Ferroptosis Suppressor Protein 1 (FSP1/AIFM2): A Quinone Reductase Linking Membrane Redox Defense and Reproductive Competence
1. Introduction
Ferroptosis suppressor protein 1 (FSP1), encoded by the AIFM2 gene and previously catalogued as AMID or PRG3, is a flavin adenine dinucleotide (FAD)-containing oxidoreductase that has moved from the margins of apoptosis research to the center of redox cell biology. The protein was originally annotated as a pro-apoptotic, p53-inducible homologue of apoptosis-inducing factor. The recent work established that it behaves as a rotenone-sensitive, type-II NADH: ubiquinone oxidoreductase rather than as a nuclease (Elguindy and Nakamaru-Ogiso, 2015). That biochemical description acquired physiological meaning in 2019, when two independent expression-cloning and CRISPR screens identified the same flavoprotein as the factor that allows cells to survive genetic or pharmacological loss of glutathione peroxidase 4 (Doll et al., 2019). The gene was renamed accordingly, and the FSP1–coenzyme Q10–NAD(P)H module is now recognized as the second mainstay of ferroptosis surveillance (Bersuker et al., 2019).
Among the redox enzymes, FSP1 remains unusual as it is not a peroxidase. It does not repair oxidized phospholipids but continuously regenerates a small lipophilic radical-trapping antioxidant inside the membrane, so that lipid peroxyl radicals are intercepted before propagation begins. This mechanistic separation explains why it operates in parallel with, rather than downstream of, the cysteine–glutathione–GPX4 axis (Jiang et al., 2021). The distinction matters clinically because tumors and injured tissues can shift between the two arms depending on nutrient supply, oxygen tension, and mutational background (Dixon and Olzmann, 2024).
The growing interest in FSP1 has increased due to several reasons, including its structural biology, which has resolved how a single flavoprotein reduces chemically diverse quinones. It was shown to sustain a warfarin-insensitive vitamin K cycle, solving a question that has been open in vitamin biology for many years. Lastly, in vivo genetic models have demonstrated that FSP1 is not merely a backup system but an independent requirement for tumor growth. Besides this progress, reproductive biologists have begun to report that FSP1 governs oocyte meiotic competence and porcine early embryonic development, extending the protein’s relevance from oncology into gamete and embryo physiology (Wang et al., 2024). This review examines FSP1 from a protein-centered perspective, covering its structure, catalysis, regulation, disease associations, reproductive functions, detection methods, and therapeutic prospects.
2. Structural Organization of FSP1
Crystallographic analysis of vertebrate FSP1 in substrate-free and ubiquinone-bound states has defined a three-part architecture; an FAD-binding domain, an NAD(P)H-binding domain, and a distinctive carboxy-terminal domain (Lv et al., 2023). The first two modules are shared with apoptosis-inducing factor and with bacterial NADH oxidoreductases, which accounts for the historical misassignment of the protein to the apoptotic machinery. The carboxy-terminal domain is the element that makes FSP1 functionally distinct. It mediates dimerization and creates two catalytic sites positioned on opposite faces of the FAD cofactor, one dedicated to quinone reduction and one supporting an unusual hydroxylation of the flavin itself.
Catalysis follows a two-step, ping-pong logic; NAD(P)H first reduces FAD to FADH2, the reduced flavin then transfers electrons to a membrane-resident quinone acceptor. Because the acceptor site tolerates a range of isoprenoid and naphthoquinone substrates, the same active site can service ubiquinone and vitamin K, an unusual dissipation for a mammalian oxidoreductase and that underlies much of the protein’s physiological versatility. Deletion or mutation of the carboxy-terminal domain abolishes both dimerization and anti-ferroptotic capacity, confirming that the quaternary arrangement is functionally load-bearing rather than incidental.
Species differences within this fold have direct experimental consequences; for example, human and mouse FSP1 share roughly 90% sequence identity, yet a small number of divergent residues determine whether a given small-molecule inhibitor engages the enzyme at all (Xavier Da Silva et al., 2023). Investigators who assume that an inhibitor validated on human protein will behave identically in rodent or livestock systems are therefore likely to misinterpret negative results. As shown in Figure 1, the domain organization of FSP1 explains both how a single flavoprotein accommodates chemically distinct quinones and why its carboxy-terminal region is indispensable for ferroptosis suppression.
Figure 1. Crystal structure of FSP1 showing the FAD-binding domain, the NAD(P)H-binding domain and the carboxy-terminal dimerization domain, with the ubiquinone-bound active site.
3. Membrane Targeting, Subcellular Distribution and Regulation of FSP1
Myristoylation of FSP1 at the N-terminal (lipid modification) during translation causes the enzyme to attach to the plasma membrane and to lipid droplets. The localization of this enzyme is not simply a matter of how it arrives in a particular cell compartment, it dictates whether or not it will be able to provide reducing equivalents to the compartment where the actual propagation of lipid peroxide occurs. Mutating the myristoylation site creates a mutant protein with enzymatic activity which has been demonstrated to localize improperly and therefore cannot prevent damage in cultured cells nor restore tumor growth in animals transplanted from a tumor.(Wu et al., 2026). The enzyme thus behaves as a membrane-tethered electron shuttle whose activity and address are equally essential.
Expression is regulated at many different levels. The gene is transcriptionally activated as a result of the binding of Nuclear Factor Erythroid-Related Factor 2 (Nrf2) to antioxidant response elements. This results in increased expression of FSP1 due to mutations in KEAP1 and confers resistance to radiotherapy and ferroptotic Cell Death (Koppula et al., 2022). Post-transcriptional regulation also includes additional regulatory processes such as microRNA-dependent control and N6-methyladenosine (m6A)-dependent modification. FSP1 is subject to degradation via Ubiquitination by TRIM family ligases, however, it can be stabilized by deubiquitinase activity. Therefore, FSP1 has been determined to be an adjustable value in terms of its cellular content and not a static one (Jin et al., 2025).
