Back

Vitamin B2 metabolism promotes FSP1 stability to prevent ferroptosis

Deol, K. K.; Harris, C. A.; Tomlinson, S. J.; Doubravsky, C. E.; Mathiowetz, A. J.; Olzmann, J. A.

2025-08-06 cell biology
10.1101/2025.08.05.668752 bioRxiv
Show abstract

Ferroptosis, a regulated form of cell death driven by excessive lipid peroxidation, has emerged as a promising therapeutic target in cancer. Ferroptosis suppressor protein 1 (FSP1) is a critical regulator of ferroptosis resistance, yet the mechanisms controlling its expression and stability remain mostly unexplored. To uncover regulators of FSP1 abundance, we conducted CRISPR-Cas9 screens utilizing a genome-edited, dual-fluorescent FSP1 reporter cell line, identifying both transcriptional and post-translational mechanisms that determine FSP1 levels. Notably, we identified riboflavin kinase (RFK) and FAD synthase (FLAD1), enzymes which are essential for synthesizing flavin adenine dinucleotide (FAD) from vitamin B2, as key contributors to FSP1 stability. Biochemical and cellular analyses revealed that FAD binding is critical for FSP1 activity. FAD deficiency, and mutations blocking FSP1-FAD binding, triggered FSP1 degradation via a ubiquitin-proteasome pathway that involves the E3 ligase RNF8. Unlike other vitamins that inhibit ferroptosis by scavenging radicals, vitamin B2 supports ferroptosis resistance through FAD cofactor binding, ensuring proper FSP1 stability and function. This study provides a rich resource detailing mechanisms that regulate FSP1 abundance and highlights a novel connection between vitamin B2 metabolism and ferroptosis resistance with implications for therapeutic strategies targeting FSP1 in cancer.

Published in Nature Structural & Molecular Biology (predicted rank #7) · training set

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.