KIF24 depletion induces Fe–S-dependent mitochondrial dysfunction and AMPK/p53/SLC7A11 mediated ferroptosis in gastric cancer

Technical Summary: KIF24 Depletion Induces Fe–S-Dependent Mitochondrial Dysfunction and AMPK/p53/SLC7A11 Mediated Ferroptosis in Gastric Cancer

Problem Statement
Gastric cancer (GC) remains a leading cause of cancer-related mortality, largely due to a scarcity of actionable therapeutic targets for advanced disease. While ferroptosis—an iron-dependent form of regulated cell death driven by lipid peroxidation—has emerged as a potential therapeutic vulnerability, aggressive tumor cells frequently enhance antioxidant defense systems to evade ferroptosis. The upstream regulators that sustain this ferroptosis resistance in GC, particularly those linking metabolic stress to the SLC7A11-dependent antioxidant network, remain incompletely defined.

Methodology
The study employed an integrative multi-omics and experimental validation strategy:

• Bioinformatics Prioritization: The authors utilized supervised and unsupervised machine learning algorithms (Support Vector Machine, k-Nearest Neighbors, Decision Tree) on transcriptomic data from The Cancer Genome Atlas (TCGA) and multiple Gene Expression Omnibus (GEO) datasets to identify ferroptosis-associated risk genes. Consensus clustering and Weighted Gene Co-expression Network Analysis (WGCNA) were used to refine candidate selection.
• Single-Cell and Spatial Analysis: Single-cell RNA sequencing (scRNA-seq) data (GSE167297) was analyzed to determine KIF24 expression distribution and cell-cell communication patterns (using CellChat) within the tumor microenvironment.
• In Vitro Functional Assays: Stable KIF24 knockdown (shRNA) and overexpression cell lines (MKN45, AGS) were generated. Phenotypic assays included CCK-8, colony formation, and cell growth curves. Ferroptosis markers were quantified via intracellular Fe²⁺ (FerroOrange), GSH (colorimetric assay), MDA (lipid peroxidation), and ROS (DCFH-DA/MitoSO). Transmission electron microscopy (TEM) was used to assess mitochondrial ultrastructure.
• Mechanistic Dissection: The study investigated the Fe–S cluster pathway by measuring mitochondrial aconitase activity and NFS1 localization via subcellular fractionation and Western blotting. Rescue experiments were conducted using pharmacological inhibitors (Ferrostatin-1, Deferoxamine, Compound C/AMPKi) and genetic silencing (siRNA against p53). Autophagy involvement was assessed via CYTO-ID staining and 3-MA treatment.
• In Vivo Validation: Subcutaneous xenograft models in nude mice and syngeneic models in immunocompetent BALB/c mice were used to evaluate tumor growth, immune infiltration (flow cytometry for CD4+/CD8+ T cells), and response to KIF24 depletion and rescue agents.
• Virtual Screening: A structure-based virtual screening campaign was performed on a library of ~19,800 compounds using homology modeling of KIF24, Glide docking, and MM-GBSA calculations to identify potential small-molecule inhibitors.

Key Contributions and Results

1. Identification of KIF24 as a Ferroptosis Resistance Factor: Machine learning prioritization identified KIF24 as a top candidate gene associated with malignant progression and ferroptosis resistance. Clinical validation confirmed KIF24 is upregulated in GC tissues and correlates with poor prognosis.
2. KIF24 Depletion Induces Ferroptosis: Functional assays demonstrated that KIF24 depletion significantly suppressed GC cell proliferation and colony formation while inducing hallmark ferroptotic features: Fe²⁺ accumulation, GSH depletion, lipid peroxidation (MDA), mitochondrial ROS production, and characteristic mitochondrial shrinkage with cristae loss. These effects were reversible by ferroptosis inhibitors (Fer-1) and iron chelators (DFO).
3. Mechanism: Fe–S-Dependent Mitochondrial Dysfunction: KIF24 loss was found to impair Fe–S cluster-dependent mitochondrial function. Specifically, it reduced mitochondrial aconitase activity and disrupted the mitochondrial localization of NFS1 (a core component of the ISC assembly machinery), leading to mitochondrial oxidative stress.
4. Signaling Axis: AMPK/p53/SLC7A11: The mitochondrial stress triggered by KIF24 depletion activated the AMPK signaling cascade. Activated AMPK phosphorylated p53, which subsequently repressed the transcription of SLC7A11 (the light chain of system xCT). This repression led to the failure of GSH synthesis and subsequent ferroptosis.
   • Rescue experiments confirmed that inhibiting AMPK (Compound C) or silencing p53 restored SLC7A11 expression, replenished GSH, and rescued cell viability.
   • The study explicitly ruled out ferritinophagy (autophagy-dependent ferritin degradation) as the primary mechanism, as autophagy inhibition did not rescue the phenotype, and ferritin levels remained unchanged.
5. Immune Microenvironment: Single-cell analysis and in vivo models suggested that KIF24 expression in epithelial cells correlates with reduced CD8+ T cell infiltration. KIF24 knockdown increased CD8+ T cell infiltration in syngeneic tumors, potentially linking ferroptosis induction to immune activation.
6. Therapeutic Candidates: Virtual screening identified several candidate compounds (including Verbascoside, Forsythoside A, TAS4464, Rutin, and Capreomycin Sulfate) with predicted binding affinity to KIF24 and cytotoxic activity against GC cells.

Significance and Claims
The paper claims to identify KIF24 as a previously unrecognized regulator of ferroptosis resistance in gastric cancer. It establishes a specific signaling axis where KIF24 maintains mitochondrial Fe–S homeostasis to suppress the AMPK/p53/SLC7A11 pathway, thereby protecting GC cells from ferroptosis.

The authors position KIF24 as a potential therapeutic target. They propose that targeting the KIF24–mitochondrial stress–AMPK–p53–SLC7A11 axis could sensitize gastric cancer cells to ferroptosis. While the study presents preliminary virtual screening hits, the authors modestly note that these compounds serve as starting points for future optimization rather than validated inhibitors, and that further studies (e.g., orthotopic models, direct target engagement) are required to fully establish the therapeutic relevance of KIF24-targeted strategies.