A second regulatory aspect of FSP1 is physical instead of transcriptional. The presence of low-complexity, intrinsically disordered segments allows FSP1 to undergo liquid-liquid phase separation; this leads to the removal of the enzyme from the membrane through condensation, which renders the enzyme non-functional but does not affect its Kcat/Km values (Nakamura et al., 2023). This finding reframed the enzyme as a spatially switchable defense system and, importantly, provided a druggable mechanism distinct from active-site inhibition.
4. FSP1 as a Glutathione-Independent Ferroptosis Defense Axis
Ferroptosis is an iron-dependent, non-apoptotic mode of death defined by unrestrained peroxidation of polyunsaturated fatty acyl phospholipids, culminating in loss of membrane integrity. Two cellular strategies oppose it. GPX4 uses glutathione to reduce phospholipid hydroperoxides to the corresponding alcohols, repairing damage after it has occurred. FSP1 instead maintains a pool of reduced ubiquinol at the membrane, which intercepts lipid peroxyl radicals and terminates the autoxidation chain before hydroperoxides accumulate (Doll et al., 2019). The two systems draw on different reducing currencies, glutathione for one and NAD(P)H for the other, which is why cysteine limitation cripples one arm while leaving the other intact.
The consequences of this architecture are visible experimentally. In many cultured cell lines, deleting FSP1 has no measurable phenotype so long as GPX4 remains functional, and the enzyme’s contribution becomes apparent only when GPX4 is inhibited (Bersuker et al., 2019). This masking led to FSP1 being described for several years as a secondary or backup axis. In genetically engineered mouse models of lung adenocarcinoma, tumor-specific deletion of the gene restricted tumorigenesis to essentially the same extent as GPX4 loss, and the effect was reversed by radical-trapping antioxidants, dietary vitamin E, or deletion of the pro-ferroptotic enzyme ACSL4 (Wu et al., 2026). Cells growing in tissue evidently require far greater buffering capacity against lipid peroxidation than cells growing in nutrient-replete culture medium.
The relationship between the two arms is therefore best described as complementary rather than hierarchical, and the microenvironment sets the balance between them (Berndt et al., 2024). Figure 2 summarizes the two axes and the point at which each intervenes in the lipid peroxidation cascade.
Figure 2. Parallel ferroptosis defense systems: the GPX4–glutathione axis and the FSP1–CoQ10–NAD(P)H axis acting on lipid peroxidation.
(Wu et al., 2026)
5. The Non-Canonical Vitamin K Cycle
Vitamin K has been understood as a cofactor for γ-glutamyl carboxylase in blood coagulation. A systematic screen of naturally occurring vitamins and their derivatives revealed a second, unrelated function; the fully reduced hydroquinone forms of phylloquinone and menaquinone are potent radical-trapping antioxidants that block lipid peroxidation and rescue cells and tissues from ferroptosis (Mishima et al., 2022). The enzyme responsible for generating those hydroquinones proved to be FSP1, which reduces vitamin K at the expense of NAD(P)H in a manner insensitive to warfarin.
This result closed two questions simultaneously; it identified the long-postulated warfarin-resistant vitamin K reductase, and it explained why high-dose vitamin K is an effective antidote to warfarin poisoning; the antidotal route bypasses the inhibited canonical cycle entirely. For ferroptosis biology, the finding widened the substrate landscape of the enzyme considerably, because FSP1 reduces both the isoprenoid quinone used by animal cells and the naphthoquinones used by plants and bacteria. Figure 3 depicts this non-canonical cycle alongside the classical carboxylation pathway.
Figure 3. The FSP1-driven non-canonical vitamin K cycle, shown alongside the canonical warfarin-sensitive cycle, and its role in suppressing phospholipid peroxidation.
6. FSP1 in Disease: Tumor Dependency, Prognosis and Microenvironment
Across malignancies of the liver, colon, pancreas, stomach, breast, lung and hematopoietic system, elevated FSP1 tends to mark aggressive disease and resistance to therapies that would otherwise trigger lipid peroxidation (Jin et al., 2025). The mechanistic basis differs by context, in KEAP1-mutant lung cancers, constitutive NRF2 signaling drives FSP1 expression and produces a phenotype refractory to radiotherapy, which can be reversed pharmacologically by targeting the quinone-reducing axis (Koppula et al., 2022).
Two developments have strengthened the case for FSP1 as a therapeutic target rather than a correlative marker. In lung adenocarcinoma, FSP1 protein rises progressively as adenomas advance to adenocarcinomas, high expression predicts poorer overall survival, and pharmacological inhibition extends survival in autochthonous, orthotopic and patient-derived models (Wu et al., 2026). In metastasizing melanoma, the hypoxic and reductive lymph node niche promotes proteasomal degradation of GPX4 and lowers glutathione, so that metastatic cells become selectively dependent on FSP1; inhibitor monotherapy suppresses growth in lymph nodes but not in subcutaneous tumors (Palma et al., 2026).
Collectively, these observations describe a context-specific vulnerability rather than a universal one. The practical implication is that FSP1 dependency must be mapped to tissue site, oxygen tension, and co-mutational background before inhibition is considered, and that in vitro screening alone will systematically underestimate the enzyme’s importance (Nakamura and Conrad, 2024).
7. FSP1 in Reproductive Biology: Oocyte Competence and Early Embryonic Development
Oocytes and preimplantation embryos are unusually exposed to lipid peroxidation. They are lipid-rich, mitochondrially active, and in livestock species, routinely cultured under oxygen tensions well above those of the reproductive tract. Reproductive relevance of the FSP1 axis was first demonstrated in the pig. Pharmacological inhibition of the enzyme during in vitro culture reduced cleavage and blastocyst rates, decreased blastocyst diameter and total cell number, and raised ferrous iron and malondialdehyde without altering GPX4 abundance, confirming that the injury proceeded through a glutathione-independent route (Wang et al., 2024). The same treatment lowered mitochondrial membrane potential and ATP content while increasing reactive oxygen species and depleting glutathione, superoxide dismutase, and catalase activity.
A complementary picture has emerged in the murine oocyte. FSP1 is expressed throughout meiotic maturation and localizes to the cytoplasm, and its abundance falls markedly in ovaries and oocytes from aged animals (Ruan et al., 2025). Inhibiting the enzyme disrupted germinal vesicle breakdown and first polar body extrusion, produced spindle defects and chromosome misalignment, sustained spindle assembly checkpoint activation and arrested meiosis, with the expected accumulation of ferrous iron, reactive oxygen species and peroxidized lipids alongside aberrant mitochondrial distribution.
