A DMT1-dependent iron-endoplasmic reticulum-extracellular matrix axis regulates cancer cell invasion

Technical Summary: A DMT1-Dependent Iron–Endoplasmic Reticulum–Extracellular Matrix Axis Regulates Cancer Cell Invasion

Problem Statement
Intracellular iron homeostasis is recognized as a regulator of cancer cell behavior, yet the mechanisms by which iron distribution influences extracellular matrix (ECM) organization and invasion remain poorly understood. While total cellular iron content is often cited as a driver of tumor progression, the specific impact of intracellular iron trafficking on structural integrity and context-dependent invasion is unclear. Conventional two-dimensional (2D) cultures fail to reproduce the spatial constraints, multicellular organization, and metabolic gradients of tumors, potentially obscuring invasion-related phenotypes. This study investigates whether the loss of divalent metal transporter 1 (DMT1/SLC11A2), a key iron transporter, disrupts intracellular iron distribution in a manner that compromises endoplasmic reticulum (ER) function and ECM integrity, thereby altering cancer cell invasion in three-dimensional (3D) tumor models.

Methodology
The study utilized MDA-MB-231 triple-negative breast cancer (TNBC) cell lines, including wild-type (WT), DMT1 knockout (KO), and stable DMT1-GFP overexpression (OE) variants. Phenotypes were compared across 2D monolayers and liquid-overlay 3D spheroids of varying sizes: small (LOS, ~300–500 µm) and large (LOL, ~500–800 µm).

Key experimental approaches included:

• Transcriptomics: RNA sequencing (RNA-seq) to characterize transcriptional baselines across 2D, LOS, and LOL conditions, and to assess DMT1-dependent gene expression changes.
• Iron Homeostasis Analysis: Atomic absorption spectroscopy (AAS) for total iron content; FerroOrange fluorescence for labile iron pool (LIP) quantification; and Western blotting for iron-regulatory proteins (FTH1, IRP2, ZIP8).
• Structural and Morphological Imaging: Optical coherence tomography (OCT) and two-photon microscopy to assess spheroid sphericity, density, and internal architecture. Second-harmonic generation (SHG) microscopy was used to visualize fibrillar collagen, while Sircol assays measured soluble collagen.
• Functional Assays: Wound-healing and Transwell Matrigel invasion assays for 2D motility; 3D invasion assays embedding spheroids in high-density Matrigel or type I collagen matrices, stimulated with TGF-β1.
• Pharmacological Perturbation: Treatment with the iron chelator deferoxamine (DFO) and the ER stress inducer tunicamycin to dissect mechanistic links.
• Clinical Data Correlation: Analysis of patient-derived transcriptomic (METABRIC) and proteomic (CPTAC) datasets to evaluate correlations between iron regulators, ER stress markers, and ECM components.

Key Results

1. Context-Dependent Transcriptional Baselines: Spheroid size and dimensionality, rather than DMT1 expression alone, defined the dominant transcriptional variance. Large (LOL) spheroids exhibited distinct hypoxia-associated and metabolic signatures compared to 2D cultures, making them the preferred model for interrogating invasion.
2. DMT1 Loss Disrupts Iron Distribution: DMT1 KO cells exhibited reduced total iron content but a paradoxical increase in the labile iron pool (LIP) in both 2D and 3D cultures. This was accompanied by upregulation of iron-regulatory proteins (FTH1, IRP2, ZIP8).
3. Structural Integrity and ECM Disruption: In 3D LOL spheroids, DMT1 KO cells formed loose, less spherical aggregates with reduced cell density, whereas DMT1 OE cells formed compact structures. This structural defect correlated with impaired collagen production: DMT1 KO cells showed reduced COL1A1 expression, decreased fibrillar collagen (SHG), and downregulated ECM-associated gene signatures, despite elevated levels of some ER-resident enzymes (PDI, P4HA1).
4. Divergent Motility Phenotypes: DMT1 loss suppressed cell motility and invasion in 2D assays (wound healing and Transwell). Conversely, in 3D matrix-embedded spheroids, DMT1 KO cells displayed significantly enhanced invasive outgrowth into surrounding Matrigel and collagen matrices compared to WT and OE cells.
5. ER Stress as a Mechanistic Link: DMT1 KO cells exhibited pronounced upregulation of ER stress markers (BiP/GRP78, ATF6, IRE1α). Pharmacological induction of ER stress with tunicamycin phenocopied the loose spheroid architecture and invasive behavior of DMT1 KO cells, suggesting ER stress is sufficient to drive ECM destabilization and invasion.
6. EMT and Clinical Correlation: In 3D culture, DMT1 KO cells showed increased expression of EMT-related proteins (integrin β1, ICAM1, TGF-β1). Analysis of patient datasets revealed coordinated relationships among iron regulation, ER stress, and ECM markers, supporting the biological relevance of the identified axis in human tumors.

Significance and Claims
The paper identifies an "iron–ER–ECM axis" through which intracellular iron homeostasis regulates cancer cell invasion. The authors claim that total cellular iron content is insufficient to predict invasive behavior; rather, the intracellular distribution of iron is the critical determinant. Specifically, DMT1 is shown to support collagen synthesis and maintain ECM integrity by preserving ER function and proper iron partitioning.

The study challenges the assumption that lowering bulk iron uniformly suppresses invasive phenotypes. Instead, it demonstrates that DMT1 loss, by disrupting iron distribution and inducing ER stress, promotes the formation of loosely aggregated spheroids and enhances invasion in 3D tumor models, despite reduced total iron levels. The findings suggest that intracellular iron trafficking, rather than simple iron abundance, is a potential therapeutic target for limiting cancer cell invasion. The authors conclude that ER stress acts as a mechanistic bridge between altered iron handling and ECM destabilization, highlighting the importance of context (2D vs. 3D) in evaluating cancer cell behavior.