of NAT (10-20 µmol/L, 24 h). (M) Representative IB images of ACSL4, LPCAT3 and PLTP in H1975 cells following co-incubation with plasma exosome from healthy individuals or LUAD patients. (N) Percentages of the exosome from healthy individuals (n = 100) and LUAD patients (n = 100) that resulted in a different fold change (FC) in ACSL4, LPCAT3 and PTLP between the H1975 cells with or without co-incubation of plasma exosome. The relevant information of healthy individual and LUAD patient is summarized in Supplementary Table S7. (O-P) Protein (O) and mRNA (P) expression of ACSL4, LPCAT3 and PLTP in control and H1975 cells with TLX2 knocked down, treating with DMSO or NAT (20 µmol/L, 24 h). (Q) Representative images of ChIP using anti-TLX2 or control IgG antibodies for analyzing TLX2 binding within ACSL4, LPCAT3 and PTLP promoters in H1975 cells. The data are shown as the mean ± SD from three biological replicates (including IB). * P < 0.05, **P < 0.01 indicate statistical significance. NS, non-significance. Data in B-E, I, J, P were analyzed using one-way ANOVA tests. Data in F, H were analyzed using Student's t-tests. Data in N were analyzed using a ?2 test. Abbreviations: AA: arachidonic acid, Ctrl: control, NC: negative control, CDO1: cysteine dioxygenase type 1, CSAD: cysteine sulfinic acid decarboxylase, cir93: circRNA_101093, FABP3: fatty acid-binding protein 3, DAPI: 2-(4-Amidinophenyl)-6-indolecarbamidine dihydrochloride, NAT: N-arachidonoyl taurine, NS: non-significance, WT: wild type, LUAD: lung adenocarcinoma, ROS: reactive oxygen species, DMSO: dimethyl sulfoxide, RSL3: ras-selective lethal small molecule 3, PLTP: phospholipid transfer protein, LPCAT3: lysophosphatidylcholine acyltrasferase 3, ACSL4: acyl-CoA synthetase long-chain family member, FC: fold change, TLX2: T-cell leukemia homeobox 2, ChIP: chromatin immunoprecipitation, h: hours, IB: immunoblotting, ANOVA: analysis of variance
Supplier Page from Abcam for Anti-H-FABP antibody [EPR6126]