These findings position FSP1 as a determinant of gamete and embryo quality, but three caveats deserve emphasis. Both studies rely on pharmacological inhibition, and inhibitor selectivity across species is known to be imperfect. Germline knockout animals lacking the gene are viable and display no gross physiological defects, which suggests that developmental requirements in vivo are more permissive than acute inhibition in culture implies (Wu et al., 2026). There are limited studies that have yet examined the enzyme in the porcine oviduct, in granulosa cells, or across breeds differing in prolificacy, so the reproductive literature remains thin relative to the oncological literature.
8. Methods for Detecting and Analyzing FSP1
FSP1 can be interrogated at transcript, protein, and functional levels. Reverse-transcription quantitative PCR remains standard for AIFM2 expression, while immunoblotting reports protein abundance and immunofluorescence resolves the membrane-associated versus condensate-associated pools that phase separation generates (Nakamura et al., 2023). Because antibody performance varies considerably between species, validation against a knockout or knockdown control is advisable rather than optional.
Loss-of-function studies use either CRISPR–Cas9 deletion or small-molecule inhibitors, and the two approaches answer different questions. Inhibitors such as iFSP1, icFSP1, viFSP1 and FSEN1 differ in mechanism and in species reactivity, so an inhibitor that is potent against the human enzyme may be inactive against the rodent or porcine orthologue (Xavier Da Silva et al., 2023). Genetic deletion avoids that ambiguity but cannot capture acute, stage-specific requirements in short-lived systems such as maturing oocytes.
Functional endpoints should establish that death is ferroptotic rather than merely oxidative. Useful readouts include C11-BODIPY lipid peroxidation imaging, malondialdehyde and labile iron quantification, and mass spectrometric measurement of the reduced-to-oxidized coenzyme Q ratio together with oxidized phosphatidylcholine and phosphatidylethanolamine species (Wu et al., 2026). The decisive control in every case is rescue by a lipophilic radical-trapping antioxidant such as liproxstatin-1 or ferrostatin-1; without it, an observed phenotype cannot be attributed to ferroptosis (Dixon and Olzmann, 2024). In reproductive models, these assays are normally paired with cleavage and blastocyst rates, polar body extrusion, spindle morphology, mitochondrial membrane potential, and ATP content (Ruan et al., 2025).
9. Therapeutic Potential and Future Perspectives
The therapeutic logic surrounding FSP1 runs in two directions. In oncology, the aim is inhibition, and the enzyme offers a more favorable safety margin than GPX4; germline deletion of Gpx4 is embryonically lethal, whereas animals lacking Fsp1 develop normally, and GPX4 inhibition is toxic to T cells in a way that would blunt antitumor immunity (Wu et al., 2026). Current inhibitors nevertheless remain limited by species selectivity, modest bioavailability and, in most cases, efficacy only when GPX4 is simultaneously compromised (Nakamura and Conrad, 2024).
In degenerative disease, ischemia–reperfusion injury and gamete preservation, the objective is the opposite; sustaining or supplementing the axis. Because the enzyme reduces both ubiquinone and vitamin K, supplementation strategies have a defined biochemical rationale rather than a purely empirical one (Mishima et al., 2022). Whether such approaches translate into improved outcomes will depend on delivering the quinone to the correct membrane compartment, which is precisely the step that myristoylation-dependent targeting governs.
For animal reproductive science, the immediate priorities are descriptive. Expression of the enzyme across follicular stages, oviductal epithelium and preimplantation embryos has not been characterized in the pig, and species-appropriate inhibitors and antibodies would need validation before mechanistic work could proceed (Wang et al., 2024). Testing whether supplementing maturation, fertilization or vitrification media with coenzyme Q10 or vitamin K derivatives improves developmental competence through this pathway would be a tractable first experiment, and comparative work across breeds differing in litter size could indicate whether the axis contributes to genetic variation in reproductive performance (Jin et al., 2025).
10. Conclusion
FSP1 has moved in a decade from a misannotated apoptotic factor to a structurally defined, membrane-anchored quinone reductase that constitutes an independent arm of cellular defense against lipid peroxidation. Its catalytic architecture accommodates both ubiquinone and vitamin K, its localization and phase behavior make it a spatially regulated switch, and its transcriptional control by NRF2 links it to the broader antioxidant programme. Recent in vivo work has overturned the view of the enzyme as a mere backup to GPX4, establishing instead a context-dependent and therapeutically exploitable requirement in tumors. Early reproductive studies indicate that the same axis supports oocyte meiotic competence and porcine early embryonic development, though the evidence rests largely on pharmacological inhibition and requires genetic confirmation. FSP1 can therefore best be regarded as a conditionally essential redox enzyme whose importance is set by tissue context, and whose reproductive biology represents a substantial and largely unexplored opportunity.
References
Berndt, C., Alborzinia, H., Amen, V.S., Ayton, S., Barayeu, U., Bartelt, A., Bayir, H., Bebber, C.M., Birsoy, K., Böttcher, J.P., et al. (2024). Ferroptosis in health and disease. Redox Biology, 75, 103211. https://doi.org/10.1016/j.redox.2024.103211.
Bersuker, K., Hendricks, J.M., Li, Z., Magtanong, L., Ford, B., Tang, P.H., Roberts, M.A., Tong, B., Maimone, T.J., Zoncu, R., Bassik, M.C., Nomura, D.K., Dixon, S.J. and Olzmann, J.A. (2019). The CoQ oxidoreductase FSP1 acts parallel to GPX4 to inhibit ferroptosis. Nature, 575, pp.688–692. https://doi.org/10.1038/s41586-019-1705-2.
Dixon, S.J. and Olzmann, J.A. (2024). The cell biology of ferroptosis. Nature Reviews Molecular Cell Biology, 25, pp.424–442. https://doi.org/10.1038/s41580-024-00703-5.
