For most of medical history, scientists thought cells died in only a handful of ways: necrosis (traumatic destruction), apoptosis (programmed cellular suicide), or senescence (permanent arrest). Over the last decade, however, researchers have uncovered a surprising number of additional death pathways, and one of the most fascinating, with some of the most exciting therapeutic implications, is ferroptosis.
Ferroptosis is a form of regulated cell death caused by the uncontrolled oxidation of fats in cell membranes. In this process, the fatty components of the cell membrane accumulate oxidative damage, ultimately leading to membrane collapse and cell death. Unlike apoptosis, ferroptosis does not require the classic cellular suicide machinery; it operates through a distinct mechanism, which may provide a powerful approach for eliminating cancer cells that evade conventional death signals.
A significant challenge is that cancer cells are highly effective at defending themselves against ferroptosis. Elucidating the mechanisms underlying these defenses and identifying strategies to disrupt them represents a major frontier in cancer biology.
A new study published in Nature Cell Biology (2026) by researchers from the University of Würzburg and multiple international institutions has made a significant contribution to this frontier, revealing an unexpected connection between vitamin B2 and cancer cells’ ability to resist ferroptotic death.
Two Gatekeepers Standing Between Cancer Cells and Ferroptosis
Understanding this discovery requires knowledge of the mechanisms by which cancer cells typically protect themselves from ferroptosis.
The primary guardian is the enzyme GPX4 (glutathione peroxidase 4), which neutralizes oxidized lipids using glutathione as a cofactor. GPX4 serves as the cell’s first line of defense against lipid peroxidation, and several promising cancer therapies aim to block this enzyme to induce ferroptosis in tumors.
GPX4 is not the sole defense mechanism. An additional, independently functioning system relies on the protein FSP1 (ferroptosis suppressor protein 1), which was identified recently. Unlike GPX4, FSP1 does not directly degrade oxidized lipids; instead, it recycles lipid-soluble antioxidants, particularly ubiquinone (CoQ10) and vitamin K, utilizing NAD(P)H as an energy source. Through continuous regeneration of these antioxidants, FSP1 prevents the oxidative chain reaction that drives ferroptosis from escalating.
In many cancers, FSP1 serves as a critical backup system. Even when GPX4 is inhibited by experimental drugs, tumors exhibiting high FSP1 activity can persist. Therefore, understanding the regulatory mechanisms of FSP1 is essential for advancing cancer therapy.
The CRISPR Screen That Found Vitamin B2
The research team designed a genetic experiment in which a cancer cell line (HT1080 cells) was engineered to rely exclusively on FSP1 for survival by deleting GPX4. Using CRISPR-Cas9, they systematically knocked out approximately 3,000 potentially druggable genes, one at a time, to identify those essential for FSP1 function.
The two strongest hits from the screen were SCD1 (already known to influence ferroptosis through fatty acid biology) and, strikingly, RFK, the gene encoding riboflavin kinase, the enzyme that converts riboflavin (vitamin B2) into the molecule FMN (flavin mononucleotide).
This finding was unexpected. Riboflavin is a fundamental B vitamin present in dairy, eggs, meat, and leafy vegetables. It is essential for energy metabolism and serves as a precursor to two critical cellular cofactors: FMN (flavin mononucleotide) and FAD (flavin adenine dinucleotide). The connection between riboflavin and ferroptosis had not been previously established.
The link proved to be direct and mechanistic: FSP1 is a flavoprotein, a protein that requires FAD as a structural and functional cofactor. Without FAD, FSP1 not only loses activity but also becomes destabilized and degraded. The riboflavin–FMN–FAD pathway is essentially the supply chain that keeps FSP1 physically intact and enzymatically functional.
Without Vitamin B2, Cancer Cell Defenses Collapse
The researchers confirmed this connection through multiple approaches. When they depleted intracellular FAD by knocking out FADS, the enzyme responsible for the final step in FAD production, FSP1, was among the most severely depleted proteins in the cell, far more so than most other flavoproteins. Computer simulations of the FSP1 protein showed that without FAD bound to it, the protein’s backbone becomes unstable, particularly in the region (residues 282–300) that normally interacts with the cofactor.
When cancer cells were cultured in riboflavin-free medium for 72 hours, they became significantly more sensitive to GPX4 inhibition, and this sensitization was specifically dependent on FSP1. Cancer cell lines deficient in FSP1 exhibited no increased vulnerability upon riboflavin withdrawal, confirming that the riboflavin effect operates exclusively through the FSP1 pathway.
A particularly important finding concerns concentrations. Standard laboratory cell culture media contain riboflavin at 500–1,000 nanomolar, far above the actual concentration in human blood plasma, which is typically 10–20 nanomolar. This discrepancy matters enormously: when the researchers cultured cells at physiologically realistic riboflavin concentrations (below 20 nM), FSP1 expression dropped dramatically, and cells became much more sensitive to ferroptosis. Previous laboratory work may have been inadvertently conducted under conditions that artificially inflate FSP1 protection, potentially obscuring how vulnerable cancer cells actually are to ferroptotic attack in the real human body.
