A retraction notice in Experimental & Molecular Medicine has withdrawn a high-profile claim about how the cancer-linked NOTCH signaling pathway interacts with mitochondria and reshapes tumor-initiating cells. The notice concerns the paper titled “NOTCH localizes to mitochondria through the TBC1D15-FIS1 interaction and is stabilized via blockade of E3 ligase and CDK8 recruitment to reprogram tumor-initiating cells,” authored by H.Y. Choi, Y. Zhu, X. Zhao and colleagues. Published in 2026 under the DOI 10.1038/s12276-026-01834-9, the retraction means that the study’s conclusions should no longer be treated as reliable evidence, even though the biological questions it raised remain important to cancer researchers.
The original paper focused on NOTCH, a family of membrane-associated receptors that regulate cell fate, development, tissue renewal and cancer progression. In its canonical role, NOTCH signaling begins when a neighboring cell presents a ligand that binds the NOTCH receptor. This interaction triggers a sequence of proteolytic cleavages, ultimately releasing the NOTCH intracellular domain. That fragment can enter the nucleus and cooperate with transcriptional regulators to activate genes involved in proliferation, survival and cellular identity. Aberrant NOTCH activity has been implicated in several malignancies, including leukemias, breast cancer, brain tumors and other solid cancers. The retracted study proposed that NOTCH also has a direct mitochondrial role, extending the pathway beyond its well-known function in the nucleus.
Mitochondria are often described as the energy-producing structures of cells, but they also control metabolism, redox balance, calcium handling and programmed cell death. Cancer cells frequently reconfigure mitochondrial activity to survive hostile conditions, maintain rapid growth and resist therapy. Tumor-initiating cells, sometimes referred to as cancer stem-like cells, are a particularly important population because they can self-renew, generate diverse tumor cell types and contribute to relapse. The paper claimed that NOTCH could be redirected to mitochondria in these cells, where it might influence the organelle’s behavior and help preserve a tumor-initiating state. If validated, such a mechanism would have connected developmental signaling, mitochondrial biology and cancer persistence in a striking way.
Central to the reported mechanism was the interaction between TBC1D15 and FIS1. FIS1 is an outer mitochondrial membrane protein associated with mitochondrial dynamics, including the division of mitochondria into smaller units. TBC1D15 is a protein involved in membrane trafficking and has been linked to regulation of mitochondrial morphology through interactions with FIS1 and related cellular machinery. According to the retracted claim, this molecular pairing served as a route by which NOTCH localized to mitochondria. In technical terms, the proposed interaction would have placed a signaling protein traditionally studied at the plasma membrane and in the nucleus within a mitochondrial protein network, potentially allowing NOTCH to affect mitochondrial organization or signaling from inside the organelle’s immediate environment.
The study also described a stabilization mechanism involving an E3 ubiquitin ligase and CDK8. E3 ligases are enzymes that attach ubiquitin molecules to specific proteins, marking them for altered activity, transport or destruction by the proteasome. This system is one of the cell’s principal methods for controlling protein abundance. CDK8, or cyclin-dependent kinase 8, is a transcriptional regulator that functions as part of the Mediator complex and can influence gene expression programs associated with development and cancer. The paper proposed that blocking the recruitment of an E3 ligase and CDK8 to NOTCH prevented its degradation or inactivation, thereby stabilizing the protein and supporting a cellular reprogramming process linked to tumor initiation.
That proposed chain of events was potentially significant because it suggested that NOTCH could operate through two interconnected layers of control: a mitochondrial localization step and a protein-stabilization step. In principle, such a pathway might help explain how tumor-initiating cells maintain both metabolic flexibility and self-renewal capacity. It also raised the possibility that disrupting the TBC1D15-FIS1 interaction, altering mitochondrial recruitment or restoring NOTCH turnover could weaken cancer cells that depend on this state. However, these therapeutic implications relied entirely on the accuracy and reproducibility of the reported molecular interactions. Once a paper is retracted, those proposed connections must be regarded as unresolved rather than as a confirmed map of cancer biology.
A retraction is different from a routine correction. Corrections usually address limited errors that do not overturn the central conclusions of a study. Retraction, by contrast, signals that the publication should not be relied upon as part of the established scientific record. The citation supplied for this case identifies the publication as a “Retraction Note,” but it does not, by itself, provide the detailed reason or reasons for the decision. Without an official explanation specifying whether the issue involved data integrity, methodology, interpretation, authorship or another concern, it would be inappropriate to infer what went wrong. The responsible scientific position is therefore to distinguish clearly between the hypotheses described in the original title and results that have been independently confirmed.
The withdrawal is especially relevant in a field where molecular claims can rapidly influence laboratory priorities, drug-development strategies and public expectations about cancer treatments. A mechanism that appears to connect NOTCH, mitochondria and tumor-initiating cells may attract attention because it offers a single narrative linking several major areas of cancer research. Yet complex biological systems are vulnerable to experimental artifacts. Apparent protein interactions can be affected by overexpression, antibody specificity, cell-line differences, subcellular fraction contamination or conditions that do not reflect tumors in living organisms. Demonstrating mitochondrial localization, for example, generally requires complementary approaches such as high-resolution imaging, biochemical fractionation with appropriate purity controls and independent confirmation using endogenous proteins.
The retraction does not erase the importance of studying NOTCH signaling, mitochondrial dynamics or tumor-initiating cells. It does mean that the specific TBC1D15-FIS1 localization model and the proposed blockade of E3 ligase and CDK8 recruitment cannot currently serve as dependable foundations for new conclusions. Future research will need to establish whether any form of NOTCH-mediated mitochondrial regulation exists, whether it occurs in physiologically relevant cancer models and how it relates to protein degradation and transcriptional control. For now, the most accurate message is one of scientific caution: an intriguing mechanism linking a major developmental pathway to cancer-cell metabolism has been removed from the trusted literature, leaving the underlying question open for rigorous investigation.
Subject of Research: NOTCH signaling, mitochondrial localization, TBC1D15-FIS1 interaction, protein stabilization and tumor-initiating cells in cancer
Article Title: Retraction Note: NOTCH localizes to mitochondria through the TBC1D15-FIS1 interaction and is stabilized via blockade of E3 ligase and CDK8 recruitment to reprogram tumor-initiating cells.
Article References: Choi, H.Y., Zhu, Y., Zhao, X. et al. Retraction Note: NOTCH localizes to mitochondria through the TBC1D15-FIS1 interaction and is stabilized via blockade of E3 ligase and CDK8 recruitment to reprogram tumor-initiating cells. Exp Mol Med (2026). https://doi.org/10.1038/s12276-026-01834-9
Image Credits: AI Generated
DOI: 10.1038/s12276-026-01834-9
Keywords: NOTCH signaling, mitochondria, TBC1D15, FIS1, E3 ligase, CDK8, tumor-initiating cells, cancer biology, retraction, molecular oncology