Doll, S., Freitas, F.P., Shah, R., Aldrovandi, M., da Silva, M.C., Ingold, I., Goya Grocin, A., Xavier da Silva, T.N., Panzilius, E., Scheel, C.H., Mourão, A., Buday, K., Sato, M., Wanninger, J., Vignane, T., Mohana, V., Rehberg, M., Flatley, A., Schepers, A., Kurz, A., White, D., Sauer, M., Sattler, M., Tate, E.W., Schmitz, W., Schulze, A., O’Donnell, V., Proneth, B., Popowicz, G.M., Pratt, D.A., Friedmann Angeli, J.P. and Conrad, M. (2019). FSP1 is a glutathione-independent ferroptosis suppressor. Nature, 575, pp.693–698. https://doi.org/10.1038/s41586-019-1707-0.
Elguindy, M.M. and Nakamaru-Ogiso, E. (2015). Apoptosis-inducing factor (AIF) and its family member protein, AMID, are rotenone-sensitive NADH:ubiquinone oxidoreductases (NDH-2). Journal of Biological Chemistry, 290, pp.20815–20826. https://doi.org/10.1074/jbc.M115.641498.
Jiang, X., Stockwell, B.R. and Conrad, M. (2021). Ferroptosis: mechanisms, biology and role in disease. Nature Reviews Molecular Cell Biology, 22, pp.266–282. https://doi.org/10.1038/s41580-020-00324-8.
Jin, Z., Zhang, Q., Pan, Y., Chen, H., Zhou, K., Cai, H. and Huang, P. (2025). Roles and prospective applications of ferroptosis suppressor protein 1 (FSP1) in malignant tumor treatment. Current Oncology, 32(8), 456. https://doi.org/10.3390/curroncol32080456.
Koppula, P., Lei, G., Zhang, Y., Yan, Y., Mao, C., Kondiparthi, L., Shi, J., Liu, X., Horbath, A., Das, M., Li, W., Poyurovsky, M.V., Olszewski, K. and Gan, B. (2022). A targetable CoQ-FSP1 axis drives ferroptosis- and radiation-resistance in KEAP1 inactive lung cancers. Nature Communications, 13, 2206. https://doi.org/10.1038/s41467-022-29905-1.
Lv, Y., Liang, C., Sun, Q., Zhu, J., Xu, H., Li, X., Li, Y.-Y., Wang, Q., Yuan, H., Chu, B. and Zhu, D. (2023). Structural insights into FSP1 catalysis and ferroptosis inhibition. Nature Communications, 14, 5933. https://doi.org/10.1038/s41467-023-41626-7.
Mishima, E., Ito, J., Wu, Z., Nakamura, T., Wahida, A., Doll, S., Tonnus, W., Nepachalovich, P., Eggenhofer, E., Aldrovandi, M., Henkelmann, B., Yamada, K.-i., Wanninger, J., Zilka, O., Sato, E., Feederle, R., Hass, D., Maida, A., Santos Dias Mourão, A., Linkermann, A., Geissler, E.K., Nakagawa, K., Abe, T., Fedorova, M., Proneth, B., Pratt, D.A. and Conrad, M. (2022). A non-canonical vitamin K cycle is a potent ferroptosis suppressor. Nature, 608, pp.778–783. https://doi.org/10.1038/s41586-022-05022-3.
Nakamura, T. and Conrad, M. (2024). Exploiting ferroptosis vulnerabilities in cancer. Nature Cell Biology, 26, pp.1407–1419. https://doi.org/10.1038/s41556-024-01425-8.
Nakamura, T., Hipp, C., Santos Dias Mourão, A., Borggräfe, J., Aldrovandi, M., Henkelmann, B., Wanninger, J., Mishima, E., Lytton, E., Emler, D., Proneth, B., Sattler, M. and Conrad, M. (2023). Phase separation of FSP1 promotes ferroptosis. Nature, 619, pp.371–377. https://doi.org/10.1038/s41586-023-06255-6.
Palma, M., Chaufan, M., Breuer, C.B., Müller, S., Sabatier, M., Fraser, C.S., Szylo, K.J., Yavari, M., Carmona, A., Kaur, M., Martins Nascentes Melo, L., Cansiz, F., Monge-Lorenzo, J., Flores, M., Mishima, E., Nakamura, T., Proneth, B., Labrado, M., Liang, Y., Cayting, N., Zheng, L., Cañeque, T., Colombeau, L., Wahida, A., Friedmann Angeli, J.P., Tasdogan, A., Hui, S., Rodriguez, R., Conrad, M., Reticker-Flynn, N.E. and Ubellacker, J.M. (2026). Lymph node environment drives FSP1 targetability in metastasizing melanoma. Nature, 649, pp.477–486. https://doi.org/10.1038/s41586-025-09709-1.
Ruan, H., Xiang, H., Liu, Y., Wang, P., Dong, L., Cao, Y., Liang, D. and Ding, Z. (2025). FSP1 regulates ferroptosis and mitochondrial function during mouse oocyte maturation. Experimental Cell Research, 447(2), 114524. https://doi.org/10.1016/j.yexcr.2025.114524.
Wang, Y.-Q., Qu, H.-X., Dong, Y.-W., Qi, J.-J., Wei, H.-K., Sun, H., Jiang, H., Zhang, J.-B., Sun, B.-X. and Liang, S. (2024). Inhibition of FSP1 impairs early embryo developmental competence in pigs. Theriogenology, 214, pp.257–265. https://doi.org/10.1016/j.theriogenology.2023.10.025.
Wu, K., Vaughan, A.J., Bossowski, J.P., Hao, Y., Ziogou, A., Kim, S.M., Kim, T.H., Nakamura, M.N., Pillai, R., Mancini, M., Rajalingam, S., Han, M., Nakamura, T., Wang, L., Chung, S., Simeone, D., Shackelford, D., Kang, Y.P., Conrad, M. and Papagiannakopoulos, T. (2026). Targeting FSP1 triggers ferroptosis in lung cancer. Nature, 649, pp.487–495. https://doi.org/10.1038/s41586-025-09710-8.
Xavier da Silva, T.N., Schulte, C., Alves, A.N., Maric, H.M. and Friedmann Angeli, J.P. (2023). Molecular characterization of AIFM2/FSP1 inhibition by iFSP1-like molecules. Cell Death & Disease, 14, 281. https://doi.org/10.1038/s41419-023-05787-z.
