MUC1 drives ferroptosis resistance in ICC via Src-mediated FSP1 deubiquitination and myristoylation

Abstract

Background: Intrahepatic cholangiocarcinoma (ICC) exhibits poor prognosis and limited therapeutic options. Ferroptosis represents a promising therapeutic strategy, yet resistance mechanisms remain poorly understood. This study investigated the role of mucin 1 (MUC1) in regulating ferroptosis sensitivity in ICC.

Methods: Bioinformatic analyses of GEO and TCGA datasets identified ferroptosis-related factors in ICC. MUC1 expression was validated in ICC cell lines and clinical specimens. Ferroptosis sensitivity was assessed through RSL3-induced cell death assays, lipid peroxidation measurements, and iron detection. Mechanistic studies employed immunoprecipitation-mass spectrometry, co-immunoprecipitation, kinase assays, and deubiquitination assays. In vivo efficacy was evaluated using subcutaneous tumor models.

Results: MUC1 was identified as a critical ferroptosis suppressor in ICC. MUC1 overexpression conferred RSL3 resistance by inhibiting lipid peroxidation and reducing ferrous iron accumulation, independent of the GPX4-glutathione pathway. Mechanistically, MUC1 recruited Src kinase, which phosphorylated deubiquitinating enzyme ubiquitin-specific protease 10 (USP10) at tyrosines 359 and 364, enhancing ferroptosis suppressor protein 1 (FSP1) deubiquitination at lysine 246 and stabilizing FSP1 protein. Concurrently, Src phosphorylated N-myristoyltransferase 1 (NMT1) at tyrosine 41, augmenting FSP1 membrane localization through myristoylation. This dual mechanism potentiated the FSP1- coenzyme Q10 (CoQ10) antioxidant system. MUC1 knockdown significantly enhanced ferroptotic sensitivity in vitro and suppressed tumor growth in vivo.

Conclusions: MUC1 orchestrates ferroptosis resistance in ICC through the Src-USP10/NMT1-FSP1 axis. Targeting this signaling cascade represents a potential therapeutic strategy for overcoming ferroptosis resistance in ICC.

Key points: MUC1 suppresses ferroptosis in ICC via Src-mediated post-translational modifications. Src phosphorylation of USP10 stabilizes FSP1 by removing K48-linked polyubiquitin. Src activates NMT1 to enhance FSP1 myristoylation and membrane localization.

Keywords: MUC1; ferroptosis; intrahepatic cholangiocarcinoma; post‐translational modification.

FIGURE 1

MUC1 is overexpressed in ICC and identified as a potential ferroptosis regulator. (A) Heatmap displaying differentially expressed genes between ICC and normal tissues identified through integration of GEO and TCGA datasets using Robust Rank Aggregation algorithm. (B) GSEA of ferroptosis‐related gene sets in MUC1‐high versus MUC1‐low ICC samples. (C) qRT‐PCR analysis of MUC1 mRNA expression in HiBEC and four ICC cell lines (RBE, HuCCT1, SNU‐1079, HCCC‐9810). (D) Western blot assessment of MUC1 protein expression in HiBEC and ICC cell lines. (E) qRT‐PCR analysis comparing MUC1 mRNA levels in 10 paired ICC tumour tissues and adjacent non‐tumour tissues. (F) Western blot analysis of MUC1 protein expression in representative paired clinical samples. (G) Representative immunohistochemical staining of MUC1 in ICC tumour tissues and adjacent non‐tumour tissues. Representative IHC images shown at 400× magnification; scale bar = 50 µm. Data are shown as the means ± SD, the significant level was identified by *p < .05; ***p < .001.