Enter Roseoflavin: A Bacterial Molecule That Acts as a Trojan Horse
Having established that riboflavin metabolism is essential for FSP1 function, the researchers turned to a natural compound that can hijack this system: roseoflavin, a riboflavin analog produced by Streptomyces bacteria.
Roseoflavin is structurally similar to riboflavin, allowing cells to import it, process it through the same metabolic pathway, and incorporate the resulting analog (roFAD) into FSP1 in place of normal FAD. The chemical modification of roFAD alters its electron-transfer properties; while it can stabilize FSP1’s structure, it cannot perform the enzymatic recycling function required for FSP1’s protective activity.
This effect functions as a Trojan horse: roseoflavin enters the cell as a nutritional building block, is converted and incorporated into FSP1, and renders the protein catalytically inactive while leaving it structurally intact. Externally, FSP1 appears unchanged, but it is no longer functional.
The functional consequences were striking. At physiologically realistic riboflavin concentrations, roseoflavin triggered ferroptosis in cancer cells at single-digit nanomolar concentrations, extremely potent by pharmacological standards. This effect was specific to FSP1: cells engineered to lack FSP1 showed no sensitization by roseoflavin, confirming the mechanism was on-target. The effect was reproduced across multiple cancer cell types, including breast cancer, lung cancer, and fibrosarcoma cells.
Importantly, when researchers purified FSP1 from cells treated with roseoflavin and assessed its enzymatic activity in a cell-free assay, they found that roseoflavin-loaded FSP1 was completely inactive and unable to perform the antioxidant recycling necessary for protection against ferroptosis.
A Broader Principle: Nutrients as Gatekeepers of Cell Death
The researchers draw a direct parallel between their findings and a well-established phenomenon in GPX4 biology: just as GPX4 requires selenium for its activity (selenium deficiency destabilizes GPX4), FSP1 requires riboflavin, which is converted to FAD. Both are micronutrients that act as functional gatekeepers for ferroptosis-resistance proteins.
This establishes a broader biological principle: micronutrient status doesn’t just affect overall health; it can fundamentally determine the extent to which cancer cells are resistant to a specific form of lethal oxidative stress. The implication is that the nutritional context, which varies between individual patients and tumor microenvironments, may determine whether ferroptosis-inducing therapies succeed or fail.
This observation may also help explain the inconsistent, sometimes paradoxical results of antioxidant supplementation in cancer clinical trials. If riboflavin and other B vitamins support FSP1-mediated protection in tumors, supplementation could inadvertently increase cancer cell resistance to ferroptotic mechanisms, warranting systematic clinical investigation.
Why This Research Matters for Cancer Therapy
Several therapeutic implications flow from these findings.
First, roseoflavin represents a genuinely novel mechanism for targeting FSP1, and the researchers note key advantages over conventional FSP1 inhibitors. Because roseoflavin enters cells through the normal riboflavin transporter (SLC52A2) and requires the riboflavin metabolic enzymes to generate its active form, any cancer cell that developed resistance to roseoflavin by mutating these components would simultaneously cripple its own riboflavin metabolism, compromising FAD and FMN production throughout the cell. This makes the evolution of resistance substantially harder than for drugs that target FSP1 directly.
Second, roseoflavin acts at nanomolar concentrations, well below physiological riboflavin levels and the concentrations of existing FSP1 inhibitors, potentially offering a larger therapeutic window.
Third, roseoflavin has already been studied as an antibiotic candidate in bacteria, providing some preliminary safety and pharmacokinetic data to build on.
This is early-stage research, conducted in cell lines and not yet in animals or humans. A considerable distance remains between these findings and clinical application. But for cancer biologists and oncology researchers, the identification of riboflavin metabolism as a druggable node in the FSP1 pathway opens a genuinely new angle of attack on one of cancer’s most important survival mechanisms.
The Takeaway
Cancer cells resist ferroptotic death through two independent systems: GPX4, which is well-studied, and FSP1, which is gaining increasing attention. The function of FSP1 depends critically on vitamin B2 (riboflavin) as a precursor for the cofactor FAD; in its absence, FSP1 degrades. The bacterial compound roseoflavin can mimic riboflavin, enter this pathway, and render FSP1 structurally intact but catalytically inactive, thereby increasing cancer cell vulnerability to ferroptosis.
The finding that standard laboratory conditions use riboflavin concentrations 25–50 times higher than what circulates in human blood is also a significant methodological insight, one that may reshape how ferroptosis research is conducted going forward.
eference
Skafar, V., de Souza, I., Ghosh, B., Ferreira dos Santos, A., Porto Freitas, F., Chen, Z., Sun, S., Donate Castillo, M., Nepachalovich, P., Seufert, L., Bothe, S., Tschuck, J., et al. & Friedmann Angeli, J.P. (2026). Riboflavin metabolism shapes FSP1-driven ferroptosis resistance. Nature Cell Biology, 28, 696–706. https://doi.org/10.1038/s41556-025-01856-x