Ferroptosis suppressor protein 1 (FSP1), encoded by the AIFM2 gene and previously catalogued as AMID or PRG3, is a flavin adenine dinucleotide (FAD)-containing oxidoreductase that has moved from the margins of apoptosis research to the center of redox cell biology. The protein was originally annotated as a pro-apoptotic, p53-inducible homologue of apoptosis-inducing factor. The recent work established that it behaves as a rotenone-sensitive, type-II NADH: ubiquinone oxidoreductase rather than as a nuclease (Elguindy and Nakamaru-Ogiso, 2015). That biochemical description acquired physiological meaning in 2019, when two independent expression-cloning and CRISPR screens identified the same flavoprotein as the factor that allows cells to survive genetic or pharmacological loss of glutathione peroxidase 4 (Doll et al., 2019). The gene was renamed accordingly, and the FSP1–coenzyme Q10–NAD(P)H module is now recognized as the second mainstay of ferroptosis surveillance (Bersuker et al., 2019).
Among the redox enzymes, FSP1 remains unusual as it is not a peroxidase. It does not repair oxidized phospholipids but continuously regenerates a small lipophilic radical-trapping antioxidant inside the membrane, so that lipid peroxyl radicals are intercepted before propagation begins. This mechanistic separation explains why it operates in parallel with, rather than downstream of, the cysteine–glutathione–GPX4 axis (Jiang et al., 2021). The distinction matters clinically because tumors and injured tissues can shift between the two arms depending on nutrient supply, oxygen tension, and mutational background (Dixon and Olzmann, 2024).
The growing interest in FSP1 has increased due to several reasons, including its structural biology, which has resolved how a single flavoprotein reduces chemically diverse quinones. It was shown to sustain a warfarin-insensitive vitamin K cycle, solving a question that has been open in vitamin biology for many years. Lastly, in vivo genetic models have demonstrated that FSP1 is not merely a backup system but an independent requirement for tumor growth. Besides this progress, reproductive biologists have begun to report that FSP1 governs oocyte meiotic competence and porcine early embryonic development, extending the protein’s relevance from oncology into gamete and embryo physiology (Wang et al., 2024). This review examines FSP1 from a protein-centered perspective, covering its structure, catalysis, regulation, disease associations, reproductive functions, detection methods, and therapeutic prospects.
2. Structural Organization of FSP1
Crystallographic analysis of vertebrate FSP1 in substrate-free and ubiquinone-bound states has defined a three-part architecture; an FAD-binding domain, an NAD(P)H-binding domain, and a distinctive carboxy-terminal domain (Lv et al., 2023). The first two modules are shared with apoptosis-inducing factor and with bacterial NADH oxidoreductases, which accounts for the historical misassignment of the protein to the apoptotic machinery. The carboxy-terminal domain is the element that makes FSP1 functionally distinct. It mediates dimerization and creates two catalytic sites positioned on opposite faces of the FAD cofactor, one dedicated to quinone reduction and one supporting an unusual hydroxylation of the flavin itself.
Catalysis follows a two-step, ping-pong logic; NAD(P)H first reduces FAD to FADH2, the reduced flavin then transfers electrons to a membrane-resident quinone acceptor. Because the acceptor site tolerates a range of isoprenoid and naphthoquinone substrates, the same active site can service ubiquinone and vitamin K, an unusual dissipation for a mammalian oxidoreductase and that underlies much of the protein’s physiological versatility. Deletion or mutation of the carboxy-terminal domain abolishes both dimerization and anti-ferroptotic capacity, confirming that the quaternary arrangement is functionally load-bearing rather than incidental.
Species differences within this fold have direct experimental consequences; for example, human and mouse FSP1 share roughly 90% sequence identity, yet a small number of divergent residues determine whether a given small-molecule inhibitor engages the enzyme at all (Xavier Da Silva et al., 2023). Investigators who assume that an inhibitor validated on human protein will behave identically in rodent or livestock systems are therefore likely to misinterpret negative results. As shown in Figure 1, the domain organization of FSP1 explains both how a single flavoprotein accommodates chemically distinct quinones and why its carboxy-terminal region is indispensable for ferroptosis suppression.
Figure 1. Crystal structure of FSP1 showing the FAD-binding domain, the NAD(P)H-binding domain and the carboxy-terminal dimerization domain, with the ubiquinone-bound active site.
3. Membrane Targeting, Subcellular Distribution and Regulation of FSP1
Myristoylation of FSP1 at the N-terminal (lipid modification) during translation causes the enzyme to attach to the plasma membrane and to lipid droplets. The localization of this enzyme is not simply a matter of how it arrives in a particular cell compartment, it dictates whether or not it will be able to provide reducing equivalents to the compartment where the actual propagation of lipid peroxide occurs. Mutating the myristoylation site creates a mutant protein with enzymatic activity which has been demonstrated to localize improperly and therefore cannot prevent damage in cultured cells nor restore tumor growth in animals transplanted from a tumor.(Wu et al., 2026). The enzyme thus behaves as a membrane-tethered electron shuttle whose activity and address are equally essential.
Expression is regulated at many different levels. The gene is transcriptionally activated as a result of the binding of Nuclear Factor Erythroid-Related Factor 2 (Nrf2) to antioxidant response elements. This results in increased expression of FSP1 due to mutations in KEAP1 and confers resistance to radiotherapy and ferroptotic Cell Death (Koppula et al., 2022). Post-transcriptional regulation also includes additional regulatory processes such as microRNA-dependent control and N6-methyladenosine (m6A)-dependent modification. FSP1 is subject to degradation via Ubiquitination by TRIM family ligases, however, it can be stabilized by deubiquitinase activity. Therefore, FSP1 has been determined to be an adjustable value in terms of its cellular content and not a static one (Jin et al., 2025).
A second regulatory aspect of FSP1 is physical instead of transcriptional. The presence of low-complexity, intrinsically disordered segments allows FSP1 to undergo liquid-liquid phase separation; this leads to the removal of the enzyme from the membrane through condensation, which renders the enzyme non-functional but does not affect its Kcat/Km values (Nakamura et al., 2023). This finding reframed the enzyme as a spatially switchable defense system and, importantly, provided a druggable mechanism distinct from active-site inhibition.