FIGURE 2

MUC1 confers resistance to ferroptosis in ICC cells through multiple mechanisms. (A) IC50 values of RSL3 (24 h treatment) in MUC1‐overexpressing and MUC1‐knockdown HuCCT1 cells compared with controls. Control experiments utilised sh‐NC (negative control short hairpin RNA), which does not target any known gene sequences and serves as an appropriate control for gene knockdown studies. (B) Cell viability of RSL3‐treated (2 µM, 24 h) sh‐MUC1 HuCCT1 cells following 1 h pre‐treatment with cell death inhibitors: ferroptosis inhibitor ferrostatin‐1 (5 µM), apoptosis inhibitor Z‐VAD‐FMK (30 µM), necroptosis inhibitor necrostatin‐1 (20 µM), autophagy inhibitor 3‐MA (5 mM) or pyroptosis inhibitor Ac‐YVAD‐cmk (20 µM). (C) Colony formation assay in HuCCT1 cells across treatment groups. (D) Ferrorrange staining for Fe²⁺ detection in HuCCT1 cells across treatment groups. Representative images shown at 200× magnification; scale bar = 25 µm. (E) Lipid peroxidation levels measured by MDA assay. (F) Intracellular ROS quantification by flow cytometry using DCFH‐DA staining. (G) Transmission electron micrographs showing mitochondrial morphology in HuCCT1 cells across treatment groups. Representative images at different magnifications: 1500× and 5000×; scale bars = 5 and 1 µm. (H) Intracellular GSH levels in HuCCT1 cells across treatment groups. (I–K) In vivo subcutaneous tumour model using HuCCT1 cells: (I) tumour growth curves, (J) representative tumour images and (K) endpoint tumour weights across treatment groups. (L) H&E, MUC1 and 4HNE immunohistochemical staining of tumour sections from different treatment groups. Representative images shown at 200× magnification; scale bar = 100 µm. Data are shown as the means ± SD, the significant level was identified by *p < .05; ***p < .001; n.s.: no significant.

FIGURE 3

MUC1 regulates ferroptosis through FSP1 stabilisation and membrane localisation. (A) Western blot analysis of ferroptosis‐related proteins in HuCCT1 cells with or without MUC1 overexpression. (B) Ratio of reduced CoQ10 to total CoQ10 in HuCCT1 cells with MUC1 knockdown or overexpression, determined by HPLC. (C) SYTOX Green staining assessing cell death in HuCCT1 cells pretreated with 4‐CBA (3 mM, 24 h) followed by different RSL3 concentration exposure with or without MUC1 overexpression. (D–F) Ferroptotic phenotypes in HuCCT1 cells treated with RSL3 (2 µM, 24 h): (D) ferrous ion accumulation by Ferrorrange staining, representative images shown at 200× magnification; scale bar = 25 µm, (E) lipid peroxidation by MDA assay and (F) ROS levels by DCFH‐DA staining. (G) Western blot analysis of FSP1 protein degradation in HuCCT1 cells treated with cycloheximide (50 µg/mL) for indicated time periods following MUC1 knockdown. (H) Western blot analysis of FSP1 protein levels in HuCCT1 cells with MUC1 knockdown treated with or without MG132 (20 µM, 6 h). (I) Immunoprecipitation assay detecting FSP1 ubiquitination levels in HuCCT1 cells with MUC1 knockdown under MG132 (20 µM, 6 h) treatment. (J) Immunofluorescence co‐localisation analysis of FSP1 (green) and ATP1A1 (red) in HuCCT1 cells with MUC1 knockdown. Representative images shown at 600× magnification; scale bar = 10 µm. Data are shown as the means ± SD, the significant level was identified by *p < .05; **p < .01; ***p < .001; n.s.: no significant.

FIGURE 4

MUC1 interacts with and activates Src kinase to regulate FSP1 expression. (A) MUC1 interactome identification in HuCCT1 cells: left panel shows Coomassie blue staining of immunoprecipitated proteins; right panel displays tabulated IP–MS results of MUC1‐associated proteins. (B) Co‐immunoprecipitation assays demonstrating interaction between MUC1 and Src in HuCCT1 cells. (C) In vitro GST pull‐down assay confirming direct binding between recombinant MUC1 and Src proteins. (D) Immunofluorescence co‐localisation analysis of MUC1 (green) and Src (red) in HuCCT1 cells. Representative images shown at 600× magnification; scale bar = 10 µm. (E) Western blot analysis of Src phosphorylation at Tyr419 in HuCCT1 cells with MUC1 manipulation. (F) Western blot analysis of FSP1 protein levels in HuCCT1 cells with MUC1 overexpression treated with or without Src inhibitors. (G) Ferrorrange staining for ferrous ion detection in HuCCT1 cells under indicated treatments. Representative images shown at 200× magnification; scale bar = 25 µm. (H) Lipid peroxidation levels measured by MDA assay. (I) Flow cytometric analysis of intracellular ROS levels using DCFH‐DA staining. (J) Representative images of subcutaneous tumours from different treatment groups: control, RSL3 treatment and RSL3 + dasatinib combination therapy. (K) Quantitative analysis of tumour weights across treatment groups. Data are shown as the means ± SD, the significant level was identified by *p < .05; **p < .01; ***p < .001.

