Copper cell death splits into two distinct routes converging on mitochondria

Copper has long been known as an essential trace element, a redox cofactor tucked into the machinery of mitochondrial respiration and antioxidant defense. But a new open-access study published in Molecular Cancer on 24 October 2026 argues that the way copper kills cells is not one story but two. A research team led by Jana Bachmann and Ossia Eichhoff reports that two copper-active compounds, elesclomol and neocuproine, trigger mechanistically distinct forms of cuproptosis, the copper-dependent cell death program, which ultimately converge on the same mitochondrial collapse.

The team built its case on the previously established mechanism in which FDX1, a mitochondrial enzyme, reduces the copper(II) delivered by elesclomol to copper(I), driving rapid oligomerization of lipoylated mitochondrial proteins and proteotoxic stress. The researchers now designate this FDX1-dependent pathway as canonical cuproptosis, or c-CPX. Neocuproine, by contrast, induces a copper-dependent death program that proceeds without FDX1, which they term non-canonical cuproptosis, or nc-CPX. Despite this independence, neocuproine still rapidly induces oligomerization of lipoylated DLAT, an early point of convergence between the two routes.

The methods behind the finding were unusually broad. The study integrated limited-proteolysis proteomics, which tracks compound-responsive changes in protein structural accessibility, with whole-genome CRISPRa activation screening, transcriptomic analysis of 436 cancer cell lines from the Cancer Cell Line Encyclopedia, mass spectrometry-based measurement of cellular and tumor copper abundance, genetic perturbation experiments, mitochondrial functional assays, and mitophagy analyses. Pathway enrichment and functional validation were then used to identify the adaptive survival programs and compound-specific copper-handling mechanisms that cells deploy under copper stress.

One of the most striking results concerns which cancer cells are vulnerable. Sensitivity to both neocuproine and elesclomol was associated with enrichment of oxidative phosphorylation transcriptional programs across the large cell-line panel. Neocuproine-sensitive cells, however, additionally displayed significantly higher mesenchymal, or epithelial-mesenchymal transition, scores across multiple independent gene-expression signatures. That links an OXPHOS-enriched, mesenchymal cell state to increased susceptibility to neocuproine-induced copper-dependent death, a candidate biomarker signature the authors flag for further investigation rather than a validated clinical predictor.

Limited-proteolysis profiling identified ATOX1, a copper(I) chaperone, among the prominent neocuproine-responsive proteins. Functional validation showed that depleting ATOX1 partially attenuates neocuproine-induced cytotoxicity but not elesclomol toxicity, while ATOX1 loss itself promotes DLAT oligomerization. Together, the authors argue, this supports a specific contribution of ATOX1-dependent copper(I) handling to the non-canonical pathway, distinguishing it from the FDX1-mediated route taken by elesclomol.

CRISPRa screening further revealed that the adaptive programs enriched during neocuproine exposure predominantly involve proteasomal and lysosomal pathways, whereas elesclomol is associated with distinct signaling and bioenergetic programs, including PI3K, JAK-STAT, and MAPK signaling, oxidative phosphorylation, and glutathione metabolism. The functional relevance of the neocuproine-associated pathways was supported by enhanced cytotoxicity when proteasomal or lysosomal function was pharmacologically inhibited, suggesting that blocking cellular cleanup systems could amplify this form of copper killing.

Despite their divergent initiation mechanisms, both canonical and non-canonical cuproptosis converge on a shared sequence: early oligomerization of lipoylated mitochondrial proteins, followed by loss of iron-sulfur proteins, respiratory-chain dysfunction, mitochondrial remodeling, impaired mitochondrial quality control, and ultimately respiratory failure. Both compounds also induce cytoplasmic and mitochondrial reactive oxygen species and late apoptotic features, but the study presents these as consequences rather than causes of the death program.

That distinction rests on a series of careful controls. Continuous exposure to N-acetylcysteine reduced the labile intracellular copper(I) signal and rescued cytotoxicity, whereas NAC pretreatment followed by washout did not protect against subsequent neocuproine or elesclomol exposure. Combined with the lack of rescue by caspase inhibition and the delayed appearance of apoptotic features, the authors conclude that ROS accumulation and apoptosis are predominantly downstream consequences rather than defining initiating mechanisms of cuproptosis.

The researchers frame canonical and non-canonical cuproptosis not as separate downstream death programs but as distinct upstream copper-handling routes that converge on a shared mitochondria-centered proteotoxic program. They propose FDX1 dependency, ATOX1-dependent copper handling, the EMT/OXPHOS cell state, and proteostasis pathways as candidate biomarkers and therapeutic vulnerabilities for further study. The scope of the evidence deserves emphasis: the findings derive from cancer cell lines, genetic screens, and preclinical assays, and the authors themselves present the biomarker and vulnerability candidates as hypotheses requiring validation, not as established clinical results. Whether these two copper routes can be exploited therapeutically in patients remains an open question, but the study provides a mechanistic framework for the next round of experiments.

Subject of Research: Cuproptosis

Article Title: FDX1-independent non-canonical cuproptosis establishes a parallel mitochondria-centric death program

Article References: Bachmann, J. J., Molvetti, M. G., Trajkovic, A., Haunerdinger, V., Kitanovic, A., Schöberl, T., Dünnebacke, M., Leytens, A., Turko, P., Stylios, D., Zillikens, H., Look, T., Villano, D., Tellenbach, F., Shukla, P., Esposito, M., Günther, D., Kleele, T., Picotti, P., … Eichhoff, O. M. (2026). FDX1-independent non-canonical cuproptosis establishes a parallel mitochondria-centric death program. Molecular Cancer. https://doi.org/10.1186/s12943-026-02821-4

Image Credits: AI Generated

DOI: 10.1186/s12943-026-02821-4

Keywords: FDX1-independent, non-canonical, cuproptosis, establishes, parallel, mitochondria-centric, death, program, scientific research