4. FSP1 as a Glutathione-Independent Ferroptosis Defense Axis
Ferroptosis is an iron-dependent, non-apoptotic mode of death defined by unrestrained peroxidation of polyunsaturated fatty acyl phospholipids, culminating in loss of membrane integrity. Two cellular strategies oppose it. GPX4 uses glutathione to reduce phospholipid hydroperoxides to the corresponding alcohols, repairing damage after it has occurred. FSP1 instead maintains a pool of reduced ubiquinol at the membrane, which intercepts lipid peroxyl radicals and terminates the autoxidation chain before hydroperoxides accumulate (Doll et al., 2019). The two systems draw on different reducing currencies, glutathione for one and NAD(P)H for the other, which is why cysteine limitation cripples one arm while leaving the other intact.
The consequences of this architecture are visible experimentally. In many cultured cell lines, deleting FSP1 has no measurable phenotype so long as GPX4 remains functional, and the enzyme’s contribution becomes apparent only when GPX4 is inhibited (Bersuker et al., 2019). This masking led to FSP1 being described for several years as a secondary or backup axis. In genetically engineered mouse models of lung adenocarcinoma, tumor-specific deletion of the gene restricted tumorigenesis to essentially the same extent as GPX4 loss, and the effect was reversed by radical-trapping antioxidants, dietary vitamin E, or deletion of the pro-ferroptotic enzyme ACSL4 (Wu et al., 2026). Cells growing in tissue evidently require far greater buffering capacity against lipid peroxidation than cells growing in nutrient-replete culture medium.
The relationship between the two arms is therefore best described as complementary rather than hierarchical, and the microenvironment sets the balance between them (Berndt et al., 2024). Figure 2 summarizes the two axes and the point at which each intervenes in the lipid peroxidation cascade.
Figure 2. Parallel ferroptosis defense systems: the GPX4–glutathione axis and the FSP1–CoQ10–NAD(P)H axis acting on lipid peroxidation.
(Wu et al., 2026)
5. The Non-Canonical Vitamin K Cycle
Vitamin K has been understood as a cofactor for γ-glutamyl carboxylase in blood coagulation. A systematic screen of naturally occurring vitamins and their derivatives revealed a second, unrelated function; the fully reduced hydroquinone forms of phylloquinone and menaquinone are potent radical-trapping antioxidants that block lipid peroxidation and rescue cells and tissues from ferroptosis (Mishima et al., 2022). The enzyme responsible for generating those hydroquinones proved to be FSP1, which reduces vitamin K at the expense of NAD(P)H in a manner insensitive to warfarin.
This result closed two questions simultaneously; it identified the long-postulated warfarin-resistant vitamin K reductase, and it explained why high-dose vitamin K is an effective antidote to warfarin poisoning; the antidotal route bypasses the inhibited canonical cycle entirely. For ferroptosis biology, the finding widened the substrate landscape of the enzyme considerably, because FSP1 reduces both the isoprenoid quinone used by animal cells and the naphthoquinones used by plants and bacteria. Figure 3 depicts this non-canonical cycle alongside the classical carboxylation pathway.
Figure 3. The FSP1-driven non-canonical vitamin K cycle, shown alongside the canonical warfarin-sensitive cycle, and its role in suppressing phospholipid peroxidation.
6. FSP1 in Disease: Tumor Dependency, Prognosis and Microenvironment
Across malignancies of the liver, colon, pancreas, stomach, breast, lung and hematopoietic system, elevated FSP1 tends to mark aggressive disease and resistance to therapies that would otherwise trigger lipid peroxidation (Jin et al., 2025). The mechanistic basis differs by context, in KEAP1-mutant lung cancers, constitutive NRF2 signaling drives FSP1 expression and produces a phenotype refractory to radiotherapy, which can be reversed pharmacologically by targeting the quinone-reducing axis (Koppula et al., 2022).
Two developments have strengthened the case for FSP1 as a therapeutic target rather than a correlative marker. In lung adenocarcinoma, FSP1 protein rises progressively as adenomas advance to adenocarcinomas, high expression predicts poorer overall survival, and pharmacological inhibition extends survival in autochthonous, orthotopic and patient-derived models (Wu et al., 2026). In metastasizing melanoma, the hypoxic and reductive lymph node niche promotes proteasomal degradation of GPX4 and lowers glutathione, so that metastatic cells become selectively dependent on FSP1; inhibitor monotherapy suppresses growth in lymph nodes but not in subcutaneous tumors (Palma et al., 2026).
Collectively, these observations describe a context-specific vulnerability rather than a universal one. The practical implication is that FSP1 dependency must be mapped to tissue site, oxygen tension, and co-mutational background before inhibition is considered, and that in vitro screening alone will systematically underestimate the enzyme’s importance (Nakamura and Conrad, 2024).
7. FSP1 in Reproductive Biology: Oocyte Competence and Early Embryonic Development
Oocytes and preimplantation embryos are unusually exposed to lipid peroxidation. They are lipid-rich, mitochondrially active, and in livestock species, routinely cultured under oxygen tensions well above those of the reproductive tract. Reproductive relevance of the FSP1 axis was first demonstrated in the pig. Pharmacological inhibition of the enzyme during in vitro culture reduced cleavage and blastocyst rates, decreased blastocyst diameter and total cell number, and raised ferrous iron and malondialdehyde without altering GPX4 abundance, confirming that the injury proceeded through a glutathione-independent route (Wang et al., 2024). The same treatment lowered mitochondrial membrane potential and ATP content while increasing reactive oxygen species and depleting glutathione, superoxide dismutase, and catalase activity.
A complementary picture has emerged in the murine oocyte. FSP1 is expressed throughout meiotic maturation and localizes to the cytoplasm, and its abundance falls markedly in ovaries and oocytes from aged animals (Ruan et al., 2025). Inhibiting the enzyme disrupted germinal vesicle breakdown and first polar body extrusion, produced spindle defects and chromosome misalignment, sustained spindle assembly checkpoint activation and arrested meiosis, with the expected accumulation of ferrous iron, reactive oxygen species and peroxidized lipids alongside aberrant mitochondrial distribution.