FIGURE 5

Src phosphorylates USP10 to regulate FSP1 stability. (A) Co‐immunoprecipitation assays demonstrating interaction between Src and USP10 in HuCCT1 cells. (B) In vitro GST pull‐down assay confirming direct binding between recombinant Src and USP10 proteins. (C–F) Domain mapping experiments identifying critical interaction regions: (C) schematic of Src domain mutants, (D) GST pull‐down assay with Src mutants, (E) schematic of USP10 domain mutants, (F) GST pull‐down assay with USP10 mutants. (G) Phos‐tag SDS‐PAGE analysis of USP10 phosphorylation status following in vitro kinase reactions with purified Src, MUC1 and Src inhibitor. (H) Co‐immunoprecipitation assays examining Src–USP10 interaction in HuCCT1 cells with MUC1 knockdown.

FIGURE 6

Src‐mediated USP10 phosphorylation and its effect on FSP1 deubiquitination. (A) Prediction of potential Src phosphorylation sites on USP10 using Netphos 3.1. (B) Molecular docking analysis of USP10 (wild‐type and phosphorylated Y359/Y364) with FSP1 protein using AlphaFold‐Multimer. (C) Western blot analysis examining His‐tagged FSP1 levels in USP10‐knockdown HEK293 cells expressing wild‐type or mutant (Y359A/Y364A) Flag‐USP10. (D and E) Assessment of FSP1 polyubiquitination in cells expressing wild‐type or mutant USP10 following MG132 treatment (10 µM, 6 h). (F) Co‐immunoprecipitation analysis of FSP1 with wild‐type or mutant USP10. (G) In vitro enzymatic activity measurement of purified wild‐type and mutant USP10 using Ub‐AMC substrate. (H) Analysis of ubiquitin chain types on FSP1 using ubiquitin vectors containing single lysine mutations. (I) Identification of FSP1 ubiquitination sites through point mutation analysis. Data are shown as the means ± SD, with statistical significance indicated as n.s.: not significant.

FIGURE 7

Src and NMT1 interaction and their effects on FSP1 localisation. (A) Co‐immunoprecipitation analysis of Src and NMT1 in HuCCT1 cells. (B) GST pull‐down assay with purified Src and NMT1 proteins. (C) Schematic illustration of Src domain constructs used for interaction mapping. (D) GST pull‐down assay using different Src domain constructs with NMT1. (E) Phos‐tag gel electrophoresis of NMT1 following in vitro kinase reactions under various conditions including Src, Src inhibitor and MUC1. (F) Phos‐tag analysis of NMT1 in cells with Src manipulation and inhibitor treatment. (G) Phos‐tag analysis comparing wild‐type and Y41A mutant NMT1 after Src kinase reaction. (H) Co‐immunoprecipitation analysis between FSP1 and either wild‐type or Y41A mutant NMT1. (I) Fluorescence microscopy visualisation of cellular myristoylation (red) and FSP1 distribution (green) in cells expressing wild‐type or Y41A mutant NMT1. Representative images shown at 600× magnification; scale bar = 10 µm.

FIGURE 8

MUC1 promotes ferroptosis resistance in ICC cells through dual regulatory mechanisms targeting FSP1. MUC1 overexpression in ICC cells results in Src kinase activation, leading to the phosphorylation of two critical regulatory proteins: (1) USP10 at tyrosine residues Y359/Y364, which enhances its deubiquitinase activity and strengthens its binding affinity toward FSP1, thereby stabilising FSP1 protein levels in the cytoplasm; and (2) NMT1 at tyrosine residue Y41, which increases its myristoyltransferase activity and facilitates N‐terminal myristoylation and subsequent membrane localisation of FSP1. Membrane‐localised FSP1 efficiently catalyses CoQ10 reduction, establishing a lipid peroxide‐trapping system that neutralises lipid peroxidation and prevents ferroptotic cell death. Through these dual regulatory mechanisms (protein stabilisation and subcellular localisation), the FSP1–CoQ10 antioxidant axis is potentiated by MUC1, conferring ferroptosis resistance to ICC cells.