These findings position FSP1 as a determinant of gamete and embryo quality, but three caveats deserve emphasis. Both studies rely on pharmacological inhibition, and inhibitor selectivity across species is known to be imperfect. Germline knockout animals lacking the gene are viable and display no gross physiological defects, which suggests that developmental requirements in vivo are more permissive than acute inhibition in culture implies (Wu et al., 2026). There are limited studies that have yet examined the enzyme in the porcine oviduct, in granulosa cells, or across breeds differing in prolificacy, so the reproductive literature remains thin relative to the oncological literature.
8. Methods for Detecting and Analyzing FSP1
FSP1 can be interrogated at transcript, protein, and functional levels. Reverse-transcription quantitative PCR remains standard for AIFM2 expression, while immunoblotting reports protein abundance and immunofluorescence resolves the membrane-associated versus condensate-associated pools that phase separation generates (Nakamura et al., 2023). Because antibody performance varies considerably between species, validation against a knockout or knockdown control is advisable rather than optional.
Loss-of-function studies use either CRISPR–Cas9 deletion or small-molecule inhibitors, and the two approaches answer different questions. Inhibitors such as iFSP1, icFSP1, viFSP1 and FSEN1 differ in mechanism and in species reactivity, so an inhibitor that is potent against the human enzyme may be inactive against the rodent or porcine orthologue (Xavier Da Silva et al., 2023). Genetic deletion avoids that ambiguity but cannot capture acute, stage-specific requirements in short-lived systems such as maturing oocytes.
Functional endpoints should establish that death is ferroptotic rather than merely oxidative. Useful readouts include C11-BODIPY lipid peroxidation imaging, malondialdehyde and labile iron quantification, and mass spectrometric measurement of the reduced-to-oxidized coenzyme Q ratio together with oxidized phosphatidylcholine and phosphatidylethanolamine species (Wu et al., 2026). The decisive control in every case is rescue by a lipophilic radical-trapping antioxidant such as liproxstatin-1 or ferrostatin-1; without it, an observed phenotype cannot be attributed to ferroptosis (Dixon and Olzmann, 2024). In reproductive models, these assays are normally paired with cleavage and blastocyst rates, polar body extrusion, spindle morphology, mitochondrial membrane potential, and ATP content (Ruan et al., 2025).
9. Therapeutic Potential and Future Perspectives
The therapeutic logic surrounding FSP1 runs in two directions. In oncology, the aim is inhibition, and the enzyme offers a more favorable safety margin than GPX4; germline deletion of Gpx4 is embryonically lethal, whereas animals lacking Fsp1 develop normally, and GPX4 inhibition is toxic to T cells in a way that would blunt antitumor immunity (Wu et al., 2026). Current inhibitors nevertheless remain limited by species selectivity, modest bioavailability and, in most cases, efficacy only when GPX4 is simultaneously compromised (Nakamura and Conrad, 2024).
In degenerative disease, ischemia–reperfusion injury and gamete preservation, the objective is the opposite; sustaining or supplementing the axis. Because the enzyme reduces both ubiquinone and vitamin K, supplementation strategies have a defined biochemical rationale rather than a purely empirical one (Mishima et al., 2022). Whether such approaches translate into improved outcomes will depend on delivering the quinone to the correct membrane compartment, which is precisely the step that myristoylation-dependent targeting governs.
For animal reproductive science, the immediate priorities are descriptive. Expression of the enzyme across follicular stages, oviductal epithelium and preimplantation embryos has not been characterized in the pig, and species-appropriate inhibitors and antibodies would need validation before mechanistic work could proceed (Wang et al., 2024). Testing whether supplementing maturation, fertilization or vitrification media with coenzyme Q10 or vitamin K derivatives improves developmental competence through this pathway would be a tractable first experiment, and comparative work across breeds differing in litter size could indicate whether the axis contributes to genetic variation in reproductive performance (Jin et al., 2025).
10. Conclusion
FSP1 has moved in a decade from a misannotated apoptotic factor to a structurally defined, membrane-anchored quinone reductase that constitutes an independent arm of cellular defense against lipid peroxidation. Its catalytic architecture accommodates both ubiquinone and vitamin K, its localization and phase behavior make it a spatially regulated switch, and its transcriptional control by NRF2 links it to the broader antioxidant programme. Recent in vivo work has overturned the view of the enzyme as a mere backup to GPX4, establishing instead a context-dependent and therapeutically exploitable requirement in tumors. Early reproductive studies indicate that the same axis supports oocyte meiotic competence and porcine early embryonic development, though the evidence rests largely on pharmacological inhibition and requires genetic confirmation. FSP1 can therefore best be regarded as a conditionally essential redox enzyme whose importance is set by tissue context, and whose reproductive biology represents a substantial and largely unexplored opportunity.
References
Berndt, C., Alborzinia, H., Amen, V.S., Ayton, S., Barayeu, U., Bartelt, A., Bayir, H., Bebber, C.M., Birsoy, K., Böttcher, J.P., et al. (2024). Ferroptosis in health and disease. Redox Biology, 75, 103211. https://doi.org/10.1016/j.redox.2024.103211.
Bersuker, K., Hendricks, J.M., Li, Z., Magtanong, L., Ford, B., Tang, P.H., Roberts, M.A., Tong, B., Maimone, T.J., Zoncu, R., Bassik, M.C., Nomura, D.K., Dixon, S.J. and Olzmann, J.A. (2019). The CoQ oxidoreductase FSP1 acts parallel to GPX4 to inhibit ferroptosis. Nature, 575, pp.688–692. https://doi.org/10.1038/s41586-019-1705-2.
Dixon, S.J. and Olzmann, J.A. (2024). The cell biology of ferroptosis. Nature Reviews Molecular Cell Biology, 25, pp.424–442. https://doi.org/10.1038/s41580-024-00703-5.
Doll, S., Freitas, F.P., Shah, R., Aldrovandi, M., da Silva, M.C., Ingold, I., Goya Grocin, A., Xavier da Silva, T.N., Panzilius, E., Scheel, C.H., Mourão, A., Buday, K., Sato, M., Wanninger, J., Vignane, T., Mohana, V., Rehberg, M., Flatley, A., Schepers, A., Kurz, A., White, D., Sauer, M., Sattler, M., Tate, E.W., Schmitz, W., Schulze, A., O’Donnell, V., Proneth, B., Popowicz, G.M., Pratt, D.A., Friedmann Angeli, J.P. and Conrad, M. (2019). FSP1 is a glutathione-independent ferroptosis suppressor. Nature, 575, pp.693–698. https://doi.org/10.1038/s41586-019-1707-0.
Elguindy, M.M. and Nakamaru-Ogiso, E. (2015). Apoptosis-inducing factor (AIF) and its family member protein, AMID, are rotenone-sensitive NADH:ubiquinone oxidoreductases (NDH-2). Journal of Biological Chemistry, 290, pp.20815–20826. https://doi.org/10.1074/jbc.M115.641498.
Jiang, X., Stockwell, B.R. and Conrad, M. (2021). Ferroptosis: mechanisms, biology and role in disease. Nature Reviews Molecular Cell Biology, 22, pp.266–282. https://doi.org/10.1038/s41580-020-00324-8.
Jin, Z., Zhang, Q., Pan, Y., Chen, H., Zhou, K., Cai, H. and Huang, P. (2025). Roles and prospective applications of ferroptosis suppressor protein 1 (FSP1) in malignant tumor treatment. Current Oncology, 32(8), 456. https://doi.org/10.3390/curroncol32080456.
Koppula, P., Lei, G., Zhang, Y., Yan, Y., Mao, C., Kondiparthi, L., Shi, J., Liu, X., Horbath, A., Das, M., Li, W., Poyurovsky, M.V., Olszewski, K. and Gan, B. (2022). A targetable CoQ-FSP1 axis drives ferroptosis- and radiation-resistance in KEAP1 inactive lung cancers. Nature Communications, 13, 2206. https://doi.org/10.1038/s41467-022-29905-1.
Lv, Y., Liang, C., Sun, Q., Zhu, J., Xu, H., Li, X., Li, Y.-Y., Wang, Q., Yuan, H., Chu, B. and Zhu, D. (2023). Structural insights into FSP1 catalysis and ferroptosis inhibition. Nature Communications, 14, 5933. https://doi.org/10.1038/s41467-023-41626-7.
Mishima, E., Ito, J., Wu, Z., Nakamura, T., Wahida, A., Doll, S., Tonnus, W., Nepachalovich, P., Eggenhofer, E., Aldrovandi, M., Henkelmann, B., Yamada, K.-i., Wanninger, J., Zilka, O., Sato, E., Feederle, R., Hass, D., Maida, A., Santos Dias Mourão, A., Linkermann, A., Geissler, E.K., Nakagawa, K., Abe, T., Fedorova, M., Proneth, B., Pratt, D.A. and Conrad, M. (2022). A non-canonical vitamin K cycle is a potent ferroptosis suppressor. Nature, 608, pp.778–783. https://doi.org/10.1038/s41586-022-05022-3.
Nakamura, T. and Conrad, M. (2024). Exploiting ferroptosis vulnerabilities in cancer. Nature Cell Biology, 26, pp.1407–1419. https://doi.org/10.1038/s41556-024-01425-8.
Nakamura, T., Hipp, C., Santos Dias Mourão, A., Borggräfe, J., Aldrovandi, M., Henkelmann, B., Wanninger, J., Mishima, E., Lytton, E., Emler, D., Proneth, B., Sattler, M. and Conrad, M. (2023). Phase separation of FSP1 promotes ferroptosis. Nature, 619, pp.371–377. https://doi.org/10.1038/s41586-023-06255-6.
Palma, M., Chaufan, M., Breuer, C.B., Müller, S., Sabatier, M., Fraser, C.S., Szylo, K.J., Yavari, M., Carmona, A., Kaur, M., Martins Nascentes Melo, L., Cansiz, F., Monge-Lorenzo, J., Flores, M., Mishima, E., Nakamura, T., Proneth, B., Labrado, M., Liang, Y., Cayting, N., Zheng, L., Cañeque, T., Colombeau, L., Wahida, A., Friedmann Angeli, J.P., Tasdogan, A., Hui, S., Rodriguez, R., Conrad, M., Reticker-Flynn, N.E. and Ubellacker, J.M. (2026). Lymph node environment drives FSP1 targetability in metastasizing melanoma. Nature, 649, pp.477–486. https://doi.org/10.1038/s41586-025-09709-1.
Ruan, H., Xiang, H., Liu, Y., Wang, P., Dong, L., Cao, Y., Liang, D. and Ding, Z. (2025). FSP1 regulates ferroptosis and mitochondrial function during mouse oocyte maturation. Experimental Cell Research, 447(2), 114524. https://doi.org/10.1016/j.yexcr.2025.114524.
Wang, Y.-Q., Qu, H.-X., Dong, Y.-W., Qi, J.-J., Wei, H.-K., Sun, H., Jiang, H., Zhang, J.-B., Sun, B.-X. and Liang, S. (2024). Inhibition of FSP1 impairs early embryo developmental competence in pigs. Theriogenology, 214, pp.257–265. https://doi.org/10.1016/j.theriogenology.2023.10.025.
Wu, K., Vaughan, A.J., Bossowski, J.P., Hao, Y., Ziogou, A., Kim, S.M., Kim, T.H., Nakamura, M.N., Pillai, R., Mancini, M., Rajalingam, S., Han, M., Nakamura, T., Wang, L., Chung, S., Simeone, D., Shackelford, D., Kang, Y.P., Conrad, M. and Papagiannakopoulos, T. (2026). Targeting FSP1 triggers ferroptosis in lung cancer. Nature, 649, pp.487–495. https://doi.org/10.1038/s41586-025-09710-8.
Xavier da Silva, T.N., Schulte, C., Alves, A.N., Maric, H.M. and Friedmann Angeli, J.P. (2023). Molecular characterization of AIFM2/FSP1 inhibition by iFSP1-like molecules. Cell Death & Disease, 14, 281. https://doi.org/10.1038/s41419-023-05787-z.
